Thomas G. Spiro

Is bound carbonyl linear or bent in heme proteins? Evidence from resonance Raman and infrared spectroscopic data

The vibrational data on heme-CO complexes are analyzed with a view toward elucidating the structural evidence for distorted FeCO units in heme proteins. The effects of FeCO distortion on the vibrational frequencies are attributable to back-bonding changes, as evidenced by the correlation of vF& with vCo. Steric restraint of perpendicular binding increases Fe CO back-donation slightly. This is opposite to the direction expected if the FeCO unit is bent but is consistent with the expected effect of two other distortion coordinates, FeCO tilting and porphyrin buckling. Larger effects on back-bonding are attributable to polar interactions of distal residues with the bound CO, especially H bonding. Substantial FeCO bending is inconsistent with the relatively small frequency separation between vFeC and dFeCO observed in all adducts so far examined; this separation is calculated to increase strongly with the bending angle due to mixing of the modes. For a given displacement of the 0 atom from the heme normal, the energy required for the three distortion coordinates is calculated to increase in the order tilting < buckling < bending. The energy is minimized, as are the individual angular displacements, if all three contribute to the actual molecular distortion, and the small net change may result from the opposing effects of bending and of tilting and buckling in vFcc. This concerted model of FeCO distortion is not inconsistent with the available X-ray crystal structure data. It is inconsistent with EXAFS analysis of MbCO at 4 K, which has yielded a small value, 127 f 4O, for the FeCO angle from the attenuation of second-shell scattering relative to a protein-free model. This attenuation might be due instead to other changes in second-shell scattering, e.g. due to porphyrin buckling. functional role in lowering the C O affinit;, thereby raising ;he threshold for CO poisoning. Whether the CO affinity is actually (2) (a) Peng, S. M.; Ibers, J. A. J . Am. Chem. SOC. 1976, 98, 8032. (b) Scheidt. W. R.; Haller. K. J.: Fons. M.: Fashiko. T.: Reed. C. A. Biochemistrv (1) (a) Kuriyan, J.; Wiltz, S.; Karplus, M.; Petsko, G. J . Mol. Bid . 1986, 192, 133-154. (b) Hanson, J. C.; Schoenborn, B. P. J . Mol. Biol. 1981, 153, 117. (c) Baldwin, J. M.; Chothia, C. J . Mol. Biol. 1979, 129, 175. (d) Baldwin, J. M. J . Mol. Biol. 1980, 136, 103. (e) Steigeman, W.; Weber, E. J . Mol. B i d . 1979, 127, 309. (0 Tucker, P. W.; Philipps, S. E. V.; Perutz. M. F.; Houtchens, R. A.; Caughey, W. S. Proc. Natl. Aidd. Sci. U S . A . 1978, 75, 1076. (g) Heidner, E. J.; Ladner, R. C.; Perutz, M. F. J. Mol. Bid . 1976, 104, 707. (h) Norvell, J. C.; Nunes, A. C.; Schoenborn, B. P. Science (Washington, D.C.) 1975, 190, 568. ( i ) Padlan, E. A.; Love, W. E. J . Biol. Chem. 1974,249,4067. (j) Huber, R.; Epp, 0.; Formanek, H. J . Mol. Biol. 1970, 52, 349. 1981, 20, 3653. (c) Caron, C.; Mitschler, A,; Riviere, G.; Ricard, L:; Schappacher, M.; Weiss, R. J . Am. Chem. SOC. 1979, 101, 7401. (3) (a) Collman, J. P.; Brauman, J. I.; Halbert, T. R.; Suslick, K. S. Proc. Natl. Acad. Sci. U.S.A. 1976, 73, 3333. (b) Collman, J. P.; Brauman, J. I.; Everson, B. L.; Sessler, J. L.; Morris, R. M.; Gibson, Q. H. J. Am. Chem. SOC. 1983, 105, 3052. 0002-7863/88/1510-6024

