The Unique Kinetics of Iron Release from Transferrin: The Role of Receptor, Lobe–Lobe Interactions, and Salt at Endosomal pH

Abstract

Transferrins are a family of bilobal iron-binding proteins that play the crucial role of binding ferric iron and keeping it in solution, thereby controlling the levels of this important metal. Human serum transferrin (hTF) carries one iron in each of two similar lobes. Understanding the detailed mechanism of iron release from each lobe of hTF during receptor-mediated endocytosis has been extremely challenging because of the active participation of the transferrin receptor (TFR), salt, a chelator, lobe–lobe interactions, and the low pH within the endosome. Our use of authentic monoferric hTF (unable to bind iron in one lobe) or diferric hTF (with iron locked in one lobe) provided distinct kinetic end points, allowing us to bypass many of the previous difficulties. The capture and unambiguous assignment of all kinetic events associated with iron release by stopped-flow spectrofluorimetry, in the presence and in the absence of the TFR, unequivocally establish the decisive role of the TFR in promoting efficient and balanced iron release from both lobes of hTF during one endocytic cycle. For the first time, the four microscopic rate constants required to accurately describe the kinetics of iron removal are reported for hTF with and without the TFR. Specifically, at pH 5.6, the TFR enhances the rate of iron release from the C-lobe (7-fold to 11-fold) and slows the rate of iron release from the N-lobe (6-fold to 15-fold), making them more equivalent and producing an increase in the net rate of iron removal from Fe₂hTF. Calculated cooperativity factors, in addition to plots of time-dependent species distributions in the absence and in the presence of the TFR, clearly illustrate the differences. Accurate rate constants for the pH and salt-induced conformational changes in each lobe precisely delineate how delivery of iron within the physiologically relevant time frame of 2 min might be accomplished.

Introduction

Human serum transferrin (hTF)¹ is a bilobal ferric-iron-binding glycoprotein. The nearly homologous N-lobe and C-lobe are connected by a short peptide linker and are further divided into two subdomains (NI/NII and CI/CII). The subdomains come together to form an iron binding cleft within each lobe.1, 2 Diferric hTF preferentially binds to specific transferrin receptors (TFRs) on the cell surface at neutral pH.³ The complex undergoes clathrin-dependent receptor-mediated endocytosis during which the clathrin-coated pit fuses with an endocytic vesicle. The pH within the endosome is lowered to ∼ 5.6, resulting in protonation of the synergistic carbonate anion and iron binding residues, which, in turn, loosens the cleft and facilitates iron release to an as yet unidentified chelator. At the low pH within the endosome, iron-free transferrin (apohTF) remains bound to the TFR and is recycled back to the cell surface. Upon exposure to the pH of serum (∼ 7.4), the complex dissociates, and released apohTF is free to bind more iron and to repeat the cycle. Entry of hTF into the cell, removal of iron from hTF, and return to the surface are completed in ∼ 2–3 min.4, 5 Because ferrous iron is transported out of the endosome by a divalent metal transporter, DMT1, there is a requirement for reduction of ferric iron (Fe³⁺).⁶ Discovery of a ferrireductase (Steap3) residing in the membrane of the endosome provided a means for accomplishing the reduction.⁷ While the TFR is known to influence the redox potential of iron bound to hTF,⁸ the weight of evidence indicates that Fe³⁺ is first released from diferric hTF and is then reduced by Steap3.⁹

Ferric iron is coordinated in a near-octahedral geometry by identical ligands in each lobe of hTF: two tyrosines, one histidine, one aspartic acid, and two oxygens from the synergistic carbonate anion, which, in turn, is anchored to a highly conserved arginine residue.¹⁰ Although the iron binding ligands are identical, the precise steps leading to iron release from each lobe differ due largely to differences in “second-shell” residues that do not directly coordinate the iron but participate in an extended and intricate hydrogen-bonding network with the primary ligands.11, 12, 13

Two lysine residues lie on opposite sides of the iron binding cleft—Lys206 in the NII subdomain and Lys296 in the NI subdomain—and are 3.04 Å apart in the iron-bound isolated hTF N-lobe structure and 9 Å apart in the apo structure of this lobe; these residues comprise the “dilysine trigger.”12, 13, 14 They share a hydrogen bond at neutral pH that is protonated at low pH and literally triggers the opening of the cleft. In the C-lobe, Lys534 (in the CII subdomain) and Arg632 (in the CI subdomain) are found in positions homologous to those of Lys206 and Lys296, respectively.¹⁴ Mutation of Lys206 to glutamate in the N-lobe or mutation of Arg632 to alanine in the C-lobe to form Lock_NhTF [recombinant diferric hTF that contains an N-terminal hexa-His tag and is nonglycosylated (mutation K206E locks iron in the N-lobe)] and Lock_ChTF [recombinant diferric hTF that contains an N-terminal hexa-His tag and is nonglycosylated (mutation R632A locks iron in the C-lobe)] constructs, respectively, completely prevents iron release from that lobe on a relevant timescale and allows targeted measurement of iron release from the opposite lobe.15, 16, 17

It is well established that the presence of salt affects the iron release properties of each lobe of hTF.¹⁸ In fact, iron release requires binding of a nonchelating anion, such as Cl⁻, to an anion binding site that is distinct from the synergistic anion binding site. Specifically, it has been suggested that residues termed KISAB (kinetically significant anion binding) sites exist in each lobe of hTF.¹⁹ To exert an effect, such anions must bind to a site (or to multiple sites) on the iron-loaded closed form of hTF. Ideally, the anion binding effect should also be pH sensitive; at neutral pH, it would exert a negative or retarding effect on iron release because it is highly desirable to retain iron until delivery within the endosome. Once iron is removed, anions may play a different role in which they bind to and stabilize the open conformation. This suggestion is substantiated by identification of sulfate binding sites in the N-lobe that are inaccessible in the iron-bound N-lobe; therefore, anions cannot exert any effect on them until the iron is removed.20, 21

