Studies on the binding of nitrogenous bases to protoporphyrin IX iron(II) in aqueous solution at high pH values. Part I. Pyridine and imidazole ligands

Protoporphyrin IX iron(II) revisited. An overview of the Mössbauer spectroscopic parameters of low-spin porphyrin iron(II) complexes

Mössbauer parameters of low-spin six-coordinate [Fe(II)(Por)L2] complexes (where Por is a synthetic porphyrin; L is a nitrogenous aliphatic, an aromatic base or a heterocyclic ligand, a P-bonding ligand, CO or CN) and low-spin [Fe(Por)LX] complexes (where L and X are different ligands) are reported. A known point charge calculation approach was extended to investigate how the axial ligands and the four porphyrinato-N atoms generate the observed quadrupole splittings (ΔEQ) for the complexes. Partial quadrupole splitting (p.q.s.) and partial chemical shifts (p.c.s.) values were derived for all the axial ligands, and porphyrins reported in the literature. The values for each porphyrin are different emphasising the importance/uniqueness of the [Fe(PPIX)] moiety, (which is ubiquitous in nature). This new analysis enabled the construction of figures relating p.c.s and p.q.s values. The relationships presented in the figures indicates that strong field ligands such as CO can, and do change the sign of the electric field gradient in the [Fe(II)(Por)L2] complexes. The limiting p.q.s. value a ligand can have and still form a six-coordinate low-spin [Fe(II)(Por)L2] complex is established. It is shown that the control the porphyrin ligands exert on the low-spin Fe(II) atom limits its bonding to a defined range of axial ligands; outside this range the spin state of the iron is unstable and five-coordinate high-spin complexes are favoured. Amongst many conclusions, it was found that oxygen cannot form a stable low-spin [Fe(II)(Por)L(O2)] complex and that oxy-haemoglobin is best described as an [Fe(III)(Por)L(O2-)] complex, the iron is ferric bound to the superoxide molecule.

The binding of nitrogen-donor ligands to the ferric and ferrous forms of cytochrome P450 enzymes

The cytochrome P450 superfamily of heme-thiolate monooxygenase enzymes can catalyse various oxidation reactions. The addition of a substrate or an inhibitor ligand induces changes in the absorption spectrum of these enzymes and UV-visible (UV-vis) absorbance spectroscopy is the most common and readily available technique used to interrogate their heme and active site environment. Nitrogen-containing ligands can inhibit the catalytic cycle of heme enzymes by interacting with the heme. Here we evaluate the binding of imidazole and pyridine-based ligands to the ferric and ferrous forms of a selection of bacterial cytochrome P450 enzymes using UV-visible absorbance spectroscopy. The majority of these ligands interact with the heme as one would expect for type II nitrogen directly coordinated to a ferric heme-thiolate species. However, the spectroscopic changes observed in the ligand-bound ferrous forms indicated differences in the heme environment across these P450 enzyme/ligand combinations. Multiple species were observed in the UV-vis spectra of the ferrous ligand-bound P450s. None of the enzymes gave rise to the isolation of a single species with a Soret band at ∼442-447 nm, indicative of a 6-coordinate ferrous thiolate species with a nitrogen-donor ligand. A ferrous species with Soret band at ∼427 nm coupled with an α-band of increased intensity was observed with the imidazole ligands. With some enzyme-ligand combinations reduction resulted in breaking of the iron‑nitrogen bond yielding a 5-coordinate high-spin ferrous species. In other instances, the ferrous form was readily oxidised back to the ferric form on addition of the ligand.

