Rt

where is the redox potential of the mediator, is the redox potential of the protein, and m is the number of electrons transferred for the mediator.

4. The mediator should be spectroscopically silent in the region of interest for the redox protein, or at least not create interferences with the protein absorbance spectrum in the oxidized or reduced form.

5. The mediator should not interact with the protein in a way that influences the protein redox behavior. (For highly allosteric proteins such as hemoglobin, care must be taken as redox behavior can often be mediator dependent.)

Compilations of mediators for protein electrochemistry are available.68'83 We have used hexaammineruthenium(II/III) chloride, Ru(NH3)6C13, as a mediator for heme protein studies.1'3'35'69'84 This mediator satisfies all the criteria outlined above, and we have demonstrated that it does not influence the redox properties of myoglobin or hemoglobin. Furthermore, its positive charge prevents interaction at the allosteric site in the hemoglobin tetramer. Hemoglobin is known to interact with various anions that can influence both the oxygenation and oxidation of the protein, whereas the protein is rather insensitive to cations.2'4 The small quantity of chloride ion due to the counterion associated with Ru(NH3)g+/3+ was shown not to influence the redox potential of most Hbs (±2 mV for a 10-fold increase in mediator-to-iron ratio).3'4'69

Although we have shown that the presence of the mediator facilitates the protein-electrode electron exchange and allows oxidation-reduction of the Hbs and Mbs without influencing their structure and integrity,3 the kinetics and mechanism of electron exchange between hemoglobin and Ru(NH3)g+/3+ have not been fully described at this point. However, it is clear that because of its large size the ruthenium complex does not enter the heme pocket.

Detection System

Various detection systems can in principle be used to investigate the redox properties of proteins. Although we are describing a UV-Vis spectroscopy-based detection system here, infrared (IR) and emission spectroscopic detection can be developed on the same principle. Extensive reviews have been published on infrared, Raman, transmission, UV-Vis, and EXAFS (extended X-ray absorption fine structure)-based spectroelectrochemistry.5'81'82,84'86-93 Because both Mbs and Hbs have readily available UV-Vis spectral properties that differ upon oxidation-reduction of the prosthetic groups, we have based our technique on UV-Vis spectroscopic detection.

Specific Examples

We illustrate the principles of anaerobic spectroelectrochemistry using an OTTLE cell by describing a typical experiment used to investigate several myoglobins. In all experiments, the changes in absorbance at one or two wavelengths associated with the oxidized and reduced species resulting from altered electrode potentials are recorded. The absorbance changes indicate changes in the concentrations of the oxidized and reduced species and are used to create a plot according to the Nernst equation [Eq. (1)]:

where Eapp is the applied potential; £1/2 is the midpoint potential, at which 50% of the protein is oxidized and 50% is reduced under our specific conditions; R is the gas constant (J K-1 moF1); 7 is temperature in degrees Kelvin; n indicates the number of electrons involved in the redox process for an ideal system with nernstian behavior; F is the Faraday constant; and [Ox] and [Red] are the concentrations of oxidized and reduced species (M). The details associated with data analysis are described in Data Analysis (below).

Heme Proteins with Nernstian Response

Myoglobins

Myoglobins are a class of metalloproteins that contain a single iron(II) in a heme prosthetic group that can undergo a reversible one-electron transfer. The absence of subunits, such as found in the hemoglobins, precludes a cooperative response to redox, thus providing us with a system that exhibits nernstian behavior.

The spectral properties of horse, sperm whale, and Aplysia myoglobins (Mbs) are described briefly here. Sperm whale and horse Mbs have similar amino acid sequences.4 They have identical absorption maxima in the Soret region with small differences in molar absorptivity. The iron oxidation state of these Mbs can be monitored at 410 nm for oxidized iron(III) Mb and 435 nm for reduced iron(II) Mb (wavelengths of highest extinction coefficients), although other probe wavelengths can be used if the sample is sufficiently concentrated. Aplysia Mb, however, has a different sequence of amino acids, particularly in the heme pocket environment,4,6,7'9 with an absorbance maximum at 435 nm in its iron(II) state and at 399 nm in its iron(III) state. This corresponds to a shift of approximately 10 nm in comparison with the maxima observed in the case of both iron(III) horse and sperm whale Mbs. This slight shift is described by Brunori et al. as being associated with a reversible equilibrium between a partially opened globular structure, which exhibits a A.max at 390 nm, and the native structure, which exhibits a /,max at 410 nm.6

