Colicin M is inactivated during import by its immunity protein

Colicin M (Cma) displays a unique activity that interferes with murein and O-antigen biosynthesis through inhibition of lipid-carrier regeneration. Immunity is conferred by a specific immunity protein (Cmi) that inhibits the action of colicin M in the periplasm. The subcellular location of Cmi was determined by constructing hybrid proteins between Cmi and the TEM--lactamase (BlaM), which confers resistance to ampicillin only when it is translocated across the cytoplasmic membrane with the aid of Cmi. The smallest Cmi'-BlaM hybrid that conferred resistance to 50 g/ml ampicillin contained 19 amino acid residues of Cmi; cells expressing Cmi'-BlaM with only five N-terminal Cmi residues were ampicillin sensitive. These results support a model in which the hydrophobic sequence of Cmi comprising residues 3–23 serves to translocate residues 24–117 of Cmi into the periplasm and anchors Cmi to the cytoplasmic membrane. Residues 8–23 are integrated in the cytoplasmic membrane and are not involved in Cma recognition. This model was further tested by replacing residues 1–23 of Cmi by the hydrophobic amino acid sequence 1–42 of the penicillin binding protein 3 (PBP3). In vivo, PBP3'-'Cmi was as active as Cmi, demonstrating that translocation and anchoring of Cmi is not sequence-specific. Substitution of the 23 N-terminal residues of Cmi by the cleavable signal peptide of BlaM resulted in an active BlaM'-'Cmi hybrid protein. The immunity conferred by BlaM'-'Cmi was high, but not as high as that associated with Cmi and PBP3'-'Cmi, demonstrating that soluble Cmi lacking its membrane anchor is still active, but immobilization in the cytoplasmic membrane, the target site of Cma, increases its efficiency. Cmi1-23 remained in the cytoplasm and conferred no immunity. We propose that the immunity protein inactivates colicin M in the periplasm before Cma can reach its target in the cytoplasmic membrane.     

The crystal structure of the dimeric colicin M immunity protein displays a 3D domain swap

Bacteriocins are proteins secreted by many bacterial cells to kill related bacteria of the same niche. To avoid their own suicide through reuptake of secreted bacteriocins, these bacteria protect themselves by co-expression of immunity proteins in the compartment of colicin destination. In Escherichia coli the colicin M (Cma) is inactivated by the interaction with the Cma immunity protein (Cmi). We have crystallized and solved the structure of Cmi at a resolution of 1.95? by the recently developed ab initio phasing program ARCIMBOLDO. The monomeric structure of the mature 10kDa protein comprises a long N-terminal α-helix and a four-stranded C-terminal β-sheet. Dimerization of this fold is mediated by an extended interface of hydrogen bond interactions between the α-helix and the four-stranded β-sheet of the symmetry related molecule. Two intermolecular disulfide bridges covalently connect this dimer to further lock this complex. The Cmi protein resembles an example of a 3D domain swapping being stalled through physical linkage. The dimer is a highly charged complex with a significant surplus of negative charges presumably responsible for interactions with Cma. Dimerization of Cmi was also demonstrated to occur in vivo. Although the Cmi-Cma complex is unique among bacteria, the general fold of Cmi is representative for a class of YebF-like proteins which are known to be secreted into the external medium by some Gram-negative bacteria.     

The biology of colicin M

This communication summarizes our present knowledge of colicin M, an unusual member of the colicin group. The gene encoding colicin M, cma, has been sequenced and the protein isolated and purified. With a deduced molecular size of 29,453 Da, colicin M is the smallest of the known colicins. The polypeptide can be divided into functional domains for cell surface receptor binding, uptake into the cell, and killing activity. To kill, the colicin must enter from outside the cell. Colicin M blocks the biosynthesis of both peptidoglycan and O-antigen by inhibiting regeneration of the bactoprenyl-P carrier lipid. Autolysis occurs as a secondary effect following inhibition of peptidoglycan synthesis. Colicin M is the only colicin known to have such a mechanism of action. Immunity to this colicin is mediated by the cmi gene product, a protein of 13,890 Da. This cytoplasmic membrane protein confers immunity by binding to and thus neutralizing the colicin. Cmi shares properties with both immunity proteins of the pore-forming and the cytoplasmically active colicins. Genes for the colicin and immunity protein are found next to each other, but in opposite orientation, on pColM plasmids. The mechanism of colicin M release is not known.     

