Effects of colicins K and E1 on the glucose phosphotransferase system.

1. Glycerol-grown cells of Escherichia coli and its mutant uncA, treated with colicin E1 or K, exhibited a several-fold higher level of alpha-methylglucoside uptake than untreated cells. This stimulation was independent of the carbon source present during the uptake test. In a mutant strain that has elevated levels of alpha-methylglucoside accumulation the addition of colicin E1 or carbonylcyanide m-chlorophenylhydrazone (CCCP) did not further enhance the uptake. 2. Colicins K and E1 decreased the apparent Km for alpha-methylglucoside uptake significantly and increased the V about twofold. The exit of the glucoside was severely inhibited by the colicins. 3. In the presence of colicins, alpha-methylglucoside is still accumulated via the phosphoenolpyruvate-phosphotransferase system since no accumulation or phosphorylation occurs in an enzyme I mutant. The colicins increased the relative intracellular concentration of phosphorylated alpha-methylglucoside, possibly by inhibiting the dephosphorylation reaction, and caused an excretion of this compound. 4. The results are interpreted as indicating that energization of the membrane has an inhibitory effect on the phosphotransferase system. Possible modes of action are discussed.     

Defective transport in S-adenosylmethionine synthetase mutants of Escherichia coli

The transport of several metabolites is decreased in mutant strains of Escherichia coli (Met K, E4 and E40), which contain decreased levels of S-adenosylmethionine synthetase. The rates and extents of uptake for lysine, leucine, methionine, and α-methylglucoside in both whole cells and membrane vesicles isolated from these mutants are 2- to 10-fold lower than in corresponding preparations from wild-type cells, although proline uptake is normal. The addition of S-adenosylmethionine to cultures of strain E40 can partially restore the rate and extent of lysine uptake. Lysine transport is lower in mutant vesicles in the presence of either d-lactate, succinate, α-hydroxylbutyrate, or NADH even though these substrates are oxidized at rates comparable to those in wild-type vesicles. This suggests that the defect is not related to the ability of vesicles to oxidize electron donors, but is very likely related to the ability of mutant vesicles to couple respiration to lysine transport. In addition, temperature-induced efflux of α-methylglucoside phosphate and dinitrophenol-induced efflux of lysine are similar in both the mutant and wild-type membranes, indicating that the barrier properties of the membrane and the activity of the lysine carrier are normal.     

Enhancement of α-methylglucoside efflux by respiration in respiratory mutants of Escherichia coli K-12

The rate of α-methylglucoside efflux from wild-type cells of Escherichia coli K-12 is enhanced by different substrates, as long as they are readily respired. A similar enhancement takes place in strains with impaired oxidative phosphorylation (unc mutants), regardless of their being able (strains AN120, N₁₄₄, and AN382) or unable (strain NR70) to energize the membrane through respiratory electron flow. The uncouplers carbonylcyanide-m-chlorophenylhydrazone and tetrachlorosalicylanilide do not diminish the efflux acceleration in wild-type strains or unc mutants. However, the stimulation of α-methylglucoside efflux does not occur in the mutant AN59 which cannot perform a normal respiratory electron transport due to a defective synthesis of ubiquinone. The failure to stimulate the efflux is observed with succinate, which is a typical substrate of respiration, as well as with substrates which can yield ATP both at respiratory and substrate levels such as gluconate or glycerol. Moreover, potassium cyanide nullifies the acceleration of α-methylglucoside efflux caused in any type of strain and by any substrate. These results show that neither ATP nor an energized state of the membrane appears to be needed for respiration to accelerate α-methylglucoside release from E. coli cells, and question the existence of any energy-requiring reaction for αMG exit, previously proposed by other authors.     

The TolA protein interacts with colicin E1 differently than with other group A colicins.

