Search

The ompAgene of Escherichia colicodes for a major protein of the outer membrane. When this gene was moved between various unrelated strains (E. coliK-12 and two clinical isolates of E. coli) by transduction, the gene was expressed very poorly. Recombinants carrying “foreign” genes produced no OmpA protein which could be detected on polyacrylamide gels and became resistant to bacteriophage K3, which uses this protein as receptor. The recombinants were sensitive to host-range mutants of K3, indicating a very low level of OmpA protein was produced. When an E. coliK-12 recombinant carrying an unexpressed foreign ompAallele was subjected to two cycles of selection for an OmpA+phenotype, a mutant strain was obtained which was sensitive to K3 and which expressed nearly normal levels of OmpA protein in the outer membrane. This strain carried mutations in the foreign ompAgene, as indicated both by genetic mapping and the alteration of a peptide in the mutant OmpA protein. The ability of the OmpA protein to bind to lipopolysaccharide (LPS) showed similar strain specificity, and the mutant OmpA protein which was expressed in an unrelated host showed enhanced ability to bind LPS from its new host. Thus, cell surface expression of the ompAgene appears to depend upon the ability of the gene product to bind LPS, suggesting that an interaction between the protein and LPS plays an essential role in biosynthesis of this outer membrane protein.

Bacteriophage Mud (Casadaban and Cohen 1979) was used to bring the transcription of the gene for ß galactosidase (lacZ) under the control of the promoter of the structural gene for colicin Ib (cia(Ib)) on a derivative of the Col plasmid ColIb.P9. Transcription of this fusion operon was stimulated by agents which damaged cellular DNA (mitomycin C, bleomycin and colicin E2). Increased transcription of the cia-lacZ operon could be detected within 13 min of the addition of these agents. In a strain bearing the tif-1 (recA441) mutation, constitutive expression of the SOS DNA repair system at 42°C also increased transcription of the cia-lacZ operon. Transcription of the cia-lacZ operon was also stimulated by inhibition of DNA gyrase activity with nalidixic acid but not with novobiocin. Transitory inhibition of protein synthesis with chloramphenicol or by proline starvation of a proline auxotroph did not stimulate cia-lacZ transcription. Transcription of the cia-lacZ operon was substantially reduced in the presence of a recA mutation, but was largely unaffected by a mutation in recB affecting the RecBC DNase or by catabolite repression. Control experiments in which the production of colicin Ib was measured confirmed that the experiments with the fusion operon gave an accurate indication as to the activity of the wild type cia gene except for the effect of catabolite repression, where we observed up to 90% reduction in colicin Ib production in strains carrying mutant crp or cya alleles. The overall results confirm previous suggestions that there was considerble similarity between the regulatory systems controlling production of colicins and the repressor-dependent regulation of lambdoid prophage induction.

The DNA sequence of the entire colicin E2 operon was determined. The operon comprises the colicin activity gene, ceaB, the colicin immunity gene, ceiB, and the lysis gene, celB, which is essential for colicin release from producing cells. A potential LexA binding site is located immediately upstream from ceaB, and a rho-independent terminator structure is located immediately downstream from celB. A comparison of the predicted amino acid sequences of colicin E2 and cloacin DF13 revealed extensive stretches of homology. These colicins have different modes of action and recognise different cell surface receptors; the two major regions of heterology at the carboxy terminus, and in the carboxy-terminal end of the central region probably correspond to the catalytic and receptor-recognition domains, respectively. Sequence homologies between colicins E2, A and E1 were less striking, and the colicin E2 immunity protein was not found to share extensive homology with the colicin E3 or cloacin DF13 immunity proteins. The lysis proteins of the ColE2, ColE1 and CloDF13 plasmids are almost identical except in the aminoterminal regions, which themselves have overall similarity with lipoprotein signal peptides. Processing of the ColE2 prolysis protein to the mature form was prevented by globomycin, a specific inhibitor of the lipoprotein signal peptidase. The mature ColE2 lysis protein was located in the cell envelope. The results are discussed in terms of the functional organisation of the colicin operons and the colicin proteins, and the way in which colicins are released from producing cells.

