The structural mechanism for iron uptake and release by transferrins

Transferrins are a group of iron-binding proteins that control the levels of iron in the body fluids of vertebrates by their ability to bind two Fe3+ and two CO3(2-). The transferrin molecule, with a molecular mass of about 80 kDa, is folded into two similarly sized homologous N- and C-lobes that are stabilized by many intrachain disulfides. As observed by X-ray crystallography, each lobe is further divided into two similarly sized domains, domain 1 and domain 2, and an Fe3+-binding site is within the interdomain cleft. Four of the six Fe3+ coordination sites are occupied by protein ligands (2 Tyr residues, 1 Asp, and 1 His) and the other two by a bidentate CO3(2-). Upon uptake and release of Fe3+, transferrins undergo a large-scale conformational change depending on a common structural mechanism: domains 1 and 2 rotate as rigid bodies around a rotation axis that passes through the two antiparallel beta-strands linking the domains. The extent of the rotation is, however, variable for different transferrin species and lobes. As a Fe3+ release mechanisms at low pH from the N-lobes of serum transferrin and ovotransferrin, the structural evidence for 'dilysine trigger mechanism' is shown. A structural mechanism for the Fe3+ release in presence of a non-synergistic anion is proposed on the basis of the sulfate-bound apo crystal structure of the ovotransferrin N-lobe. Domain-opened structures with the coordinated Fe3+ by the two tyrosine residues are demonstrated in fragment and intact forms, and their functional implications as a possible intermediate for iron uptake and release are discussed.     

Removal of iron from diferric rabbit serum transferrin by rabbit reticulocytes.

When radioiron-labelled transferrin with 55Fe located predominantly in the N-terminal iron-binding site and 59Fe predominantly in the C-terminal iron-binding site was incubated with rabbit reticulocytes, both radioisotopes of iron were removed at similar rates. Electrophoresis of transferrin samples taken during the course of an incubation, in polyacrylamide gels containing 6 M-urea, showed that iron was removed in a pairwise fashion, giving rise to iron-free transferrin.     

The influence of pH on the equilibrium distribution of iron between the metal-binding sites of human transferrin.

The dependence of the metal-binding properties of transferrin on pH in the pH 6--9 range was investigated by urea/polyacrylamide-gel electrophoresis. Equations are presented for calculating the relative values of the four conditional site constants for the stepwise binding of iron to the two sites of transferrin and for calculating the equilibrium distribution of the protein among the four principal forms, apotransferrin, the C-terminal and N-terminal monoferric transferrins and diferric transferrin. The relative affinity of iron for the two sites and the co-operativity of iron-binding follow characteristic "pH titration' curves. A mathematical model that can account for the former behaviour is presented. In both cases the metal-binding sites are affected by the ionization of functional groups with apparent pKa values near physiological pH approx. 7.4. There is strong positive co-operatively in the release of protons from these groups. The results indicate that pH must be accurately controlled in studies of the differential properties of the two sites of the transferrin molecule.     

Studies of the binding of different iron donors to human serum transferrin and isolation of iron-binding fragments from the N- and C-terminal regions of the protein.

1. Trypsin digestion of human serum transferrin partially saturated with iron(III)-nitrilotriacetate at pH 5.5 or pH 8.5 produces a carbohydrate-containing iron-binding fragment of mol.wt. 43000. 2. When iron(III) citrate, FeCl3, iron (III) ascorabate and (NH4)2SO4,FeSO4 are used as iron donors to saturate the protein partially, at pH8.5, proteolytic digestion yields a fragment of mol.wt. 36000 that lacks carbohydrate. 3. The two fragments differ in their antigenic structures, amino acid compositions and peptide 'maps'. 4. The fragment with mol.wt. 36000 was assigned to the N-terminal region of the protein and the other to the C-terminal region. 5. The distribution of iron in human serum transferrin partially saturated with various iron donors was examined by electrophoresis in urea/polyacrylamide gels and the two possible monoferric forms were unequivocally identified. 6. The site designated A on human serum transferrin [Harris (1977) Biochemistry 16, 560--564] was assigned to the C-terminal region of the protein and the B site to the N-terminal region. 7. The distribution of iron on transferrin in human plasma was determined.     

Crystallographic and biochemical analyses of the metal-free Haemophilus influenzae Fe3+-binding protein.

