Effects of inhibitory ligands on the aerobic carbon monoxide complex of cytochrome c oxidase.

1. In the presence of both CO and O2, ox heart cytochrome c oxidase forms a 607 nm-peak intermediate distinct from both the cytochrome a2+a3 2+CO and the cytochrome a3+a3 2+CO ('mixed-valence') CO complexes. 2. This aerobic CO compound is stable towards ferricyanide addition, but decomposed on treatment with ferric cytochrome a2 ligands such as formate, cyanide and azide. 3. Addition of formate or cyanves rise to a complex with alpha-peak at 598 nm, not identical with any azide complex of the free enzyme, but possibly a cytochrome a3 2+NO complex produced by oxidative attack of partially reduced O2 on the azide. 4. The results support the idea that although the initial reaction of oxygen is with cytochrome a3 2+, the next step is not an oxidation of the ferrous cytochrome a3, but a transfer of O2 to a neighbouring group, such as Cu+, to give Cu2+O2- or similar complexes. 5. The aerobic CO complex is then identified as a3+a3 2+COCu2+O2-; a similar compound ('Compound C') is formed by photolysis of a3+a3 2+CO (the 'mixed-valence' CO complex) in the presence of oxygen at low temperatures.     

Characterization of the Ni-Fe-C complex formed by reaction of carbon monoxide with the carbon monoxide dehydrogenase from Clostridium thermoaceticum by Q-band ENDOR.

Q-Band ENDOR studies on carbon monoxide dehydrogenase (CODH) from the acetogenic bacterium Clostridium thermoaceticum provided unambiguous evidence that the reaction of CO with CODH produces a novel metal center that includes at least one nickel, at least three iron sites, and the carbon of one CO. The 57Fe hyperfine couplings determined by ENDOR are similar to the values used in simulation of the M?ssbauer spectra [Lindahl et al. (1989) J. Biol. Chem. 265, 3880-3888]. EPR simulation using these AFe values is equally good for a 4Fe or a 3Fe center. The 13C ENDOR data are consistent with the binding of a carbon atom to either the Ni or the Fe component of the spin-coupled cluster. The 13C hyperfine couplings are similar to those determined earlier for the C0-bound form of the H cluster of the Clostridium pasteurianum hydrogenase, proposed to be the active site of hydrogen activation [Telser et al. (1987) J. Biol. Chem. 262, 6589-5694]. The 61 Ni ENDOR data are the first nickel ENDOR recorded for an enzyme. The EPR simulation using the ENDOR-derived hyperfine values for 61Ni is consistent with a single nickel site in the Ni-Fe-C complex. On the basis of our results and the M?ssbauer data [Lindahl et al. (1989) J. Biol. Chem. 265, 3880-3888], we propose the stoichiometry of the components of the Ni-Fe-C complex to be Ni1Fe3-4S greater than or equal to 4C1, with four acid-labile sulfides.     

Crystal structure of the carbon monoxide complex of human cytoglobin

Cytoglobin (Cgb) is a vertebrate heme‐containing globin‐protein expressed in a broad range of mammalian tissues. Unlike myoglobin, Cgb displays a hexa‐coordinated (bis‐hystidyl) heme iron atom, having the heme distal His81(E7) residue as the endogenous sixth ligand. In the present study, we crystallized human Cgb in the presence of a reductant Na₂S₂O₄ under a carbon monoxide (CO) atmosphere, and determined the crystal structure at 2.6 Å resolution. The CO ligand occupies the sixth axial position of the heme ferrous iron. Eventually, the imidazole group of His81(E7) is expelled from the sixth position and swings out of the distal heme pocket. The flipping motion of the His81 imidazole group accompanies structural readjustments of some residues (Gln62, Phe63, Gln72, and Ser75) in both the CD‐corner and D‐helix regions of Cgb. On the other hand, no significant structural changes were observed in other Cgb regions, for example, on the proximal side. These structural alterations that occurred as a result of exogenous ligand (CO) binding are clearly different from those observed in other vertebrate hexa‐coordinated globins (mouse neuroglobin, Drosophila melanogaster hemoglobin) and penta‐coordinated sperm whale myoglobin. The present study provides the structural basis for further discussion of the unique ligand‐binding properties of Cgb. Proteins 2011. © 2011 Wiley‐Liss, Inc.     