The 421-residue protein TolA is required for the translocation of group A colicins (colicins E1, E2, E3, A, K, and N) across the cell envelope of Escherichia coli. Mutations in TolA can render cells tolerant to these colicins and cause hypersensitivity to detergents and certain antibiotics, as well as a tendency to leak periplasmic proteins. TolA contains a long alpha-helical domain which connects a membrane anchor to the C-terminal domain, which is required for colicin sensitivity. The functional role of the alpha-helical domain was tested by deletion of residues 56 to 169 (TolA delta1), 166 to 287 (TolA delta2), or 54 to 287 (TolA delta3) of the alpha-helical domain of TolA, which removed the N-terminal half, the C-terminal half, or nearly the entire alpha-helical domain of TolA, respectively. TolA and TolA deletion mutants were expressed from a plasmid in an E. coli strain producing no chromosomally encoded TolA. Cellular sensitivity to the detergent deoxycholate was increased for each deletion mutant, implying that more than half of the TolA alpha-helical domain is necessary for cell envelope stability. Removal of either the N- or C-terminal half of the alpha-helical domain resulted in a slight (ca. 5-fold) decrease in cytotoxicity of the TolA-dependent colicins A, E1, E3, and N compared to cells producing wild-type TolA when these mutants were expressed alone or with TolQ, -R, and -B. In cells containing TolA delta3, the cytotoxicity of colicins A and E3 was decreased by a factor of >3,000, and K+ efflux induced by colicins A and N was not detectable. In contrast, for colicin E1 action on TolA delta3 cells, there was little decrease in the cytotoxic activity (<5-fold) or the rate of K+ efflux, which was similar to that from wild-type cells. It was concluded that the mechanism(s) by which cellular uptake of colicin E1 is mediated by the TolA protein differs from that for colicins A, E3, and N. Possible explanations for the distinct interaction and unique translocation mechanism of colicin E1 are discussed.     

Energy requirement for the initiation of colicin action in Escherichia coli

1. Starved cells of a strain of Escherichia coli and its mutant uncA, treated with colicin K, E2 or E3, remained fully rescuable upon trypsin treatment (stage I in colicin action). The transition to stage II in colicin action (cells no longer rescuable by trypsin) was promoted by the addition of either glucose or d-lactate.2. Aerobically glucose-grown cells of the normal strain were irreversibly killed by colicin K, E2 or E3 under anaerobic conditions, while similarly treated cells of its mutant uncA remained fully rescuable. The stage I-stage II transition in colicin action was blocked in normal cells under anaerobic conditions when succinate was the sole carbon source.3. Arsenate alone had little effect on the progression of the stage I-stage II transition in normal cells, treated with colicin K. However, this transition was abolished in the presence of both arsenate and anaerobic conditions.4. The initiation of colicin action could be coupled to the anaerobic electron transfer systems formate dehydrogenase-nitrate reductase and α-glycerophosphate dehydrogenase-fumarate reductase.5. These results indicate that an energized state of the cytoplasmic membrane is required for the initiation of colicin action and that no high-energy phosphorylated compounds are necessary.     

Interaction of Colicins with Bacterial Cells III. Colicin-tolerant Mutations in Escherichia coli

Mutants that adsorb certain colicins without being killed, i.e., tolerant mutants (tol), were isolated from Escherichia coli K-12 strains. Selection was done either with colicin K or E2. Several groups of mutants showing different phenotypes were found, and some of them showed tolerance to both K and E colicins, which have different receptors. Many of these mutants mapped near gal. Typical mutants from group II, III, and IV were studied in more detail. The mutant loci were contransducible with gal by phage P1. The linkage order was deduced to be tol-gal-λ. In partially diploid strains, these mutant loci are recessive to wild-type alleles. Temperature-dependent conditionally tolerant mutants were also isolated. Two groups were found: the first was tolerant to E2 and E3 at 40 C, but sensitive at 30 C; the second was tolerant to E2 at 30 C, but sensitive at 40 C. Experiments done with these mutants suggest that these mutations affect the heat lability of some protein that is necessary for the response of cells to colicins. Conditionally lethal tolerant mutants were isolated which at 40 C were tolerant to E2 and E3 and could not grow, but which at 30 C were fully sensitive and grew normally. The mutation mapped near malA. The tolerance at 40 C is not due to a consequence of an inactivation of general cellular metabolism, but presumably is a cause of the subsequent inhibition of cellular growth. The results suggest that some protein components involved in the response to colicin are also vital to normal cellular growth.     