The biology of colicin M

Abstract This communication summarizes our present knowledge of colicin M, an unusual member of the colicin group. The gene encoding colicin M, cma , has been sequenced and the protein isolated and purified. With a deduced molecular size of 29 453 Da, colicin M is the smallest of the known colicins. The polypeptide can be divided into functional domains for cell surface receptor binding, uptake into the cell, and killing activity. To kill, the colicin must enter from outside the cell. Colicin M blocks the biosynthesis of both peptidoglycan and O-antigen by inhibiting regeneration of the bactoprenyl-P carrier lipid. Autolysis occurs as a secondary effect following inhibition of peptidoglycan synthesis. Colicin M is the only colicin known to have such a mechanism of action. Immunity to this colicin is mediated by the cmi gene product, a protein of 13 890 Da. This cytoplasmic membrane protein confers immunity by binding to and thus neutralizing the colicin. Cmi shares properties with both immunity proteins of the pore-forming and the cytoplasmically active colicins. Genes for the colicin and immunity protein are found next to each other, but in opposite orientation, on pColM plasmids. The mechanism of colicin M release is not known.     

Mapping functional domains of colicin M

Colicin M (Cma) lyses Escherichia coli cells by inhibiting murein biosynthesis through hydrolysis of the phosphate ester between C(55)-polyisoprenol and N-acetylmuramyl (MurNAc)-pentapeptide-GlcNAc in the periplasm. To identify Cma functional domains, we isolated 54 point mutants and small deletion mutants and examined their cytotoxicity levels. Activity and uptake mutants were distinguished by osmotic shock, which transfers Cma into the periplasm independent of the specific FhuA receptor and the Ton system. Deletion of the hydrophobic helix α1, which extends from the compact Cma structure, abolished interference with the antibiotic albomycin, which is transported across the outer membrane by the same system as Cma, thereby identifying α1 as the Cma site that binds to FhuA. Deletion of the C-terminal Lys-Arg strongly reduced Cma translocation across the outer membrane after binding to FhuA. Conversion of Asp226 to Glu, Asn, or Ala inactivated Cma. Asp226 is exposed at the Cma surface and is surrounded by Asp225, Asp229, His235, Tyr228, and Arg236; replacement of each with alanine inactivated Cma. We propose that Asp226 directly participates in phosphate ester hydrolysis and that the surrounding residues contribute to the active site. These residues are strongly conserved in Cma-like proteins of other species. Replacement of other conserved residues with alanine inactivated Cma; these mutations probably altered the Cma structure, as particularly apparent for mutants in the unique open β-barrel of Cma, which were isolated in lower yields. Our results identify regions in Cma responsible for uptake and activity and support the concept of a three-domain arrangement of Cma.     

Activation of colicin M by the FkpA prolyl cis-trans isomerase/chaperone

Colicin M (Cma) is specifically imported into the periplasm of Escherichia coli and kills the cells. Killing depends on the periplasmic peptidyl prolyl cis-trans isomerase/chaperone FkpA. To identify the Cma prolyl bonds targeted by FkpA, we replaced the 15 proline residues individually with alanine. Seven mutant proteins were fully active; Cma(P129A), Cma(P176A), and Cma(P260A) displayed 1%, and Cma(P107A) displayed 10% of the wild-type activity. Cma(P107A), Cma(P129A), and Cma(P260A), but not Cma(P176A), killed cells after entering the periplasm via osmotic shock, indicating that the former mutants were translocation-deficient; Cma(P129A) did not bind to the FhuA outer membrane receptor. The crystal structures of Cma and Cma(P176A) were identical, excluding inactivation of the activity domain located far from Pro-176. In a new peptidyl prolyl cis-trans isomerase assay, FkpA isomerized the Cma prolyl bond in peptide Phe-Pro-176 at a high rate, but Lys-Pro-107 and Leu-Pro-260 isomerized at only <10% of that rate. The four mutant proteins secreted into the periplasm via a fused signal sequence were toxic but much less than wild-type Cma. Wild-type and mutant Cma proteins secreted or translocated across the outer membrane by energy-coupled import or unspecific osmotic shock were only active in the presence of FkpA. We propose that Cma unfolds during transfer across the outer or cytoplasmic membrane and refolds to the active form in the periplasm assisted by FkpA. Weak refolding of Cma(P176A) would explain its low activity in all assays. Of the four proline residues identified as being important for Cma activity, Phe-Pro-176 is most likely targeted by FkpA.     

