HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Muralidhara, B. K. Articles by Hirose, M. Search for Related Content PubMed PubMed Citation Articles by Muralidhara, B. K. Articles by Hirose, M. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 17, 12463-12469, April 28, 2000

Anion-mediated Iron Release from Transferrins
THE KINETIC AND MECHANISTIC MODEL FOR N-LOBE OF OVOTRANSFERRIN^*

B. K. Muralidhara and Masaaki Hirose^§

From the Research Institute for Food Science, Kyoto University, Uji, Kyoto 611 0011, Japan

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Iron release process of ovotransferrin N-lobe (N-oTf) to anion/chelators has been resolved using kinetic and mechanistic approach. The iron release kinetics of N-oTf were measured at the endosomal pH of 5.6 with three different anions such as nitrilotriacetate, pyrophosphate, and sulfate using stopped flow spectrofluorimetric method, all yielding clear biphasic progress curves. The two observed rate constants and the corresponding amplitudes obtained from the double exponential curve fit to the biphasic curves varied depending on the type and concentration of anions. Several possible models for the iron release kinetic mechanism were examined on the basis of a newly introduced quantitative equation. Results from the curve fitting analyses were consistent with a dual pathway mechanism that includes the competitive iron release from two different protein states, namely, X and Y, with the respective first order rate constants of K₁ and K₂ (X, domain closed holo N-oTf; Y, anion induced different conformer of holo N-oTf). The reversible interconversions of X to Y and Y to X are driven by the second order rate constant k₃ and the first order rate constant K₄, respectively. The obtained rate constants were greatly variable for the three anions depending on the synergistic or nonsynergistic nature. In the light of the anion-binding sites of N-oTf located crystallographically, the compatible mechanistic model that includes competitive anion binding to the iron coordination sites and to a specific anion site is suggested for the dual pathway iron release mechanism.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Transferrins are a class of bilobal proteins that bind, transport, and release various metals such as iron, aluminum, copper, and bismuth of which the key interest is to study the uptake and release of iron by transferrins because it is abundant and also plays a significant role in many biological processes (1, 2). Each lobe of transferrin binds one Fe³⁺ ion in the interdomain cleft in presence of a synergistic anion (biologically, carbonate) with similar chemistry but with different affinity (3, 4). Although the main core of structure is similar in both the lobes (N and C), the chemistry is remarkably dissimilar toward the interaction of nonsynergistic anions (5), disulfide bridges (6), conformational and thermal stability (7, 8), and iron binding and release characteristics (9-13). Physiologically, low molecular weight chelators/anions play a vital role in the iron exchange reactions with transferrins, and the rate of iron release from the N-lobe of transferrins is more facile compared with its counterpart, either in the full-length molecule or in the independent half molecule (9, 14). The molecular details of origin of such differences are poorly understood. However, a significant difference has been the dilysine trigger in case of N-lobe and kinetically significant anion-binding site in C-lobe, each of which are proposed to be absent in the respective counterparts (13, 15, 16). Addition of simple nonsynergistic anions such as chloride accelerates the chelator mediated iron release from C-lobe but retards release from N-lobe either in full-length structure or in the independent lobe (12, 13, 16, 17). Anion effect was also shown to vary with pH-dependent conformational change in N-hTf where chloride retards iron release at high pH but accelerates at low pH (18). The exact location of nonsynergistic anion-binding site(s) on either lobe is/are not fully known. A recent site-directed mutagenesis study shows that the Lys²⁹⁶ of the dilysine pair, Lys²⁰⁶-Lys²⁹⁶, is one of the anion-binding site residues on N-hTf¹ (19). The mechanism by which anions regulate iron release in both the lobes is not clearly understood.

Considerable interest has been focused to delineate the mechanism of iron release from transferrins in vitro, mainly by using small molecular weight chelators (9, 11-14, 16-20), and, rarely, by using receptors (21-23). Because the iron release mechanism seems to be very complex with diferric transferrin, studies are now being undertaken using either independent lobes or selectively iron-loaded lobe on the full-length transferrin (12-26). The iron-binding site on each lobe of transferrin involves two tyrosine residues, one each of aspartic acid and histidine residues. The remaining two ionic coordinates of iron are shared by a synergistic bicarbonate anion that anchors using an arginine residue of the protein. Hence, recently a number of iron release studies are focused using mutants of these functional sites of transferrin or its isolated lobes to delineate the mechanism (13, 17, 22, 23, 25). The concept of iron release mechanism proposed by Bates and co-workers (27, 28) involving a "mixed ligand" intermediate between "domain closed" holo form and "domain opened" apo form is widely accepted, where the rate-limiting step is the conformational change because of anion binding followed by the rapid removal of bound iron. A recent report emphasizes the slow protonation of iron-binding residues of ovotransferrin for the iron release process at acidic pH in the absence of anionic chelators (29).

