| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 281, Issue 35, 25541-25550, September 1, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
12
From the
Unité des Membranes Bactériennes, CNRS URA 2172 Département de Microbiologie and Unité de Résonance Magnétique Nucléaire des Biomolécules CNRS URA 2185 Département de Biologie Structurale et de la Chimie, Institut Pasteur, 75724 Paris Cedex 15 France, ||CNRS FRE 2737, Université de la Méditerranée (Aix-Marseille II), Faculté de Pharmacie 13385 Marseille Cedex 05 France, and ¶Department of Chemistry, Biochemistry, and Molecular Biology, North Dakota State University, Fargo, North Dakota 58105-5516
Received for publication, April 18, 2006 , and in revised form, June 9, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-sheet (seven strands) on one side of the protein and four 
-helices on the other (6). The affinity of HasA for heme is due to two axial heme iron ligands (His-32 and Tyr-75) and a third residue (His-83) stabilizing the Tyr-75-Fe bond (4). The binding of HasA to the HasR receptor depends on two strands making independent contact with the receptor (7). The HasR receptor belongs to the TonB-class outer membrane-dependent receptors. The three-dimensional structures of these receptors are organized in a "plug and barrel" motif (2). Homology modeling of HasR, based on structural and sequential alignments with receptors of known structure, is consistent with HasR having the same organization (8). Heme receptors have several conserved features; two conserved histidine residues and a so-called FRAP box (9). HasR has similar features, with two conserved histidine residues that are close to each other, as shown by the model; one at the apex of the "plug" and the other one on an extracellular loop (8). Mutations of these conserved residues have been shown to be deleterious in the heme receptor HemR in Yersinia enterocolitica (9). The HasR receptor can use different substrates, either free heme up to a concentration of 10-6 M or the heme-HasA complex. The system is more efficient for the heme-HasA complex, allowing growth at a heme concentration 100 times lower than for free heme (10). For both substrates heme transport depends upon turnover of the inner membrane TonB·ExbBD protein complex and the proton motive force across the cytoplasmic membrane. Higher concentrations of the TonB complex and greater proton motive force are required for heme uptake involving HasA than for uptake of free heme (11). The precise cascade of events that includes recognition of the hemophore by the receptor, internalization of heme, and release of the empty hemophore is not precisely known. Our in vivo studies have clearly shown that heme transport is dependent upon the TonB complex and the pmf and that hemophore recycling is also dependent upon the TonB complex and the pmf. However, it is not known whether the TonB complex is required for transfer of the heme from the heme-loaded hemophore to the apo-receptor. We have also shown that recycling of HasA requires heme and a higher TonB complex concentration and pmf than uptake of free heme. In vivo, the HasA·HasR complex is irreversible in the absence of an energy source, the TonB complex, and of heme (5). In the current study the first steps of heme acquisition, the binding of the receptor to its various partners (heme, apo- and holoHasA) and the transfer of heme from the hemophore to the receptor, have been investigated. We have shown that heme transfer from the holohemophore to the aporeceptor does not require the TonB complex and that protein-protein interactions between HasA and HasR drive the heme transfer. Furthermore, we have shown that heme binding to HasR involves two conserved histidine residues.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
hemA and C600hemAexbBD::Tn10 transformed with the various HasR constructs were used for growth tests with exogenous heme-loaded hemophore, heme, and hemoglobin, as previously described (7); for those experiments, HasA variants were used at a 80% heme charge at 40 and 4 µM, and HasR expression was induced by arabinose (20 mg/liter).
Plasmid Construction—PFR2H1 and pFR2H2 were constructed using pFR2 as a template in a mutagenic PCR using the following complementary mutagenic oligonucleotides: Hasrhis1, TTCAAAAGAGCGGCGCTGGCCAACGTAATG; Hasrhis1r, CATTACGTTGGCCAGCGCCGCTCTTTTGAA; Hasrhis2mod, ACCAACGGCAGCGCGGCCAGTTCTTCCACG; Hasrhis2rmod, CGTGGAAGAACTGGCCGCGCTGCCGTTGGT. The mutation was verified by sequencing and reintroduced into an otherwise WT pFR2 plasmid using EcoRI and BsrGI for H1 and BstZ17I and BsiWI for H2. For the construction of the double H1H2 mutant, the EcoRI-BsrGI fragment from pFR2H1 was exchanged for the same fragment from pFR2H2 and yielded pFR2H1H2.
