Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer_journal/the-nmr-contribution-to-protein-protein-networking-in-fe-s-protein-vcnU0qiKUk
- Publisher
- Springer Journals
- Copyright
- Copyright © 2018 by The Author(s)
- Subject
- Life Sciences; Biochemistry, general; Microbiology
- ISSN
- 0949-8257
- eISSN
- 1432-1327
- DOI
- 10.1007/s00775-018-1552-x
- Publisher site
- See Article on Publisher Site
Abstract
Iron–sulfur proteins were among the first class of metalloproteins that were actively studied using NMR spectroscopy tai - lored to paramagnetic systems. The hyperfine shifts, their temperature dependencies and the relaxation rates of nuclei of cluster-bound residues are an efficient fingerprint of the nature and the oxidation state of the Fe–S cluster. NMR significantly contributed to the analysis of the magnetic coupling patterns and to the understanding of the electronic structure occurring in [2Fe–2S], [3Fe–4S] and [4Fe–4S] clusters bound to proteins. After the first NMR structure of a paramagnetic protein was obtained for the reduced E. halophila HiPIP I, many NMR structures were determined for several Fe–S proteins in different oxidation states. It was found that differences in chemical shifts, in patterns of unobserved residues, in internal mobility and in thermodynamic stability are suitable data to map subtle changes between the two different oxidation states of the protein. Recently, the interaction networks responsible for maturing human mitochondrial and cytosolic Fe–S proteins have been largely characterized by combining solution NMR standard experiments with those tailored to paramagnetic systems. We show here the contribution of solution NMR in providing a detailed molecular view of “Fe–S interactomics”. This contribu- tion was particularly effective when protein–protein interactions are weak and transient, and thus difficult to be characterized at high resolution with other methodologies. Keywords NMR spectroscopy · Hyperfine interactions · Fe–S proteins · Interactomics · CIA machinery · ISC machinery Introduction used as source of information on protein oxidation states and on the number and nature of heme ligands [2, 3]. Soon The NMR spectrum of cytochrome c, collected by Kowal- after that, NMR spectroscopy was applied on other para- sky in 1965, was the first high-resolution NMR spectrum magnetic proteins such as single iron rubredoxins [4] and on of a paramagnetic protein published ever [1]. The hyperfine Fe–S cluster containing proteins [5, 6]. In combination with shifts induced onto the methyl resonances of heme by the EPR, Mössbauer and magnetic susceptibility measurements, 3+ 1 paramagnetic Fe ion were large enough to circumvent res- H NMR spectroscopy significantly contributed, since the olution problems and permitted, for the r fi st time, the identi - early days of research on Fe–S proteins, to elucidate the fication of “individual” proton resonances, which have been electronic structure and magnetic coupling among the iron ions in Fe–S clusters [7–11]. Small electron transfer proteins such as rubredoxins, ferredoxins and HiPIPs are paradig- The original version of this article was revised due to a matic examples of how solution NMR can easily identify dif- retrospective Open Access order. ferent types of Fe–S clusters and different oxidation states. Indeed, the number of iron ions, their oxidation states and * Lucia Banci the magnetic couplings among them determine NMR spectra banci@cerm.unifi.it that differ one another in terms of signal linewidths, chemi - * Mario Piccioli cal shifts and number of observed signals. piccioli@cerm.unifi.it Magnetic Resonance Center CERM, University of Florence, Via Luigi Sacconi 6, Sesto Fiorentino, 50019 Florence, Italy Department of Chemistry, University of Florence, Via della Lastruccia 3, Sesto Fiorentino, 50019 Florence, Italy Vol.:(0123456789) 1 3 666 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 2+ [2Fe–2S] clusters is only due to coupling of the nuclear Elucidation of electronic structure of Fe–S spin with the electron spin excited levels, which can be clusters in proteins significantly populated at room temperature. These pro- teins feature a broad signal, usually unresolved, in the From the NMR spectroscopy point of view, there is a vari- 40–30 ppm range and individual βCH protons from Cys ety of possible behaviors and patterns. The least favora- bound to the two ions cannot be identified nor sequence ble situation is that occurring in oxidized single iron ion specifically assigned (Fig. 1c) [15, 23]. such as in rubredoxins (Fig. 1a). The first H NMR spec- Upon reduction, the added electron can be localized on a trum was reported by Moura and coworkers. Here an iso- 3+ single iron ion, and therefore, the iron pair is described as a lated, high spin Fe ion (S = 5/2) gives a contribution to 3+ 2+ Fe –Fe pair, or the extra electron can be delocalized over transverse nuclear relaxation rates of the βCH protons the cluster [19, 24]. For localized valence cases, as observed of iron-bound cysteines as large as 80 kHz [12]. Albeit in plant-type ferredoxins, when antiferromagnetic coupling their very fast nuclear relaxation rates, which determine occurs, the electron spin relaxation rates of both iron ions large linewidths, Cys βCH signals are observable, thanks increase, and therefore, the coupled nuclear spins relax to their large hyperfine shifts, being well outside the dia- 3+ slower. As a consequence, NMR signals become sharper magnetic envelope. The reduction of Fe ion to high spin, 2+ 2+ than those in the oxidized [2Fe–2S] form, in particular for S = 2, Fe , gives a significant decrease in the observed 2+ βCH bound to the purely Fe ion (Fig. 1d). According to linewidths and an increase in the chemical