Apart from the location of nonsynergistic anion-binding sites, there is a lacuna in the quantitative evaluation of the kinetic data with a suitable model applicable for different anions. The kinetic inequivalence between the two lobes in addition to nonsynergistic anion requirement and inter-lobe interaction makes the process much more complex in diferric transferrins. At large, the kinetic process of iron release was evaluated by using observed rate constant (k_obs), which is obtained by single exponential curve fit analysis of the raw data (11-14, 16-26). Several anion systems have shown saturation kinetics of k_obs values with respect to ligand concentration (9, 17, 20, 25); some have first order kinetics (25, 30), and several others appear to follow both saturation and first order kinetics (12, 25, 26). The saturation kinetics (hyperbolic) obtained with N-hTf showed first order kinetics (linear) for its R124A mutant for the same anion (25). However from the kinetic analysis methods used the origin of such differential behavior was not clear. In this present complexity iron release kinetic mechanism of N-oTf was studied using three anions of different chemical nature: pyrophosphate, a chelator and nonsynergistic anion; NTA, a chelator and synergistic anion; and sulfate, a nonchelator and nonsynergistic anion. We have opted for isolated N-oTf because the solution and crystallographic structural details of iron binding and the intermediate complex of iron uptake are reported from this laboratory (15, 31-33). The kinetic data was obtained using the stopped flow spectrofluorimetric method at the endosomal pH of 5.6 and single anion system at a time, for the iron release process of three different anions. Several kinetic models were analyzed by the newly introduced method to obtain the rate constants for the quantitative evaluation of the iron release kinetics. The crystallographic structure of nonsynergistic anion bound apo N-oTf (33) along with the holo (15) and iron uptake intermediate (32) structures of N-oTf were critically analyzed to derive a mechanistic model compatible with the dual pathway kinetic mechanism.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Preparation of Protein-- N-oTf was prepared by the established procedure (34). The protein concentration was measured in the apo form, and iron-bound holo form was prepared freshly in 100 mM Hepes, pH 7.4, using (Fe³⁺-NTA) complex as the iron donor and a large excess of bicarbonate as the synergistic anion to replace bound NTA. Holo N-oTf was then equilibrated to 50 mM Mes, pH 5.6, by dialysis, and the bound iron was confirmed spectrophotometrically to retain a 1:1 molar ratio with the protein. An N-oTf concentration of 5.5 µM was used in all the iron release experiments.

Preparation of Anions-- Guaranteed grade of chemicals (pyrophosphoric acid from Wako Pure Chemicals; NTA and Na₂SO₄ from Nacalai Tesque), nitric acid-washed glassware and double distilled, deionized, and 40 µM filtered water were used. Extreme care was taken to avoid any metal contamination or alteration of pH in all the solutions. All of the anion stock solutions were prepared freshly just before use in 50 mM Mes, and the pH was adjusted to 5.60 using NaOH and readjusted to 5.60 for minor changes, if any, upon dilution of anion stock solutions.

Iron Release Experiments-- The principle of nonradiative resonance energy transfer from excited aromatic chromophores to the visible charge transfer band of bound iron in transferrin (35), which was successfully used to measure the iron release kinetics in transferrins, was adopted for our studies with minor modifications (12, 21). The kinetic data was precisely obtained by 1:1 sample mixing on a SX.18MV stopped flow reaction analyzer (Applied Photophysics Ltd., UK), which has the dead time of 4 ms. The excitation and emission wavelengths, 288 and 330 nm, respectively, were set on two different monochromators with a bandpass of 9.3 nm. A photomultiplier voltage of 600 mV with the reaction temperature of 30 ± 0.1 °C maintained using an external water bath and 50 mM Mes, pH 5.6, because the reaction buffer did not contribute to iron release from holo N-oTf in the absence of added anions used in the present study. The instrument conditions applied did not denature either apo or holo N-oTf even after 500 s of exposure, which was confirmed by the respective fluorescence emission properties. The data were collected using "oversampling mode," which allows fast and accurate pickup and averaging of data points with improved signal-to-noise ratio over the extended time of measurements. A minimum of 1000 data points were collected per trace on a time scale of 0-500 s, and at least three identical traces were averaged for each anion concentration. The averaged traces were best fit using the Levenberg-Marquardt algorithm to the following double exponential curve fit equation using the built-in software (version 4.36) on an Acorn Risc computer interfaced to the main instrument.