Mutation analysis of the receptor for colicins E1–E7. A pilot study

Thirty eight mutant clones of the colicin indicator strainEscherichia coli K 12 ROW, selected by their insensitivity to any of the colicins El–E7, were isolated. Comparison of their sensitivity-resistance patterns to colicins El–E7 enabled us to draw a rough preliminary map of the receptor for E colicins. In this receptor, the highly specific binding site for colicin El partially overlaps with the domain shared by all colicins E2 through E7. A specific binding site of this domain appears to be common for colicins E3 and E6; a part of the E3 and E6 binding site is also common for colicins E4 and E5 and a small, least specific, part also for colicins E2 and E7. Using colicin assay experiments, the binding capacity of coliein E receptor mutants could be estimated. A decreased, but not completely lost ability of certain mutants to bind colicins E, correlated to their lowered sensitivity to them, was found. Thus the phenomenon of partial colicin resistance was established, showing that colicin sensitivity—resistance is not a qualitative but a quantitative marker.     

Increased production of the outer membrane receptors for colicins B, D and M by Escherichia coli under iron starvation.

Growth of $\text{E. coli}$ K-12 under severe iron stress results in increased production of the outer membrane receptors for colicins B, D, Ib and M. The increase in colicin receptor activity coincides with the appearance of large amounts of two high molecular weight proteins in the outer membrane of the cells. These proteins are identified as the outer membrane receptors for colicins B and D and for colicin M. Mutants lacking a functional outer membrane receptor for colicins B and D are defective in the uptake of iron complexed with the siderochrome enterochelin, and are thus comparable with $\text{tonA}$ mutants which lack a functional receptor for colicin M and are defective in the uptake of iron complexed with ferrichrome (6). The colicin B and D receptor may therefore function in the uptake of ferri-enterochelin.     

Effects of Colicins E1 and K on Permeability to Magnesium and Cobaltous Ions

The energy-dependent exchange of intracellular Mg(2+) with extracellular Mg(2+) or Co(2+) is inhibited by colicin E1 and, less strongly, by colicin K. Treatment with either colicin causes a net loss of intracellular Mg(2+). This loss begins immediately in cells treated with colicin E1, but in colicin K-treated cells the onset of Mg(2+) loss is delayed 1 to 10 min, depending upon the temperature and the multiplicity of colicin K. Both colicins differ from chemical inhibitors of energy-yielding metabolism; energy poisons block transport of Mg(2+) and Co(2+), but both colicins increase passive permeability to Mg(2+) and Co(2+). Inhibitors of energy-yielding metabolism (and of Mg(2+) exchange) block the initiation of Mg(2+) loss by either colicin, but do not stop colicin-promoted efflux once it has begun. Colicin E1 added before colicin K prevents the more rapid Mg(2+) efflux characteristic of colicin K-treated cells. Quantitative comparisons of the effects of colicins E1 and K upon permeability to Mg(2+) and Co(2+) lead us to conclude that the two colicins are not identical in their mode of action.     

Purification and molecular properties of a new colicin.

The process of isolation and purification of a new colicin isolated from a Citrobacter strain is described. Escherichia coli sensitive cells are protected by vitamin B12 from the action of this bacteriocin; this suggests that it belongs to the E group of colicins. Therefore, we have called it colicin E4. It has a molecular weight of 56 000 and two molecular forms of isoelectric points 9.4 and 8.2 are separated in electrofocusing on polyacrylamide gels. It has a sedimentation coefficient of 3.4 S and the absorption coefficient A1(280%) nm is 6.23 cm(-1). Using an antibody raised against pure colicin E4, no cross-reaction was detected against colicins A, E1 or K. The physiological effect of colicin E4 on sensitive cells is very similar to that of colicins E1, K or I which disrupt the energized membrane state.     

Energy-coupled colicin transport through the outer membrane of Escherichia coli K-12: mutated TonB proteins alter receptor activities and colicin uptake

Abstract The current model of TonB-dependent colicin transport through the outer membrane of Escherichia coli proposes initial binding to receptor proteins, vectorial release from the receptors and uptake into the periplasm from where the colicins, according to their action, insert into the cytoplasmic membrane or enter the cytoplasm. The uptake is energy-dependent and the TonB protein interacts with the receptors as well as with the colicins. In this paper we have studied the uptake of colicins B and Ia, both pore-forming colicins, into various tonB point mutants. Colicin Ia resistance of the tonB mutant (G186D, R204H) was consistent with a defective Cir receptor-TonB interaction while colicin Ia resistance of E. coli expressing TonB of Serratia marcescens , or TonB of E. coli carrying a C-terminal fragment of the S. marcescens TonB, seemed to be caused by an impaired colicin Ia-TonB interaction. In contrast, E. coli tonB (G174R, V178I) was sensitive to colicin Ia and resistant to colicin B unless TonB, ExbB and ExbD were overproduced which resulted in colicin B sensitivity. The differential effects of tonB mutations indicate differences in the interaction of TonB with receptors and colicins.     

