Deconstructing the electron transfer chain in a complex molybdoenzyme: Assimilatory nitrate reductase from Neurospora crassa.

Nitrate reductase (NR) from the fungus Neurospora crassa is a complex homodimeric metallo-flavoenzyme, where each protomer contains three distinct domains; the catalytically active terminal molybdopterin cofactor, a central heme-containing domain, and an FAD domain which binds with the natural electron donor NADPH. Here, we demonstrate the catalytic voltammetry of variants of N. crassa NRs on a modified Au electrode with the electrochemically reduced forms of benzyl viologen (BV2+) and anthraquinone sulfonate (AQS-) acting as artificial electron donors. The biopolymer chitosan used to entrap NR on the electrode non-covalently and the enzyme film was both stable and highly active. Electrochemistry was conducted on two distinct forms; one lacking the FAD cofactor and the other lacking both the FAD and heme cofactors. While both enzymes showed catalytic nitrate reductase activity, removal of the heme cofactor resulted in a more significant effect on the rate of nitrate reduction. Electrochemical simulation was carried out to enable kinetic characterisation of both the NR:nitrate and NR:mediator reactions.

Methods for determining the reduction potentials of flavin enzymes.

A key factor for flavoenzyme activity is the reduction potential of the bound flavin. The reduction potentials of protein-bound flavins span approximately a 500-mV range consistent with flavoenzymes having critical roles in metabolism and a variety of biological processes. Redox potentials of flavoenzymes have traditionally been determined using an electrode-based system with either direct or indirect electrochemical coupling between the protein and the working electrode. An electrode independent method, however, is also now commonly used and involves calculating the unknown flavin reduction potential of the protein from the known reduction potential of a reference or indicator dye. Here, the "classic" potentiometric method and the xanthine/xanthine oxidase methods are described. Both methods rely on equilibrium between protein-bound flavin and redox dyes. The potentiometric method measures the equilibrated redox potential of the protein-dye mixture whereas the xanthine/xanthine oxidase technique relies on slow continuous enzymatic reduction to maintain a constant equilibrium between the protein and the dyes. Because electrochemical equipment is not required, the xanthine/xanthine oxidase method is more accessible and convenient for researchers seeking to determine reduction potentials of flavoproteins or other biological redox centers such as hemes. The xanthine/xanthine oxidase method has been used to determine flavin reduction potentials from +132 to -417mV, demonstrating it is suitable for characterizing the redox properties of most flavoproteins.

Reciprocal regulation of sulfite oxidation and nitrite reduction by mitochondrial sulfite oxidase.

The oxygen-independent nitrate-nitrite-nitric oxide (NO) pathway is considered as a substantial source of NO in mammals. Dietary nitrate/nitrite are distributed throughout the body and reduced to NO by the action of various enzymes. The intermembrane spaced (IMS), molybdenum cofactor-dependent sulfite oxidase (SO) was shown to catalyze such a nitrite reduction. In this study we asked whether the primary function of SO - sulfite oxidation - and its novel function - nitrite reduction - impact each other. First, we utilized benzyl viologen as artificial electron donor to investigate steady state NO synthesis by SO and found fast (kcat = 14 s-1) nitrite reduction of SO full-length and its isolated molybdenum domain at pH 6.5. Next, we determined the impact of nitrite on pre-steady state kinetics in SO catalysis and identified nitrite as a pH-dependent inhibitor of SO reductive and oxidative half reaction. Finally, we report on the time-dependent formation of the paramagnetic Mo(V) species following nitrite reduction and demonstrate that sulfite inhibits nitrite reduction. In conclusion, we propose a pH-dependent reciprocal regulation of sulfite oxidation and nitrite reduction by each substrate, thus facilitating quick responses to hypoxia induced changes in the IMS, which may function in protecting the cell from reactive oxygen species production.

Rapid small intestinal permeability assay based on riboflavin and lactulose detected by bis-boronic acid appended benzyl viologens.

