Differential release of periplasmic versus cytoplasmic enzymes from Escherichia coli B by polymixin B

Subcellular Localization of Signal Peptide Fusion Proteins Expressed in E. coli

For expression of some proteins in Escherichia coli, export to the periplasmic space is preferred over conventional expression in the cytosol. Export can be accomplished by fusing the coding sequence to DNA encoding a signal peptide (e.g., using pET-22b), which is cleaved by the bacterial signal peptidase as the protein is exported into the space between the inner and outer membranes of E. coli This protocol uses osmotic shock to release polypeptides from the periplasm. Although not quantitative, it should provide preliminary information on the cellular location of signal peptide fusion proteins.

Antibiotic Resistance by Enzymatic Modification of Antibiotic Targets

Antibiotic resistance remains a significant threat to modern medicine. Modification of the antibiotic target is a resistance strategy that is increasingly prevalent among pathogens. Examples include resistance to glycopeptide and polymyxin antibiotics that occurs via chemical modification of their molecular targets in the cell envelope. Similarly, many ribosome-targeting antibiotics are impaired by methylation of the rRNA. In these cases, the antibiotic target is subjected to enzymatic modification rather than genetic mutation, and in many instances the resistance enzymes are readily mobilized among pathogens. Understanding the enzymes responsible for these modifications is crucial to combat resistance. Here, we review our current understanding of enzymatic modification of antibiotic targets as well as discuss efforts to combat these resistance mechanisms.

Soluble versions of outer membrane cytochromes function as exporters for heterologously produced cargo proteins

This study reveals that it is possible to secrete truncated versions of outer membrane cytochromes into the culture supernatant and that these proteins can provide a basis for the export of heterologously produced proteins. Different soluble and truncated versions of the outer membrane cytochrome MtrF were analyzed for their suitability to be secreted. A protein version with a very short truncation of the N-terminus to remove the recognition sequence for the addition of a lipid anchor is secreted efficiently to the culture supernatant, and moreover this protein could be further truncated by a deletion of 160 amino acid and still is detectable in the supernatant. By coupling a cellulase to this soluble outer membrane cytochrome, the export efficiency was measured by means of relative cellulase activity. We conclude that outer membrane cytochromes of S. oneidensis can be applied as transporters for the export of target proteins into the medium using the type II secretion pathway.

Lipopolysaccharide-Dependent Membrane Permeation and Lipid Clustering Caused by Cyclic Lipopeptide Colistin

Polyanionic lipopolysaccharides (LPS) play an important role in regulating the permeability of the outer membrane (OM) of Gram-negative bacteria. Impairment of the LPS-enriched OM is essential in initiating the bactericidal activity of polymyxins. We are interested in how colistin (polymyxin E) affects the membrane permeability of LPS/phospholipid bilayers. Our vesicle leakage experiment showed that colistin binding enhanced bilayer permeability; the maximum increase in the bilayer permeability was positively correlated with the LPS fraction. Addition of magnesium ions abolished the effect of LPS in enhancing bilayer permeabilization. To describe the vesicle leakage behavior from a structural perspective, we performed liquid atomic force microscopy (AFM) measurements on planar lipid bilayers. We found that colistin caused the formation of nano- and macroclusters that protruded from the bilayer by ∼2 nm. Moreover, cluster development was promoted by increasing the fraction of LPS or colistin concentration but inhibited by magnesium ions. To explain our experimental data, we proposed a lipid clustering model where colistin binds to LPS to form large-scale complexes segregated from zwitterionic phospholipids. The discontinuity (and thickness mismatch) at the edge of LPS-colistin clusters will create a passage that allows solutes to permeate through. The proposed model is consistent with all data obtained from our leakage and AFM experiments. Our results of LPS-dependent membrane restructuring provided useful insights into the mechanism that could be used by polymyxins in impairing the permeability barrier of the OM of Gram-negative bacteria.

Novobiocin Enhances Polymyxin Activity by Stimulating Lipopolysaccharide Transport

Gram-negative bacteria are challenging to kill with antibiotics due to their impenetrable outer membrane containing lipopolysaccharide (LPS). The polymyxins, including colistin, are the drugs of last resort for treating Gram-negative infections. These drugs bind LPS and disrupt the outer membrane; however, their toxicity limits their usefulness. Polymyxin has been shown to synergize with many antibiotics including novobiocin, which inhibits DNA gyrase, by facilitating transport of these antibiotics across the outer membrane. Recently, we have shown that novobiocin not only inhibits DNA gyrase but also binds and stimulates LptB, the ATPase that powers LPS transport. Here, we report the synthesis of novobiocin derivatives that separate these two activities. One analog retains LptB-stimulatory activity but is unable to inhibit DNA gyrase. This analog, which is not toxic on its own, nevertheless enhances the lethality of polymyxin by binding LptB and stimulating LPS transport. Therefore, LPS transport agonism contributes substantially to novobiocin-polymyxin synergy. We also report other novobiocin analogs that inhibit DNA gyrase better than or equal to novobiocin, but bind better to LptB and therefore have even greater LptB stimulatory activity. These compounds are more potent than novobiocin when used in combination with polymyxin. Novobiocin analogs optimized for both gyrase inhibition and LPS transport agonism may allow the use of lower doses of polymyxin, increasing its efficacy and safety.

