Skip to main content
NCBI home page
Protein Science : A Publication of the Protein Society logoLink to Protein Science : A Publication of the Protein Society
. 2007 Feb;16(2):250–260. doi: 10.1110/ps.062572607

Covalent heme attachment in Synechocystis hemoglobin is required to prevent ferrous heme dissociation

Julie A Hoy 1, Benoit J Smagghe 1, Puspita Halder 1, Mark S Hargrove 1
PMCID: PMC2203299  PMID: 17242429

Abstract

Synechocystis hemoglobin contains an unprecedented covalent bond between a nonaxial histidine side chain (H117) and the heme 2-vinyl. This bond has been previously shown to stabilize the ferric protein against denaturation, and also to affect the kinetics of cyanide association. However, it is unclear why Synechocystis hemoglobin would require the additional degree of stabilization accompanying the His117–heme 2-vinyl bond because it also displays endogenous bis-histidyl axial heme coordination, which should greatly assist heme retention. Furthermore, the mechanism by which the His117–heme 2-vinyl bond affects ligand binding has not been reported, nor has any investigation of the role of this bond on the structure and function of the protein in the ferrous oxidation state. Here we report an investigation of the role of the Synechocystis hemoglobin His117–heme 2-vinyl bond on structure, heme coordination, exogenous ligand binding, and stability in both the ferrous and ferric oxidation states. Our results reveal that hexacoordinate Synechocystis hemoglobin lacking this bond is less stable in the ferrous oxidation state than the ferric, which is surprising in light of our understanding of pentacoordinate Hb stability, in which the ferric protein is always less stable. It is also demonstrated that removal of the His117–heme 2-vinyl bond increases the affinity constant for intramolecular histidine coordination in the ferric oxidation state, thus presenting greater competition for the ligand binding site and lowering the observed rate and affinity constants for exogenous ligands.

Keywords: hemoglobin, hexacoordinate, Synechocystis, ligand binding, ferrous, heme affinity, folding

The past few decades have revealed that hemoglobins are found not only in vertebrates, but also in archaea, bacteria, animals, and plants (Weber and Vinogradov 2001; Wittenberg et al. 2002; Kundu et al. 2003b; Wu et al. 2003). A common function of vertebrate Hb is to transport oxygen from the lungs to tissues through the vascular system. However, Hbs have many other potential functions including ligand scavenging, detoxification, small molecule sensing, signal transduction, and gene regulation (Appleby 1984; Minning et al. 1999; Burr et al. 2000; Hou et al. 2000; Poole and Hughes 2000; Wittenberg et al. 2002; Hankeln et al. 2005), while the functions of many Hbs remain to be identified.

A number of newly discovered Hbs display unorthodox structural characteristics. Plants, animals, and some bacteria have Hbs that carry out reversible intramolecular coordination of the heme iron by a “distal” histidine side chain (Burmester et al. 2000; Trent et al. 2001b; Weber and Vinogradov 2001; Burmester et al. 2002; Trent and Hargrove 2002; Wittenberg et al. 2002; Kundu et al. 2003b; Wu et al. 2003). These “hexacoordinate” Hbs (hxHbs) are likely the scaffold from which oxygen transport Hbs evolved in plants and animals (Vinogradov et al. 2006). The “proximal” histidine is tightly coordinated to the heme iron (as in oxygen transport Hbs), but the distal histidine reversibly coordinates the ligand binding site and competes with the binding of exogenous ligands like oxygen and nitric oxide. Recent studies in plant and animal systems have described structural components that regulate affinity constants for distal histidine binding, showing that the strength of histidine coordination greatly affects affinity and kinetic constants for exogenous ligand binding (Smagghe et al. 2006a,b).

Another nontraditional group of Hbs, the “truncated” Hbs (trHbs), is found in eubacteria, bacteria, single-celled eukaryotes, and plants (Couture et al. 2000; Scott and Lecomte 2000; Hvitved et al. 2001; Watts et al. 2001; Wittenberg et al. 2002). They can be 20–40 residues shorter than the majority of Hbs and comprise an abbreviated globin architecture surrounding the heme that forms a “2-on-2” fold (Pesce et al. 2000; Wittenberg et al. 2002). Cyanobacteria contain a trHb that is unique even within this family. The Hb from Synechocystis sp. PCC 6803 (SynHb) is hexacoordinate, truncated, and bears an unprecedented covalent bond between the non-axial His117 and a heme porphyrin vinyl group (Vu et al. 2002).

Extensive research has investigated the role of this covalent heme–protein link since its 2002 discovery (Vu et al. 2002). Prior to identification of this bond, the existence of two forms of the protein had been noted, and experiments were carefully performed only on the form of the protein that, in retrospect, lacked the covalent bond (Lecomte et al. 2001). These experiments included observation of slow heme release (Scott and Lecomte 2000; Lecomte et al. 2001), thermal denaturation (Lecomte et al. 2001), and measurement of the electrochemical midpoint potential (Lecomte et al. 2001). After the covalent bond was discovered as the source of heterogeneity, partial NMR structural analysis indicated little global conformational change among the ferric tertiary structures of the two forms of wild-type protein and the H117A mutant protein (which is incapable of forming the covalent bond) (Falzone et al. 2002; Vu et al. 2004a, 2004b). Thermal and acid denaturation studies of the ferric forms of these three proteins found the wild-type protein with the covalent bond to be more stable than the mutant or wild-type lacking the bond (Lecomte et al. 2001; Vu et al. 2004a,b). Finally, cyanide binding to the ferric form of the H117A mutant was found to be decreased compared to wild-type protein with an undetermined covalent bond contribution (Vu et al. 2004a).

The above studies suggested that the His117–heme vinyl bond does not affect the overall conformation of the ferric protein, but does affect protein stability, heme affinity, and the ability of SynHb to bind exogenous ligands in the ferric oxidation state. The most plausible explanation for the lowered cyanide affinity seen in the mutant lacking the covalent bond is increased competition for the binding site due to an increase in the equilibrium constant for distal histidine coordination to the ferric heme iron. Since heme affinity (and thus protein stability) would increase with this tighter heme coordination, it is odd that the His117–heme vinyl bond would be necessary to stabilize the protein in the ferric oxidation state. Furthermore, small molecule model-heme studies affirm that bis-histidyl heme coordination may be tighter in the ferric than the ferrous oxidation state (Nesset et al. 1996; Safo et al. 1997). If this chemical principle holds true in SynHb, the protein should be more susceptible to heme loss and subsequent unfolding in the ferrous oxidation state compared to the ferric. Consequently, the largest effects of removal of the His117–heme vinyl covalent bond might be exposed in the ferrous protein, a factor not investigated previously.