Individual domains of colicins confer specificity in colicin uptake, in pore-properties and in immunity requirement.

Six different hybrid colicins were constructed by recombining various domains of the two pore-forming colicins A and E1. These hybrid colicins were purified and their properties were studied. All of them were active against sensitive cells, although to varying degrees. From the results, one can conclude that: (1) the binding site of OmpF is located in the N-terminal domain of colicin A; (2) the OmpF, TolB and TolR dependence for translocation is also located in this domain; (3) the TolC dependence for colicin E1 is located in the N-terminal domain of colicin E1; (4) the 183 N-terminal amino acid residues of colicin E1 are sufficient to promote E1AA uptake and thus probably colicin E1 uptake; (5) there is an interaction between the central domain and C-terminal domain of colicin A; (6) the individual functioning of different domains in various hybrids suggests that domain interactions can be reconstituted in hybrids that are fully active, whereas in others that are much less active, non-proper domain interactions may interfere with translocation; (7) there is a specific recognition of the C-terminal domains of colicin A and colicin E1 by their respective immunity proteins.     

Importation of nuclease colicins into E coli cells: endoproteolytic cleavage and its prevention by the immunity protein

A major group of colicins comprises molecules that possess nuclease activity and kill sensitive cells by cleaving RNA or DNA. Recent data open the possibility that the tRNase colicin D, the rRNase colicin E3 and the DNase colicin E7 undergo proteolytic processing, such that only the C-terminal domain of the molecule, carrying the nuclease activity, enters the cytoplasm. The proteases responsible for the proteolytic processing remain unidentified. In the case of colicin D, the characterization of a colicin D-resistant mutant shows that the inner membrane protease LepB is involved in colicin D toxicity, but is not solely responsible for the cleavage of colicin D. The lepB mutant resistant to colicin D remains sensitive to other colicins tested (B, E1, E3 and E2), and the mutant protease retains activity towards its normal substrates. The cleavage of colicin D observed in vitro releases a C-terminal fragment retaining tRNase activity, and occurs in a region of the amino acid sequence that is conserved in other nuclease colicins, suggesting that they may also require a processing step for their cytotoxicity. The immunity proteins of both colicins D and E3 appear to have a dual role, protecting the colicin molecule against proteolytic cleavage and inhibiting the nuclease activity of the colicin. The possibility that processing is an essential step common to cell killing by all nuclease colicins, and that the immunity protein must be removed from the colicin prior to processing, is discussed.     

Strong function-related homology between the pore-forming colicins K and 5.

Sequence determination of the Escherichia coli colicin K determinant revealed identity with the E. coli colicin 5 determinant in the immunity and lysis proteins, strong homologies in the pore-forming region (93.7%) and the Tsx receptor-binding region (77%) of the colicins, and low levels of homology (20.3%) in the N-terminal region of the colicins. This latter region is responsible for the Tol-dependent uptake of colicin K and the Ton-dependent uptake of colicin 5 in the respective colicins. During evolution, the DNA encoding colicin activity and binding to the Tsx receptor was apparently recombined with two different DNA fragments that determined different uptake routes, leading to the differences observed in colicin K and colicin 5 import.     

Outer Membrane-Dependent Transport Systems in Escherichia coli: Effect of Repression or Cessation of Colicin Receptor Synthesis on Colicin Receptor Activities

Proteins in the outer membrane of gram-negative bacteria serve as general porins or as receptors for specific nutrient transport systems. Many of these proteins are also used as receptors initiating the processes of colicin or phage binding and uptake. The functional activities of several outer membrane proteins in Escherichia coli K-12 were followed after cessation or repression of their synthesis. Cessation of receptor synthesis was accomplished with a thermolabile suppressor activity acting on amber mutations in btuB (encoding the receptor for vitamin B(12), the E colicins, and phage BF23) and in fepA (encoding the receptor for ferric enterochelin and colicins B and D). After cessation of receptor synthesis, cells rapidly became insensitive to the colicins using that receptor. Treatment with spectinomycin or rifampin blocked appearance of insensitive cells and even increased susceptibility to colicin E1. Insensitivity to phage BF23 appeared only after a lag of about one division time, and the receptors remained functional for B(12) uptake throughout. Therefore, possession of receptor is insufficient for colicin sensitivity, and some interaction of receptor with subsequent uptake components is indicated. Another example of physiological alteration of colicin sensitivity is the protection against many of the tonB-dependent colicins afforded by provision of iron-supplying siderophores. The rate of acquisition of this nonspecific protection was found to be consistent with the repression of receptor synthesis, rather than through direct and immediate effects on the tonB product or other components of colicin uptake or action.     

