Modulating carbon monoxide binding affinity and kinetics in myoglobin: the roles of the distal histidine and the heme pocket docking site

 Myoglobin has long served as a model system for understanding the relations between protein structure, dynamics, and function. Its ability to discriminate between toxic CO and vital O₂, two small ligands that are almost equivalent in size and dipole moment, has attracted much attention. To understand discrimination and reversible ligand-binding in Mb, both the bound state and the "docked" state that leads to binding need to be studied. We have reported previously the nearly linear Fe–C–O geometry of bound CO and the nearly orthogonal geometry of docked CO [Lim et al. (1995), Science 269 : 962]. With the exception of X-ray structures, a preponderance of evidence points to a nearly linear Fe–C–O geometry and calls into question the proposal that the highly conserved distal histidine forces CO to bind in a nonoptimal geometry. The differences between the bound CO structures determined using IR and X-ray methods might arise from a water molecule hydrogen bonded to the distal histidine in some of the unit cells. Discrimination by Mb is manifested not only thermodynamically but also kinetically. Time-resolved CO rebinding studies that compare Mb with microperoxidase suggest that the heme pocket docking site in Mb exerts steric control of the ligand rebinding rate, slowing the rate of CO binding by a factor of more than 10⁴. Received, accepted: 23 May 1997     

On the significance of hydrogen bonds for the discrimination between CO and O2 by myoglobin

 Quantum chemical geometry optimisations have been performed on realistic models of the active site of myoglobin using density functional methods. The energy of the hydrogen bond between the distal histidine residue and CO or O₂ has been estimated to be 8 kJ/mol and 32 kJ/mol, respectively. This 24 kJ/mol energy difference accounts for most of the discrimination between CO and O₂ by myoglobin (about 17 kJ/mol). Thus, steric effects seem to be of minor importance for this discrimination. The Fe—C and C—O vibrational frequencies of CO-myoglobin have also been studied and the results indicate that CO forms hydrogen bonds to either the distal histidine residue or a water molecule during normal conditions. We have made several attempts to optimise structures with the deprotonated nitrogen atom of histidine directed towards CO. However, all such structures lead to unfavourable interactions between the histidine and CO, and to ν_CO frequencies higher than those observed experimentally. Received: 7 July 1998 / Accepted: 26 October 1998     

Myoglobin and CO: structure, energetics, and disorder

 The physiological significance of steric inhibition of CO binding to heme proteins is a fundamentally quantitative issue, since it has meaning only relative to the enhancement of O₂ binding by the protein. Previously, difficulties in reconciling structural and energetic data have hindered a quantitative assessment of the energetic cost of distorting the bound CO. Recent progress on both fronts suggests that the energetic cost of CO distortion in myoglobin is quite small. However, distortion of the surrounding protein to accommodate the heme-CO complex may contribute significantly to the free energy of CO binding. Polarized IR measurements on single crystals of MbCO not only yield precise structural information on the CO geometry, but also address the limitations of conventional structural refinement by characterizing conformational disorder in the crystalline state. Received, accepted: 23 May 1997     

Resonance Raman enhancement of the Mn-N-O bending mode in nitrosyl manganese "strapped" and "open" heme complexes