The crystal structure of the iron-free (apo) form of the Haemophilus influenzae Fe(3+)-binding protein (hFbp) has been determined to 1.75 A resolution. Information from this structure complements that derived from the holo structure with respect to the delineation of the process of iron binding and release. A 21 degrees rotation separates the two structural domains when the apo form is compared with the holo conformer, indicating that upon release of iron, the protein undergoes a change in conformation by bending about the central beta-sheet hinge. A surprising finding in the apo-hFbp structure was that the ternary binding site anion, observed in the crystals as phosphate, remained bound. In solution, apo-hFbp bound phosphate with an affinity K(d) of 2.3 x 10(-3) M. The presence of this ternary binding site anion appears to arrange the C-terminal iron-binding residues conducive to complementary binding to Fe(3+), while residues in the N-terminal binding domain must undergo induced fit to accommodate the Fe(3+) ligand. These observations suggest a binding process, the first step of which is the binding of a synergistic anion such as phosphate to the C-terminal domain. Next, iron binds to the preordered half-site on the C-terminal domain. Finally, the presence of iron organizes the N-terminal half-site and closes the interdomain hinge. The use of the synergistic anion and this iron binding process results in an extremely high affinity of the Fe(3+)-binding proteins for Fe(3+) (nFbp K'(eff) = 2.4 x 10(18) M(-1)). This high-affinity ligand binding process is unique among the family of bacterial periplasmic binding proteins and has interesting implications in the mechanism of iron removal from the Fe(3+)-binding proteins during FbpABC-mediated iron transport across the cytoplasmic membrane.     

Functional heterogeneity and pH-dependent dissociation properties of human transferrin.

Human diferric transferrin was partially labeled with 59Fe at low or neutral pH (chemically labeled) and by replacement of diferric iron previously donated to rabbit reticulocytes (biologically labeled). Reticulocyte 59Fe uptake experiments with chemically labeled preparations indicated that iron bound at near neutral pH was more readily incorporated by reticulocytes than iron bound at low pH. The pH-dependent iron dissociation studies of biologically labeled transferrin solutions indicated that Fe3+, bound at the site from which the metal was initially utilized by the cells, dissociated between pH 5.8 and 7.4. In contrast, lower pH (5.2--5.8) was required to effect dissociation of iron that has remained bound to the protein after incubation with reticulocytes. These findings suggest that each human transferrin iron-binding site has different acid-base iron-binding properties which could be related to the observed heterogenic rabbit reticulocyte iron-donating properties of human transferrin and identifies that the near neutral iron-binding site initially surrenders its iron to these cells.     

Drosophila melanogaster transferrin. Cloning, deduced protein sequence, expression during the life cycle, gene localization and up-regulation on bacterial infection.

Drosophila melanogaster transferrin cDNA was cloned from an ovarian cDNA library by using a PCR fragment amplified by two primers designed from other dipteran transferrin sequences. The clone (2035 bp) encodes a protein of 641 amino acids containing a signal peptide of 29 amino acids. Like other insect transferrins, Drosophila transferrin appears to have a functional iron-binding site only in the N-terminal lobe. The C-terminal lobe lacks iron-binding residues found in other transferrins, and has large deletions which make it much smaller than functional C-terminal lobes in other transferrins. In-situ hybridization using a digoxigenin labeled transferrin cDNA probe revealed that the gene is located at position 17B1-2 on the X chromosome. Northern blot analysis showed that transferrin mRNA was present in the larval, pupal and adult stages, but was not detectable in the embryo. Iron supplementation of the diet resulted in lower levels of transferrin mRNA. When adult flies were inoculated with bacteria (Escherichia coli), transferrin mRNA synthesis was markedly increased relative to controls.     

Functional heterogeneity and pH-dependent dissociation properties of human transferrin

Human diferric transferrin was partially labeled with ⁵⁹Fe at low or neutral pH (chemically labeled) and by replacement of diferric iron previously donated to rabbit reticulocytes (biologically labeled). Reticulocyte ⁵⁹ uptake experiments with chemically labeled preparations indicated that iron bound at near neutral ph was more readily incorporated by reticulocytes than iron bound at low pH. The pH-dependent iron dissociation studies of biologically labeled transferrin solutions indicated that Fe³⁺, bound at the site from which the metal was initially utilized by the cells, dissociated between pH 5.8 and 7.4. In contrast, lower pH (5.2–5.8) was required to effect dissociation of iron that had remained bound to the protein after incubation with reticulocytes. These findings suggest that each human transferrin iron-binding site has different acid-base iron-binding properties which could be related to the observed heterogenic rabbit reticulocyte iron-binding properties of human transferrin and identifies that the near neutral iron-donating site initially surrenders its iron to these cells.     

