Spectroscopic Observations of Ferric Enterobactin Transport
2003; Elsevier BV; Volume: 278; Issue: 2 Linguagem: Inglês
10.1074/jbc.m210360200
ISSN1083-351X
AutoresZhenghua Cao, Paul Warfel, Salete M. Newton, Phillip E. Klebba,
Tópico(s)Bacterial Genetics and Biotechnology
ResumoWe characterized the uptake of ferric enterobactin (FeEnt), the native Escherichia coli ferric siderophore, through its cognate outer membrane receptor protein, FepA, using a site-directed fluorescence methodology. The experiments first defined locations in FepA that were accessible to covalent modification with fluorescein maleimide (FM) in vivo; among 10 sites that we tested by substituting single Cys residues, FM labeled W101C, S271C, F329C, and S397C, and all these exist within surface-exposed loops of the outer membrane protein. FeEnt normally adsorbed to the fluoresceinated S271C and S397C mutant FepA proteins in vivo, which we observed as quenching of fluorescence intensity, but the ferric siderophore did not bind to the FM-modified derivatives of W101C or F329C. These in vivo fluorescence determinations showed, for the first time, consistency with radioisotopic measurements of the affinity of the FeEnt-FepA interaction; K d was 0.2 nm by both methods. Analysis of the FepA mutants with AlexaFluor680, a fluorescein derivative with red-shifted absorption and emission spectra that do not overlap the absorbance spectrum of FeEnt, refuted the possibility that the fluorescence quenching resulted from resonance energy transfer. These and other data instead indicated that the quenching originated from changes in the environment of the fluor as a result of loop conformational changes during ligand binding and transport. We used the fluorescence system to monitor FeEnt uptake by live bacteria and determined its dependence on ligand concentration, temperature, pH, and carbon sources and its susceptibility to inhibition by the metabolic poisons. Unlike cyanocobalamin transport through the outer membrane, FeEnt uptake was sensitive to inhibitors of electron transport and phosphorylation, in addition to its sensitivity to proton motive force depletion. We characterized the uptake of ferric enterobactin (FeEnt), the native Escherichia coli ferric siderophore, through its cognate outer membrane receptor protein, FepA, using a site-directed fluorescence methodology. The experiments first defined locations in FepA that were accessible to covalent modification with fluorescein maleimide (FM) in vivo; among 10 sites that we tested by substituting single Cys residues, FM labeled W101C, S271C, F329C, and S397C, and all these exist within surface-exposed loops of the outer membrane protein. FeEnt normally adsorbed to the fluoresceinated S271C and S397C mutant FepA proteins in vivo, which we observed as quenching of fluorescence intensity, but the ferric siderophore did not bind to the FM-modified derivatives of W101C or F329C. These in vivo fluorescence determinations showed, for the first time, consistency with radioisotopic measurements of the affinity of the FeEnt-FepA interaction; K d was 0.2 nm by both methods. Analysis of the FepA mutants with AlexaFluor680, a fluorescein derivative with red-shifted absorption and emission spectra that do not overlap the absorbance spectrum of FeEnt, refuted the possibility that the fluorescence quenching resulted from resonance energy transfer. These and other data instead indicated that the quenching originated from changes in the environment of the fluor as a result of loop conformational changes during ligand binding and transport. We used the fluorescence system to monitor FeEnt uptake by live bacteria and determined its dependence on ligand concentration, temperature, pH, and carbon sources and its