Resonance Raman spectra of the MnII-NO moiety in synthetic nitrosyl manganese heme complexes with and without steric hindrance are reported. The "strapped" hemes having a hydrocarbon strap (variable length) across one face of the heme hinder the perpendicular bonding of a linear ligand. These complexes were employed to investigate the effects of ligand distortion (primarily tilting) on Mn-NO stretching, Mn-N-O bending, and N-O stretching modes. It is demonstrated that ligand distortion in the MnII-NO system is a valid mechanism for causing the resonance enhancement of the Mn-N-O bending mode, similar to that observed in the FeII-CO system (Yu, N.-T., E. A. Kerr, B. Ward, and C. K. Chang. 1983. Biochemistry. 22:4534-4540). More interesting is the observation of the delta(Mn-N-O) enhancement caused by the tilting of the trans Mn-N epsilon bond in the "open" heme complexes (e.g., heme-5 and proto-1X dimethylester) with 1,2-dimethylimidazole or piperidine as a base. The nu(Mn-NO) and nu(N-O) modes exhibit an increase and a decrease, respectively, as the strap length decreases (hence the steric hindrance increases). Both nu(Mn-NO) and nu(N-O) frequencies are insensitive to the strength of the trans base. The results from "strapped" and "open" model heme systems imply that the Mn-N-O geometry is essentially linear and perpendicular in the nitrosyl complexes of monomeric manganese insect hemoglobin CTT IV and sperm whale myoglobin. The unusually low nu(N-O) frequency in the manganese myoglobin complex may be caused by the distal histidine-NO interaction. The delta(Mn-N-O) enhancement in both nitrosyl manganese CTT IV and nitrosyl manganese myoglobin may be caused by a tilting of the Mn"-Nf (proximal histidine) bond.     

Interaction of the active site of the Ni–Fe–Se hydrogenase from Desulfovibrio vulgaris Hildenborough with carbon monoxide and oxygen inhibitors

The study of Ni–Fe–Se hydrogenases is interesting from the basic research point of view because their active site is a clear example of how nature regulates the catalytic function of an enzyme by the change of a single residue, in this case a cysteine, which is replaced by a selenocysteine. Most hydrogenases are inhibited by CO and O₂. In this work we studied these inhibition processes for the Ni–Fe–Se hydrogenase from Desulfovibrio vulgaris Hildenborough by combining catalytic activity measurements, followed by mass spectrometry or chronoamperometry, with Fourier transform IR spectroscopy experiments. The results show that the CO inhibitor binds to Ni in both conformations of the active site of this hydrogenase in a way similar to that in standard Ni–Fe hydrogenases, although in one of the CO-inhibited conformations the active site of the Ni–Fe–Se hydrogenase is more protected against the attack by O₂. The inhibition of the Ni–Fe–Se hydrogenase activity by O₂ could be explained by oxidation of the terminal cysteine ligand of the active-site Ni, instead of the direct attack of O₂ on the bridging site between Ni and Fe.     

Molecular orbital study of the hydroxylation of benzene and monofluorobenzene catalysed by iron-oxo porphyrin π cation radical complexes

 The reaction mechanism for the hydroxylation of benzene and monofluorobenzene, catalysed by a ferryl-oxo porphyrin cation radical complex (compound) is described by electronic structure calculations in local spin density approximation. The active site of the enzyme is modelled as a six-coordinated (Por⁺)Fe(IV)O a_2u complex with imidazole or H₃CS^– as the axial ligand. The substrates under study are benzene and fluorobenzene, with the site of attack in para, meta and ortho position with respect to F. Two reaction pathways are investigated, with direct oxygen attack leading to a tetrahedral intermediate and arene oxide formation as a primary reaction step. The calculations show that the arene oxide pathway is distinctly less probable, that hydroxylation by an H₃CS^––coordinated complex is energetically favoured compared with imidazole, and that the para position with respect to F is the preferred site for hydroxylation. A partial electron transfer from the substrate to the porphyrin during the reaction is obtained in all cases. The resulting charge distribution and spin density of the substrates reveal the transition state as a combination of a cation and a radical σ-adduct intermediate with slightly more radical character in the case of H₃CS^– as axial ligand. A detailed analysis of the orbital interactions along the reaction pathway yields basically different mechanisms for the modes of substrate–porphyrin electron transfer and rupture of the Fe–O bond. In the imidazole-coordinated complex an antibonding π*(Fe–O) orbital is populated, whereas in the H₃CS^––coordinated system a shift of electron density occurs from the Fe–O bond region into the Fe–S bond. Received: 1 July 1995 / Accepted: 18 December 1995     

Will the real FeCO please stand up?