(Eq. 1)

Z(t) is the fraction of iron-released (apo) form as detected by the change in fluorescence emission signal at time t, and A₁ and A₂ are the fractions of the fluorescence amplitudes (A₁+ A₂ = 1) for the observed rate constants r₁ and r₂ (r₁ > r₂), respectively. For the curve fit of raw data of all the three anions at different concentrations the normalized variance was between 6 × 10⁷ and 6 × 10⁸. The data derived from this primary analysis (r₁, r₂, A₁, and A₂) were plotted as a function of anion concentrations, and the curve fitting analyses by Equations 3-8 (see "Results") were carried out using the Levenberg-Marquardt algorithm (KaleidaGraph, Synergy Software).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Biphasic Time Course for the Anion-dependent Iron Release Process-- Iron release kinetics of N-oTf were studied using three anions of different chemical nature: NTA, which can act both as a chelator and a synergistic anion; pyrophosphate, a strong nonsynergistic chelator anion; and sulfate, a simple nonsynergistic anion that can form a coordination complex with iron. In Fig. 1 is shown a typical example of a clear biphasic progress curve of iron release from N-oTf to 100 mM NTA (A), 10 mM pyrophosphate (B), and 200 mM sulfate (C). The profiles can be demarcated by an initial rapid step followed by a slow hyperbolic step, both of which vary with the type and concentration of anion. For NTA the initial rapid process is significantly slow compared with the other two anions, and to emphasize the initial rapid phase the data within 20 s is shown as respective insets in Fig. 1. The initial rapid process is very dominant in the case of pyrophosphate even below 5 mM concentrations but did not induce complete iron removal within the time frame of 500 s used in all the experiments. A similar observation was made in the case of sulfate, but in the case of NTA the initial rapid process was not clearly distinguishable below 10 mM concentration. For data analysis the anion concentrations that induce complete iron release within 500 s were considered. Sulfate and NTA below 20 mM and pyrophosphate below 10 mM did not induce complete iron removal.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Biphasic kinetics. Typical biphasic time course data of the iron release process of N-oTf in 50 mM Mes, pH 5.6, containing 100 mM NTA (A; for clarity every tenth data point is plotted), 10 mM pyrophosphate (B; for clarity every second point is plotted), and 200 mM sulfate (C; for clarity every second point is plotted). The relative fluorescence emission signal (RFES) was monitored by stopped flow fluorometry. The data were fit to a double exponential curve fit equation (Equation 1), and the best fit is represented by the solid line. The initial burst phase pattern of the corresponding anion concentrations are shown as respective insets and plotted between 0 and 20 s for clear comparison.

Biphasic progress curves of complete iron release from single iron-binding site of transferrin, either in the isolated N-lobe or in the selectively iron loaded N-lobe of full-length transferrin, have been observed for the first time with N-oTf. Iron release studies either with N-hTF or hTf-Fe_N were performed at different pH values, and only monophasic progress curves were obtained and curve fit to single exponential (16-20, 22-26). Iron release from C-lobe of ovotransferrin at pH 5.6 showed monophasic progress curves with different types and concentrations of anions under identical experimental conditions used for N-oTf.² The monophasic progress curves at acidic pH were also observed with hTf-Fe_C (11, 13, 16).

Effect of Anion Concentration on the Observed Rate Constants-- In Fig. 1, the solid line represents the best fit line to Equation 1 in all the three anion cases. The normalized variance for all the double exponential curve fits were in the order of 10⁷ to 10⁸, indicating the excellent fit. The two observed rate constants (r₁ and r₂) obtained from the curve fit are plotted as a function of NTA (A), pyrophosphate (B), and sulfate concentration (C) in Fig. 2. Only anion concentrations that yield complete iron removal were considered for analysis, and in the case of sulfate, higher concentrations were required to achieve complete iron removal. The larger rate constants (r₁) are increased almost linearly with anion concentration in all the three cases, and they are about one order lower in case of NTA compared with sulfate and pyrophosphate, at identical concentrations, although the values for pyrophosphate are more than twice compared with sulfate. The smaller rate constants (r₂) did not follow a linear relationship with anion concentration; instead, they follow an apparent hyperbolic pattern in all the three anion cases. The values are of the same order for NTA and sulfate but are of about five times smaller compared with pyrophosphate.