Polymyxin: Alternative Mechanisms of Action and Resistance

Antibiotic resistance among pathogenic bacteria is an ever-increasing issue worldwide. Unfortunately, very little has been achieved in the pharmaceutical industry to combat this problem. This has led researchers and the medical field to revisit past drugs that were deemed too toxic for clinical use. In particular, the cyclic cationic peptides polymyxin B and colistin, which are specific for Gram-negative bacteria, have been used as "last resort" antimicrobials. Before the 1980s, these drugs were known for their renal and neural toxicities; however, new clinical practices and possibly improved manufacturing have made them safer to use. Previously suggested to primarily attack the membranes of Gram-negative bacteria and to not easily select for resistant mutants, recent research exploring resistance and mechanisms of action has provided new perspectives. This review focuses primarily on the proposed alternative mechanisms of action, known resistance mechanisms, and how these support the alternative mechanisms of action.

A globally distributed mobile genetic element inhibits natural transformation of Vibrio cholerae

Natural transformation is one mechanism of horizontal gene transfer (HGT) in Vibrio cholerae, the causative agent of cholera. Recently, it was found that V. cholerae isolates from the Haiti outbreak were poorly transformed by this mechanism. Here, we show that an integrating conjugative element (ICE)-encoded DNase, which we name IdeA, is necessary and sufficient for inhibiting natural transformation of Haiti outbreak strains. We demonstrate that IdeA inhibits this mechanism of HGT in cis via DNA endonuclease activity that is localized to the periplasm. Furthermore, we show that natural transformation between cholera strains in a relevant environmental context is inhibited by IdeA. The ICE encoding IdeA is globally distributed. Therefore, we analyzed the prevalence and role for this ICE in limiting natural transformation of isolates from Bangladesh collected between 2001 and 2011. We found that IdeA(+) ICEs were nearly ubiquitous in isolates from 2001 to 2005; however, their prevalence decreased to ∼40% from 2006 to 2011. Thus, IdeA(+) ICEs may have limited the role of natural transformation in V. cholerae. However, the rise in prevalence of strains lacking IdeA may now increase the role of this conserved mechanism of HGT in the evolution of this pathogen.

Polymyxin B-Induced Release of Low-Molecular-Weight, Heat-Labile Enterotoxin from Escherichia coli

Polymyxin B-induced release of enterotoxin from Escherichia coli strain H-10407 was demonstrated. Incubation of E. coli cells derived from 6-h cultures with polymyxin caused the rapid release of enterotoxin with a molecular weight of approximately 20,000, as estimated by the gel filtration technique. The rapidity of the release of enterotoxin indicates that it probably resides in the periplasmic space of the cell. The low-molecular-weight enterotoxin possessed vascular permeability factor and diarrheagenic activities, both of which were found to be heat-labile. The permeability factor activity of this enterotoxin was neutralized by antisera prepared against crude E. coli enterotoxin, Vibrio cholerae enterotoxin (choleragen), and V. cholerae toxoid (choleragenoid), respectively. Supernatant fluids of 6-h E. coli cultures did not contain this molecular form of enterotoxin but did contain very high-molecular-weight, heat-labile enterotoxin. Incubation of cells derived from older (18 h) cultures with polymyxin caused the release of both low- (20,000) and high-molecular-weight forms of enterotoxin. We concluded that either the 20,000-dalton form of heat-labile enterotoxin is not released by E. coli under in vitro growth conditions or that enterotoxin released in this form is rapidly destroyed or inactivated.

Periplasmic electron transfer via the c-type cytochromes MtrA and FccA of Shewanella oneidensis MR-1

Dissimilatory microbial reduction of insoluble Fe(III) oxides is a geochemically and ecologically important process which involves the transfer of cellular, respiratory electrons from the cytoplasmic membrane to insoluble, extracellular, mineral-phase electron acceptors. In this paper evidence is provided for the function of the periplasmic fumarate reductase FccA and the decaheme c-type cytochrome MtrA in periplasmic electron transfer reactions in the gammaproteobacterium Shewanella oneidensis. Both proteins are abundant in the periplasm of ferric citrate-reducing S. oneidensis cells. In vitro fumarate reductase FccA and c-type cytochrome MtrA were reduced by the cytoplasmic membrane-bound protein CymA. Electron transfer between CymA and MtrA was 1.4-fold faster than the CymA-catalyzed reduction of FccA. Further experiments showing a bidirectional electron transfer between FccA and MtrA provided evidence for an electron transfer network in the periplasmic space of S. oneidensis. Hence, FccA could function in both the electron transport to fumarate and via MtrA to mineral-phase Fe(III). Growth experiments with a DeltafccA deletion mutant suggest a role of FccA as a transient electron storage protein.

Production of Shiga toxin by Shiga toxin-expressing Escherichia coli (STEC) in broth media: from divergence to definition

AIMS

To determine the suitability of eight different commercial broth media for Shiga toxin (Stx) production.

METHODS AND RESULTS

Shiga toxin-producing Escherichia coli (STEC) strains producing Stx1 or Stx2 were grown at 37 degrees C (250 rev min(-1)) for 24 h in brain heart infusion broth, E. coli broth, Evans medium, Luria-Bertani broth, Penassay broth, buffered-peptone water, syncase broth and trypticase soy broth. Toxin production was measured by enzyme-linked immunosorbent assay (ELISA) in polymyxin-treated cell pellets and/or supernatants of cultures, ELISA optical densities reached 1 when isolates were grown for 2-4 h in E. coli broth in the presence of antibiotic. Besides, a collection of STEC-expressing Stx strains was evaluated and the Stx production was assayed in the supernatants and in polymyxin-treated pellets of bacterial growth after 4 h of cultivation in E. coli broth in the presence of antibiotic.