X-ray structure and site-directed mutagenesis analysis of the Escherichia coli colicin M immunity protein

Colicin M (ColM), which is produced by some Escherichia coli strains to kill competitor strains from the same or related species, was recently shown to inhibit cell wall peptidoglycan biosynthesis through enzymatic degradation of its lipid II precursor. ColM-producing strains are protected from the toxin that they produce by coexpression of a specific immunity protein, named Cmi, whose mode of action still remains to be identified. We report here the resolution of the crystal structure of Cmi, which is composed of four β strands and four α helices. This rather compact structure revealed a disulfide bond between residues Cys31 and Cys107. Interestingly, these two cysteines and several other residues appeared to be conserved in the sequences of several proteins of unknown function belonging to the YebF family which exhibit 25 to 35% overall sequence similarity with Cmi. Site-directed mutagenesis was performed to assess the role of these residues in the ColM immunity-conferring activity of Cmi, which showed that the disulfide bond and residues from the C-terminal extremity of the protein were functionally essential. The involvement of DsbA oxidase in the formation of the Cmi disulfide bond is also demonstrated.     

Studies on the deoxyribonucleic acid-bearing portion of the cell envelope of Escherichia coli.

The envelope components of nuclear bodies which were obtained from Escherichia coli W7 by a mild lysis method were investigated. By using 2,6-diaminopimelic acid (DAP) as precursor which is incorporated only into peptidoglycan in this strain it was found that the particles contained about 14% of the murein layer of the cell. The percentage of phosphatidylethanolamine was enriched at the cost of the other phospholipids in the nuclear bodies compared to whole cells. If lipids were labelled with 3H-palmitic acid the cytoplasmic and the outer membrane could be found after isopycnic centrifugation; however, when the cells were incubated with chloramphenicol, only the outer membrane was seen. The peptidoglycan and the proteins could be assigned only to the outer membrane. The DNA is also bound to the outer membrane. From these results it was concluded that (1) in all lysis methods the cytoplasmic membrane is more easily dissolved than the outer layers of the envelope, and (2) that there is a firm binding between DNA and the outer membrane in vivo.     

The bacteriocin lactococcin A specifically increases permeability of lactococcal cytoplasmic membranes in a voltage-independent, protein-mediated manner.

Lactococcin A is a bacteriocin produced by Lactococcus lactis. Its structural gene has recently been cloned and sequenced (M. J. van Belkum, B. J. Hayema, R. E. Jeeninga, J. Kok, and G. Venema, Appl. Environ. Microbiol. 57:492-498, 1991). Purified lactococcin A increased the permeability of the cytoplasmic membrane of L. lactis and dissipated the membrane potential. A significantly higher concentration of lactococcin A was needed to dissipate the membrane potential in an immune strain of L. lactis. Lactococcin A at low concentrations (0.029 microgram/mg of protein) inhibited secondary and phosphate-bond driven transport of amino acids in sensitive cells and caused efflux of preaccumulated amino acids. Accumulation of amino acids by immune cells was not affected by this concentration of lactococcin A. Lactococcin A also inhibited proton motive force-driven leucine uptake and leucine counterflow in membrane vesicles of the sensitive strain but not in membrane vesicles of the immune strain. These observations indicate that lactococcin A makes the membrane permeable for leucine in the presence or absence of a proton motive force and that the immunity factor(s) is membrane linked. Membrane vesicles of Clostridium acetobutylicum, Bacillus subtilis, and Escherichia coli were not affected by lactococcin A, nor were liposomes derived from phospholipids of L. lactis. These results indicate that lactococcin A acts on the cytoplasmic membrane and is very specific towards lactococci. The combined results obtained with cells, vesicles, and liposomes suggest that the specificity of lactococcin A may be mediated by a receptor protein associated with the cytoplasmic membrane.     