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Curve fitting analysis using observed rate constants. The variation of observed rate constants r1 (a panels) and r2 (b panels), obtained from double exponential curve fit of the raw data to Equation 1, as a function of NTA (A), pyrophosphate (B), and sulfate (C) concentration. The r1 and r2 values were curve fit by nonlinear least square method to Equations 3-6. The best fit is represented by the dashed line, and the corresponding correlation coefficient (R) is also shown with in each plot.

The fractional amplitude data was obtained from the double exponential curve fit of the biphasic raw data to Equation 1 yielding two amplitudes, A₁ and A₂ for the observed rate constants r₁ and r₂, respectively. In Fig. 3 is shown the pattern of variation of fractional amplitude as a function of NTA (A) and pyrophosphate (B) concentrations. In both the cases the amplitude of larger rate constant (A₁) decreases, whereas the amplitude for smaller rate constant (A₂) increases with anion concentration. Fractional amplitudes at lower sulfate concentration did not follow a specific pattern, and at concentrations above 50 mM both of the amplitudes were almost identical, although the trend of A₁ decreasing and A₂ increasing with anion concentration was maintained (data not shown). The standard error for observed rate constants and amplitudes obtained from the curve fit to Equation 1 were less than 1 and 0.5% of the corresponding values, respectively.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Curve fitting analysis using amplitude data. Fractional amplitudes A1 and A2 obtained from double exponential curve fits of the raw fluorescence data to Equation 1 are plotted as functions of NTA (A) and pyrophosphate (B) concentration. Both A1 and A2 were curve fit to Equations 7 and 8, respectively, and the best fit is represented by the dashed line, yielding essentially the same curve fit results. The corresponding correlation coefficients (R) are also shown. The errors of the data obtained from amplitude curve fit analysis (Table II) highlight the sensitivity of the analysis because of the complexity in the curve fit equation for amplitude.

Model Presentation for Biphasic Kinetics and Data Evaluation-- Biphasic iron release process from N-oTf can be generally represented as shown in Scheme 1. 

where X represents the carbonate bound, domain closed holo-form and Y represents a protein state with the bound carbonate and iron but in different conformational state from X. The reversible interconversion of X and Y are driven by the first order rate constants, K₃ and K₄. Iron is released from X and Y with the respective first order rate constants of K₁ and K₂, giving rise to a common iron-free (apo) form represented by Z. Z(t), which is defined as the fraction of Z at an iron releasing time t, can be expressed by the following double exponential equation.

(Eq. 2)

where X₀ and Y₀ (X₀ + Y₀ = 1) are the fractions of X and Y at the initial time (t = 0), respectively. The observed rate constants, r₁ and r₂, are related to the first order rate constants K₁ to K₄ as shown in the following equations.

(Eq. 3)

(Eq. 4)

(Eq. 5)

(Eq. 6)

The amplitudes A₁ and A₂ in Equation 1 can be related to the amplitude terms in Equation 2.

(Eq. 7)

(Eq. 8)

A detailed analysis was done to find out whether each of the rate constants K₁ to K₄ is a true first order rate constant or a pseudo-first order rate constant that corresponds to the product of second order rate constant (k₁, k₂, k₃, or k₄) and anion concentration (e.g. if K₁ is a pseudo-first order rate constant, then K₁ = k₁ × [anion]). As a primary search, we plotted 2 and  as a function of anion concentrations, using the observed rate constants r₁ and r₂ obtained in Fig. 2 and the relations 2 = r₁ + r₂ and  = r₁ × r₂. In Fig. 4 is shown that both  (b panels) and 2 (a panels) values clearly follow a linear relationship of non-zero positive slope with NTA (A), pyrophosphate (B), and sulfate (C) concentrations, which is indicated by the linear least square fit line (solid line) and the respective correlation coefficient (R). The linear relation with a non-zero slope for the 2 plot implies, on the basis of Equation 5, that at least one of the rate constants is a pseudo-first order rate constant. Equation 6 includes the product terms of two first order rate constants (K₁K₂, K₁K₄, and K₂ K₃); for the cases that the two rate constants in combination (K₁ and K₂, K₁ and K₄, or K₂ and K₃) are both pseudo-first order rate constants, Equation 6 results in the inclusion of square term(s) for anion concentration. These cases can therefore be excluded according to the linear relation for the  plot (Fig. 4). Consequently, on the basis of the linear relations in Fig. 4, the possible iron release mechanisms can be restricted to six models only as depicted in Table I. The models include both single pathway and dual pathway mechanisms along with pre-equilibrium or anion-dependent equilibrium for the conversion of X to Y. 