YieJ (CbrC) Mediates CreBC-Dependent Colicin E2 Tolerance in Escherichia coli

Colicin E2-tolerant (known as Cet2) Escherichia coli K-12 mutants overproduce an inner membrane protein, CreD, which is believed to cause the Cet2 phenotype. Here, we show that overproduction of CreD in a Cet2 strain results from hyperactivation of the CreBC two-component regulator, but CreD overproduction is not responsible for the Cet2 phenotype. Through microarray analysis and gene knockout and overexpression studies, we show that overexpression of another CreBC-regulated gene, yieJ (also known as cbrC), causes the Cet2 phenotype.Colicins are protein antibiotics that have various modes of action. They are usually encoded on plasmids and, in many cases, alongside genes encoding colicin immunity factors, which protect colicin-producing cells from the colicin they produce. Of the enzymatic (E) colicins, some carry nuclease activity, including colicin E2, colicin E9, and colicin E3. These three proteins bind to susceptible cells via the surface protein BtuB (the vitamin B₁₂ importer) and, through a series of events that are poorly understood, cross the cell envelope to enter the cytoplasm, where they degrade nucleic acids: colicins E2 and E9 target DNA; colicin E3 targets rRNA (11).Cells can readily become tolerant of E colicins. Mutants usually have lost either the colicin receptor or some protein involved in colicin import. Loss-of-function mutations in btuB confer tolerance of high levels of colicins E2, E9, and E3. Almost 40 years ago, Escherichia coli mutants having a colicin E2-tolerant (Cet2) phenotype were identified. The Cet2 phenotype confers tolerance of colicins E2 and E9 only, while cells remain susceptible to colicin E3, and BtuB is intact (8, 9). Cet2 mutants were shown to overproduce an inner membrane protein (26), and the cet2 mutation was found to be dominant in trans and mapped at 99.9 min on the E. coli chromosome (8, 9). Using the Cet2 mutant RB208 as a source of genomic DNA, a clone able to transform E. coli cells to a Cet2 phenotype was identified. Since this clone carried a gene predicted to encode an inner membrane protein with properties identical to those overproduced in Cet2 mutants, the gene was named cet (15).The cet gene is the last gene in the four-gene cre locus, so cet is also known as creD. The other genes in this locus are creA (hypothetical open reading frame [ORF]); creB, encoding a response regulator; and creC, encoding a sensor kinase. CreB and CreC form a classical two-component regulatory system, and we recently showed that CreBC are activated upon fermentation of glucose in minimal medium or during aerobic growth on minimal medium containing fermentation products, such as pyruvate, lactate, or acetate, as the sole carbon and energy source (10). CreBC controls the expression of a number of genes (the Cre regulon), some of which encode metabolic functions but several of which are hypothetical. One of the most tightly controlled Cre regulon genes is creD (5).We have previously shown that the Cet2 strain RB208 has a point mutation in creC but that creD itself is wild type (5). Since the RB208 genomic clone capable of transforming cells to a Cet2 phenotype carries the whole cre locus, not just creD (15), our hypothesis is that the Cet2 phenotype of the transformant was due to a trans-dominant mutation in the cloned creC mutant allele activating one or more Cre regulon genes and that the Cet2 phenotype may or may not be caused by overexpression of creD. The aims of the experiments described in this paper were to test our hypothesis that the Cet2 phenotype is caused by activating mutations in CreBC and to definitively identify the Cre regulon gene that encodes the colicin E2 tolerance (Cet) protein.     