Characterization of monoferric fragments obtained by tryptic cleavage of bovine transferrin.

1. The electrophoretically fast (F) and slow (S) fragments obtained by tryptic cleavage of bovine iron-saturated transferrin differed in carbohydrate content and peptide 'maps'. 2. A fragment capable of binding one Fe3+ ion per molecule was isolated after brief tryptic digestion of bovine apotransferrin and shown closely to resemble the S fragment obtained from the iron-saturated protein. 3. Fragments F and S are probably derived from the N- and C-terminal halves of the transferrin molecule respectively. 4. Bovine transferrin could donate iron to rabbit reticulocytes, but the monoferric fragments possessed little iron-donating ability.     

The preparation and partial characterization of N-terminal and C-terminal iron-binding fragments from rabbit serum transferrin.

Two iron-binding fragments of Mr 36 000 and 33 000 corresponding to the N-terminal domain of rabbit serum transferrin were prepared. One iron-binding fragment of Mr 39 000 corresponding to the C-terminal domain was prepared. The N-terminal amino acid sequence of rabbit serum transferrin is: Val-Thr-Glu-Lys-Thr-Val-Asn-Trp-?-Ala-Val-Ser. One glycan unit is presented in rabbit serum transferrin and it is located in the C-terminal domain.     

High-affinity binding by the periplasmic iron-binding protein from Haemophilus influenzae is required for acquiring iron from transferrin

The periplasmic iron-binding protein, FbpA (ferric-ion-binding protein A), performs an essential role in iron acquisition from transferrin in Haemophilus influenzae. A series of site-directed mutants in the metal-binding amino acids of FbpA were prepared to determine their relative contribution to iron binding and transport. Structural studies demonstrated that the mutant proteins crystallized in an open conformation with the iron atom associated with the C-terminal domain. The iron-binding properties of the mutant proteins were assessed by several assays, including a novel competitive iron-binding assay. The relative ability of the proteins to compete for iron was pH dependent, with a rank order at pH 6.5 of wild-type, Q58L, H9Q>H9A, E57A>Y195A, Y196A. The genes encoding the mutant FbpA were introduced into H. influenzae and the resulting strains varied in the level of ferric citrate required to support growth on iron-limited medium, suggesting a rank order for metal-binding affinities under physiological conditions comparable with the competitive binding assay at pH 6.5 (wild-type=Q58L>H9Q>H9A, E57A>Y195A, Y196A). Growth dependence on human transferrin was only obtained with cells expressing wild-type, Q58L or H9Q FbpAs, proteins with stability constants derived from the competition assay >2.0x10(18) M(-1). These results suggest that a relatively high affinity of iron binding by FbpA is required for removal of iron from transferrin and its transport across the outer membrane.     

Effect of synthetic carrier ampholytes on saturation of human serum transferrin.

We have investigated the effect in solution of synthetic carrier ampholytes on the saturation of human serum transferrin. By spectrophotometric titrations of human serum transferrin with various Fe3+-carrier ampholyte solutions, we demonstrated that under these conditions carrier ampholytes behave as typical chelators, their binding curves being very similar to that obtained with disodium nitrilotriacetate. On performing titration experiments at three different pH values, carrier ampholytes act like nitrilotriacetate at pH 7.5, but the former are more effective iron donors at pH 8.4 and worse iron donors at pH 5.2. Spectrophotometric titrations of isolated C-terminal and N-terminal fragments obtained from human serum transferrin by thermolysin cleavage show no differences between them, and no differences with respect to the whole protein except that they contain half the number of binding sites. In order to determine a site-specificity of iron in the presence of ampholytes, the classical urea/polyacrylamide-gel-electrophoresis technique was adopted. Under saturating conditions carrier ampholyte solutions act mostly on the C-terminal site, whereas desaturating agents remove iron preferentially from the N-terminal site. Our findings support the hypothesis that Ampholine may chelate Fe3+ as well as many other compounds.     

