Biomedical subjects
Publications and source records attributed to Edward I Solomon.
Elemental sulfur (S8) was found to react with [(TMPA)CuI(CH3CN)]+ to form the trans-mu-1,2 end-on disulfide complex [(TMPA)Cu-S-S-Cu(TMPA)]2+. The X-ray structure of this centrosymmetric disulfide complex shows a Cu(1)-S(1) bond length of 2.280(2) A and a S(1)-S(1A) bond length of 2.044(4) A.
The spectroscopic properties and electronic structure of the four-coordinate high-spin [FeIII(L3)(OOtBu)]+ complex (1; L3 = hydrotris(3-tert-butyl-5-isopropyl-1-pyrazolyl)borate; tBu = tert-butyl) are investigated and compared to the six-coordinated high-spin [Fe(6-Me3TPA)(OHx)(OOtBu)]x+ system (TPA = tris(2-pyridylmethyl)amine, x = 1 or 2) studied earlier [Lehnert, N.; Ho, R. Y. N.; Que, L., Jr.; Solomon, E. I. J. Am. Chem. Soc. 2001, 123, 12802-12816]. Complex 1 is characterized by Raman features at 889 and 830 cm-1 which are assigned to the O-O stretch (mixed with the symmetric C-C stretch) and a band at 625 cm-1 that corresponds to nu(Fe-O). The UV-vis spectrum shows a charge-transfer (CT) transition at 510 nm from the alkylperoxo pi v* (v = vertical to C-O-O plane) to a d orbital of Fe(III). A second CT is identified from MCD at 370 nm that is assigned to a transition from pi h* (h = horizontal to C-O-O plane) to an Fe(III) d orbital. For the TPA complex the pi v* CT is at 560 nm while the pi h* CT is to higher energy than 250 nm. These spectroscopic differences between four- and six-coordinate Fe(III)-OOR complexes are interpreted on the basis of their different ligand fields. In addition, the electronic structure of Fe-OOPtn complexes with the biologically relevant pterinperoxo ligand are investigated. Substitution of the tert-butyl group in 1 by pterin leads to the corresponding Fe(III)-OOPtn species (2), which shows a stronger electron donation from the peroxide to Fe(III) than 1. This is related to the lower ionization potential of pterin. Reduction of 2 by one electron leads to the Fe(II)-OOPtn complex (3), which is relevant as a model for potential intermediates in pterin-dependent hydroxylases. However, in the four-coordinate ligand field of 3, the additional electron is located in a nonbonding d orbital of iron. Hence, the pterinperoxo ligand is not activated for heterolytic cleavage of the O-O bond in this system. This is also evident from the calculated reaction energies that are endothermic by at least 20 kcal/mol.
The geometric and electronic structure of NO bound to reduced protocatechuate 3,4-dioxygenase and its substrate (3,4-dihydroxybenzoate, PCA) complex have been examined by X-ray absorption (XAS), UV-vis absorption (Abs), magnetic circular dichroism (MCD), and variable temperature variable field (VTVH) MCD spectroscopies. The results are compared to those previously published on model complexes described as [FeNO]7 systems in which an S = 5/2 ferric center is antiferromagnetically coupled to an S = 1 NO-. XAS pre-edge analysis indicates that the Fe-NO units in FeIIIPCD[NO-] and FeIIIPCD[PCA,NO-] lack the greatly increased pre-edge intensity representative of most [FeNO]7 model sites. Furthermore, from extended X-ray absorption fine structure (EXAFS) analysis, the FeIIIPCD[NO-] and FeIIIPCD[PCA,NO-] active sites are shown to have an Fe-NO distance of at least 1.91 A, approximately 0.2 A greater than those found in the model complexes. The weakened Fe-NO bond is consistent with the overall lengthening of the bond lengths and the fact that VTVH MCD data show that NO(-)-->FeIII CT transitions are no longer polarized along the z-axis of the zero-field splitting tensor. The weaker Fe-NO bond derives from the strong donor interaction of the endogenous phenolate and substrate catecholate ligands, which is observed from the increased intensity in the CT region relative to that of [FeNO]7 model complexes, and from the shift in XAS edge position to lower energy. As NO is an analogue of O2, the effect of endogenous ligand donor strength on the Fe-NO bond has important implications with respect to O2 activation by non-heme iron enzymes.