Methods

Sample and Reagent Preparation. A stock solution of the electrochemical mediator, Ru(NH3)6C13, is prepared in a 0.05 M 4-morpholinepropanesulfonic acid (MOPS) buffer solution adjusted to pH 7.1 to give a concentration of 4.5 to 5.5 mJi. MOPS is selected as the buffer for its noncomplexing nature and stability, as well as the absence of spectral and electrochemical interferences. KNO3 (0.2 M) is used in this particular set of experiments as the background electrolyte. The presence of a background electrolyte facilitates solution conductivity necessary for an electrochemical experiment. It is important to be systematic in the choice of the electrolyte in order to minimize the different kinds of species in solution. For example, because the bridge solution that connects the reference electrode and the working solution is composed of 0.2 M KC1 (as described in Spectroelectrochemi-cal Technique and Cell Design above), a potassium salt is used as the background electrolyte. The presence of K+ on both sides of the agar gel connection minimizes the errors associated with a junction potential across the reference-working solution interface. Nanopure water is used at all times and all solutions are stored under an N2 atmosphere at 4°.

For each experiment, a solution containing 0.2 M KN03, 1 mM Ru(NH3)6Cl3, and 0.05 M MOPS at pH 7.1 in a 5-ml pear-shaped flask is connected to a vacuum line for repeated pump purging with N2, followed by addition of Mb and additional pump purging with gentle swirling to minimize bubbling. Final concentrations are typically 0.06-0.08 mM in heme.

Spectroelectrochemical Experiment. Spectroelectrochemical experiments for myoglobins described here are carried out in the previously described anaerobic OTTLE cell (Fig. 1). In a typical experiment, about 0.5 ml of the working solution (protein, mediator, buffer with background electrolyte) is injected at the bottom of the OTTLE cell via a gas-tight syringe. The cell is then placed in the temperature-controlled cell holder of a spectrophotometer linked to a potentiostat [e.g., a Cary 2300 UV-Vis-NIR (near infrared) spectrophotometer (Varian, Palo Alto, CA) and a Princeton Applied Research (Oakridge, TN) model 75 potentiostat]. Spectra are collected from 340 to 700 nm, with specific emphasis on the Soret region. Absorbance changes are monitored at 410 nm [absorbance maximum for iron(III) Mb] and 435 nm [absorbance maximum for iron(II) Mb], The full region between 340 and 700 nm is recorded and the five isosbestic points (420, 462, 522, 606, and 662 nm) are scrutinized to detect any problems associated with the nature or concentration of the protein (e.g., denaturation). The absorbances of the fully oxidized (A0) and fully reduced (Ar) Mb are obtained by applying a potential of +400 and —250 mV [vs. normal hydrogen electrode (NHE)], respectively, and the absorbance is recorded when the system reaches equilibrium (15 to 45 min may be required to obtain a stable equilibrium absorbance reading, depending on the system). The optical path length can vary from cell to cell, but can be precisely determined for each experiment by using the Soret band absorbance, the known concentration of the working solution, and the extinction coefficients of the protein. The concentration is determined independently, typically after addition of dithionite to a portion of the unused working solution. The extinction coefficients for the reduced species [deoxy-Mb; iron(II) Mb] are as follows: £AApiySia = 113mAf 1 cm = 121mA/ 'cm '. and s^ whale =

115 mM_1 cm-1.4,6 The extinction coefficients for the oxidized species [met-Mb; iron(III) Mb] at 435 nm are negligible.

A typical increment of 20 mV is applied to the system, starting at approximately +300 mV (fully oxidized met-Mb) and ending at —120 mV (vs. NHE) (fully reduced deoxy-Mb). Although most experiments are performed by proceeding from fully oxidized to fully reduced Mb, the system is reversible under our experimental conditions and can be performed in either direction. From the recorded spectral changes (Fig. 2) and applied potential, Nernst plots are developed as described below.