The TraB protein, which mediates the intermycelial transfer of the Streptomyces plasmid pSN22, has functional NTP-binding motifs and is localized to the cytoplasmic membrane

The traB gene on the Streptomyces conjugative plasmid pSN22 is required for intermycelial plasmid transfer and the mobilization of chromosomal markers (Cma). The predicted amino acid sequence of TraB contains one Walker type-A and two type-B NTP-binding motifs. Site-directed mutagenesis revealed that the type-A motif and one of the type-B motifs, 109 amino acid residues downstream of the type-A motif, were essential for both plasmid transfer and Cma. The second type-B sequence could be changed without any phenotypic effect. A modified traB gene was constructed, resulting in the production of a functional protein with an amino-terminal c-Myc epitope tag for immunological analysis. This protein was associated with the cytoplasmic membrane, suggesting that TraB is a membrane protein that uses energy from ATP hydrolysis to transport DNA between mycelia. The c-Myc tagging of TraB decreased the efficiency of intramycelial plasmid spread, suggesting that TraB is involved in both inter- and intramycelial transfer processes.     

The use of mutants in the study of structure-function relationships in cloacin DF13.

A bacteriocin from cells with a mutant Clo DF13 plasmid (cloacin clp03 . immunity protein complex) and a bacteriocin from cells containing the recombinant plasmic Clo DF13 :: Tn901 (cloacin pJN82) have been isolated. Both bacteriocins like wild-type cloacin DF13, are still able to inhibit in vitro protein synthesis, but their in vivo killing activity is absent. Comparison of some physicochemical characteristics of the cloacin clp03 . immunity protein complex and wild-type cloacin complex showed no significant differences. From a comparison of the binding capacity to specific receptors on sensitive cells, the translocation through the cell wall, and the interaction with cytoplasmic membranes, it could be concluded that the cloacin clp03 complex is hampered in its translocation from the outer membrane receptor site to the cytoplasmic membrane, resulting in the observed lack in killing activity. Cloacin pJN82 is shortened at the C-terminal of the molecule by approximately ten amino acid residues. Together with its loss of in vivo killing activity it has lost its capacity to bind immunity protein. Since the immunity protein probably not only provides cloacin-producing cells with "immunity" but is also involved in the translocation of the bacteriocin to the interior of sensitive cells, the absence of this protein is probably the reason for the lack of killing activity of cloacin pJN82. The implications of these findings for the topography of the cloacin molecule as suggested by de Graaf et al. (de Graaf, F.K., Stukart, M.J., Boogerd, F.C. and Metselaar, K. (1978) Biochemistry, in press) are discussed.     

The use of mutants in the study of structure-function relationships in cloacin DF13

A bacteriocin from cells with a mutant Clo DF13 plasmid (cloacin clp03· immunity protein complex) and a bacteriocin from cells containing the recombinant plasmic Clo DF13 :: Tn901 (cloacin pJN82) have been isolated. Both bacteriocins like wild-type cloacin DF13, are still able to inhibit in vitro protein synthesis, but their in vivo killing activity is absent. Comparison of some physicochemical characteristics of the cloacin clp03 · immunity protein complex and wild-type cloacin complex showed no significant differences.From a comparison of the binding capacity to specific receptors on sensitive cells, the translocation through the cell wall, and the interaction with cytoplasmic membranes, it could be concluded that the cloacin clp03 complex is hampered in its translocation from the outer membrane receptor site to the cytoplasmic membrane, resulting in the observed lack in killing activity.Cloacin pJN82 is shortened at the C-terminal of the molecule by approximately ten amino acid residues. Together with its loss of in vivo killing activity it has lost its capacity to bind immunity protein. Since the immunity protein probably not only provides cloacin-producing cells with “immunity” but is also involved in the translocation of the bacteriocin to the interior of sensitive cells, the absence of this protein is probably the reason for the lack of killing activity of cloacin pJN82.The implications of these findings for the topography of the cloacin molecule as suggested by de Graaf et al. (de Graaf, F.K., Stukart, M.J., Boogerd, F.C. and Metselaar, K. (1978) Biochemistry, in press) are discussed.     

Immunocytochemical Identification and Localization of Active and Inactive alpha-Amylase and Pullulanase in Cells of Clostridium thermosulfurogenes EM1

