Spectroscopic study of [Fe2O2(5-Et3-TPA)2]3+: nature of the Fe2O2 diamond core and its possible relevance to high-valent binuclear non-heme enzyme intermediates

The spectroscopic properties and electronic structure of an Fe(2)(III,IV) bis-mu-oxo complex, [Fe(2)O(2)(5-Et(3)-TPA)(2)](ClO(4))(3) where 5-Et(3)-TPA = tris(5-ethyl-2-pyridylmethyl)amine, are explored to determine the molecular origins of the unique electronic and geometric features of the Fe(2)O(2) diamond core. Low-temperature magnetic circular dichroism (MCD) allows the two features in the broad absorption envelope (4000-30000 cm(-)(1)) to be resolved into 13 transitions. Their C/D ratios and transition polarizations from variable temperature-variable field MCD saturation behavior indicate that these divide into three types of electronic transitions; t(2) --> t(2) involving excitations between metal-based orbitals with pi Fe-O overlap (4000-10000 cm(-)(1)), t(2)/t(2) --> e involving excitations to metal-based orbitals with sigma Fe-O overlap (12500-17000 cm(-)(1)) and LMCT (17000-30000 cm(-)(1)) and allows transition assignments and calibration of density functional calculations. Resonance Raman profiles show the C(2)(h)() geometric distortion of the Fe(2)O(2) core results in different stretching force constants for adjacent Fe-O bonds (k(str)(Fe-O(long)) = 1.66 and k(str)(Fe-O(short)) = 2.72 mdyn/A) and a small ( approximately 20%) difference in bond strength between adjacent Fe-O bonds. The three singly occupied pi-metal-based orbitals form strong superexchange pathways which lead to the valence delocalization and the S = (3)/(2) ground state. These orbitals are key to the observed reactivity of this complex as they overlap with the substrate C-H bonding orbital in the best trajectory for hydrogen atom abstraction. The electronic structure implications of these results for the high-valent enzyme intermediates X and Q are discussed.     

Direct detection of oxygen intermediates in the non-heme Fe enzyme taurine/alpha-ketoglutarate dioxygenase

The reaction of substrate-bound taurine/alpha-ketoglutarate dioxygenase with O2 has been studied using cryogenic continuous-flow spectroscopy. Transient absorption spectra acquired at -38 degrees C show an exponential decay of a 318-nm chromophore with an apparent rate of 1.3 s-1. The observed optical changes and their kinetics are consistent with the profile of an Fe(IV) species detected recently by M?ssbauer spectroscopy (Price et al., Biochemistry 2003, 42, 7497-7508). Resonance Raman measurement upon excitation at 363.7 nm reveal at least two oxygen isotope-sensitive vibrations at 821/787 cm-1 and 583/555 cm-1 for 16O and 18O derivatives, respectively. An additional mode is likely to be obscured by an ethylene glycol vibration at 865 cm-1 and/or 1089 cm-1. The 821 cm-1 vibration is assigned to the stretching mode of Fe(IV)=O species on the basis of its frequency and isotopic shift amplitude. The 583 cm-1 band is likely to originate from an Fe-O2 precursor of the Fe(IV)=O species, although its structural details are unclear at present.     

Nature of the peroxo intermediate of the W48F/D84E ribonucleotide reductase variant: implications for O2 activation by binuclear non-heme iron enzymes

Analysis of the spectroscopic signatures of the R2-W48F/D84E biferric peroxo intermediate identifies a cis mu-1,2 peroxo coordination geometry. DFT geometry optimizations on both R2-W48F/D84E and R2-wild-type peroxo intermediate models including constraints imposed by the protein also identify the cis mu-1,2 peroxo geometry as the most stable peroxo intermediate structure. This study provides significant insight into the electronic structure and reactivity of the R2-W48F/D84E peroxo intermediate, structurally related cis mu-1,2 peroxo model complexes, and other enzymatic biferric peroxo intermediates.     

End-on and side-on peroxo derivatives of non-heme iron complexes with pentadentate ligands: models for putative intermediates in biological iron/dioxygen chemistry

