Demonstration of ferric L-parabactin-binding activity in the outer membrane of Paracoccus denitrificans.

Under low-iron conditions, Paracoccus denitrificans excretes a catecholamine siderophore, L-parabactin, to sequester and utilize iron. In this report, we demonstrate the presence of stereospecific high-affinity ferric L-parabactin-binding activity associated with P. denitrificans membranes grown in low-iron medium. Isolated outer membrane components were shown to be three to four times higher in specific activity for ferric L-parabactin. The same amount of binding activity existed whether or not the radiolabel was present in the metal (55Fe) or the ligand (3H) portion of ferric parabactin chelate, suggesting that binding was to the intact complex. Ion-exchange chromatography of a Triton X-100-solubilized outer membrane mixture on DEAE-cellulose resulted in a 10-fold increase in binding activity relative to that present in whole membranes. Polypeptide profiles by sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the products of each stage of the purification showed that binding activity copurified with one or more of the low-iron-induced outer membrane proteins in the 80-kilodalton (kDa) region. Membrane proteins and [55Fe]ferric L-parabactin electrophoresed in nondenaturing gels demonstrated the presence of membrane component(s) which stereo-specifically bound ferric L-parabactin, thus providing independent confirmation of the binding assay results. Moreover, when the band labeled by [55Fe]ferric L-parabactin was excised and profiled by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, 80-kDa polypeptides were the major components present. These results demonstrate the presence of a high-affinity ferric L-parabactin receptor in P. denitrificans membranes and suggest that one or more of the 80-kDa low-iron-induced polypeptides are components of the ferric L-parabactin receptor.     

Iron uptake with ferripyochelin and ferric citrate by Pseudomonas aeruginosa.

Pyochelin is an iron-binding compound produced by Pseudomonas aeruginosa and demonstrates siderophore activity by its involvement in iron transport. During the transport process, an energy-independent association of [55Fe]ferripyochelin with bacteria occurred within the initial 30 s of reaction, followed by an energy-dependent accumulation of iron. The energy-independent association with iron appeared to be at the surface of the bacteria because the iron could be washed from the cells with thioglycolate, whereas accumulated iron was not washed from the bacteria. Energy-independent association of iron with bacteria and energy-dependent accumulation of iron in the presence of ferripyochelin varied concomitantly in cells grown under various conditions, but pyochelin synthesis appeared to be controlled separately. 55Fe complexed with citrate was also taken up by P. aeruginosa with a lower level of initial cell association. Bacterial mechanisms for iron uptake from ferric citrate were present in cells grown in a variety of media and were in lowest levels in cells grown in citrate. The synthesis of bacterial components for iron uptake from ferric citrate and from ferripyochelin was inhibited by high concentrations of iron supplied in growth media.     

Recognition and transport of ferric enterobactin in Escherichia coli.

The specificity of the outer membrane protein receptor for ferric enterobactin transport in Escherichia coli and the mechanism of enterobactin-mediated transport of ferric ions across the outer membrane have been studied. Transport kinetic and inhibition studies with ferric enterobactin and synthetic structural analogs have mapped the parts of the molecule important for receptor binding. The ferric complex of the synthetic structural analog of enterobactin, 1,3,5-N,N',N'-tris-(2,3-dihydroxybenzoyl)triaminomethylbenzene (MECAM), was transported with the same maximum velocity as was ferric enterobactin. A double-label transport assay with [59Fe, 3H]MECAM showed that the ligand and the metal are transported across the outer membrane at an identical rate. Under the growth conditions used, large fractions of the transported complexes were available for exchange across the outer membrane when a large excess of extracellular complex was added to the cell suspension; at least 60% of the internalized [59Fe]enterobactin exchanged with extracellular [55Fe]enterobactin. Internalized [59Fe, 3H]MECAM was released from the cell as the intact complex when either unlabeled Fe-MECAM or Fe-enterobactin was added extracellularly. The results suggest a mechanism of active transport of unmodified coordination complex across the outer membrane with possible accumulation in the periplasm.     