Data Analysis

The set of absorbance data obtained for the system at equilibrium at various electrode potentials can be converted to the concentration ratio of oxidized to

340 390 440 490 540 590 640 690 wavelength (nm)

Fig. 2. Representative data set obtained by spectroelectrochemistry during the oxidation of horse Mb, showing a decrease in the absorbance at 435 nm in conjunction with an increase in the absorbance at 410 nm. Inset: Enlarged area showing the presence of four of the five isosbestic points at 462, 522, 606, and 662 nm, respectively.

340 390 440 490 540 590 640 690 wavelength (nm)

Fig. 2. Representative data set obtained by spectroelectrochemistry during the oxidation of horse Mb, showing a decrease in the absorbance at 435 nm in conjunction with an increase in the absorbance at 410 nm. Inset: Enlarged area showing the presence of four of the five isosbestic points at 462, 522, 606, and 662 nm, respectively.

where A(E) is the absorbance of the solution at equilibrium for any applied potential £app, A0 is the absorbance of the fully oxidized protein (at +400 mV vs. NHE), and Ar is the absorbance of the fully reduced protein (at —250 mV vs. NHE).

The [Ox]/[Red] ratio is plotted as a function of the applied potential £app, according to a rearranged form of the Nernst equation [Eq. (1)]. This results in a direct determination of £1/2, the potential at which 50% of the protein is oxidized and 50% is reduced, at log[Ox]/[Red] = 0. Representative Nernst plots for horse

Fig. 3. Representative Nernst plots for horse Mb in 0.05 M MOPS, 1 mM Ru(NH3)6Cl3, 1.0 M KNO3 at pH 7.1, 20°. Data analysis was performed at two different wavelengths, 410 nm (•) and 435 nm (O), and shows the overlap of the Nernst plots in comparison with a representative Nernst plot for Aplysia Mb (a) obtained at 435 nm.

Fig. 3. Representative Nernst plots for horse Mb in 0.05 M MOPS, 1 mM Ru(NH3)6Cl3, 1.0 M KNO3 at pH 7.1, 20°. Data analysis was performed at two different wavelengths, 410 nm (•) and 435 nm (O), and shows the overlap of the Nernst plots in comparison with a representative Nernst plot for Aplysia Mb (a) obtained at 435 nm.

and Aplysia myoglobins are illustrated in Fig. 3. Although only one wavelength needs to be monitored to obtain a Nernst plot, it is good practice, when possible, to monitor the wavelengths of maximum absorbance of the reduced and oxidized forms of the protein (i.e., absorbances at 410 and 435 nm in the case of the myoglobins). The Nernst plot derived from each monitored absorbance change should overlap to give the same Em and n values (Fig. 3). In addition, if a protein has characteristic isosbestic points for its reduced and oxidized forms, denaturation and a change in concentration of the protein can be monitored by evaluating how "clean" these isosbestic points are as the oxidation or reduction experiment is performed (Fig. 2). This is an important spectroscopic tool that we have used systematically to evaluate the quality of our data for both the Mbs and Hbs.

The results obtained for sperm whale, horse, and Aplysia Mbs are presented in Figs. 3 and 4. As shown for solutions of a single protein, we obtained linear Nernst plots throughout the applied potential range with an n value of 1, consistent with a well-behaved one-electron transfer process for each of these myoglobins. These

Fig. 4. Nernst plots for sperm whale (•) and Aplysia (O) myoglobins, and a 1:1 mixture of Aplysia : sperm whale (♦) myoglobins. The x axis represents applied potential versus Ag/AgCl according to a rearrangement of Eq. (1), where £/58.1 = E^F/RT. Conditions: 1 mMRuiNH^Ch, 0.06 m M heme, 0.2 M KN03, 0.05 M MOPS, pH 7.1, and 20°; data collected at À = 435 nm. Simu

Fig. 4. Nernst plots for sperm whale (•) and Aplysia (O) myoglobins, and a 1:1 mixture of Aplysia : sperm whale (♦) myoglobins. The x axis represents applied potential versus Ag/AgCl according to a rearrangement of Eq. (1), where £/58.1 = E^F/RT. Conditions: 1 mMRuiNH^Ch, 0.06 m M heme, 0.2 M KN03, 0.05 M MOPS, pH 7.1, and 20°; data collected at À = 435 nm. Simu lated data (—) calculated according to Eq. (12), where sTa = £'

„435 Aplysia

.435 sperm whale

= 115mM cm

= 113mA/ 'cm 1, andfc = 0.3mm. [Reprinted from C. H. Taboy, C.Bonaventura, andA.L.