Bacteriogenic Manganese Oxides

01.50/0

Cobalt(I) porphyrin catalysis of hydrogen production from water

Microorganisms control the redox cycling of manganese in the natural environment. Although the homogeneous oxidation of Mn(II) to form manganese oxide minerals is slow, solid MnO(2) is the stable form of manganese in the oxygenated portion of the biosphere. Diverse bacteria and fungi have evolved the ability to catalyze this process, producing the manganese oxides found in soils and sediments. Other bacteria have evolved to utilize MnO(2) as a terminal electron acceptor in respiration. This Account summarizes the properties of Mn oxides produced by bacteria (bacteriogenic MnO(2)) and our current thinking about the biochemical mechanisms of bacterial Mn(II) oxidation. According to X-ray absorption spectroscopy and X-ray scattering studies, the MnO(2) produced by bacteria consists of stacked hexagonal sheets of MnO(6) octahedra, but these particles are extremely small and have numerous structural defects, particularly cation vacancies. The defects provide coordination sites for binding exogenous metal ions, which can be adsorbed to a high loading. As a result, bacterial production of MnO(2) influences the bioavailability of these metals in the natural environment. Because of its high surface area and oxidizing power, bacteriogenic MnO(2) efficiently degrades biologically recalcitrant organic molecules to lower-molecular-mass compounds, spurring interest in using these properties in the bioremediation of xenobiotic organic compounds. Finally, bacteriogenic MnO(2) is reduced to soluble Mn(II) rapidly in the presence of exogenous ligands or sunlight. It can therefore help to regulate the bioavailability of Mn(II), which is known to protect organisms from superoxide radicals and is required to assemble the water-splitting complex in photosynthetic organisms. Bioinorganic chemists and microbiologists have long been interested in the biochemical mechanism of Mn(IV) oxide production. The reaction requires a two-electron oxidation of Mn(II), but genetic and biochemical evidence for several bacteria implicate multicopper oxidases (MCOs), which are only known to engage one-electron transfers from substrate to O(2). In experiments with the exosporium of a Mn(II)-oxidizing Bacillus species, we could trap the one-electron oxidation product, Mn(III), as a pyrophosphate complex in an oxygen-dependent reaction inhibited by azide, consistent with MCO catalysis. The Mn(III) pyrophosphate complex can further act as a substrate, reacting in the presence of the exosporium to produce Mn(IV) oxide. Although this process appears to be unprecedented in biology, it is reminiscent of the oxidation of Fe(II) to form Fe(2)O(3) in the ferritin iron storage protein. However, it includes a critical additional step of Mn(III) oxidation or disproportionation. We shall continue to investigate this biochemically unique process with purified enzymes.

Hemoglobin: resonance Raman spectra.

Cobalt complexes of three water-soluble porphyrins have been examined as catalysts of H/sub 2/O reduction to H/sub 2/. They have been shown to catalyze H/sub 2/ production via controlled-potential electrolysis (-0.95 V vs. SCE at Hg-pool electrode; 0.1 M trifluoroacetic acid) at rates approximately 10-fold greater than background and with nearly 100% current efficiency. Reversible cyclic voltammograms were observed in dry Me/sub 2/SO, with Co(III)/Co(II) and Co(II)/Co(I) potentials near their expected values. Addition of water (0.5-2%) increased the cathodic peak and decreased the anodic peak at the Co(II)/Co(I) couple, consistent with electrocatalytic H/sub 2/O reduction. In aqueous buffers, catalytic currents were observed for CoTMPyP, which increased with decreasing pH, but at a rate less than expected, due to porphyrin adsorption. Adsorption was more pronounced for CoTMAP, which showed no catalytic current except at very low pH. The rate of Co/sup I/TMAP reaction with H/sub 2/O, however, is extremely fast as shown by spectrophotometry in dry acetonitrile; a lower limit for the Co(I)-H/sub 2/O reaction rate constant was estimated to be approximately 10/sup 4/ M/sup -1/s/sup -1/. Co(I) reactivity and cobalt hydride stability are discussed in comparison with the reactivity and stability of related compounds.

Cobalt(I) porphyrin catalysts of hydrogen production from water

Abstract Raman spectra have been recorded for several hemoglobin derivatives in dilute solution. They exhibit a complex set of bands, arising from vibrations of the heme groups. Two of the bands offer promise as structural probes, inasmuch as their intensities correlate inversely with the degree to which the iron atoms are out of plane.

Resonance Raman Spectroscopy as a Probe of Heme Protein Structure and Dynamics

Cobalt complexes of three water-soluble porphyrins have been examined as catalysts of H/sub 2/O reduction to H/sub 2/. They have been shown to catalyze H/sub 2/ production via controlled-potential electrolysis (-0.95 V vs. SCE at Hg-pool electrode; 0.1 M trifluoroacetic acid) at rates approximately 10-fold greater than background and with nearly 100% current efficiency. Reversible cyclic voltammograms were observed in dry Me/sub 2/SO, with Co(III)/Co(II) and Co(II)/Co(I) potentials near their expected values. Addition of water (0.5-2%) increased the cathodic peak and decreased the anodic peak at the Co(II)/Co(I) couple, consistent with electrocatalytic H/sub 2/O reduction. In aqueous buffers, catalytic currents were observed for CoTMPyP, which increased with decreasing pH, but at a rate less than expected, due to porphyrin adsorption. Adsorption was more pronounced for CoTMAP, which showed no catalytic current except at very low pH. The rate of Co/sup I/TMAP reaction with H/sub 2/O, however, is extremely fast as shown by spectrophotometry in dry acetonitrile; a lower limit for the Co(I)-H/sub 2/O reaction rate constant was estimated to be approximately 10/sup 4/ M/sup -1/s/sup -1/. Co(I) reactivity and cobalt hydride stability are discussed in comparison with the reactivity and stability of related compounds.