Magnetic, vibrational, and optical techniques are combined with density functional calculations to elucidate the electronic structure of the diamagnetic mononuclear side-on CuII-superoxo complex. The electronic nature of its lowest singlet/triplet states and the ground-state diamagnetism are explored. The triplet state is found to involve the interaction between the Cu xy and the superoxide pi v * orbitals, which are orthogonal to each other. The singlet ground state involves the interaction between the Cu xy and the in-plane superoxide pi v * orbitals, which have a large overlap and thus strong bonding. The ground-state singlet/triplet states are therefore fundamentally different in orbital origin and not appropriately described by an exchange model. The ground-state singlet is highly delocalized with no spin polarization.
Low-temperature EPR spectroscopy with frequencies between 95 and 345 GHz and magnetic fields up to 12 T has been used to study metal sites in proteins or inorganic complexes and free radicals. The high-field EPR method was used to resolve g-value anisotropy by separating it from overlapping hyperfine couplings. The presence of hydrogen bonding interactions to the tyrosyl radical oxygens in ribonucleotide reductases were detected. At 285 GHz the g-value anisotropy from the rhombic type 2 Cu(II) signal in the enzyme laccase has its g-value anisotropy clearly resolved from slightly different overlapping axial species. Simple metal site systems with S>1/2 undergo a zero-field splitting, which can be described by the spin Hamiltonian. From high-frequency EPR, the D values that are small compared to the frequency (high-field limit) can be determined directly by measuring the distance of the outermost signal to the center of the spectrum, which corresponds to (2 S-1)* mid R: Dmid R: For example, D values of 0.8 and 0.3 cm(-1) are observed for S=5/2 Fe(III)-EDTA and transferrin, respectively. When D values are larger compared to the frequency and in the case of half-integer spin systems, they can be obtained from the frequency dependence of the shifts of g(eff), as observed for myoglobin in the presence ( D=5 cm(-1)) or absence ( D=9.5 cm(-1)) of fluoride. The 285 and 345 GHz spectra of the Fe(II)-NO-EDTA complex show that it is best described as a S=3/2 system with D=11.5 cm(-1), E=0.1 cm(-1), and g(x)= g(y)= g(z)=2.0. Finally, the effects of HF-EPR on X-band EPR silent states and weak magnetic interactions are demonstrated.
Cobalt(II) porphyrins were studied to determine the influence of distal site metalation and superstructure upon dioxygen reactivity in active site models of cytochrome c oxidase (CcO). Monometallic, Co(II)(P) complexes when ligated by an axial imidazole react with dioxygen to form reversible Co-superoxide adducts, which were characterized by EPR and resonance Raman (RR). Unexpectedly, certain Co porphyrins with Cu(I) metalated imidazole pickets do not form mu-peroxo Co(III)/Cu(II) products even though the calculated intermetallic distance suggests this is possible. Instead, cobalt-porphyrin-superoxide complexes are obtained with the distal copper remaining as Cu(I). Moreover, distal metals (Cu(I) or Zn(II)) greatly enhance the stability of the dioxygen adduct, such that Co superoxides of bimetallic complexes demonstrate minimal reversibility. The "trapping" of dioxygen by a second metal is attributed to structural and electrostatic changes within the distal pocket upon metalation. EPR evidence suggests that the terminal oxygen in these bimetallic Co-superoxide systems is H-bonded to the NH of an imidazole picket amide linker, which may contribute to enthalpic stabilization of the dioxygen adduct. Stabilization of the dioxygen adduct in these bimetallic systems suggests one possible role for the distal copper in the Fe/Cu bimetallic active site of terminal oxidases, which form a heme-superoxide/copper(I) adduct upon oxygenation.