Clostridium thermosulfurogenes EM1 formed blebs, i.e., protrusions still in contact with the cytoplasmic membrane, that originated from the cytoplasmic membrane during growth in batch culture and continuous culture. They could be observed squeezed between the cell wall and cytoplasmic membrane in cells with seemingly intact wall layers (surface layer and peptidoglycan layer) as well as in cells with wall layers in different states of degradation caused by phosphate limitation or high dilution rates. Blebs were found to turn into membrane vesicles by constriction in cases when the cell wall was heavily degraded. Bleb and vesicle formation was also observed in the absence of substrates that induce alpha-amylase and pullulanase synthesis. No correlations existed between bleb formation and the presence of active enzyme. Similar blebs could also be observed in a number of other gram-positive bacteria not producing these enzymes, but they were not observed in gram-negative bacteria. For immunoelectron-microscopic localization of alpha-amylase and pullulanase in C. thermosulfurogenes EM1, two different antisera were applied. One was raised against the enzymes isolated from the culture fluid; the other was produced against a peptide synthesized, as a defined epitope, in analogy to the N-terminal amino acid sequence (21 amino acids) of the native extracellular alpha-amylase. By using these antisera, alpha-amylase and pullulanase were localized at the cell periphery in samples taken from continuous culture or batch culture. In samples prepared for electron microscopy by freeze substitution followed by ultrathin sectioning, blebs could be seen, and the immunolabel pinpointing alpha-amylase enzyme particles was seen not only randomly distributed in the cell periphery, but also lining the surface of the cytoplasmic membrane and the blebs. Cells exhibiting high or virtually no enzyme activity were labeled similarly with both antisera. This finding strongly suggests that alpha-amylase and pullulanase may occur in both active and inactive forms, depending on growth conditions.     

Degradation process of Mycobacterium leprae cells in infected tissue examined by the freeze-substitution method in electron microscopy

Mycobacterium leprae cells (strain Thai-53) harvested from infected mouse foot pads were examined by electron microscopy using the freeze-substitution technique. The population of M. leprae cells from the infected tissue consisted of a large number of degraded cells and a few normal cells. These thin sectioned cell profiles could be categorized into four groups depending on the alteration of the membrane structures, and the degradation process is considered to occur in stages, namely from stages 1 to 3. These are the normal cells with an asymmetrical membrane, a seemingly normal cell but with a symmetrical membrane (stage 1), a cell possessing contracted and highly concentrated cytoplasm with a membrane (stage 2), and a cell that has lost its membrane (stage 3). The peptidoglycan layer was found to remain intact in these cell groups.     

Granular layer in the periplasmic space of gram-positive bacteria and fine structures of Enterococcus gallinarum and Streptococcus gordonii septa revealed by cryo-electron microscopy of vitreous sections

High-resolution structural information on optimally preserved bacterial cells can be obtained with cryo-electron microscopy of vitreous sections. With the help of this technique, the existence of a periplasmic space between the plasma membrane and the thick peptidoglycan layer of the gram-positive bacteria Bacillus subtilis and Staphylococcus aureus was recently shown. This raises questions about the mode of polymerization of peptidoglycan. In the present study, we report the structure of the cell envelope of three gram-positive bacteria (B. subtilis, Streptococcus gordonii, and Enterococcus gallinarum). In the three cases, a previously undescribed granular layer adjacent to the plasma membrane is found in the periplasmic space. In order to better understand how nascent peptidoglycan is incorporated into the mature peptidoglycan, we investigated cellular regions known to represent the sites of cell wall production. Each of these sites possesses a specific structure. We propose a hypothetic model of peptidoglycan polymerization that accommodates these differences: peptidoglycan precursors could be exported from the cytoplasm to the periplasmic space, where they could diffuse until they would interact with the interface between the granular layer and the thick peptidoglycan layer. They could then polymerize with mature peptidoglycan. We report cytoplasmic structures at the E. gallinarum septum that could be interpreted as cytoskeletal elements driving cell division (FtsZ ring). Although immunoelectron microscopy and fluorescence microscopy studies have demonstrated the septal and cytoplasmic localization of FtsZ, direct visualization of in situ FtsZ filaments has not been obtained in any electron microscopy study of fixed and dehydrated bacteria.     

Plasmid-determined immunity of Escherichia coli K-12 to colicin Ia Is mediated by a plasmid-encoded membrane protein.

The colicin Ia structural (cia) and immunity (iia) genes of plasmid pColIa-CA53 have been cloned into the cloning vector pBR322. These two genes are closely linked, and both of them can be isolated on a deoxyribonucleic acid fragment approximately 4,800 base pairs long. An analysis of the polypeptides synthesized in ultraviolet-irradiated cells containing these cloned genes led to the conclusion that the iia gene product is a polypeptide with a molecular weight of approximately 14,500. Insertion of transposon Tn5 into the iia gene led to a concomitant loss of the immune phenotype and the ability to produce this protein. Fractionation of ultraviolet-irradiated cells harboring a plasmid carrying the iia gene showed that the immunity protein is a component of the inner (cytoplasmic) membrane. Furthermore, the mechanism of immunity to colicin Ia appears to operate at the level of the cytoplasmic membrane. This conclusion is based on our finding that membrane vesicles prepared from colicin Ia-immune cells could be depolarized by colicins E1 and Ib but not by colicin Ia.     