Studies on the binding of CO to low-spin [Fe(II)(Por)L2] complexes: an aid to understanding the binding of CO to haemoglobin and myoglobin

The visible and Mössbauer spectra of [Fe(II)(Por)L₂] and [Fe(II)(Por)L(CO)] complexes (where Por = protoporphyrin IX (PPIX) or tetra(p-sulfophenyl)porphyrin (TPPS) and L = an aliphatic or aromatic nitrogenous base) are reported and discussed. The results are compared to those of previously reported [Fe(II)(Por)L(CO)] complexes (where Por = PPIX, TPPS, PMXPP, TPP, OMTBP and OEP; L = a nitrogenous aromatic ligand) and HbCO (where Hb = haemoglobin) and MyCO (where My = myoglobin). A new approach, to extracting information from the Mössbauer parameters has been developed by plotting those of the [Fe(II)(Por)L₂] complexes against those of [Fe(II)(Por)L(CO)] complexes for the same ligands, has yielded a series of trend lines that show a significant dependence on both the nature of the porphyrin and also of the nitrogenous ligand. Different trend lines were found for aromatic nitrogenous ligands to aliphatic nitrogenous ligands showing that the porphyrins could donate different amounts of charge to the Fe(II) cations as the L ligand changed, and hence, they display electron sink properties. From the plots, it was shown that haemoglobin and myoglobin both bind CO very strongly compared to the model complexes studied herein. Using the reported structural and Mössbauer data for the [Fe(II)(Por)L₂] and [Fe(II)(Por)L(CO)] complexes, it proved possible and instructive to plot the Mössbauer parameters against a number of the bond lengths around the Fe(II) cations. The interpretation of the resulting trend lines both supported and facilitated the extension of our findings enabling further understanding of the geometry of the bonding in CO haemoglobin and CO myoglobin.

Studies on the binding of nitrogenous bases to protoporphyrin IX iron(II) in aqueous solution at high pH values

Studies are reported on the formation of low-spin six-coordinate [Fe(PPIX)L₂] complexes from iron(II) protoporphyrin where L is one of a series of nitrogenous ligands (aliphatic, aromatic or heterocyclic). The bonding constants have been determined by titration of the metal complex with these ligands and are compared in relation to previous studies. The adduct formation was monitored utilising optical spectroscopy. In addition, Mӧssbauer spectroscopic experiments were conducted to monitor the electronic environment around the central iron atom in these complexes. The two complementary spectroscopic methods indicated that all nitrogen ligands formed low-spin octahedral complexes. The magnitude of the overall binding constants (β₂ values) are discussed and related to (a) the pK_a values of the free ligands and (b) the Mössbauer parameter ΔE_Q, which represents the quadrupole splitting of the haem iron. The β₂ and ΔE_Q values are also discussed in terms of the structure of the ligand. Cooperative binding was observed for nearly all the ligands with Hill coefficients close to 2 for iron(II) protoporphyrin; one of these ligands displayed a much greater affinity than any we previously studied, and this was a direct consequence of the structure of the ligand. Overall conclusions on these and previous studies are drawn in terms of aliphatic ligands versus aromatic ring structures and the absence or presence of sterically hindered nitrogen atoms. The implications of the work for the greater understanding of haem proteins in general and in particular how the nitrogenous ligand binding results are relevant to and aid the understanding of the binding of inhibitor molecules to the cytochrome P₄₅₀ mono-oxygenases (for therapeutic purposes) are also discussed. Changes in the electronic absorption spectra of five-coordinate [Fe(II)(PPIX)(2-MeIm)] that occurred as the temperature was lowered from room temperature to 78° K.

The structures and stabilities of the complexes of biologically available ligands with Fe(III)-porphine: an ab initio study