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Linear relationship of 2 and  values. The sum (2, a panels) and the product (, b panels) of the observed rate constants r1 and r2 obtained from double exponential curve fit of the raw data to Equation 1 are shown as a function of NTA (A), pyrophosphate (B), and sulfate (C) concentration. The 2 and  plots were curve fit by linear least square method, and the best fit is represented by the solid line. The corresponding correlation coefficient (R) for each linear fit is shown.

We examined the six models by curve fitting analyses using Equations 3-6 for the observed rate constants (Fig. 2) and using Equations 7 and 8 for the fractional amplitudes (Fig. 3). Getting reliable and reproducible data is, however, almost impossible by using as many as four unknown parameters (K₁ to K₄). The number of unknown parameters were minimized to two by utilizing the additional advantage of 2 and  plots in Fig. 4 with the relations described under situation for 2 and  plots in Table I. For example, K₂ and K₃ can be directly obtained from the two slopes for the models 1, 4, and 6, hence only K₁ and K₄ remain as the unknown parameters. For the other models, some of the four parameters can be related with each other using the data of two slopes, and hence two parameters can be treated as unknown.

The r₁ and r₂ (Fig. 2) and amplitude (Fig. 3) data were evaluated by curve fitting analyses using Equations 3-8 for refining the model of iron release. Data did not fit well for the single pathway mechanism in the pre-equilibrium state as shown in model 4 where the iron release can occur only through Y form. Models 5 and 6 involve the anion-dependent state of equilibrium for iron release; the curve fit for both models yielded very poor results. This confirms the unlikely single pathway mechanism of iron release for N-oTf.

In the dual pathway mechanism of iron release there is one possibility for pre-equilibrium process (presence of both X₀ and Y₀ at t = 0) as represented by model 3. Good fits were obtained for r₁ and r₂ plots but did not yield comparable values of K₁ and K₄ between the two plots within the limits of the error bar, and fit for amplitude data yielded larger error. Models 2 and 1 are the anion-dependent equilibrium cases (X₀ = 1 and Y₀ = 0 at t = 0) where curve fit for model 2 did not yield good results. The data were best fit to model 1 for all the three plots (r₁, r₂, and amplitude). In Fig. 2 is shown the plots of observed rate constants; r₁ (a panels) and r₂ (b panels) for the three anions that are curve fit to Equations 3-6, and the best fit is represented by the dashed line. The curve fit for r₁ followed an apparently linear relation with concentrations of all the three anion, whereas for r₂ there is a clear hyperbolic pattern seen. The correlation coefficients (R) for curve fitting analyses were excellent (shown within each plot), and moreover the values obtained for K₁ and K₄ from the two plots were remarkably consistent with minimum errors. This was the case irrespective of the type of anion used, although a slightly larger error was obtained for sulfate case. The curve fit for amplitude data of NTA and pyrophosphate is shown by the dashed line in Fig. 3, which yielded agreeable values in comparison with the values obtained from the respective r₁ and r₂ plots. The summary of all the rate constants obtained for the model 1 from the present analysis is shown in Table II. Considering the complexity of the curve fit equation for amplitude (Equations 7 and 8), the minor errors in the r₁ and r₂ values are amplified in the analysis, and hence there is deviation in the curve fit values from the other two plots. However, amplitude curve fit analysis provides a good reevaluation method for the values obtained from r₁ and r₂ plots.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Addition of nonsynergistic anion such as chloride was shown to significantly effect the iron release kinetics from both the lobes of transferrin (36). Anions bind strongly but differentially to both the lobes of the apo structure of transferrins (5, 37-40). With respect to the synergistic anion-binding site, a second class of anion-binding sites was proposed to be involved in inducing the conformational change, which is the rate-limiting step in the iron release pathway proposed by Bates and co-workers (27, 28). The prerequisite nature of nonsynergistic anion binding for domain opening and subsequent iron release was clearly shown in case of hTf-Fe_N (41), hTf-Fe_C (16, 41), and diferric hTf (24). The speculated location of nonsynergistic anion-binding site on hTf-Fe_C was termed as kinetically significant anion-binding site (42), and Lys⁵⁶⁹ was suggested to act as one such possible site (13). One of the recent report using site-directed mutagenesis approach showed that Lys²⁹⁶ of the dilysine pair (Lys²⁰⁶-Lys²⁹⁶) was one of the anion-binding residues of N-hTf (19). Recently, 1.9 Å resolution crystallographic structure of the sulfate anion-binding sites on N-oTf was determined in this laboratory, and two anion-binding sites (sites 1 and 2) in the interdamain cleft were shown to play key roles in the domain opening and synergistic carbonate anion release mechanism (33). In brief, the groups involved in the binding of anion are Ser⁹¹-OG and His²⁵⁰-NE at site 1 and Arg¹²¹-NE, Arg¹²¹-NH2, and Ser¹²²-N at site 2, and from the perspective of an iron-binding/release mechanism we name these two sites as regulatory anion-binding sites.