Protein Expression and Purification, Biochemical Techniques—HisHasA was purified as previously described (7) with the following modifications. HisHasA from a soluble fraction of a 500-ml culture was loaded on a Q-Sepharose column (10 ml) and eluted with a 0 to 1 M NaCl gradient in 50 mM Tris-HCl at pH 7.5, allowing for separation of heme-loaded from heme-free protein. The heme-free protein-containing fractions were then combined and loaded on Ni-NTA-agarose and eluted with an imidazole gradient. This yielded pure apoHisHasA, which was purified using size exclusion chromatography. ApoHis-HasA could be loaded with heme using the theoretical molar extinction coefficient for His-HasA of 21,350 M-1 cm-1 at 276 nm.
ApoHasA WT was purified from cells grown in M9 minimal medium as previously described (12). ApoHasA76t and the heme pocket mutants were purified from a culture supernatant of Escherichia coli MC4100 harboring the relevant plasmids (grown in LB at 30 °C) by anion exchange chromatography followed by size exclusion chromatography.
Wild-type and mutant HasR were partially purified from crude membrane preparations obtained after French press treatment; membranes from 50 g of wet cells were resuspended and solubilized in 200 ml of 50 mM Tris-HCl at pH 7.5, 1% ZW3-14 (n-tetradecyl-N,N-dimethyl-3-ammonio-1-propanesulfonate; Calbiochem), 5 mM MgCl2, and protease inhibitors (Roche Applied Science mixture, EDTA free). After 1 h of incubation at 4 °C, the suspension was centrifuged (50,000 x g for 1 h). The supernatant, which mostly contained inner membrane proteins, was discarded, and the pellet was resuspended in the same volume of 50 mM Tris-HCl at pH 7.5, 2% ZW3-14, 20 mM EDTA, and protease inhibitors. After 1 h of incubation at 4 °C, the suspension was centrifuged (50,000 x g for 1 h), and the supernatant was frozen at 77 K in liquid nitrogen. The pellet was resuspended in the same volume of 50 mM Tris-HCl at pH 7.5, 2.5% ZW3-14, 20 mM EDTA, and protease inhibitors. After 1 h of incubation at 4 °C, the suspension was centrifuged (50,000 x g for 1 h), and the supernatant was frozen at 77 K in liquid nitrogen. The combined supernatants from the second and third solubilization were loaded at 2 ml/min on a 20-ml Q-Sepharose fast-flow column equilibrated with buffer A (50 mM Tris-HCl at pH 7.5, 0.02% ZW3-14, 20 mM EDTA) and extensively washed with the same buffer. The bound proteins were then eluted with 20 column volumes of a linear gradient of buffer A and buffer B (50 mM Tris-HCl at pH 7.5, 0.02% ZW3-14, 1 M NaCl) at a flow rate of 2 ml/min at 15 °C. The fractions (4 ml) were tested for the presence of the receptor. and the most enriched fractions were combined and concentrated on an Amicon centriflow filter (cutoff molecular mass of 100 kDa). A second purification by chromatography under the same conditions (but without EDTA) was carried out, and the concentrated fractions were purified on a size exclusion column (Superdex 200pg 320-ml column 60 x 2.6 or Superose 6 24-ml column 300/10; Amersham Biosciences) in 25 mM Tris-HCl at pH 7.5, 0.02% ZW3-14, 150 mM NaCl. The routine yield was about 10 mg of purified HasR from 10 g of cells. For purification of the holo receptor, hemin (10-5 M) was added either to the culture 1 h before harvesting or to the membrane preparation before detergent extraction. HoloHasR eluted at a higher salt concentration than for apoHasR for the anion exchange chromatography (not shown).
