Spectroscopic and theoretical approaches for studying radical reactions in class I ribonucleotide reductase

Extending the Range of Distances Accessible by 19F Electron-Nuclear Double Resonance in Proteins Using High-Spin Gd(III) Labels

Fluorine electron-nuclear double resonance (19F ENDOR) has recently emerged as a valuable tool in structural biology for distance determination between F atoms and a paramagnetic center, either intrinsic or conjugated to a biomolecule via spin labeling. Such measurements allow access to distances too short to be measured by double electron-electron resonance (DEER). To further extend the accessible distance range, we exploit the high-spin properties of Gd(III) and focus on transitions other than the central transition (|-1/2⟩ ↔ |+1/2⟩), that become more populated at high magnetic fields and low temperatures. This increases the spectral resolution up to ca. 7 times, thus raising the long-distance limit of 19F ENDOR almost 2-fold. We first demonstrate this on a model fluorine-containing Gd(III) complex with a well-resolved 19F spectrum in conventional central transition measurements and show quantitative agreement between the experimental spectra and theoretical predictions. We then validate our approach on two proteins labeled with 19F and Gd(III), in which the Gd-F distance is too long to produce a well-resolved 19F ENDOR doublet when measured at the central transition. By focusing on the |-5/2⟩ ↔ |-3/2⟩ and |-7/2⟩ ↔ |-5/2⟩ EPR transitions, a resolution enhancement of 4.5- and 7-fold was obtained, respectively. We also present data analysis strategies to handle contributions of different electron spin manifolds to the ENDOR spectrum. Our new extended 19F ENDOR approach may be applicable to Gd-F distances as large as 20 Å, widening the current ENDOR distance window.

The stability of oxygen-centered radicals and its response to hydrogen bonding interactions

The stability of various alkoxy/aryloxy/peroxy radicals, as well as TEMPO and triplet dioxygen (3 O2 ) has been explored at a variety of theoretical levels. Good correlations between RSEtheor and RSEexp are found for hybrid DFT methods, for compound schemes such as G3B3-D3, and also for DLPNO-CCSD(T) calculations. The effects of hydrogen bonding interactions on the stability of oxygen-centered radicals have been probed by addition of a single solvating water molecule. While this water molecule always acts as a H-bond donor to the oxygen-centered radical itself, it can act as a H-bond donor or acceptor to the respective closed-shell parent.

Fate of oxygen species from O2 activation at dimetal cofactors in an oxidase enzyme revealed by 57Fe nuclear resonance X-ray scattering and quantum chemistry

Oxygen (O₂) activation is a central challenge in chemistry and catalyzed at prototypic dimetal cofactors in biological enzymes with diverse functions. Analysis of intermediates is required to elucidate the reaction paths of reductive O₂ cleavage. An oxidase protein from the bacterium Geobacillus kaustophilus, R2lox, was used for aerobic in-vitro reconstitution with only ⁵⁷Fe(II) or Mn(II) plus ⁵⁷Fe(II) ions to yield [FeFe] or [MnFe] cofactors under various oxygen and solvent isotopic conditions including ^16/18O and H/D exchange. ⁵⁷Fe-specific X-ray scattering techniques were employed to collect nuclear forward scattering (NFS) and nuclear resonance vibrational spectroscopy (NRVS) data of the R2lox proteins. NFS revealed Fe/Mn(III)Fe(III) cofactor states and Mössbauer quadrupole splitting energies. Quantum chemical calculations of NRVS spectra assigned molecular structures, vibrational modes, and protonation patterns of the cofactors, featuring a terminal water (H₂O) bound at iron or manganese in site 1 and a metal-bridging hydroxide (μOH^-) ligand. A procedure for quantitation and correlation of experimental and computational NRVS difference signals due to isotope labeling was developed. This approach revealed that the protons of the ligands as well as the terminal water at the R2lox cofactors exchange with the bulk solvent whereas ¹⁸O from ¹⁸O₂ cleavage is incorporated in the hydroxide bridge. In R2lox, the two water molecules from four-electron O₂ reduction are released in a two-step reaction to the solvent. These results establish combined NRVS and QM/MM for tracking of iron-based oxygen activation in biological and chemical catalysts and clarify the reductive O₂ cleavage route in an enzyme.