Demonstration of a Ferric Vibrioferrin-Binding Protein in the Outer Membrane of Vibrio parahaemolyticus

Under iron-restricted conditions, Vibrio parahaemolyticus produces a siderophore, vibrioferrin, accompanying expression of two major outer membrane proteins of 78 and 83 kDa. Autoradiographic analysis of nondenaturing polyacrylamide gel electrophoregrams of outer membrane preparations previously incubated with [⁵⁵Fe]ferric vibrioferrin revealed a single radiolabeled band, in which the 78-kDa protein was detected predominantly by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The antiserum against the purified 78-kDa protein partially inhibited Fe-VF binding to isolated OMPs. The 78-kDa protein was cleaved by the treatment of whole cells with proteinase K, indicating that a portion of this protein is exposed on the surface of the outer membrane. The treated cells lost most of their iron uptake activity mediated by vibrioferrin. These results suggest that the ferric vibrioferrin-binding protein of 78 kDa may function as the receptor for ferric vibrioferrin involved in the initial step of vibrioferrin-mediated iron uptake. Immunoblot analysis using the antiserum against the 78-kDa protein demonstrated that the molecular mass and antigenic properties of the protein were highly conserved among V. parahaemolyticus strains examined. The antiserum also recognized an iron-repressible outer membrane protein of 78 kDa from iron-restricted V. alginolyticus strains, some of which appeared to produce vibrioferrin.     

Isolation of a membrane associated iron chelator from Pseudomonas aeruginosa

A membrane associated iron chelator (MAIC) has been extracted with ethanol from the membranes of Pseudomonas aeruginosa, and isolated on thin-layer chromatograms. Also extracted from the membranes is the ferrated form of MAIC, FeMAIC. When cell-bound or in the complete ethanol extract of membranes, MAIC binds iron from exogenous iron sources forming FeMAIC. Methanol solutions of each compound exhibit similar absorption spectra with strong absorption in the ultraviolet, indicating the aromatic structure of the compounds. Colorimetric reactions reveal the presence of a phenolic moiety in these compounds. MAIC and FeMAIC are extracted from the membranes of cells grown in media supplemented with iron or in media containing significant trace levels of iron. Transport studies revealed that neither iron-fed nor iron-starved cells transport detectable levels of radiolabeled iron from exogenous iron sources, yet low amounts of 55FeMAIC are extracted from the membranes of cells incubated with [55Fe]ferric chelators. The MAIC may serve as an iron transporter in these cells, or may serve to bind iron following its transport into the cell via another mechanism.     

Iron uptake from ferric citrate by Vibrio anguillarum

We report here that Vibrio anguillarum possesses a non-inducible active transport system which can efficiently supply iron to the cell from ferric citrate, independently of the siderophore-based mechanisms. The strains tested were able to grow in CM9 medium in iron-restricted conditions when ferric citrate was present in the medium. Moreover, the presence of ferric citrate inhibited the production of siderophores in the strains tested. V. anguillarum cells and isolated membranes could incorporate ⁵⁵Fe³⁺ complexed by citrate, without a difference between cells grown in the presence or absence of ferric citrate. The presence of 2,4-dinitrophenol, ferrozine, ferricyanide, trypsin, as well as low temperature produced a marked decrease or total inhibition of ⁵⁵Fe³⁺ uptake by the cells. All these results suggest that iron uptake from ferric citrate in V. anguillarum must be an energy-dependent process not induced by the presence of iron or citrate in the medium, mediated by a membrane protein(s), which may require an iron reduction step to function.     

Iron (III) hydroxamate transport into Escherichia coli. Substrate binding to the periplasmic FhuD protein

Due to its extreme insolubility, Fe3+ is not transported as a monoatomic ion. In microbes, iron is bound to low molecular weight carriers, designated siderophores. For uptake into cells of Escherichia coli Fe3+ siderophores have to be translocated across two membranes. Transport across the outer membrane is receptor-dependent and energy-coupled; transport across the cytoplasmic membrane seems to follow a periplasmic binding protein-dependent transport mechanism. In support of this notion we demonstrate specific binding of the Fe3+ hydroxamate compounds ferrichrome, aerobactin, and coprogen, which are transported via the Fhu system, to the periplasmic FhuD protein, and no binding of the transport inactive ferrichrome A, ferric citrate, and iron sulfate. About 10(4) ferrichrome molecules were bound to the FhuD protein of cells which overproduced plasmid-encoded FhuD. Binding depended on transport across the outer membrane mediated by the FhuA receptor and the TonB protein. Binding to FhuD was supported by the exclusive resistance of FhuD to proteinase K in the presence of the transport active hydroxamates. The overproduced precursor form of the FhuD protein was not protected by the Fe3+ hydroxamates indicating a conformation different to the mature form. The FhuD protein apparently serves as a periplasmic carrier for Fe3+ hydroxamates with widely different structures.     

Investigations of iron uptake in Halobacterium salinarum

The iron transport in the extremely halophilic Euryarchaeon Halobacterium salinarum JW5 was investigated. Experiments to detect endogenous siderophores from H. salinarum failed, but it was able to utilize exogenous siderophores. Measurement of the uptake of (55)Fe and [(14)C]citrate gave evidence only for the accumulation of iron. Two additional membrane proteins could be detected in iron-starved cells, one in iron-repleted membranes and one that is up-regulated there. Respiratory rates of iron-starved membranes after the addition of succinate and NADH differed considerably from iron-repleted ones. Furthermore, both types of membrane exhibited different degrees of inhibition by cyanide.     