Fe(III) porphine complexes and a limited number of Fe(II) porphine complexes were investigated at the 'MP2/LB'//B3LYP/SB level of theory, where SB and LB represent small and large basis sets, 6-31+G(d) and 6-311+G(2df,2p), respectively. Solvation effects were incorporated by the IEFPCM procedure. Most of the ligands whereby the heme prosthetic group is bound in biological systems were modeled in the study. These include H(2)O, Im (imidazole), CH(3)NH(2), CH(3)CO(2)(-), CH(3)S(-), CH(3)PhO(-), OH(-), and Cl(-). Fe(III) porphine, 2(+), and the pentacoordinated complexes, 2(+)(Im), 2(+)(CH(3)NH(2)), and 2(+)(H(2)O), have quartet ground states. The pentacoordinated complexes with negatively charged ligands all have high spin hextet ground states. All of the hexacoordinated complexes have low spin doublet ground states, with the exception of 2(+)(H(2)O)(2) and 2(+)(H(2)O)(Im) which have intermediate spin quartet ground states. None of the pentacoordinated complexes, 2(+)(OH(-)), 2(+)(CH(3)PhO(-)), and 2(+)(CH(3)S(-)), are predicted to form stable hexacoordinated complexes in water with any of the ligands of the present study. The most stable species in water is 2(+)(OH(-)). The hydroxide may be displaced by CH(3)PhOH and CH(3)SH at physiological pH, and by Cl(-), CH(3)CO(2)(-), and Im under acidic conditions, but not by CH(3)NH(3)(+). The relevance of the present results for the pH-dependent transitions of cytochrome c and the fragments, NAcMP8, and NAcMP11, the resting state of cytochrome P450, and the bonding interactions between heme and Aβ, is discussed.

Insights into porphyrin chemistry provided by the microperoxidases, the haempeptides derived from cytochrome c

The water-soluble haem-containing peptides obtained by proteolytic digestion of cytochrome c, the microperoxidases, have been used to explore aspects of the chemistry of iron porphyrins, and as mimics for some reactions catalysed by the haemoproteins, including the reactions catalysed by the peroxidases and the cytochromes P450. The preparation of the microperoxidases, their physical and chemical properties including their electronic structure, the kinetics and thermodynamics of their reactions with ligands, electrochemical studies and examples of their uses as haemoproteins mimics, is reviewed.

The coordination of imidazole and substituted pyridines by the hemeoctapeptide N-acetyl-ferromicroperoxidase-8 (FeIINAcMP8)

The N-terminus acetylated ferric hemeoctapeptide from cytochrome c, N-acetylmicroperoxidase-8 (Fe(III)-NAcMP8) can be reduced by dithionite in aqueous solution to produce Fe(II)-NAcMP8. The UV-Vis spectrum has a broad Soret band and relatively poorly defined Q bands which is consistent with a mixture of a five-coordinate high spin species with His as the axial ligand and a six-coordinate, predominantly high spin species with His/H(2)O as axial ligands. There are two spectroscopically observable pK(a)s at 8.7+/-0.1 and 10.9+/-0.2 which are attributed to ionization of a heme propionic acid group and coordinated H(2)O, respectively; a pK(a) > or = 14 is due to ionization of the proximal His ligand. Equilibrium constants were determined by UV-Vis spectrophotometry at 25.0+/-0.2 degrees C and 0.5 M ionic strength (NaClO(4)) for the coordination of imidazole and a number of substituted pyridines, and complement available data for the ferric hemepeptide, allowing a comparison to be made of the affinity of an iron porphyrin with Fe in the +2 and +3 oxidation states towards these ligands. Imidazole is coordinated more strongly by the ferric porphyrin (log K=4.08) than by the ferrous porphyrin (log K=3.40). The equilibrium constants for coordination of pyridines by the ferric and ferrous porphyrins increase and decrease, respectively, with increasing ligand basicity. Values determined by cyclic voltammetry show the same dependence on the identity of the ligand. In the ferric porphyrin, the stability of the complex increases with the basicity of the ligand and hence its ability to donate electron density onto the metal. In the case of the more electron rich ferrous porphyrin, greater stability occurs with pyridine ligands that have an electron withdrawing group and hence can accept electron density from the metal. This is consistent with the midpoint reduction potentials E(1/2) of the pyridine complexes determined by cyclic voltammetry; E(1/2) is linearly dependent on, and becomes more negative with an increase in, ligand basicity. Log K for coordination of pyridines by the ferrous hemepeptide correlates well with the energy of the ligand frontier orbital with pi symmetry, suggesting that pi-bonding effects are significant in determining the strength of binding of pyridines by a ferrous porphyrin.