The rate of iron release from hTF can be measured by an increase in the intrinsic Trp fluorescence (with a small contribution from Tyr residues) that occurs upon iron removal. hTF has eight Trp residues: three in the N-lobe and five in the C-lobe. Ferric iron within each binding cleft strongly quenches Trp fluorescence through radiationless transfer of electronic excited-state energy.²² This energy is transferred via a Tyr-to-Fe³⁺ charge transfer absorption band at 470 nm,²³ which overlaps the Trp fluorescence emission band. Additionally, the charge transfer band results in a disruption of the π-to-π⁎ transition energy of the liganding Tyr residues, leading to an increase in the UV absorbance overlapping the intrinsic Trp fluorescence.²² The decrease in absorbance (at 470 nm) or the increase in fluorescence signal has been utilized to derive rate constants associated with the iron release process. The recovery of the intrinsic fluorescence signal from Trp (and, to a much smaller extent, Tyr) can be monitored as iron is removed from hTF. Additionally, the large conformational changes associated with iron removal impact specific Trp residues that are very sensitive to alterations in their local environment.24, 25 Thus, the increase in the intrinsic Trp signal is ascribed to a combination of unquenching by loss of iron, triggering the large conformational changes in hTF and more localized changes in the immediate environment of the Trp residues. Recent studies from our laboratory have determined the contributions of the individual Trp residues in each lobe to the iron release signal,26, 27 with no contribution from the 22 Trp residues in the dimer of the soluble portion of the transferrin receptor (sTFR; residues 121–760, expressed as a recombinant entity that contains an N-terminal hexa-His tag).²⁸

Early studies by Bali et al.²⁹ and Bali and Aisen30, 31 provided the first insights into the mechanistic role of TFR in iron removal from diferric hTF. A time-based steady-state fluorescence approach to monitoring iron release using the increase in the intrinsic Trp fluorescence from hTF was pioneered by the Aisen laboratory.³² In their series of kinetic studies, iron release to the chelator pyrophosphate was measured with monoferric N-lobe and monoferric C-lobe, diferric, and mixed-metal transferrins, with kinetically inert Co³⁺ introduced into one lobe and with Fe³⁺ introduced into the other lobe of the protein. Experiments were performed in the presence or in the absence of full-length TFR isolated from placenta and solubilized at pH 5.6 using detergent micelles. Despite the technical challenges of this work (assuring that the metal was in the assigned lobe and remained there during the experiment, the low yield of TFR from the placenta, the requirement of detergent for its solubilization, and the instability of TFR at pH 5.6), the authors were able to conclude that, in the absence of the TFR, iron is released from the N-lobe, followed by the C-lobe, and that binding to the TFR induced a switch in this order.29, 30, 31 Iron release from both lobes was observed to occur at comparable rates on the seconds timescale.

More recently, el Hage Chahine and Pakdaman³³ and Hemadi et al.³⁴ carried out pH jump chemical relaxation studies at 4.3 ≤ pH ≤ 6.5 in which iron was removed from diferric and monoferric C-lobe transferrins using acetate as competing ligand in the presence and in the absence of detergent-solubilized TFR from placenta. Contrary to the findings of the Aisen laboratory, it was concluded that iron is preferentially removed from the N-lobe in both instances. Moreover, in the presence of the receptor, a rapid kinetic event on a milliseconds timescale was assigned to removal of iron from the N-lobe, whereas a much slower kinetic event on a seconds timescale was assigned to removal of iron from the C-lobe.

In the present work, we have addressed these issues by exploiting recombinant technology, including site-directed mutagenesis, to produce Fe₂hTF (recombinant diferric hTF that contains an N-terminal hexa-Histag and is nonglycosylated), authentic monoferric constructs {in which iron can bind in only one lobe or the other: Fe_NhTF [recombinant monoferric N-lobe hTF (mutations Y426F and Y517F preclude iron binding in the C-lobe) that contains an N-terminal hexa-His tag and is nonglycosylated] and Fe_ChTF [recombinant monoferric C-lobe hTF (mutations Y95F and Y188F preclude iron binding in the N-lobe) that contains an N-terminal hexa-His tag and is nonglycosylated]} and diferric locked constructs (in which iron can be removed from only one lobe or the other: Lock_NhTF and Lock_ChTF), as well as the sTFR (eliminating the need for detergent), to allow an unambiguous assignment of events related to iron release from hTF. The use of a sensitive stopped-flow spectrofluorimeter has provided data with a high signal-to-noise ratio, allowing the observation of early kinetic events not previously detected using the less sensitive steady-state format. Precise fitting of progress curves was achieved with equations describing the kinetic processes occurring during iron removal. In addition to iron release, we were able to assign rate constants to conformational changes within the individual lobes of hTF, to interactions with the sTFR, and to salt effects. Building on our recent qualitative study of iron release from these constructs and a model presented for iron release from Fe₂hTF in the absence of the sTFR,¹⁷ we now present a comprehensive model for iron release from Fe₂hTF in the presence of the sTFR that more fully describes this complicated system. We provide accurate rate constants and irrefutable evidence that a critical role of the sTFR is to balance the rates of iron release from both lobes so that removal from Fe₂hTF occurs efficiently during one cycle of endocytosis. We also offer a compelling argument for why hTF is bilobal. Although our findings are in general accord with the early studies of Bali et al.²⁹ and Bali and Aisen,³⁰,³¹ they are more comprehensive and provide valuable new insights into this complex system.