Binding domains of colicins E1, E2 and E3 in the receptor protein btub ofEscherichia coli

Eight reagents specifically modifying amino acids were applied to cells of a standardEscherichia coli colicin indicator strain to followin vivo changes of its binding capacity for colicins E1–E3 and hence the binding domains (epitopes) for them in the outer membrane receptor protein BtuB. The effect of these reagents was also investigated in a mutant strain carrying an extensive BtuB deletion. The following differences of the binding epitopes could be ascertained.Colicin E1: Blockage of OH-groups, just as N-substitution of His and modification of Arg and Trp enhance binding of colicin E1. In the deleted receptor, also abolition of carboxylic anion bonds enhances its affinity for colicin E1. It follows that colicin E1 is bound, most of all, to the hydrophobic domain A (loops 1+2) of BtuB.Colicins E2 and E3: both exert rather analogous binding parameters. In contrast to E1, O-substitution of Ser and Thr dramatically decreases the E2 and E3 binding, similarly to modification of Lys. There is also a clear difference in the binding affinity of the domain for E2 and/or E3 and for E1 following modifications of their Arg and His. Colicins E2 and E3 are rather bound to the hydrophilic domain B (loops 5–7) of the receptor. In this respect, interactions of colicins E2 and E3 with deeper parts of A and B domains (Trp, several Arg, Lys and His residues) exhibited subtle differences. Acidic pH (4.5–6.0) shows a positive, while pH 7.0–8.5 a rather negative impact on the receptor-binding function for the colicins. It was clearly demonstrated that there is just a partial difference between the binding behavior of colicins E1, E2 and/or E3.     

Studies of colicin action on wall-less stable L-forms of Escherichia coli. I. Degree of attachment and of killing effect on rods and stable L-form cells.

Escherichia coli strains B and K12 W 1655 F+ are able to bind more lethal units of colicins E2, E3, G, H, Ia, and K+ X per one stable L-form cell (of the protoplast type) than per one rod cell; colicin D is bound in a higher amount on E. coli B rods. This pattern remains unchanged, if the same colicins are attached on chloroform-killed cells of both forms. Rods of both E. coli strains are more sensitive to colicins D, E2, E3, K + X (as--in the strain B--to colicin Ia) than cells of the respective L-forms. In the strain W 1655 F+ both cell forms are equally highly sensitive to colicin Ia. The stable L-forms of both strains are much more sensitive to colicins G and H than the rods. Thus the Gram-negative cell wall decreases the probability of a colicin molecule to get attached to its receptor in the cytoplasmic membrane. On the other hand, in E. coli cells the attachment of most colicin molecules to the wall receptors increases the probability of their biological effect. There is no such effect of the wall-attachment on the action of colicins G or H. The strain B is tolerant to colicin E2, while being resistant to E3; thus the cytoplasmic membrane receptor sites for them are not identical.     

Interaction of the Colicin K Bactericidal Toxin with Components of Its Import Machinery in the Periplasm of Escherichia coli