susceptibility to inhibition by the metabolic poisons. Unlike cyanocobalamin transport through the outer membrane, FeEnt uptake was sensitive to inhibitors of electron transport and phosphorylation, in addition to its sensitivity to proton motive force depletion. outer membrane ferric enterobactin fluorescein maleimide lipopolysaccharide iodoacetamide fluorescein Luria Bertani AlexaFluor680 maleimide third surface loop 4-morpholinepropanesulfonic acid carbonyl cyanidep-chlorophenylhydrazone 2,4-dinitrophenol proton motive force bovine serum albumin Gram-negative bacteria recognize and transport a variety of ferric siderophores (1Neilands J.B. Annu. Rev. Microbiol. 1982; 36: 285-309Crossref PubMed Scopus (332) Google Scholar, 2Neilands J.B. J. Biol. Chem. 1995; 270: 26723-26726Abstract Full Text Full Text PDF PubMed Scopus (1202) Google Scholar) through outer membrane (OM)1 receptor proteins that function as ligand-gated porins (LGP) (3Rutz J.M. Liu J. Lyons J.A. Goranson J. Armstrong S.K. McIntosh M.A. Feix J.B. Klebba P.E. Science. 1992; 258: 471-475Crossref PubMed Scopus (145) Google Scholar). Some of the hallmarks of these transport processes are the specificity with which LGP select their iron-containing ligands (4Wayne R. Neilands J.B. J. Bacteriol. 1975; 121: 497-503Crossref PubMed Google Scholar, 5Wayne R. Frick K. Neilands J.B. J. Bacteriol. 1976; 126: 7-12Crossref PubMed Google Scholar, 6Di Masi D.R. White J.C. Schnaitman C.A. Bradbeer C. J. Bacteriol. 1973; 115: 506-513Crossref PubMed Google Scholar, 7Scott D.C. Cao Z.Qi, Z. Bauler M. Igo J.D. Newton S.M. Klebba P.E. J. Biol. Chem. 2001; 276: 13025-13033Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar), the high affinity of the receptor-ligand interactions (8Fiss E.H Stanley-Samuelson P. Neilands J.B. Biochemistry. 1982; 31: 4517-4522Crossref Scopus (31) Google Scholar, 9Newton S.M. Igo J.D. Scott D.C. Klebba P.E. Mol. Microbiol. 1999; 32: 1153-1165Crossref PubMed Scopus (77) Google Scholar), conformational changes in the receptors during ligand binding (10Payne M.A. Igo J.D. Cao Z. Foster S.B. Newton S.M. Klebba P.E. J. Biol. Chem. 1997; 272: 21950-21955Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 11Cao Z., Qi, Z. Sprencel C. Newton S.M. Klebba P.E. Mol. Microbiol. 2000; 37: 1306-1317Crossref PubMed Scopus (40) Google Scholar, 12Scott D.C. Newton S.M. Klebba P.E. J. Bacteriol. 2002; 184: 4906-4911Crossref PubMed Scopus (30) Google Scholar, 13Ferguson A.D. Chakraborty R. Smith B.S. Esser L. van der Helm D. Deisenhofer J. Science. 2002; 295: 1715-1719Crossref PubMed Scopus (301) Google Scholar) and transport (14Jiang X. Payne M.A. Cao Z. Foster S.B. Feix J.B. Newton S.M. Klebba P.E. Science. 1997; 276: 1261-1264Crossref PubMed Scopus (91) Google Scholar), 2Z. Cao, S. M. Newton, and P. E. Klebba, submitted for publication.2Z. Cao, S. M. Newton, and P. E. Klebba, submitted for publication. the requirement for the accessory proteins TonB (16Wang C.C. Newton A. J. Biol. Chem. 1971; 246: 2147-2151Abstract Full Text PDF PubMed Google Scholar), ExbB, and ExbD (17Guterman S.K. J. Bacteriol. 1973; 114: 1217-1224Crossref PubMed Google Scholar, 18Guterman S.K. Dann L. J. Bacteriol. 1973; 114: 1225-1230Crossref PubMed Google Scholar), and the need for cellular energy to accomplish active transport of the metal-containing ligands through the OM (19Wang C.C. Newton A. J. Bacteriol. 1969; 98: 1142-1150Crossref PubMed Google Scholar, 20Pugsley A.P. Reeves P. J. Bacteriol. 1976; 127: 218-228Crossref PubMed Google Scholar, 21Reynolds P.R. Mottur G.P. Bradbeer C. J. Biol. Chem. 1980; 255: 4313-4319Abstract Full Text PDF PubMed Google Scholar, 22Bradbeer C. J. Bacteriol. 