Cytoplasmic membrane vesicles of Escherichia coli. A simple method for preparing the cytoplasmic and outer membranes.

A simple preparative method is described for isolation of the cytoplasmic and outer membranes from E. coli. The characteristics of both membrane fractions were studied chemically, biologically, and morphologically. Spheroplasts of E. coli K-12 strain W3092, prepared by treating cells with EDTA-lysozyme [EC 3.2.1.17], were disrupted in a French press. The crude membrane fraction was washed with 3 mM EDTA-10% (w/v) sucrose, pH 7.2, and the cytoplasmic membranes and outer membranes were separated by sucrose isopycnic density gradient centrifugation. The crude membrane fraction contained approximately 10% of the protein of the whole cells, 0.3% of the DNA, 0.7% of the RNA, 0.3% of the peptidoglycan, and about 30% of the lipopolysaccharide. The cytoplasmic membrane fraction was rich in phospholipid, while the outer membrane fraction contained much lipopolysaccharide and carbohydrate; the relative contents of lipopolysaccharide and carbohydrate per mg protein in the cytoplasmic membrane fraction were 12 and 40%, respectively, of the contents in the outer membrane fraction. Cytochrome b1, NADH oxidase, D-lactate dehydrogenase [EC 1.1.1.28], succinate dehydrogenase [EC 1.3.99.1], ATPase [EC 3.5.1.3], and activity for concentrative uptake of proline were found to be localized mainly in the cytoplasmic membranes; their specific activities in the outer membrane fraction were 1.5 to 3% of those in the cytoplasmic membrane fraction. In contrast, a phospholipase A appeared to be localized mainly in the outer membranes and its specific activity in the cytoplasmic membrane fraction was only 5% of that in the outer membrane fraction. The cytoplasmic and outer membrane fractions both appeared homogeneous in size and shape and show vesicular structures by electron microscopy. The advantages of this method for large scale preparation of the cytoplasmic and outer membrane fractions are discussed.     

Toxicity of the Colicin M Catalytic Domain Exported to the Periplasm Is FkpA Independent