Mimicking Class I b Mn2 -Ribonucleotide Reductase: A MnII2 Complex and Its Reaction with Superoxide

A fascinating discovery in the chemistry of ribonucleotide reductases (RNRs) has been the identification of a dimanganese (Mn₂ ) active site in class I b RNRs that requires superoxide anion (O₂^.- ), rather than dioxygen (O₂ ), to access a high-valent Mn₂ oxidant. Complex 1 ([Mn₂ (O₂ CCH₃ )(N-Et-HPTB)](ClO₄ )₂ , N-Et-HPTB=N,N,N',N'-tetrakis(2-(1-ethylbenzimidazolyl))-2-hydroxy-1,3-diaminopropane) was synthesised in high yield (90 %). 1 was reacted with O₂^.- at -40 °C resulting in the formation of a metastable species (2). 2 displayed electronic absorption features (λ_max =460, 610 nm) typical of a Mn-peroxide species and a 29-line EPR signal typical of a Mn^II Mn^III entity. Mn K-edge X-ray absorption near-edge spectroscopy (XANES) suggested a formal oxidation state change of Mn^II₂ in 1 to Mn^II Mn^III for 2. Electrospray ionisation mass spectrometry (ESI-MS) suggested 2 to be a Mn^II Mn^III -peroxide complex. 2 was capable of oxidizing ferrocene and weak O-H bonds upon activation with proton donors. Our findings provide support for the postulated mechanism of O₂^.- activation at class I b Mn₂ RNRs.

A >200 meV Uphill Thermodynamic Landscape for Radical Transport in Escherichia coli Ribonucleotide Reductase Determined Using Fluorotyrosine-Substituted Enzymes

Escherichia coli class Ia ribonucleotide reductase (RNR) converts ribonucleotides to deoxynucleotides. A diferric-tyrosyl radical (Y₁₂₂•) in one subunit (β2) generates a transient thiyl radical in another subunit (α2) via long-range radical transport (RT) through aromatic amino acid residues (Y₁₂₂ ⇆ [W₄₈] ⇆ Y₃₅₆ in β2 to Y₇₃₁ ⇆ Y₇₃₀ ⇆ C₄₃₉ in α2). Equilibration of Y₃₅₆•, Y₇₃₁•, and Y₇₃₀• was recently observed using site specifically incorporated unnatural tyrosine analogs; however, equilibration between Y₁₂₂• and Y₃₅₆• has not been detected. Our recent report of Y₃₅₆• formation in a kinetically and chemically competent fashion in the reaction of β2 containing 2,3,5-trifluorotyrosine at Y₁₂₂ (F₃Y₁₂₂•-β2) with α2, CDP (substrate), and ATP (effector) has now afforded the opportunity to investigate equilibration of F₃Y₁₂₂• and Y₃₅₆•. Incubation of F₃Y₁₂₂•-β2, Y₇₃₁F-α2 (or Y₇₃₀F-α2), CDP, and ATP at different temperatures (2–37 °C) provides ΔE°′(F₃Y₁₂₂•–Y₃₅₆•) of 20 ± 10 mV at 25 °C. The pH dependence of the F₃Y₁₂₂• ⇆ Y₃₅₆• interconversion (pH 6.8–8.0) reveals that the proton from Y₃₅₆ is in rapid exchange with solvent, in contrast to the proton from Y₁₂₂. Insertion of 3,5-difluorotyrosine (F₂Y) at Y₃₅₆ and rapid freeze-quench EPR analysis of its reaction with Y₇₃₁F-α2, CDP, and ATP at pH 8.2 and 25 °C shows F₂Y₃₅₆• generation by the native Y₁₂₂•. F_nY-RNRs (n = 2 and 3) together provide a model for the thermodynamic landscape of the RT pathway in which the reaction between Y₁₂₂ and C₄₃₉ is ∼200 meV uphill.

Protonation State of MnFe and FeFe Cofactors in a Ligand-Binding Oxidase Revealed by X-ray Absorption, Emission, and Vibrational Spectroscopy and QM/MM Calculations

