Iron-sulfur clusters (ISC) were present in the most ancient organisms and have adapted to aerobic life. They constitute one of the most widespread and important class of proteins, being involved in numerous essential biological processes: electron transfer in respiratory chain, iron and superoxide sensing, and a large panel of enzymatic functions such as hydrolysis of substrates, bond formations (DNA synthesis, RNA modification, biotin synthesis, ...) to mention only a few ^[7-8]. They are constituted by assemblies of iron and sulfide ions anchored to protein cysteinate residues ^[9]. Although various nuclearities have been found, the [2Fe-2S] and [4Fe-4S] clusters are the most commonly encountered. They can exist in several oxidation states but again the [2Fe-2S]^+/2+ and [4Fe-4S]^+/2+ are the most common. In a simple (maybe simplistic) view the [2Fe-2S]⁺ cluster can be described as a pair of strongly antiferromagnetically coupled Fe²⁺–Fe³⁺ ions bridged by two sulfide S^2– ions, that possesses an overall spin S = 1/2 and is EPR active. Its oxidized form [2Fe-2S]²⁺ involves a strongly antiferromagnetically coupled pair of ferric ions Fe³⁺–Fe³⁺ with a resulting overall spin S = 0. It is EPR silent as the cluster [4Fe-4S]²⁺. Indeed in the latter, the four iron ions assembly can be viewed as an antiferromagnetically coupled dimer of the above Fe²⁺–Fe³⁺ pairs, thus leading to an overall spin S = 0. One-electron reduction to the [4Fe-4S]⁺ state that is composed of one ferric and three ferrous ions restores an overall spin S = 1/2 and the associated EPR spectrum. As a consequence of these diverse electronic states, every one of these clusters exhibits a distinct Mössbauer signature that is used to identify protein active sites and to monitor their functional changes ^[9].


Biogenesis of iron-sulfur proteins and dysfunction

Figure 1: Time course of the formation of Fe-S clusters on IscU in the absence (A) and in the presence (B) of bacterial frataxin (CyaY). Mössbauer spectra were recorded at 4.2 K. Samples were frozen immediately after reaction initiation (black marks) or after an incubation time of 5 (grey), 15 (orange) or 30 min (red).

Iron-sulfur clusters (ISC) can be assembled in vitro from iron salts and sulfides ^[9], but in vivo their assembly requires a complex protein machinery. Several such machineries have been described over the past ten years but their functioning is not fully understood yet ^[10-11]. Of special interest is the Isc machinery that is conserved from bacteria to human. The Isc machinery involves a cysteine desulfurase (IscS or Nfs1 in eukaryotes) to generate sulfide anions and a scaffold protein (IscU or Isu in eukaryotes) where the cluster is assembled. The two proteins and an auxiliary protein (Isd11) form a complex with a particular protein, frataxin (FXN), the role of which is still hotly debated. Addition of iron ions, electrons provided by a ferredoxin, and cysteine to initiate sulfide production, to the complex induces the formation of [2Fe-2S]²⁺ and [4Fe-4S]²⁺ clusters within the scaffold protein. The strong interest of the community for frataxin comes from the fact that a defect in its gene has been identified as the cause for the neurodegenerative disease known as Friedreich ataxia ^[12-13]. It has now been evidenced that FXN deficiency results in the inability of assembling ISC, and later in the accumulation of Fe aggregates.
From biochemical and optical monitoring of ISC synthesis in a bacterial system (IscU) it was concluded that frataxin negatively regulates ISC synthesis, whereas an inverse effect was evidenced in a eukaryotic system ^[14]. In collaboration with A. Pastore (National Institute for Medical Research, London, United-Kingdom), we used Mössbauer spectroscopy to gain more molecular information on frataxin negative role in prokaryotes by following the respective formations of [2Fe-2S]²⁺ and [4Fe-4S]²⁺ clusters in IscU. Figure 1 illustrates the time course of ISC formation over half an hour in the absence (A) and in the presence (B) of bacterial frataxin CyaY. It confirms the earlier observation on frataxin negative role in bacterial system, and reveals that this effect does not change the ratio of the individual clusters [2Fe-2S]²⁺ and [4Fe-4S]²⁺ nor the reductive transformation of the former into the latter ^[15].