Crumbliss, Bioelectrochem. Bioenerget. 48, 79 (1999), with permission from Elsevier Science.]

spectroelectrochemical results are in close agreement with Em values previously determined by Potentiometrie titration in the presence of redox mediators.6'7-95

Heme Proteins with Nonnernstian Response

Although the preceding examples represent three well-behaved Nernst plots with a slope corresponding to an n value of 1 for met-Mb reduction to deoxy-Mb,

95 J. F. Taylor and V. E. Morgan,./. Biol. Chem. 144, 15 (1942).

it is important to be alert for possible deviations from linearity in the Nernst plot. Such deviations can signal a poorly designed or malfunctioning experiment, or may indicate the presence of a more complex system. In the latter case, there are two scenarios that can lead to a nonnernstian response. The first case (scenario 1) involves two (or more) noninteracting redox groups. This results in a deviation from linearity of the slope in the Nernst plot and is sometimes referred to as negative cooperativity. However, this behavior is, in fact, associated with the presence of two noninteracting electroactive species with different E\a values. This scenario can occur within a unique protein possessing two noninteracting electroactive centers or within a mixture of two or more redox-active proteins, as long as a difference in their respective E\n values is present. Below, we show that a theoretical model describing this behavior accurately predicts and precisely the experimental data obtained for a mixture of Aplysia and sperm whale myoglobins, for which a difference in £1/2 of 80 mV has been demonstrated.84 The second case (scenario 2) involves two (or more) redox centers interacting with one another and is observed experimentally when studying the hemoglobins. Intracellular hemoglobin typically possesses four interdependent heme centers that have the ability to communicate with each other. These heme-heme interactions aid in the efficient uptake and release of dioxygen, a phenomenon known as cooperativity. The presence of a slope corresponding to n > 1 (nonnernstian response) for the hemoglobins is linked to the intrinsic electronic properties of the protein and gives information regarding the ability of the four prosthetic groups to influence each other electronically.

These two examples (scenarios 1 and 2) of nonnernstian response are different in nature and cannot be overlooked when analyzing the results from a spectro-electrochemical experiment.

Noninteracting Electroactive Centers: Development of Model

The following model is designed to predict the behavior of a Nernst plot when a mixture of two noninteracting but well-behaved nernstian electroactive species is studied by spectroelectrochemistry (scenario 1 described above). Two cases are developed. In one case the two species have identical spectra, but different midpoint potentials, and in the other case the spectra and midpoint potentials are both different.

For a mixture of two noninteracting species A and B we can represent the oxidized and reduced forms as Oa and 0B, and Ra and Rb, respectively. Beer's law at a fixed applied potential and fixed wavelength may be used to describe the total absorbance at equilibrium, Ae, as shown in Eq. (3):

Ae = AeA + AeB = (£rA[RA] + £oa[Oa])& + (£rB[RB] + £0b[Ob])6 (3)

where AeA and Aeb are the equilibrium absorbance values for species A and B at each applied potential, respectively; eoA, erA, e0B> and £rs are the extinction coefficients of species A and B in their oxidized and reduced forms, respectively; and b is the optical path length through the OTTLE cell. This relationship may then be used to derive an expression for the ratio of the mixture of oxidized to reduced species in terms of absorbance values as shown in Eqs. (4) and (5):

where E Ar and E A0 are the sums of the absorbances of the fully reduced and fully oxidized species A and B, respectively, and E Ae is the sum of the equilibrium absorbances of the oxidized and reduced forms of species A and B at a specific applied potential.

Because the oxidized Mb species have a negligible absorbance at 435 nm, Eq. (5) may be simplified to Eq. (6).

