Intramolecular Electron Transfer Pathway Research Articles

Structure–function studies of glutamate synthases: A class of self‐regulated iron‐sulfur flavoenzymes essential for nitrogen assimilation

Glutamate synthases play with glutamine synthetase an essential role in nitrogen assimilation processes in microorganisms, plants, and lower animals by catalyzing the net synthesis of one molecule of L-glutamate from L-glutamine and 2-oxoglutarate. They exhibit a modular architecture with a common subunit or region, which is responsible for the L-glutamine-dependent glutamate synthesis from 2-oxoglutarate. Here, a PurF- (Type II- or Ntn-) type amidotransferase domain is coupled to the synthase domain, a (beta/alpha)8 barrel containing FMN and one [3Fe-4S]0,+1 cluster, through a approximately 30 angstroms-long intramolecular tunnel for the transfer of ammonia between the sites. In bacterial and eukaryotic GltS, reducing equivalents are provided by reduced pyridine nucleotides thanks to the stable association with a second subunit or region, which acts as a FAD-dependent NAD(P)H oxidoreductase and is responsible for the formation of the two low potential [4Fe-4S]+1,+2 clusters of the enzyme. In photosynthetic cells, reduced ferredoxin is the physiological reductant. This review focus on the mechanism of cross-activation of the synthase and glutaminase reactions in response to the bound substrates and the redox state of the enzyme cofactors, as well as on recent information on the structure of the alphabeta protomer of the NADPH-dependent enzyme, which sheds light on the intramolecular electron transfer pathway between the flavin cofactors.

Oxidative stabilization of iso‐1‐cytochromecby redox‐inspired protein engineering

Iso-1-cytochrome c, as any other hemeprotein, is able to react with hydrogen peroxide and to engage in the peroxidase cycle. However, peroxidases are irreversibly inactivated by their substrate, hydrogen peroxide. The oxidative inactivation of hemeproteins is mechanism based and arises as the consequence of unproductive electron abstraction reactions. Protein elements, such as the porphyrin ring or the protein backbone, act as simultaneous and competing electron sources even in the presence of exogenous reducing substrates, leading to a decline in activity. It is hypothetically possible to alter the intramolecular electron transfer pathways by direct replacement of low redox potential residues around the active site; as a consequence, the inactivation process would be delayed or even suppressed. To demonstrate this hypothesis, a redox-inspired strategy was implemented until an iso-1-cytochrome c variant fully stable at catalytic concentrations of hydrogen peroxide was obtained. This variant, harboring the N52I,W59F,Y67F,K79A,F82G substitutions, preserved the catalytic performance of the parental protein but achieved a 15-fold higher total-turnover number. The phenotype of this variant was reflected in the stability of its electronic components, allowing identification of a protein-based radical intermediate mechanistically similar to Compound I of classical peroxidases. The results presented here clearly demonstrate that redox-inspired protein engineering is a useful tool for the rational modulation of intramolecular electron transfer networks.

Light-Induced Reduction of Bovine Adrenodoxin via the Covalently Bound Ruthenium(II) Bipyridyl Complex: Intramolecular Electron Transfer and Crystal Structure

Bovine adrenodoxin (Adx) plays an important role in the electron-transfer process in the mitochondrial steroid hydroxylase system of the bovine adrenal cortex. Using electron paramagnetic resonance (EPR) spectroscopy, we showed that photoreduction of the [2Fe-2S] cluster of Adx via (4'-methyl-2,2'-bipyridine)bis(2,2'-bipyridine)ruthenium(II) [Ru(bpy)2(mbpy)] covalently attached to the protein surface can be used as a new approach to probe the "shuttle" hypothesis for the electron transfer by Adx. The 1.5 A resolution crystal structure of a 1:1 Ru(bpy)2(mbpy)-Adx(1-108) complex reveals the site of modification, Cys95, and allows to predict the possible intramolecular electron-transfer pathways within the complex. Photoreduction of uncoupled Adx, mutant Adx(1-108), and Ru(bpy)2(mbpy)-Adx(1-108) using safranin T as the mediating electron donor suggests that two electrons are transferred from the dye to Adx. The intramolecular photoreduction rate constant for the ruthenated Adx has been determined and is discussed according to the predicted pathways.