Colicins A, E1, E2, E3, E4, E5, E6, and E7 exhibited reduced activity against BtuB mutants of Escherichia coliK-12 and also against wild-type cells in the presence of vitamin B12. Plasmids encoding representatives of these colicins were specifically immune to high levels of the homologous colicin. Col+cells grown in media containing mitomycin C accumulated large amounts of colicin polypeptide. ColE2+, ColE3+, ColE4+, ColE5+, and ColE6+cultures also synthesized large amounts of second, lower-molecular-weight protein under these conditions. Colicins E2 through E7, but not A or E1, reacted with antiserum raised against purified colicin E3. Colicins E2 and E7 induced synthesis of β-galactosidase encoded by lacZunder the control of the colicin Ib gene promotor on a derivative of Col plasmid ColIb.P9. This promotor is usually active only when the cells are treated with agents which damage DNA or block replication. Plasmids encoding various mutant forms of colicin E3 (M. Mock and M. Schwartz, J. Bacteriol. 142:384-390, 1980) recombined with ColE2, ColE4, ColE5, or ColE6 plasmids at a frequency of 10−4per cell to produce a colicin active against ColE2+, E4+, E5+, or E6+cells. ColE5 and ColE6 plasmids recombined with ColE3 plasmids bearing mutations affecting colicin E3 receptor recognition, envelope penetration, and catalytic activities. ColE2 and ColE4 plasmids recombined only with ColE3 plasmids bearing mutations affecting receptor recognition and envelope penetration. Recombinants between mutant ColE3 plasmids and ColA, ColE1, or ColE7 plasmids were not detected. We propose the designation BtuB group for the colicins described here, and we divide the group into two classes comprising colicins A and E1, which act on the cytoplasmic membrane, and the related colicins E2 through E7, which have known or putative nuclease activities.

Colicins B-K260 and D-CA23 were purified by ammonium sulfate precipitation, gel filtration, and ion-exchange chromatography and were compared with respect to a number of physical and chemical properties. Both colicins were shown to be proteins and were found to have similar molecular weights, isoelectric points and amino acid compositions. The two colicins also have substantial antigenic similarities but are distinguished by the presence of non-cross-reacting antigens and by differences in stability and in sensitivity to heat and reducing conditions. In addition, strains of Escherichia coliK-12 producing colicins B-K260 and D-CA23 are not cross-immune. The similarities noted between the two colicins are compatible with their use of a common cell surface receptor while having different modes of action.

Strains of Escherichia coli K-12 carrying mutations at either of two distinct loci (nmpA and nmpB) produce a new outer membrane pore protein which is not present in wild type cells. Mutations at either of these loci result in sensitivity to phage TC45, which can use this new protein as its receptor, and the new protein (the NmpAB protein) appears to be identical in both NmpA and NmpB mutants. In order to determine whether both of these loci contain structural genes for the NmpAB protein, strains carrying secondary mutations at either of these loci which produced altered proteins were sought by selecting for resistance to phage TC45. Mutants which produced proteins with altered electrophoretic mobility and altered peptide maps were isolated from strains carrying both nmpA and nmpB mutations, and these secondary mutations mapped at the same sites as the original mutations leading to production of the NmpAB protein. This suggests that both loci contain structural genes. Strains mutant at nmpB which can no longer produce the protein can mutate at the nmpA locus to produce the new protein, indicating that both genes can exist in the same cell. Since the altered proteins of mutant strains could be distinguished from one another, we attempted to construct strains in which both nmpA and nmpB were expressed. In all cases only the protein produced by the nmpB mutation was produced, indicating some form of cooperative regulation of the two genes.

A 2,250 base pair Sau3A restriction digest fragment of the colicin plasmid ColE2-P9 was shown to contain all the information necessary for colicin production, immunity to colicin E2 and mitomycin-induced lysis in host cells. Transposon Tn5 insertions in the colicin structural gene (ceaB) completely abolished colicin production and mitomycin-induced immunity protein production and lysis. This indicates the existence of a SOS-regulated colicin E2 operon in which ceaB:: Tn5 mutations exert a polar effect on genes concerened with lysis and production of immunity protein. However, cells harbouring these plasmids were immune to colicin E2, a phenomenon which is probably explained by the existence of an internal, SOS-independent promotor in front of the immunity gene (ceiB). A Tn5 insertion in a site (celB) removed from ceaB and ceiB produced a colicin+ immunity+ lysis- phenotype in host cells treated with mitomycin. These cells produced abnormally low amounts of colicin both in the presence and absence of mitomycin induction. Another difference between these cells and those harbouring wild-type ColE2 plasmids is that the vast majority of the colicin they produced remained within the cells. Cells carrying the wild-type ColE2 plasmid were shown to produce four additional polypeptides when grown in the presence of mitomycin. The two small polypeptides (Mr 5,000 and 3,000) were specifically absent from mitomycin-treated cells harbouring the plasmid with the celB:: Tn5 mutation, whereas all four polypeptides (the other two being the colicin and immunity proteins) were not detected in mitomycin-treated cultures of cells harbouring the plasmids with ceaB:: Tn5 mutations. The possible role of the Mr 5,0000 and 3,000 polypeptides in the regulation of colicin synthesis, cell lysis and colicin release is discussed.