For N-oTf, the crystallographic data of three different structural states are now available. The structures are the holo form (15), an intermediate form (32), and the sulfate anion bound apo form (33). The apo crystal was soaked in NTA-Fe³⁺ solution complex, which assumes an open structure but with bound iron and NTA to form an intermediate form in the iron uptake/release process (32). The iron atom is held by the coordination of two protein ligands of Tyr⁹²-OH and Tyr¹⁹¹-OH, and the other iron coordination sites are shared by direct interaction of NTA anion. In this coordinating intermediate complex, the regulatory anion-binding site groups make unique interactions as summarized in Table III. As an important observation for the coordinating intermediate structure, site 1 is still occupied by sulfate anion, but no sulfate anion exists on site 2, despite free situation of Arg¹²¹-NE. This may be due, at least in part, to the occupation of Ser¹²²-N by NTA-O4. Although the apo structure with saturated NTA in the absence of Fe³⁺ is not available at present, the absence of SO₄² binding in site 2 for the intermediate form strongly suggests that this complex does not receive another NTA in site 2. A putative N-oTf-Fe³⁺-NTA noncoordinating complex in which the anion occupies site 2 without coordinating iron would not receive the iron-coordination attack by another NTA. This incompatible situation of anion for the iron coordination and for the site 2 occupation may be the structural basis for the dual pathway kinetic mechanism of iron release.

Fig. 5 displays the feasible mechanistic scheme compatible with the kinetic data for the dual pathway mechanism of iron-release from N-oTf. The holo form with domain closed conformation X is transformed into putative CO₃²- and iron-loaded open form Y cooperatively by an anion (A) binding to each regulatory anion-binding site. This form corresponds to the mixed ligand intermediate, a quaternary complex of (A·N-oTf·Fe³⁺·CO₃²) with open-like structure similar to the one proposed by Bates and co-workers for full-length transferrin on the basis of reversible iron uptake analysis using anionic chelators (27, 28). The structural states X and Y interconvert reversibly, and the equilibrium is driven by anion concentration. State Y is then transformed into the noncoordinating intermediate Y₁ by carbonate anion release. The state X also undergoes alternative domain opening by anion binding to site 1 and direct anion coordination to iron. The reaction accompanies carbonate anion release and yields the coordinating intermediate X₁, which corresponds to the intermediate form (Table III). In the parallel intermediate Y₁ both Arg¹²¹-NE and Ser¹²²-N of site 2 are occupied by anion without coordinating iron. The structural state X₁ by no means can receive anion at site 2; similarly, the state Y₁ cannot have direct iron coordination by another anion. These states then undergo iron release by the attack of H₂O molecules (33) to Tyr⁹²-OH and Tyr¹⁹¹-OH yielding the common apo form, Z. The iron release through coordinating intermediate X₁ corresponds to the kinetic K₁ pathway and that through noncoordinating intermediate Y₁ to the K₂ pathway. The absence of the anion concentration dependence of K₂ (Table II) is consistent with the fact that no anion uptake is involved in the Y to Y₁ and Y₁ to Z conversions. For K₁ pathway, we hypothesize that X₁ to Z conversion is the rate-limiting step. This hypothesis comes from the observation of large variations of K₁ value with the different chemical nature of anions used. When two chelator anions are compared, the K₁ value of pyrophosphate is about 140 times greater than that of NTA (Table II), which can be accounted for by the nonsynergistic nature of the farmer and the synergistic nature of NTA. A synergistic anion should form, through anion to protein group interactions as demonstrated by the structural evidence (32), a stable coordinating intermediate X₁ from which iron should be released with a decreased rate. It is therefore very likely that the overall rate for K₁ pathway largely depends on the rate of X₁ to Z conversion in which no anion uptake is involved.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Schematic representation of the feasible mechanistic model for the dual pathway kinetics of iron release from N-oTf. In the structural representation, iron coordinating bonds are shown by arrows, and the other bonds are shown by dotted lines. Domain opening is depicted similar to a stereo representation and hence should not assumed to be restricted only to the movement of domain 1. The structural transformations from X to Z and Y to Z are kinetically single step precesses with the respective first order rate constants K1 and K2, and the intermediate states X1 and Y1 are assigned to the mechanistic model only. A represents anion and other details are described in the text.