A Photoelectronic Switching Device Using a Mixed Self-Assembled Monolayer

A cathodic-anodic biway photoelectronic device has been successfully constructed using a self-assembled monolayer (SAM). The SAM consists of two kinds of photofunctional thiol derivatives, a ruthenium complex-viologen linked compound (RuVS) and a phthalocyanine derivative (PcS), on a gold electrode. Structural characterization of the SAM has been carried out by absorption spectroscopy, cyclic voltammetry, and differential pulse voltammetry. Photocurrent responses were measured in the presence of methyl viologen (MV2+) and oxygen as electron acceptors and triethanolamine (TEOA) as a sacrificial reagent. For the SAM of RuVS alone, intramolecular electron transfer (ET) was superior to intermolecular ET, resulting in anodic photocurrents even in the presence of MV2+ and oxygen at 0 V vs Ag/AgCl. On the contrary, only cathodic photocurrents were observed at 0 V for the SAM of PcS alone. Photocurrents from the mixed SAM of RuVS and PcS were roughly the sum of individual photocurrents from RuVS and PcS. In fact, photocurrents from the mixed SAM of RuVS and PcS were observed in the anodic direction below approximately 550 nm, and in the cathodic direction above approximately 550 nm at 0 V vs Ag/AgCl. In the case of the mixed SAM of RuS (ruthenium complex disulfide) and PcS, only cathodic photocurrents were observed at 0 V vs Ag/AgCl, due to the lack of an intramolecular ET pathway. The results indicate that in the mixed SAM of RuVS and PcS both dyes can individually function for opposite photocurrent generation. We have also applied the mixed SAM as a photoelectronic logic device by using two LEDs (470 and 640 nm). The system clearly operated as an XOR logic device.

Crystal Structures of the Met148Leu and Ser86Asp Mutants of Rusticyanin from Thiobacillus ferrooxidans: Insights into the Structural Relationship with the Cupredoxins and the Multi Copper Proteins

The crystal structures of the Met148Leu and Ser86Asp mutants of rusticyanin are presented at 1.82 and 1.65 Å resolution, respectively. Both of these structures have two molecules in the asymmetric unit compared to the one present in the crystal form of the native protein. This provides an opportunity to investigate intramolecular electron transfer pathways in rusticyanin. The redox potential of the Met148Leu mutant (∼800 mV) is elevated compared to that of the native protein (∼670 mV at pH 3.2) while that of the Ser86Asp mutant (∼623 mV at pH 3.2) is decreased. The effect of the Ser86Asp mutation on the hydrogen bonding near the type 1 Cu site is discussed and hence its role in determining acid stability is examined. The type 1 Cu site of Met148Leu mimics the structural and biochemical characteristics of those found in domain II of ceruloplasmin and fungal laccase. Moreover, the native rusticyanin's cupredoxin core and the type 1 Cu site closely resemble those found in ascorbate oxidase and nitrite reductase. Structure based phylogenetic trees have been re-examined in view of the additional structural data on rusticyanin and fungal laccase. We confirm that rusticyanin is in the same class as nitrite reductase domain 2, laccase domain 3 and ceruloplasmin domains 2, 4 and 6.

Synthesis, Structural Characterization, and Reactivity of Organolanthanide Complexes Derived from a New, Versatile Boron-Bridged Ligand, iPr2NB(C9H7)(C2B10H11)

New boron-bridged ligands incorporating both indenyl and carboranyl moieties have been prepared. Reaction of iPr2NB(C9H7)Cl with 1 equiv of Li2C2B10H10 gave, after treatment with 1 equiv of n-BuLi, the dilithium salt [iPr2NB(C9H6)(C2B10H10)]Li2(OEt2)2 (1), which was conveniently converted into its neutral counterpart iPr2NB(C9H7)(C2B10H11) (2) via reaction with excess C5H6. Mixing 1 and 2 in a 1:1 molar ratio quantitatively afforded the monolithium salt [iPr2NB(C9H6)(C2B10H11)]Li(THF)2 (3). Treatment of LnI2 with 1 equiv of 1 generated the trivalent organolanthanide complexes meso-[{η5:σ-iPr2NB(C9H6)(C2B10H10)}2Ln][Li(S)n] (Ln = Sm, Yb; S = THF, n = 4; S = DME, n = 3), which was also prepared in a much higher yield by reaction of LnI2 with 1 equiv of 1 followed by treatment with 1 equiv of 3. Both inter- and intramolecular electron-transfer pathways are proposed for these reactions. Reaction of LnCl3 with 1 or 2 equiv of 1 gave the same ionic complexes meso-[{η5:σ-iPr2NB(C9H6)(C2B10H10)}2Ln][Li(S)n] (Ln = Nd, Y, Yb; S = THF, n = 4; S = DME, n = 3). Silylamine elimination reactions of 2 and Ln[N(SiHMe2)2]3(THF)2 resulted in clear formation of [η5:σ-iPr2NB(C9H6)(C2B10H10)]LnN(SiHMe2)2(THF)2 (Ln = Nd (8), Er (9), Y (10)). Treatment of 8 with Me3NHCl or Me3SiCl led to the isolation of 2 and NdCl3(DME)2 (11), respectively. The new complexes were fully characterized by various spectroscopic data and element analyses. Some were further confirmed by single-crystal X-ray analyses. Complex 8 is an active catalyst for the polymerization of methyl methacrylate (MMA) in toluene, affording syn-rich poly(MMA)s.