Escherichia coli K-12 cells carrying the high copy number plasmid ColE2-P9 and a sfiA-lacZ gene fusion exhibit abnormally high levels of SOS-regulated fsfiA-lacZ expression. Increased sfiA-lacZ expression is caused by the action of colicin E2, which is a DNase, rather than by the presence of multiple copies of a binding site for LexA protein, the repressor for the sfiA and colicin E2 genes. Expression of sfiA-lacZ was reduced to normal levels if the ColE2+ strain lacked the outer membrane colicin E2 receptor protein (BtuB) or if they carried an increased number of colicin E2 immunity genes. The results suggest that cultures of ColE2+ strains contain a small number of cells which produce colicin which can then enter other, non-producing cells in the culture and cause sufficient damage to the DNA to induce the SOS system. The levels of colicin E2 immunity in the producing cells is presumably sufficient to prevent extensive lethal effects of the colicin, but insufficient to prevent limited endonuclease activity. An important consequence of this phenomenon is that the DNase action of colicin E2 can stimulate its own production.

To study the structure-function relationship of outer membrane pore proteins of E. coli K12, a hybrid gene was constructed in which the DNA encoding amino acid residues 2–73 of the mature PhoE protein is replaced by the homologous part of the related ompF gene. The product of this gene is incorporated normally into the outer membrane. It was characterized with respect to its pore activity and its phage receptor and colicin receptor properties. It is concluded (i) that the preference of the PhoE protein pore for negatively charged solutes is partly determined by the amino terminal 73 amino acids, (ii) that part of the receptor site of PhoE protein for phage TC45 is located in this part of the protein, (iii) that colicin N uses OmpF protein as (part of) its receptor, (iv) that the specificity of OmpF protein as a colicin N receptor is completely located within the 80 amino terminal amino acid residues, whereas the specificity of this protein as a colicin A receptor is completely located within the 260 carboxy terminal amino acid residues, and (v) that the amino terminal 73 amino acid residues of PhoE protein span the membrane at least once.

Nucleotide sequencing of part of the plasmid pCHAP4, which encodes the ca. 42000 Da putative poreforming colicin N, confirmed previous results indicating that the colicin N immunity gene (cni) and the colicin release or lysis gene (cnl) are located immediately downstream from the colicin N structural gene (cna) in the order cna-cni-cnl. The cni gene is transcribed in the opposite direction to cna and probably encodes an Mr 15239 Da protein. The putative immunity protein was detected among the [35S]methionine-labelled proteins produced by minicells carrying cni cloned under lac promoter control, and when the gene was subcloned into expression vectors under the control of a bacteriophage T7 promoter. Deletion of the region immediately upstream from cni completely abolished colicin N immunity, presumably because the natural promoter had been deleted. cnl is in the same operon as cna, and encodes a typical Col plasmid pro-lysis protein comprising a signal peptide and a 34 residue mature polypeptide with high homology to all but one of the other known Col lysis proteins, including the fatty acylated amino-terminal cysteine residue which was specifically labelled with 3H-palmitate. Cell fractionation studies indicated that the cnl gene product was located predominantly in the outer membrane.

DNA sequence analysis, Tnpho and Tntac-1, mutagenesis, deletion analysis, expression under bacteriophage T7 gene 10 promoter control, subcellular fractionation and complementation tests were used to study the function of DNA located in the centre of thepulC-O operon fromKlebsiella oxytoca strain UNF5023. The characterized region of the operon includes five genes (pulG,pulH,pulI,pulJ andpulK) coding for apparently integral inner membrane proteins which are required for pullulanase secretion. The results presented here and previously show that thepulC-O operon contains at least 11 pullulanase secretion genes.