shifts of the the theoretical model developed for two magnetically cou- βCH resonances (Fig. 1b), which are, therefore, oxida- pled metal ions [15, 20], the isotropic shifts decrease with tion state-dependent spectral parameters [13, 14]. In this increasing temperature (Curie-type behavior) for cysteines way, NMR can be exploited to obtain information on iron 3+ bound to the F e ion and increase with increasing tem- oxidation states in rubredoxins. perature (anti Curie-type behavior) for cysteines bound to In the case of Fe–S clusters, the magnetic coupling 2+ the Fe ion. Therefore, the fitting of the experimental tem- between the iron ions determines various electron spin perature dependence can provide a direct measure of the energy levels whose separation depends on the magnetic magnetic exchange coupling constant J [15]. The sequence- coupling constants [7, 15–18]. The coupling of the nuclear specific assignment of Cys residues bound to the cluster is spins with these multiple electron spin levels significantly the crucial step for the identification of the oxidation states affect both the chemical shifts and the relaxation rates [15, of individual iron ions. Pioneering NOE experiments ele- 19, 20]. As a consequence of this coupling, NMR signals gantly showed that, in [2Fe–2S] ferredoxins from plants are sharper than those observed for isolated iron ions, and and algae, the more reducible iron ion of the pair is that the contact shifts experienced by cluster-bound residues closer to the protein surface [25]. The Cys ligands bound to are usually smaller than those observed in rubredoxins. the more reducible ion form a larger number of hydrogen Consequently, the NMR spectra are dramatically differ - bonds than those bound to the other iron ion, in agreement ent when changing oxidation state or cluster composition. with previous proposals [26]. This valence-localized model The hyperfine shifts, their temperature dependencies and holds also in the case of Rieske ferredoxins [27]. the relaxation rates of nuclei of cluster-bound residues In the case of ion pairs with delocalized valence, as allowed NMR spectroscopists to elucidate the magnetic observed in mammalian ferredoxins, the iron ions have much coupling patterns occurring in [2Fe–2S], [3Fe–4S] and slower electron spin relaxation rates than in the localized [4Fe–4S] clusters. This contributed to the understanding valence pairs [28]. The pattern of chemical shifts can still of the electronic structure of Fe–S clusters in proteins, be described with the model successfully used to account for also thanks to the fact that NMR provides information at the NMR properties of valence localized [2Fe–2S] ferre- room temperature. We would like to recall here that the doxins, but nuclear relaxation is much faster thus determin- Fe–S clusters are usually always characterized by only two ing much broader lines often undetectable for H signals redox states differing by a single electron. Let us briefly (Fig. 1e). Sequence-specific assignment of metal bind- overview the different cases. 13 15 2+ ing residues is only possible via a combination of C, N In [2Fe–2S] clusters, the oxidized [2Fe–2S] state 3+ and H experiments [23]. contains two antiferromagnetically coupled Fe ions that Discovered about 10 years later than the other Fe–S clus- give rise to a S = 0 ground state [21]. Antiferromagnetic ters [29], the [3Fe–4S] clusters are available in two oxidation coupling between two identical ions does not contribute states as well but, at variance with the [2Fe–2S] case, have significantly to reduce the electron spin relaxation times, paramagnetic ground states. The oxidized form, [3Fe–4S] , and therefore, the βCH signals are quite broad [15, 22]. 3+ contains three high spin F e ions. A total ground electron Their contact shifts are small compared to those in rubre- spin S = 1/2 level is observed, arising from slight inequiva- doxins, in agreement with the fact that the contact shift in lence among the three J values and from spin frustration ij 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 667 Fig. 1 1D H NMR spectra of different Fe–S cluster types. 400 MHz 1D H NMR spectra 3+ 2+ of Fe (a) and Fe (b) C. pasteurianum rubredoxin, acquired at 308 K (adapted from [13]); c 200 MHz 1D H NMR 2+ spectrum of [2Fe–2S] P. umbilicalis ferredoxin, acquired at 303 K [15]; d 360 MHz 1D H NMR spectrum of [2Fe–2S] P. umbilicalis ferredoxin, recorded at 303 K [25]; e 400 MHz 1D H NMR spectrum of [2Fe–2S] human ferredoxin, acquired at 303 K [201]; f 500 MHz 1D H NMR spectrum of [3Fe–4S] P. furiosus ferredoxin, recorded at 303 K [33]; 600 MHz 1D H 2+ NMR spectra of [4Fe–4S] , g 3+ [40] and [4Fe–4S] , i [60] E. halophila HIPIP II, recorded at 300 K; h 600 MHz 1D H NMR spectrum of [4Fe–4S] C. acidi urici ferredoxin, acquired at 298 K [43] 1 3 668 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 Fig. 2 Magnetic coupling and electronic distribution in [3Fe–4S] and [4Fe–4S] clusters. Schematic representation of the spin frustration in a [3Fe–4S] cluster (a) and of the coupling scheme in a [4Fe–4S] cluster (b). c–e Electronic distribution 3+ in the [4Fe–4S] clusters of HiPIPs: c the extra electron can be unevenly distributed among the iron ions Fe, Fe and Fe ; d 1 3 4 a chemical equilibrium between two different electronic distri- butions in the cluster, where 2.5+ 2.5+ mixed-valence Fe –Fe iron ion pairs are represented as 3+ grey squares and purely Fe – 3+ Fe pairs are represented as white squares; e illustration of the resonance between two limit 3+ 2+ formulas. Fe and Fe ions are represented as white and black squares, respectively 2+ [30, 31]. As shown in Fig. 2a, when J > J = J, Fe and the situation observed in [2Fe–2S] case; however, nuclear 12 13 23 