In vivo degradation of secreted fusion proteins by the Escherichia coli outer membrane protease OmpT.

The Escherichia coli outer membrane protease OmpT (protease VII) has been shown to degrade several proteins in vitro, but its function in vivo is uncertain. We demonstrate that OmpT participates in the degradation of a fusion protein secreted into the periplasmic space. A strain with mutations in degP (K.L. Strauch and J. Beckwith, Proc. Natl. Acad. Sci. USA 85:1576-1580, 1988) and ompT exhibits a cumulative decrease in protein degradation and should be useful for the expression of proteolytically sensitive secreted proteins.     

Lysis of Escherichia coli by beta-lactams which bind penicillin-binding proteins 1a and 1b: inhibition by heat shock proteins.

The heat shock proteins (HSPs) of Escherichia coli were artificially induced in cells containing the wild-type rpoH+ gene under control of a tac promoter. At 30 degrees C, expression of HSPs produced cells that were resistant to lysis by cephaloridine and cefsulodin, antibiotics that bind penicillin-binding proteins (PBPs) 1a and 1b. This resistance could be reversed by the simultaneous addition of mecillinam, a beta-lactam that binds PBP 2. However, even in the presence of mecillinam, cells induced to produce HSPs were resistant to lysis by ampicillin, which binds all the major PBPs. Lysis of cells induced to produce HSPs could also be effected by imipenem, a beta-lactam known to lyse nongrowing cells. These effects suggest the existence of at least two pathways for beta-lactam-dependent lysis, one inhibited by HSPs and one not. HSP-mediated lysis resistance was abolished by a mutation in any one of five heat shock genes (dnaK, dnaJ, grpE, GroES, or groEL). Thus, resistance appeared to depend on the expression of the complete heat shock response rather than on any single HSP. Resistance to lysis was significant in the absence of the RelA protein, implying that resistance could not be explained by activation of the stringent response. Since many environmental stresses promote the expression of HSPs, it is possible that their presence contributes an additional mechanism toward development in bacteria of phenotypic tolerance to beta-lactam antibiotics.     

Substrate specificity of the Escherichia coli outer membrane protease OmpT

OmpT is a surface protease of gram-negative bacteria that has been shown to cleave antimicrobial peptides, activate human plasminogen, and degrade some recombinant heterologous proteins. We have analyzed the substrate specificity of OmpT by two complementary substrate filamentous phage display methods: (i) in situ cleavage of phage that display protease-susceptible peptides by Escherichia coli expressing OmpT and (ii) in vitro cleavage of phage-displayed peptides using purified enzyme. Consistent with previous reports, OmpT was found to exhibit a virtual requirement for Arg in the P1 position and a slightly less stringent preference for this residue in the P1' position (P1 and P1' are the residues immediately prior to and following the scissile bond). Lys, Gly, and Val were also found in the P1' position. The most common residues in the P2' position were Val or Ala, and the P3 and P4 positions exhibited a preference for Trp or Arg. Synthetic peptides based upon sequences selected by bacteriophage display were cleaved very efficiently, with kcat/Km values up to 7.3 x 10(6) M(-1) s(-1). In contrast, a peptide corresponding to the cleavage site of human plasminogen was hydrolyzed with a kcat/Km almost 10(6)-fold lower. Overall, the results presented in this work indicate that in addition to the P1 and P1' positions, additional amino acids within a six-residue window (between P4 and P2') contribute to the binding of substrate polypeptides to the OmpT binding site.     

Cell cycle-independent lysis of Escherichia coli by cefsulodin, an inhibitor of penicillin-binding proteins 1a and 1b.

Cefsulodin lyses actively growing Escherichia coli by binding specifically to penicillin-binding proteins (PBPs) 1a and 1b. Recent findings (F. García del Portillo, M. A. de Pedro, D. Joseleau-Petit, and R. D'Ari, J. Bacteriol. 171:4217-4221, 1989) have linked cefsulodin-induced lysis to septation during the first division cycle after a nutritional shift-up or chromosome replication realignment. We synchronized cells by membrane filtration to determine whether cefsulodin-induced lysis depended on septation in normally growing cells. Populations of newly divided cells were allowed to grow for variable lengths of time. Cefsulodin was added to these synchronous cultures, which represented points in two to three rounds of the cell cycle. Since the cell numbers were small, a new lysis assay was developed that was based on the release of DNA measured by fluorometry. Lysis occurred at a constant time after addition of the antibiotic, regardless of the time in the cell cycle at which the addition was made. Thus, cefsulodin-induced lysis is not linked to septation or to any other cell cycle-related event.     