Here we demonstrate that removal of the His117–heme vinyl covalent bond in SynHb affects distal histidine coordination of the heme iron by facilitating hexacoordination in the ferric oxidation state by a factor of ~10 compared to the wild-type protein. This is the root of the slower binding kinetics for ferric ligands in SynH117A, and also results in lowered equilibrium affinity constants for azide and cyanide. Surprisingly, the effects of the H117A mutation on protein stability and heme affinity are much more pronounced in the ferrous oxidation state compared to the ferric. This extraordinary observation is counter to our understanding of heme affinity in pentacoordinate hemoglobins (Hargrove and Olson 1996), but is consistent with studies of coordination affinity in bis-histidyl model heme compounds. Finally, crystal structures of wild-type and H117A SynHb affirm that the effects of this covalent bond on protein chemistry are not obvious in protein structure in either the ferric or ferrous oxidation sates.

Results

Crystal structures of ferric and ferrous SynHb and SynH117A

Figure 1 presents an overlay of the crystal structures of SynHb (with the covalent bond) and SynH117A in both ferric and ferrous oxidation states. The backbone superposition of the ferric structures in Figure 1A shows minimal backbone deviation (RMSD = 0.45 Å), but small changes in the orientation of the heme and axial histidines, in agreement with the partial structural observations obtained by NMR (Vu et al. 2004b). The superposition in Figure 1B shows that, relative to the heme, mutation of H117 causes outward movement of the H-helix with minor movements of the F- and G-helices (RMSD = 1.08 Å). Similar results are found in the comparisons of the ferrous forms as shown in Figure 1C (backbone superposition) and Figure 1D (heme superposition), with additional movement of the short A-helix. Therefore, the His117–heme vinyl bond has minimal effects on overall structure compared to the wild-type protein in both the ferric and ferrous oxidation states. However, the H117A mutation and the consequent movement of the H-helix opens space on the proximal side of the heme that allows solvent entry into heme pocket from the protein surface in both oxidation states (Fig. 1E,F). A closer examination of the changes in heme stereochemistry is summarized in the supplemental materials. It appears that there are no obvious structural differences between wild-type and SynH117A in either the ferrous or ferric oxidation states that rationalize the observed functional differences between these two proteins or the presence of the covalent bond between His117 and the heme 2-vinyl.

Figure 1.

Figure 1.

Open in a new tab

SynHb (gray) versus SynH117A (colored by RMSD from SynHb, with dark blue corresponding to smallest deviation). Ferric structures overlaid by backbone (A) and heme (B). Ferrous structures overlaid by backbone (C) and heme (D). Axial histidines are shown, along with the amino acid at position 117. Mutation causes a solvent pocket to open on the proximal side of the heme, as seen in the superposition of SynHb (red) with SynH117A (blue) in the ferric (E) and ferrous (F) forms. The amino acid at position 117 is shown for each protein. The X-ray crystal structure of the ferric form of SynHb was solved previously (Hoy et al. 2004).

Ligand binding in the ferrous oxidation state

The reactions associated with ligand binding in HxHbs are shown in Scheme 1. For the case of negligible ligand dissociation, the equation for the observed rate of the exogenous ligand binding reaction assuming a steady-state equilibrium between HbH and HbP is given by Equation 1.

Scheme 1.

Scheme 1.

Open in a new tab
graphic file with name 250equ1.jpg

Here, H, P, and L refer to Hb in the hexacoordinate, pentacoordinate, and ligand bound states, respectively. The rate constants k −H and k H are the dissociation and association rates for the distal histidine, while k′L[L] and kL are the ligand bimolecular association and dissociation rate constants.

Flash photolysis was used to measure the CO bimolecular association rate constant of the SynH117A mutant protein for comparison to that published previously for wild-type SynHb (Hvitved et al. 2001; Smagghe et al. 2006b); this value, k′CO = 150 μM−1sec−1, is only slightly greater than that of the wild-type protein (see Table 1 for values and supplemental data for more information). The influence of intramolecular histidine coordination becomes apparent when comparing kinetic experiments initiated by flash photolysis to those measured by rapid mixing (Smagghe et al. 2006b). Figure 2A shows time courses for SynH117A binding with CO following rapid mixing. These data are consistent with those of wild-type SynHb in that the total expected amplitude is observed for the reaction at each [CO], and the observed rate constant increases with [CO] (Smagghe et al. 2006b). This indicates that k′CO > k −H2 and k H2 (where k −H2 and k H2 are the dissociation and association constants for the distal histidine in the ferrous state), and that the fraction of hexacoordinate Hb is near 1. Both of these observations are consistent with wild-type ferrous SynHb (Smagghe et al. 2006b).

Table 1.

Ligand binding constants and redox potentials for wild-type and H117A SynHb

graphic file with name 250tbl1.jpg

Open in a new tab

Figure 2.

Figure 2.

Open in a new tab

Rapid mixing kinetic measurement of CO binding. (A) Change in absorbance over time for SynH117A mixed with 500, 200, 100, 50, and 25 mM CO (from top to bottom). (B) k obs versus CO concentration for SynHb (filled circles) compared to SynH117A (open circles).

The [CO] dependence of the observed rate constants (from rapid mixing) in Figure 2A is plotted in Figure 2B. The asymptote approached in this curve is k −H2, and a fit to Equation 1 provides k H2 (Hargrove 2000). These values are reported in Table 1, and reflect only modest (approximately threefold) increases for each in the mutant protein. Since both k H2 and k −H2 increase to approximately the same degree in the mutant compared to the wild-type protein, the resulting values of K H2 (the association equilibrium constant for the distal histidine, equivalent to k H/k −H) are nearly the same for each. These results demonstrate that the His117–heme vinyl bond has only minor effects on hexacoordination and exogenous ligand binding in ferrous SynHb.

Spectroelectrochemistry

Electrochemistry can be used to measure changes that occur in the strength of hexacoordination upon change in oxidation state (Halder et al. 2006; Smagghe et al. 2006b). In hexacoordinate Hbs, a shift to more-negative reduction potentials (compared to pentacoordinate Hbs) reveals that intramolecular histidine coordination of the heme iron is tighter in the ferric compared to the ferrous oxidation state (Dewilde et al. 2001; Halder et al. 2006; Smagghe et al. 2006b). Since previous work reported the midpoint reduction potential for only the wild-type protein without the covalent bond (Lecomte et al. 2001), we examined the midpoint potential for SynHb with the covalent bond and the SynH117A protein to determine if the His117–heme vinyl bond influences hexacoordination differentially in the ferric versus the ferrous oxidation state. The fraction of reduced protein for both SynHb and the H117A mutant protein as a function of the observed cell potential are shown in Figure 3A. The midpoint potential value (given in Table 1) is vastly lower for SynH117A (−280 mV) than for SynHb (−195 mV).

Figure 3.

Figure 3.