BACKGROUND: Although organoboronic acids are efficient high-throughput sugar sensors, they have not been pursued for gut permeability studies. A modification of the lactulose/mannitol assay is described by which small intestinal permeability is assessed at the time of urine collection using a lactulose/riboflavin ratio. METHODS: Volunteers ingested 50mg riboflavin and either 5 g mannitol or 10 g lactulose. Urine was collected for 6 hrs. Riboflavin was assayed by autofluorescence. Riboflavin was removed by C18 solid phase extraction. Lactulose and mannitol were then assayed using 1,1'-bis(2-boronobenzyl)-4,4'-bipyridinium (4,4'oBBV) coupled to the fluorophore HPTS. RESULTS: The temporal profile over 6 hrs for riboflavin paralleled mannitol. Riboflavin recovery in urine was 11.1 ± 1.9 % (mean ± SEM, n=7), similar to mannitol. There was selective binding of 4,4'oBBV to lactulose, likely involving cooperativity between the fructose and galactose moieties. Lower limits of detection and quantification were 90 and 364 µM. The lactulose assay was insensitive to other permeability probes (e.g., sucrose, sucralose) while tolerating glucose or lactose. This assay can be adapted to automated systems. Stability of 4,4'oBBV exceeds 4 years. CONCLUSIONS: Riboflavin measured by autofluorescence combined with lactulose measured with 4,4'oBBV represents a useful new chemistry for rapid measurement of intestinal permeability with excellent stability, cost and throughput benefits.

Photochromic hybrid containing in situ-generated benzyl viologen and novel trinuclear [Bi3Cl14]5-: improved photoresponsive behavior by the π···π interactions and size effect of inorganic oligomer.

Two new member of (V)((2n+2)/2)[Bi(2n)Cl(8n+2)] series hybrids, (BzV)2[Bi2Cl10] (1) and (BzV)5[Bi3Cl14]2·(C6H5CH2)2O (2) (where BzV(2+) = N,N'-dibenzyl-4,4'-bipyridinium and (C6H5CH2)2O = dibenzyl ether) have been obtained, and compound 2 contains an unprecedented discrete trimer [Bi3Cl14](5-) counterion. The novel in situ-synthesized symmetric viologen cation with aromatic groups on both sides of 4,4'-bipy would provide more opportunities to create π···π interactions to optimize the photochromic property of the hybrid, and different bismuthated-halide oligomers enable us to discuss the size effect in this series of compounds. Both 1 and 2 are photochromic, and their photoresponsive rate is faster than that of reported viologen-metal halide hybrids. Experimental and theoretical data illustrated that the size of the inorganic oligomer can significantly influence the photoresponsive rate of the viologen dication, and the π···π interaction behaves as not only a powerful factor to stabilize the viologen monocation radical but also the second electron-transfer pathway, from a π-conjugated substituent to a viologen cation, for the photochromic process.

Mediated electrochemistry of nitrate reductase from Arabidopsis thaliana.

Herein we report the mediated electrocatalytic voltammetry of the plant molybdoenzyme nitrate reductase (NR) from Arabidopsis thaliana using the established truncated molybdenum-heme fragment at a glassy carbon (GC) electrode. Methyl viologen (MV), benzyl viologen (BV), and anthraquinone-2-sulfonic acid (AQ) are employed as effective artificial electron transfer partners for NR, differing in redox potential over a range of about 220 mV and delivering different reductive driving forces to the enzyme. Nitrate is reduced at the Mo active site of NR, yielding the oxidized form of the enzyme, which is reactivated by the electro-reduced form of the mediator. Digital simulation was performed using a single set of enzyme dependent parameters for all catalytic voltammetry obtained under different sweep rates and various substrate or mediator concentrations. The kinetic constants from digital simulation provide new insight into the kinetics of the NR catalytic mechanism.

A simple chip free-flow electrophoresis for monosaccharide sensing via supermolecule interaction of boronic acid functionalized quencher and fluorescent dye.

Here, a simple micro free-flow electrophoresis (µFFE) was developed for fluorescence sensing of monosaccharide via supermolecule interaction of synthesized boronic acid functionalized benzyl viologen (ο-BBV) and fluorescent dye. The µFFE contained two open electrode cavities and an ion-exchange membrane was sandwiched between two polymethylmethacrylate plates. The experiments demonstrated the following merits of developed µFFE: (i) up to 90.5% of voltage efficiency due to high conductivity of ion-exchange membrane; (ii) a strong ability against influence of bubble produced in two electrodes due to open design of electrode cavities; and (iii) reusable and washable separation chamber (45 mm × 17 mm × 100 µm, 77 µL) avoiding the discard of µFFE due to blockage of solute precipitation in chamber. Remarkably, the µFFE was first designed for the sensing of monosaccharide via the supermolecule interaction of synthesized ο-BBV, fluorescent dye, and monosaccharide. Under the optimized conditions, the minimum concentration of monosaccharide that could be detected was 1 × 10(-11) M. Finally, the developed device was used for the detection of 0.3 mM glucose spiked in human urine. All of the results demonstrated the feasibility of monosaccharide detection via the µFFE.