 The paradigm that nature protects us from CO poisoning by forcing the bound CO to bend over in heme proteins, thereby reducing its binding affinity, is now in textbooks, but is nevertheless problematic. Results from vibrational spectroscopy give no evidence for bent CO, although X-ray crystallography continues to indicate appreciable distortions in myoglobin. However, the energetic significance of the discrepancy is doubtful, since new Density Functional Theory calculations indicate that much less energy is required to distort the CO than had been thought, perhaps 2 kcal/mol or less. Binding studies on site-directed mutants of myoglobin suggest that steric hindrance by the distal histidine is worth ca. 1 kcal/mol. While the distal histidine does account for the discrimination by Mb against CO and in favor of O₂, most of the effect is due to its H-bond with bound O₂. Received, accepted: 23 May 1997     

Resonance Raman Studies of Co—O2 and O—O Stretching Vibrations in Oxy-Cobalt Hemes

Strong evidence suggests that the stretching vibration of the bound oxygen can be perturbed by an accidentally degenerate porphyrin ring mode, resulting in two split frequencies. In the Co(II)(TpivPP) (pyridine) ¹⁸O₂ complex, we demonstrate that the ν(¹⁸O—¹⁸O) mode, after being shifted from its ν(¹⁶O—¹⁶O) value at 1,156 cm^-1, undergoes a resonance interaction with the 1,080 cm^-1 porphyrin mode, giving rise to two lines at 1,067 and 1,089 cm^-1. In the O₂ complex of Co(II) mesoporphyrin IX-substituted sperm whale myoglobin, we observed a dramatic intensity increase at 1,132 cm^-1 upon ¹⁶O₂ → ¹⁸O₂ substitution, which is due to the reappearance of the 1,132-cm^-1 porphyrin mode after the removal of resonance conditions. A decrease in O₂ binding affinity, caused by the proximal base tension, corresponds to an increase in the Co—O₂ stretching frequency. The ν(Co—O₂) at 527 cm^-1 for the low affinity Co(II)(TpivPP)(1,2-Me₂Im) O₂ complex is 11 cm^-1 higher than the 516-cm^-1 value for the high affinity complex (with N-MeIm replacing 1,2-Me₂Im). However, in the corresponding iron complexes the reverse behavior is observed, i.e., the ν(Fe—O₂) decreases for the (1,2-Me₂Im) complex. There is a 24-cm^-1 difference in the Co—O₂ stretching frequencies between Co(II)(TpivPP)(N-MeIm)O₂ (at 516 cm^-1) and oxy meso CoMb (at 540 cm^-1), suggesting a protein induced distortion of the Co—O—O linkage. However, the values for ν(Fe—O₂) are nearly identical between Fe(II)(TpivPP)(N-MeIm)O₂ (at 571 cm^-1) and oxy Mb (at 573 cm^-1), indicating that O₂ binds to myoglobin in the same manner as in the sterically unhindered “picket fence” complex. Evidence is presented that suggests the presence of two dioxygen stretching frequencies due to two different conformers in each of the N-MeIm and 1,2-Me₂Im complex of oxy Co(II)(TpivPP).     

Structural characterization of a carbon monoxide adduct of a heme–DNA complex

Abstract  

The structure of a carbon monoxide (CO) adduct of a complex between heme and a parallel G-quadruplex DNA formed from a single repeat sequence of the human telomere, d(TTAGGG), has been characterized using ¹H and ¹³C NMR spectroscopy and density function theory calculations. The study revealed that the heme binds to the 3′-terminal G-quartet of the DNA though a π–π stacking interaction between the porphyrin moiety of the heme and the G-quartet. The π–π stacking interaction between the pseudo-C ₂-symmetric heme and the C ₄-symmetric G-quartet in the complex resulted in the formation of two isomers possessing heme orientations differing by 180° rotation about the pseudo-C ₂ axis with respect to the DNA. These two slowly interconverting heme orientational isomers were formed in a ratio of approximately 1:1, reflecting that their thermodynamic stabilities are identical. Exogenous CO is coordinated to heme Fe on the side of the heme opposite the G-quartet in the complex, and the nature of the Fe–CO bond in the complex is similar to that of the Fe–CO bonds in hemoproteins. These findings provide novel insights for the design of novel DNA enzymes possessing metalloporphyrins as prosthetic groups.     