	Figure 2: Mössbauer spectra recorded at 4.2 K of mutant mitochondria: A. Δyfh1, B. Δssq1, C. Δggc1, D. Δggc1-YFH1 and E. Δggc1-FTN.

The possibility that FXN is an iron storage protein was advanced in a number of papers and therefore it was of interest to investigate the nature of the Fe aggregates formed in various mutant strains impaired in their ability to synthesize or transfer ISC (collaboration with E. Lesuisse, Institut Jacques Monod, Paris, France). Traces A-C in Figure 2 are the Mössbauer spectra recorded at 4.2 K on a mutant yeast mitochondria depleted in frataxin (Yfh1) or in proteins involved in ISC transfer (Ssq1, Ggc1). The signal is constituted of one slightly asymmetric quadrupole doublet with δ = 0.52(1) mm.s^–1, ΔEQ = 0.63(2) mm.s^–1 and = 0.52/0.50(2) mm.s^–1, typical of a high-spin ferric iron bound to oxygen/nitrogen in an octahedral arrangement. Further experiments with applied magnetic fields confirmed the nature of the aggregate formed in all mutants as polydisperse nanoparticles of iron phosphate ^[16]. Interestingly these aggregates differ strongly from those formed in the well-known iron storage protein ferritin (FTN) ^[17]. Moreover the observation that there is no difference of Fe aggregates in the two strains Δggc1 and Δggc1-YFH1 that overexpresses frataxin (Figure 2C and D) is not consistent with an Fe storage function ^[18]. This conclusion is further supported by the fact that the strain Δggc1-FTN that overexpresses ferritin exhibits a Mössbauer spectrum mostly identical to those of the other strains, the only difference being the presence of a signal (arrow in Figure 2E) assigned to the high-energy line of a quadrupole doublet with δ ≈ 1.23 mm.s^–1 and ΔEQ ≈ 2.98 mm.s^–1, of a high spin Fe^II signal associated to the labile Fe pool ^[19].


Iron-sulfur enzymes

	Figure 3: Mössbauer spectra of as-isolated NadA recorded at 4.2 K in the absence (left) and in the presence of the inhibitor DTHPA (right).

Two main classes of Fe-S enzymes have been described: hydrolytic enzymes (the paradigm for this class being aconitase) and enzymes involved in radical processes, among them the vast family of the so-called "Radical SAM" enzymes using the S-AdenosylMethionine cofactor ^[8]. Most enzymes of these classes have in common an active center constituted by a [4Fe-4S]²⁺ center bound to the protein chain by three cysteine residues thus leaving a single Fe ion available for substrate interaction. Mössbauer spectroscopy has proven very efficient to detect this particular Fe ion and monitor its coordination.
Our very recent study of quinolinate synthase (NadA), a non radical-SAM enzyme, has provided a nice illustration of the potential of Mössbauer spectroscopy in this respect (collaboration with S. Ollagnier de Choudens, LCBM, Grenoble, France) ^[20]. NadA is a key enzyme in the synthesis of Nicotinamide Adenine Dinucleotide, an essential cofactor of many biological redox reactions. The Mössbauer spectrum of the enzyme is depicted in Figure 3. Biological studies have shown that the enzyme is strongly inhibited by 4,5-dithiohydroxyphthalic acid (DTHPA), but the molecular basis for this inhibition was unknown. The Mössbauer spectrum of the enzyme in the presence of DTHPA immediately shows that the inhibitor interacts with the [4Fe-4S]²⁺ cluster through one single Fe ion. Indeed, in the absence of DTHPA (left part of Figure 3), the spectrum recorded is characteristic of classical [4Fe-4S]²⁺ clusters and can be reproduced with the superposition of two slightly different quadrupole doublets (component 1: δ = 0.44 mm.s^–1, ΔEQ= 1.25 mm.s^–1; component 2: δ = 0.45 mm.s^–1, ΔEQ = 0.95 mm.s^–1) of equal intensities associated with the two delocalized antiferromagnetically coupled Fe²⁺–Fe³⁺ pairs. Binding of a ligand to one ion of a pair induces a localization of the valences within this pair which can eventually leads to the pair (component 2, left part of Figure 3) appearing as two distinct quadrupole doublets (components 2 and 3, right part of Fig. 3): component 1 (δ = 0.48 mm.s^–1, ΔEQ = 1.20 mm.s^–1), component 2 (δ = 0.44 mm.s^–1, ΔEQ = 1.04 mm.s^–1) and component 3 (δ = 0.63 mm.s^–1, ΔEQ = 2.02 mm.s^–1). Comparison of the Mössbauer parameters with those of a model complex and to DFT calculated values suggests that the inhibitor chelates one unique Fe ion of the cluster.