Variability in the posttranslational processing of penicillin-binding protein 1b among different strains of Escherichia coli

Screening of a number of unrelated strains of Escherichia coli confirms the existence of at least two patterns of molecular forms for penicillin-binding protein 1b in E. coli cell envelopes. Our data support that the beta-form of this protein is produced by posttranslational modification of the alpha-form and suggest that the absence of the beta-form in some strains is due to a strain-dependent variability in the alpha-form processing mechanism.     

Nucleotide sequences of genes encoding penicillin-binding proteins from Streptococcus pneumoniae and Streptococcus oralis with high homology to Escherichia coli penicillin-binding proteins 1a and 1b.

The nucleotide sequence of a 3,378-bp DNA fragment of Streptococcus pneumoniae that included the structural gene for penicillin-binding protein (PBP) 1a (ponA), which encodes 719 amino acids, was determined. Homologous DNA fragments from an S. oralis strain were amplified with ponA-specific oligonucleotides. The 2,524-bp S. oralis sequence contained the coding region for the first 636 amino acids of a PBP. The coding sequence differed by 437 nucleotides (27%) and one additional triplet, resulting in 87 amino acid substitutions (14%), from S. pneumoniae PBP 1a. Both PBPs are highly homologous to bifunctional high-M(r) Escherichia coli PBPs 1a and 1b.     

Lipopolysaccharide regions involved in the activation of Escherichia coli outer membrane protease OmpT.

OmpT is an integral outer membrane protease of Escherichia coli. Overexpression of OmpT in E. coli and subsequent in vitro folding of the produced inclusion bodies yielded protein with a native-like structure. However, enzymatically active protease was only obtained after addition of the outer membrane lipid lipopolysaccharide (LPS). OmpT is the first example of an enzyme that requires LPS for activity. In this study, we investigated the nature of this activation. Circular dichroism analysis showed that binding of LPS did not lead to large structural changes. Titration of OmpT with LPS and determining the resulting OmpT activity with a fluorimetric assay yielded a dissociation constant of 10-4 m for E. coli K-12 LPS. Determining the dissociation constants for different LPS chemotypes revealed that a fully acylated lipid A part is minimally required for activation of OmpT. The heptose-bound phosphates in the inner core region were also important for activation. The affinity for LPS was not dependent on the concentration of substrate, neither was affinity for the substrate influenced by the concentration of LPS. This indicated that LPS most likely does not act at the level of substrate binding. We hypothesize that LPS induces a subtle conformational change in the protein that is required for obtaining a native active site geometry.     

Colicin cleavage by OmpT protease during both entry into and release from Escherichia coli cells.

Proteolysis of colicins A, E1, E2, and E3 was observed after they were added to whole cells carrying a functional ompT gene. Recombinant plasmid pML19 containing the ompT gene enabled two mutant strains to cleave the added colicins. On the other hand, two colicin A recombinants were split after release from the wild-type bacteria that produced them but not from ompT mutant cells.     

Topographical distribution of penicillin-binding proteins in the Escherichia coli membrane.

The penicillin-binding proteins (PBPs) found in the membranes of Escherichia coli X925 minicells (primarily cell ends or septa) were compared with those found in rod-shaped cells (primarily sidewalls) in an effort to determine whether certain PBPs are unevenly distributed over the bacterial cell membrane. The seven major PBPs of E. coli were all present in minicell membranes. PBP 1B was altered in minicells, however, appearing as two bands on sodium dodecyl sulfate-polyacrylamide gels rather than the usual three. PBP 2, which is needed for longitudinal growth of the cell but not for septum formation, was significantly reduced in minicell membranes. This observation is consistent with the fact that minicells contain very little sidewall material and raises the possibility that the specialized function of PBP 2 may be determined or regulated by its uneven topographical distribution in the membrane. None of the PBPs appeared to be selectively enriched in minicell membranes.     