CONCLUSIONS

The most suitable medium for Stx production was E. coli broth when the bacterial isolates were grown for 4 h in the presence of ciprofloxacin and the Stx production is detected in the supernatant.

SIGNIFICANCE AND IMPACT OF THE STUDY

This study presents the first comprehensive comparison of different broth media with regard to Stx production to establish optimal culture conditions for STEC detection in routine diagnostic laboratories.

Polymyxin B induces lysis of marine pseudoalteromonads

Polymyxin B (PMB) is a cationic antibiotic that interacts with the envelopes of gram-negative bacterial cells. The therapeutic use of PMB was abandoned for a long time due to its undesirable side effects; however, the spread of resistance to currently used antibiotics has forced the reevaluation of PMB for clinical use. Previous studies have used enteric bacteria to examine the mode of PMB action, resulting in a somewhat limited understanding of this process. This study examined the effects of PMB on marine pseudoalteromonads and demonstrates that the frequently accepted view that "what is true for Escherichia coli is true for all bacteria" does not hold true. We show here that in contrast to the growth inhibition observed for enteric bacteria, PMB induces lysis of pseudoalteromonads, which is not prevented by high concentrations of divalent cations. Furthermore, we demonstrate that a high membrane voltage is required for the interaction of PMB with the cytoplasmic membranes of pseudoalteromonads, further elucidating the mechanisms by which PMB interacts with the cell envelopes of those gram-negative bacteria.

Production, characterization, and application of antibodies against heat-labile type-I toxin for detection of enterotoxigenic Escherichia coli

Strains of enterotoxigenic Escherichia coli (ETEC) are responsible for significant rates of morbidity and mortality among children, particularly in developing countries. The majority of clinical and public health laboratories are capable of isolating and identifying Salmonella, Shigella, Campylobacter, and Escherichia coli O157:H7 from stool samples, but ETEC cannot be identified by routine methods. The method most often used to identify ETEC is polymerase chain reaction for heat-stable and heat-labile enterotoxin genes, and subsequent serotyping, but most clinical and public health laboratories do not have the capacity or resources to perform these tests. In this study, polyclonal rabbit and monoclonal mouse IgG2b antibodies against ETEC heat-labile toxin-I (LT) were characterized and the potential applicability of a capture assay was analyzed. IgG-enriched fractions from rabbit polyclonal and the IgG2b monoclonal antibodies recognized LT in a conformational shape and they were excellent tools for detection of LT-producing strains. These findings indicate that the capture immunoassay could be used as a diagnostic assay of ETEC LT-producing strains in routine diagnosis and in epidemiological studies of diarrhea in developing countries as enzyme linked immunosorbent assay techniques remain as effective and economical choice for the detection of specific pathogen antigens in cultures.

Modification of the glycolipid-binding specificity of vero cytotoxin by polymyxin B and other cyclic amphipathic peptides

Polymyxin B, an amphipathic cyclic decapeptide produced by Bacillus polymyxa, is routinely used in the extraction of the components from the periplasmic space of gram-negative bacteria. Vero cytotoxin 1 (VT1) is an Escherichia coli-elaborated subunit toxin which binds to the glycolipid globotriosylceramide (Gal-alpha 1-4-Gal beta 1-4-Glc-ceramide [Gb3]) and has been strongly implicated in the etiology of the hemolytic uremic syndrome and hemorrhagic colitis. We now show by in vitro glycolipid-binding assays that in the presence of low concentrations of polymyxin B, globotetraosylceramide (GalNAc beta 1-3Gal alpha 1-4Gal beta 1-4Glc-ceramide [Gb4]) is also recognized by both the VT1 B (binding) subunit and holotoxin. Melittin, a 26-amino-acid cyclic peptide of similar amphipathic nature, produced the same effect, whereas a hydrophobic blocking agent did not. Triton X-100 did not increase binding of VT1 to Gb4 but prevented glycolipid binding in toto at concentrations above 0.5%. Caution is therefore advised in the analysis of VT1 glycolipid binding in the presence of amphipathic peptides.

The action of various venoms on Escherichia coli

The antibacterial activity of honeybee venom (Apis mellifera), three snake venoms (Naja naja sputatrix, Vipera russellii and Crotalus adamanteus) and the polypeptide melittin was investigated against Escherichia coli. Minimum inhibitory concentration values, cell lysis and alterations in cell permeability were determined and action against E. coli was in the order: A. mellifera venom greater than melittin greater than N. naja sputatrix venom much greater than V. russellii venom greater than C. adamanteus venom. Cellular damage by A. mellifera and N. naja sputatrix venoms was evident in electron micrographs.

Selective action of mycobacillin on the uptake of releasable cell materials by Aspergillus niger

The uptake of normally releasable (i.e. releasable in the absence of the antibiotic) cell constituents (namely lysine, proline, ATP, Pi, Na+, K+ and Ca2+) by sensitive cells of Aspergillus niger that occurs in the absence of mycobacillin is gradually enhanced with increase in concentration of the antibiotic until the uptake attains the maximum. With still higher concentrations the uptake decreases until it becomes the same as in the control without mycobacillin. Uptake follows saturation kinetics both in the absence and in the presence of the antibiotic. Mycobacillin significantly increases Vmax. for uptake with any effect on Km, Mycobacillin has no action on the uptake of non-releasable materials.