Complexes of apoHisHasA and apoHasR were purified; an excess of apoHisHasA resuspended in 25 mM Tris-HCl at pH 7.5, 0.02% ZW3-14, 150 mM NaCl was mixed with partially purified apoHasR in the same buffer. After 1 h of incubation at 4 °C, the preparation was mixed with Ni-NTA-agarose in the same buffer and incubated for a further hour. Unbound material was washed with the same buffer containing 20 mM imidazole. Bound material was eluted with the same buffer containing 0.5 M imidazole. The resulting concentrated imidazole eluate was purified on a size exclusion column (Superose 6, 24 ml, 0.5-ml fractions were collected). Control experiments showed that apoHasR did not bind to the Ni-NTA affinity resin under these same conditions (not shown). Absorption spectra (UV-visible) were recorded on a Beckman DU-800 spectrophotometer with 1 cm optical path cuvettes.
Gel electrophoresis was carried out according to Laemmli (13) either at room temperature or at 4 °C for the gel used for heme peroxidase revelation by Enhanced Chemiluminescence (ECL) (14); in that last case the gel was transferred on a nitrocellulose membrane before development. Amino acid analysis was performed after 6 N HCl hydrolysis for 24 h at 110 °C by the Laboratoire de Chimie Organique at the Institut Pasteur.
Resonance Raman Spectroscopy—Resonance Raman (rR) spectra were obtained from 10 to 55 µM HasA, HasR, and HasA·HasR samples, which were contained in a 5-mm NMR tube spinning at 
20 Hz. Raman scattering was excited using either 413.1 nm emission from a Kr+ laser (10 milliwatt). UV-visible absorbance spectra were recorded before and after the rR experiments to ensure that the samples were not irreversibly altered in the laser beam. No spectral artifacts attributable to laser-induced sample damage were observed. Spectra were recorded at ambient temperature using 135° backscattering geometry with the laser beam focused to a line. Scattered light was collected with an f1 lens and filtered with a holographic notch filter to attenuate Rayleigh scattered light. The polarization of the scattered light was then scrambled, and the spot image was f-matched to a 0.63-m spectrograph fitted with a 2400 groove/mm grating and a LN2-cooled CCD camera. The spectrometer was calibrated using the Raman bands of toluene and N,N-dimethylformamide as external frequency standards.
Isothermal Titration Calorimetry (ITC)—Titrations were performed at 25 °C using a MicroCal MCS titration calorimeter (MicroCal Inc., Northampton, MA) (15). Proteins and hemin were in 20 mM sodium phosphate at pH 7, 0.08% ZW3-14. Samples were thoroughly degassed before use by stirring under vacuum. All injections were carried out at 3-min intervals. Due to heme adsorption, the calorimeter cell and the micro syringe used for injections were extensively washed after each experiment. The heat of dilution of the ligand was measured either by injecting the ligand into the buffer alone or by injecting more ligand into the cell after saturation. The value obtained was subtracted from the heat of reaction to give the effective heat of binding. Because apoHasR contains some contaminants, all of the ITC control titrations were carried out with membrane preparation not expressing HasR and purified in the same manner at the equivalent concentration of contaminant than that present in apoHasR sample, i.e. few 10-6 M. No higher signal than that corresponding to the heat dilution of ligand was observed. For the titration experiments the concentrations of wild-type and mutant receptors ranged between 2 and 10 µM, and the ligand was set to 10-20 times this value.
The resulting titration data were analyzed and fitted using the Origin for ITC software package supplied by Microcal to obtain the stoichiometry (n), the dissociation constants (Kd), and the enthalpy (H) changes of binding. For the fit any constraints on the stoichiometry and H were not fixed. ITC titrations of HasR with hemin showed 20% of variation of stoichiometry around 1. This was due to the imprecision in the determination of hemin concentration in the presence of detergent and to the fact that hemin is known to be very dynamic in nature inside micelles (16).