Enzymes with a dimetal-carboxylate cofactor catalyze reactions among the top challenges in chemistry such as methane and dioxygen (O₂) activation. Recently described proteins bind a manganese-iron cofactor (MnFe) instead of the classical diiron cofactor (FeFe). Determination of atomic-level differences of homo- versus hetero-bimetallic cofactors is crucial to understand their diverse redox reactions. We studied a ligand-binding oxidase from the bacterium Geobacillus kaustophilus (R2lox) loaded with a FeFe or MnFe cofactor, which catalyzes O₂ reduction and an unusual tyrosine-valine ether cross-link formation, as revealed by X-ray crystallography. Advanced X-ray absorption, emission, and vibrational spectroscopy methods and quantum chemical and molecular mechanics calculations provided relative Mn/Fe contents, X-ray photoreduction kinetics, metal-ligand bond lengths, metal-metal distances, metal oxidation states, spin configurations, valence-level degeneracy, molecular orbital composition, nuclear quadrupole splitting energies, and vibrational normal modes for both cofactors. A protonation state with an axial water (H₂O) ligand at Mn or Fe in binding site 1 and a metal-bridging hydroxo group (μOH) in a hydrogen-bonded network is assigned. Our comprehensive picture of the molecular, electronic, and dynamic properties of the cofactors highlights reorientation of the unique axis along the Mn-OH₂ bond for the Mn1(III) Jahn-Teller ion but along the Fe-μOH bond for the octahedral Fe1(III). This likely corresponds to a more positive redox potential of the Mn(III)Fe(III) cofactor and higher proton affinity of its μOH group. Refined model structures for the Mn(III)Fe(III) and Fe(III)Fe(III) cofactors are presented. Implications of our findings for the site-specific metalation of R2lox and performance of the O₂ reduction and cross-link formation reactions are discussed.

Non-enzymatic ribonucleotide reduction in the prebiotic context

Model studies of prebiotic chemistry have revealed compelling routes for the formation of the building blocks of proteins and RNA, but not DNA. Today, deoxynucleotides required for the construction of DNA are produced by reduction of nucleotides catalysed by ribonucleotide reductases, which are radical enzymes. This study considers potential non-enzymatic routes via intermediate radicals for the ancient formation of deoxynucleotides. In this context, several mechanisms for ribonucleotide reduction, in a putative H2 S/HS(.) environment, are characterized using computational chemistry. A bio-inspired mechanistic cycle involving a keto intermediate and HSSH production is found to be potentially viable. An alternative pathway, proceeding through an enol intermediate is found to exhibit similar energetic requirements. Non-cyclical pathways, in which HSS(.) is generated in the final step instead of HS(.) , show a markedly increased thermodynamic driving force (ca. 70 kJ mol(-1) ) and thus warrant serious consideration in the context of the prebiotic ribonucleotide reduction.

Transfer hydrogenation in open-shell nucleotides - a theoretical survey

The potential of a larger number of sugar models to act as dihydrogen donors in transfer hydrogenation reactions has been quantified through the calculation of hydrogenation energies of the respective oxidized products. Comparison of the calculated energies to hydrogenation energies of nucleobases shows that many sugar fragment radicals can reduce pyrimidine bases such as uracil in a strongly exothermic fashion. The most potent reducing agent is the C3' ribosyl radical. The energetics of intramolecular transfer hydrogenation processes has also been calculated for a number of uridinyl radicals. The largest driving force for such a process is found for the uridin-C3'-yl radical, whose rearrangement to the C2'-oxidized derivative carrying a dihydrouracil is predicted to be exothermic by 61.1 kJ/mol in the gas phase.

Hydrogen transfer in SAM-mediated enzymatic radical reactions

S-adenosylmethionine (SAM) plays an essential role in a variety of enzyme-mediated radical reactions. One-electron reduction of SAM is currently believed to generate the C5'-desoxyadenosyl radical, which subsequently abstracts a hydrogen atom from the actual substrate in a catalytic or a non-catalytic fashion. Using a combination of theoretical and experimental bond dissociation energy (BDE) data, the energetics of these radical processes have now been quantified. SAM-derived radicals are found to react with their respective substrates in an exothermic fashion in enzymes using SAM in a stoichiometric (non-catalytic) way. In contrast, the catalytic use of SAM appears to be linked to a sequence of moderately endothermic and exothermic reaction steps. The use of SAM in spore photoproduct lyase (SPL) appears to fit neither of these general categories and appears to constitute the first example of a SAM-initiated radical reaction propagated independently of the cofactor.