The Nernst equation for species A and B [Eqs. (7) and (8)],

may be substituted for [0A] and [Ob] in Eq. (3). Using the mass balance Eqs. (9) and (10), we can relate the applied and midpoint potentials E and £1/2 to the absorbance levels via Eq. (11). The symbols cA and cB represent the total concentrations of species A and B, respectively.

cB = [Rb] + [Ob] (10) EAe = [(£rAfecA)/(l +exp{nA[£ - EmA)]F/RT})]

Finally, Eq. (12) is derived by substituting Eq. (11) into Eq. (6):

[Ox]/[Red] = -1 + {EAr/[(erAfecA)/(l +exp{«A[£ - Em(A)]F/RT})

Data simulations derived from the log plots of Eq. (12) are shown as solid lines in Figs. 4, 5 and 6.

Figures 5 and 6 represent simulations of two factors influencing Nernst plots obtained from spectroelectrochemical data. In Fig. 5, the influence of two nonin-teracting systems with identical spectra, but different midpoint potentials (AEm), on the Nernst plot is explored. This situation can be related to studying a system possessing two heterogeneous electroactive centers that are spectroscopically

Fig. 5. Simulation showing the influence of A£i/2 on a log plot of Eq. (12). For each case, £rA = £rB- (•) £i/2(A) = 50 mV, £i/2(B) = 50 mV, A £1/2 = OmV; (o) Emw = 10 mV, £i/2(b> = 90mV, AEi/2 = 80mV; (♦) Em(A) = -50mV, £i/2(B) = 150mV, AEm = 200mV. [Reprinted from C. H. Taboy, C. Bonaventura, and A. L. Crumbliss, Bioelectrochem. Bioenerget. 48, 79 (1999), with permission from Elsevier Science.]

Fig. 5. Simulation showing the influence of A£i/2 on a log plot of Eq. (12). For each case, £rA = £rB- (•) £i/2(A) = 50 mV, £i/2(B) = 50 mV, A £1/2 = OmV; (o) Emw = 10 mV, £i/2(b> = 90mV, AEi/2 = 80mV; (♦) Em(A) = -50mV, £i/2(B) = 150mV, AEm = 200mV. [Reprinted from C. H. Taboy, C. Bonaventura, and A. L. Crumbliss, Bioelectrochem. Bioenerget. 48, 79 (1999), with permission from Elsevier Science.]

equivalent, but do not influence the electronic properties of one another. As the difference between the midpoint potentials (AEia) of the two independent species increases, a clear increase in the deviation from linearity around the midpoint potential can be observed, as predicted on theoretical grounds.5 For this hypothetical system, where eA = £b, it is important to note that the minimum slope is still found at the midpoint potential (Em).

Figure 6 illustrates the additional effect on the Nernst plot for a mixture of two species with homologous spectra that differ in their molar absorptivity coefficients at the probe wavelength (e.g., in our experimental examples at 435 nm, £a ^ £b) and with different midpoint potentials; A Em = 100mV. Two

Fig. 6. Simulation showing the influence of the extinction coefficient for species A (£,a) and B (£rB) on a log plot of Eq. (12). For all data simulated, Ev2(A) = +50 mV, Ei/2(B) = —50mV, AE1/2 = 100 mV. Single species A (T); single species B (V). Mixtures (1: 1) of species A and B with the following ratios of erA to £rB: (•) 1:9; (O) 1:3; (A) 2:3; (A) 1:1; (♦) 9:1. [Reprinted from C. H. Taboy, C. Bonaventura, and A. L. Crumbliss, Bioelectrochem. Bioenerget. 48, 79 (1999), with permission from Elsevier Science.]

Fig. 6. Simulation showing the influence of the extinction coefficient for species A (£,a) and B (£rB) on a log plot of Eq. (12). For all data simulated, Ev2(A) = +50 mV, Ei/2(B) = —50mV, AE1/2 = 100 mV. Single species A (T); single species B (V). Mixtures (1: 1) of species A and B with the following ratios of erA to £rB: (•) 1:9; (O) 1:3; (A) 2:3; (A) 1:1; (♦) 9:1. [Reprinted from C. H. Taboy, C. Bonaventura, and A. L. Crumbliss, Bioelectrochem. Bioenerget. 48, 79 (1999), with permission from Elsevier Science.]

important points are illustrated in Fig. 6: (1) One of the redox species must have an extinction coefficient at the probe wavelength that is much larger than the other (e.g., £a/£es = 1 : 9 or 9 : 1) in order for the Nernst plot to become tangential to the respective single-component plot at extreme applied potentials; and (2) if there is a difference between the extinction coefficients for the two species at the probe wavelength (êa and êb), the minimum slope will not be present at E\p_, but will shift as a function of the ratio of gA to eB.