Cytochrome c: Resonance Raman spectra

Our understanding of metalloporphyrin resonance Raman spectra has advanced to the point where it is possible to obtain detailed information about the structure of the heme group in situ in heme proteins. The porphyrin skeletal mode frequencies can be analyzed in terms of the ligation and spin state of the heme and may provide information about protein-induced stresses. The high-frequency region of the spectrum also contains bands due to vibrations of the porphyrin peripheral substituents, which are potentially monitors of the protein contacts. In the low-frequency region, it is possible to locate bands, at least in some states of the heme protein, which are associated with vibrations of the axial ligands. They give direct information about the nature of the bonding to exogenous ligands or to the proximal protein residue. Thus, a variety of evidence is potentially available in the resonance Raman spectra from which a fairly complete picture of the heme site can be assembled for a particular protein in its various functional states. Detailed studies have been pursued for paradigmatic heme proteins, including myoglobin, hemoglobin, cytochrome c, horseradish peroxidase, and cytochrome oxidase. These studies provide a substantial data base from which the exploration of lesser known systems can be launched. Another extension of current knowledge to new frontiers is in the time domain, since pulsed lasers now make it feasible to carry out time-resolved resonance Raman studies on heme protein reactions. Time-resolved resonance Raman spectroscopy is capable of elucidating the temporal evolution of heme structure and provides a link between heme chemistry and protein dynamics. This link is being elucidated for hemoglobin and cytochrome c, where specific heme intermediates have been identified following ligand photodissociation or electron transfer.

Resonance-Raman evidence for anomalous heme structures in cytochrome c′ from Rhodopseudomonas palustris

Abstract Resonance Raman spectra have been recorded for cytochrome c in both reduced and oxidized forms. The reduced form exhibits intense resonance scattering, and its spectrum is similar to that of oxyhemoglobin. The oxidized form gives much weaker scattering, but shows new bands which do not appear in the spectrum of the reduced form, and may represent significant frequency shifts upon oxidation. The situation is in contrast to that observed for hemoglobin, for which the change in oxidation state produces very little change in the resonance spectrum.

Raman antiresonance: De‐enhancement of Raman intensity by forbidden electronic transitions

Abstract Resonance Raman spectra have been obtained for cytochrome c′ from Rhodopseudomonas palustris, using laser excitation frequencies in the region of both the Soret band and the α-β bands. The porphyrin-ring vibrational frequencies are compared with those of other heme proteins which are in well-defined spin and oxidation states. In alkaline solution (pH > 11.5) both oxidized and reduced forms show normal low-spin heme vibrational patterns. At lower pH the reduced form (high spin) shows a pattern similar to that of deoxyhemoglobin, i.e. that of a high-spin five-coordinate ferroheme. The lower pH forms of the oxidized protein exhibit anomalous spectra, with frequencies which are closer to expected low-spin than high-spin values, although their magnetic moments, while intermediate, are closer to expected high-spin values. Mixtures of high- and low-spin forms can be ruled out by the absence of two sets of vibrational frequencies, which are seen for hydroxymethemoglobin, an authentic mixed-spin heme protein. It is suggested that the heme groups are closer to being planar than are normal high-spin hemes, and that the spin state may be intermediate, s = 3 2 . At pH 10.3 the inferred structure is less planar than at pH 6.9, consistent with trends in the absorption spectrum. The depolarization ratios of the polarized approx. 1370-cm−1 Raman band indicate a lowering of 4-fold symmetry of the heme chromophore in the lower pH forms of both oxidized and reduced cytochrome c′, consistent with the observed splittings of the Soret band.

New light on NO bonding in Fe(III) heme proteins from resonance Raman spectroscopy and DFT modeling.

The intensities of the Raman lines of the transition metal complexes Co(en)33+, PdCl4=, PdBr4=, and PdI4=, have been investigated as a function of excitation energy in the vicinity of vibronically allowed ligand field electronic transitions. Whereas resonance enhancement usually results in local maxima in Raman intensities at the energies of allowed electronic transitions, we observe minima at the energies of these electronically forbidden transitions. These minima are ascribable to interference between the weak scattering from the forbidden electronic states and strong preresonance scattering from higher energy allowed electronic states. The excitation profiles can be reproduced with a three‐state scattering model, using reasonable parameters.