Evidence for Anisotropic Coupling between the Protein Environment and the Copper Site in Azurin from Resonance Raman Spectroscopy

A key problem in biological electron transfer is the independent measurement of the parameters upon which the rate constant for electron transfer depends. In this paper, we report evidence that quantitative resonance Raman spectroscopy may be such an independent probe of the electronic coupling parameter in proteins. The resonance Raman cross-sections of four species of azurin have been measured throughout the ca. 600-nm absorption band. The total resonance Raman intensity, as quantified by the cross-sections, is very similar for the four azurins in the 200-700 cm - 1 region. However, the resonance Raman intensity is distributed differentially among the vibrational modes, sometimes also accompanied by small frequency shifts. Correlations are attempted between this re-distribution of resonance Raman intensity and Cu-S bond length, Cu out-of-plane distance, backbone structure, and protein environment. Of these four parameters, the only significant correlation exists with the protein environment. A detailed analysis of the residues within 10 A of the copper site demonstrates a strong, anisotropic coupling between the copper site and Trp48, a residue previously implicated in intramolecular electron transfer pathways of azurin.

Characterization of the Intramolecular Electron Transfer Pathway from 2-Hydroxyphenazine to the Heterodisulfide Reductase fromMethanosarcina thermophila

Heterodisulfide reductase (HDR) is a component of the energy-conserving electron transfer system in methanogens. HDR catalyzes the two-electron reduction of coenzyme B-S-S-coenzyme M (CoB-S-S-CoM), the heterodisulfide product of the methyl-CoM reductase reaction, to free thiols, HS-CoB and HS-CoM. HDR from Methanosarcina thermophila contains two b-hemes and two [Fe(4)S(4)] clusters. The physiological electron donor for HDR appears to be methanophenazine (MPhen), a membrane-bound cofactor, which can be replaced by a water-soluble analog, 2-hydroxyphenazine (HPhen). This report describes the electron transfer pathway from reduced HPhen (HPhenH(2)) to CoB-S-S-CoM. Steady-state kinetic studies indicate a ping-pong mechanism for heterodisulfide reduction by HPhenH(2) with the following values: k(cat) = 74 s(-1) at 25 degrees C, K(m) (HPhenH(2)) = 92 microm, K(m) (CoB-S-S-CoM) = 144 microm. Rapid freeze-quench EPR and stopped-flow kinetic studies and inhibition experiments using CO and diphenylene iodonium indicate that only the low spin heme and the high potential FeS cluster are involved in CoB-S-S-CoM reduction by HPhenH(2). Fe-S cluster disruption by mersalyl acid inhibits heme reduction by HPhenH(2), suggesting that a 4Fe cluster is the initial electron acceptor from HPhenH(2). We propose the following electron transfer pathway: HPhenH(2) to the high potential 4Fe cluster, to the low potential heme, and finally, to CoB-S-S-CoM.