1 Fe form an antiferromagnetically coupled pair; as a conse- relaxation is much slower and NMR signals are relatively quence, Fe cannot be antiferromagnetically coupled to both sharp and easy to be identified [32]. Indeed, temperature Fe and F e and remains with S = 5/2, while the F e –Fe iron dependence of the βCH protons signals of the iron bound 1 2 2 3 2 2+ pair has a subspin S = 2. Observed hyperfine shifts are in the cysteines is similar to the situation of [2Fe–2S] case, i.e., 40–0 ppm range (Fig. 1f) and therefore, they are similar to signals from cysteines bound to the “frustrated” Fe ion have 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 669 a Curie temperature dependence, while those arising from The relatively sharp linewidths of the signals made possible the Fe –Fe pair (S = 2) have an opposite behavior with tem- their sequence-specific assignment, while their temperature 1 2 perature. Therefore, sequence-specific assignment and iron dependence allowed us to identify the oxidation state (i.e., 2+ 2.5+ identification within the scheme of Fig. 2a was achieved Fe or Fe ) of each individual iron ion. and contributed significantly to the understanding of the Individual oxidation states of iron ions in [4Fe–4S] 3+ magnetic coupling scheme in [3Fe–4S] clusters [33–35]. clusters were identified for the first time in [4Fe–4S] In the reduced [3Fe–4S] state, the extra electron is delo- proteins, which represent the most favorable situation for 3+ calized on a ferromagnetically coupled iron pair [36]. The solution NMR studies. In [4Fe–4S] proteins, the elec- total S = 2 ground state [37] is such that NMR signals from tron spin relaxation times of the paramagnetic centers are 2+ + cluster bound residues are too broad and/or shifted too far shorter than those in the [4Fe–4S] and [4Fe–4S] clus- to be detected. ters, because of larger magnetic couplings among iron ions. In [4Fe–4S] clusters, there are three possible/available Therefore, signals are sharper and easier to be sequence 3+/2+/1+ states, [4Fe–4S] . The magnetic coupling scheme specifically assigned (Fig. 1i). Theoretical models involving increases in complexity, as six J magnetic coupling con- double exchange contributions [18], spin frustration [53] and ij stants are needed to describe the system (Fig. 2b). In the asymmetric model compounds [54] contributed to the under- 2+ 3+ [4Fe–4S] case, the situation is described by two, identi- standing of the electronic structure of [4Fe–4S] clusters in 3+ 2+ cal, valence delocalized, F e –Fe pairs, that are antifer- proteins, which have been described as a pair of two purely 3+ 3+ romagnetically coupled each other [38]. The electron spin ferric (Fe –Fe ) iron ions and a delocalized valence pair 2.5+ 2.5+ energy levels diagram has a diamagnetic S = 0 ground state (formally Fe –Fe ), which are antiferromagnetically 2.5+ and the cluster has four equivalent iron ions, formally Fe . coupled each other giving rise to a S = 1/2 ground state [55, 2+ As in the case of oxidized [2Fe–2S] , paramagnetism arises 56], essentially having a magnetic coupling scheme analo- from excited states [5, 6]. Typical spectrum is depicted in gous to the [4Fe–4S] case. 3+ Fig. 1g. The observed contact shifts are smaller than those The favorable NMR properties of the [4Fe–4S] cluster 2+ in [2Fe–2S] proteins, indicating that, at room temperature, and the availability of a series of homologous proteins from the excited levels are less populated, and therefore, larger J different bacterial sources, characterized by high reduc- 2+ values than in the [2Fe–2S] case are operative [39, 40]. tion potential values, spanning the + 450/+ 50 mV range Compared to [2Fe–2S] ferredoxins, shorter electron spin (termed High Potential Iron Protein or HiPIPs) provided relaxation times determine sharper signals for Cys βCH / NMR spectroscopists with an exemplary case. Sequence- αCH protons, which made possible the sequence-specific, specific assignments of Cys residues bound to the cluster stereospecific assignment of all eight βCH signals of the were performed, and the two iron ions constituting the mixed iron bound cysteines [40–42]. It was also found that contact valence pair and those forming the purely ferric pair were shifts of Cys βCH protons depend on the Fe–S–C–H dihe- identified [40– 42, 57–60]. It was observed that the electronic dral angle [43]; this angular dependence was successfully distribution within the cluster varies from one protein to converted into structural constraints within solution struc- another, which can be described either with a “low symme- ture calculations (see later) [44]. try distribution” in which the extra electron (considering that 2+ 3+ 3+ 3+ The [4Fe–4S] state can be either oxidized to [4Fe–4S] a [4Fe–4S] cluster can be formally viewed as 4Fe plus or reduced to [4Fe–4S] ; in small electron transfer proteins, one electron) is unevenly distributed among the iron ions the number of hydrogens bonds with the sulfur atoms of the (Fig. 2c), or with a chemical equilibrium between two dif- cluster is the driving force for stabilizing one of the pos- ferent electronic distributions within the cluster, the position sible oxidation state pairs [45, 46], while water and peptide of the equilibrium being determined by the electric field pro- dipoles [47], electrostatic energy [48] and aromatic residues duced by the charges of the protein atoms around the cluster 3+ around the cluster [49, 50] provide a fine tuning of the reduc- and by the metal ligands (Fig. 2d) [60]. The [4Fe–4S] sys- tion potential. tem can also be described by introducing a double exchange Upon cluster reduction to [4Fe–4S] , the ground state is term [61], describing the resonance delocalization of the paramagnetic. The electronic