Studies on the role of transferrin and endocytosis in the uptake of Fe3+ from Fe-nitrilotriacetate by mouse duodenum

Addition of iron-binding proteins (human serum transferrin, mouse serum transferrin, human lactoferrin) to the luminal fluid in tied-off segments of mouse intestine in vivo led to reduced 59Fe3+ absorption from 59Fe3+-nitrilotriacetate when compared to 59Fe3+-nitrilotriacetate alone. Assay of transferrin in luminal fluid from tied segments revealed only trace amounts of immunoreactivity. The levels of luminal transferrin are unaltered in chronic hypoxia where iron absorption is significantly enhanced. Studies in vitro revealed that NH4Cl, dansylcadavarine, para-chloromercuribenzoate and trinitrobenzenesulphonate have no effect on initial 59Fe3+ uptake rates from 59Fe3+-nitrilotriacetate, while N-ethylmaleimide (1 mM) caused a 40% inhibition. In vivo 59Fe3+ uptake was unaffected by preincubation of tied-off segments with colchicine (5 mM) for up to 2 h. These results suggest that receptor-mediated endocytosis of transferrin is not a significant mechanism in the uptake of luminal Fe3+ by mouse duodenum.     

Human melanotransferrin (p97) has only one functional iron-binding site.

The iron-binding properties of melanotransferrin, the tumour-associated antigen also known as p97, have been investigated by UV/visible and fluorescence spectroscopy, amino acid sequence comparison, and modelling. These show that, in contrast to other transferrins, melanotransferrin binds only one Fe3+ ion per molecule. The binding properties of its N-terminal site are similar to other transferrins, but its C-terminal site does not bind iron at all. The differences can be related to specific amino acid changes in the C-terminal site.     

Metal substitution in transferrins: the crystal structure of human copper-lactoferrin at 2.1-A resolution.

The structural consequences of binding a metal other than iron to a transferrin have been examined by crystallographic analysis of human copper-lactoferrin, Cu2Lf. X-ray diffraction data were collected from crystals of Cu2Lf, using a diffractometer, to 2.6-A resolution, and oscillation photography on a synchrotron source, to 2.1-A resolution. The structure was refined crystallographically, by restrained least-squares methods, starting with a model based on the isomorphous diferric structure from which the ligands, metal ions, anions, and solvent molecules had been deleted. The final model, comprising 5321 protein atoms (691 residues), 2 Cu2+ ions, 2 (bi)carbonate ions, and 308 solvent molecules has good stereochemistry (rms deviation of bond lengths from standard values of 0.018 A) and gives a crystallographic R value of 0.196 for 43,525 reflections in the range 7.5-2.1-A resolution. The copper coordination is different in the two binding sites. In the N-terminal site, the geometry is square pyramidal, with equatorial bonds to Asp 60, Tyr 192, His 253, and a monodentate anion and a longer apical bond to Tyr 92. In the C-terminal site, the geometry is distorted octahedral, with bonds to Asp 395, Tyr 435, Tyr 528, and His 597 and an asymmetrically bidentate anion. The protein structure is the same as for the diferric protein, Fe2Lf, demonstrating that the closure of the protein domains over the metal is the same in each case irrespective of whether Fe3+ or Cu2+ is bound and that copper could be transported and delivered to cells equally well as iron. The differences in metal coordination are achieved by small movements of the metal ion and anion within each binding site, which do not affect the protein structure.     

Spectroscopic and thermodynamic studies on the binding of gadolinium(III) to human serum transferrin

A wide variety of thermodynamic, kinetic, and spectroscopic studies have demonstrated differences between the two metal-binding sites of transferrin. In the present investigation, we have further assessed these differences with respect to the binding of gadolinium, evaluated by UV difference spectrophotometry, electron paramagnetic resonance (EPR) titration, EPR difference spectroscopy in conjunction with urea gel electrophoresis, and equilibrium dialysis. Combinations of these studies establish that only one site of the protein binds Gd(III) sufficiently firmly to be characterized. In order to reveal which of the two sites accepts Gd(III), we made use of monoferric transferrins preferentially loaded with Fe(III) at either site in EPR spectroscopic studies. Because of the overlap of signals, difference spectroscopy was required to distinguish resonances arising from Fe(III) and Gd(III) specifically complexed to the protein. When iron is bound to the C-terminal site, leaving the N-terminal site free for binding of gadolinium, the difference spectrum shows no evidence of specific binding. However, when iron is bound to the N-terminal site, the difference spectrum shows a resonance line at g' = 4.1 indicative of specific binding, thus implicating the C-terminal site in the binding of Gd(III). The effective stability constant for the binding of Gd(III) to this site of transferrin at pH 7.4 and ambient pCO2 is 6.8 X 10(6) M-1. At physiological pCO2, the formation of nonbinding carbonato complexes of Gd(III) precludes a substantial role for transferrin in the transport of the lanthanide in vivo.     