Colicins are bacterial antibiotic toxins produced by Escherichia coli cells and are active against E. coli and closely related strains. To penetrate the target cell, colicins bind to an outer membrane receptor at the cell surface and then translocate their N-terminal domain through the outer membrane and the periplasm. Once fully translocated, the N-terminal domain triggers entry of the catalytic C-terminal domain by an unknown process. Colicin K uses the Tsx nucleoside-specific receptor for binding at the cell surface, the OmpA protein for translocation through the outer membrane, and the TolABQR proteins for the transit through the periplasm. Here, we initiated studies to understand how the colicin K N-terminal domain (KT) interacts with the components of its transit machine in the periplasm. We first produced KT fused to a signal sequence for periplasm targeting. Upon production of KT in wild-type strains, cells became partly resistant to Tol-dependent colicins and sensitive to detergent, released periplasmic proteins, and outer membrane vesicles, suggesting that KT interacts with and titrates components of its import machine. Using a combination of in vivo coimmunoprecipitations and in vitro pulldown experiments, we demonstrated that KT interacts with the TolA, TolB, and TolR proteins. For the first time, we also identified an interaction between the TolQ protein and a colicin translocation domain.Colicins are bacterial toxins produced by Escherichia coli strains and are active against E. coli or related strains (17). These bacterial antibiotic toxins play an important role in the E. coli colonization of environmental niches, including the mammal gastrointestinal tract (25, 32, 49, 50). The classification of colicins is based on differences in the mechanisms of action, such as pore formation (colicins A, B, E1, K, Ia, N, 5, etc.), degradation of nucleic acids (including DNases [colicins E2, E7, and E9], 16S RNases [colicins E3, E4, and E6], or tRNases [colicins D and E5]), or degradation of lipid II (colicin M) (17, 34). Colicins are also categorized depending on their import machines: colicins using the Tol proteins are classified as group A (colicins A, E1 to E9, K, N, etc.), whereas colicins using the ExbBD-TonB proteins are classified as group B (colicins B, D, Ia, M, 5, etc.). However, the transport across the periplasm is only one of the three steps of the mechanism of action. Colicins bind to an outer membrane receptor and are translocated through the outer membrane and the periplasm (14, 35, 55, 56). Finally, the C-terminal domain (responsible for the activity) is translocated to its final destination (inner membrane or cytoplasm) depending on its mechanism of action. Colicins are divided into three different structural and functional domains that correspond to the three steps of the mechanism of action: the N-terminal domain is required for translocation, the central domain is involved in receptor binding, and the C-terminal domain carries the activity (4, 5). During the translocation step, the N-terminal domain of the colicin interacts with components of the import machine: colicins A, E1, and N interact with the TolA protein; colicins A, E3, E7, and E9 interact with the TolB protein; and colicins A and E3 interact with TolR (6, 12, 13, 15, 21, 23, 26, 27, 30, 39, 48, 54). In some cases, the domains of the Tol proteins involved in colicin binding have been identified. Reciprocally, the regions of colicins in interaction with the Tol proteins have been delineated. In colicin A, the TolA binding sequence (ABS) is contained within residues 37 to 98 (13, 30), in which a SYNT motif (residues 57 to 60) has been shown to be essential for TolA binding (18, 46). The TolB box and the TolR binding sequences have also been identified in colicin A (27, 30). The TolB box is well conserved within TolB-dependent colicins, including colicins A and E2 to E9, and is composed of residues DG[T,S]GWSSE (12, 13). These residues form a loop penetrating within the TolB beta-propeller (39, 57), mimicking the TolB-Pal interaction (9, 10). Interestingly, the Tol-dependent, pore-forming colicin K does not possess a TolB box (see Fig. ​Fig.1A),1A), raising the hypothesis that its translocation might be TolB independent or that colicin K interacts with TolB differently than do other TolB-dependent colicins. In this study, we tested the Tol requirements for colicin K translocation and showed that colicin K requires the TolA, TolB, TolQ, and TolR proteins. Production of the N-terminal domain of colicin K in the periplasm of wild-type (WT) cells induces specific tol defects and tolerance to Tol-dependent colicins and bacteriophage, suggesting that the colicin K N-terminal domain binds and titrates the Tol proteins. Further in vivo coimmunoprecipitation and in vitro pulldown experiments demonstrated interactions between the colicin K N-terminal domain and the TolA, TolB, and TolR proteins. For the first time, we also identified an interaction between a colicin translocation domain and the fourth component of the Tol complex, the TolQ protein.Open in a separate windowFIG. 1.In the absence of an identifiable TolB-binding sequence, colicin K translocation is TolB dependent. (A) Sequence alignment of colicin K and three TolB-dependent colicins (A, E2, and E9). Conserved residues are indicated by red letters. The characterized TolB binding sequence is indicated by the green box (defined in references 12 and 27). (B) Colicin spot assays using serial dilutions of colicins A (TolB dependent), E1 (TolB independent), and K on a wild-type (WT) strain and its tolB derivative (from left to right, 100, 10, 1, and 0.1 ng of colicins have been spotted, respectively).     

A contribution to the problem of a common receptor of the E-group colicins

The question of a common receptor for colicins E1, E2 and E3 was studied by comparing the kinetics of their action in different colicin mixtures with that of each colicin alone.The rate of specific adsorption of colicins was studied in two ways: by assaying the decreasing amount of free colicin in the solution (direct) and by determining the numbers of surviving colony-forming bacteria (indirect). At the same multiplicity, the rate of adsorption and inhibitory effect varied for each colicin tested (E1, E2, E3 and K).These differences were the basis of our study on the inhibitory effects of mixtures of two colicins added either simultaneously or successively.The results were conclusive: E1 and K bind to receptor sites different from a common receptor site for colicins E2 and E3. Thus colicin E1 should be excluded from the E group. It is suggested to sign it J as previously.The authors wish to thank Dr. B. marda for his mathematical advice.