Infrared studies of carbon monoxide binding to carbon monoxide dehydrogenase/acetyl-CoA synthase from Moorella thermoacetica

Carbon monoxide dehydrogenase/acetyl-CoA synthase (CODH/ACS) is a bifunctional enzyme that catalyzes the reversible reduction of carbon dioxide into carbon monoxide and the coupled synthesis of acetyl-CoA from the carbon monoxide produced. Exposure of CODH/ACS from Moorella thermoacetica to carbon monoxide gives rise to several infrared bands in the 2100-1900 cm(-1) spectral region that are attributed to the formation of metal-coordinated carbon monoxide species. Infrared bands attributable to M-CO are not detected in the as-isolated enzyme, suggesting that the enzyme does not contain intrinsic metal-coordinated CO ligands. A band detected at 1996 cm(-1) in the CO-flushed enzyme is assigned as arising from CO binding to a metal center in cluster A of the ACS subunit. The frequency of this band is most consistent with it arising from a terminally coordinated Ni(I) carbonyl. Multiple infrared bands at 2078, 2044, 1970, 1959, and 1901 cm(-1) are attributed to CO binding at cluster C of the CODH subunit. All infrared bands attributed to metal carbonyls decay in a time-dependent fashion as CO(2) appears in the solution. These observations are consistent with the enzyme-catalyzed oxidation of carbon monoxide until it is completely depleted from solution during the course of the experiments.     

Reversible acidic-alkaline transition of the carbon monoxide complex of cytochrome c peroxidase

The Soret absorption band of the ferrous carbon monoxide (CO) complex of cytochrome c peroxidase exhibited a blue shift from 423.7 to 420 nm upon an increase in pH from 6.5 to 8.5. The spectral change was reversible with an isosbestic point at 422 nm. The pH dependence of this spectral change gave a sigmoidal curve fitted well to a theoretical curve of a cooperative release of two protons with a pK value of 7.5, indicating the existence of the acidic and alkaline forms of the ferrous CO enzyme. Upon irradiation of light flash (100 J of power and 30-microseconds), the heme-bound CO was readily dissociated in both acidic and alkaline forms with a quantum yield of approximately unity. On the other hand, the rate of recombination of the dissociated CO with the heme iron was significantly different between these two forms; the recombination rate constants were 1.1 X 10(3) and 3.0 X 10(4) M-1 S-1 at 25 degrees C for the acidic and alkaline forms, respectively. At intermediate pH values, kinetics of recombination were biphasic, consisting of the slow and fast processes with the appropriate rate constants mentioned above. When the fraction of the fast process was plotted against pH, the pH profile coincided with the spectrophotometric pH titration curve described above. Thus, it was concluded that the acidic and alkaline forms of the enzyme were responsible for the slow and fast processes, respectively. In infrared spectroscopy, the acidic form showed a narrow CO stretching band at 1922 cm-1 with a half-band width of 12.5 cm-1, while the alkaline form exhibited a broad CO-stretching band at 1948 cm-1 with a half-band width of 33 cm-1. Significance of these results are discussed in relation to the structure of the heme vicinity on the CO complex of cytochrome c peroxidase.     

Resonance Raman studies of dioxygen and carbon monoxide binding to imidazole-appended hemes.

Resonance Raman spectroscopy has been employed to probe the effects of proximal base strain on the bonding of O2 and CO in three synthetic hemins with covalently linked imidazole ligands. The strain is introduced by varying the length of the imidazole-containing side chain and by restricting the side chain flexibility with a phenyl ring. These hemins are abbreviated as "long," "short," and "stiff" hemins, respectively. In the deoxy state, the iron-imidazole stretching frequencies [nu(Fe--N epsilon)] for long, short, and stiff hemins are detected at 200, 207, and 204 cm-1, respectively. The strain induced in the iron-imidazole bond by the short hemin results in a higher nu(Fe--N epsilon) frequency, in contrast to the strain induced by sterically hindered 2-methylimidazole or 1,2-dimethylimidazole complexes in which the Fe--N epsilon bond is tilted and lengthened, but the imidazole ring remains perpendicular to the heme plane. However, in the short hemin, the plane of the imidazole ring may not be perpendicular to the plane of the porphyrin, altering the amount of pi-interaction (hence the strength of Fe--N epsilon bond) and the nature of normal mode containing Fe--N epsilon bond stretching. Upon CO binding, we have observed the nu(Fe--CO) stretching frequencies at 497 (long), 499 (short), and 496 cm-1 (stiff), somewhat lower than those reported by Mitchell et al. (Inorg. Chem., 1985, 24:967) for the chelated-heme X CO complexes (i.e., 501-506 cm-1). This is the first report of an iron-oxygen-associated vibration observed in solution for an unprotected heme.(ABSTRACT TRUNCATED AT 250 WORDS)     

Influence of carbon monoxide on hemoglobin-oxygen binding.