1993; 175: 3146-3150Crossref PubMed Scopus (147) Google Scholar). This latter energy requirement is atypical in a bilayer that contains open channels (23Nikaido H. Vaara M. Microbiol. Rev. 1985; 49: 1-32Crossref PubMed Google Scholar, 24Nikaido H. Rosenberg E.Y. J. Gen. Physiol. 1981; 77: 121-135Crossref PubMed Scopus (158) Google Scholar, 25Nikaido H. Rosenberg E.Y. J. Bacteriol. 1983; 153: 241-252Crossref PubMed Google Scholar, 26Luckey M. Nikaido H. Proc. Natl. Acad. Sci. U. S. A. 1980; 77: 167-171Crossref PubMed Scopus (183) Google Scholar) and therefore cannot sustain an ion gradient. At present the pathways through which both energy and TonB participate in metal transport are unresolved (27Postle K. J. Bioenerg. Biomembr. 1993; 25: 591PubMed Google Scholar, 28Letain T.E. Postle K Mol. Microbiol. 1997; 24: 271Crossref PubMed Scopus (132) Google Scholar, 29Klebba P.E. Newton S.M. Curr. Opin. Microbiol. 1998; 1: 238-247Crossref PubMed Scopus (81) Google Scholar). The crystal structures of FepA (30Buchanan S.K. Smith B.S. Venkatramani L. Xia D. Esser L. Palnitkar M. Chakraborty R. van der Helm D. Deisenhofer J. Nat. Struct. Biol. 1999; 6: 56-63Crossref PubMed Scopus (486) Google Scholar), FhuA (31Locher K.P. Rees B. Koebnik R. Mitschler A. Moulinier L. Rosenbusch J.P. Moras D. Cell. 1998; 95: 771-778Abstract Full Text Full Text PDF PubMed Scopus (453) Google Scholar, 32Ferguson A.D. Hofmann E. Coulton J.W. Diederichs K. Welte W. Science. 1998; 282: 2215-2220Crossref PubMed Scopus (661) Google Scholar), and FecA (13Ferguson A.D. Chakraborty R. Smith B.S. Esser L. van der Helm D. Deisenhofer J. Science. 2002; 295: 1715-1719Crossref PubMed Scopus (301) Google Scholar) revealed that their N-terminal 150 amino acids fold into a 4-stranded β-sheet domain that lodges within otherwise typical transmembrane β-barrels, creating a third mechanistic paradox: how do ligands pass through such closed pores? Site-directed biophysical methodologies defined many of the biochemical properties of FepA (10Payne M.A. Igo J.D. Cao Z. Foster S.B. Newton S.M. Klebba P.E. J. Biol. Chem. 1997; 272: 21950-21955Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 11Cao Z., Qi, Z. Sprencel C. Newton S.M. Klebba P.E. Mol. Microbiol. 2000; 37: 1306-1317Crossref PubMed Scopus (40) Google Scholar, 14Jiang X. Payne M.A. Cao Z. Foster S.B. Feix J.B. Newton S.M. Klebba P.E. Science. 1997; 276: 1261-1264Crossref PubMed Scopus (91) Google Scholar, 33Liu J. Rutz J.M. Klebba P.E. Feix J.B. Biochemistry. 1994; 33: 13274-13283Crossref PubMed Scopus (46) Google Scholar) by derivatization of the genetically engineered mutant FepAE280C with fluorescent or paramagnetic labels. Residue E280C exists on the external surface of the third surface loop (L3) (30Buchanan S.K. Smith B.S. Venkatramani L. Xia D. Esser L. Palnitkar M. Chakraborty R. van der Helm D. Deisenhofer J. Nat. Struct. Biol. 1999; 6: 56-63Crossref PubMed Scopus (486) Google Scholar, 33Liu J. Rutz J.M. Klebba P.E. Feix J.B. Biochemistry. 1994; 33: 13274-13283Crossref PubMed Scopus (46) Google Scholar) of the receptor, and nitroxide spin labels attached to it reflected structural changes in FepA when FeEnt binds and different motion when it passes through the OM protein (14Jiang X. Payne M.A. Cao Z. Foster S.B. Feix J.B. Newton S.M. Klebba P.E. Science. 1997; 276: 1261-1264Crossref PubMed Scopus (91) Google Scholar). Similarly, the binding of either FeEnt or ColB quenched fluorescein labels attached to purified FepAE280C, which allowed determination of the thermodynamic and kinetic properties of the binding reaction (10Payne M.A. Igo J.D. Cao Z. Foster S.B. Newton S.M. Klebba P.E. J. Biol. Chem. 1997; 272: 21950-21955Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 11Cao Z., Qi, Z. Sprencel C. Newton S.M. Klebba P.E. Mol. Microbiol. 