The autocatalytic step is an integral part of the hydrogenase cycle.

We earlier proved the involvement of an autocatalytic step in the oxidation of H(2) by HynSL hydrogenase from Thiocapsa roseopersicina, and demonstrated that two enzyme forms interact in this step. Using a modified thin-layer reaction chamber which permits quantitative analysis of the concentration of the reaction product (reduced benzyl viologen) in the reaction volume during the oxidation of H(2), we now show that the steady-state concentration of the product displays a strong enzyme concentration dependence. This experimental fact can be explained only if the previously detected autocatalytic step occurs inside the catalytic enzyme-cycle and not in the enzyme activation process. Consequently, both interacting enzyme forms should participate in the catalytic cycle of the enzyme. As far as we are aware, this is the first experimental observation of such a phenomenon resulting in an apparent inhibition of the enzyme. It is additionally concluded that the interaction of the two enzyme forms should result in a conformational change in the enzyme-substrate form. This scheme is very similar to that of prion reactions. Since merely a few molecules are involved at some point of the reaction, this process is entirely stochastic in nature. We have therefore developed a stochastic calculation method, calculations with which lent support to the conclusion drawn from the experiment.

A highly sensitive water-soluble system to sense glucose in aqueous solution.

A glucose sensing switch is formed by water soluble conjugated polymer (PP-S-BINOL) and boronic acid-functionalized benzyl viologen (o-BBV). The two-component system shows a high sensitivity for glucose sensing with a 17-fold increase in the fluorescence intensity in the presence of 100 mM glucose.

Electro-fluorescence switching of bis-imidazolium ionic liquids.

Room-temperature ionic liquids (RTILs) containing bis-imidazolium salts were explored to control their optical properties of them in an organic device. The neat bis-imidazolium salts showed ionic conductivity of 3.5 x 10(-4) S/cm at room-temperature and the electrochemical window was exhibited within +/-2.5 V in a two electrode cell. The bis-imidazolium salts were transparent yellow and showed fluorescence upon excitation with light in the range of 360 to 500 nm. A two electrode organic ionic liquid cell was fabricated using a mixture of PEO200Blm-TFSI and electroactive molecules to control the emission property of the ionic liquid by electrochemical methods. The first example of the ionic liquid containing electro-fluorescence switch was explored with an electroactive benzyl viologen (BzV) blend of bis-imidazolium ionic liquid.

Measuring the pH dependence of hydrogenase activities.

The pH dependences of activities of homogenous hydrogenases of Thiocapsa roseopersicina and Desulfomicrobium baculatum in the reaction of hydrogen uptake in solution in the presence of benzyl viologen and the pH dependences of catalytic currents of hydrogen oxidation by electrodes on which these hydrogenases were immobilized were compared. Maximal activities of the hydrogenases from T. roseopersicina and D. baculatum in the reaction hydrogen uptake in solution were observed at pH 9.5 and 8.5, respectively. However, the steady-state current caused by catalytic uptake of hydrogen was maximal for the T. roseopersicina hydrogenase-containing electrode at pH 5.5-6.5 under overvoltage of 30-60 mV, whereas for electrodes with D. baculatum hydrogenase it was maximal at pH 6.0-6.5. Analysis of these data suggests that pH-dependent changes in the hydrogenase activities in solution during hydrogen uptake are due not only to the effect of proton concentration on the enzyme conformation or protonation of certain groups of the enzyme active center, but they are rather indicative of changes in free energy of the reaction accompanying changes in pH.

Nanoparticle films as electrodes: voltammetric sensitivity to the nanoparticle energy gap.