Carbon versus Silicon Bridges. Synthesis of a New Versatile Ligand and Its Applications in Organolanthanide Chemistry

A new carbon-bridged versatile ligand Me2C(C9H7)(C2B10H11) (1) has been designed and successfully prepared by treatment of Li2C2B10H10 with 1 equiv of 6,6-dimethylbenzofulvene followed by hydrolysis with a saturated NH4Cl aqueous solution. 1 can be conveniently converted into the monoanion [Me2C(C9H6)(C2B10H11)]Li (2) and the dianion [Me2C(C9H6)(C2B10H10)]Li2 (3) by treatment with 1 or 2 equiv of n-BuLi, respectively. Both NaNH2 and NaH can only convert 1 into the monoanion, but cannot deprotonate the CH proton of the carborane cage in 1. These results differ significantly from those of a closely related analogue, Me2Si(C9H7)(C2B10H11). Treatment of SmI2 with 1 equiv of 3, followed by reaction with 1 equiv of 2, gave the redox product rac-[Li(DME)2][{η5:σ-Me2C(C9H6)(C2B10H10)}2Sm] (4). 4 can also be prepared by reaction of SmI2 with 1 equiv of 3 in a relatively lower yield. These two reactions may undergo different pathways, an intramolecular electron-transfer pathway for the former and an intermolecular electron-transfer pathway for the latter. The latter reaction can be accelerated by addition of CS2 or PhC⋮CPh, which led to the isolation of rac-[Li(THF)4][{η5:σ-Me2C(C9H6)(C2B10H10)}2Sm] (5). Unlike the SmI2 case, an equimolar reaction between 3 and YbI2 afforded the Yb(II) compound [η5:σ-Me2C(C9H6)(C2B10H10)]Yb(DME)2 (6). 6 can react with 1 equiv of 2 to generate a C−H bond reduction product, rac-[Li(DME)3][{η5:σ-Me2C(C9H6)(C2B10H10)}2Yb]·C6H5CH3 (7). Reaction of LnCl3 with 1 or 2 equiv of 2 yielded organolanthanide dichloride and monochloride compounds, respectively, [η5-Me2C(C9H6)(C2B10H11)]GdCl2(THF)2 (15) and [η5-Me2C(C9H6)(C2B10H11)]2LnCl(THF)(OEt2) (Ln = Y (8), Yb (9)). Treatment of 9 with 1 or 2 equiv of MeLi gave deprotonation products rac-[{η5:σ-Me2C(C9H6)(C2B10H10)}{η5-Me2C(C9H6)(C2B10H11)}]Yb(μ-Cl)Li(DME)2 (10) and rac-[Li(DME)2][{η5:σ-Me2C(C9H6)(C2B10H10)}2Yb] (11), respectively. Reaction of LnCl3 with 2 equiv of 3 also afforded ionic compounds rac-[Li(DME)2][{η5:σ-Me2C(C9H6)(C2B10H10)}2Ln] (Ln = Yb (11), Nd (12), Er (13)). Recrystallization of 7 from a mixed solvent of toluene/DME (10:1) gave meso-[Li(DME)3][{η5:σ-Me2C(C9H6)(C2B10H10)}2Yb]·2C6H5CH3 (14). All of these compounds were fully characterized by various spectroscopic and elemental analyses. The molecular structures of 4−7, 11, 12, and 14 have been confirmed by single-crystal X-ray analyses. The structural analyses reveal that the anions in 7 (or 11) and 14 are one pair of diastereomers.

Myrothecium verrucaria bilirubin oxidase and its mutants for potential copper ligands.

Bilirubin oxidase (EC:1.3.3.5) purified from a culture medium of Myrothecium verrucaria MT-1 (authentic enzyme) catalyzes the oxidation of bilirubin to biliverdin in vitro and recombinant enzyme (wild type) was obtained by using an overexpression system of the bilirubin oxidase gene with Aspergillus oryzae harboring an expression vector. The absorption and ESR spectra showed that both bilirubin oxidases are multicopper oxidases containing type 1, type 2, and type 3 coppers similar to laccase, ascorbate oxidase, and ceruloplasmin. Site-directed mutagenesis has been performed for the possible ligands of each type of copper. In some mutants, Cys457 --> Val, Ala, His94 --> Val, and His134.136 --> Val, type 1 and type 2 copper centers were perturbed completely and the enzyme activity was completely lost. Differing from the holoenzyme, these mutants showed type 3 copper signals. However, the optical and magnetic properties characteristic of type 1 copper were retained even by mutating one of the type 1 copper ligands, i.e., a mutant, Met467 --> Gly, showed a weak but apparent enzyme activity. A double mutant His456.458 --> Val had only type 1 Cu, showing a blue band at 600 nm (epsilon = 1.6 x 10(3)) and an ESR signal with very narrow hyperfine splitting (A parallel = 7.2 x 10(-)3 cm-1). Since the type 2 and type 3 coppers are not present, the mutant did not show enzyme activity. These results strongly imply that the peculiar sequence in bilirubin oxidase, His456-Cys457-His458, forms an intramolecular electron-transfer pathway between the type 1 copper site and the trinuclear center composed of the type 2 and type 3 copper sites.