Mononuclear iron(III) species with end-on and side-on peroxide have been proposed or identified in the catalytic cycles of the antitumor drug bleomycin and a variety of enzymes, such as cytochrome P450 and Rieske dioxygenases. Only recently have biomimetic analogues of such reactive species been generated and characterized at low temperatures. We report the synthesis and characterization of a series of iron(II) complexes with pentadentate N5 ligands that react with H(2)O(2) to generate transient low-spin Fe(III)-OOH intermediates. These intermediates have low-spin iron(III) centers exhibiting hydroperoxo-to-iron(III) charge-transfer bands in the 500-600-nm region. Their resonance Raman frequencies, nu(O)(-)(O), near 800 cm(-)(1) are significantly lower than those observed for high-spin counterparts. The hydroperoxo-to-iron(III) charge-transfer transition blue-shifts and the nu(O)(-)(O) of the Fe-OOH unit decreases as the N5 ligand becomes more electron donating. Thus, increasing electron density at the low-spin Fe(III) center weakens the O-O bond, in accord with conclusions drawn from published DFT calculations. The parent [(N4Py)Fe(III)(eta(1)-OOH)](2+) (1a) ion in this series (N4Py = N,N-bis(2-pyridylmethyl)-N-bis(2-pyridyl)methylamine) can be converted to its conjugate base, which is demonstrated to be a high-spin iron(III) complex with a side-on peroxo ligand, [(N4Py)Fe(III)(eta(2)-O(2))](+) (1b). A detailed analysis of 1a and 1b by EPR and M?ssbauer spectroscopy provides insights into their electronic properties. The orientation of the observed (57)Fe A-tensor of 1a can be explained with the frequently employed Griffith model provided the rhombic component of the ligand field, determined by the disposition of the hydroperoxo ligand, is 45 degrees rotated relative to the octahedral field. EXAFS studies of 1a and 1b reveal the first metrical details of the iron-peroxo units in this family of complexes: [(N4Py)Fe(III)(eta(1)-OOH)](2+) has an Fe-O bond of 1.76 A, while [(N4Py)Fe(III)(eta(2)-O(2))](+) has two Fe-O bonds of 1.93 A, values which are in very good agreement with results obtained from DFT calculations.     

High conversion of olefins to cis-diols by non-heme iron catalysts and H2O2

Efficient and highly stereoselective oxidation of olefins to cis-diols as the major product is obtained by using biomimetic non-heme FeII catalysts in combination with H2O2.     

On the nature of oxoiron (IV) intermediate in dioxygen activation by non-heme enzymes

We discuss electronic properties of the molecular systems with the short distance Fe–O unit, which are presumably formed as reaction intermediates during oxygen activation by non-heme enzymes. By performing an analysis of electronic densities in terms of multiconfigurational expansions of wavefunctions with localized orbitals the electronic properties of the Fe–O moiety in two model complexes are compared. The first one refers to the enzymatic intermediate, and the second biomimetic complex models a synthetic compound [Fe(O)(TMC)(NCCH₃)](OTf)₂ with a terminal Fe–O unit, which is experimentally characterized as the Fe(IV)=O species. We show that the orbital pictures of the FeO unit in both model complexes share common features. According to these simulations, the non-heme enzymatic intermediates may be assigned to the systems with the oxidation state of Fe between III and IV, as recently proposed for the TauD enzyme in experimental spectroscopic studies.     

Stereospecific alkane hydroxylation by non-heme iron catalysts: mechanistic evidence for an Fe(V)=O active species.

High-valent iron-oxo species have frequently been invoked in the oxidation of hydrocarbons by both heme and non-heme enzymes. Although a formally Fe(V)=O species, that is, [(Por(*))Fe(IV)=O](+), has been widely accepted as the key oxidant in stereospecific alkane hydroxylation by heme systems, it is not established that such a high-valent state can be accessed by a non-heme ligand environment. Herein we report a systematic study on alkane oxidations with H(2)O(2) catalyzed by a group of non-heme iron complexes, that is, [Fe(II)(TPA)(CH(3)CN)(2)](2+) (1, TPA = tris(2-pyridylmethyl)amine) and its alpha- and beta-substituted analogues. The reactivity patterns of this family of Fe(II)(TPA) catalysts can be modulated by the electronic and steric properties of the ligand environment, which affects the spin states of a common Fe(III)-OOH intermediate. Such an Fe(III)-peroxo species is high-spin when the TPA ligand has two or three alpha-substituents and is proposed to be directly responsible for the selective C-H bond cleavage of the alkane substrate. The thus-generated alkyl radicals, however, have relatively long lifetimes and are susceptible to radical epimerization and trapping by O(2). On the other hand, 1 and the beta-substituted Fe(II)(TPA) complexes catalyze stereospecific alkane hydroxylation by a mechanism involving both a low-spin Fe(III)-OOH intermediate and an Fe(V)=O species derived from O-O bond heterolysis. We propose that the heterolysis pathway is promoted by two factors: (a) the low-spin iron(III) center which weakens the O-O bond and (b) the binding of an adjacent water ligand that can hydrogen bond to the terminal oxygen of the hydroperoxo group and facilitate the departure of the hydroxide. Evidence for the Fe(V)=O species comes from isotope-labeling studies showing incorporation of (18)O from H(2)(18)O into the alcohol products. (18)O-incorporation occurs by H(2)(18)O binding to the low-spin Fe(III)-OOH intermediate, its conversion to a cis-H(18)O-Fe(V)=O species, and then oxo-hydroxo tautomerization. The relative contributions of the two pathways of this dual-oxidant mechanism are affected by both the electron donating ability of the TPA ligand and the strength of the C-H bond to be broken. These studies thus serve as a synthetic precedent for an Fe(V)=O species in the oxygen activation mechanisms postulated for non-heme iron enzymes such as methane monooxygenase and Rieske dioxygenases.     