Figure 4: Mössbauer spectra of reconstituted RimO recorded at 4.2 K showing the contribution of the "classical cluster" (blue) and the differentiated cluster (red).

A number of radical-SAM enzymes have been discovered recently as key players in the modification of tRNA and ribosomal proteins in processes aimed at improving the accuracy of the genetic machinery. These enzymes commonly associate two [4Fe4S]²⁺ clusters, one interacting with the SAM cofactor to initiate the generation of the adenosyl radical while the other interacts with the substrate ^[8]. We recently characterized by Mössbauer spectroscopy one such enzyme RimO that is able to perform the thiomethylation of an aspartate residue of a ribosomal protein S12 (collaboration with M. Atta, LCBM, Grenoble, France) ^[21]. Interestingly the Mössbauer spectrum of the protein depicted in Figure 4 reveals the inequivalence of the two [4Fe-4S]²⁺ clusters with the presence in one of them of a unique Fe site as found for NadA (see above). The first cluster can be taken into account with a single quadrupole doublet with usual parameters averaging the two pair components (δ = 0.45 mm.s^–1, ΔEQ = 1.04 mm.s^–1). By contrast, the second cluster requires three components to be simulated as was the case for NadA in the presence of the inhibitor: component 1 (δ = 0.48 mm.s^–1, ΔEQ = 1.15 mm.s^–1), component 2 (δ = 0.3 mm.s^–1, ΔEQ = 0.9 mm.s^–1) and component 3 (δ = 0.60 mm.s^–1, ΔEQ = 2.07 mm.s^–1). This observation raises the question of the nature of the peculiar ligand of the unique Fe ion of this cluster. As previously, the distinction of the coordination of this Fe ion constitutes a powerful tool to investigate the reactivity of the enzyme and to address the question of its mechanism.

For the past three decades understanding how molecular oxygen is activated and transferred to inactive substrates and duplicating this reaction has constituted one of the main challenges posed to chemists with implications ranging from basic knowledge to far-reaching industrial applications. Oxygen atom transfer is the reaction catalyzed by enzymes called oxygenases, the paradigm of which are the cytochromes P450, a class of heme-thiolate proteins. After more than thirty years of intense research, the active form of cytochromes P450 has been demonstrated recently to be an Fe^IV=O complex of a porphyrin radical cation ^[22]. In the nonheme world, the most fascinating enzyme is undoubtedly methane monooxygenase that uses an oxo-bridged diiron^IV unit to oxidize methane into methanol. Trying to elaborate catalysts that can rival these enzymes in their oxidation power and selectivity is thus a very active research domain.

High-valent iron catalysts
The first non-heme Fe^IV=O complexes were characterized almost a decade ago in systems that were deactivated by the use of aromatic amine ligands ^[23]. Mössbauer spectroscopy revealed that these systems were characterized by an S = 1 spin state whereas all corresponding protein intermediates harbored a spin S = 2 state and a higher reactivity. Improving the reactivity of these model complexes and understanding the structure-reactivity relationships has become a major endeavor and Mössbauer spectroscopy has proved to be a tool of choice to characterize the electronic structures of trapped intermediates.
The influence of steric hindrance/ distortion on the reactivity of the Fe^IV=O complex of tetranitrogen macrocylic ligands has been investigated by replacing four methyl groups 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetradecane ligand (TMC) by four benzyl groups (TBC) (collaboration with W. Nam, Ewha Womans University, Seoul, South Korea; and E. I. Solomon, Stanford University, USA) ^[24]. Figure 5 illustrates the Mössbauer spectra of the complex Fe^IV=O(TBC) recorded in different temperature (from 4.2 to 78 K) and applied field (from 60 mT to 7 T) conditions to investigate its electronic structure. Simultaneous fitting of all spectra allowed to conclude that the complex possesses a spin S = 1.