Adenosylcobinamide Plus Exogenous, Sterically Hindered, Putative Axial Bases: A Reinvestigation into the Cause of Record Levels of Co−C Heterolysis

A reinvestigation of an earlier Ph.D. thesis (Sirovatka, J. M. Ph.D. Thesis, Colorado State University, Fort Collins, CO, 1999) is reported herein. That thesis examined the thermolysis reaction of AdoCbi(+)BF(4)(-) in ethylene glycol solution with exogenous bases, N-methylimidazole (N-Me-Im) and the sterically hindered 1,2-dimethylimidazole, (1,2-Me(2)-Im), 2-methylpyridine (2-Me-py), and 2,6-dimethylpyridine (2,6-Me(2)-py). In the present work, multiple purities of each base have been utilized as a check to see if impurities in the nitrogenous bases are causing the observed homolysis and heterolysis product distributions as others have implied (Trommel, J. S.; Warncke, K.; Marzilli, L. G. J. Am. Chem. Soc. 2001, 123, 3358). The "impurity hypothesis" is disproven by a series of results, including the following: N-Me-Im displays an invariant 52 +/- 10% heterolysis and the 1,2-Me(2)-Im system displays an invariant 83 +/- 7% heterolysis as a function of different base purification methods. Moreover, 2-Me-py and 2,6-Me(2)-py also display an invariant approximately 16 +/- 5% heterolysis as a function of different purification methods. What is responsible for the high levels of Co-C heterolysis in the AdoCbi(+) plus sterically bulky base thermolyses was uncovered via a revisitation of our four, earlier alternative hypotheses for the enhanced Co-C heterolysis (Sirovatka, J. M.; Finke, R. G. Inorg. Chem. 1999, 38, 1697). Our prior number one alternative hypothesis is shown to be correct: the added bases simply deprotonate the ethylene glycol solvent, forming ethylene glycolate anion and base-H(+)() as the key agents behind the previously ill-understood Co-C heterolyses. Also reported are Co(II)Cbi(+) titrations with five bases (1,2-Me(2)-Im, N-Me-Im, pyridine, 2-MePy, and 2,6-Me(2)-py). These experiments confirm Marzilli and co-workers' (op. cit.) results by showing that sterically hindered bases do not bind to Co(II)Cbi(+); therefore, Co(II)Cbi(+) EPR literature showing binding of bulky pyridines is erroneous as is the previously reported binding of bulky pyridine bases to Co(II)Cbi(+) by UV-vis spectroscopy (Sirovatka, J. Ph.D. Thesis, op. cit.). Also reported is our current best synthesis and purification of AdoCbi(+)BF(4)(-), work that builds off our 1987 synthesis of AdoCbi(+)BF(4)(-) (Hay, B. P.; Finke, R. G. J. Am. Chem. Soc. 1987, 109, 8012). Finally, the multiple, compounding errors which have caused problems in this project are listed, notably errors in the protein X-ray crystallography literature, the EXAFS literature, the Co(II)Cbi(+) plus bulky-bases EPR literature, the misleading B(12)-model literature, the erroneous experimental work (Sirovatka, op. cit.) and thus incorrect conclusions in one of our prior papers, as well as the erroneous implications in parts of the Marzilli and co-workers paper (op. cit.). It is hoped that a forthright reporting of these errors will help others avoid similar mistakes in the future when studying complex, bioinorganic systems.