Using experimentally calibrated density functional calculations on models of the active site of soybean lipoxygenase 1 (SLO-1), insight has been obtained into the coordination flexibility of the iron active site and its molecular mechanism of catalysis. The ferrous form of SLO-1 shows a variation in coordination number in solution that is related to a weakly coordinating Asn694 ligand. From the calculations it is determined that the weak Fe-O(694) bond associated with this coordination flexibility is due to a sideways tilted geometry of Asn694 that is imposed on the site by the protein. Release of this constraint (by altering the hydrogen bonding network) leads to a pure six-coordinate site. In contrast, the ferric form of the enzyme stays five-coordinate. In this case, deprotonation of a coordinated water gives a strong hydroxo donor in the cis position to Asn694, weakening the Fe-O(694) bond. Hence, Asn694 is a stronger ligand to the reduced relative to the oxidized site. Using these experimentally calibrated models, the reaction energy for H-atom transfer in SLO-1 has been calculated to be about -18 kcal/mol. The observed change in coordination number going from five-coordinate in ferric to six-coordinate in ferrous SLO-1 increases the reduction potential of the iron active site. Hence, the protein adjusts the active site for optimal reactivity. Analysis of the electronic structure along the reaction coordinate shows that the H-atom transfer in SLO-1 actually corresponds to a proton-coupled electron transfer (PCET). The transferred electron does not localize on the proton, but tunnels directly from the substrate to the ferric active site in a concerted proton tunneling-electron tunneling (PTET) process. The covalently linked Fe-O-H-C bridge in the transition state lowers the energy barrier and provides an efficient superexchange pathway for this tunneling. The thermal barrier for the PTET process is estimated from the calculations to be about +15 kcal/mol including zero-point energy corrections. This corresponds to a thermal reaction rate of k(therm) approximately 1 s(-1). In comparison, the rate of proton tunneling can be as high as 2 x 10(9) s(-1) under these conditions.
The spectroscopic properties, electronic structure, and reactivity of the low-spin Fe(III)-hydroperoxo complex [Fe(N4Py)(OOH)](2+) (1, N4Py = N,N-bis(2-pyridylmethyl)-N-bis(2-pyridyl)methylamine) are investigated in comparison to those of activated bleomycin (ABLM). Complex 1 is characterized by Raman features at 632 (Fe-O stretch) and 790 cm(-1) (O-O stretch), corresponding to a strong Fe-O bond (force constant 3.62 mdyn/A) and a weak O-O bond (3.05 mdyn/A). The UV-vis spectrum of 1 shows a broad absorption band around 550 nm that is assigned to a charge-transfer transition from the hydroperoxo to a t(2g) d orbital of Fe(III) using resonance Raman and MCD spectroscopies and density functional (DFT) calculations. Compared to low-spin [Fe(TPA)(OH(x))(OO(t)Bu)](x+)(TPA = tris(2-pyridylmethyl)amine, x = 1 or 2), an overall similar Fe-OOR bonding results for low-spin Fe(III)-alkylperoxo and -hydroperoxo species. Correspondingly, both systems show similar reactivities and undergo homolytic cleavage of the O-O bond. From the DFT calculations, this reaction is more endothermic for 1 due to the reduced stabilization of the .OH radical compared to .O(t)Bu and the absence of the hydroxo ligand that helps to stabilize the resulting Fe(IV)=O species. In contrast, ABLM has a somewhat different electronic structure where no pi donor bond between the hydroperoxo ligand and iron(III) is present [Neese, F.; Zaleski, J. M.; Loeb-Zaleski, K.; Solomon, E. I. J. Am. Chem. Soc. 2000, 122, 11703]. Possible reaction pathways for ABLM are discussed in relation to known experimental results.