situation can be described by extra electron as depicted in Fig. 2e. It was found that, given 2.5+ 2.5+ the combination of a mixed valence F e –Fe pair and by the consensus sequence for cluster binding to HiPIPs, i.e., 2+ 2+ a purely ferrous pair, F e –Fe , which are antiferromagneti- Cys -X-X-Cys -X -Cys -X -Cys , the iron ion bound to 1 2 n 3 m 4 cally coupled with each other to give an S = 1/2 ground state Cys (Fe ) is the most reducible iron ion and it is in a mixed 4 4 [38, 51]. While nuclear relaxation rates are very similar to valence state in all the investigated HiPIPs from different 2+ those observed in the [4Fe–4S] case [52], the magnitudes bacterial sources; F e and Fe share the extra electron and 1 3 of the observed contact shifts as well as their temperature- interchange their character when passing from one protein dependence are quite distinctive. Signals from Cys βCH / to another, while F e is the less reducible iron ion and has a 2 2 αCH protons are spread over a 65–5 ppm range (Fig. 1h). purely ferric character throughout the series (Fig. 2c, d) [60]. 1 3 670 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 dependent structural rearrangements are too small to be Solution structures of Fe–S proteins observable [64], consistent with the low reorganization and beyond energy present in electron transfer proteins [95]. Redox- dependent structural differences were instead observed in The first solution structure of a paramagnetic protein [2Fe–2S] putidaredoxin, using a combination of diamag- determined by NMR was that of E. halophila HiPIP I, con- 2+ netic restraints, paramagnetic restraints and residual dipo- taining a reduced [4Fe–4S] cluster [62]. This structure lar couplings measured in orienting systems [96]. represented a breakthrough: until then, there was a com- Chemical shift differences were widely explored to map mon and somehow dogmatic belief that NMR structures transient interactions, as will be discussed in the next sec- of paramagnetic proteins were impossible to obtain, due tion. However, it should be always taken into account that, in to the lack of information in the proximity of the paramag- paramagnetic proteins, chemical shift differences can arise netic center. Indeed, scalar and dipolar connectivites are from both structural changes, and changes of the hyperfine quenched by the presence of unpaired electron spins, but contributions (either a change of the hyperfine coupling con- tailored 1D NOE, 2D TOCSY/NOESY experiments pro- stant or a change of the magnetic susceptibility anisotropy vide sufficient structural information to obtain low RMSD tensor). The two effects need to be disentangled for a proper structures also around the Fe–S cluster. Furthermore, the analysis of the available NMR information. In the case of hyperfine interaction is, per se, a source of additional Fe–S proteins, the hyperfine contributions to the chemical constraints, which can be implemented within standard shifts are limited to Fe–S cluster-bound ligands and to those structure calculations programs to circumvent and possibly residues that are hydrogen-bonded to the cluster. Hyperfine compensate the loss of structural information [63]. shifts are often more immediate and sensitive than NMR After the first NMR structure of reduced E. halophila structures to monitor chemical events. For example, the tran- HiPIP I, many NMR structures were determined for sev- sition from a native state to high energy species in an unfold- eral Fe–S proteins in different oxidation states [44, 64–73]. ing process, and the way how the folding/unfolding process For all of them, tailored approaches were used to compen- is triggered by the electron transfer in electron-transfer pro- sate the quenching of the cross peaks arising from scalar teins, can be followed by simple 1D H NMR experiments and dipolar couplings, and to obtain structures with a low [97]. In the case of two [4Fe–4S] clusters containing ferre- RMSD values in the proximity of the cluster. Recently, doxins, it was possible to measure the inter-cluster electron NMR structures of the holo form of some Fe–S proteins self-exchange rates and compare them with the exchange involved into the mitochondrial ISC assembly machin- rates observed between oxidized, partly reduced, and fully ery [74] and of the NO sensing protein Wb1, containing 2+ reduced states [43]. Hyperfine shifts on N nuclei were also a [4Fe–4S] cluster, have been solved [75]. Structures used to monitor the hydrogen-bonding network of residues of Fe–S proteins in their apo form or with a diamagnetic surrounding the cluster [27, 98]. metal ion replacing the Fe–S cluster, determined with Many experimental approaches have been developed to the standard/classic solution NMR approaches, are also collect structural information on paramagnetic proteins, and available [76–82]. The latter approaches were also used many of them turned out to be suitable for Fe–S proteins: to obtain solution NMR structures of holo proteins, but dihedral angle constraints from Karplus-type equations, without structurally defining the area surrounding the clus- derived by considering the through-bond unpaired electron ter [83–85], i.e., no coordinates are given for the cluster spin density delocalization onto the ligands [43], T and T atoms, and for the residues coordinating and/or belonging 1 2 relaxation-based constraints [68, 99], and C direct detec- to the cluster environment. Hybrid approaches, where the tion-based approaches [100–103] were successfully applied absence of direct structural information in the proximity 3+/2+ to Fe–S proteins. The magnetic anisotropy of the Fe of the paramagnetic center was compensated by the use of ions in a tetrahedral environment is quite low, thus limiting