Hemin uptake in Porphyromonas gingivalis: Omp26 is a hemin-binding surface protein.

A 26-kDa outer membrane protein (Omp26) has been proposed to play a role in hemin acquisition by Porphyromonas gingivalis (T. E. Bramanti and S. C. Holt, J. Bacteriol. 174:5827-5839, 1992). We studied [55Fe]hemin uptake in P. gingivalis grown under conditions of hemin starvation (Omp26 expressed on the outer membrane surface) and hemin excess (Omp26 not expressed on surface). [55Fe]hemin uptake occurred rapidly in hemin-starved cells which incorporated up to 70% of total [55Fe]hemin within 3 min. P. gingivalis grown under hemin-starved conditions or treated with the iron chelator 2,2'-bipyridyl to induce an iron stress took up six times more [55Fe]hemin than hemin-excess-grown cells. Polyclonal monospecific anti-Omp26 antibody added to hemin-starved cells inhibited [55Fe]hemin uptake by more than 50%, whereas preimmune serum had no effect. [55Fe]hemin uptake in hemin-starved P. gingivalis was inhibited (36 to 67%) in the presence of equimolar amounts of unlabeled hemin, protoporphyrin IX, zinz protoporphyrin, and Congo red dye but was not inhibited in the presence of non-hemin-containing iron sources. Heat shock treatment (45 degrees C) of hemin-excess-grown P. gingivalis (which cases translocation of Omp26 to the surface) increased [55Fe]hemin uptake by threefold after 3 min in comparison with cells grown at 37 degrees C. However, no [55Fe] hemin uptake beyond 3 min was observed in either hemin-excess-grown or hemin-starved cells exposed to heat shock. In experiments using heterobifunctional cross-linker analysis, hemin and selected porphyrins were cross-linked to Omp26 in hemin-starved P. gingivalis, but no cross-linking was seen with hemin-excess-grown cells. However, cross-linking of hemin to Omp26 was observed after heat shock treatment of hemin-excess-grown cells. Finally, anti-Omp26 antibody inhibited cross-linked of hemin to Omp26. These findings indicate that hemin binding and transport into P.gingivalis cell mediated by Omp26.     

Iron supply of Escherichia coli with polymer-bound ferricrocin.

Uptake of ferric iron from ferricrocin was studied in Escherichia coli using a polymer-coupled ferricrocin that was unable to penetrate into the cell. Ferricrocinyl polyethylene glycol succinate (Mr 7000 -- 8500) promoted growth of E. coli K-12 AB2847 aroB under iron-limiting conditions. In iron-starved cells, uptake of 55Fe could be demonstrated; the amount of iron accumulated amounted to 10% of that observed with free ferricrocin. The iron supply by ferricrocin bound to polyethylene glycol was strictly dependent upon the functions expressed by the tonA and the tonB genes, as was the iron uptake promoted by free ferricrocin. Polymer-bound ferricrocin protected cells against colicin M and phage T5 by competition for the common tonA-coded outer membrane receptor protein. In addition, the rate of iron transport via the negatively charged ferricrocinyl succinate was as fast as via the neutral ferricrocin molecule. No ligand was found associated with the cells. Penetration of chelator beyond receptor is not necessary for siderophore-mediated iron uptake. It is concluded that sufficient amounts of iron can be released from the polymer complex to satisfy growth requirements.     

Identification of the ferrioxamine B receptor,FoxB, inEscherichia coli K12

The photoreactivep-azidobenzoyl analog of ferrioxamine B was used to show that ferrioxamine-B-mediated iron transport is separate and distinct from coprogen-mediated iron transport inEscherichia coli. Photolysis of this analog inhibited uptake of [⁵⁹Fe]ferrioxamine B but not [⁵⁹Fe]coprogen or [⁵⁹Fe]ferrichrome. Conversely, photolysis of thep-azidobenzoyl analog of coprogen B inhibited uptake of [⁵⁹Fe]coprogen but not [⁵⁹Fe]ferrioxamine B or [⁵⁹Fe]ferrichrome. Photolabeling of outer membranes withp-azidobenzoyl-[⁵⁹Fe]ferrioxamine B resulted in the labeling of two iron-regulated peptides with molecular masses of about 66 and 26 kDa. Expression of these peptides was increased when ferrioxamine B was the sole iron source. Both peptides were present in outer membrane preparations of thefhuF mutant H1717, but the 66 kDa peptide was not inducible. These results are evidence for an outer membrane receptor inE. coli unique for linear ferrioxamines.     