3Fe-4S] to [4Fe-4S] cluster conversion in Desulfovibrio fructosovorans [NiFe] hydrogenase by site-directed mutagenesis.

The role of the high potential [3Fe-4S]1+,0 cluster of [NiFe] hydrogenase from Desulfovibrio species located halfway between the proximal and distal low potential [4Fe-4S]2+,1+ clusters has been investigated by using site-directed mutagenesis. Proline 238 of Desulfovibrio fructosovorans [NiFe] hydrogenase, which occupies the position of a potential ligand of the lacking fourth Fe-site of the [3Fe-4S] cluster, was replaced by a cysteine residue. The properties of the mutant enzyme were investigated in terms of enzymatic activity, EPR, and redox properties of the iron-sulfur centers and crystallographic structure. We have shown on the basis of both spectroscopic and x-ray crystallographic studies that the [3Fe-4S] cluster of D. fructosovorans hydrogenase was converted into a [4Fe-4S] center in the P238 mutant. The [3Fe-4S] to [4Fe-4S] cluster conversion resulted in a lowering of approximately 300 mV of the midpoint potential of the modified cluster, whereas no significant alteration of the spectroscopic and redox properties of the two native [4Fe-4S] clusters and the NiFe center occurred. The significant decrease of the midpoint potential of the intermediate Fe-S cluster had only a slight effect on the catalytic activity of the P238C mutant as compared with the wild-type enzyme. The implications of the results for the role of the high-potential [3Fe-4S] cluster in the intramolecular electron transfer pathway are discussed.

Intramolecular “Through-Bond” and “Through-Space” Electron Transfer Pathways in Covalently Linked Porphyrin-Quinone Molecules

Abstract In order to understand electron transfer (ET) pathways through bonds and through space, a porphyrin-spacer-benzoquinone molecule, where the spacer is spiro[4.4]nonane and a phenyl group is inserted between the space of the redox pair, was prepared. On the basis of MM2 calculation, the edge-to-edge distances between the phenyl and the porphyrin or quinone rings are in the range of 2.5—4.2 Å. Therefore, it was expected that there are two possible intramolecular ET pathways in the compound, i.e., through bond and through space. The ET rates for charge separation process (kcs) were obtained on the basis of fluorescence lifetime. The kcs value of the above compound is almost identical with that of a reference compound in which no inserted phenyl ring is present. Therefore, it was concluded that the inserted π-system between the redox pair is not used as a stepping stone in ET pathways; in other words, ET in the present molecules takes place in a through-bond mechanism. To compete with fast ET with through-bond pathways in covalently linked donor–acceptor systems where the two ET pathways are possible, through-space ET may function under quite limited conditions.

Complete coordination of the four Fe-S centers of the beta subunit from Escherichia coli nitrate reductase. Physiological, biochemical, and EPR characterization of site-directed mutants lacking the highest or lowest potential [4Fe-4S] clusters.

The beta subunit of the nitrate reductase A from Escherichia coli contains four groups of cysteine residues (I-IV) which are thought to bind the four iron-sulfur centers (1-4) of the enzyme. The fourth Cys residue of each group was replaced by Ala by site-directed mutagenesis, which led to the C26A, C196A, C227A, and C263A mutants. Physiological and biochemical effects of the mutations were investigated on both the membrane-bound and the soluble forms of the enzyme. In addition, detailed redox titrations of the mutants were monitored by EPR spectroscopy. The C196A and C227A mutations resulted in the full loss of the four Fe-S clusters and of the Mo-cofactor, leading to inactive enzymes. In contrast, the C26A and C263A mutants retained significant nitrate reductase activities. The EPR analysis showed that the highest redox potential [4Fe-4S] cluster (center 1) was selectively removed by the C263A mutation and that the C26A replacement likely eliminated the lowest potential [4Fe-4S] cluster (center 4). In both mutants, the three remaining Fe-S clusters kept the same spectral and redox properties as in the wild type enzyme. These results enabled the determination of the Cys ligands of center 1 to be completed and led to a proposed model for the coordination of the four Fe-S centers by the four Cys groups of the beta subunit. In this model, the four clusters are organized in two pairs, (center 1, center 4) and (center 2, center 3), which is in good agreement with the magnitude of intercenter magnetic interactions observed by EPR and with the stability of the different mutants. The possible implications on the intramolecular electron transfer pathway are discussed.