The oxygen dissociation curve and Bohr effect were measured in normal whole blood as a function of carboxyhemoglobin concentration [HbCO]. pH was changed by varying CO2 concentration (CO2 Bohr effect) or by addition of isotonic NaOH or HCl at constant PCO2 (fixed acid Bohr effect). As [HbCO] varied through the range of 2, 25, 50, and 75%, P50 was 26.3, 18.0, 11.6, and 6.5 mmHg, respectively. CO2 Bohr effect was highest at low oxygen saturations. This effect did not change as [HbCO] was increased. However, as [HbCO] was increased from 2 to 75%, the fixed acid Bohr factor increased in magnitude from -0.20 to -0.80 at very low oxygen saturations. The effect of molecular CO2 binding (carbamino) on oxygen affinity was eliminated at high [HbCO]. These results are consistent with the initial binding of O2 or CO to the alpha-chain of hemoglobin. The results also suggest that heme-heme interaction is different for oxygen than for carbon monoxide.     

Mitochondrial oxygenation of carbon monoxide.

A variety of mitochondria have been observed to oxidize 13CO to 13CO2 in the presence of dioxygen, and on the basis of earlier studies [Young & Caughey (1986) Biochemistry 25, 152-161; Young (1981) Ph.D. Dissertation, Colorado State University] this activity is attributed to cytochrome c oxidase. Implications of these findings in respect of some aspects of the pathological biochemistry of CO poisoning are discussed.     

Thin-layer microcalorimetric studies of oxygen and carbon monoxide binding to hemoglobin and myoglobin.

A thin-layer gas-solution microcalorimeter has been developed to study the binding reactions of gaseous ligands with ligand binding macromolecules. We have measured the enthalpy of binding oxygen and carbon monoxide to horse myoglobin, human hemoglobin A0 and sperm whale myoglobin in phosphate buffer at pH 7.6, with the enzyme reducing system of Hayashi. Reactions of human hemoglobin were also done under various buffer conditions in order to elucidate the Bohr effect. These binding reactions were found not to exhibit a detectable enthalpy change over the temperature range of 10 degrees C to 25 degrees C. The enzyme reducing system was shown to react with oxygen in a manner that releases a substantial amount of heat. This problem was corrected by using a minimum amount and by placing the buffer and enzyme system in the reference cell effectively cancelling the oxygen enzyme reaction heat as well as the heat of gas dissolution. It was also demonstrated that glucose-6-phosphate, one of the reducing system components, in 50 mM concentrations can influence the heat of binding oxygen and carbon monoxide to hemoglobin. This effect was shown to be absent in the myoglobins and also with hemoglobin at glucose-6-phosphate concentrations less than 5 mM.     

Synthesis of acetyl coenzyme A by carbon monoxide dehydrogenase complex from acetate-grown Methanosarcina thermophila.

The carbon monoxide dehydrogenase (CODH) complex from Methanosarcina thermophila catalyzed the synthesis of acetyl coenzyme A (acetyl-CoA) from CH3I, CO, and coenzyme A (CoA) at a rate of 65 nmol/min/mg at 55 degrees C. The reaction ended after 5 min with the synthesis of 52 nmol of acetyl-CoA per nmol of CODH complex. The optimum temperature for acetyl-CoA synthesis in the assay was between 55 and 60 degrees C; the rate of synthesis at 55 degrees C was not significantly different between pHs 5.5 and 8.0. The rate of acetyl-CoA synthesis was independent of CoA concentrations between 20 microM and 1 mM; however, activity was inhibited 50% with 5 mM CoA. Methylcobalamin did not substitute for CH3I in acetyl-CoA synthesis; no acetyl-CoA or propionyl coenzyme A was detected when sodium acetate or CH3CH2I replaced CH3I in the assay mixture. CO could be replaced with CO2 and titanium(III) citrate. When CO2 and 14CO were present in the assay, the specific activity of the acetyl-CoA synthesized was 87% of the specific activity of 14CO, indicating that CO was preferentially incorporated into acetyl-CoA without prior oxidation to free CO2. Greater than 100 microM potassium cyanide was required to significantly inhibit acetyl-CoA synthesis, and 500 microM was required for 50% inhibition; in contrast, oxidation of CO by the CODH complex was inhibited 50% by approximately 10 microM potassium cyanide.     

Effect on the heart muscle of experimental carbon monoxide poisoning.

The electron microscopic technique failed to reveal ultrastructural alterations specific for carbon monoxide poisoning. In this contract, the specificity of the lead salt detection method of calcium determination is discussed.     

Function and CO binding properties of the NiFe complex in carbon monoxide dehydrogenase from Clostridium thermoaceticum.