Multinuclear (13C, 17O, 57Fe) NMR studies of carbonmonoxy heme proteins and synthetic model compounds

13C, 17O and 57Fe NMR spectra of several carbonmonoxy hemoprotein models with varying polar and steric effects of the distal organic superstructure, constraints of the proximal side, and porphyrin ruffling are reported. Both heme models and heme proteins obey a similar excellent linear delta(13C) versus nu(C-O) relationship which is primarily due to modulation of pi-back-bonding from the Fe d(pi) to CO pi* orbital by the distal pocket polar interactions. The lack of correlation between delta(13C) and delta(17O) suggests that the two probes do not reflect a similar type of electronic and structural perturbation. delta(17O) is not primarily influenced by the local distal field interactions and does not correlate with any single structural property of the Fe-C-O unit; however, atropisomerism and deformation of the porphyrin geometry appear to play a significant role. 57Fe shieldings vary by nearly 900 ppm among various hemes and an excellent correlation was found between delta(57Fe) and the absolute crystallographic average displacement of the meso carbon atoms, /Cm/, relative to the porphyrin core mean plane. The excellent correlation between iron-57 shieldings and the average shieldings of the meso carbons of the porphyrin skeleton of TPP derivatives suggests that the two probes reflect a similar type of electronic and structural perturbation which is primarily porphyrin ruffling.     

Myoglobin discriminates between O2, NO, and CO by electrostatic interactions with the bound ligand

 Most biological substrates have distinctive sizes, shapes, and charge distributions which can be recognized specifically by proteins. In contrast, myoglobin must discriminate between the diatomic gases O₂, CO, and NO which are apolar and virtually the same size. Selectivity occurs at the level of the covalent Fe-ligand complexes, which exhibit markedly different bond strengths and electrostatic properties. By pulling a water molecule into the distal pocket, His64(E7)¹ inhibits the binding of all three ligands by a factor of ∼10 compared to that observed for protoheme-imidazole complexes in organic solvents. In the case of O₂ binding, this unfavorable effect is overcome by the formation of a strong hydrogen bond between His64(E7) and the highly polar FeO₂ complex. This favorable electrostatic interaction stabilizes the bound O₂ by a factor of ∼1000, and the net result is a 100-fold increase in overall affinity compared to model hemes or mutants with an apolar residue at position 64. Electrostatic interaction between FeCO and His64 is very weak, resulting in only a two- to three-fold stabilization of the bound state. In this case, the inhibitory effect of distal pocket water dominates, and a net fivefold reduction in K _CO is observed for the wild-type protein compared to mutants with an apolar residue at position 64. Bound NO is stabilized ∼tenfold by hydrogen bonding to His64. This favorable interaction with FeNO exactly compensates for the tenfold inhibition due to the presence of distal pocket water, and the net result is little change in K _NO when the distal histidine is replaced with apolar residues. Thus, it is the polarity of His64 which allows discrimination between the diatomic gases. Direct steric hindrance by this residue plays a minor role as judged by: (1) the independence of K _O2, K _CO, and K _NO on the size of apolar residues inserted at position 64, and (2) the observation of small decreases, not increases, in CO affinity when the mobility of the His64 side chain is increased. Val68(E11) does appear to hinder selectively the binding of CO. However, the extent is no more than a factor of 2–5, and much smaller than electrostatic stabilization of bound O₂ by the distal histidine. Received, accepted: 23 May 1997     

Effects of local protein environment on the binding of diatomic molecules to heme in myoglobins. DFT and dispersion-corrected DFT studies

The heme-AB binding energies (AB = CO, O₂) in a wild-type myoglobin (Mb) and two mutants (H64L, V68N) of Mb have been investigated in detail with both DFT and dispersion-corrected DFT methods, where H64L and V68N represent two different, opposite situations. Several dispersion correction approaches were tested in the calculations. The effects of the local protein environment were accounted for by including the five nearest surrounding residues in the calculated systems. The specific role of histidine-64 in the distal pocket was examined in more detail in this study than in other studies in the literature. Although the present calculated results do not change the previous conclusion that the hydrogen bonding by the distal histidine-64 residue plays a major role in the O₂/CO discrimination by Mb, more details about the interaction between the protein environment and the bound ligand have been revealed in this study by comparing the binding energies of AB to a porphyrin and the various myoglobins. The changes in the experimental binding energies from one system to another are well reproduced by the calculations. Without constraints on the residues in geometry optimization, the dispersion correction is necessary, since it improves the calculated structures and energetic results significantly.     