Differential responses of Escherichia coli cells expressing cytoplasmic domain mutants of penicillin-binding protein 1b after impairment of penicillin-binding proteins 1a and 3

Penicillin-binding protein 1b (PBP1b) is the major high-molecular-weight PBP in Escherichia coli. Although it is coded by a single gene, it is usually found as a mixture of three isoforms which vary with regard to the length of their N-terminal cytoplasmic tail. We show here that although the cytoplasmic tail seems to play no role in the dimerization of PBP1b, as was originally suspected, only the full-length protein is able to protect the cells against lysis when both PBP1a and PBP3 are inhibited by antibiotics. This suggests a specific role for the full-length PBP1b in the multienzyme peptidoglycan-synthesizing complex that cannot be fulfilled by either PBP1a or the shorter PBP1b proteins. Moreover, we have shown by alanine-stretch-scanning mutagenesis that (i) residues R(11) to G(13) are major determinants for correct translocation and folding of PBP1b and that (ii) the specific interactions involving the full-length PBP1b can be ascribed to the first six residues at the N-terminal end of the cytoplasmic domain. These results are discussed in terms of the interactions with other components of the murein-synthesizing complex.     

Topographical and functional investigation of Escherichia coli penicillin-binding protein 1b by alanine stretch scanning mutagenesis.

Penicillin-binding proteins (PBPs) are the targets of beta-lactam antibiotics. We have used a systematic five-alanine substitution method (called ASS [alanine stretch scanning] mutagenesis) to investigate the functional or structural role of various stretches of amino acids in the PBP1b of Escherichia coli. To probe the specific activity of each variant, the antibiotic discs assay was used with strain QCB1 (delta ponB) in the presence of cefaloridine, which totally inhibits the complementing action of PBP1a. This in vivo test has been combined with a quick and efficient in vitro test of the penicillin-binding activity of each of these variants with fluorescent penicillin. This approach has enabled us to show an unexpected role of the N-terminal and C-terminal tails of PBP1b. Moreover, we have established the correct position in PBP1b of the SMN motif that, with the SXXK and the KTG motifs, constitutes the signature of the penicilloyl serine transferases family. Finally, we have shown that the transglycosylase and the transpeptidase domains are separated by an inert linker region, where substitutions and insertions can be made without hindering the in vivo and in vitro activity of the protein.     

Comparison of Escherichia coli K-12 outer membrane protease OmpT and Salmonella typhimurium E protein.

The predicted amino acid sequence of OmpT, an Escherichia coli outer membrane protease, was found to be highly homologous to that predicted for the pgtE gene product of Salmonella typhimurium. In this paper, it is shown that pgtE codes for a protein functionally homologous to OmpT as judged by its ability to proteolyze T7 RNA polymerase and to localize in the outer membrane of E. coli.     

Approaching the physiological functions of penicillin-binding proteins in Escherichia coli

A rigid shell of peptidoglycan encases and shapes bacteria and is constructed and maintained by a diverse set of enzymes, among which are the penicillin-binding proteins (PBPs). Although a great deal has been learned about how these proteins synthesize and modify peptidoglycan, the physiological functions of the multitude of bacterial PBPs remain enigmatic. We approached this problem by combining PBP mutations in a comprehensive manner and screening for effects on biochemical processes involving the passage of proteins or nucleic acids across the cell wall. The results indicate that the PBPs or their peptidoglycan product do have significant biological functions, including roles in determination of cell shape, in phage resistance, in induction of capsule synthesis, and in regulation of autolysis.     

Comparative studies of penicillin-binding proteins in Pseudomonas aeruginosa and Escherichia coli.

Penicillin-binding proteins in Pseudomonas aeruginosa were compared with those of Escherichia coli. These in P. aeruginosa were found exclusively in the cytoplasmic membrane fraction (fraction soluble in sodium N-lauroyl sarcosinate). Sodium dodecyl sulfate/acrylamide gel electrophoresis of the proteins bound to [14C]penicillin G resulted in the separation of six major bands and several minor bands. The proteins in these bands are referred to as proteins 1A, 1B, 2, 3, 4 and 5 in order of increasing electrophoretical mobility. The electrophoretic mobilities and other properties of penicillin-binding proteins in P. aeruginosa and E. coli were compared and correlated. Fundamentally they seem to be very similar in the two bacteria, but proteins 1A and 1B in P. aeruginosa seem to correspond respectively to proteins 1B and 1A in E. coli, and protein 6 seems to be missing or present in only small amount in P. aeruginosa. In addition, the affinities of currently developed beta-lactam antibiotics to each protein of P. aeruginosa and E. coli were examined in relation to the morphological changes of the cells induced by these antibiotics and their antibacterial potencies. Mecillinam showed high affinity to only protein 2 in both P. aeruginosa and E. coli. At a minimal inhibitory concentration, it converted cells of both P. aeruginosa and E. coli from rods to spherical cells, although its minimal inhibitory concentration was much higher for P. aeruginosa than for E. coli.     