2000; 37: 1306-1317Crossref PubMed Scopus (40) Google Scholar). One of the conspicuous findings of those experiments was that purification reduced the affinity of FepA for FeEnt (9Newton S.M. Igo J.D. Scott D.C. Klebba P.E. Mol. Microbiol. 1999; 32: 1153-1165Crossref PubMed Scopus (77) Google Scholar, 10Payne M.A. Igo J.D. Cao Z. Foster S.B. Newton S.M. Klebba P.E. J. Biol. Chem. 1997; 272: 21950-21955Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar) and the affinity of FhuA for its ligand, ferrichrome, (35Locher K.P. Rosenbusch J.P. Eur. J. Biochem. 1997; 247: 770-775Crossref PubMed Scopus (49) Google Scholar, 7Scott D.C. Cao Z.Qi, Z. Bauler M. Igo J.D. Newton S.M. Klebba P.E. J. Biol. Chem. 2001; 276: 13025-13033Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar) at least 100-fold. In this report we applied fluorescence methodologies to FeEnt transport in vivo; the sensitivity and specificity of the technique provided not only an explanation for the discrepancies in affinity that were observed in vivo andin vitro but also more detailed information on the ligand internalization reaction through the OM, including its temperature, pH, and energy dependence. Escherichia coli K12 strains KDF541 (F−, thi, entA, pro, trp, rpsL, recA, fepA, fhuA, cir) (3Rutz J.M. Liu J. Lyons J.A. Goranson J. Armstrong S.K. McIntosh M.A. Feix J.B. Klebba P.E. Science. 1992; 258: 471-475Crossref PubMed Scopus (145) Google Scholar) and KDF571 (KDF541, but tonB) (3Rutz J.M. Liu J. Lyons J.A. Goranson J. Armstrong S.K. McIntosh M.A. Feix J.B. Klebba P.E. Science. 1992; 258: 471-475Crossref PubMed Scopus (145) Google Scholar) were hosts for fepA + pUC19 derivatives pITS449 (36Armstrong S.A Francis C.L. McIntosh M.A. J. Biol. Chem. 1990; 265: 14536-14575Abstract Full Text PDF PubMed Google Scholar) and pT944 (11Cao Z., Qi, Z. Sprencel C. Newton S.M. Klebba P.E. Mol. Microbiol. 2000; 37: 1306-1317Crossref PubMed Scopus (40) Google Scholar). Both plasmids encode wild type FepA, but pT944 contains a series of genetically engineered restriction sites that facilitate cloning procedures and restriction fragment exchange. pMF19 (provided by M. A. Valvano) (37Feldman M.F. Valvano M.A. J. Biol. Chem. 1999; 274: 35129-35138Abstract Full Text Full Text PDF PubMed Scopus (174) Google Scholar) is a pEXT21 derivative that carries wbbL, the structural gene for a rhamnosyltransferase involved in O16 LPS biosynthesis. Fluorescent reagents were purchased from Molecular Probes. Enterobactin was purified from E. coli strain AN102 (39Klebba P.E. McIntosh M.A. Neilands J.B. J. Bacteriol. 1982; 149: 880-888Crossref PubMed Google Scholar), and FeEnt was prepared and purified as previously described (40Pollack J.R. Neilands J.B. Biochem. Biophys. Res. Commun. 1970; 38: 989-992Crossref PubMed Scopus (302) Google Scholar). Its concentration was determined from absorbance at 495 nm (ε495nmmM = 5.6) (41Murphy C.K. Kalve V.I. Klebba P.E. J. Bacteriol. 1990; 172: 2736-2746Crossref PubMed Google Scholar). Rabbits were immunized weekly for one month with 5-IAF (Molecular Probes) conjugated to BSA and subsequently bled through the ear. Sera were titered by enzyme-linked immunosorbent assay (> 10,000) and Western immunoblot against ovalbumin-IAF conjugates. We used QuikChange (Stratagene) for site-directed mutagenesis (7Scott D.C. Cao Z.Qi, Z. Bauler M. Igo J.D. Newton S.M. Klebba P.E. J. Biol. Chem. 2001; 276: 13025-13033Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar, 12Scott D.C. Newton S.M. Klebba P.E. J. Bacteriol. 2002; 184: 4906-4911Crossref PubMed Scopus (30) Google Scholar) on pITS449 or pT944 and designated the mutant FepA proteins FepAXnY, where X represents the one-letter abbreviation for the wild type residue at position n and Y represents the one-letter abbreviation for the substituted amino acid. We tested the functionality of each FepA mutant by evaluating its expression (42Newton S.M. Allen J.S. Cao Z., Qi, Z. Jiang X. Sprencel C. Igo J.D. Foster S.B. Payne M.A. Klebba P.E. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 4560-4565Crossref PubMed Scopus (52) Google Scholar), susceptibility to colicins B and D (40Pollack J.R. Neilands J.B. Biochem. Biophys. Res. Commun. 1970; 38: 989-992Crossref PubMed Scopus (302) Google Scholar),59FeEnt binding (9Newton S.M. Igo J.D. Scott D.C. Klebba P.E. Mol. Microbiol. 1999; 32: 1153-1165Crossref PubMed Scopus (77) Google Scholar), and FeEnt transport in qualitative siderophore nutrition assays (5Wayne R. Frick K. Neilands J.B. J. Bacteriol. 1976; 126: 7-12Crossref PubMed Google Scholar) and quantitative determinations of59FeEnt uptake (9Newton S.M. Igo J.D. Scott D.C. Klebba P.E. Mol. Microbiol. 1999; 32: 1153-1165Crossref PubMed Scopus (77) Google Scholar). For fluorescein labeling of live bacteria, we grew the cells to stationary phase in LB broth (43Miller J.H. Experiments in Molecular Genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N. Y.1972Google Scholar) and subcultured at 1% into MOPS medium (44Neidhardt F.C. Bloch P.L. Smith D.F. J. Bacteriol. 1974; 119: 736-747Crossref PubMed Google Scholar) containing appropriate nutritional supplements at 37 °C with shaking to mid-log phase. For labeling in energy-deficient conditions, after growth in LB and subculture in MOPS the bacteria were collected by centrifugation and resuspended in the same volume of MOPS media but without glucose or casamino acids, and with half the usual supplementation of required amino acids. The culture was then incubated with shaking for 10 h at 37 °C. Bacteria were collected by centrifugation (5000 × g, 20 min), washed with and resuspended in Tris-buffered saline (TBS), pH 7.4, to a final concentration of 5 × 108 cells/ml. Fluorescein maleimide (FM) or AlexaFluor680 maleimide (AM) in dimethyl formamide (less than 0.1% of final volume) was added to a final concentration of 5 μm and incubated at room temperature for 30 min in the dark with shaking. The labeled cells were centrifuged, washed three times with 25 ml of ice-cold TBS plus 0.05% Tween-20, washed once with and resuspended in ice-cold TBS, and stored on ice for immediate use. We evaluated labeling specificity with anti-FepA and anti-IAF immunoblots of cell lysates and with fluorescence emission scans of labeled experimental and control bacterial cultures. In the former case, lysates from 108cells were subjected to 10% SDS-PAGE and Western immunoblot with anti-BSA-IAF sera; the reactions were quantified on a Storm Scanner (Molecular Dynamics) after development with 125I-Protein A (42Newton S.M. Allen J.S. Cao Z., Qi, Z. Jiang X. Sprencel C. Igo J.D. Foster S.B. Payne M.A. Klebba P.E. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 4560-4565Crossref PubMed Scopus (52) Google Scholar). The immunoblots showed that expression of FepA did not significantly vary under the culture conditions we employed. The anti-BSA-IAF serum was equally effective against proteins and cells modified by FM, demonstrating its recognition of the fluorescein moiety. The characteristic orange color of fluorescein facilitated qualitative evaluation of labeling specificity by visual inspection of treated cell pellets. Strains expressing wild type FepA, which was not labeled by either IAF or FM, were not colored after treatment with the reagents, whereas bacteria expressing mutant FepA proteins containing single, genetically engineered cysteines in accessible locations were identified by the orange color of their cell pellets. To quantitatively determine fluorescein labeling specificity, we performed emission scans