Open in a new tab

Electrochemistry and ferric binding data for SynHb (filled circles) and SynH117A (open circles). (A) SynHb has a higher reduction midpoint potential than SynH117A. (B) Optical cyanide binding kinetics showing slower ligand binding to SynH117A. (C) SynHb and SynH117A equilibrium binding curves. Values found from the fits (solid lines) in each panel are given in Table 1.

The relationship between the equilibrium constant for hexacoordination (K H) in the ferric (H3) and ferrous (H2) forms and the midpoint reduction potential (E mid) for the pentacoordinate (pent) and hexacoordinate (hex) forms is given by Equation 2 (Halder et al. 2006).

graphic file with name 250equ2.jpg

If we assume that the difference in midpoint potentials between the wild-type and H117A proteins is due only to histidine coordination (that is, E mid,pent is the same in both proteins), an estimate of the effect of the covalent bond on hexacoordination (K H3/K H2) can be calculated by Equation 3, which is derived by division of Equation 2 for the wild-type protein by that for SynH117A (Halder et al. 2006).

graphic file with name 250equ3.jpg

This equation correlates the effects of hexacoordination in each oxidation state with the observed reduction potential in each protein. An increased value of K H3/K H2 is predicted for the mutant based on the large negative shift in reduction potential observed in Figure 3A. In fact, using the measured values given in Table 1 to calculate the right-hand side of Equation 3 gives a value of 27, affirming the increase in ferric hexacoordination upon removal of the covalent bond.

Ferric ligand binding

The increase in K H3/K H2 predicted above can be tested using equilibrium and kinetic ligand binding experiments. The effect of a change in K H3 on these rate constants has been described previously in other systems (Trent et al. 2001b; Smagghe et al. 2006b), and the equations are shown here for cyanide serving as the ligand.

graphic file with name 250equ4.jpg
graphic file with name 250equ5.jpg

K CN,obs and k CN,obs are the observed equilibrium affinity and association rate constants (respectively). The right-hand side of each equation relates these values to the equilibrium affinity and association rate constants for the pentacoordinate state of the protein (K CN,pent and k′CN,pent) along with the effect of hexacoordination under conditions of prior equilibrium (this assumes that k′CN[CN] << k −H3 and k H3) (Espenson 2002).

Kinetic and equilibrium constants for azide and cyanide binding were measured for the H117A and wild-type proteins (Fig. 3B,C). The observed rate constants for cyanide association are linear and very slow, indicating that the bimolecular rate constant for the reaction with the pentacoordinate form of the Hb (k′CN) is << k H3 and k −H3, as required in the equations above (Smagghe et al. 2006b). A linear fit to these data provides k CN,obs for each protein showing that the value for SynH117A is reduced ~10-fold compared to the wild-type protein (Table 1), which is consistent with the value reported previously (Vu et al. 2004a). This is accompanied by a proportionate decrease in the association equilibrium constant (K CN,obs) for cyanide. Likewise, the wild-type protein binds azide (albeit with a much smaller K N3,obs than for cyanide), but the mutant protein displayed no binding until concentrations higher than 4 M azide (data not shown).

GdmCl unfolding

The foremost hypothesis for the role of the His117–heme vinyl bond is stabilization of the protein through heme retention. Vu et al. (2004a) showed that, in fact, the ferric protein is less stable without the covalent bond. However, while there is reason to believe that hxHbs might be even less stable in the ferrous oxidation state than the ferric (Nesset et al. 1996; Safo et al. 1997), ferrous protein stability has not been examined. Therefore, we have measured GdmCl-induced unfolding of these proteins in both oxidation states. These experiments are shown in Figures 4 and 5, with unfolding midpoints, ΔG O, and m values listed in Table 2.

Figure 4.

Figure 4.

Open in a new tab

GdmCl denaturation. Ferric (3+, filled circles, left axis) versus ferrous (2+, open circles, right axis) protein denaturation curves for horse heart myoglobin (A), SynHb (B), and SynH117A (C). Values found from the fits (solid lines) in each panel are given in Table 2.

Figure 5.

Figure 5.

Open in a new tab

Spectral analysis of GdmCl denaturation. Optical spectra are shown without addition of GdmCl (solid lines) and at high [GdmCl] (dashed lines) for the ferric (A) and ferrous (B) forms of pentacoordinate horse heart myoglobin, the ferric (C) and ferrous (D) forms of hexacoordinate SynHb with the covalent bond, and ferric (E) and ferrous (F) forms of hexacoordinate SynH117A. Panels E and F also contain the high [GdmCl] spectra (dotted lines) from the hhMb graphs of A and B for comparison.

Table 2.

GdmCl denaturation midpoints and free energies

graphic file with name 250tbl2.jpg

Open in a new tab

Figure 4A provides a control with pentacoordinate horse heart Mb (hhMb), demonstrating that the ferrous protein is more stable than the ferric; mainly due to the tighter heme–proximal histidine bond in the ferrous protein (Hargrove et al. 1996a,b; Hargrove and Olson 1996). GdmCl unfolding of Mb yields free heme and unfolded apoglobin (Hargrove and Olson 1996; Tang et al. 1998), evident from the absorbance spectra for ferrous and ferric hhMb at 5 M GdmCl (Fig. 5A,B). Wild-type SynHb, on the other hand, is equally stable in either oxidation state (Fig. 4B); in fact, the spectral transition associated with GdmCl titration of wild-type SynHb does not yield free heme, but rather pentacoordinate complexes in both oxidation states (Fig. 5C,D). This is most obvious in the ferrous oxidation state (Fig. 5D, inset), in which the spectrum in 7 M GdmCl is a single broad absorbance band at 550 nm, characteristic of a ferrous pentacoordinate heme complex (Arredondo-Peter et al. 1997; Smagghe et al. 2006b).

The results for the SynH117A mutant protein are quite surprising (Fig. 4C). This protein is less stable in both oxidation states compared with wild type SynHb, but unlike pentacoordinate hhMb, is most unstable in the ferrous oxidation state. Furthermore, the final absorbance spectra at high [GdmCl] are nearly identical to those of hhMb, indicating that heme is dissociating from this protein that lacks the His117–heme vinyl bond (Fig. 5E,F). This suggests that the His117–heme vinyl covalent bond is the dominant contributor to heme affinity compared to the axial His-heme coordinate bonds.

Heme dissociation

If heme affinity is the principal factor affecting protein stability, it should be possible to directly measure its dissociation. This is the case for ferric (but not ferrous) Mb, in which transfer to the heme-scavenger protein (apoH64Y/V68F Mb) has been used to measure rate constants for heme dissociation (Hargrove et al. 1996b), demonstrating that this protein loses heme at a rate of ~0.007 h−1 at pH 7.0. Previous work demonstrated very slow heme loss from ferric wild-type SynHb without the covalent bond (Scott and Lecomte 2000; Lecomte et al. 2001). Figure 6 shows heme dissociation experiments for SynHb (containing the covalent bond) and the SynH117A mutant protein in both the ferrous and ferric oxidation states. In each case, absorbance spectra are shown just after mixing with apoH64Y/V68F Mb and 2000 min after mixing. For the wild-type protein in both oxidation states, and for the ferric SynH117A mutant protein, no appreciable heme transfer is observed (Fig. 6 A–C). However, the ferrous SynH117A mutant protein loses heme at a rate of 0.012 min−1, ~100-times faster than ferric sperm whale myoglobin under the same conditions (Fig. 6D,E). Thus, the ferrous form of the protein is susceptible to heme loss, and therefore loss of function, without the added protection of the covalent bond.