Studies of ribonucleotide reductase in crucian carp-an oxygen dependent enzyme in an anoxia tolerant vertebrate

The enzyme ribonucleotide reductase (RNR) catalyzes the conversion of ribonucleotides to deoxyribonucleotides, the precursors for DNA. RNR requires a thiyl radical to activate the substrate. In RNR of eukaryotes (class Ia RNR), this radical originates from a tyrosyl radical formed in reaction with oxygen (O(2)) and a ferrous di-iron center in RNR. The crucian carp (Carassius carassius) is one of very few vertebrates that can tolerate several months completely without oxygen (anoxia), a trait that enables this fish to survive under the ice in small ponds that become anoxic during the winter. Previous studies have found indications of cell division in this fish after 7 days of anoxia. This appears nearly impossible, as DNA synthesis requires the production of new deoxyribonucleotides and therefore active RNR. We have here characterized RNR in crucian carp, to search for adaptations to anoxia. We report the full-length sequences of two paralogs of each of the RNR subunits (R1i, R1ii, R2i, R2ii, p53R2i and p53R2ii), obtained by cloning and sequencing. The mRNA levels of these subunits were measured with quantitative PCR and were generally well maintained in hypoxia and anoxia in heart and brain. We also report maintained or increased mRNA levels of the cell division markers proliferating cell nuclear antigen (PCNA), brain derived neurotrophic factor (BDNF) and Ki67 in anoxic hearts and brains. Electron paramagnetic resonance (EPR) measurements on in vitro expressed crucian carp R2 and p53R2 proteins gave spectra similar to mammalian RNRs, including previously unpublished human and mouse p53R2 EPR spectra. However, the radicals in crucian carp RNR small subunits, especially in the p53R2ii subunit, were very stable at 0°C. A long half-life of the tyrosyl radical during wintertime anoxia could allow for continued cell division in crucian carp.

High-valent [MnFe] and [FeFe] cofactors in ribonucleotide reductases

Ribonucleotide reductases (RNRs) are essential for DNA synthesis in most organisms. In class-Ic RNR from Chlamydia trachomatis (Ct), a MnFe cofactor in subunit R2 forms the site required for enzyme activity, instead of an FeFe cofactor plus a redox-active tyrosine in class-Ia RNRs, for example in mouse (Mus musculus, Mm). For R2 proteins from Ct and Mm, either grown in the presence of, or reconstituted with Mn and Fe ions, structural and electronic properties of higher valence MnFe and FeFe sites were determined by X-ray absorption spectroscopy and complementary techniques, in combination with bond-valence-sum and density functional theory calculations. At least ten different cofactor species could be tentatively distinguished. In Ct R2, two different Mn(IV)Fe(III) site configurations were assigned either L(4)Mn(IV)(μO)(2)Fe(III)L(4) (metal-metal distance of ~2.75Å, L = ligand) prevailing in metal-grown R2, or L(4)Mn(IV)(μO)(μOH)Fe(III)L(4) (~2.90Å) dominating in metal-reconstituted R2. Specific spectroscopic features were attributed to an Fe(IV)Fe(III) site (~2.55Å) with a L(4)Fe(IV)(μO)(2)Fe(III)L(3) core structure. Several Mn,Fe(III)Fe(III) (~2.9-3.1Å) and Mn,Fe(III)Fe(II) species (~3.3-3.4Å) likely showed 5-coordinated Mn(III) or Fe(III). Rapid X-ray photoreduction of iron and shorter metal-metal distances in the high-valent states suggested radiation-induced modifications in most crystal structures of R2. The actual configuration of the MnFe and FeFe cofactors seems to depend on assembly sequences, bound metal type, valence state, and previous catalytic activity involving subunit R1. In Ct R2, the protonation of a bridging oxide in the Mn(IV)(μO)(μOH)Fe(III) core may be important for preventing premature site reduction and initiation of the radical chemistry in R1.

Modeling of the ribonucleotide reductases substrate reaction. Hydrogen atom abstraction by a thiyl free radical and detection of the ribosyl-based carbon radical by pulse radiolysis

The 1,4-anhydro-5-deoxy-6-thio-D-ribo-hexofuranitol (1) was prepared from 1,2-O-isopropylidene-α-D-glucose in 10 steps. In a key step treatment of the 1,2-O-isopropylidenehexofuranose derivative with BF3/Et3SiH effected deacetonization and reductive deoxygenation at carbon 1. Pulse radiolysis experiments with 6-thiohexofuranitol 1 and its disulfide derivative demonstrated formation of the ribosyl-based carbon-centered radical upon generation of 6-thiyl radical in basic medium. The proposed [1,5]-hydrogen shift abstraction with generation of the C3 radical mimics the initial substrate reaction of RNRs. The reversible H-atom transfer has been quantified and was correlated with the computed rate constants for the internal H atom abstraction from C1, C2, C3 and C4 by the thiyl radical. The energy barrier for the H3 and H4 abstractions were calculated to be most favorable with the corresponding barriers of 11.1 and 11.2 kcal/mol, respectively.