Synthetic ColE1 plasmids carrying genes for penicillin-binding proteins in Escherichia coli

Clarke and Carbon's collection of 2000 Escherichia coli strains which harbor ColE1 plasmids carrying small random segments of the E. coli chromosome was screened for the correction of mutational defects in penicillin-binding proteins (PBPs): ponA (PBP-1a), ponB (PBP-1b), dacB (PBP-4), and pfv (PBP-5). We found plasmids carrying chromosomal segments containing ponA⁺-aroB⁺ (pLC29-47), ponB⁺-tonA⁺ (pLC4-43, pLC4-44, and pLC19-19), and argG⁺-dacB⁺ (pLC10-46 and pLC18-38). Characters of these plasmids were analyzed. Two other plasmids (pLC26-6 and pLC4-14) previously found to correct ftsI mutation (Y. Nishimura, Y. Takeda, A. Nishimura, H. Suzuki, M. Inouye, and Y. Hirota (1977)Plasmid1, 67–77) were also investigated further. Restriction maps of chromosomal DNAs carried by pLC29-47, pLC4-44, pLC19-19, pLC18-38, pLC26-6, and pLC4-14 were constructed. The regions of ponB-tonA on pLC4-44 and pLC19-19, and of leuA-ftsI-murE and F on pLC26-6 were located on the restriction maps. Although both pLC26-6 and pLC4-14 corrected a thermosensitive mutation, ftsI, which causes a defect in cell division due to abnormal PBP-3, only pLC26-6 led to restoration of PBP-3 production by an ftsI mutant, while pLC4-14 did not. Restriction and heteroduplex analyses of pLC26-6 and pLC4-14 have shown the absence of nucleotide sequence homology between them. The plasmids, pLC29-47 carrying ponA⁺ and pLC4-43, pLC4-44, and pLC19-19 carrying ponB⁺ led the host cell to overproduce the respective PBP.     

Escherichia coli outer membrane protease OmpT confers resistance to urinary cationic peptides

Escherichia coli OmpT, located in the outer membrane, has been characterized as a plasminogen activator, with the ability to hydrolyze protamine and block its entry. In this investigation, a complex of low molecular weight cationic peptides purified from human urine by a combination of membrane ultrafiltration and weak cation exchange chromatography was characterized. The impact of OmpT on E. coli resistance to urinary cationic peptides was investigated by testing ompT knockout strains. The ompT mutants were more susceptible to urinary cationic peptides than ompT⁺ strains, and this difference was abolished by complementation of the mutants with pUC19 carrying the ompT gene. The urinary protease inhibitor ulinastatin greatly decreased the resistance of the ompT⁺ strains. Overall, the data indicate that OmpT may help E. coli persist longer in the urinary tract by enabling it to resist the antimicrobial activity of urinary cationic peptides.     

Daughter cell separation by penicillin-binding proteins and peptidoglycan amidases in Escherichia coli

As one of the final steps in the bacterial growth cycle, daughter cells must be released from one another by cutting the shared peptidoglycan wall that separates them. In Escherichia coli, this delicate operation is performed by several peptidoglycan hydrolases, consisting of multiple amidases, lytic transglycosylases, and endopeptidases. The interactions among these enzymes and the molecular mechanics of how separation occurs without lysis are unknown. We show here that deleting the endopeptidase PBP 4 from strains lacking AmiC produces long chains of unseparated cells, indicating that PBP 4 collaborates with the major peptidoglycan amidases during cell separation. Another endopeptidase, PBP 7, fulfills a secondary role. These functions may be responsible for the contributions of PBPs 4 and 7 to the generation of regular cell shape and the production of normal biofilms. In addition, we find that the E. coli peptidoglycan amidases may have different substrate preferences. When the dd-carboxypeptidase PBP 5 was deleted, thereby producing cells with higher levels of pentapeptides, mutants carrying only AmiC produced a higher percentage of cells in chains, while mutants with active AmiA or AmiB were unaffected. The results suggest that AmiC prefers to remove tetrapeptides from peptidoglycan and that AmiA and AmiB either have no preference or prefer pentapeptides. Muropeptide compositions of the mutants corroborated this latter conclusion. Unexpectedly, amidase mutants lacking PBP 5 grew in long twisted chains instead of straight filaments, indicating that overall septal morphology was also defective in these strains.     