The formation of the complex during the ITC experiment was verified by SDS-PAGE and UV-visible spectroscopy. Analysis of the samples by optical absorbance after the ITC experiments was carried out after removal of excess ligand by filtration using a membrane having a molecular mass cut off of 100 kDa. The same protocol was applied for Raman spectroscopy when needed. Ultrafiltration through a membrane having a molecular mass cut off of 100 kDa passes uncomplexed holo-HasA and free heme, thereby allowing them to be separated from the complex.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
278 is in reasonable agreement with that calculated on the basis of amino acid composition (1.4 x 105 M-1 cm-1). Because our apoHasR preparation contains some contaminants, its absorbance in the preparation was estimated to account for 
of the total absorbance at 278 nm. Its concentration in subsequent experiments was estimated using the calculated 278 of 1.4 x 105 M-1 cm-1 and multiplying the result by 2/3. We have previously shown that both holoHasA and apoHasA bind to cells expressing the HasR receptor. These associations occur with similar affinities and are independent of the TonB complex (5). Herein we address whether analogous behavior occurs between HasA and the purified receptor. Because HasR is purified in the presence of detergent, all of these experiments were carried out in detergent solution to maintain solubility of HasR.
Both ApoHasA and HoloHasA Form a 1:1 Stoichiometric Complex with ApoHasR—Stable complexes form upon mixing of both apo- and holoHasA with apoHasR. These complexes can be isolated chromatographically, but because of impurities in our apoHasR preparations, it was difficult to determine their precise stoichiometries. The impurities, which might account for 
of the absorbance of the apoHasR preparation, do not react with HasA (see ITC section). These impurities cannot, therefore, account for spurious determination of the stoichiometry as we carried it out. A His6 tag HasA (HisHasA), having the same affinity for heme, spectral characteristics, and competence to deliver heme to the receptor as wild-type HasA (8), was prepared. It was purified to homogeneity in both its holo and apo forms by affinity chromatography. In the presence of apoHasR, a stable HisHasA·HasR complex formed and was separated from HisHasA by a single gel filtration step (Fig. 2, A and B). Amino acid analyses of HisHasA and HasR complexes made with both apo- and holo-HisHasA were performed. The compositions of both complexes were consistent with 1:1 stoichiometry (supplemental Table S1). It is reasonable to suggest that a similar complex is formed with the non-tagged version of HasA.
UV-Visible Absorbance Spectra of HoloHasR and the Complex Formed between HoloHasA and ApoHasR Are Identical— The absorbance spectrum of the purified complex obtained between holoHisHasA and apoHasR differs markedly from that of holoHasA and is virtually identical to that of holoHasR (Fig. 2C). This shows that heme in the HisHasA·HasR complex has an environment similar to that in holoHasR.
|
Resonance Raman Spectra of holoHasR and the Complex Formed between HoloHasA and ApoHasR Are Identical and Differ from That of HoloHasA—As a means of further testing our hypothesis that heme is transferred from HasA to HasR, we have used resonance rR spectroscopy to compare and contrast the heme environments in holoHasR, holoHasA, and the HasA·HasR complex. The vibrational signatures of the heme, which are provided by rR, constitute a sensitive probe of heme structure, conformation, and environment in proteins.
rR spectra were recorded for holoHasA, holoHasR, and the complex formed between holo-HasA and apoHasR (Fig. 4). Oxidation state, spin state, and ligation state marker bands for heme proteins are observed in the high frequency region of their rR spectra. Two 
3 bands (1476 and 1503 cm-1) are observed for holoHasA, consistent with the protein existing in both 6-coordinate high spin and 6-coordinate low spin forms as reported previously (18). Upon the addition of excess apo-HasR, the 3 band attributed to 6-coordinate low spin heme shifts from 1503 to 1501 cm-1, where it appears in the spectrum of holoHasR. The rR spectrum of this complex is characteristic of ferric 6-coordinate bishistidine complexes. Accordingly, the oxidation state/porphyrin 
* marker band, 4, for holoHasA, holoHasR, and the HasA·HasR complex is observed at 1372 cm-1, typical of ferric hemes. A small band at about 1476 cm-1 persists after the transfer reaction has reached equilibrium. The spectrum in Fig. 4 is that of a 1:2 mixture of holoHasA and apoHasR. The spectrum of a 1:3 mixture exhibits the same small band. This band persists even after the complex is washed with an excess of buffer via ultra filtration. Hence, the 3 band at 1476-cm-1 is intrinsic to the complex and corresponds to neither uncomplexed holoHasA nor free heme.