Spectroscopic methods combined with density functional calculations are used to develop a detailed bonding description of the mu(4)-sulfide bridged tetranuclear Cu(Z) cluster in N(2)O reductase. The ground state of Cu(Z) has the 1Cu(II)/3Cu(I) configuration. The single electron hole dominantly resides on one Cu atom (Cu(I)) and partially delocalizes onto a second Cu atom (Cu(II)) via a Cu(I)-S-Cu(II) sigma/sigma superexchange pathway which is manifested by a Cu(II) --> Cu(I) intervalence transfer transition in absorption. The observed excited-state spectral features of Cu(Z) are dominated by the S --> Cu(I) charge-transfer transitions and Cu(I) based d-d transitions. The intensity pattern of individual S --> Cu(I) charge-transfer transitions reflects different bonding interactions of the sulfur valence orbitals with the four Cu's in the Cu(Z) cluster, which are consistent with the individual Cu-S force constants obtained from a normal coordinate analysis of the Cu(Z) resonance Raman frequencies and profiles. The Cu(I) d orbital splitting pattern correlates with its distorted T-shaped ligand field geometry and accounts for the observed low g( parallel ) value of Cu(Z) in EPR. The dominantly localized electronic structure description of the Cu(Z) site results from interactions of Cu(II) with the two additional Cu's of the cluster (Cu(III)/Cu(IV)), where the Cu-Cu electrostatic interactions lead to hole localization with no metal-metal bonding. The substrate binding edge of Cu(Z) has a dominantly oxidized Cu(I) and a dominantly reduced Cu(IV). The electronic structure description of Cu(Z) provides a strategy to overcome the reaction barrier of N(2)O reduction at this Cu(I)/Cu(IV) edge by simultaneous two-electron transfer to N(2)O in a bridged binding mode. One electron can be donated directly from Cu(IV) and the other from Cu(II) through the Cu(II)-S-Cu(I) sigma/sigma superexchange pathway. A frontier orbital scheme provides molecular insight into the catalytic mechanism of N(2)O reduction by the Cu(Z) cluster.
The activation of dioxygen (O(2)) by Cu(I) complexes is an ubiquitous process in biology and industrial applications. In tyrosinase, a binuclear copper enzyme, a mu-eta(2):eta(2)-peroxodicopper(II) species is generally accepted to be the active oxidant. Reported here is the characterization and reactivity of a stable mu-eta(2):eta(2)-peroxodicopper(II) complex at -80 degrees C using a secondary diamine ligand, N,N'-di-tert-butyl-ethylenediamine (DBED). The spectroscopic characteristics of this complex (UV-vis, resonance Raman) prove to be strongly dependent on the counteranion employed and not on the solvent, suggesting an intimate interaction of the counteranions with the Cu-O(2) cores. This interaction is also supported by solution EXAFS data. This new complex exhibits hydroxylation reactivity by converting phenolates to catechols, proving to be a functional model of tyrosinase. Additional interest in this Cu/O(2) species results from the use of Cu(I)-DBED as a polymerization catalyst of phenols to polyphenylene oxide (PPO) with O(2) as the terminal oxidant.
The multicopper oxidases contain at least four copper atoms and catalyze the four-electron reduction of O(2) to H(2)O at a trinuclear copper cluster. An intermediate, termed native intermediate, has been trapped by a rapid freeze-quench technique from Rhus vernicifera laccase when the fully reduced form reacts with dioxygen. This intermediate had been described as an oxygen-radical bound to the trinuclear copper cluster with one Cu site reduced. XAS, however, shows that all copper atoms are oxidized in this intermediate. A combination of EXAFS, multifrequency EPR, and VTVH MCD has been used to understand how this fully oxidized trinuclear Cu cluster relates to the fully oxidized resting form of the enzyme. It is determined that in the native intermediate all copper atoms of the cluster are bridged by the product of full O(2) reduction. In contrast, the resting form has one copper atom of the cluster (the T2 Cu) magnetically isolated from the others. The native intermediate decays to the resting oxidized form with a rate that is too slow to be in the catalytic cycle. Thus, the native intermediate appears to be the catalytically relevant fully oxidized form of the enzyme, and its role in catalysis is considered.