at 20 °C with an excitation wavelength of 490 nm and a 5 s integration time. We compared scans for fluorescein-labeled strains expressing wild type and mutant FepA proteins and used scans of untreated bacteria to establish the background fluorescence. Using an SLM-AMINCO 8000 fluorimeter (Rochester, NY) upgraded to 8100 functionality, we recorded the fluorescence intensities of bacteria (5 × 107 cells) suspended in 2 ml of either TBS or MOPS media and equilibrated at temperature. For fluorescence determinations of FeEnt binding affinity, we added various concentrations of FeEnt to the cell suspension and recorded fluorescence intensity with excitation and emission wavelengths of 490 nm and 518 nm, respectively, for fluorescein and 679 nm and 702 nm for AlexaFluor680. We accounted for background fluorescence and volume changes and analyzed the data with the bound (1-F/F0)versus total function of GraFit 4 (Erithacus Software Ltd., Middlesex, UK). To study the effect of inhibitors on FeEnt uptake, 5 × 107FM-labeled, energy-starved cells were first incubated for 40 min at 37 °C with shaking in TBS buffer plus glucose (0.4%) and the corresponding inhibitors. The cells were transferred to the sample cuvette, equilibrated at the measurement temperature, and the uptake time course was recorded. OM proteins contain few unpaired cysteines, and we tested the possibility that genetically engineered single Cys residues at sites of interest may be covalently modified in live bacteria with -SH-specific fluorescent probes. Although we previously modified residue E280C in FepA L3 with nitroxide compounds (14Jiang X. Payne M.A. Cao Z. Foster S.B. Feix J.B. Newton S.M. Klebba P.E. Science. 1997; 276: 1261-1264Crossref PubMed Scopus (91) Google Scholar, 33Liu J. Rutz J.M. Klebba P.E. Feix J.B. Biochemistry. 1994; 33: 13274-13283Crossref PubMed Scopus (46) Google Scholar, 34Klug C.S., Su, W. Liu J. Klebba P.E. Feix J.B. Biochemistry. 1995; 31: 14230-14236Crossref Scopus (39) Google Scholar), we could not label it with fluorescent probes (45Cao Z. Ferric Enterobactin Transport through FepA in Escherichia coli from Hydrophobic Stacking to Charge InteractionsPh.D. thesis. University of Oklahoma, 1999Google Scholar), probably because of the inaccessibility of sites close to the hydrophilic OM surface to hydrophobic molecules like fluorescein (Fig. 1 A). Therefore, from the FepA crystal structure we designed and introduced 10 more individual, unpaired Cys residues at positions either on the cell surface or within the periplasm: W101C, S150C, S211C, Y260C, S271C, F329C, S397C, S423C, S575C, and S595C. Among these 10 target Cys residues, FM specifically labeled W101C, S271C, F329C, and S397C, which all reside in cell surface-exposed loops of FepA (Fig. 1 A) above the level of the LPS core sugars; W101 lies in the second loop of the N-domain, whereas S271, F329, and S397 exist in loops 3, 4, and 5, respectively, of the C-domain. FM also specifically modified, at lower levels, sites S63C and S150C, which reside in NL1 and on the periplasmic rim of the FepA β-barrel, respectively. Further study showed that two circumstances affected the specificity and efficiency of the fluorescence labeling procedure: the nature of the LPS O-antigen and the bacterial culture conditions. In our initial experiments the bacteria adsorbed fluorescence probes, but we saw little difference in the fluorescence intensity of FM-treated KDF541/pFepAS271C and KDF541/pITS449 (fepA +). The use of two different reagents, IAEDANS and coumarin maleimide, did not remedy this lack of specificity (data not shown). KDF541, the E. coli strain that was the usual host for plasmids, produces rough LPS without any O-antigen, and we considered the possibility that LPS was a