An insertion mutation in the Escherichia coli dsbA gene, coding for periplasmic disulfide oxidoreductase, dramatically reduces the level of OmpF porin protein in the cell envelope. Studies with chromosomal ompF-lacZ operon and gene fusions indicate that this is due to reduced ompF transcription. Identical effects resulted from growth in medium containing the reducing agent dithiothreitol, but the combined effects of the reducing agent and dsbA were no greater than the effects of each individually. The dsbA mutation did not prevent normal regulation of ompF transcription by the local anaesthetic procaine or by osmolarity. OmpF does not contain a cysteine residue, and the sole cysteine residue in the cytoplasmic membrane regulator of ompF transcription, EnvZ, is predicted to be located on the cytoplasmic side of the membrane, and is therefore unlikely to be involved in the effects of the dsbA mutation. The effects of the dsbA mutation and of the reducing agent on ompF transcription may be due to the failure to from an essential disulfide bond in an as yet unidentified envelope protein that affects ompF transcription.

Protein 1, a major protein of the outer membrane of Escherichia coli, has been shown to be the pore allowing the passage of small hydrophilic solutes across the outer membrane. In E. coliK-12 protein 1 consists of two subspecies, 1a and 1b, whereas in E. coliB it consists of a single species which has an electrophoretic mobility similar to that of 1a. K-12 strains mutant at the ompBlocus lack both proteins 1a and 1b and exhibit multiple transport defects, resistance to toxic metal ions, and tolerance to a number of colicins. Mutation at the tolFlocus results in the loss of 1a, in less severe transport defects, and more limited colicin tolerance. Mutation at the parlocus causes the loss of protein 1b, but no transport defects or colicin tolerance. Lysogeny of E. coliby phage PA-2 results in the production of a new major protein, protein 2. Lysogeny of K-12 ompBmutants resulted in dramatic reversal of the transport defects and restoration of the sensitivity to colicins E2 and E3 but not to other colicins. This was shown to be due to the production of protein 2, since lysogeny by phage mutants lacking the ability to elicit protein 2 production did not show this effect. Thus, protein 2 can function as an effective pore. ompBmutations in E. coliB also resulted in loss of protein 1 and similar multiple transport defects, but these were only partially reversed by phage lysogeny and the resulting production of protein 2. When the ompBregion from E. coliB was moved by transduction into an E. coliK-12 background, only small amounts of proteins 1a and 1b were found in the outer membrane. These results indicate that genes governing the synthesis of outer membrane proteins may not function interchangeably between K-12 and B strains, indicating differences in regulation or biosynthesis of these proteins between these strains.

ABSTRACTIn Gram-negative bacteria, the Lol and Bam machineries direct the targeting of lipidated and nonlipidated proteins, respectively, to the outer membrane (OM). Using Pseudomonas aeruginosastrains with depleted levels of specific Bam and Lol proteins, we demonstrated a variable dependence of different OM proteins on these targeting pathways. Reduction in the level of BamA significantly affected the ability of the β-barrel membrane protein OprF to localize to the OM, while the targeting of three secretins that are functionally related OM proteins was less affected (PilQ and PscC) or not at all affected (XcpQ). Depletion of LolB affected all lipoproteins examined and had a variable effect on the nonlipidated proteins. While the levels of OprF, PilQ, and PscC were significantly reduced by LolB depletion, XcpQ was unaffected and was correctly localized to the OM. These results suggest that certain β-barrel proteins such as OprF primarily utilize the complete Bam machinery. The Lol machinery participates in the OM targeting of secretins to variable degrees, likely through its involvement in the assembly of lipidated Bam components. XcpQ, but not PilQ or PscC, was shown to assemble spontaneously into liposomes as multimers. This work raises the possibility that there is a gradient of utilization of Bam and Lol insertion and targeting machineries. Structural features of individual proteins, including their β-barrel content, may determine the propensity of these proteins for folding (or misfolding) during periplasmic transit and OM insertion, thereby influencing the extent of utilization of the Bam targeting machinery, respectively.IMPORTANCETargeting of lipidated and nonlipidated proteins to the outer membrane (OM) compartment in Gram-negative bacteria involves the transfer across the periplasm utilizing the Lol and Bam machineries, respectively. We show that depletion of Bam and Lol components in Pseudomonas aeruginosadoes not lead to a general OM protein translocation defect, but the severity (and therefore, Lol and Bam dependence), varies with individual proteins. XcpQ, the secretin component of the type II secretion apparatus, is translocated into the OM without the assistance of Bam or Lol machineries. The hypothesis that XcpQ, after secretion across the cytoplasmic membrane, does not utilize the OM targeting machineries was supported by demonstrating that in vitro-synthesized XcpQ (but not the other P. aeruginosasecretins) can spontaneously incorporate into lipid vesicles. Therefore, the requirement for ancillary factors appears to be, in certain instances, dictated by the intrinsic properties of individual OM proteins, conceivably reflecting their propensities to misfold during periplasmic transit.