Characteristic vibration patterns of odor compounds from bread-baking volatiles upon protein binding: density functional and ONIOM study and principal component analysis

As the mechanism underlying the sense of smell is unclear, different models have been used to rationalize structure–odor relationships. To gain insight into odorant molecules from bread baking, binding energies and vibration spectra in the gas phase and in the protein environment [7-transmembrane helices (7TMHs) of rhodopsin] were calculated using density functional theory [B3LYP/6-311++G(d,p)] and ONIOM [B3LYP/6-311++G(d,p):PM3] methods. It was found that acetaldehyde (“acid” category) binds strongly in the large cavity inside the receptor, whereas 2-ethyl-3-methylpyrazine (“roasted”) binds weakly. Lys296, Tyr268, Thr118 and Ala117 were identified as key residues in the binding site. More emphasis was placed on how vibrational frequencies are shifted and intensities modified in the receptor protein environment. Principal component analysis (PCA) suggested that the frequency shifts of C–C stretching, CH₃ umbrella, C = O stretching and CH₃ stretching modes have a significant effect on odor quality. In fact, the frequency shifts of the C–C stretching and C = O stretching modes, as well as CH₃ umbrella and CH₃ symmetric stretching modes, exhibit different behaviors in the PCA loadings plot. A large frequency shift in the CH₃ symmetric stretching mode is associated with the sweet-roasted odor category and separates this from the acid odor category. A large frequency shift of the C–C stretching mode describes the roasted and oily-popcorn odor categories, and separates these from the buttery and acid odor categories.     

Computer simulation of the interaction of α3hclix of λ Cro protein with DNA and origin of sequence specific recognition

The conformation of the α3 helix of Cro protein (residues 27–36) of bacteriophageλ is optimised by the damped least square minimization technique, with the steric constraint that Cα atom positions should match the crystallographic data available to date. On the basis of minimization of total interaction and conformation energy, models for complexes of this peptide sequence with heptanucleotide duplexes from native and altered O_R3 operator are obtained in the major groove of B DNA. Analysis of the energetics for 3 sequences of the DNA show that binding strength is derived mainly from the interaction of side chains of the peptide with DNA. Sequence specificity (maximum difference in binding energy for different DNA sequences) is due to hydrogen bonding interaction. A small amount of sequence specificity is derived from non-bonded interaction also. Stereochemical aspects of peptide DNA interaction and their role in DNA recognition are discussed in this paper.     

Ultrahigh resolution structures of nitrophorin 4: heme distortion in ferrous CO and NO complexes

Nitrophorin 4 (NP4), a nitric oxide (NO)-transport protein from the blood-sucking insect Rhodnius prolixus, uses a ferric (Fe3+) heme to deliver NO to its victims. NO binding to NP4 induces a large conformational change and complete desolvation of the distal pocket. The heme is markedly nonplanar, displaying a ruffling distortion postulated to contribute to stabilization of the ferric iron. Here, we report the ferrous (Fe2+) complexes of NP4 with NO, CO, and H2O formed after chemical reduction of the protein and the characterization of these complexes by absorption spectroscopy, flash photolysis, and ultrahigh-resolution crystallography (resolutions vary from 0.9 to 1.08 A). The absorption spectra, both in solution and in the crystal, are typical for six-coordinated ferrous complexes. Closure and desolvation of the distal pocket occurs upon binding CO or NO to the iron regardless of the heme oxidation state, confirming that the conformational change is driven by distal ligand polarity. The degree of heme ruffling is coupled to the nature of the ligand and the iron oxidation state in the following order: (Fe3+)-NO > (Fe2+)-NO > (Fe2+)-CO > (Fe3+)-H2O > (Fe2+)-H2O. The ferrous coordination geometry is as expected, except for the proximal histidine bond, which is shorter than typically found in model compounds. These data are consistent with heme ruffling and coordination geometry serving to stabilize the ferric state of the nitrophorins, a requirement for their physiological function. Possible roles for heme distortion and NO bending in heme protein function are discussed.     