homology models, have also been used [86–89]. to a few cases the use of paramagnetism-induced residual The analysis of the structures of the same protein in two dipolar couplings and pseudocontact shifts as a source of different oxidation states for several Fe–S proteins showed structural information [104, 105], while cross-correlation that differences in chemical shifts, unobserved residues, phenomena [106] are not, at least so far, suitable for exploi- internal mobility and thermodynamic stability are suit- tation in Fe–S proteins. able data to map subtle changes between the two different In more recent years, the use of NMR allowed the charac- oxidation states [90–94]. On the contrary, no structural terization of various aspects of Fe–S proteins in the studies differences are observed in the solution structures at the of the complex machineries responsible for the biogenesis available resolution. Even the solution structures of some of Fe–S proteins. Structural properties, recognition patterns HiPIPs, which are available for both oxidation states with characterized by weak transient protein–protein interac- a very high resolution, indicated that oxidation-state tions, and transient metal binding sites can be successfully 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 671 addressed by NMR [107]. Small electron transfer proteins resolution by other methodologies [110, 111]. The cel- extensively studied in the previous decades have been used lular pathways responsible for the maturation iron–sulfur as model systems for more complex cases, in which confor- proteins in humans are sequential, multistep processes, mational flexibility and protein–protein interactions make overall comprising at least 30 interacting proteins involved the investigation more challenging. The combination of 2D in two distinct machineries: the mitochondrial Fe–S cluster HSQC tailored to paramagnetic systems and inversion recov- (ISC) assembly machinery, and the cytosolic Fe–S pro- ery (IR) lead to the development of IR-HSQC-antiphase tein assembly (CIA) machinery, respectively composed (AP) [108, 109], a 2D experiment designed to provide of 17 and 13 proteins [112–114]. By combining standard both additional assignment and relaxation-based structural and tailored solution NMR experiments, it was possible information for those cases in which information from con- to describe structural and mechanistic aspects of the Fe–S tact shifts alone cannot be obtained. The use of this pulse cluster transfer and assembly processes occurring in these sequence (Fig. 3a) turned out to be particularly helpful for two machineries, which involve the formation of transient 2+ 1 systems such as [2Fe–2S] proteins, characterized by effi- complexes. 1D H NMR experiments optimized to study cient paramagnetic relaxation and by the absence of hyper- the proximity of paramagnetic centers can provide infor- fine shifts for all residues other than cluster ligands. The mation on the type of Fe–S cluster(s) bound or assembled inversion recovery delay τ selects the signals according to T on a target protein or in a protein–protein complex, as well 1 15 relaxation, while a customized choice of the coherence trans- as on the redox state(s) of the cluster(s); standard H– N 1 15 fer delay δ allows us to optimize it according to T relaxa- HSQC and H– N IR-HSQC-AP NMR experiments can tion properties (Fig. 3b). Therefore, many signals affected by detect weak and transient protein–protein interactions paramagnetic relaxation are buried under the bulk diamag- through NH chemical shift mapping upon titration of one netic envelope. In the case of the CIAPIN1 domain of human N-labeled partner (apo or holo) with the other unlabeled 2+ protein anamorsin, the [2Fe–2S] clusters are expected to protein (apo or holo) (Fig. 4). These NMR data allowed us induce minimal hyperfine shift and sizable paramagnetic to identify the protein–protein interface on the interacting nuclear relaxation on H spins that do not belong to metal proteins, to have a good estimate of their affinity, of the coordinating cysteine residues, but are within a ~ 10 Å dis- stoichiometry of the interaction, and of the binding speci- tance from each of the two iron ions. Consistent with these ficity (Fig. 4). It was also possible to obtain a structural 1 15 expectations, the H– N-HSQC spectrum of the protein, model of protein complexes that bind a Fe–S cluster, via recorded under standard conditions (Fig. 3d), shows only an experimentally driven docking approach (exploiting the 71 out of 108 expected backbone NH signals. About 30% HADDOCK program [115]), by integrating NMR chemi- of the resonances remain unobserved due to paramagnetic cal shift perturbation analysis with other experimental data broadening or exchange contributions. As shown in Fig. 3c, derived from EPR, NMR tailored to paramagnetic systems 10 additional HN signals, completely absent in standard and mutagenesis [116]. 