Radical stability and its role in synthesis and catalysis

The thermodynamic stability of carbon-centered radicals may be defined in quantitative terms using the hydrogen transfer reaction shown in . The stability values obtained in this way for substituted systems may be understood as the stabilizing or destabilizing influence of substituents on the neighboring radical center. This approach can be easily adapted to oxygen- or sulfur-centered radicals as expressed in eqn (b). [Formula: see text] The stability values obtained in this way do not only serve as a quantitative basis for the discussion of substituent effects, but also allow for quantitative estimates of reaction energies for hydrogen transfer reactions. These occur as key steps in a multitude of synthetically useful radical-chain processes in apolar solution, in enzyme-mediated non-chain processes in biological systems, and in the oxidative degradation of a variety of biomolecules. The review will highlight the usefulness of radical stability values for the rationalization of successful (and not so successful) synthetic radical reactions as well as the potential design of new radical reactions.

Biochemical evidence for the tyrosine involvement in cationic intermediate stabilization in mouse beta-carotene 15, 15'-monooxygenase

Background

β-carotene 15,15'-monooxygenase (BCMO1) catalyzes the crucial first step in vitamin A biosynthesis in animals. We wished to explore the possibility that a carbocation intermediate is formed during the cleavage reaction of BCMO1, as is seen for many isoprenoid biosynthesis enzymes, and to determine which residues in the substrate binding cleft are necessary for catalytic and substrate binding activity. To test this hypothesis, we replaced substrate cleft aromatic and acidic residues by site-directed mutagenesis. Enzymatic activity was measured in vitro using His-tag purified proteins and in vivo in a β-carotene-accumulating E. coli system.

Results

Our assays show that mutation of either Y235 or Y326 to leucine (no cation-π stabilization) significantly impairs the catalytic activity of the enzyme. Moreover, mutation of Y326 to glutamine (predicted to destabilize a putative carbocation) almost eliminates activity (9.3% of wt activity). However, replacement of these same tyrosines with phenylalanine or tryptophan does not significantly impair activity, indicating that aromaticity at these residues is crucial. Mutations of two other aromatic residues in the binding cleft of BCMO1, F51 and W454, to either another aromatic residue or to leucine do not influence the catalytic activity of the enzyme. Our ab initio model of BCMO1 with β-carotene mounted supports a mechanism involving cation-π stabilization by Y235 and Y326.

Conclusions

Our data are consistent with the formation of a substrate carbocation intermediate and cation-π stabilization of this intermediate by two aromatic residues in the substrate-binding cleft of BCMO1.

The effect of nitric oxide on vaccinia virus-encoded ribonucleotide reductase

Growth inhibition of the DNA virus vaccinia (VACV) by NO is known to occur at the level of DNA synthesis. This inhibition is partially reversed by addition of deoxyribonucleosides, suggesting that NO or NO-related species inhibit viral ribonucleotide reductase (RR). However, the effect of NO on VACV-encoded RR or other DNA-synthesizing enzymes has not been demonstrated. In order to study the effects of NO on VACV-encoded RR, DNA polymerase (DNA pol) and thymidine kinase (TK), we generated a VACV recombinant expressing murine macrophage iNOS under control of a VACV early/late promoter p7.5. Using this recombinant, we demonstrate that expression of iNOS and the resulting production of NO inhibit activity of the viral RR, but not of viral DNA pol and TK. This NO-mediated inhibition of viral RR occurred around the same time as the increase of ADP levels, while it preceded the block in VACV DNA synthesis and the decrease of ATP levels. In addition, we tested the effects of DPTA/NONOate on the growth of different VACV mutants. Fold-inhibition of the growth of VACV deletion mutant for TK was comparable to that of wild-type VACV. VACV containing amplification of the gene for the small subunit of RR appeared to be least sensitive to DPTA/NONOate, while VACV deletion mutant for the large subunit of RR was most sensitive. The results provide a direct evidence for NO-mediated inhibition of VACV-encoded RR.