Quantum-instanton evaluation of the kinetic isotope effects

A general quantum-mechanical method for computing kinetic isotope effects is presented. The method is based on the quantum-instanton approximation for the rate constant and on the path-integral Metropolis-Monte Carlo evaluation of the Boltzmann operator matrix elements. It computes the kinetic isotope effect directly, using a thermodynamic integration with respect to the mass of the isotope, thus avoiding the more computationally expensive process of computing the individual rate constants. The method should be more accurate than variational transition-state theories or the semiclassical instanton method since it does not assume a single tunneling path and does not use a semiclassical approximation of the Boltzmann operator. While the general Monte Carlo implementation makes the method accessible to systems with a large number of atoms, we present numerical results for the Eckart barrier and for the collinear and full three-dimensional isotope variants of the hydrogen exchange reaction H + H2 --> H2 + H. In all seven test cases, for temperatures between 250 and 600 K, the error of the quantum instanton approximation for the kinetic isotope effects is less than approximately 10%.     

Enhanced selectivity in non-heme iron catalysed oxidation of alkanes with peracids: evidence for involvement of Fe(IV)=O species

Catalytic alkane oxidation with high selectivity using peracids and an (N4Py)Fe complex is presented and the role of [(N4Py)Fe(IV)=O]2+ species, molecular oxygen and hydroxyl radicals in the catalysis is discussed.     

Deuterium kinetic isotope effects in microsolvated gas-phase E2 reactions

This work describes the first experimental studies of deuterium kinetic isotope effects (KIEs) for the gas-phase E2 reactions of microsolvated systems. The reactions of F(-)(H(2)O)(n) and OH(-)(H(2)O)(n), where n = 0, 1, with (CH(3))(3)CX (X = Cl, Br), as well as the deuterated analogs of the ionic and neutral reactants, were studied utilizing the flowing afterglow-selected ion flow tube technique. The E2 reactivity is found to decrease with solvation. Small, normal kinetic isotope effects are observed for the deuteration of the alkyl halide, while moderately inverse kinetic isotope effects are observed for the deuteration of the solvent. Minimal clustering of the product ions is observed, but there are intriguing differences in the nature and extent of the clustering process. Electronic structure calculations of the transition states provide qualitative insight into these microsolvated E2 reactions.     

Studies of iron(II) and iron(III) complexes with fac-N2O, cis-N2O2 and N2O3 donor ligands: models for the 2-His 1-carboxylate motif of non-heme iron monooxygenases

Enzymes in the oxygen-activating class of mononuclear non-heme iron oxygenases (MNOs) contain a highly conserved iron center facially ligated by two histidine nitrogen atoms and one carboxylate oxygen atom that leave one face of the metal center (three binding sites) open for coordination to cofactor, substrate, and/or dioxygen. A comparative family of [Fe(II/III)(N(2)O(n))(L)(4-n))](±x), n = 1-3, L = solvent or Cl(-), model complexes, based on a ligand series that supports a facially ligated N,N,O core that is then modified to contain either one or two additional carboxylate chelate arms, has been structurally and spectroscopically characterized. EPR studies demonstrate that the high-spin d(5) Fe(III)g = 4.3 signal becomes more symmetrical as the number of carboxylate ligands decreases across the series Fe(N(2)O(3)), Fe(N(2)O(2)), and Fe(N(2)O(1)), reflecting an increase in the E/D strain of these complexes as the number of exchangeable/solvent coordination sites increases, paralleling the enhanced distribution of electronic structures that contribute to the spectral line shape. The observed systematic variations in the Fe(II)-Fe(III) oxidation-reduction potentials illustrate the fundamental influence of differential carboxylate ligation. The trend towards lower reduction potential for the iron center across the [Fe(III)(N(2)O(1))Cl(3)](-), [Fe(III)(N(2)O(2))Cl(2)](-) and [Fe(III)(N(2)O(3))Cl](-) series is consistent with replacement of the chloride anions with the more strongly donating anionic O-donor carboxylate ligands that are expected to stabilize the oxidized ferric state. This electrochemical trend parallels the observed dioxygen sensitivity of the three ferrous complexes (Fe(II)(N(2)O(1)) < Fe(II)(N(2)O(2)) < Fe(II)(N(2)O(3))), which form μ-oxo bridged ferric species upon exposure to air or oxygen atom donor (OAD) molecules. The observed oxygen sensitivity is particularly interesting and discussed in the context of α-ketoglutarate-dependent MNO enzyme mechanisms.     