determining factor in the specificity of the fluorescence labeling reactions. The use of deep rough mutant strains did not improve the labeling specificity of surface-exposed Cys residues (data not shown). Although Western blots revealed specific covalent modification of some of the Cys mutants, FM nonspecifically adsorbed to deep rough strains, which was apparent in the yellow color of the cell pellets that was not diminished by repeated washing. These results suggested that in deep rough strains, and to a lesser extent in rough strains, the fluorescent reagents penetrated into the OM bilayer and resisted washing procedures. The converse experiment confirmed this inference; the use of bacterial cells synthesizing full-length LPS minimized nonspecific adsorption of fluorescent dyes. We achieved this result by introducing pMF19, which carries a rhamnosyl transferase that allows production of the LPS O-chain (wbbL+) (37Feldman M.F. Valvano M.A. J. Biol. Chem. 1999; 274: 35129-35138Abstract Full Text Full Text PDF PubMed Scopus (174) Google Scholar). Strains harboring pMF19 and one of the four accessible Cys substitution mutants of FepA were specifically labeled by FM, as seen by fluorescence intensity measurements of live bacteria and in Western immunoblots (Fig. 1 B). We measured the ability of the four surface-localized, Cys substitution mutants of FepA to bind 59FeEnt before and after fluoresceination. Adsorption of the ferric siderophore was weak and barely detectable in W101C- and F329C-FM; attachment of fluorescein at these sites sterically hindered the adsorption of the siderophore. These data concurred with the prior conclusion that residue Phe329 exists in close proximity to, or is a component of, an initial FeEnt binding site (B1) (11Cao Z., Qi, Z. Sprencel C. Newton S.M. Klebba P.E. Mol. Microbiol. 2000; 37: 1306-1317Crossref PubMed Scopus (40) Google Scholar); the crystal structure shows W101C in the middle of the FepA vestibule (Fig. 1 A). On the other hand, FM modification of S271C or S397C did not interfere with FeEnt recognition and binding; bacteria expressing either of these mutant proteins manifested normal (subnanomolar) binding affinities (Fig. 2) even after fluoresceination. These experiments showed, for the first time, correspondence between the binding affinities measured in vivo by radioisotopic and fluorescence methodologies (K d = 0.3 nm). In subsequent experiments we exclusively studied the S271C site, located at the extremity of L3. Ligand binding diminished the fluorescence of probes attached to FepA at E280C, and our modifications and analyses of S271C and S397C showed the same quenching phenomenon in vivo. Previous determinations of binding kinetics found a biphasic association reaction between purified FepAE280C-FM and FeEnt and ColB (10Payne M.A. Igo J.D. Cao Z. Foster S.B. Newton S.M. Klebba P.E. J. Biol. Chem. 1997; 272: 21950-21955Abstract Full Text Full Text PDF PubMed Scopus (46) Google Scholar, 11Cao Z., Qi, Z. Sprencel C. Newton S.M. Klebba P.E. Mol. Microbiol. 2000; 37: 1306-1317Crossref PubMed Scopus (40) Google Scholar). We interpreted these data as initial ligand adsorption to an external site (B1), followed by movement to a second site deeper in the vestibule (B2). The ability of E280C-FM to reflect these two binding phases presumably arose from changes in the local environment of the fluor that occurred as the receptor protein underwent conformational dynamics during ligand adsorption. However, because its absorption spectrum overlaps the emission spectrum of fluorescein, the possibility existed that during its binding FeEnt quenched fluorescence by energy transfer between the excited fluor and the ferric siderophore. Studies with a red-shifted fluorescein derivative