Figure 5. Left: DFT calculated structure of Fe^IV=O(TBC). Right: Mössbauer spectra of Fe^IV=O(TBC) (vertical bars) measured at (A-C) 4.2 K, (D) 41 K and (E) 78 K in a magnetic field of (A) 60 mT, (B) 4 T and (C-E) 7 T. The solid (green) curve is a Spin Hamiltonian simulation with the following set of parameters: δ = 0.22 mm.s–1, ΔE_Q = 0.97 mm.s^–1, D = 29.5 cm–1, E/D = 0, g_x,y,z = 2.3, 2.3, 2.0 and A_x,y,z/g_Nß_N = –18, –18, –2 T. Reprinted by permission from American Chemical Society: Journal of the American Chemical Society, 2012, 134: 11791, copyright (2012).

In a similar study, it was shown that the reaction of the Fe^II complex of the ligand Me₃NTB (Me₃NTB = tris((N-methylbenzimidazol-2-yl)methyl)amine) with m-chloroperbenzoic acid, a classical oxygen donor, produces the corresponding Fe^IV=O species, that has been characterized by Mössbauer spectroscopy as an intermediate spin system S = 1 (collaboration with W. Nam, Seoul, South Korea). Reactivity studies showed that this species is one of the most active catalysts reported so far for oxygen transfer reactions to unactivated aliphatic substrates ^[25].

Diiron systems

The dioxygenase MiaE
This enzyme belongs to a group of enzymes involved in improving the efficiency and fidelity in genome decoding through tRNA modifications. It catalyzes the posttranscriptional allylic hydroxylation of 2-methylthio-N-6-isopentenyl adenosine, a difficult oxygenation reaction of a methyl group (Figure 6A). Mössbauer spectroscopy proved invaluable to show that the enzyme active site comprises a diiron site similar to that of Methane MonoOxygenase (collaboration with M. Atta, LCBM, Grenoble, France) ^[26]. Traces a and b on Figure 6B reproduce Mössbauer spectra of the as-isolated enzyme. They revealed the presence of at least three components that were identified thanks to a temperature and field-dependent study. The major component (54 %) which contributes the central quadrupole doublet is a diferric species that experiences a moderate antiferromagnetic coupling, revealed by the broadening of the doublet at 77 K (trace d on Figure 6B), as found in µ-hydroxodiferric entities. The shoulders that flank both sides of the central doublets are assigned to a µ-oxodiferric species from its Mössbauer parameters and its strong antiferromagnetically coupling contributing 16 % of the total Fe content. The remaining 30 % of the iron correspond to a paramagnetic species that contributes at higher velocities (trace c on Figure 6B) and behaves as a mixed-valent Fe^II-Fe^III pair. Its presence was confirmed by EPR spectroscopy.



Figure 6. A: Reaction catalyzed by MiaE. B: Mössbauer spectra of MiaE2H (vertical bars) measured at 4.2 K in a magnetic field of 60 mT applied parallel to the -beam (a) or 22 mT applied perpendicular to the γ-beam (b) or at 77 K and zero applied field (d). Spectrum c is obtained by subtraction (a-b). The solid blue lines are spin-Hamiltonian simulations. Contributions from the oxodiferric and mixed valence clusters are shown in pink and green, respectively.

Mixed-valent Fe^IIFe^III model complexes
Diiron complexes have been used for decades to try and model the structure and the reactivity of various enzymes with a diiron site. They are invaluable to investigate the potential cooperativity between the two sites involved in hydrolytic or redox reactions. The complex illustrated in Figure 7A is a model of the mixed valence state of hemerythrin that reproduces its pH behavior. It comprises a ferric (left) and a ferrous ion (right) bridged by a phenolate and two carboxylates. Their terminal ligation differs: the ferric site is bound to a bis(picolyl)amine group whereas an aniline ligand replaces a pyridine on the ferrous ion. The Mössbauer spectrum of Figure 7Ba shows the two quadrupole doublets of the two ions. Interestingly, when the complex is treated by triethylamine its spectroscopic properties change drastically as shown in Figure 7Bb. The main change that occurs in the Mössbauer spectrum concerns the ferric site whose quadrupole splitting parameter increases dramatically from ΔEQ = 0.39 mm.s^–1 to ΔEQ = 1.77 mm.s^–1. This change reveals that the ferric ion is bound to an anionic ligand in trans position with respect to the bridging phenolate. This in turn indicates that the aniline ligand has been deprotonated and that an internal electron transfer has occurred within the diiron pair ^[27]. In other words, the deprotonation of the aniline has induced a valence interchange, as illustrated in Figure 7C. Further studies based on electrochemical techniques have shown that this proton-coupled intervalence transfer is a concerted process ^[28]. Apart from its bioinorganic interest this compound can be viewed as a proton induced redox switch.