Electron-transfer reactions of redox solutes at electrode/solution interfaces are facilitated when their formal potentials match, or are close to, the energy of an electronic state of the electrode. Metal electrodes have a continuum of electronic levels, and redox reactions occur without restraint over a wide span of electrode potentials. This paper shows that reactions on electrodes composed of films of metal nanoparticles do have constraints when the nanoparticles are sufficiently small and molecule-like so as to exhibit energy gaps, and resist electron transfers with redox solutes at potentials within the energy gap. When solute formal potentials are near the electronic states of the nanoparticles in the film, electron-transfer reactions can occur. The electronic states of the nanoparticle film electrodes are reflected in the formal potentials of the electrochemical reactions of the dissolved nanoparticles at naked metal electrodes. These ideas are demonstrated by voltammetry of aqueous solutions of the redox solutes methyl viologen, ruthenium hexammine, and two ferrocene derivatives at films on electrodes of 1.1 nm core diameter Au nanoparticles coated with protecting monolayers of phenylethanethiolate ligands. The methyl viologen solute is unreactive at the nanoparticle film electrode, having a formal potential lying in the nanoparticle's energy gap. The other solutes exhibit electron transfers, albeit slowed by the electron hopping resistance of the nanoparticle film. The nanoparticles are not linked together, being insoluble in the aqueous medium; a small amount of an organic additive (acetonitrile) facilitates observing the redox solute voltammetry.

Characterization of catalytic properties of hydrogenase isolated from the unicellular cyanobacterium Gloeocapsa alpicola CALU 743.

The main catalytic properties of the Hox type hydrogenase isolated from the Gloeocapsa alpicola cells have been studied. The enzyme effectively catalyzes reactions of oxidation and evolution of H2 in the presence of methyl viologen (MV) and benzyl viologen (BV). The rates of these reactions in the interaction with the physiological electron donor/acceptor NADH/NAD+ are only 3-8% of the MV(BV)-dependent values. The enzyme interacts with NADP+ and NADPH, but is more specific to NAD+ and NADH. Purification of the hydrogenase was accompanied by destruction of its multimeric structure and the loss of ability to interact with pyridine nucleotides with retained activity of the hydrogenase component (HoxYH). To show the catalytic activity, the enzyme requires reductive activation, which occurs in the presence of H2, and NADH accelerates this process. The final hydrogenase activity depends on the redox potential of the activation medium (E(h)). At pH 7.0, the enzyme activity in the MV-dependent oxidation of H2 increased with a decrease in E(h) from -350 mV and reached the maximum at E(h) of about -390 mV. However, the rate of H2 oxidation in the presence of NAD+ in the E(h) range under study was virtually constant and equal to 7-8% of the maximal rate of H2 oxidation in the presence of MV.

An autocatalytic step in the reaction cycle of hydrogenase from Thiocapsa roseopersicina can explain the special characteristics of the enzyme reaction.

A moving front has been observed as a special pattern during the hydrogenase-catalyzed reaction of hydrogen uptake with benzyl viologen as electron acceptor in a thin-layer reaction chamber. Such fronts start spontaneously and at random times at different points of the reaction chamber; blue spheres are seen expanding at constant speed and amplitude. The number of observable starting points depends on the hydrogenase concentration. Fronts can be initiated by injecting either a small amount of completed reaction mixture or activated hydrogenase, but not by injecting a low concentration of reduced benzyl viologen. These characteristics are consistent with an autocatalytic reaction step in the enzyme reaction. The special characteristics of the hydrogen-uptake reaction in the bulk reaction (a long lag phase, and the enzyme concentration dependence of the lag phase) support the autocatalytic nature. We conclude that there is at least one autocatalytic reaction step in the hydrogenase-catalyzed reaction. The two possible autocatalytic schemes for hydrogenase are prion-type autocatalysis, in which two enzyme forms interact, and product-activation autocatalysis, where a reduced electron acceptor and an inactive enzyme form interact. The experimental results strongly support the occurrence of prion-type autocatalysis.

Toxicity of redox cycling pesticides in primary mesencephalic cultures.