that does not overlap the absorption spectrum of FeEnt (AM: 679 nm and 702 nm, respectively), refuted this explanation (Fig. 2); during FeEnt binding, FepA proteins modified with AM exhibited equivalent reductions in intensity to those modified with FM. The comparable quenching of AM, without the possibility of energy transfer to FeEnt, indicated that the reductions in intensity did not derive from close proximity of the ferric siderophore to the fluor. Rather, the data favored the notion that binding triggers conformational dynamics that alter the environment of the reporter molecules. At physiological temperatures we expected a decrease in the polarization of fluorescent labels attached to the surface loops of FepA because of increased Brownian motion (46Einstein A. Ann. Phys. 1906; 19: 371Crossref Scopus (685) Google Scholar, 47Langevin P. Comptes Rendus. 1908; 146: 530Google Scholar). The polarization of FepAS397C-FM (in L5) decreased as the temperature of the system was raised, but that of FepAS271C-FM (in L3) increased (Fig. 3). These data confirmed that the temperature affected the motion of the attached fluor; L3 became less mobile and L5 became more mobile in response to increased temperature. These states were reversible as the temperature of the system decreased again. When exposed to FeEnt at 20 °C, KDF541/pMF19/pS271C-FM bound and transported it, and we spectroscopically monitored these reactions. Subsequent to its binding, we observed FeEnt transport as a recovery of fluorescence intensity, but only when the extracellular free ligand was depleted by its uptake into the cells. At that time, when the population of receptor proteins was vacated, its original fluorescence intensity returned. The consumption of FeEnt was necessary to observe fluorescence changes because, when present, it bound and quenched again. Thus in the presence of sufficient excess of the ferric siderophore, we saw no renewal of fluorescence.2 When it occurred, the resurgence of fluorescence paralleled the kinetics of FeEnt uptake, independently measured in 59Fe uptake experiments (Fig. 4). In those studies, the increase in fluorescence intensity occurred between 1200 and 1400 s. The corresponding radioisotope uptake experiment at 20 °C showed that accumulation of 59FeEnt stopped at the same time as a result of depletion of the substrate from the media. So the recovery of fluorescence intensity precisely mirrored the depletion of the ligand from the culture. Consistent with prior observations of the energy-dependent uptake reaction (9Newton S.M. Igo J.D. Scott D.C. Klebba P.E. Mol. Microbiol. 1999; 32: 1153-1165Crossref PubMed Scopus (77) Google Scholar, 48Leong J. Neilands J.B. J. Bacteriol. 1976; 126: 823-831Crossref PubMed Google Scholar, 49Ecker D. Matzanke B. Raymond K.N. J. Bacteriol. 1986; 167: 666-675Crossref PubMed Google Scholar, 50Pugsley A.P. Reeves P J. Bacteriol. 1977; 130: 26-36Crossref PubMed Google Scholar, 51Thulasiraman P. Newton S.M., Xu, J. Raymond K.N. Mai C. Hall A. Montague M.A. Klebba P.E. J. Bacteriol. 1998; 180: 6689-6696Crossref PubMed Google Scholar), the spectroscopic measurement of FeEnt transport was affected by media composition, FeEnt concentration, and temperature. It occurred more slowly in bacteria deprived of nutritional requirements, either in minimal media lacking auxotrophic amino acid supplements or suspended in TBS (Fig. 5 A). As discussed above, at 20 °C the transport of FeEnt was visible as a return of fluorescence intensity, and the concentration-dependence of the phenomenon was striking. For bacteria exposed to 2, 5, 10, and 20 nmFeEnt, the lag time preceding transport was 250, 550, 1000, and 2050 s, respectively (Fig. 5 B). Furthermore, AM-labeled cells showed t
Referência(s)