|
|
(CCcCd) and vinyl (CCaCb) groups in holoHasR have been tentatively assigned to the bands at 376 and 417 cm-1 by comparison with rR assignments for bisimidazole heme model complexes and proteins (19, 20). Moreover, the low frequency rR spectra of holoHasR and the HasA·HasR complex are indistinguishable. Thus, the propionate and vinyl substituents of the hemes in authentic holoHasR and in the HasA·HasR complex are in comparable environments. Given that (a) the heme in holo-HasA exists as an equilibrium mixture of high and low spin states and (b) the heme in authentic holoHasR is completely low spin, there are two possible explanations for the features in the rR spectrum of the HasA·HasR complex. One is that the heme remains associated with HasA when it complexes with apoHasR, and it is the protein-protein interaction that drives the spin state change to convert all the heme to a low spin form. The second explanation is that the heme is transferred from holoHasA to HasR upon protein-protein complexation. The latter is supported by the identity between the rR signatures of holoHasR and the HasA·HasR complex. The aforementioned similarities in both the axial ligand field and peripheral heme environment constitute compelling circumstantial evidence that heme is transferred from its binding site in HasA to a binding site in HasR upon complex formation.
ITC Studies of the Interaction of HasR with Its Different Ligands, Heme/ApoHasA/HoloHasA—We have used isothermal titration calorimetry to understand the molecular events involved in this transfer and to characterize the interaction between the three partners, HasA, HasR, and heme.
HasR/Heme—the ITC titration of apoHasR with hemin displays a very weak enthalpic signal. The data fit to a single-site (or set of equivalent sites) model with a stoichiometry close to 1, an affinity constant of 5(±3) x 106 M-1, and a H of -18 ± 3 kJ·mol-1 (Table 1). Because the apoHasR preparations were not pure, a control experiment on similar preparation from a strain that does not express HasR was carried out. No signal apart from that attributable to dilution was detected.
|
H is -105 kJ·mol-1, and the affinity constant is >109 M-1. In addition, the structural fingerprints of 15N-labeled holoHasA in 20 mM sodium phosphate buffer, pH 7, with or without detergent have been compared by 1H,15N heteronuclear single quantum correlation spectrum. These spectra are identical, showing that the presence of detergent does not modify the holoHasA structure (data not shown).
|
H of -201 ± 16 kJ·mol-1, and an affinity constant >109 M-1, too large to be deduced precisely from ITC titration (Fig. 5A). Whatever the exact value, it shows the very high affinity of the receptor for the hemophore, which is at least 10-fold higher than the apparent Ka estimated in vivo. Again there was no signal with the control preparation, showing the specificity of the signal.
HoloHasA/ApoHasR—The ITC titration of apoHasR by holoHasA gave an enthalpic signal lower than that observed for apoHasA-apoHasR (Fig. 5B). Fitting of the data by a single-site model indicated a 1:1 stoichiometry and an observed enthalpy change (Hobsd) of -105 kJ·mol-1. The UV-visible spectral properties of the complex so formed after removal of the excess of ligand shows that heme transfer accompanies the formation of the complex from holoHasA to apoHasR. The observed ITC signal is, thus, the sum of the protein-protein interaction signal, that of extracting heme from HasA and that of transferring heme to apoHasR. Hence, the observed enthalpy signal varies on the basis of the equation Hobsd = Hbinding - Hheme-HasA + Hheme-HasR. Hence, Hbinding = Hobsd + Hheme-HasA - Hheme-HasR = -105 kJ·mol-1 - 105 kJ·mol-1 + 18 kJ·mol-1, and the Hbinding of holoHasA to apo-HasR given by this equation is -192 kJ·mol-1. This value is comparable with that obtained by ITC for the binding of apo-HasA and apoHasR (-201 kJ·mol-1). No difference was found for in vivo binding of both forms of HasA to HasR, and no structural difference was observed in the regions of HasA shown to be involved in interaction with HasR in the presence or in the absence of heme.4 All those results are consistent with the fact that the same polar interactions are involved in the binding of both forms of HasA (apo/holo) to HasR. The interaction of HasA with HasR involves two distinct strands on HasA, and the very large H value suggests that several residues are involved in the interaction.