Providing a chemical basis toward understanding the histidine base-on motif of methylcobalamin-dependent methionine synthase: an improved purification of methylcobinamide, plus thermodynamic studies of methylcobinamide binding exogenous imidazole and pyridine bases

Reported herein are the synthesis and improved purification of MeCbi(+).BF(4)(-) leading to 95% pure product. The availability of this higher purity MeCbi(+).BF(4)(-) has, in turn, allowed a study of the K(assoc), DeltaH, and DeltaS for exogenous imidazole and pyridine bases binding to MeCbi(+) in ethylene glycol and buffered aqueous solution. The results show that (1) the bases studied have larger K(assoc) values (where measurable) when binding to MeCbi(+) than when binding to AdoCbi(+) under analogous conditions; (2) comparison of the thermodynamic binding parameters for py and N-MeIm show that these bases bind similarly, within experimental error to MeCbi(+), contrary to what was seen earlier with AdoCbi(+); (3) the bases follow the expected trend, with the base with the highest pK(a) of those studied, 4-Me(2)Npy, exhibiting the highest K(assoc) value (K(assoc)(25 degrees C) = 18.0 +/- 0.3 M(-1)) and the base of lowest pK(a), py, exhibiting the lowest detectable K(assoc) value (K(assoc) (25 degrees C) = 6.2 +/- 0.4 M(-1)); (4) there is no detectable binding (K(assoc) = 0.07 M(-1)) for 2-Mepy or 2,6-Me(2)py with MeCbi(+); and (5) the base that is closest to the biologically relevant axial His759 residue in methionine synthase, N-MeIm, exhibits an unusual DeltaH value for the formation of MeCbi(+).N-MeIm, results interpreted as offering further support for the presence of sigma plus pi effects when imidazole bases bind to alkylcobinamides. The results of these studies allow the percentage of base-on methylcobinamide, MeCbi(+).base, to be calculated as a function of temperature and added base. As such, they provide necessary background information for RS(-) + MeCbi(+).base and other methionine synthase chemical precedent studies.

Coenzyme B(12) Axial-Base Chemical Precedent Studies. Adenosylcobinamide Plus Sterically Hindered Axial-Base Co-C Bond Cleavage Product and Kinetic Studies: Evidence for the Dominance of Axial-Base Transition-State Effects and for Co-N(Axial-Base) Distance-Dependent, Competing sigma and pi Effects

Adenosylcobinamide (AdoCbi(+)) plus the sterically hindered bases 1,2-dimethylimidazole, 2-methylpyridine, and 2,6-dimethylpyridine, as well as control experiments with imidazolate and 4-methylimidazolate, have been investigated to provide chemical precedent for the benzimidazole base-off, protein histidine imidazole base-on form of adenosylcobalamin (AdoCbl, also coenzyme B(12)). This imidazole base-on form of AdoCbl was observed in the recent X-ray crystallographic structural study of methylmalonyl-CoA (MMCoA) mutase; of interest to the present work is the fact that MMCoA mutase contains a long, ca. 2.5 Å, Co-N(imidazole) axial bond, at least in the enzyme's crystallographically characterized Co(II)/Co(III) state and conformation. In the present studies, upper limits for the axial-base binding K(assoc) parameters to form [AdoCbi.bulky base](+) BF(4)(-) have been obtained; these thermodynamic studies reveal that sterically hindered bases do not bind detectably to AdoCbi(+) in the ground state, which results in negligible ground-state free-energy stabilization via the formation of [AdoCbi.bulky base](+). The sterically hindered bases do, however, bind to Co(II)Cbi(+), a good energetic model of the [Ado. - - -.CoCbi](+) homolysis transition state. Kinetic studies demonstrate that the sterically hindered bases are involved in the rate-determining step of Co-C bond homolysis, accelerating it by 200-fold; hence, Co-C cleavage does occur via the low-level and otherwise nondetectable amount of [AdoCbi.bulky base](+) formed in solution. Product studies reveal (i) that both Co-C heterolysis and homolysis occur, and (ii) that there is no simple correlation between the ratio of Co-C heterolysis to homolysis and the Co-N(axial-base) bond length. Overall, the results provide strong evidence for the dominance of axial-base transition-state effects on Co-C bond cleavage, and reveal a subtle interplay of sigma and pi effects as a function of the Co-N(axial-base) bond length.