Colicin M (ColM) is a bactericidal protein that kills sensitive cells by hydrolyzing lipid II, involved in the biosynthesis of cell wall peptidoglycan. It recognizes FhuA on the outer leaflet, and its translocation through the outer membrane depends on the energized Ton complex in the inner membrane. To be active in the periplasm, ColM must be translocated through the outer membrane and then interact with FkpA, a periplasmic protein that exhibits both cis- and trans-peptidylprolyl isomerase (PPiase) and chaperon activities. In an attempt to directly target ColM to the periplasm of the producing bacteria, we fused the presequence of OmpA to ColM (sp-ColM). We found that expression of this hybrid protein in an Escherichia coli strain devoid of ColM immunity protein (Cmi) was bactericidal. We showed that sp-ColM was correctly expressed, processed, and associated with the inner membrane. sp-ColM toxicity was related to its enzymatic activity and did not rely on the TonB import proteins or the FhuA receptor. The presence of both activity domains of FkpA was still required for sp-ColM activity. Analyses of deletion mutants of sp-ColM show that the domain required for toxicity corresponds to the C-terminal last 153 amino acids of ColM. Like the full-length protein, this domain is not active in the presence of the immunity protein Cmi. On the other hand, it does not require FkpA for toxic activity.Colicins are plasmid-encoded secreted bacterial toxins that kill Escherichia coli and other closely related bacteria (7). Colicins are classified into three groups, according to the way in which they kill sensitive cells. The first group consists of pore-forming colicins, including colicins A (ColA), B, E1, N, and Ia, all of which form ionic voltage-gated channels into the inner membrane of sensitive cells (20). The second group consists of the endonuclease colicins, which digest the DNA or RNA in the cytoplasm of the target cell. These colicins include ColE2 to -E9 and ColD (17). The last group includes only ColM, which specifically cleaves the bond between the lipid moiety and the pyrophosphoryl group of the peptidoglycan lipid II intermediate, located at the periplasmic side of the inner membrane (11). Any plasmid carrying a colicin gene also codes for a specific immunity protein that protects the producing cell itself from colicin''s toxic activity. Immunity proteins to pore-forming colicins and to ColM are found in the inner membrane, whereas immunity proteins to endonuclease colicins are soluble and are found in the cytoplasm.The lethal action of colicins can be divided into three steps. First, they bind to a receptor located in the outer membrane of the target cell (15, 19). Second, they are translocated into the periplasm through interaction with the Tol or Ton import machinery located in the inner membrane. The Ton import machinery consists of the three membrane proteins TonB, ExbB, and ExbD. Finally, colicins reach their target molecule. Colicins have three functional domains; the N-terminal and central domains are involved in the import of colicins through the E. coli envelope, and the C-terminal domain is involved in the toxic activity of the protein (5).ColM binds to the FhuA outer membrane receptor protein. Its translocation into the cell depends on the interaction of FhuA with TonB, as well as the proton motive force of the inner membrane (6). A consensus sequence of seven residues, designated the TonB box, has been identified in the N-terminal region of ColM. This sequence interacts with the TonB protein, which may provide the energy required for import of ColM across the outer membrane (6). ColM is atypical among colicins for different reasons. First, ColM has an unusual mode of action through causing lysis by inhibiting peptidoglycan synthesis. Although its enzymatic activity and three-dimensional crystal structure have been reported (11, 29), the amino acids that comprise its active site have not yet been clearly identified (4). Second, the toxic activity of the imported ColM depends upon the presence of a specific periplasmic protein (FkpA) (14), a chaperone, and a cis- and trans-peptidylprolyl isomerase (PPiase) (2). Last, although it has been shown that all colicins share their organization in three domains, the three-dimensional model of ColM reveals a unique fold with no similarity to other colicins (29).Here we describe a new approach to investigate the in vivo activity of ColM which can help to more precisely delineate the different domains of ColM and identify the minimal sequence that retains toxic activity. Although the substrate of ColM is present in the inner leaflet of the E. coli cytoplasmic membrane, ColM is not active in the cytoplasm and must enter the cell from the outside to be toxic in the periplasm (13). We designed a hybrid protein composed of ColM fused to a prokaryotic signal sequence (sp-ColM). We showed that sp-ColM is exported into the periplasm of the producing cells and is toxic for cells that do not produce the immunity protein. We also showed that the hybrid protein is functional in vivo and requires the presence of FkpA but does not depend on the presence of FhuA or the Ton translocation machinery. Using this new approach for testing the activity of ColM in vivo, we designed different deletion mutants of ColM that were tested for toxicity, FkpA dependence, and sensitivity to the Cmi immunity protein. We showed that the ColM toxic domain (sp-C) comprises the last 153 amino acid residues of the C terminus, which is longer than the last 130 residues reported previously (29). These results are consistent with those of a recent in vitro study of truncated forms of ColM (4). Our results also show that Cmi protects the cells against sp-C and that sp-C does not require FkpA to kill the producing cells. The use of sp-ColM is therefore of great value for investigating how ColM kills cells in vivo and how ColM interacts with FkpA and the immunity protein.     

A colicin M derivative containing the lipoprotein signal sequence is secreted and renders the colicin M target accessible from inside the cells

Colicin M is only released in very low amounts by cells harbouring this plasmid encoded colicin, due to the lack of a release (lysis) protein. A fusion gene (lpp'cma) was constructed which determined two proteins: Lpp'-Cma composed of the signal sequence of the murein lipoprotein (Lpp) and colicin M (Cma), and unaltered colicin M. Cells expressing the fusion gene released 50% of the total colicin M into the culture medium, compared to 1% found in the medium of cells synthesizing only colicin M. The release resulted from partial cell lysis caused by colicin M since a colicin M tolerant strain remained unaffected. Lpp'-Cma thus mimics phenotypically the action of colicin release proteins but displays a different lysis mechanism. In strains defective in components of the colicin M uptake system, Lpp'-Cma caused lysis as effectively as in uptake proficient strains. Apparently, Lpp'-Cma renders the colicin M target site accessible from inside the cell which stands in contrast to the action of colicin M which is only bactericidal when provided from outside.Abbreviation bp base pairs