In conclusion, iron release kinetics of N-oTf was studied at the endosomal pH using NTA, pyrophosphate, and sulfate anions, and in all the three cases clear biphasic kinetics were observed compared with monophasic kinetics of N-hTf and hTF-Fe_N (16-20, 22-26). The new kinetic approach derived allows critical evaluation of the process using first order rate constants, which differ with the chemical nature of anions used. The proposed model better explains the anion-dependent formation of open-like structure from which iron release occurs by dual pathway mechanism. The two anion-binding sites located in the apo form (33) and their coordinating structural situations in the holo (15) and intermediate forms (32) are successfully used to derive the mechanistic model to support and substantiate the dual pathway kinetic model for N-oTf.

	ACKNOWLEDGEMENTS

We thank Prof. Tokuji Ikeda and Dr. Shuji Adachi (Kyoto University) and Dr. Eizo Tatsumi (Japan International Research Center for Agricultural Sciences) for helpful discussions about kinetic analysis.

	FOOTNOTES

^* This work was supported in part by a grant-in-aid for scientific research from the Ministry of Education, Science, Sports, and Culture of Japan.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Recipient of a Postdoctoral Fellowship for Foreign Researchers of the Japan Society for the Promotion of Science. Present address: Dept. of Protein Chemistry and Technology, Central Food Technological Research Institute, Mysore 570013, India.

^§ To whom correspondence should be addressed: Research Inst. for Food Science, Kyoto University, Uji, Kyoto 611 0011, Japan. Tel.: 81-774-38-3734; Fax: 81-774-38-3735.

² B. K. Muralidhara and M. Hirose, manuscript in preparation.

	ABBREVIATIONS

The abbreviations used are: N-hTf, N-lobe of human serum transferrin; N-oTf, N-lobe of ovotransferrin; hTf-Fe_C, C-terminal monoferric human serum transferrin; hTf-Fe_N, N-terminal monoferric human serum transferrin; NTA, nitrilotriacetate; hTf, human serum transferrin; Mes, 2-(N-morpholino)ethanesulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				T. E. Adams, A. B. Mason, Q.-Y. He, P. J. Halbrooks, S. K. Briggs, V. C. Smith, R. T. A. MacGillivray, and S. J. Everse The Position of Arginine 124 Controls the Rate of Iron Release from the N-lobe of Human Serum Transferrin. A STRUCTURAL STUDY J. Biol. Chem., February 14, 2003; 278(8): 6027 - 6033. [Abstract] [Full Text] [PDF]

				K. Mizutani, B. K. Muralidhara, H. Yamashita, S. Tabata, B. Mikami, and M. Hirose Anion-mediated Fe3+ Release Mechanism in Ovotransferrin C-lobe. A STRUCTURALLY IDENTIFIED SO42- BINDING SITE AND ITS IMPLICATIONS FOR THE KINETIC PATHWAY J. Biol. Chem., September 14, 2001; 276(38): 35940 - 35946. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Muralidhara, B. K. Articles by Hirose, M. Search for Related Content PubMed PubMed Citation Articles by Muralidhara, B. K. Articles by Hirose, M. Social Bookmarking                      What's this?