Fet3p is a multicopper oxidase that uses four copper ions (one type 1, one type 2, and one type 3 binuclear site) to couple substrate oxidation to the reduction of O(2) to H(2)O. The type 1 Cu site shuttles electrons between the substrate and the type 2/type 3 Cu sites which form a trinuclear Cu cluster that is the active site for O(2) reduction. This study extends the spectroscopic and reactivity studies that have been conducted with type 1-substituted Hg (T1Hg) laccase to Fet3p and a mutant of Fet3p in which the trinuclear Cu cluster is perturbed. To examine the reaction between the trinuclear Cu cluster and O(2), the type 1 Cu Cys(484) was mutated to Ser, resulting in a type 1-depleted (T1D) form of the enzyme. Additional His to Gln mutations were made at the trinuclear cluster to further probe specific contributions to reactivity. One of these mutants (His(126)Gln) produces the first stable but perturbed trinuclear Cu cluster (T1DT3' Fet3p). Spectroscopic characterization (absorption, circular dichroism, magnetic circular dichroism, and electron paramagnetic resonance) of the resting trinuclear sites in T1D and T1DT3' Fet3p reveal that the His(126)Gln mutation changes the electronic structure of both the type 3 and type 2 Cu sites. The trinuclear clusters in T1D and T1DT3' Fet3p react with O(2) to produce peroxide intermediates analogous to that observed in T1Hg laccase. Spectroscopic data on the peroxide intermediates in the three forms provide further insight into the structure of this intermediate. In T1D Fet3p, the decay of this peroxide intermediate is pH-dependent, and the rate of decay is 10-fold higher at low pH. In T1DT3' Fet3p, the decay of the peroxide intermediate is pH-independent and is slow at all pH's. This change in the pH dependence provides new insight into the mechanism of intermediate decay involving reductive cleavage of the O-O bond.
Previous studies of ferrous wild-type phenylalanine hydroxylase, [Fe(2+)]PAH(T)[], have shown the active site to be a six-coordinate distorted octahedral site. After the substrate and cofactor bind to the enzyme ([Fe(2+)]PAH(R)[L-Phe,5-deaza-6-MPH(4)]), the active site converts to a five-coordinate square pyramidal structure in which the identity of the missing ligand had not been previously determined. X-ray absorption spectroscopy (XAS) at the Fe K-edge further supports this coordination number change with the binding of both cosubstrates to the enzyme, and determines this to be due to the loss of a water ligand.
1-Aminocyclopropane-1-carboxylic acid (ACC) oxidase (ACCO) catalyzes the last step in the biosynthesis of the gaseous plant hormone ethylene, which is involved in development, including germination, fruit ripening, and senescence. ACCO is a mononuclear non-heme ferrous enzyme that couples the oxidation of the cosubstrate ascorbate to the oxidation of substrate ACC by dioxygen. In addition to substrate and cosubstrate, ACCO requires the activator CO(2) for continuous turnover. NIR circular dichroism and magnetic circular dichroism spectroscopies have been used to probe the geometric and electronic structure of the ferrous active site in ACCO to obtain molecular-level insight into its catalytic mechanism. Resting ACCO/Fe(II) is coordinatively saturated (six-coordinate). In the presence of CO(2), one ferrous ligand is displaced to yield a five-coordinate site only when both the substrate ACC and cosubstrate ascorbate are bound to the enzyme. The open coordination position allows rapid O(2) activation for the oxidation of both substrates. In the absence of CO(2), ACC binding alone converts the site to five-coordinate, which would react with O(2) in the absence of ascorbate and quickly deactivate the enzyme. These studies show that ACCO employs a general strategy similar to other non-heme iron enzymes in terms of opening iron coordination sites at the appropriate time in the reaction cycle and define the role of CO(2) as stabilizing the six-coordinate ACCO/Fe(II)/ACC complex, thus preventing the uncoupled reaction that inactivates the enzyme.
Copper(I)-dioxygen interactions are of great interest due to their role in biological O2-processing as well as their importance in industrial oxidation processes. We describe here the study of systems which lead to new insights concerning the factors which govern Cu(II)-mu-eta2:eta2 (side-on) peroxo versus Cu(III)-bis-mu-oxo species formation. Drastic differences in O2-reactivity of Cu(I) complexes which differ only by a single -CH3 versus -H substituent on the central amine of the tridentate ligands employed are observed. [Cu(MeAN)]B(C6F5)4 (1) (MeAN = N,N,N',N',N'-pentamethyl-dipropylenetriamine) reacts with O2 at -80 degrees C to form almost exclusively the side-on peroxo complex [{CuII(MeAN)}2(O2)]2+ (3) in CH2Cl2, tetrahydrofuran, acetone, and diethyl ether solvents, as characterized by UV-vis and resonance Raman spectroscopies. In sharp contrast, [Cu(AN)]B(C6F5)4 (2) (AN = 3, 3'-iminobis(N,N-dimethyl-propylamine) can support either Cu2O2 structures in a strongly solvent-dependent manner. Extreme behavior is observed in CH2Cl2 solvent, where 1 reacts with O2 giving 3, while 2 forms exclusively the bis-mu-oxo species [{CuIII(AN)}2(O)2]2+ (4Oxo). Stopped-flow kinetics measurements also reveal significant variations in the oxygenation reactions of 1 versus 2, including the observations that 4Oxo forms much faster than does 3; the former decomposes quickly, while the latter is quite stable at 193 K. The solvent-dependence of the bis-mu-oxo versus side-on peroxo preference observed for 2 is opposite to that reported for other known copper(I) complexes; the factors which may be responsible for the unusual behavior of 1/O2 versus 2/O2 (possibly N-H hydrogen bonding in the AN chemistry) are suggested. The factors which affect bis-mu-oxo versus side-on peroxo formation continue to be of interest.