Lincomycin stimulates synthesis of TEM-2 beta-lactamase by Escherichia coli

Lincomycin increased the TEM-2 beta-lactamase activity of Escherichia coli K-12 cells carrying plasmid RP4 at a concentration which slightly inhibited cell growth. In a control culture beta-lactamase activity reached its maximal level in late log phase, whereas when lincomycin was present beta-lactamase activity continued to increase into the stationary phase. Lincomycin (100 micrograms/ml) inhibited both cell growth and protein synthesis by about 35% but stimulated beta-lactamase activity 2.5-fold per ml of culture and about 4-fold per cell after 20 h of growth. The amount of beta-lactamase produced in each culture was also compared by densitophotometry of a stained sodium dodecyl sulfate-polyacrylamide gel. The relative values were in good agreement with the relative enzyme activities, indicating that the stimulatory effect of lincomycin was due to an increase in the amount of beta-lactamase protein. Inactivation of beta-lactamase appeared to be faster when lincomycin was present. This was determined by measuring the decrease in beta-lactamase activity when phenethyl alcohol was present to prevent maturation of the enzyme. There was no significant difference in plasmid copy number between the cells grown in the presence or absence of lincomycin. These results indicate that lincomycin stimulates transcription, translation, or translocation of beta-lactamase.

Leakage of periplasmic proteins from Escherichia coli mediated by polymyxin B nonapeptide

The effects of polymyxin B and polymyxin B nonapeptide (PMBN) on cell envelope integrity in Escherichia coli were compared. Both compounds caused loss of proteins from E. coli K-12 3300(pBR322), although PMBN released less protein than did polymyxin B. The origin of the released protein was determined both by polyacrylamide gel electrophoresis and by using specific enzyme markers (beta-lactamase in periplasm, beta-galactosidase in cytoplasm). The proteins released by both compounds were derived principally from the periplasm, accompanied in the case of polymyxin B by a low level of cytoplasmic proteins. Although polymyxin B and PMBN both caused release of periplasmic proteins, the individual proteins released by the compounds differed. The periplasmic fraction contained six principal polypeptides with molecular weights between 62,000 (polypeptide 1) and 29,000 (polypeptide 6). Polypeptide 6 was identified as the pBR322-encoded beta-lactamase, but the other proteins were not specifically identified. Polymyxin B caused considerable release of polypeptides 1, 2, and 5 with some release of polypeptides 4 and 6. PMBN released polypeptide 1 (trace), 3, 4, and 6 (trace). Scanning electron microscopy showed that polymyxin B and PMBN both caused surface damage in E. coli. However, polymyxin B produced greater morphological changes than PMBN.

Binding and neutralization of bacterial lipopolysaccharide by colistin nonapeptide

Polymyxin nonapeptides, proteolytic derivatives of polymyxin antibiotics, are less toxic than their parent compounds but retain some of their antibacterial activities. To confirm and expand observations that polymyxin nonapeptides have anti-endotoxin activity, we studied the ability of colistin nonapeptide to bind to bacterial lipopolysaccharide (LPS) and to inhibit the effects of LPS on Limulus amoebocyte lysate and lymphocyte mitogenicity. Colistin nonapeptide was purified by high-pressure liquid chromatography and was demonstrated to bind to LPS by equilibrium dialysis. The ability of colistin nonapeptide to render E. coli ATCC 25922 cells sensitive to erythromycin was abrogated by 50% after incubation with E. coli O18 LPS in a ratio by weight of LPS to colistin nonapeptide of 3.9:1. The presence of 4 micrograms of colistin nonapeptide or colistin per ml increased by 130- and 800-fold, respectively, the concentration of E. coli O113 LPS required to produce 50% gelation of Limulus amoebocyte lysate as measured by a spectrophotometric assay. Neutralization of LPS by colistin nonapeptide was time and concentration dependent. In contrast to the neutralization seen with LPS derived from a colistin-sensitive organism, colistin nonapeptide neutralized very little LPS extracted from a strain of Serratia marcescens that was resistant to colistin. Colistin nonapeptide also inhibited LPS-induced [3H]thymidine uptake by splenic lymphocytes, but its activity was less than 1/10 that of colistin. We conclude that colistin nonapeptide binds to LPS and possesses antiendotoxin activity. However, the anti-endotoxin activity of the nonapeptide is considerably less than that of its parent compound, colistin.

Effects of polymyxin B sulfate and polymyxin B nonapeptide on growth and permeability of the yeast Saccharomyces cerevisiae

Polymyxin B, a toxic, membrane-affecting antibiotic, can be rendered harmless to yeast cells by enzymatic removal of its fatty acyl moiety. The remaining cyclic peptide portion, polymyxin B nonapeptide, has no significant effect on growth and viability but it drastically reduces mating efficiency. In addition, the cyclic peptide enhances sensitivity of cells to several drugs, presumably by increasing membrane permeability. Mutants resistant to polymyxin B are simultaneously less responsive to the combination of the nonapeptide and the drugs. This indicates that the peptide portion of polymyxin B is the moiety responsible for the permeability changes. The resistance is inherited as a simple recessive trait. The mutation has been mapped to chromosome XV of Saccharomyces cerevisiae.