Figure 6.

Figure 6.

Open in a new tab

Heme transfer to H64Y/V68F apomyoglobin. Optical spectra initially (lines) and after 2000 min (dotted lines) for ferric SynHb (A), ferrous SynHb (B), ferric SynH117A (C), and ferrous SynH117A (D) reveal heme loss only for ferrous SynH117A (E).

Discussion

Interest in hemoglobin stability originated with efforts to understand hemoglobin pathologies that resulted in heme release and protein degradation. Thus, the vast majority of research in hemoglobin stability has focused on myoglobin and red blood cell hemoglobin, which are pentacoordinate with only the proximal His coordinating the heme iron. The earliest studies linked heme release with the ferric oxidation state, which led to the view that the ferrous–His bond is much stronger (Banerjee 1962; Bunn and Jandl 1968). Later work established that contributions to stability comprise globin folding and heme affinity in a relatively complex, interdependent relationship between the strength of the coordinate bond, hydrophobic interactions between the globin and the porphyrin, and solvent exclusion from the heme pocket (Hargrove et al. 1994a, 1996a,b; Hargrove and Olson 1996; Tang et al. 1998).

Stability and heme affinity in hexacoordinate hemoglobins is not yet understood at this level, though a logical extension from work with Mb would suggest that having two coordinate His-heme iron bonds would further stabilize bound heme and foster protein stability. It is therefore surprising that heme retention in hexacoordinate SynHb would require additional support from the His117–heme vinyl bond. The present results indicate that this bond is necessitated by lowered heme affinity in the ferrous oxidation state; this is counterintuitive to the situation with pentacoordinate Hbs, where it is the ferric oxidation state that has a lower affinity for heme. Possible reasons for this fundamental difference between SynHb and pentacoordinate Hbs, along with the consequences of this unusual covalent bond on ligand binding are discussed below.

The role of the His117–heme vinyl bond in protein stability

The His117–heme vinyl bond is the dominant force holding heme in SynHb in both the ferrous and ferric oxidation states. This conclusion is most evident from the fact that even when the wild-type protein unfolds (presumably leading to the pentacoordinate spectral transition in Fig. 5C,D), the heme is retained. But if the covalent bond is broken, heme dissociation accompanies unfolding, which occurs under much milder conditions (Fig. 5E,F). The cause of the decrease in stability in the absence of the His117–heme vinyl bond could have both enthalpic and entropic origins. The loss of hydrophobic clustering in the heme pocket upon heme dissociation could lower the enthalpy of the folded state, and heme dissociation upon unfolding could increase the entropy of the unfolded state. However, the experiments presented here do not distinguish between these contributions.

Vu et al. (2004b) report that decoordination of both axial histidines accompanies protein unfolding in both ferric wt SynHb forms (with and without the covalent bond) due to the presence of isosbestic points in the spectral transitions. However, their ferric, unfolded spectra are similar to those reported here, including the difference seen with and without the covalent bond. The similarity of the spectrum lacking the covalent bond to that of unfolded hhMb, in which it has been demonstrated that heme is dissociated, supports heme dissociation in SynHb lacking the covalent bond. The spectra of unfolded SynHb containing the covalent bond, however, does not look like that of hhMb, suggesting that heme is retained. Furthermore, the presence of isosbestic points does not imply any particular final coordination state, but only suggests a two-state transition. However, differences in unfolded states could also result from the different methods of denaturation used in these two reports.

The most surprising discovery from the experiments illustrated in Figures 4, 5, and 6 is that, in the absence of the His117–heme vinyl bond, the ferrous hexacoordinate complex is significantly less stable than the ferric. This suggests the possibility that the influence of oxidation state on heme affinity in hexacoordinate Hbs is opposite to that in pentacoordinate Hbs like Mb, at least in the case of SynHb. Coordination studies of model heme compounds are consistent with the behavior of SynHb. Coordination of the heme iron is stronger in the ferric form than the ferrous due to the associated change in spin state of the iron upon binding of the second ligand (Nesset et al. 1996; Safo et al. 1997).

These results raise questions about the chemical nature of Hb stability in pentacoordinate versus hexacoordinate hemoglobins. Belief that the ferrous heme–His bond is stronger in Mb originated with the observation that heme transfer does not occur in this oxidation state (Hargrove et al. 1994a, 1996a,b, 1997; Hargrove and Olson 1996). However, there is no obvious explanation for why the single ferrous His–heme bond is stronger than the ferric bond in pentacoordinate Hbs, or why the opposite might be the case in hxHbs. It will be important to investigate whether this phenomenon is specific to SynHb, or if it is also seen in larger hexacoordinate Hbs lacking a covalent bond between the porphyrin ring and globin. It is possible that the truncated nature of SynHb requires the His117–heme vinyl bond in lieu of sufficient hydrophobic “waterproofing” of the proximal heme pocket (Liong et al. 2001). This hypothesis is consistent with the increase in solvation of the H117A proximal heme pocket seen in the crystal structures of the mutant protein in both oxidation states (Fig. 1E,F).

Effects of the His117–heme vinyl bond on hexacoordination and exogenous ligand binding

While the details of the effects of the His117–heme vinyl bond on protein stability were surprising, a role related to heme affinity was not unanticipated. A more curious result is the effect of the covalent bond on ligand binding. We have shown here that removal of this bond is felt more acutely on the ferric heme than the ferrous due to a higher affinity constant for hexacoordination in the ferric oxidation state. The ferrous hexacoordination equilibrium constant (K H2) is only approximately threefold larger in the H117A mutant protein, mainly due to an increase in the rate constant for His coordination (k H2) (Table 1). A much larger effect is observed on the redox potential, which drops by 85 mV in the H117A protein compared to wild-type SynHb. A shift of this magnitude suggests the possibility of much tighter hexacoordination in the ferric oxidation state (K H3); specifically, a 27-fold higher ratio of K H3/K H2 compared to wild-type SynHb. In fact, this 27-fold increase in K H3/K H2 combined with the threefold increase in K H2 for SynH117A predicts that K H3 for the mutant should be ~80-times greater than that of SynHb according to Equation 3.