The exact Ka between holoHasA and apoHasR cannot be directly measured by ITC titration because of the heme transfer reaction occurring during the ITC measurement. The apparent Ka between holoHasA and apoHasR is in the order of 5 x 107 M-1.
Role of the Two Conserved HasR Histidine in Heme Uptake and Heme Transfer from HasA to HasR—To gain further insight into the heme transfer reaction between holoHasA and apoHasR, both HasR and HasA mutants were constructed and studied in vivo and in vitro. Like many heme receptors, HasR contains two conserved histidine residues, one in the plug His-189 (hereinafter referred to as H1) and the other one in the -barrel His-603 (hereinafter referred to as H2). Mutation of these residues in HemR, the heme receptor from Y. enterocolitica or HmuR from Porphyromonas gingivalis, led to an inactive receptor (9, 21). Homology modeling of HasR using the already crystallized receptors structures as templates indicated that those two residues might be in close proximity at the apex of the receptor, H1 at the top of the plug and H2 in an extracellular loop and facing the plug (8). Both were mutagenized into alanine and the two single mutants, and the double mutants were constructed. All three proteins were expressed to comparable levels without evidence for degradation. They were correctly localized to the outer membrane and shown by dot-blot to interact with HasA (data not shown). All three proteins could be purified using the same protocol as for wild-type apoHasR, and the same molar extinction coefficient values were used to estimate their concentrations.
In vivo experiments indicated that both single mutants were still able to take up free heme but with a lower efficiency than the wild-type receptor, whereas the double mutant was no longer able to take up free heme (Table 2). This heme uptake was TonB complex-dependent (not shown). This indicates that one of the two histidines is sufficient to bind and internalize heme, but higher heme concentration than that necessary for WT-receptor was required for growth. This suggests that the affinity of these receptors for heme is affected by the mutation. The ITC titration of these simple mutants with hemin in our experimental conditions did not show higher enthalpic signal than that of dilution. This is compatible with the loss of one iron coordination and the very weak enthalpic signal observed for the wild-type receptor with heme. It is expected that the double mutant has no significant affinity for heme.
|
H of binding of apoHasA·HasRH1H2 is lower (-136 kJ·mol-1 versus -210 kJ·mol-1 for apoHasA-apoHasR), and the affinity constant is 3 x 107 M-1, whereas that of the interaction of apoHasA with the wild-type receptor is >109 M-1 (Table 1, Fig. 5C). This significant difference in binding, which could be attributable to structural modifications of this mutant or at least of the parts of the receptor involved in the binding with HasA, precluded further analysis of the heme transfer from thermodynamic parameters. HasRH1 and HasRH2 mutants are still active heme transporters but cannot acquire heme from wild-type HasA. If the heme transfer step from HasA to HasR depends upon the respective affinities of HasA and HasR for heme, one might then expect that HasA mutants having lower heme affinities (mutated in the three residues involved in the heme axial ligation His-32, Tyr-75, and His-83) could still serve as heme sources for the HasR mutants studied here, since none of them are per se required for heme transport. Hence, the abilities of these HasA heme pocket mutants to serve as heme sources for the WT and mutant receptors were explored. The results of these experiments are shown in Table 2. Under our conditions, the