Nucleotide sequence of the thermostable direct hemolysin gene of Vibrio parahaemolyticus

The gene encoding the thermostable direct hemolysin of Vibrio parahaemolyticus was characterized. This gene (designated tdh) was subcloned into pBR322 in Escherichia coli, and the functional tdh gene was localized to a 1.3-kilobase HindIII fragment. This fragment was sequenced, and the structural gene was found to encode a mature protein of 165 amino acid residues. The mature protein sequence was preceded by a putative signal peptide sequence of 24 amino acids. A putative tdh promoter, determined by its similarity to concensus sequences, was not functional in E. coli. However, a promoter that was functional in E. coli was shown to exist further upstream by use of a promoter probe plasmid. A 5.7-kilobase SalI fragment containing the structural gene and both potential promoters was cloned into a broad-host-range plasmid and mobilized into a Kanagawa phenomenon-negative V. parahaemolyticus strain. In contrast to E. coli, where the hemolysin was detected only in cell lysates, introduction of the cloned gene into V. parahaemolyticus resulted in the production of extracellular hemolysin.

Antibacterial action of colistin (polymyxin E) against Mycobacterium aurum

Mycobacterium aurum was susceptible to the antibiotic colistin (polymyxin E),which had an MIC of 5 micrograms/ml and an apparent bactericidal effect at concentrations above 50 micrograms/ml. Treatment of actively growing cells with sublethal concentrations of colistin (15 micrograms/ml) resulted in synchronized cell division once the antibiotic was removed. Under conditions of synchronized cell growth, one cycle of DNA replication lasted 120 min and one cycle of cell division lasted about 180 min. Although the antibiotic treatment during synchronization experiments did not produce apparent changes in the bacterial envelope, it was accompanied by the accumulation of a polysaccharide-like substance in the bacterial cytoplasm which gradually decreased after the removal of antibiotic and by an increase in the number of mesosomes at 3 h after antibiotic removal. This step was closely linked to the doubling time of bacteria. Lethal concentrations of colistin of 50 and 100 micrograms/ml, which caused about 90 and 99% cell death, respectively, produced significant cytoplasmic membrane injuries, patchy appearance of the cell wall outer polysaccharide layer, and little cell lysis. These data indicate that the cytoplasmic membrane is a site of action of colistin and raise a question as to whether an outer bilayer exists in mycobacteria, at least functionally.

Improved GM1-enzyme-linked immunosorbent assay for detection of Escherichia coli heat-labile enterotoxin

Previously described GM1 ganglioside enzyme-linked immunosorbent assays (GM1-ELISA) for the detection of Escherichia coli heat-labile enterotoxin (LT) showed sensitivity equal to the Y-1 adrenal cell assay when anti-LT (a reagent not commercially available) was used. However, when antitoxin to immunologically related (commercially available) cholera toxin was substituted, a marked loss in sensitivity occurred. We modified the GM1-ELISA that employed anti-cholera toxin to make it comparable in sensitivity to the Y-1 adrenal cell assay. When five media commonly used for LT production were compared, Mundell's Casamino Acids medium was shown to be significantly superior. Lincomycin (45 micrograms/ml) added to E. coli cultures significantly increased net optical densities in the GM1-ELISA, a direct measure of the amount of LT. Treatment of broth cultures or bacterial cell pellets with polymyxin B or extension of culture time to 48 h also significantly increased net optical density by allowing enhanced release of periplasmic LT. A major innovation involved the direct culture of E. coli strains in GM1-coated wells of microtiter plates followed by ELISA. This direct culture method GM1-ELISA (DCM-GM1-ELISA) saved not only assay time, but also materials and reagents. The net optical densities that result from this assay allow the test to be read visually without a spectrophotometer. Three independent observers read plates with E. coli tested by DCM-GM1-ELISA. Thirty-four of 35 adrenal cell-positive strains (97% sensitivity) and 30 of 30 LT-negative control E. coli strains (100% specificity) were identified by all three observers reading coded plates. The DCM-GM1-ELISA provides a simple, practical and efficient assay for LT for less sophisticated laboratories.

Polycations as outer membrane-disorganizing agents

The outer membrane-disorganizing effect of a short (10-min) treatment with polycationic agents was studied with smooth Salmonella typhimurium used as a test organism. The polycationic agents were the protamine salmine, a lysine polymer with 20 lysine residues (lysine20), and the deacylated polymyxin B derivative polymyxin B nonapeptide. Two different types of outer membrane-disorganizing were found. Protamine and lysine20 released 20 to 30% of the lipopolysaccharide from the outer membrane and sensitized the bacteria to the anionic detergent sodium dodecyl sulfate but did not (under these conditions) make the bacteria permeable to the hydrophobic probes fusidic acid and actinomycin D. In contrast, polymyxin B nonapeptide did not release lipopolysaccharide or sensitize the bacteria to sodium dodecyl sulfate but made the outer membrane permeable to the hydrophobic probes. None of the agents was bactericidal under the conditions used or caused any leakage of periplasmic beta-lactamase. Polymyxin B was used as a reference and showed characteristic outer membrane-disorganizing action. In thin-section electron microscopy, polymyxin B nonapeptide caused the appearance of long, narrow, finger-like projections on the outer membrane. Protamine and lysine20 caused a distinctly wrinkled appearance of the outer membrane but no projections.

Release of Shiga toxin from Shigella dysenteriae 1 by polymyxin B

Release of Shiga toxin from Shigella dysenteriae 1 was found to occur after exposure to polymyxin B. The amount of toxin released was dependent on both the polymyxin concentration and time of incubation. An immunoblot characterization of the Shiga toxin released by polymyxin treatment demonstrated that it is electrophoretically similar to purified Shiga toxin and to Shiga toxin present in crude bacterial sonicates of S. dysenteriae 1 cells.