The effect on K H3 can also be evaluated by measuring rate and affinity constants for exogenous ligand binding in the ferric oxidation state. The influence of K H3 on these reactions is described by Equations 4 and 5, respectively. In each case, the observed binding constant is reduced by a factor ~K H3. Therefore, K H3,H117A/K H3,Syn Hb should equal the ratio of k CN,obs,SynHb/k CN,obs,H117A, which should also be equal to K CN,obs,SynHb/K CN,obs,H117A. In fact, these values (14 and 10, respectively) are quite similar, and their correspondence supports assignment of this value to the increase in K H3 associated with breaking the His117–heme vinyl bond in SynHb.

However, these values are much lower than the value of 80 predicted by Equation 3, suggesting that the shift in redox potential for SynH117A compared to the wild-type protein derives from factors other than just effects of the covalent bond on axial histidine heme coordination. This conclusion is also supported by the fact that the redox potential for wild-type SynHb in which the protein is produced without the His117–heme vinyl bond is −150 mV (Lecomte et al. 2001), 45 mV higher than when this bond is present (Table 1), suggesting that the H117A mutation has additional effects beyond simply preventing this covalent bond. These effects might include movement of the H-helix (Fig. 1B,D), and the resulting increase in solvent accessibility to the proximal heme pocket (Fig. 1E,F).

In conclusion, the His117–heme vinyl bond in SynHb is not required to achieve the global structural architecture of the wild-type protein in either the ferric or ferrous oxidation states, although minor structural perturbations do affect hexacoordination of the heme. Removal of the covalent linkage also affects exogenous ligand binding, particularly in the ferric oxidation state. The different kinetic and equilibrium constants exhibited by SynHb with and without this bond can both be explained by changes in the affinity constant for intramolecular distal His coordination for the ferric heme (K H3). Furthermore, the ferrous protein lacking this bond is much less stable against denaturation than the ferric, and both oxidation states differ from wild-type SynHb in that heme is lost during unfolding of the H117A mutant protein. In contrast, when this bond is present in SynHb, even very high concentrations of GdmCl (7M) do not dissociate the heme from the globin in either oxidation state. Therefore, this covalent bond is likely present to prevent heme dissociation from the ferrous hexacoordinate complex of SynHb.

Materials and methods

Protein expression and purification

The expression and purification methods used for SynHb (containing the covalent bond) and SynH117A have been described in detail (Hvitved et al. 2001; Trent et al. 2001a; Smagghe et al. 2006b). Briefly, both proteins were cloned into a pET-29a (Novagen) vector and overexpressed in Escherichia coli BL21-CP (Invitrogen) for wt SynHb, and C41 (Avidis) for SynH117A. Following cell lysis, purification included ammonium sulphate fractionation, phenyl Sepharose hydrophobic exchange, DEAE Sepharose anion-exchange, and G-75 Sephadex size-exclusion chromatography. The purified protein was oxidized with an excess of potassium ferricyanide followed by desalting on a G-25 Sephadex column in 0.01 M potassium phosphate (pH 7.0). The purity of the resulting ferric hemoglobin was found to have a Soret/A280 absorbance ratio >4.7. The wild-type SynHb protein purified and used in all experiments below contains the covalent bond (Hoy et al. 2004; Trent et al. 2004).

Structure determination

Crystals of ferric SynHb and SynH117A were grown as described previously (Hoy et al. 2004). Ferrous crystals were produced by reducing ferric crystals for 5 min in 10 μL of well buffer containing 100 μM sodium dithionite. Before mounting, the crystals were extensively washed in well buffer lacking sodium dithionite to remove excess reductant. To confirm that the crystal remained in the ferrous oxidation state during diffraction, the crystal was removed from the goniometer after data collection and immediately dissolved in 100 μL of 100 mM potassium phosphate buffer (pH 7.0), equilibrated with CO. The presence of CO-bound Hb in this sample was observed spectrally with an Ocean-Optics USB2000 UV-VIS spectrophotometer, confirming that the crystal was still reduced, as ferric SynHb does not bind this ligand. As a control, this procedure (minus the initial treatment with dithionite prior to data collection) was repeated on ferric crystals and no CO-bound Hb absorption spectrum was observed. An additional check on oxidation state is the observation that ferrous crystals shatter in the presence of CO while ferric crystals do not. Finally, sulfur dioxide molecules (the reducing agent arising from sodium dithionite) (Weiland et al. 2004) are seen in complex with the protein in the electron density of the reduced crystals (see supplemental materials for comparative solvent electron density demonstrating the presence of sulfur dioxide).

Diffraction data for all three structures were collected on a Rigaku/MSC home source generator and solved by molecular replacement using the ferric SynHb structure as a model (1RTX.pdb). All three crystals were in space group P212121 with four monomers in the asymmetric unit. A table of data collection and refinement statistics for the datasets of ferrous SynHb, ferric SynH117A, and ferrous SynH117A is included in the supplemental materials. Atomic coordinates have been deposited in the Protein Data Bank (www.rcsb.org, PDB ID 2HZ1, 2HZ2, and 2HZ3 for ferrous SynHb, ferric SynH117A, and ferrous SynH117A, respectively). Figure 1 was created using the programs SPDBV (ExPASy), POV-Ray (Persistence of Vision Raytracer Pty. Ltd.), and SURFNET using a 1.4 Å probe (Laskowski 1995; Guex and Peitsch 1997).

Carbon monoxide binding

Flash photolysis was used to measure bimolecular carbon monoxide association rate constants as described previously (Hargrove 2000). CO binding was measured by rapid mixing using a BioLogic SFM 400 stopped flow reactor coupled to a MOS 250 spectrophotometer. Experiments were conducted at 20°C in 100 mM potassium phosphate (pH 7.0), using the method of Smagghe et al. (2006b). Time courses were collected at different CO concentrations, with at least three kinetic traces collected and averaged for each CO concentration. Data analysis and figure generation for ligand binding experiments were performed with the software Igor Pro (Wavemetrics, Inc.).

Spectroelectrochemistry

Potentiometric titrations were performed using the method of Altuve et al. (2004) as described in detail previously (Halder et al. 2006; Smagghe et al. 2006b). In brief, ferric Hbs (~30 μM) were titrated stepwise with a sodium dithionite solution (~40 mM) under anaerobic conditions. All reactions used saturated calomel (reference) and platinum (working) electrodes, and were carried out at 25°C in argon-saturated 0.1 M potassium phosphate (pH 7.0). Reduction was monitored by recording absorbance spectrum in the visible region (500–700 nm), and the corresponding cell potential was noted for each addition of sodium dithionite after equilibrium was reached. All midpoint potentials are reported with respect to a standard hydrogen electrode (SHE).

Midpoint potentials were extracted from the change in absorbance by fitting to the following equation (Bogumil et al. 1994):

graphic file with name 250equ6.jpg

F reduced is the fraction of reduced protein, given as the normalized change in absorbance at 560 nm, E obs is the observed cell potential, and E mid is the midpoint potential obtained by fitting the experimental data to Equation 6.