wild-type receptor can use HasA wild type and the mutants tested whose affinity for heme ranges from 5.3 x 1010 M-1 to 6 x 104 M-1. Whereas neither wild-type HasA (Ka = 5.3 x 1010 M-1) nor HasA-H32A (Ka = 1 x 1010 M-1) can serve as an heme source for the H2 mutant of the receptor, all the other HasA mutants with lower affinities starting with HasA-H83A (Ka = 2 x 108 M-1) could. A similar situation was found with the H1 mutant where none of the HasA mutants could serve as the heme source except HasA-Y75A-H83A (Ka = 1.8 x 106 M-1) and HasA-H32A-Y75A (Ka = 5.9 x 104 M-1). Our results indicate that the mutation of one of the two histidines does not abolish the capacity of HasR to acquire heme either free or bound to HasA but only decreases it. Moreover the two histidines are not equivalent since the mutation of H2 (barrel histidine) results in a milder defect in heme acquisition as compared with that of H1 (plug histidine). Hence, it is concluded that the two histidines are responsible for heme fixation of the receptor and are not equivalent in the functioning of the receptor.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
HasR is one such outer membrane receptor involved in heme transport across the outer membrane. This heme transport process is an active one since it involves the TonB complex and the proton-motive force across the inner membrane (11). Although HasR transports only heme, it is able to recognize three substrates from the extracellular side, either free heme or apo or holohemophore. In the case of heme uptake from the hemophore, the TonB complex is required both at the heme transport step across the outer membrane through the receptor and at the dissociation of the empty hemophore from the receptor. The latter step is the most energy-consuming (11).
In this study we showed that HasR binds one molecule of heme with an affinity of 5 x 106 M-1. In holoHasR, the Raman and UV-visible spectral signatures are consistent with bishistidine axial ligation of a low spin heme iron. Individual mutation of the two conserved histidines in HasR decreases its capacity for heme acquisition, thereby establishing His-189 in the plug and His-603 in the barrel as heme ligands in HasR. The affinity of HasR for heme is much lower than the affinity of siderophores for their cognate receptors and is consistent with our in vivo observations of growth stimulation by heme of a heme auxotroph strain expressing the HasR receptor. However, the affinity for heme of HasR is on the same order as that of other heme receptors. The present study shows that HasR forms stable 1:1 complexes with both the apo and holo forms of HasA with affinities >109 M-1. This makes the heme acquisition by HasR via HasA very efficient and advantageous compared with the heme receptors of other bacteria.
The most striking result of our study is that heme transfers from purified holoHasA to apoHasR in vitro in the absence of any source of energy external to the HasA·HasR complex. This transfer of heme is supported by several independent lines of evidence. Changes in the UV-visible and rR spectra are consistent with formation of a new bis-His-ligated heme having a peripheral heme environment distinct from that in holoHasA. The thermodynamic parameters of interaction between holo-HasA and apoHasR, as compared with those of the self-association of the individual components and their association with heme, are consistent with this conclusion. These lines of evidence suggest that the protein-protein interaction between HasA and HasR drives heme transfer despite the unfavorable difference of affinity for heme. Analysis of HasA·HasR complexes obtained with mutants of both HasA and HasR also corroborated this conclusion.