Shigella dysenteriae 1 cytotoxin: periplasmic protein releasable by polymyxin B and osmotic shock

Treatment of Shigella dysenteriae 1 either with the antibiotic polymyxin B or by osmotic shock resulted in the release of 80 to 90% of the cytotoxin activity of the organism. Under the conditions employed, the release of toxin activity was accompanied by the appearance of a periplasmic enzyme, 5'-nucleotidase. There was no significant release of cytoplasmic contents, assessed by measurement of glucose-6-phosphate dehydrogenase activity. The release of cytotoxin and 5'-nucleotidase by polymyxin B were both dependent on the duration of incubation with, and the concentration of, the antibiotic. In terms of specific activity (cytotoxin activity per milligram of protein), the polymyxin B and osmotic shock extracts were 20- to 30-fold more active than crude toxin preparation derived from a whole-cell lysate. The data strongly support a periplasmic location for Shiga cytotoxin and the utility of the polymyxin B extraction to obtain starting material for toxin purification.

Periplasmic enzymes in Bdellovibrio bacteriovorus and Bdellovibrio stolpii

When cells of either Bdellovibrio bacteriovorus 109J or Bdellovibrio stolpii UKi2 were subjected to osmotic shock by treatment with sucrose-EDTA and MgCl2 solutions, only trace amounts of proteins or enzyme activities were released into the shock fluid. In contrast, when nongrowing cells were converted to motile, osmotically stable, peptidoglycan-free spheroplasts by penicillin treatment, numerous proteins were released into the suspending fluid. For both species, this suspending fluid contained substantial levels of 5'-nucleotidase, purine phosphorylase, and deoxyribose-phosphate aldolase. Penicillin treatment also released aminoendopeptidase N from B. bacteriovorus, but not from B. stolpii. Penicillin treatment did not cause release of cytoplasmic enzymes such as malate dehydrogenase. The data indicated that bdellovibrios possess periplasmic enzymes or peripheral enzymes associated with the cell wall complex. During intraperiplasmic bdellovibrio growth, periplasmic and cytoplasmic enzymes of the Escherichia coli substrate cell were not released upon formation of the spherical bdelloplast during bdellovibrio penetration. Most of the E. coli enzymes were retained within the bdelloplast until later in the growth cycle, when they became inactivated or released into the suspending buffer or both.

Detection of toxins produced by vibrio fluvialis

The results of studies with cell-free extracts and culture supernatant fluids of Vibrio fluvialis (a recently recognized, potential enteric pathogen for humans) grown in the absence and presence of lincomycin indicated that the bacterium could produce (i) a factor which causes CHO cell elongation (CEF) similar to that elicited by V. cholerae enterotoxin and by the heat-labile enterotoxin of Escherichia coli, (ii) cytolysin(s) active against erythrocytes, (iii) nonhemolytic, CHO cell-killing factor(s), and (iv) protease(s) active against azocasein. The CEF was heat labile and ammonium sulfate precipitable, and it had an isoelectric point (estimated by sucrose density gradient electrofocusing) and molecular weight (estimated by gel filtration) of about 5.1 and 135,000, respectively.

Outer membrane permeability barrier disruption by polymyxin in polymyxin-susceptible and -resistant Salmonella typhimurium

In contrast to their polymyxin-susceptible parent strains, polymyxin-resistant Salmonella typhimurium mutants (pmrA strains) did not lose their outer membrane permeability barrier to macromolecules such as lysozyme and periplasmic proteins upon polymyxin treatment. The sensitization of pmrA strains to deoxycholate-induced lysis required 10-times-higher polymyxin concentrations than did the sensitization of the parent strains. These findings indicate that the pmrA mutation affects the outer membrane and decreases its susceptibility to polymyxin. By contrast, the pmrA mutants did not differ from their parents in the uptake of gentian violet after treatment with polymyxin, suggesting a degree of specificity in the pmrA effect in the outer membrane.

Isolation and characterization of minicell-producing mutants of Shigella spp

Minicells are small, anucleate cells resulting from aberrant cell divisions at the polar ends of bacilli. We have isolated minicell-producing mutant strains of Shigella flexneri 2a (MC-I) and Shigella dysenteriae 1 (MC-V) after mutagenesis with N-methyl-N'-nitro-N-nitrosoguanidine. Microscopically, broth cultures of MC-I and MC-V were found to contain free minicells, normal cells, and filamentous cells with polar, attached minicells. Both strains retained their ability to provoke keratoconjunctivitis in guinea pigs and to invade HeLa cells. Purified suspensions of minicells containing less than one whole cell per 10(6) minicells were obtained by a combination of differential sedimentation and density gradient centrifugation (5 to 30% [wt/vol] linear sucrose gradients). Each MC-I minicell contained about 0.005 times the amount of deoxyribonucleic acid of one normal S. flexneri. The MC-V minicell had about 0.003 times the amount of deoxyribonucleic acid of one whole S. dysenteriae cell. Purified MC-V minicells were treated with polymyxin B to release Shiga toxin. Shiga toxin was readily detected in MC-V minicells by means of a microtiter HeLa cell cytotoxicity assay. Our findings indicate that such a minicell-producing alteration in the cell division cycle of shigellae has not significantly affected their virulence.