Adding 1,10-phenanthroline to carbon monoxide dehydrogenase from Clostridium thermoaceticum results in the complete loss of the NiFeC EPR signal and the CO/acetyl-CoA exchange activity. Other EPR signals characteristic of the enzyme (the gav = 1.94 and gav = 1.86 signals) and the CO oxidation activity are completely unaffected by the 1,10-phenanthroline treatment. This indicates that there are two catalytic sites on the enzyme; the NiFe complex is required for catalyzing the exchange and acetyl-CoA synthase reactions, while some other site is responsible for CO oxidation. The strength of CO binding to the NiFe complex was examined by titrating dithionite-reduced enzyme with CO. During the titration, the NiFeC EPR signal developed to a final spin intensity of 0.23 spin/alpha beta. The resulting CO titration curve (NiFeC spins/alpha beta vs CO pha beta) was fitted using two reactions: binding of CO to the oxidized NiFe complex, and reduction of the CO-bound species to a form that exhibits the NiFeC signal. Best fits yielded apparent binding constants between 6000 and 14,000 M-1 (Kd = 70-165 microM). This sizable range is due to uncertainty whether CO binds to all or only a small fraction (approximately 23%) of the NiFe complexes. Reduction of the CO-bound NiFe complex is apparently required to activate it for catalysis. The electron used for this reduction originates from the CO oxidation site, suggesting that delivery of a low-potential electron to the CO-bound NiFe complex is the physiological function of the CO oxidation reaction catalyzed by this enzyme.     

Microbial growth on carbon monoxide

The utilization of carbon monoxide as energy and/or carbon source by different physiological groups of bacteria is described and compared. Utilitarian CO oxidation which is coupled to the generation of energy for growth is achieved by aerobic and anaerobic eu- and archaebacteria. They belong to the physiological groups of aerobic carboxidotrophic, facultatively anaerobic phototrophic, and anaerobic acetogenic, methanogenic or sulfate-reducing bacteria. The key enzyme in CO oxidation is CO dehydrogenase which is a molybdo iron-sulfur flavoprotein in aerobic CO-oxidizing bacteria and a nickel-containing iron-sulfur protein in anaerobic ones. In carboxidotrophic and phototrophic bacteria, the CO-born CO₂ is fixed by ribulose bisphosphate carboxylase in the reductive pentose phosphate cycle. In acetogenic, methanogenic, and probably in sulfate-reducing bacteria, CODH/acetyl-CoA synthase directly incorporates CO into acetyl-CoA.In plasmid-harbouring carboxidotrophic bacteria, CO dehydrogenase as well as enzymes involved in CO₂ fixation or hydrogen utilization are plasmid-encoded. Structural genes encoding CO dehydrogenase were cloned from carboxidotrophic, acetogenic and methanogenic bacteria. Although they are clustered in each case, they are genetically distinct.Soil is a most important biological sink for CO in nature. While the physiological microbial groups capable of CO oxidation are well known, the type and nature of the microorganisms actually representing this sink are still enigmatic. We also tried to summarize the little information available on the nutritional and physicochemical requirements determining the sink strength. Because CO is highly toxic to respiring organisms even in low concentrations, the function of microbial activities in the global CO cycle is critical.     

Kinetic studies on the binding affinity of human hemoglobin for the 4th carbon monoxide molecule, L4.

L4, the affinity of hemoglobin for the 4th CO molecule, has been determined for human adult hemoglobin (HbA) as a function of pH and the presence of organic phosphates by measuring the kinetic parameters for the reaction. l'4, the rate of combination of CO with the triliganded molecule, was measured by flash photolysis while l4, the rate of CO dissociation for the ligand-saturated molecule, was measured by ligand replacement. L4 is pH-dependent and affected by 2,3-diphosphoglycerate. Additionally, this pH dependence of the high affinity state is largely eliminated by carboxypeptidase A digestion. L4 for human fetal hemoglobin (HbF) in phosphate buffers was also determined and found to be pH-dependent. These results cannot be reconciled within the framework of the two-state allosteric model. Additional structures in the conformational equilibrium due to either intermediates in the T to R transition or two or more R states must exist.     

Studies on carbon monoxide binding by shark haemoglobin.

The kinetics of the reactions of Pacific-porbeagle haemoglobin with CO were studied by flash-photolysis and stopped-flow methods, and the equilibrium binding curves for CO were measured in spectrophotometric titrations. Measurements were made in the pH range 6-8 and in the temperature range 0-40 degrees C. The results are discussed in terms of the allosteric model proposed by Monod, Wyman & Changeux [(1965) J. Mol. Biol. 12, 88-118]. Within this framework the results indicate that in the R-state the haem groups fall into two classes of different reactivity with different spectral characteristics, but that in the T-state the groups may be essentially equivalent. The physiological importance of the temperature-insensitivity of the equilibrium ligand-binding curves for porbeagle haemoglobin is discussed.