Localization of penicillin-binding protein 1b in Escherichia coli: immunoelectron microscopy and immunotransfer studies.

We report the localization of penicillin-binding protein 1b (PBP 1b) in Escherichia coli KN126 and in an overproducing construct containing plasmid pHK231. We used PBP 1b-specific antiserum for the immunoelectron microscopy of ultrathin sections of whole cells and for immunoelectrophoresis of cytoplasm and isolated membrane fractions. We studied ultrathin sections of both glutaraldehyde-fixed cells that had been embedded after progressively lowering the temperature and cryofixed cells that had been freeze-substituted in Lowicryl K4M and HM20. Most of the PBP 1b-specific label was observed in the inner membrane (IM) and the adjacent cytoplasm, much less was observed in the outer membrane (OM); appreciable amounts were also seen in the bulk cytoplasm. Distribution and intensity of label were both temperature dependent: temperature shift-up to 37 degrees C, causing PBP 1b overproduction in the construct, showed a statistically highly significant increase in label of the IM, including a cytoplasmic zone (of at least 30 nm in depth) adjacent to the IM, a zone we termed the membrane-associated area. Concomitant with the temperature shift-up, a decrease in label density was observed in the bulk cytoplasm. Increased label was also found in IM-OM contact areas (zones of membrane adhesion). The periplasm did not show significant label. Western blotting (immunoblotting) revealed PBP 1b in most of the isolated membrane fractions; however, the highest label density was found in membrane fractions of intermediate density, supporting the suggestion of an increased concentration of PBP 1b in the membrane adhesion zones. In summarizing, we propose that PBP 1b is present in the membrane-associated area of the cytoplasm, from where proteins (such as PBP 1b or thioredoxin) gain access to their specific insertion sites in the envelope. The use of several methods of immunoelectron microscopy provided the first unequivocal evidence for localization of PBP 1b at membrane adhesion sites. Since such sites are specifically labeled with anti-PBP 1b serum, we hypothesize that they contain parts of the machinery for assembly and growth of the murein layer.     

Specific location of penicillin-binding proteins within the cell envelope of Escherichia coli.

This communication deals with the location of penicillin-binding proteins in the cell envelope of Escherichia coli. For this purpose, bacterial cells have been broken by various procedures and their envelopes have been fractioned. To do so, inner (cytoplasmic) and outer membranes were separated by isopycnic centrifugation in sucrose gradients. Some separation methods (Osborn et al., J. Biol. Chem. 247:3962-3972, 1972; J. Smit, Y. Kamio, and H. Nikaido, J. Bacteriol. 124:942-958, 1975) revealed that penicillin-binding proteins are not exclusively located in the inner membrane. They are also found in the outer membrane (A. Rodríguez-Tébar, J. A. Barbas, and D. Vásquez, J. Bacteriol. 161:243-248, 1985). Under the milder conditions for cell rupture used in this work, an intermembrane fraction, sedimenting between the inner and outer membrane, can be recovered from the gradients. This fraction has a high content of both penicillin-binding proteins and phospholipase B activity and may correspond to the intermembrane adhesion sites (M. H. Bayer, G. P. Costello, and M. E. Bayer, J. Bacteriol. 149:758-769, 1982). We postulate that this intermembrane fraction is a labile structure that contains a high amount of all penicillin-binding proteins which are usually found in both the inner and outer membranes when the adhesion sites are destroyed by the cell breakage and fractionation procedures.     

Interaction of monoclonal antibodies with the enzymatic domains of penicillin-binding protein 1b of Escherichia coli.

Monoclonal antibodies (MAbs) against four different antigenic determinants of penicillin-binding protein (PBP) 1b were used to study the transglycosylase and transpeptidase activities of PBP 1b. Enzyme kinetics in the presence of and without the MAbs were determined, and the synthesized murein was analyzed. Two MAbs against the transglycosylase domain of PBP 1b appeared to inhibit this reaction. One MAb inhibited only the transpeptidase reaction, and one inhibited both enzymatic activities of PBP 1b. The latter two MAbs bound to the transpeptidase domain of PBP 1b. The following major conclusions were deduced from the results. (i) Transpeptidation is the rate-limiting step of the reaction cascade, and it is dependent on the product of transglycosylation. (ii) PBP 1b has only one type of transpeptidase activity, i.e., a penta-tetra transpeptidase activity. (iii) PBP 1b is probably a globular protein which has two intimately associated enzymatic domains.