Clearly, the Nernst plots for a mixture of two noninteracting redox systems of different spectra and midpoint potentials will deviate from ideality (linearity).

If the system is functioning properly the shapes of the Nernst plots are diagnostic of the presence of two or more heterogeneous electroactive centers, as illustrated in Figs. 5 and 6.

Noninteracting Electroactive Centers: 1:1 Mixture of Myoglobins

We can experimentally verify the model calculations described above, using a mixture of Mbs. When a 1:1 mixture of horse and sperm whale Mbs is studied by spectroelectrochemistry, a well-behaved Nernst plot is obtained with a slope corresponding to n = 1 and £1/2 = 17 mV, which is equivalent to the £"1/2 obtained for the individual species (results not shown). This is the expected result as these two Mbs have identical spectra and Em values when isolated from each other and act therefore as a single homogeneous protein, spectrally and electronically "identical" to each other.

Mixtures containing Aplysia and sperm whale Mb, however, give a reproducible nonnernstian plot, as anticipated by the model described in the previous section. This is due to a difference in their respective midpoint potentials (A£i/2 = 80 mV). Nernst plots for Aplysia and sperm whale Mb alone, and in 1:1 mixture, are presented in Fig. 4 along with a simulation for the mixture.

These results demonstrate that a mixture of two noninteracting (and therefore noncooperative) heme proteins with identical £1/2 values (horse and sperm whale Mbs) gives a well-behaved linear nernstian response at both 410 and 435 nm (Amax for the oxidized and reduced forms of Mb), with a slope corresponding to n = 1 and £1/2 overall equal to the £i/2 of each independent Mb. By comparison, a mixture of two noninteracting heme proteins with notably different £1/2 values gives a nonnernstian response. This nonnernstian behavior, illustrated in Fig. 4, demonstrates the influence on Nernst plots of mixtures containing noninteracting heme sites with significant £i/2 differences. Protein heterogeneity will lead to a Nernst plot with minimum slope of less than unity (n < 1). This depression in n gives important information regarding the composition of the protein or mixture studied. A representative example illustrating this lack of homogeneity in a structure was presented by Malmstrom, who rationalized the observed value of« = 0.5 reported in early work for cytochrome c oxidase oxidation-reduction by developing a two-site model based on a difference between the £1/2 value for each site.96

Interacting Electroactive Centers: Hemoglobins

Spectroelectrochemical results obtained from hemoglobin samples differ widely from the nonnernstian behavior observed for myoglobin mixtures, and for other noninteracting electroactive centers. Nernst plots for human Hb Ao actually show a drastic increase in their slope, corresponding to n values as high as 2.0

96 B. G. Malmstrom, Q. Rev. Biophys. 6, 389 (1974).

in some cases. For systems with subunit-subunit interactions the Nernst plot slope (n) no longer strictly corresponds to the number of electrons transferred (as is also the case for noninteracting systems with chain heterogeneity, as discussed above). In this section, we summarize specific representative responses obtained when studying human hemoglobin (Hb A0). More extensive results on Hb Ao and other hemoglobins are presented elsewhere.1,3'35'69'84'85

The experimental design for evaluation of the redox behavior of the hemoglobins (Hbs) is similar to that described for the Mbs. The extinction coefficient and specific Soret band maximum are shifted relative to Mb, with maxima at 406 nm for iron(III) Hb and at 430 nm for iron(II) Hb and isosbestic points at 415, 455, 524, and 598 nm. An example of the nonnernstian response obtained for Hb Aq is presented in Fig. 7, along with a well-behaved nernstian response for

Fig. 7. Nernst plot for sperm whale myoglobin (O) and human hemoglobin (Hb Ao) (•). The x axis represents applied potential versus NHE according to a rearrangement of Eq. (1) where E/58.1 = EmF/RT. E\n is indicated for Hb A0. Conditions: 0.06-0.08 m M heme, 0.05 M MOPS, 1 m M Ru(NH3)6Cl3, 0.2 M KNO3, pH 7.1, and 20°. [Reprinted from C. H. Taboy, C. Bonaventura, and A. L. Crumbliss, Bioelectrochem. Bioenerget. 48, 79 (1999), with permission from Elsevier Science.]