An iron-dependent and transferrin-mediated cellular uptake pathway for plutonium

Plutonium is a toxic synthetic element with no natural biological function, but it is strongly retained by humans when ingested. Using small-angle X-ray scattering, receptor binding assays and synchrotron X-ray fluorescence microscopy, we find that rat adrenal gland (PC12) cells can acquire plutonium in vitro through the major iron acquisition pathway--receptor-mediated endocytosis of the iron transport protein serum transferrin; however, only one form of the plutonium-transferrin complex is active. Low-resolution solution models of plutonium-loaded transferrins derived from small-angle scattering show that only transferrin with plutonium bound in the protein's C-terminal lobe (C-lobe) and iron bound in the N-terminal lobe (N-lobe) (Pu(C)Fe(N)Tf) adopts the proper conformation for recognition by the transferrin receptor protein. Although the metal-binding site in each lobe contains the same donors in the same configuration and both lobes are similar, the differences between transferrin's two lobes act to restrict, but not eliminate, cellular Pu uptake.     

Functional equivalence of iron bound to human transferrin at low pH or high pH.

Human transferrin was labeled with 59Fe at one of its two metal-binding sites (designated A) at pH 6.0. 55Fe was then added to site B at pH 7.5. Both isotopes of iron were taken up in equal proportions by human reticulocytes. These experiments do not support the hypothesis that each binding site of transferrin has a different physiologic function.     

The synergistic anion-binding sites of human transferrin: chemical and physiological effects of site-directed mutagenesis

A defining feature of all transferrins is the absolute dependence of iron binding on the concomitant binding of a synergistic anion, normally but not necessarily carbonate. Acting as a bridging ligand between iron and protein, it completes the coordination requirements of iron to lock the essential metal in its binding site. To investigate the role of the synergistic anion in the iron-binding and iron-donating properties of human transferrin, a bilobal protein with an iron binding site in each lobe, we have selectively mutated the anion-binding threonine and arginine ligands that form an essential part of the electrostatic and hydrogen-bonding network holding the synergistic anion to the protein. Preservation of either ligand is sufficient to maintain anion binding, and therefore iron binding, in the mutated lobe. Arginine is a stronger ligand than threonine, and its loss weakens carbonate and therefore iron binding, but maintains the ability of nitrilotriacetate to serve as a carbonate surrogate. Replacement of both ligands abolishes anion binding and consequently iron binding in the affected lobe. Loss of anion binding in either lobe results in a monoferric protein binding iron in normal fashion only in the opposite lobe. Both monoferric proteins are capable of transferrin receptor-dependent binding and iron donation to K562 cells, but with diminished receptor occupancy by the protein bearing iron only in the N-lobe.     

Binding of vanadate to human serum transferrin

Human serum transferrin specifically and reversibly binds 2 equiv of vanadate at the two metal-binding sites of the protein. The vanadium(V)-transferrin complex can be formed either by the addition of vanadate to apotransferrin or by the air oxidation of the vanadyl(IV)-transferrin complex. The formation of the vanadium complex can be blocked by loading the apotransferrin with iron(III), and bound vanadium can be displaced from the protein by the subsequent addition of either gallium(III) or iron(III). The binding constant for the second equiv of vanadate is 10(6.5) in 0.1 M hepes, pH 7.4 at 25 degrees C. The binding constant for the first equiv of vanadate is probably very similar, although no quantitative value could be determined. Although transferrin reacts with the vanadate anion, studies on the transferrin model compound ethylenebis(o-hydroxyphenylglycine) indicate that at pH 9.5, the vanadium is binding at the metal-binding site as a dioxovanadium(V) cation coordinated to two phenolic residues at each binding site. This bound cation appears to be protonated over the pH range 9.5-6.5, as shown by changes in the difference uv spectrum of the transferrin complex, to produce an oxohydroxo species. Further decreases in the pH lead to dissociation of the vanadium-transferrin complex.