Two-step concerted mechanism for alkane hydroxylation on the ferryl active site of methane monooxygenase

 A two-step concerted mechanism for the conversion of methane to methanol catalyzed by soluble methane monooxygenase (sMMO) is discussed. We propose that the enzymatic reaction mechanism is essentially the same as that of the gas-phase methane-methanol conversion by the bare FeO⁺ complex. In the initial stage of our mechanism, the ferryl (Fe—O) "iron" active site of intermediate Q and substrate methane come into contact to form the initial Q (CH₄) complex with an OFe—CH₄ bond. The C—H bonds of methane are significantly weakened by the formation of a five-coordinate carbon species, through orbital interactions between a C _3v - or D _2d -distorted methane and the Fe—O active site. The important transition state for an H atom abstraction exhibits a four-centered structure. The generated intermediate involves an HO—Fe—CH₃ moiety, and it is then converted into the final product complex including methanol as a ligand through a methyl migration that occurs via a three-centered transition state. The two-step concerted mechanism is consistent with recent experiments on regioselectivity of enzyme-catalyzed alkane hydroxylations. Received: 15 September 1997 / Accepted: 20 December 1997     

Ligand binding equilibrium and kinetic measurements on the dimeric myoglobin of Busycon canaliculatum and the comparative ligand binding of diverse non-cooperative heme proteins

The rate constants and delta H degrees for the non-cooperative dimeric Busycon myoglobin are: oxygen, k' = 4.75 X 10(7) M-1 sec-1, k = 71 sec-1, and CO, l'= 3.46 X 10(5) M-1 sec-1, l = 0.0052 sec-1 at 20 degrees C, pH 7, delta H degrees = -3 kcal/mol for O2 and CO.2. Log-log plots of k vs K for oxygen and of l' vs L for CO binding for numerous non-cooperative hemoglobins and myoglobins point to a large steric influence of the protein on heme ligation reactions. Many of the proteins behave as "R" state for one ligand, but "T" for the other.     

XANES of carboxy and cyanomet-myoglobin The role of the distal histidine in the bent Fe−C−O configuration

The ligand bonding geometry of carboxy-and cyanomet-myoglobin (MbCO and MbCN) has been measured by the XANES method (X-ray Absorption Near Edge Structure). A comparison between the ligand bonding geometry of carboxy- and cyanomet-myoglobin and of chelated protoheme methyl ester shows that the bent Fe–C–O configuration is the same in both systems. Therefore, we suggest that this configuration is not associated with any steric contraint imposed by the side chains of the aminoacid residues at the distal side of the heme pocket.     

Harmonic and anharmonic dynamics of Fe-CO and Fe-O(2) in heme models

We present density-functional molecular dynamics simulations of FeP(Im)(AB) heme models (AB = CO, O(2), Im = imidazole) as a way of sketching the dynamic motion of the axial ligands at room temperature. The FeP(Im)(CO) model is characterized by an essentially upright FeCO unit, undergoing small deviations with respect to its linear equilibrium structure (bending and tilting up to 10 degrees and 7 degrees, often occur). The motion of the carbon monoxide ligand is found to be quite complex and fast, its projection on the porphyrin plane sampling all the porphyrin quadrants in a short time ( approximately 0.5 ps). Simultaneously, the imidazole ligand rotates slowly around the Fe-N(epsilon) bond. In contrast to carbon monoxide, the oxygen ligand in FeP(Im)(O(2)) prefers a conformation where the projection of the O-O axis on the porphyrin plane bisects one of the porphyrin quadrants. A transition to other quadrants takes place through an O-O/Fe-N(p) overlapping conformation, within 4-6 ps. Further details of these mechanisms and their implications are discussed.