1 15 NMR experiments, are present in the H– N IR-HSQC-AP For several years, solution NMR has been extensively experiment. Furthermore, three backbone HN signals, barely exploited to investigate molecular aspects of the ISC detectable in standard experiments, significantly increase machinery of bacteria [117, 118]. In particular, E. coli their intensity. Acquisition of spectra with different IR has emerged as the model organism providing the great- 1 15 delays and the analysis of integrated intensity of the H– N est insights into the mechanistic details for Fe–S cluster resonances allowed us to measure the T values for 12 out biosynthesis and delivery to target proteins in the ISC of 13 H signals. This has provided precious information on machinery. In those studies, solution NMR data have the relative positions of related residues with respect to the been quite often integrated with other techniques, such [2Fe–2S] cluster, which has been used to obtain a structural as small-angle X-ray scattering (SAXS), optical spectros- model of the CIAPIN1 domain of anamorsin [99]. copies, Mössbauer, X-ray crystallography and molecular dynamic simulations [119, 120]. An NMR-based inte- grated approach has been also applied to the structural The contribution of solution NMR characterization of isolated human proteins of the ISC and to understand molecular aspects of Fe–S CIA assembly machineries [121–125], and, more recently, cluster trafficking and assembly in humans especially by our group, to investigate interaction networks involving proteins of the human ISC assembly and CIA The main advantage of solution NMR in the characteriza- machineries [99, 126–132]. In the following sections, we tion of cellular pathways involving numerous interacting present how solution NMR studies contributed to describ- proteins, consists in the possibility of investigating, at the ing protein–protein interactions in both human ISC and atomic level, weak transient protein–protein interactions, CIA assembly machineries. which are de facto difficult to be characterized at high 1 3 672 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 1 15 Fig. 3 The H– N IR-HSQC- AP NMR experiment: a new tool for paramagnetic Fe–S proteins. Schematic draw- ings of the pulse sequence 1 15 of the H– N IR-HSQC-AP NMR experiment. The inver- sion recovery delay τ and the coherence transfer delay δ must be chosen according to, respectively, T and T relaxa- 1 2 tion properties of the signals of interest. b Efficiency of an INEPT transfer function at dif- ferent H T values: b 100 ms, c 10 ms, d 5 ms, e 2 ms, f 1 ms, g 0.5 ms. Relaxation is neglected in a. Letters have been drawn at the correspondence of the maximum values for each transfer function. A dashed line is shown at the 2.65 ms of 1 15 INEPT step (94 Hz for H– N J coupling). A solid line is shown in correspondence of 10% transfer efficiency. The latter is a limit threshold below which direct excitation of N spins should replace the INEPT step in the first part of the sequence. 1 15 c Optimized H– N IR-HSQC- AP experiment vs. d standard 1 15 H– N HSQC experiment acquired on 500 MHz at 298 K 2+ on the [2Fe–2S] -CIAPIN1 domain of human anamorsin 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 673 ferredoxins recognizing the multi-component complex are Solution NMR spectroscopy in the human ISC assembly machinery close to the [2Fe–2S] cluster, and that no interaction occurs between apo forms of ferredoxins and the HMW complex. The human proteins of the ISC assembly machinery are all Solution NMR largely contributed to investigate the sub- sequent steps of the ISC assembly machinery, being able to soluble proteins located in the mitochondrial matrix. In the current working model, a [2Fe–2S] cluster is de novo syn- provide a detailed model of how the [2Fe–2S] clusters, de novo synthetized in the multi-component complex, couple thesized on the scaffold protein ISCU2 by a high molecular weight complex (HMW complex hereafter), composed of with each other to form a [4Fe–4S] cluster. This process involves four interacting proteins as mentioned above, i.e., five proteins (ISCU2, NFS1, frataxin, ISD11 and the acyl carrier protein) [133–135]. On the basis of yeast in vivo data GLRX5, ISCA1, ISCA2 and IBA57. The crystal structure of [2Fe–2S] GLRX5 shows a homo- [136–138], it has been proposed that the subsequent step in the human ISC assembly process consists of the transfer of tetrameric structural organization (a dimer of dimers), where two [2Fe–2S] clusters are coordinated by four protein sub- the newly synthetized cluster to the mitochondrial monothiol glutaredoxin, GLRX5, which acts as a Fe–S cluster transfer units and four GSH molecules and buried in the tetramer [153]. In such conformation, the two clusters are not easily protein, inserting the cluster into mitochondrial [2Fe–2S] protein targets. GLRX5 can also transfer the [2Fe–2S] clus- accessible by cluster receiving apo proteins, a circumstance that impairs any possible cluster transfer process. This struc- ter to the protein complex acting late in the ISC assembly machinery for generating [4Fe–4S] clusters [139]. The tural organization would make the proposed chaperone func- tion of GLRX5 difficult to occur. However, it was shown by assembly of the [4Fe–4S] cluster is accomplished by two homologous proteins (ISCA1 and ISCA2), which contain NMR that apo GLRX5 is monomeric in solution, and that it undergoes dimerization only upon cluster binding [128]. three conserved cysteine residues in a CX CGC sequence motif, and by a third protein (IBA57), whose function in the These data indicate that the tetrameric state observed in the crystal structure of [2Fe–2S] GLRX5 is likely determined process is still unknown. It has been shown that these three proteins are strictly required for the maturation of mitochon- by crystallization conditions, thus making the tetrameric state of [2Fe–2S] GLRX5 of poor functional relevance to drial [4Fe–4S] proteins, but not necessary for the maturation of mitochondrial [2Fe–2S] proteins in eukaryotes [140–144]. the cluster transfer process. The combination of standard 1 15 1 15 H– N HSQC and H– N IR-HSQC-AP NMR experi- Once assembled by ISCAs, the [4Fe–4S] cluster is inserted into mitochondrial [4Fe–4S] protein targets, being this pro- ments allowed the identification of the residues affected by [2Fe–2S] cluster binding [128]. By mapping the chemi- cess often dependent on other ISC accessory proteins, such as NFU1, BOLA3 and NUBPL [81, 145–147]. cal shift variations between apo and [2Fe–2S] GLRX5 on the