Spectroscopy coupled with density functional calculations has been used to define the spin state, oxidation states, spin distribution, and ground state wave function of the mu4-sulfide bridged tetranuclear CuZ cluster of nitrous oxide reductase. Initial insight into the electronic contribution to N2O reduction is developed, which involves a sigma superexchange pathway through the bridging sulfide.
The geometric and electronic structure of the high-spin ferric active site of protocatechuate 3,4-dioxygenase (3,4-PCD) has been examined by absorption (Abs), circular dichroism (CD), magnetic CD (MCD), and variable-temperature-variable-field (VTVH) MCD spectroscopies. Density functional (DFT) and INDO/S-CI molecular orbital calculations provide complementary insight into the electronic structure of 3,4-PCD and allow an experimentally calibrated bonding scheme to be developed. Abs, CD, and MCD indicate that there are at least seven transitions below 35 000 cm(-1) which arise from tyrosinate ligand-to-metal-charge transfer (LMCT) transitions. VTVH MCD spectroscopy gives the polarizations of these LMCT bands in the principal axis system of the D-tensor, which is oriented relative to the molecular structure from the INDO/S-CI calculations. Three transitions are associated with the equatorial tyrosinate and four with the axial tyrosinate. This large number of transitions per tyrosinate is due to the pi and importantly the sigma overlap of the two tyrosinate valence orbitals with the metal d orbitals and is governed by the Fe-O-C angle and the Fe-O-C-C dihedral angles. The previously reported crystal structure indicates that the Fe-O-C angles are 133 degrees and 148 degrees for the equatorial and axial tyrosinate, respectively. Each tyrosinate has transitions at different energies with different intensities, which correlate with differences in geometry that reflect pseudo-sigma bonding to the Fe(III) and relate to reactivity. These factors reflect the metal-ligand bond strength and indicate that the axial tyrosinate-Fe(III) bond is weaker than the equatorial tyrosinate-Fe(III) bond. Furthermore, it is found that the differences in geometry, and hence electronic structure, are imposed by the protein. The consequences to catalysis are significant because the axial tyrosinate has been shown to dissociate upon substrate binding and the equatorial tyrosinate in the enzyme-substrate complex is thought to influence asymmetric binding of the chelated substrate moiety via a strong trans influence which activates the substrate for reaction with O2.
Frontier molecular orbital (FMO) theory coupled with density functional calculations has been applied to investigate the chemical reactivity of three key bioinorganic Cu(n)-O(2) complexes, the mononuclear end-on hydroperoxo-Cu(II), the side-on bridged mu-eta(2):eta(2)-O(2)(2-) Cu(II)(2) dimer and the bis-mu-oxo Cu(III)(2) dimer. Two acceptor orbitals (sigma* and pi*) of each complex and two types of donating substrates (sigma-substrate, phosphine; pi-substrate, alkylbenzene) are considered in the electrophilic attack mechanism. The angular dependences of different reaction pathways are determined using FMO theory and the angular overlap model. Including steric effects, the sigma*/sigma and pi*/pi pathways are found more reactive than the corresponding cross sigma*/pi and pi*/sigma pathways which have poor donor-acceptor orbital overlaps in the sterically constrained substrate access region.