Figure 7. A: X-ray structure of the mixed valence aniline complex where the Fe^III and Fe^II ions are shown in pink and green, respectively. The two hydrogen atoms of the coordinated aniline group are shown in light pink. B: Experimental (hatched marks) and simulated (solid lines) Mössbauer spectra of an acetonitrile solution of the aniline complex before (a) and after (b) the addition of 1.5 eq of NEt₃. The ferric and ferrous contributions are indicated as pink and green traces, respectively. C: Intervalence exchange induced by aniline (de)protonation. Reprinted by permission from American Chemical Society: Inorganic Chemistry, 2011, 50: 6408, copyright (2011).

High nuclearity Fe complexes are interesting in many respects among which understanding the interaction pathways occurring in small clusters that constitute basic elements of larger aggregates is of current interest for molecular and nanomaterials. The X-ray structure of the complex [Fe₄(µ₄-O) (µ-OMe)₄(bisi)₄](ClO₄)₂•4MeOH (Hbisi = N-(benzimidazol-2-yl)salicylaldimine) is depicted in Figure 8A. It reveals an original arrangement of four Fe^III ions in a slightly ruffled square where they are bridged by four µ2-methoxido anions and a central µ4-oxido anion that gives rise to two magnetic interaction pathways, J₁ and J₂, respectively (inset in Figure 8B). The temperature dependence of the magnetic susceptibility was successfully simulated within this coupling scheme with the following values of the exchange constants: J₁ = –1.4 cm^–1, and J₂ = –19.2 cm^–1 (Figure 8B). Figure 8C illustrates the Mössbauer spectra recorded at various temperature and applied magnetic fields. They could be simultaneously simulated in the fast relaxation mode assuming a unique S = 0 system (traces a and b) or a symmetric dinuclear system with 5/2 local spins (traces a, b and c) with the following parameters δ = 0.515 mm.s^–1, ΔEQ = 1.011 mm.s^–1, π = 0.33, Γ_fwhm = 0.28 mm.s^–1, J₂ = –18.4 cm^–1 and a_iso = –20.12 T. As a consequence, at low temperatures, the tetranuclear cluster can be described as the sum of two identical and symmetric dinuclear high-spin Fe^III units that are moderately antiferromagnetically coupled through the µ4-oxido bridge. In addition, each Fe^III ion of a pair interacts weakly with the two sites of the second pair through the µ₄-oxido-µ-methoxido bridging pattern (collaboration with J. Reedijk, Leiden University, The Netherlands) ^[29].


Figure 8. A. Structural, magnetic (B) and Mössbauer (C) properties of [Fe₄(µ₄-O)(µ-OMe)₄(bisi)₄](ClO₄)₂•4MeOH: X-ray structure (A), temperature dependence of the magnetic susceptibility (B) and Mössbauer spectra (C) recorded at zero field and 4.2 K (a), 7 T and 4.2 K (b) and 7 T and 50 K (c). Reprinted by permission from American Chemical Society: Inorganic Chemistry, 2010, 49: 2427, copyright (2010).




Oxygen atom transfer from molecular oxygen and in ambient conditions to a poorly reactive substrate is still a chemical challenge in spite of several decades of intense research. This reaction is catalyzed by monooxygenases whose paradigm is constituted by cytochromes P450. Their action mechanism has been very recently characterized definitely and shown to involve a high oxidation state Fe species associated to an oxidized heme (porphyrin radical cation). Similar high valent iron species are implied also in non heme monooxygenases. In particular, the active form of methane monooxygenase, non heme enzyme that transforms methane into methanol, involves a pair of Fe^IV ions. Detection and characterization of such species in enzymes and oxidation catalysts bears thus a strong interest and gives rise to a hot competition. The unavoidable method to characterize these species is Mössbauer spectroscopy owing to their specific spectroscopic signatures.