Such phenomena may also occur in other receptors that recognize heme or iron-containing proteins, and a reasonable hypothesis is that substrate transfer will be spontaneous due to protein-protein interaction and independent of TonB. In this respect, it should be mentioned that apo and holo human transferrin recognize the Neisseria transferrin receptor TbpA in vitro and that the H is much lower for holoTf than for apoTf (28). The lower H seen for holoTf-TbpA may be the result of a protein-protein interaction and of the extraction of iron from transferrin and of its binding to the receptor. However, iron transfer from transferrin (Ka = 1018 M-1) to the receptor has not been studied either in vivo or in vitro (29). It has been proposed that bilobate structures might provide adequate environment for such binding and release of cofactors, in particular in the case of transferrin or hemopexin (30). In the HasA case, the contribution of the polar interactions in the binding process between HasA and HasR seems to be equivalent and heme-independent. In any case, the HasA structure might also be well adapted for binding and release of the cofactor. In HasA, heme is coordinated by the N of His-32 on one side and by the O
of Tyr-75 on the other side. These two residues and a third, His-83, which play important roles in the binding of heme, are located on two loops separated by three strands, one of which contains a binding site for HasR (7). Furthermore, the structural organization of HasA places one of the strands involved in interaction with HasA less than 10 Å from each of the heme ligands. These spatial relationships between the heme and HasR binding sites of HasA may facilitate the propagation of local conformational changes driven by the HasA·HasR interaction to the heme ligands of HasA and promote the heme delivery from HasA to HasR. In this respect it is worth mentioning that a HasA mutant isolated during the course of a systematic pentapeptide insertion mutagenesis (7) bore an insertion at position 76 close to Tyr-75, one of the heme ligands. This mutant (HasA76t) is able to bind heme with high affinity (Ka = 7 x 108 M-1) and is able to bind to the receptor, as shown by dot-blot analysis on intact cells expressing the receptor (31), but is unable to serve as a heme source for the wild-type receptor. In vitro with purified apoHasR, HasA76t makes complexes comparable with the wild-type HasA but without the characteristic spectral changes, suggesting that heme is not transferred to HasR. The insertion of this pentapeptide in mutant HasA76t at position n + 1 in the loop containing Tyr-75 heme ligand seems to inhibit whatever conformational reorganization is normally induced by HasR and required for heme transfer. Nevertheless, the binding of HasA to HasR seems to be irreversible in vitro, consistent with the fact that ejection of apoHasA requires energy provided by the TonB complex. This suggests that some structural changes of HasR and/or HasA should occur to allow the dissociation. Moreover, attempts to dissociate the complex in vitro at high ionic strength chemically or thermally have so far been unsuccessful. The sites of interaction between HasA and HasR have been defined on HasA (7), and one expects that definition of those sites on HasR will help understanding the heme transfer process as well as the ejection of hemophore under the action of the TonB complex.
A similar movement of cofactor/substrate upon protein-protein interaction has been reported for ADP/phosphate release upon interaction of myosin and actin, where the closing of the actin binding cleft is coupled to the opening of the nucleotide binding pocket (32). Although it is reasonable to suspect that heme transfer reactions similar to that reported here would be common, heme transfer that is apparently driven by protein-protein interaction has only been described for the heme transport systems of Pseudomonas aeruginosa (33) and Streptococcus pyogenes (34). Among the systems studied to date, HasA/HasR is unique in that its thermodynamic parameters are well quantified. This system is, therefore, well suited for elucidating further details of the thermodynamic and structural bases of the mechanism by which heme is transferred between these proteins.
|
FOOTNOTES |
|---|
The on-line version of this article (available at http://www.jbc.org) contains supplemental Table S1.
1 Supported by a fellowship from the Fondation pour la Recherche Médicale.
2 To whom correspondence should be addressed: Unité des Membranes Bactériennes, Département de Microbiologie, Institut Pasteur, 25-28, rue du Dr. Roux, 75724 Paris Cedex 15 France. Tel.: 33-1-40-61-32-76; Fax: 33-1-45-68-87-90; E-mail: pdelep{at}pasteur.fr.
3 The abbreviations used are: WT, wild type; Ni-NTA, nickel-nitrilotriacetic acid; rR, resonance Raman; ITC, isothermal titration calorimetry; ZW3-14, (n-tetradecyl-N,N-dimethyl-3-ammonio-1-propanesulfonate).
4 N. Wolff, N. Izadi-Pruneyre, J. Couprie, M. Habeck, J. Linge, W. Rieping, C. Wandersman, M. Nilges, M. Delepierre, and A. Lecroisey, manuscript in preparation.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Zhu, G. Xie, M. Liu, J. S. Olson, M. Fabian, D. M. Dooley, and B. Lei Pathway for Heme Uptake from Human Methemoglobin by the Iron-regulated Surface Determinants System of Staphylococcus aureus J. Biol. Chem., June 27, 2008; 283(26): 18450 - 18460. [Abstract] |