Metal-catalyzed phosphodiester cleavage: secondary 18O isotope effects as an indicator of mechanism

Information about the transition states of metal-catalyzed hydrolysis reactions of model phosphate compounds has been obtained through determination of isotope effects (IEs) on the hydrolysis reactions. Metal complexation has been found to significantly alter the transition state of the reaction from the alkaline hydrolysis reaction, and the transition state is quite dependent on the particular metal ion used. For the diester, ethyl p-nitrophenyl phosphate, the nonbridge 18O effect for the hydrolysis reactions catalyzed by Co(III) 1,5,9-triazacyclononane and Eu(III) were 1.0006 and 1.0016, respectively, indicative of a slightly associative transition state and little net change in bonding to the nonbridge oxygen. The reaction catalyzed by Zn(II) 1,4,7,10-tetraazacyclododecane had an 18O nonbridge IE of 1.0108, showing the reaction differs significantly from the reaction of the noncomplexed diester and resembles the reactions of triesters. Reaction with Co(III) 1,4,7,10-tetraazacyclododecane showed an inverse effect of 0.9948 reflecting the effects of bonding of the diester to the Co(III). Lanthanide-catalyzed hydrolysis has been observed to have unusually large 15N effects. To further investigate this effect, the 15N effect on the reaction catalyzed by Ce(IV) bis-Tris propane solutions at pH 8 was determined to be 1.0012. The 15N effects were also measured for the reaction of the monoester p-nitrophenyl phosphate by Ce(IV) bis-Tris propane (1.0014) and Eu(III) bis-Tris propane (1.0012). These smaller effects at pH 8 indicate that a smaller negative charge develops on the nitrogen during the hydrolysis reaction.     

Primary and secondary kinetic isotope effects in E2 elimination reactions

Primary and secondary deuterium kinetic isotope effects have been measured for elimination of LCl (L = H or D) from some substituted 1,2-diaryl-1-chloroethanes. Although changes in these effects are in agreement with theoretical predictions, the high values of (E_H-E_D)_β and the significantly low A_H/A_D may suggest that either proton tunnelling or an internal return mechanism is complicating this E2 elimination.     

Distinguishing rate-limiting electron versus H-atom transfers in Cu2O2-mediated oxidative N-dealkylations: application of inter- versus intramolecular kinetic isotope effects

Copper-dioxygen adducts are important biological oxidants. To gain a better understanding of the underlying chemistries of such species, we report on a series of Cu2II-O2 complexes, [{CuII(MePY2)R'}2(O2)](B(C6F5)4)2 (1R') (where (MePY2)R' is a 4-pyridyl substituted bis[2-(2-(4-R'-pyridyl)ethyl]methylamine; R' = H, MeO, Me2N; Zhang, C. X.; et al. J. Am. Chem. Soc. 2003, 125, 634-635), which readily oxidize exogenous substrates. In this study, we explore the mechanism by which 1R' facilitates the oxidative N-dealkylation of para-substituted N,N-dimethylanilines (R-DMA; R = MeO, Me, H, CN). In the case of 1H, the linear free-energy correlation plot (rho = -2.1) and intramolecular deuterium kinetic isotope effect (KIEintra, using p-R-(C6H4)-N(CH3)(CD3)) profile suggest that R-DMA oxidation occurs through rate-limiting electron transfer (ET). This mechanism was further enforced by comparison of KIEintra versus the intermolecular KIE (KIEinter, using p-R-(C6H4)-N(CH3)2 versus p-R-(C6H4)-N(CD3)2). It was found that KIEinter < KIEintra, suggesting an ET process. In the case of both 1MeO and 1Me2N, the KIEintra profile and linear free-energy correlation plots (rho = -0.49 and -0.99 for 1Me2N and 1MeO with especially poor fitting for the latter) are inconclusive in distinguishing between a rate-limiting ET or hydrogen atom transfer (HAT) pathway. Comparisons of KIEinter versus KIEintra demonstrate a switch in mechanism from ET to HAT for 1Me2N and 1MeO oxidation of R-DMA as R-DMA is made less reducing. In the case of 1Me2N, MeO-DMA and Me-DMA are oxidized via a rate-limiting ET (KIEinter < KIEintra), while H-DMA and CN-DMA are oxidized through a HAT pathway (KIEinter approximately KIEintra). For 1MeO, oxidation occurs through an ET pathway for MeO-, Me-, and H-DMA (KIEinter < KIEintra), while CN-DMA is oxidized though a HAT process (KIEinter approximately KIEintra). Copper complex attributes, which may contribute to the mechanistic observations, are suggested.