Figure 6 : A. Réaction de formation de l'espèce active 3 dans la catalyse d'oxydation du méthane à partir de la bis-porphyrine de fer 1. B. Spectres Mössbauer de l'espèce active enregistrés à 4,2 K avec les champs magnétiques indiqués sur la figure.

In collaboration with the group of Dr A. Sorokin (IRCE Lyon) we have studied species 3 (Figure 6A) that is able to oxidize methane into methanol, what is extremely rare for molecular compounds. Species 3 is generated by reaction of the Fe bis-porphyrin (1 in figure 6A) with a peracid (m-chloroperbenzoic acid, m-CPBA) in an organic solvent at very low temperature (- 90°C). We have established that 3 is an oxo bis-Fe^IV species with a porphyrin oxidized in radical cation. This conclusion derives from the Mössbauer studied realized with an applied magnetic field (Figure 6B) which revealed that the bis-FeIV entity has a spin S = 0 and interacted only weakly with the radical present in one of the two porphyrins. ^[6] This electronic structure is reminiscent of that of cytochrome P450 active species that associated a Fe^IV oxo center with an oxidized porphyrin. It differs from it by the presence of an additional Fe^IV which further increases the oxidizing ability allowing methane oxidation, a reaction that cytochromes P450 can not achieve. Similar studies have been realized in collaboration with the korean group of Pr. W. Nam (Seoul) on non heme systems of similar reactivity. <sup>[7, 8]


[6]</sup> Kudrik et al., Nature Chem., 2012, 4(12): 1024
^[7] Seo et al., Chem. Sci., 2011, 2: 1039
^[8] Wilson et al., J. Am. Chem. Soc., 2012, 134: 11791




Many metalloproteins possess multimetallic active sites. In case of iron, complicate arrangements are observed and understanding their behavior require studying structurally similar synthetic compounds whose properties are perfectly established. Along these lines, we study bi- or tetranuclear complexes to investigate their magnetic and spectroscopic properties. We have thus studied the magnetic behavior and spectroscopic properties of a binuclear iron-manganese complex illustrated on figure 7A that reproduces the active sites of recently discovered enzymes that react with molecular oxygen. To account for the properties of this compound in the oxidation state Fe^IIMn^II, which is active in EPR, the Mössbauer program developed with the support of the local group of computer programmers (GIPSE at the institute) has been extended to include the simultaneous simulation of Mössbauer and EPR spectra (Figure 7C) recorded at 4.2 K in several applied magnetic fields. The developed methodology and the obtained results will provide benchmarks for proteins studies. ^[9] Related work was done on a tetranuclear Fe complex in collaboration with the group of Pr. J. Reedijk (Leiden, The Netherlands) ^[10]



Figure 7. A. Structure cristalline du complexe binucléaire Fe^IIIMn^II. B. Spectre RPE du complexe Fe^IIMn^II enregistré à 4,2 K en bande X. C. Spectres Mössbauer du complexe Fe^IIMn^II enregistré à 1,4 K (spectres a, c, e) et 4,2 K (spectres b, d, f) avec les champs magnétiques indiqués sur la figure.

^[9] Carboni et al., Inorg. Chem., 2012, 51: 10447
^[10] Murali et al., Inorg. Chem., 2010, 49; 2427



Of course this short overview has not covered all of the activities, past or present, of the platform, and starting from this basis future work will engage deeper in reactivity studies of high-valent intermediates as well as the consideration of more complicated biological systems.
This brief survey has also illustrated how the transmission of expertise from physics to chemistry and biology has been successfully achieved in CEA-Grenoble, since it has allowed Mössbauer activity in Grenoble not only to survive but, after inputs of new people and investments, to progress steadily and engage in worldwide collaborations. Such inputs and investments were necessary to expand on the foundations led by Jean-Louis Oddou who transmitted his knowledge until he retired in 2010.