Lincomycin increases synthetic rate and periplasmic pool size for cholera toxin

Increased enterotoxigenicity of Vibrio cholerae 569B grown with low concentrations of lincomycin, previously described in terms of increased extracellular biological activity (capillary permeability factor and fluid accumulation in ligated rabbit ileal loops), was further characterized. Polyacrylamide gel electrophoresis and single radial immunodiffusion showed that lincomycin-stimulated cells produced increased molar quantities of cholera toxin (CT) both extra- and intracellularly. The intracellular CT was released in comparable amounts by sonication, deoxycholate extraction, and polymyxin B treatment. Polymyxin B release of CT was nearly complete under conditions wherein only 6% of total cellular beta-galactosidase was released, implying a periplasmic pool of CT in stimulated cells. No intracellular choleragenoid (CT subunit B) was found in stimulated cells by polymyxin B release. No proteolysis of 14C-labeled CT was detected after prolonged incubation with sonicated nonstimulated cultures or sonicated concentrated cells. These data support the conclusion that the stimulatory effect of lincomycin involves an increase in the rate of synthesis of the CT molecule, and argue against alternative models involving inhibition of putative normal degradation of CT, increased release of otherwise cell-bound CT, or activation of inactive, or less active, forms of CT.

Additivity of action between polysorbate 80 and polymyxin B towards spheroplasts of Pseudomonas aeruginosa NCTC 6750

When polymyxin B and polysorbate 80 were used together against spheroplasts of Pseudomonas aeruginosa, the activities were found to be additive. These substances have previously been reported to act synergistically against P. aeruginosa, but little or no intrinsic activity towards intact cells has been attributed to polysorbate 80. We suggest that in addition to enhancing polymyxin B penetration to the cytoplasmic membrane, polysorbate 80 may also act as an antimicrobial agent when polymyxin-induced damage to the outer membrane facilitates the surfactant's passage through the cell envelope.

Inhibition of Escherichia coli growth and respiration by polymyxin B covalently attached to agarose beads

Polymyxin B was attached to agarose beads by stable covalent bonds and the antimicrobial activity of the immobilized peptide was examined. Polymyxin-agarose inhibited the growth of Escherichia coli and Pseudomonas aeruginosa, but not Bacillus subtilis. In addition, the respiration of E. coli, E. coli spheroplasts, and B. subtilis protoplasts was inhibited by immobilized polymyxin, whereas the respiration of B. subtilis was unaffected by polymyxin-agarose. The activity of polymyxin-agarose was not due to the release of free peptide from the derivative. These data indicate that polymyxin can inhibit the growth and respiration of gram-negative bacteria by interacting with the outer surface of these cells. It is proposed that perturbation of outer membrane structure by polymyxin-agarose indirectly affected the selective permeability of the inner membrane and inhibited respiration. The results of this study emphasize the importance of outer membrane structural integrity for the normal functions of gram-negative bacteria.

Aminopeptidase N from Escherichia coli. Unusual interactions with the cell surface

The subcellular localization of aminopeptidase N (previously called aminoendopeptidase) has been investigated. This enzyme was found to be partially released (30-40%) by osmotic shock or by converting Escherichia coli K10 cells to spheroplasts. However, in all other E. coli strains (K12, B/r, MRE 600, ML 308) tested, this enzyme is not released at all by these procedures and thus behaves like a cytoplasmic enzyme. The crypticity of aminopeptidase N is surprisingly low, 75-85% of the enzyme activity is directly assayable in intact cells of any E. coli strain. Various inhibitors of transport systems do not interfer with this assay. Aminopeptidase activity could also be assayed in spheroplasts, even when an insolubilized substrate was used, which suggests a surface location of this enzyme. As well, N-ethylmaleimide (0.4 mM), under conditions which do not allow penetration in the cytoplasm, caused 70% inhibition of aminopeptidase N. Binding of 125I-labeled antiaminopeptidase N antibody to spheroplasts (from K12 strain) was used to assay the orientation of aminopeptidase N in the membrane. This enzyme is exposed on the outer surface of the cytoplasmic membrane. Confirmation of this orientation was obtained by comparing the accessibility of aminopeptidase, alkaline phosphatase and beta-galactosidase to fluorescamine in intact cells. Only 16% of the total beta-galactosidase was labeled with this fluorescent reagent whereas 44-45% of the aminopeptidase N and 59% of the alkaline phosphatase were labeled. Electron microscopic visualization of insolubilized reaction products of aminopeptidase N within the cells showed that these products are located at the poles of the cells. Neither mutant cells which were devoid of aminopeptidase N activity nor parental strains with the enzyme activity inhibited with phenylmercuric chloride contained the characteristic black caps. Thus, it appears that the periplasm is enlarged at the poles of the cells and that the reaction product is mainly located in these places. Investigation of the type of interactions of aminopeptidase N with the plasma membrane only revealed that aminopeptidase N has mainly an electrostatic interaction with the outer surface, probably mediated by magnesium ion bridges. Additional interactions are involved since disruption of the integrity of the cytoplasmic membrane is required to totally release this enzyme.