A loss of nigrostriatal dopaminergic neurons is the primary neurodegenerative feature of Parkinson's disease. Paraquat, a known redox cycling herbicide, has recently been shown to kill selectively nigrostriatal dopaminergic cells in the mouse model. The purpose of this study was to test the ability of paraquat and other redox cycling pesticides to damage dopaminergic neurons in primary mesencephalic cultures. Addition of paraquat, diquat, or benzyl viologen to mesencephalic cultures induced morphological changes (e.g., dystrophic neuronal processes) consistent with dopaminergic cell injury. The three pesticides also caused cell death as assessed by a reduction of the number of tyrosine hydroxylase-immunoreactive neurons and a dose-dependent decrease in [(3)H]dopamine uptake. Quite interestingly, diquat and benzyl viologen were significantly more toxic than paraquat, probably reflecting their more pronounced ability to trigger redox cycling reactions. The data support a role of redox cycling as a mechanism of dopaminergic cell degeneration and suggest that the property of redox cycling should be taken into consideration when evaluating putative environmental risk factors for Parkinson's disease.

Reactions of H2, CO, and O2 with active [NiFe]-hydrogenase from Allochromatium vinosum. A stopped-flow infrared study.

The Ni-Fe site in the active membrane-bound [NiFe]-hydrogenase from Allochromatium vinosum can exist in three different redox states. In the most oxidized state (Ni(a)-S) the nickel is divalent. The most reduced state (Ni(a)-SR) likewise has Ni(2+), while the intermediate state (Ni(a)-C) has Ni(3+). The transitions between these states have been studied by stopped-flow Fourier transform infrared spectroscopy. It is inferred from the data that the Ni(a)-S --> Ni(a)-C* and Ni(a)-C* --> Ni(a)-SR transitions induced by dihydrogen require one of the [4Fe-4S] clusters to be oxidized. Enzyme in the Ni(a)-S* state with all of the iron-sulfur clusters reduced reacts with dihydrogen to form the Ni(a)-SR state in milliseconds. By contrast, when one of the cubane clusters is oxidized, the Ni(a)-S state reacts with dihydrogen to form the Ni(a)-C state with all of the iron-sulfur clusters reduced. The competition between dihydrogen and carbon monoxide for binding to the active site was dependent on the redox state of the nickel ion. Formation of the Ni(a)-S.CO state (Ni(2+)) by reacting CO with enzyme in the Ni(a)-SR and Ni(a)-S states (Ni(2+)) is considerably faster than its formation from enzyme in the Ni(a)-C* (Ni(3+)) state. Excess oxygen converted hydrogen-reduced enzyme to the inactive Ni(r)* state within 158 ms, suggesting a direct reaction at the Ni-Fe site. With lower O(2) concentrations the formation of intermediate states was observed. The results are discussed in the light of the present knowledge of the structure and mechanism of action of the A. vinosum enzyme.

Diphenylene iodonium as an inhibitor for the hydrogenase complex of Rhodobacter capsulatus. Evidence for two distinct electron donor sites.

The photosynthetic bacterium Rhodobacter capsulatus synthesises a membrane-bound [NiFe] hydrogenase encoded by the H2 uptake hydrogenase (hup)SLC structural operon. The hupS and hupL genes encode the small and large subunits of hydrogenase, respectively; hupC encodes a membrane electron carrier protein which may be considered as the third subunit of the uptake hydrogenase. In Wolinella succinogenes, the hydC gene, homologous to hupC, has been shown to encode a low potential cytochrome b which mediates electron transfer from H2 to the quinone pool of the bacterial membrane. In whole cells of R. capsulatus or intact membrane preparation of the wild type strain B10, methylene blue but not benzyl viologen can be used as acceptor of the electrons donated by H2 to hydrogenase; on the other hand, membranes of B10 treated with Triton X-100 or whole cells of a HupC- mutant exhibit both benzyl viologen and methylene blue reductase activities. We report the effect of diphenylene iodonium (Ph2I), a known inhibitor of mitochondrial complex I and of various monooxygenases on R. capsulatus hydrogenase activity. With H2 as electron donor, Ph2I inhibited partially the methylene blue reductase activity in an uncompetitive manner, and totally benzyl viologen reductase activity in a competitive manner. Furthermore, with benzyl viologen as electron acceptor, Ph2I increased dramatically the observed lagtime for dye reduction. These results suggest that two different sites exist on the electron donor side of the membrane-bound [NiFe] hydrogenase of R. capsulatus, both located on the small subunit. A low redox potential site which reduces benzyl viologen, binds Ph2I and could be located on the distal [Fe4S4] cluster. A higher redox potential site which can reduce methylene blue in vitro could be connected to the high potential [Fe3S4] cluster and freely accessible from the periplasm.

Picosecond transient absorption spectroscopy in the blue spectral region of photosystem I.