Cyanide binding kinetics

Time courses for cyanide binding were measured using a Varian Cary 50 spectrophotometer. Cyanide concentrations ranging from 0 M to 0.4 M were used, and each cyanide concentration was prepared using 100 mM potassium phosphate (pH 7.0), in a separate cuvette before protein addition, and the reaction was monitored for up to 12 h. The change in Soret absorbance was plotted versus time and fitted to a single exponential to obtain k obs at each [CN]. The linear dependence of this value is the observed bimolecular (pseudo first-order) rate constant (k′CN).

Equilibrium ferric ligand binding

Since the kinetic schemes for ligand binding in hxHbs are complex, it is important to measure affinities using equilibrium methods as well (Kundu et al. 2003a). Equilibrium association constants for cyanide and azide were measured by equilibrium titration. For both experiments, the protein concentrations were ~7 μM (well below the respective K D values). The samples were allowed to sit for 30 min following mixing to ensure equilibrium. Absorbance spectra were measured during ligand binding, and the position of the Soret peak was monitored to calculate fractional saturation. The peak shifted from ~408/410 nm (for the unliganded ferric protein) to ~415/418 nm (for the ligand-bound species) for SynHb/SynH117A, and the inflection of the first derivative of each spectrum was used to find the peak maximum. Equilibrium binding curves were fit to the following equation to extract equilibrium association constants:

graphic file with name 250equ7.jpg

where [L] is the ligand concentration, K CN,obs is the equilibrium association constant, and λL and λ3+ are the Soret peak wavelengths for the ferric ligand bound and unliganded ferric forms of the protein, respectively.

Guanidinium chloride unfolding

Guanidinium chloride (GdmCl) unfolding experiments were performed according to the method of Pace et al. (1990). A stock solution of 7.5 M GdmCl, 30 mM MOPS (pH 7.0) was used to prepare cuvettes varying in [GdmCl] from 0 to 5 M; 10 μM Hb was added to each cuvette. For ferrous experiments, all solutions contained 100 μM sodium dithionite and were purged with nitrogen. Samples were equilibrated for 1–2 h, then absorbance spectra were measured. Soret absorbance values were plotted as a function of GdmCl concentration, and the ΔG O and m values of the unfolding curves (Table 2) were determined according to a two-state model for denaturation (F ↔ U) using Equation 8 (Pace et al. 1990; Hargrove et al. 1994a; Hargrove and Olson 1996; Mok et al. 1996).

graphic file with name 250equ8.jpg

where A is absorbance, F refers to folded protein, U refers to unfolded protein, ΔG O is the Gibbs free energy, and m is the linear dependence of ΔG on GdmCl concentration. The unfolding midpoint values were calculated by dividing ΔG O by m, and are in good correspondence to those estimated by visual inspection of the unfolding curves.

Heme loss assay

Rate constants for heme dissociation were measured using the H64Y/V68F apomyoglobin method described by Hargrove et al. (1994b). For experiments with ferrous Hbs, all cuvettes and solutions were purged with nitrogen and the reaction buffers contained 100 μM sodium dithionite. Spectral changes were monitored over the course of 2000 min. For ferric proteins, heme loss is accompanied by an increased absorbance at 600 nm. For ferrous proteins, heme loss shifts the Soret peak to 434 for MbH64Y/V68F. The change in absorbance at the Soret peak as a function of time was fit with a single exponential to extract the rate constant for heme loss.

Electronic supplemental material

Supplemental materials include a table of data collection and refinement statistics, a table providing a closer examination of the changes in heme stereochemistry that occur among the ferric and ferrous crystal structures of SynHb and SynH117A, a figure providing a comparative look at solvent densities in the above structures to demonstrate the presence of sulfur dioxide in the reduced crystals, and a figure presenting flash photolysis data.

Acknowledgments

This work was made possible by the National Institutes of Health Award R01-GM065948, and support from the Iowa State University Plant Sciences Institute.

Footnotes

Supplemental material: see www.proteinscience.org

Reprint requests to: Mark Hargrove, 4114 Molecular Biology Building, Ames, IA 50011, USA; e-mail: msh@iastate.edu; fax: (515) 294-0520.

Abbreviations: Hb, hemoglobin; hxHbs, hexacoordinate hemoglobins; trHbs, truncated hemoglobins; SynHb, Synechocystis hemoglobin; His, histidine; SynH117A, SynHb mutant H117A; CO, carbon monoxide; CN, cyanide anion; N3, azide anion; GdmCl, Guanidinium Chloride; Mb, myoglobin; RMSD, root-mean-square deviation; obs, observed; hhMb, horse heart myoglobin.