crystallographic structure of [2Fe–2S] GLRX5 [153], In the de novo formation of a [2Fe–2S] cluster, the major- ity of the proteins of the human ISC assembly machinery we found that the regions affected by cluster binding are in a 10-Å radius sphere centered on the [2Fe–2S] cluster, forms permanent interactions featuring tight binding affini- ties [133–135, 148]. The human HMW complex assembling which bridges the two subunits of the dimer. Backbone NH signals of 11 residues located inside this sphere were not the [2Fe–2S] cluster has been, indeed, isolated as a stable 1 15 unit from E. coli cells. Due to its very high molecular mass, detected in the standard H– N HSQC experiment, but 9 1 15 of them were recovered through the H– N IR-HSQC-AP X-ray crystallography and cryo-EM are the most appropri- ate techniques to structurally characterize the complex at NMR experiment [109]. Their H T values increase with −6 increasing distance from the cluster with the expected r the atomic level and have, indeed, successfully generated structural models of the complex or sub-complex forms dependence. Overall, the NMR data indicate that the dimeric state of [2Fe–2S] GLRX5 in solution adopts essentially the [133–135]. In this first step of the human ISC machinery, solution NMR was crucial to investigate protein–protein same structural arrangement as observed for the dimer in the 1 15 crystal structure. However, H– N IR-HSQC-AP and 1D interactions that occur between the HMW complex and mitochondrial [2Fe–2S] ferredoxins. Through these inter- C NMR experiments tailored to paramagnetic systems also showed that dimeric [2Fe–2S] GLRX5 exists in solution as actions, electrons are provided to the HMW complex by a mitochondrial [2Fe–2S] ferredoxin for generating the a mixture of two species in equilibrium each other (GLRX5 and GLRX5 ), as two sets of signals for the Fe–S ligand Cys [2Fe–2S] cluster on ISCU2 [74, 132, 149, 150]. As it often occurs in protein–protein interactions driving electron 67, and for Ser 70 were identified [128]. Also in standard 1 15 H– N HSQC experiments, the six residues surrounding transfer [151, 152], the interaction between the HMW com- plex and [2Fe–2S] ferredoxins is transient and with a µM the “paramagnetic sphere” have two sets of NH signals, both sets having chemical shifts different from those of the apo range affinity; therefore, solution NMR was the successful approach to characterize the process at the molecular level protein. By mapping the residues experiencing two sets of signals on the dimeric structure of [2Fe–2S] GLRX5, we [74, 132]. In such studies, it was shown that the regions of 1 3 674 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 675 ◂Fig. 4 Solution NMR as a tool to investigate weak, transient pro- occur at the dimer interface, suggesting an intrinsic pro- tein–protein interactions in Fe–S protein maturation pathways. Weak pensity for the subunits to be swapped with other protein and transient protein–protein interactions are detected by following partners. This is, indeed, what occurs by mixing apo ISCA2 1 15 backbone NH chemical shift changes occurring in a standard H– N 1 15 with apo ISCA1. This process has been characterized by HSQC NMR experiment and in a H– N IR-HSQC-AP NMR experi- 15 1 15 ment, upon titrating N-labeled protein with the unlabeled protein solution NMR, performing H– N HSQC NMR titrations of 1 15 partner and vice versa. Standard H– N HSQC experiments allow N-labeled apo ISCA1 or ISCA2, with, respectively, unla- the identification of protein–protein interacting regions far from the beled apo ISCA2 or ISCA1 [82]. The two proteins form paramagnetic Fe–S cluster (showed in cyano), and to estimate the 1 15 in solution a stable heterodimeric complex by exchanging, dissociation constant (K ) of the observed interaction. H– N IR- HSQC-AP NMR experiment allows to identify protein–protein inter- slowly on the NMR time scale, one subunit of the ISCA2 acting regions close to the paramagnetic Fe–S cluster (showed in dimer with one subunit of ISCA1, which, as isolated apo yellow). 1D H NMR experiment provides information on the kind protein, is present in solution as a monomer–dimer equilib- of Fe–S cluster(s) bound or assembled on a target protein or protein– rium (Fig. 5). A well-defined surface of interaction, which protein complex, on the redox state(s) of the cluster(s), and on the cluster ligands involves the subunit–subunit interface of homo-dimeric apo ISCA2, was also identified. The NMR data provided a clear evidence that a thermodynamically favored heterodimeric observed that they are all around the iron-bound GSH mol- adduct between ISCA2 and ISCA1 is formed at the expense ecule. Among these eight residues, the two charged ones, of the homodimeric species (Fig. 5). This is in agreement Lys 101 and Asp 123, are particularly important to establish with in vivo data showing a tight interaction between ISCA1 electrostatic interactions with the glycine carboxylate and and ISCA2 [158]. the glutamate amine groups of GSH, respectively. All the The 1D H spectrum of holo human ISCA2 as purified available NMR data indicate that two dimeric species of from E. coli cells showed a set of broad signals character- 2+ [2Fe–2S] GLRX5 exist in solution, differing in the bind- istic of [2Fe–2S] cluster binding [82]. In contrast, upon ing mode of the GSH molecule (Fig. 5). Apparently, in the chemical reconstitution, the 1D H spectrum is consistent crystal structure, [2Fe–2S] GLRX5 adopted one of the two with the presence of a mixture of [4Fe–4S] cluster-bound 2+ forms. The possible functional relevance of the presence of dimeric species and a minor [2Fe–2S] cluster-bound these two species in equilibrium has then been addressed by species [82]. Backbone