Use of polymyxin B, levallorphan, and tetracaine to isolate novel envelope mutants of Escherichia coli

Mutants of Escherichia coli were isolated by their resistance to the bacteriocidal effects of the membrane-active drugs polymyxin B, levallorphan, and tetracaine. The mutants were examined for additional changes in cellular physiology evoked by the lesions; many polymyxin-resistant strains had a concomitant increased sensitivity to anionic detergents, and several strains of each type had concomitant alterations in generation time and morphology. Mutants of each class (polymyxin resistant, tetracaine resistant, and levallorphan resistant) were transduced into recipient strains. The levallorphan resistance site (lev) was located at approximately 9 min on the E. coli chromosome. Polymyxin (pmx) and tetracaine (tec) resistance loci were also transduced. The lev and tec strains had a slight prolongation of generation time, in contrast with their isogenic wild-type strains. The tec transductant produced long filaments in the absence of tetracaine and had an altered colonial morphology, it reverted at high frequency, with the morphological abnormalities reverting along with the tetracaine resistance. The pmx transductant had an increased sensitivity to levallorphan and to anionic detergents. In contrast, both lev and tec mutants were more resistant to acriflavine than was the wild type or the pmx transductant. The pmx, lev, and tec loci differed in sensitivity to mitomycin C; the lev strain was more resistant, the tec strain was more sensitive, and the pmx strain was much more sensitive than the wild type. There was no difference in sensitivity to several other dyes and detergents, colicins, or T bacteriophage between the transductant and isogenic wild-type strains. Thus, lev, tec, and pmx loci confer more subtle alterations in the permeability barrier than do lipopolysaccharide-deficient mutants previously studied.

Action of polymyxin B on bacterial membranes. Binding capacities for polymyxin B of inner and outer membranes isolated from Salmonella typhimurium G30

Radioactive mono-N-acetyl-14C-polymyxin B or natural polymyxin B are within 60 s absorbed by isolated inner (cytoplasmic) and outer membranes from Salmonella typhimuriumG30. The sigmoidal binding isotherms indicate saturation of inner and outer membranes with approximately 30 and 60 nmoles polymyxin B bound per mg membrane, respectively. Based on the known content of these membranes in lipopolysaccharide, phosphatidylglycerol, cardiolipin and phosphatidylethanolamine, a calculation of the theoretical binding capacities yields almost identical values if lipopolysaccharide, phosphatidylglycerol and cardiolipin are assumed to function as the actual binding sites for the antibiotic in the isolated membranes. The excellent agreement between theoretical evaluation and experimental determination of polymyxin B-binding capacities leaves little doubt that the named anionic compounds are the chemoreceptors for the cationic antibiotic. This is further substantiated by very similar binding and killing kinetics of polymyxin B.

Control of the synthesis of alkaline phosphatase and the phosphate-binding protein in Escherichia coli

Using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and immunological techniques, we have compared the synthesis of the phoA protein (alkaline phosphatase) and the phoS protein (phosphate-binding protein) in response to the level of phosphate in the medium in different genetic backgrounds containing the known alkaline phosphatase control mutations. Both proteins are produced in excess phosphate media in a phoR1a- strain, whereas neither protein is produced in a phoB- strain even under derepression conditions. In four different phoR1c- strains, however, the phoA product cannot be detected in extracts of cells obtained from any growth condition, whereas the phoS product is produced in both excess and limiting phosphate media. It is not yet known if phoR1c- mutants are a special class of mutations within the phoB gene or whether they occur in a separate cistron involved in alkaline phosphatase regulation. From these results we conclude that the expression of the phoA gene is not always co-regulated with expression of the phoS gene product. We have determined that the phoS protein is a component of periplasmic protein band P4 described by Morris et al. (1974). The phoS product lacks sulfur-containing amino acids and is extractable by treatment with polymyxin sulfate. The other component of band P4 contains methionine and/or cysteine and is not extracted by polymyxin sulfate treatment. Like the phoS and phoA proteins, its synthesis is sensitive to the concentration of phosphate in the growth medium. In addition, the existence of a new class of periplasmic proteins synthesized at maximum rate in high phosphate media is demonstrated.

Activity of three murein hydrolases during the cell division cycle of Escherichia coli K-12 as measured in toluene-treated cells

The specific activities of three murein hydrolases, carboxypeptidase I, carboxypeptidase II, and amidase were studied with respect to cell division in toluene-treated cells of Escherichia coli K-12. Carboxypeptidase I and amidase activities were constant throughout the division cycle in cells of D11/lac+pro+. Detectable carboxypeptidase II activity varied and was highest at the time of division by a factor of three. Carboxypeptidase II specific activity was also correlated with cell division in BUG 6, a temperature-sensitive mutant (J.N Reeve, D.J. Groves, and D.J. Clark, 1970). Fifteen minutes after shifting BUG 6 from 42 C (nondividing conditions) to 32 C (dividing conditions), there was a rapid resumption of cell division, accompanied by a 10-fold increase in the specific activity of carboxypeptidase II. These results demonstrate a correlation between detectable carboxypeptidase II activity and cell division as reflected by activity in toluene-treated cells. The subcellular location of carboxypeptidase II, a soluble enzyme was found to be periplasmic since it was released by tris(hydroxymethyl)-aminomethane-ethylenediaminetetraacetate treatment and osmotic shock, two methods known to release periplasmic enzymes.

Stimulation by organic solvents and detergents of conversion of L-sorbose to L-sorbosone by Gluconobacter melanogenus IFO 3293

Treatment of Gluconobacter melanogenus IFO 3293 cells with benzene, carbon tetrachloride, cyclohexane, deoxycholate, toluene, or xylene stimulated their conversion of L-sorbose to L-sorbosone two- to threefold. The degree of stimulation depended upon the length of exposure time to the agent and the age of the G. melanogenus cells. A rapid decrease in viability of the cells and degradation of cell RNA was noted after treatment with the effective agents. The G. melanogenus cells were unable to absorb L-sorbose actively after toluene treatment.