References

  1. Altuve, A., Wang, L., Benson, D.R., and Rivera, M. 2004. Mammalian mitochondrial and microsomal cytochromes b(5) exhibit divergent structural and biophysical characteristics. Biochem. Biophys. Res. Commun. 314: 602–609. [DOI] [PubMed] [Google Scholar]
  2. Appleby, C.A. 1984. Leghemoglobin and rhizobium respiration. Annu. Rev. Plant Physiol. 35: 443–478. [Google Scholar]
  3. Arredondo-Peter, R., Hargrove, M.S., Sarath, G., Moran, J.F., Lohrman, J., Olson, J.S., and Klucas, R.V. 1997. Rice hemoglobins. Gene cloning, analysis, and O2-binding kinetics of a recombinant protein synthesized in Escherichia coli . Plant Physiol. 115: 1259–1266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Banerjee, R. 1962. Thermodynamic study of the heme-globin association. I. Dissociation equilibrium of metmyoglobin: thermodynamic data. Biochim. Biophys. Acta 64: 368–384. [DOI] [PubMed] [Google Scholar]
  5. Bogumil, R., Hunter, C.L., Maurus, R., Tang, H.L., Lee, H., Lloyd, E., Brayer, G.D., Smith, M., and Mauk, A.G. 1994. FTIR analysis of the interaction of azide with horse heart myoglobin variants. Biochemistry 33: 7600–7608. [DOI] [PubMed] [Google Scholar]
  6. Bunn, H.F. and Jandl, J.H. 1968. Exchange of heme among hemoglobins and between hemoglobin and albumin. J. Biol. Chem. 243: 465–475. [PubMed] [Google Scholar]
  7. Burmester, T., Weich, B., Reinhardt, S., and Hankeln, T. 2000. A vertebrate globin expressed in the brain. Nature 407: 520–523. [DOI] [PubMed] [Google Scholar]
  8. Burmester, T., Ebner, B., Weich, B., and Hankeln, T. 2002. Cytoglobin: A novel globin type ubiquitously expressed in vertebrate tissues. Mol. Biol. Evol. 19: 416–421. [DOI] [PubMed] [Google Scholar]
  9. Burr, A.H., Hunt, P., Wagar, D.R., Dewilde, S., Blaxter, M.L., Vanfleteren, J.R., and Moens, L. 2000. A hemoglobin with an optical function. J. Biol. Chem. 275: 4810–4815. [DOI] [PubMed] [Google Scholar]
  10. Couture, M., Das, T.K., Savard, P.Y., Ouellet, Y., Wittenberg, J.B., Wittenberg, B.A., Rousseau, D.L., and Guertin, M. 2000. Structural investigations of the hemoglobin of the cyanobacterium Synechocystis PCC6803 reveal a unique distal heme pocket. Eur. J. Biochem. 267: 4770–4780. [DOI] [PubMed] [Google Scholar]
  11. Dewilde, S., Kiger, L., Burmester, T., Hankeln, T., Baudin-Creuza, V., Aerts, T., Marden, M.C., Caubergs, R., and Moens, L. 2001. Biochemical characterization and ligand binding properties of neuroglobin, a novel member of the globin family. J. Biol. Chem. 276: 38949–38955. [DOI] [PubMed] [Google Scholar]
  12. Espenson, J.H. 2002. Chemical kinetics and reaction mechanisms, 2nd ed. McGraw-Hill College, Blacklick, OH.
  13. Falzone, C.J., Christie, V.B., Scott, N.L., and Lecomte, J.T. 2002. The solution structure of the recombinant hemoglobin from the cyanobacterium Synechocystis sp. PCC 6803 in its hemichrome state. J. Mol. Biol. 324: 1015–1029. [DOI] [PubMed] [Google Scholar]
  14. Guex, N. and Peitsch, M.C. 1997. SWISS-MODEL and the Swiss-PdbViewer: An environment for comparative protein modeling. Electrophoresis 18: 2714–2723. [DOI] [PubMed] [Google Scholar]
  15. Halder, P., Trent III, J.T., and Hargrove, M.S. 2006. The influence of the protein matrix on intramolecular histidine ligation in ferric and ferrous hexacoordinate hemoglobins. Proteins DOI 10.1002/prot.21210. [DOI] [PubMed]
  16. Hankeln, T., Ebner, B., Fuchs, C., Gerlach, F., Haberkamp, M., Laufs, T.L., Roesner, A., Schmidt, M., Weich, B., and Wystub, S., et al. 2005. Neuroglobin and cytoglobin in search of their role in the vertebrate globin family. J. Inorg. Biochem. 99: 110–119. [DOI] [PubMed] [Google Scholar]
  17. Hargrove, M.S. 2000. A flash photolysis method to characterize hexacoordinate hemoglobin kinetics. Biophys. J. 79: 2733–2738. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hargrove, M.S. and Olson, J.S. 1996. The stability of holomyoglobin is determined by heme affinity. Biochemistry 35: 11310–11318. [DOI] [PubMed] [Google Scholar]
  19. Hargrove, M.S., Krzywda, S., Wilkinson, A.J., Dou, Y., Ikeda-Saito, M., and Olson, J.S. 1994a. Stability of myoglobin: A model for the folding of heme proteins. Biochemistry 33: 11767–11775. [DOI] [PubMed] [Google Scholar]
  20. Hargrove, M.S., Singleton, E.W., Quillin, M.L., Ortiz, L.A., Phillips Jr, G.N., Olson, J.S., and Mathews, A.J. 1994b. His64(E7)→Tyr apomyoglobin as a reagent for measuring rates of hemin dissociation. J. Biol. Chem. 269: 4207–4214. [DOI] [PubMed] [Google Scholar]
  21. Hargrove, M.S., Barrick, D., and Olson, J.S. 1996a. The association rate constant for heme binding to globin is independent of protein structure. Biochemistry 35: 11293–11299. [DOI] [PubMed] [Google Scholar]
  22. Hargrove, M.S., Wilkinson, A.J., and Olson, J.S. 1996b. Structural factors governing hemin dissociation from metmyoglobin. Biochemistry 35: 11300–11309. [DOI] [PubMed] [Google Scholar]
  23. Hargrove, M.S., Whitaker, T., Olson, J.S., Vali, R.J., and Mathews, A.J. 1997. Quaternary structure regulates hemin dissociation from human hemoglobin. J. Biol. Chem. 272: 17385–17389. [DOI] [PubMed] [Google Scholar]
  24. Hou, S., Larsen, R.W., Boudko, D., Riley, C.W., Karatan, E., Zimmer, M., Ordal, G.W., and Alam, M. 2000. Myoglobin-like aerotaxis transducers in Archaea and Bacteria. Nature 403: 540–544. [DOI] [PubMed] [Google Scholar]
  25. Hoy, J.A., Kundu, S., Trent III, J.T., Ramaswamy, S., and Hargrove, M.S. 2004. The crystal structure of Synechocystis hemoglobin with a covalent heme linkage. J. Biol. Chem. 279: 16535–16542. [DOI] [PubMed] [Google Scholar]
  26. Hvitved, A.N., Trent III, J.T., Premer, S.A., and Hargrove, M.S. 2001. Ligand binding and hexacoordination in synechocystis hemoglobin. J. Biol. Chem. 276: 34714–34721. [DOI] [PubMed] [Google Scholar]
  27. Kundu, S., Premer, S.A., Hoy, J.A., Trent III, J.T., and Hargrove, M.S. 2003a. Direct measurement of equilibrium constants for high-affinity hemoglobins. Biophys. J. 84: 3931–3940. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kundu, S., Trent III, J.T., and Hargrove, M.S. 2003b. Plants, humans and hemoglobins. Trends Plant Sci. 8: 387–393. [DOI] [PubMed] [Google Scholar]
  29. Laskowski, R.A. 1995. SURFNET: A program for visualizing molecular surfaces, cavities, and intermolecular interactions. J. Mol. Graph. 13: 323–330. [DOI] [PubMed] [Google Scholar]