Isolation of a ferritin from Bacteroides fragilis

A ferritin was isolated from the obligate anaerobe Bacteroides fragilis. Estimated molecular masses were 400 kDa for the holomer and 16.7 kDa for the subunits. A 30-residue N-terminal amino acid sequence was determined and found to resemble the sequences of other ferritins (human H-chain ferritin, 43% identity; Escherichia coli gen-165 product, 37% identity) and to a lesser degree, bacterioferritins (E. coli bacterioferritin, 20% identity). The protein stained positively for iron, and incorporated 59Fe when B. fragilis was grown in the presence of [59Fe]citrate. However, the isolated protein contained only about three iron atoms per molecule, and contained no detectable haem. This represents the first isolation of a ferritin protein from bacteria. It may alleviate iron toxicity in the presence of oxygen.     

Localization of cytochromes in the outer membrane of Desulfovibrio vulgaris (Hildenborough) and their role in anaerobic biocorrosion

A protocol was developed whereby the outer and cytoplasmic membranes of the sulfate-reducing bacterium Desulfovibrio vulgaris (Hildenborough) were isolated and partially characterized. The isolated outer membrane fractions from cultures grown under high (100 ppm) and low (5 ppm) Fe2+ conditions were compared by SDS-PAGE electrophoresis, and showed that several protein bands were derepressed under the low iron conditions, most notably at 50 kDa, and 77.5 kDa. Outer membrane isolated from low iron cultured cells was found to contain two proteins, 77.5 kDa and 62.5 kDa in size, that reacted with a heme-specific stain and were referred to as high molecular weight cytochromes. Studies conducted on the low iron isolated outer membrane by a phosphate/mild steel hydrogen evolution system showed that addition of the membrane fraction caused an immediate acceleration in H2 production. A new model for the anaerobic biocorrosion of mild steel is proposed.     

Ferric citrate transport in Escherichia coli requires outer membrane receptor protein fecA.

Mutants of Escherichia coli K-12 AB2847 and of E. coli K-12 AN92 were isolated which were unable to grow on ferric citrate as the sole iron source. Of 22 mutants, 6 lacked an outer membrane protein, designated FecA protein, which was expressed by growing cells in the presence of 1 mM citrate. Outer membranes showed an enhanced binding of radioactive iron, supplied as a citrate complex, depending on the amount of FecA protein. The FecA protein was the most resistant of the proteins involved in ferric irion iron translocation across the outer membrane (FhuA = TonA, FepA, Cir, or 83K proteins) to the action of pronase P. It is also shown that previously isolated fec mutants (G. C. Woodrow et al., J. Bacteriol. 133:1524-1526, 1978) which are cotransducible with argF all lack the FecA protein. They were termed fecA to distinguish them from the other ferric citrate transport mutants, now designated fecB, which mapped in the same gene region at 7 min but were not cotransducible with ArgF. E. coli W83-24 and Salmonella typhimurium, which are devoid of a citrate-dependent iron transport system, lacked the FecA protein. It is proposed that the FecA protein participates in the transport of ferric citrate.     

Regulation of HeLa cell transferrin receptors

HeLa cells were found to have a single class of non-interacting receptors specific for transferrin. Both apotransferrin and diferric transferrin competed equally with 125I-diferric transferrin for receptor binding. Transferrin binding was temperature-dependent and reversible. Binding of transferrin to cells exhibited a KD of 27 nM with a maximum binding capacity of 1.8-3.7 x 10(6) molecules/cell. Cells grown in the presence of diferric transferrin or in the presence of ferric ammonium citrate exhibited a concentration- and time-dependent decrease in 125I-diferric transferrin binding. The decrease in binding activity reflected a reduction in receptor number rather than an alteration in ligand receptor affinity. Growth of cells in saturating concentrations of apotransferrin did not cause a decrease in receptor number. When iron-treated cells were removed to media free of ferric ammonium citrate, the receptor number returned to control values by 40 h. When receptors were removed with trypsin, cells grown and maintained in ferric ammonium citrate-supplemented media demonstrated a rate of receptor reappearance 47% that of control cells grown in ferric ammonium citrate-free media. Cells grown in media supplemented with diferric transferrin or ferric ammonium citrate exhibited an increase in cytosolic iron content. The transferrin receptor number returned to normal after cells were removed to unsupplemented media, despite persistent elevation of cytosolic iron content. Increased iron content did not appear to be the sole factor determining receptor number.     