Fig. 7. Nernst plot for sperm whale myoglobin (O) and human hemoglobin (Hb Ao) (•). The x axis represents applied potential versus NHE according to a rearrangement of Eq. (1) where E/58.1 = EmF/RT. E\n is indicated for Hb A0. Conditions: 0.06-0.08 m M heme, 0.05 M MOPS, 1 m M Ru(NH3)6Cl3, 0.2 M KNO3, pH 7.1, and 20°. [Reprinted from C. H. Taboy, C. Bonaventura, and A. L. Crumbliss, Bioelectrochem. Bioenerget. 48, 79 (1999), with permission from Elsevier Science.]

sperm whale Mb, illustrating the sigmoidal shape of the Nernst plot for Hb Ao. Four parameters are used to describe Hb responses: the midpoint potential (£1/2) and the potential at which the maximum slope of the Nernst plot is observed (£max) and, because the slope of the Nernst plot is constantly changing, we also define the n value at the midpoint potential as «1/2 and the n value at the maximum slope as "max- These parameters permit comparison of each data set as a function of the specific midpoint potential (Em) of the system and its level of cooperativity (as described by the slope at the midpoint potential, «1/2), and give a general sense of the asymmetry of the curve (\E\n — Emax|).

For the hemoglobins, as in the case of any nonnernstian response, the meaning of the midpoint slope («1/2) is not straightforward, and in any case should not be strictly interpreted as the number of electrons involved in the redox process. It can, however, be used to evaluate, within a set of experiments, the effect or influence of the medium composition on site-site interactions. For instance, in the case of hemoglobins, the presence of anions in the medium influences the apparent level of "cooperativity" of the system and gives fundamental clues regarding the mechanism used by the protein to propagate or relay important information between its surface (medium-protein interaction) and its embedded heme groups.84 As shown in many previous studies, the globin chain exerts a profound influence on the redox potential of the heme site of hemoglobins (Hbs) and myoglobins (Mbs), and protects them from rapid oxidation, which, in turn, allows for the reversibility observed in dioxygen binding to the prosthetic group.4'n"14'22,24,34'65'70'71'1AJ1-79-97-'03

Oxygenation and oxidation-reduction studies of iron(II)/iron(III) centers for diverse Hbs have reported interesting interrelationships between these two processes.2'4'10'69 One of the important observations we have made regarding the parallels between these two processes links the shift from the T [deoxy, iron(II)] to the R (oxygenated) or R-like [met, iron(III)] conformation of the Hb tetramer to the cooperativity observed in both oxygenation and oxidation processes. Numerous studies have shown that structural changes stabilizing either the T or the R conformation of the Hb tetramer are typically reflected by comparable alterations of both oxidation and oxygenation processes.4'11 69 Preferential binding to the low-affinity

97 A. Szabo and M. Karplus, J. Mol. Biol. 72, 163 (1972).

98 K. D. Vandegriff and R. M. Winslow, "Blood Substitutes: Physiological Basis of Efficacy." Birkhauser, Boston, 1995.

99 J. P. Collman, R. R. Gagne, C. A. Reed, T. R. Haibert, G. Lang, and W. T. Robinson, J. Am. Chem. Soc. 97, 1427 (1975).

100 X. Hu and T. G. Spiro, Biochemistry 36, 15701 (1997).

101 H. X. Zhou, J. Biol. Inorg. Chem. 2, 119 (1997).

103 F. A. Tezcan, J. R. Winkler, and H. B. Gray, J. Am. Chem. Soc. 120, 13383 (1998).

0 0

Post a comment