^[01] Nagy DL. Mössbauer effect: A dual method for myriad applications. Hyperfine Interactions, 2008, 182(1-3): 5-13

^[02] Chappert J. Cryostat à température variable pour effet Mössbauer. Journal de Physique, 1965, 26 A: 183

^[03] Chappert J, Teillet J and Varret F. Recent developments in high field Mössbauer spectroscopy. Journal of Magnetism and Magnetic Materials, 1979, 11: 200

^[04] Blaise A, Chappert JL and Girardet JL. Observation par mesures magnétiques et effet Mössbauer d'un antiferromagnétisme de grains fins dans la ferritine. Comptes Rendus Hebdomadaires des Séances de l'Académie des Sciences, 1965, 261: 2310-2313

^[05] Buisson G, Deronzier A, Duee E, Gans P, Marchon JC and Regnard JR. Iron(III)-porphyrin π-cation radical complexes. Molecular structures and magnetic properties. Journal of the American Chemical Society, 1982, 104(24): 6793–6796

^[06] Moulis JM, Auric P, Gaillard J and Meyer J. Unusual features in EPR and Mössbauer spectra of the 2[4Fe-4Se]⁺ ferredoxin from Clostridium pasteurianum. Journal of Biological Chemistry, 1984, 259(18): 11396-11402

^[07] Fontecave M. Iron-sulfur clusters: Ever-expanding roles. Nature Chemical Biology, 2006, 2: 171-174

^[08] Marsh EN, Patterson DP and Li L. Adenosyl radical: Reagent and catalyst in enzyme reactions. Chembiochem, 2010, 11(5): 604-621

^[09] Beinert H, Holm RH and Münck E. Iron-Sulfur clusters: Nature's modular, multipurpose structures. Science, 1997, 277(5326): 653-659

^[10] Lill R and Mühlenhoff U. Maturation of iron-sulfur proteins in eukaryotes: Mechanisms, connected processes, and diseases. Annual Review of Biochemistry, 2008, 77: 669-700

^[11] Py B and Barras F. Building Fe-S proteins: Bacterial strategies. Nature Reviews. Microbiology, 2010, 8(6):436-446

^[12] Schmucker S and Puccio H. Understanding the molecular mechanisms of Friedreich's ataxia to develop therapeutic approaches. Human Molecular Genetics, 2010, 19(R1): R103-R110

^[13] Stemmler TL, Lesuisse E, Pain D and Dancis A. Frataxin and mitochondrial FeS cluster biogenesis. Journal of Biological Chemistry, 2010, 285(35): 26737-26743

^[14] Adinolfi S, Iannuzzi C, Prischi F, Pastore C, Iametti S, Martin SR, Bonomi F and Pastore A. Bacterial frataxin CyaY is the gatekeeper of iron-sulfur cluster formation catalyzed by IscS. Nature Structural & Molecular Biology, 2009, 16(4): 390-6

^[15] Iannuzzi C, Adinolfi S, Howes BD, Garcia-Serres R, Clemancey M, Latour JM, Smulevich G and Pastore A. The role of CyaY in iron sulfur cluster assembly on the E. coli IscU scaffold protein. PLoS One, 2011, 6(7): e21992

^[16] Lesuisse E, Santos R, Matzanke BF, Knight SA, Camadro JM and Dancis A. Iron use for haeme synthesis is under control of the yeast frataxin homologue (Yfh1). Human Molecular Genetics, 2003, 12(8):879-889

^[17] Papaefthymiou GC. The Mössbauer and magnetic properties of ferritin cores. Biochimica and Biophysica Acta, 2010, 1800(8): 886-897

^[18] Seguin A, Sutak R, Anne Laure B, Garcia-Serres R, Oddou J-L, Lefevre S, Santos R, Dancis A, Camadro J-M, Latour J-M and Lesuisse E. Evidence that yeast frataxin is not an iron storage protein in vivo. Biochimica et Biophysica Acta (BBA) - Molecular Basis of Disease, 2010, 1802(6): 531-538

^[19] Sutak R, Seguin A, Garcia-Serres R, Oddou JL, Dancis A, Tachezy J, Latour JM, Camadro JM and Lesuisse E. Human mitochondrial ferritin improves respiratory function in yeast mutants deficient in iron-sulfur cluster biogenesis, but is not a functional homologue of yeast frataxin. MicrobiologyOpen, 2012, 1(2):95-104

^[20] Chan A, Clémancey M, Mouesca JM, Amara P, Hamelin O, Latour JM and Ollagnier de Choudens S. Studies of inhibitor binding to the [4Fe-4S] cluster of quinolinate synthase. Angewandte Chemie International Edition, 2012, 51(31): 7711-7714