  30. Lecomte, J.T., Scott, N.L., Vu, B.C., and Falzone, C.J. 2001. Binding of ferric heme by the recombinant globin from the cyanobacterium Synechocystis sp. PCC 6803. Biochemistry 40: 6541–6552. [DOI] [PubMed] [Google Scholar]
  31. Liong, E., Dou, Y., Scott, E., Olson, J., and Phillips Jr., G. 2001. Waterproofing the heme pocket. Role of proximal amino acid side chains in preventing hemin loss from myoglobin. J. Biol. Chem. 276: 9093–9100. [DOI] [PubMed] [Google Scholar]
  32. Minning, D.M., Gow, A.J., Bonaventura, J., Braun, R., Dewhirst, M., Goldberg, D.E., and Stamler, J.S. 1999. Ascaris haemoglobin is a nitric oxide-activated “deoxygenase.”. Nature 401: 497–502. [DOI] [PubMed] [Google Scholar]
  33. Mok, Y.K., de Prat Gay, G., Butler, P.J., and Bycroft, M. 1996. Equilibrium dissociation and unfolding of the dimeric human papillomavirus strain-16 E2 DNA-binding domain. Protein Sci. 5: 310–319. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Nesset, M.J.M., Shokhirev, N.V., Enemark, P.D., Jacobson, S.E., and Walker, F.A. 1996. Models of the cytochromes. redox properties and thermodynamic stabilities of complexes of “hindered” iron(III) and iron(II) tetraphenylporphyrinates with substituted pyridines and imidazoles. Inorg. Chem. 35: 5188–5200. [Google Scholar]
  35. Pace, C.N., Shirley, B.A., and Thomson, J.A. 1990. Protein structure, pp. 311–329. IRL Press/Oxford University Press, Oxford, England.
  36. Pesce, A., Couture, M., Dewilde, S., Guertin, M., Yamauchi, K., Ascenzi, P., Moens, L., and Bolognesi, M. 2000. A novel two-over-two α-helical sandwich fold is characteristic of the truncated hemoglobin family. EMBO J. 19: 2424–2434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Poole, R.K. and Hughes, M.N. 2000. New functions for the ancient globin family: Bacterial responses to nitric oxide and nitrosative stress. Mol. Microbiol. 36: 775–783. [DOI] [PubMed] [Google Scholar]
  38. Safo, M.K., Nesset, M.J.M., Walker, F.A., Debrunner, P.G., and Scheidt, W.R. 1997. Models of the cytochromes. Axial ligand orientation and complex stability in iron(II) porphyrinates: The case of the noninteracting dπð orbitals. J. Am. Chem. Soc. 119: 9438–9448. [Google Scholar]
  39. Scott, N.L. and Lecomte, J.T. 2000. Cloning, expression, purification, and preliminary characterization of a putative hemoglobin from the cyanobacterium Synechocystis sp. PCC 6803. Protein Sci. 9: 587–597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Smagghe, B.J., Kundu, S., Hoy, J.A., Halder, P., Weiland, T.R., Savage, A., Venugopal, A., Goodman, M., Premer, S., and Hargrove, M.S. 2006a. Role of phenylalanine B10 in plant nonsymbiotic hemoglobins. Biochemistry 45: 9735–9745. [DOI] [PubMed] [Google Scholar]
  41. Smagghe, B.J., Sarath, G., Ross, E., Hilbert, J.L., and Hargrove, M.S. 2006b. Slow ligand binding kinetics dominate ferrous hexacoordinate hemoglobin reactivities and reveal differences between plants and other species. Biochemistry 45: 561–570. [DOI] [PubMed] [Google Scholar]
  42. Tang, Q., Kalsbeck, W.A., Olson, J.S., and Bocian, D.F. 1998. Disruption of the heme iron-proximal histidine bond requires unfolding of deoxymyoglobin. Biochemistry 37: 7047–7056. [DOI] [PubMed] [Google Scholar]
  43. Trent III, J.T. and Hargrove, M.S. 2002. A ubiquitously expressed human hexacoordinate hemoglobin. J. Biol. Chem. 277: 19538–19545. [DOI] [PubMed] [Google Scholar]
  44. Trent III, J.T., Hvitved, A.N., and Hargrove, M.S. 2001a. A model for ligand binding to hexacoordinate hemoglobins. Biochemistry 40: 6155–6163. [DOI] [PubMed] [Google Scholar]
  45. Trent III, J.T., Watts, R.A., and Hargrove, M.S. 2001b. Human neuroglobin, a hexacoordinate hemoglobin that reversibly binds oxygen. J. Biol. Chem. 276: 30106–30110. [DOI] [PubMed] [Google Scholar]
  46. Trent III, J.T., Kundu, S., Hoy, J.A., and Hargrove, M.S. 2004. Crystallographic analysis of synechocystis cyanoglobin reveals the structural changes accompanying ligand binding in a hexacoordinate hemoglobin. J. Mol. Biol. 341: 1097–1108. [DOI] [PubMed] [Google Scholar]
  47. Vinogradov, S.N., Hoogewijs, D., Bailly, X., Arredondo-Peter, R., Gough, J., Dewilde, S., Moens, L., and Vanfleteren, J.R. 2006. A phylogenomic profile of globins. BMC Evol. Biol. 6: 31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Vu, B.C., Jones, A.D., and Lecomte, J.T. 2002. Novel histidine-heme covalent linkage in a hemoglobin. J. Am. Chem. Soc. 124: 8544–8545. [DOI] [PubMed] [Google Scholar]
  49. Vu, B.C., Nothnagel, H.J., Vuletich, D.A., Falzone, C.J., and Lecomte, J.T. 2004a. Cyanide binding to hexacoordinate cyanobacterial hemoglobins: Hydrogen-bonding network and heme pocket rearrangement in ferric H117A Synechocystis hemoglobin. Biochemistry 43: 12622–12633. [DOI] [PubMed] [Google Scholar]
  50. Vu, B.C., Vuletich, D.A., Kuriakose, S.A., Falzone, C.J., and Lecomte, J.T. 2004b. Characterization of the heme-histidine cross-link in cyanobacterial hemoglobins from Synechocystis sp. PCC 6803 and Synechococcus sp. PCC 7002. J. Biol. Inorg. Chem. 9: 183–194. [DOI] [PubMed] [Google Scholar]
  51. Watts, R.A., Hunt, P.W., Hvitved, A.N., Hargrove, M.S., Peacock, W.J., and Dennis, E.S. 2001. A hemoglobin from plants homologous to truncated hemoglobins of microorganisms. Proc. Natl. Acad. Sci. 98: 10119–10124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Weber, R.E. and Vinogradov, S.N. 2001. Nonvertebrate hemoglobins: Functions and molecular adaptations. Physiol. Rev. 81: 569–628. [DOI] [PubMed] [Google Scholar]
  53. Weiland, T.R., Kundu, S., Trent 3rd, J.T., Hoy, J.A., and Hargrove, M.S. 2004. Bis-histidyl hexacoordination in hemoglobins facilitates heme reduction kinetics. J. Am. Chem. Soc. 126: 11930–11935. [DOI] [PubMed] [Google Scholar]
  54. Wittenberg, J.B., Bolognesi, M., Wittenberg, B.A., and Guertin, M. 2002. Truncated hemoglobins: A new family of hemoglobins widely distributed in bacteria, unicellular eukaryotes, and plants. J. Biol. Chem. 277: 871–874. [DOI] [PubMed] [Google Scholar]
  55. Wu, G., Wainwright, L.M., and Poole, R.K. 2003. Microbial globins. Adv. Microb. Physiol. 47: 255–310. [DOI] [PubMed] [Google Scholar]

Articles from Protein Science : A Publication of the Protein Society are provided here courtesy of The Protein Society

RESOURCES