Density functional theory study of Fe(IV) d-d optical transitions in active-site models of class I ribonucleotide reductase intermediate X with vertical self-consistent reaction field methods

The Fe(IV) d-d transition energies for four active-site structural models of class I ribonucleotide reductase (RNR) intermediate X have been calculated using broken-symmetry density functional theory incorporated with the Slater transition state vertical self-consistent reaction field methodology. Our model I (Figure 1), which contains two mu-oxo bridges, one terminal water, and one bidentate carboxylate group, yields the best Fe(IV) d-d transition energies compared with experiment. Our previous study (J. Am. Chem. Soc. 2005, 127, 15778-15790) also shows that most of the other calculated properties of model I in both native and mutant Y122F forms, including geometries, spin states, pKa's, 57Fe, 1H, and 17O hyperfine tensors, and 57Fe Mössbauer isomer shifts and quadrupole splittings, are also the best in agreement with the available experimental data. This model is likely to represent the active-site structure of the intermediate state X of RNR.

Long-range distance determinations in biomacromolecules by EPR spectroscopy

Electron paramagnetic resonance (EPR) spectroscopy provides a variety of tools to study structures and structural changes of large biomolecules or complexes thereof. In order to unravel secondary structure elements, domain arrangements or complex formation, continuous wave and pulsed EPR methods capable of measuring the magnetic dipole coupling between two unpaired electrons can be used to obtain long-range distance constraints on the nanometer scale. Such methods yield reliably and precisely distances of up to 80 A, can be applied to biomolecules in aqueous buffer solutions or membranes, and are not size limited. They can be applied either at cryogenic or physiological temperatures and down to amounts of a few nanomoles. Spin centers may be metal ions, metal clusters, cofactor radicals, amino acid radicals, or spin labels. In this review, we discuss the advantages and limitations of the different EPR spectroscopic methods, briefly describe their theoretical background, and summarize important biological applications. The main focus of this article will be on pulsed EPR methods like pulsed electron-electron double resonance (PELDOR) and their applications to spin-labeled biosystems.

High-field pulsed electron-electron double resonance spectroscopy to determine the orientation of the tyrosyl radicals in ribonucleotide reductase

Class I ribonucleotide reductases (RNRs) are composed of two subunits, R1 and R2. The R2 subunit contains the essential diferric cluster-tyrosyl radical (Y.) cofactor, and R1 is the site of the conversion of nucleoside diphosphates to 2'-deoxynucleoside diphosphates. It has been proposed that the function of the tyrosyl radical in R2 is to generate a transient thiyl radical (C439.) in R1 over a distance of 35 A, which in turn initiates the reduction process. EPR distance measurements provide a tool with which to study the mechanism of radical initiation in class I RNRs. These types of experiments at low magnetic fields and frequencies (0.3 T, 9 GHz) give insight into interradical distances and populations. We present a pulsed electron-electron double resonance (PELDOR) experiment at high EPR frequency (180-GHz electron Larmor frequency) that detects the dipolar interaction between the Y.s in each protomer of RNR R2 from Escherichia coli. We observe a correlation between the orientation-dependent dipolar interaction and their resolved g-tensors. This information has allowed us to define the relative orientation of two radicals embedded in the active homodimeric protein in solution. This experiment demonstrates that high-field PELDOR spectroscopy is a powerful tool with which to study the assembly of proteins that contain multiple paramagnetic centers.

Orientation of the tyrosyl radical in Salmonella typhimurium class Ib ribonucleotide reductase determined by high field EPR of R2F single crystals

The R2 protein of class I ribonucleotide reductase (RNR) generates and stores a tyrosyl radical, located next to a diferric iron center, which is essential for ribonucleotide reduction and thus DNA synthesis. X-ray structures of class Ia and Ib proteins from various organisms served as bases for detailed mechanistic suggestions. The active site tyrosine in R2F of class Ib RNR of Salmonella typhimurium is located at larger distance to the diiron site, and shows a different side chain orientation, as compared with the tyrosine in R2 of class Ia RNR from Escherichia coli. No structural information has been available for the active tyrosyl radical in R2F. Here we report on high field EPR experiments of single crystals of R2F from S. typhimurium, containing the radical Tyr-105*. Full rotational pattern of the spectra were recorded, and the orientation of the g-tensor axes were determined, which directly reflect the orientation of the radical Tyr-105* in the crystal frame. Comparison with the orientation of the reduced tyrosine Tyr-105-OH from the x-ray structure reveals a rotation of the tyrosyl side chain, which reduces the distance between the tyrosyl radical and the nearest iron ligands toward similar values as observed earlier for Tyr-122* in E. coli R2. Presence of the substrate binding subunit R1E did not change the EPR spectra of Tyr-105*, indicating that binding of R2E alone induces no structural change of the diiron site. The present study demonstrates that structural and functional information about active radical states can be obtained by combining x-ray and high-field-EPR crystallography.