Picosecond transient absorption difference spectroscopy in the blue wavelength region (380-500 nm) was used to study the early electron acceptors in photosystem I. Samples were photosystem I core particles with about 100 chlorophylls per reaction center isolated from the cyanobacterium Synechocystis sp. PCC 6803. After excitation at 590 nm at room temperature, decay-associated spectra (DAS) were determined from global analysis in the blue region, yielding two transient components and one nondecaying component. A 3 ps decay phase is interpreted as primarily due to antenna excited-state redistribution. A 28 ps decay phase is interpreted as due to overall excited-state decay by electron transfer. The nondecaying component is ascribed to the difference spectrum of P(700) and the quinone or A(1) electron acceptor (P(700)(+)A(1)(-) - P(700)A(1)). Decay curves on the millisecond time scale at different wavelengths were measured with an autoxidizable artificial electron acceptor, benzyl viologen, and the (P(700)(+) - P(700)) difference spectrum was constructed. The (A(1)(-) - A(1)) difference spectrum was obtained by taking the difference between the above two difference spectra. A parallel picosecond experiment under strongly reducing conditions was also done as a control experiment. These conditions stabilize the electron on an earlier acceptor, A(0). The nondecaying component of the DAS at low potential was assigned to (P(700)(+)A(0)(-) - P(700)A(0)) since the electron-transfer pathway from A(0) to A(1) was blocked. The [(P(700)(+)A(0)(-) - P(700)A(0)) - (P(700)(+) - P(700))] subtraction gives a spectrum, interpreted as the (A(0)(-) - A(0)) difference spectrum of a chlorophyll a molecule, consistent with previous studies. The (A(1)(-) - A(1)) spectrum resolved on the picosecond time scale shows significant differences with similar spectra measured on longer time scales. These differences may be due to electrochromic effects and spectral evolution.

Reactions of dimethylsulfoxide reductase from Rhodobacter capsulatus with dimethyl sulfide and with dimethyl sulfoxide: complexities revealed by conventional and stopped-flow spectrophotometry.

Improved assays for the molybdenum enzyme dimethylsulfoxide reductase (DMSOR) with dimethyl sulfoxide (DMSO) and with dimethyl sulfide (DMS) as substrates are described. Maximum activity was observed at pH 6.5 and below and at 8.3, respectively. Rapid-scan stopped-flow spectrophotometry has been used to investigate the reduction of the enzyme by DMS to a species previously characterized by its UV-visible spectrum [McAlpine, A. S., McEwan, A. G., and Bailey, S. (1998) J. Mol. Biol. 275, 613-623], and its subsequent reoxidation by DMSO. Both these two-electron reactions were faster than enzyme turnover under steady-state conditions, indicating that one-electron reactions with artificial dyes were rate-limiting. Second-order rate constants for the two-electron reduction and reoxidation reactions at pH 5.5 were (1.9 +/- 0.1) x 10(5) and (4.3 +/- 0.3) x 10(2) M-1 s-1, respectively, while at pH 8.0, the catalytic step was rate-limiting (62 s-1). Kinetically, for the two-electron reactions, the enzyme is more effective in DMS oxidation than in DMSO reduction. Reduction of DMSOR by DMS was incomplete below approximately 1 mM DMS but complete at higher concentrations, implying that the enzyme's redox potential is slightly higher than that of the DMS-DMSO couple. In contrast, reoxidation of the DMS-reduced state by DMSO was always incomplete, regardless of the DMSO concentration. Evidence for the existence of a spectroscopically indistinguishable reduced state, which could not be reoxidized by DMSO, was obtained. Brief reaction (less than approximately 15 min) of DMS with DMSOR was fully reversible on removal of the DMS. However, in the presence of excess DMS, a further slow reaction occurred aerobically, but not anaerobically, to yield a stable enzyme form having a lambdamax at 660 mn. This state (DMSORmod) retained full activity in steady-state assays with DMSO, but was inactive toward DMS. It could however be reconverted to the original resting state by reduction with methyl viologen radical and reoxidation with DMSO. We suggest that in this enzyme form two of the dithiolene ligands of the molybdenum have dissociated and formed a disulfide. The implications of this new species are discussed in relation both to conflicting published information for DMSOR from X-ray crystallography and to previous spectroscopic data for its reduced forms.