The effects of p-hydroxymercuribenzoic acid modification and heat treatment on the CuA reduction potential of cytochrome c oxidase.

p-Hydroxymercuribenzoic acid modification of cytochrome c oxidase converts the CuA center into a type 2 copper site while heat treatment of the oxidase in lauryl maltoside can transform CuA almost stoichiometrically (greater than 90%) to a blue type 1 copper site. These modifications of the protein have previously been shown to have a profound effect on the dioxygen reduction and proton pumping activities of the enzyme (Li, P. M., Morgan, J. E., Nilsson, T., Ma, M., and Chan, S. I. (1988) Biochemistry 27, 7538; Nilsson, T., Gelles, J., Li, P. M., and Chan, S. I. (1988) Biochemistry 27, 296; Sone, N., and Nicholls, P. (1984) Biochemistry 23, 6550). In this work, the intrinsic reduction potentials and the midpoint reduction potentials of the "CuA" site in both these modified oxidases have been measured under various conditions in order to clarify the intramolecular electron transfer pathways in these systems. The study reveals that the CuA intrinsic reduction potential decreases by almost 150 mV upon p-hydroxymercuribenzoic acid modification whereas it increases by 100 mV upon heat treatment. In addition, the redox interactions between CuA and the remaining metal centers are perturbed upon CuA modification. It is argued that these results bear on the role of CuA in the proton-pumping paradigm of cytochrome c oxidase.

Ferredoxin electron transfer site on cytochrome c3. Structural hypothesis of an intramolecular electron transfer pathway within a tetra-heme cytochrome.

To specify electron exchanges involving Desulfovibrio desulfuricans Norway tetra-heme cytochrome c3, the chemical modification of arginine 73 residue, was performed. Biochemical and biophysical studies have shown that the modified cytochrome retains its ability to both interact and act as an electron carrier with its redox partners, ferredoxin and hydrogenase. Moreover, the chemical modification effects on the cytochrome c3 1H NMR spectrum were similar to that induced by the presence of ferredoxin. This suggests that arginine 73 is localized on the cytochrome c3 ferredoxin interacting site. The identification of heme 4, the closest heme to arginine 73, as the ferredoxin interacting heme helps us to hypothesize about the role of the three other hemes in the molecule. A structural hypothesis for an intramolecular electron transfer pathway, involving hemes 4, 3 and 1, is proposed on the basis of the crystal structures of D. vulgaris Miyazaki and D. desulfuricans Norway cytochromes c3. The unique role of some structural features (alpha helix, aromatic residues) intervening between the heme groups, is proposed.

Long-range intramolecular electron transfer in aromatic radical anions and binuclear transition metal complexes

Intramolecular electron transfer (ET) over distances up to about 10 Å between states in which the electron is localized on donor and acceptor groups by interaction with molecular or external solvent nuclear motion occurs, in particular, in two classes of systems. The excess electron in anionic radicals containing two aromatic end groups connected by a flexible polymethylene chain or a rigid cyclohexane frame is thus trapped on either aromatic end group, and ET between these groups can be detected by ESR techniques. Intramolecular ET also occurs in binuclear transition metal complexes in which the coupling between the metal centers [(Ru(II)/Ru(III) and Ru(II)/Co(III) couples] is sufficiently weak (class I or II mixed valence compounds). The ET mechanism can involve either direct transfer between the donor and acceptor groups or a higher order mechanism in which ET proceeds through intermediate states corresponding to the localization of the electron on the interconnecting fragments. We have derived rate expressions for the different conceivable intramolecular ET pathways. In particular, we have incorporated the nuclear motion not only in the initial and final states but also in the intermediate states. We have furthermore provided criteria for the distinction between direct ET, ET through high-energy (superexchange), and low-energy (’’radical’’ intermediate mechanism) intermediate states on the basis of experimental rate data and illustrated this with data for organic radical anions, and for intramolecular and inner sphere ET for transition metal complexes.