Utilization of enterobactin and other exogenous iron sources by Haemophilus influenzae, H. parainfluenzae and H. paraphrophilus

The ability of Haemophilus influenzae, H. parainfluenzae and H. paraphrophilus to utilize iron complexes, iron-proteins and exogenous microbial siderophores was evaluated. In a plate bioassay, all three species used not only ferric nitrate but also the iron chelates ferric citrate, ferric nitrilotriacetate and ferric 2,3-dihydroxybenzoate. Each Haemophilus species examined also used haemin, haemoglobin and haem-albumin as iron sources although only H. influenzae could acquire iron from transferrin or from haemoglobin complexed with haptoglobin. None of the haemophili obtained iron from ferritin or lactoferrin or from the microbial siderophores aerobactin or desferrioxamine B. However, the phenolate siderophore enterobactin supplied iron to both H. parainfluenzae and H. paraphrophilus, and DNA isolated from both organisms hybridized with a DNA probe prepared from the Escherichia coli ferric enterobactin receptor gene fepA. In addition, a monospecific polyclonal antiserum raised against the E. coli 81 kDa ferric enterobactin receptor (FepA) recognized an iron-repressible outer membrane protein (OMP) in H. parainfluenzae of between 80 and 82 kDa (depending on the strain). This anti-FepA serum did not cross-react with any of the OMPs of H. paraphrophilus or H. influenzae. The OMPs of each Haemophilus species were also probed with antisera raised against the 74 kDa Cir or 74 kDa IutA (aerobactin receptor) proteins of E. coli. Apart from one H. parainfluenzae strain (NCTC 10665), in which an OMP of about 80 kDa cross-reacted with the anti-IutA sera, no cross-reactivity was observed between Cir, IutA and the OMPs of H. influenzae, H. parainfluenzae or H. paraphrophilus.(ABSTRACT TRUNCATED AT 250 WORDS)     

Escherichia coli iron enterobactin uptake monitored by Mössbauer spectroscopy.

Iron uptake by Escherichia coli under aerobic conditions of iron deficiency is mediated by a highly stable ferric enterobactin [Fe(ent)3-] siderophore complex. M?ssbauer spectroscopy has been used to monitor the fate of the iron as 57Fe(ent) was taken up by the cells. Osmotic shock experiments were used to distinguish between the iron present in the periplasmic space and that in the cytoplasm of the cell. Iron delivery by a synthetic analog of enterobactin, 1,3,5-N,N',N'- tris-(2,3-dihydroxybenzoyl)triaminomethylbenzene (MECAM), was also studied. Although Fe-MECAM was transported at the same rate as was Fe(ent) across the outer membrane and was apparently accumulated in the periplasmic space, the subsequent behaviors of Fe(ent) and Fe-MECAM were very different. After more than 30 min, a major fraction of the iron originally absorbed as ferric enterobactin appeared as Fe(II), apparently in the cytoplasm of the cell. However, little iron was delivered to the cytoplasm by the MECAM complex. The differences in specificity of these two stages of iron uptake by E. coli are discussed.     

Citrate utilization by Escherichia coli: plasmid- and chromosome-encoded systems

Citrate utilization plasmids have previously been identified in atypical Escherichia coli isolates. A different citrate-utilizing (Cit+) variant of E. coli K-12 arose as a consequence of two chromosomal mutations (B. G. Hall, J. Bacteriol. 151:269-273, 1982). The processes controlling the transport of citrate in both a Cit+ chromosomal mutant and a Cit+ plasmid system were studied. Both systems were found to be inducible in growth experiments. In transport assays with whole cells, citrate-grown cells accumulated [1,5-14C]citrate at two to three times the rate of uninduced cells. Only the Vmax was affected by induction, and the Km for whole cells remained at 67 microM citrate for the chromosomal strain and 120 microM citrate for the plasmid-conferred system. There was no detectable accumulation of radioactivity with [6-14C]citrate, because of rapid metabolism and the release of 14CO2. Energy-dependent citrate transport was found with membrane vesicles obtained from both the chromosome-conferred and the plasmid Cit+ systems. The vesicle systems were inhibited by valinomycin and carbonyl cyanide m-chloro-phenylhydrazone but not by nigericin and monensin. In contrast to whole cells, the vesicle systems were resistant to Hg2+ and showed identical kinetics with [1,5-14C]citrate and [6-14C]citrate. H+ appeared to be important for citrate transport in whole cells and membranes. Monovalent cations such as Na+ and K+, divalent cations such as Mg2+ and Mn2+, and anions such as PO4(3-), SO4(2-), and NO3- were not required. The two systems differed in inhibition by citrate analogs.