^[21] Arragain S, Garcia-Serres R, Blondin G, Douki T, Clemancey M, Latour JM, Forouhar F, Neely H, Montelione GT, Hunt JF, Mulliez E, Fontecave M and Atta M. Post-translational modification of ribosomal proteins: Structural and functional characterization of RimO from Thermotoga maritima, a radical-SAM methylthiotransferase. Journal of Biological Chemistry, 2010, 285(8): 5792-5801

^[22] Rittle J and Green MT. Cytochrome P450 compound I: capture, characterization, and C-H bond activation kinetics. Science, 2010, 330(6006): 933-937

^[23] Rohde JU, In JH, Lim MH, Brennessel WW, Bukowski MR, Stubna A, Münck E, Nam W and Que L Jr. Crystallographic and spectroscopic characterization of a nonheme Fe(IV)-O complex. Science, 2003, 299(5609): 1037-1039

^[24] Wilson SA, Chen J, Hong S, Lee YM, Clemancey M, Garcia-Serres R, Nomura T, Ogura T, Latour JM, Hedman B, Hodgson KO, Nam W and Solomon EI. [Fe^IV=O(TBC)(CH₃CN)]²⁺: Comparative reactivity of iron(IV)-oxo species with constrained equatorial cyclam ligation. Journal of the American Chemical Society, 2012, 134(28): 11791-11806

^[25] Seo MS, Kim NH, Cho KB, So JE, Park SK, Clemancey M, Garcia-Serres R, Latour JM, Shaik S and Nam W. A mononuclear nonheme iron(IV)-oxo complex which is more reactive than cytochrome P450 model compound I. Chemical Science, 2011, 2(6): 1039-1045

^[26] Mathevon C, Pierrel F, Oddou JL, Garcia-Serres R, Blondin G, Latour JM, Menage S, Gambarelli S, Fontecave M and Atta M.tRNA-modifying MiaE protein from Salmonella typhimurium is a nonheme diiron monooxygenase. Proceedings of the National Academy of Sciences USA, 2007, 104(33): 13295-13300

^[27] Goure E, Thiabaud G, Carboni M, Gon N, Dubourdeaux P, Garcia-Serres R, Clemancey M, Oddou JL, Robin AY, Jacquamet L, Dubois L, Blondin G and Latour JM. Reversible (De)protonation-induced valence inversion in mixed-valent diiron(II,III) complexes. Inorganic Chemistry, 2011, 50(14): 6408-6410

^[28] Balasubramanian R, Blondin G, Canales JC, Costentin C, Latour JM, Robert M and Saveant JM. Proton-coupled intervalence charge transfer: Concerted processes. Journal of the American Chemical Society, 2012, 134(4): 1906-1909

^[29] Murali M, Nayak S, Costa JS, Ribas J, Mutikainen I, Turpeinen U, Clemancey M, Garcia-Serres R, Latour JM, Gamez P, Blondin G and Reedijk J. Discrete tetrairon(III) cluster exhibiting a square-planar Fe₄(μ₄-O) core: Structural and magnetic properties. Inorganic Chemistry, 2010, 49(5): 2427-2434

Kudrik EV, Afanasiev P, Alvarez LX, Dubourdeaux P, Clémancey M, Latour JM, Blondin G, Bouchu D, Albrieux F, Nefedov SE and Sorokin AB. An N-bridged high-valent diiron-oxo species on a porphyrin platform that can oxidize methane. Nature Chemistry, 2012, 4(12): 1024-1029

Carboni M, Clemancey M, Molton F, Pecaut J, Lebrun C, Dubois L, Blondin G and Latour JM. Biologically relevant heterodinuclear iron-manganese complexes. Inorganic Chemistry, 2012, 51(19): 10447-10460
Inorganic Chemistry, 2012, 51(21): 12053 [Correction]

Perche-Letuvée P, Kathirvelu V, Berggren G, Clemancey M, Latour JM, Maurel V, Douki T, Armengaud J, Mulliez E, Fontecave M, Garcia-Serres R, Gambarelli S and Atta M. 4-demethylwyosine synthase from Pyrococcus abyssi is a radical-S-adenosyl-L-methionine enzyme with an additional [4Fe-4S]⁺² cluster that interacts with the pyruvate co-substrate. Journal of Biological Chemistry, 2012, 287(49): 41174-41185