N relaxation data indicated that characterizing the interaction between [2Fe–2S] GLRX5 and [2Fe–2S] or [4Fe–4S] cluster binding does not alter the its protein partners ISCAs. quaternary structure of ISCA2 [82], at variance with what In all available bacterial apo and holo structures, ISCA observed in the bacterial homologues [154–157]. Since each proteins are either dimeric or tetrameric, and show differ - subunit of the ISCA2 dimer has three potential Fe–S cluster ent symmetries and different cluster ligands [154– 157]. This ligands (i.e., the conserved Cys 79, Cys 144, Cys 146), 1D conformational variability observed in bacterial ISCAs sug- H NMR experiments have been perfomed to identify the gests that it is not very appropriate to transfer the structural two pairs of [2Fe–2S] cluster ligands. The [2Fe–2S] clus- and cluster coordination information acquired on bacterial ter binding properties of Cys-to-Ser single mutants for each proteins to human ISCAs. A detailed structural characteriza- conserved cysteine were compared with those of the wild- tion of the human ISCAs is thus required before proceeding type protein and their H NMR spectra analyzed [129]. From with protein–protein interaction studies with their partner this study, it emerged that ISCA2 coordinates the oxidized 2+ proteins. In solution, apo ISCA2 is a symmetric dimer, [2Fe–2S] cluster with two Cys 79, provided by each of with a well-structured α–β domain encompassing residues the two subunits of the ISCA2 homodimer, with Cys 146 50–140, and a completely unstructured C-terminus of 15 from one subunit of the homodimer and with Cys 144 from residues (Fig. 5) [82]. Backbone N NMR relaxation data the other subunit of the homodimer. NMR data acquired on 13 15 showed a global rigidity of the α–β domain, but also identi- ( C, N) Cys selectively labeled wild-type ISCA2 protein fied a certain degree of backbone flexibility in the loop con- suggested that this cluster coordination is also conserved taining the Fe–S ligand Cys 79, located in the first position once ISCA2 binds a [4Fe–4S] cluster [129]. of the conserved CX CGC motif. Also, the C-terminal tail Cluster transfer between [2Fe–2S] GLRX5 and the of ISCA2, encompassing residues 141–154, and containing apo form of ISCA1 and ISCA2 was then characterized the other two conserved Fe–S binding Cys residues (Cys 144 by performing NMR titrations. The NMR data indicated and Cys 146), was found to be highly flexible. Therefore, that cluster transfer occurs unidirectionally from GLRX5 NMR analysis indicated that all the regions containing the to apo ISCA1 and ISCA2 and that the [2Fe–2S] GLRX5 conserved Cys residues can easily undergo structural rear- form is preferentially reacting relative to [2Fe–2S] GL RX5 rangements to bind the cluster. Severe line broadening of a [128]. N NMR relaxation data showed that ISCA1 and 1 15 few backbone NH signals observed in the H– N HSQC ISCA2 receive the [2Fe–2S] cluster from [2Fe–2S] GLRX5 map of apo ISCA2 indicates that dynamic processes also in their dimeric state. NMR data also showed that cluster 1 3 676 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 Fig. 5 The NMR contribution to the investigation of [4Fe–4S] clus- ISCA1–ISCA2 hetero-dimeric complex via an associative process ters formation in the mitochondrial iron–sulfur cluster assembly that involves a transient protein–protein intermediate; [2Fe–2S] machinery. On the basis of standard and paramagnetic systems- GLRX5 is more reactive than [2Fe–2S] GLRX5 to donate the clus- b a tailored NMR experiments, we provide a model for the transfer of ter; ISCA1–ISCA2 hetero-dimeric complex is obtained in solution 2+ two [2Fe–2S] clusters from GLRX5 to ISCA1–ISCA2 hetero- by exchanging one subunit of the ISCA2 symmetric dimer with one 2+ dimeric complex: in solution dimeric [2Fe–2S] GLRX5 has two subunit of ISCA1, which, as isolated protein, is present in solution in 2+ states in equilibrium with each other, differing in the binding mode a monomer–dimer equilibrium; the two [2Fe–2S] clusters received of the GSH molecules ([2Fe–2S] GLRX5 and [2Fe–2S] GLRX5 ); preferentially by [2Fe–2S] GLRX5 are reductively coupled on a b b 2+ 2+ dimeric [2Fe–2S] GLRX5 specifically transfers the cluster to apo ISCA1–ISCA2 hetero-dimeric complex to form a [4Fe–4S] cluster transfer occurs via the formation of a low-populated pro- ISCAs proteins, and thus plays a functional role in the clus- tein–protein complex, with an interacting surface involving ter transfer mechanism (Fig. 5). the GLRX5 region surrounding Cys 67 [128]. Finally, apo Solution NMR was also applied to the investigation of GLRX5 does not interact with apo ISCA1 or ISCA2, as [2Fe–2S] cluster transfer from GLRX5 to the heterodi- no significant spectral changes are observed when the apo meric ISCA1–ISCA2 complex. NMR data showed that 2+ 2+ proteins are mixed together, indicating that the [2Fe–2S] two [2Fe–2S] clusters are transferred from [2Fe–2S] cluster is essential for the formation of a weakly interact- GLRX5 to the ISCA1–ISCA2 heterodimeric complex, and ing protein–protein adduct. In conclusion, a model for the that the transfer occurs via the formation of a low-popu- transfer of the [2Fe–2S] cluster from GLRX5 to ISCA1 lated, transient protein–protein complex [82]. This sup- and ISCA2 can be proposed on the basis of the NMR data: ports a general cluster transfer mechanism occurring via (1) dimeric [2Fe–2S] GLRX5 has two states in equilib- an associative process between GLRX5 and either homodi- rium with each other, [2Fe–2S] GLRX5 and [2Fe–2S] meric ISCA1 or homodimeric ISCA2 or the heterodimeric GLRX5 ; (2) dimeric [2Fe–2S] GLRX5 specifically trans- ISCA1–ISCA2 complex. By monitoring NMR chemical 2+ 15 fers the [2Fe–2S] cluster to apo ISCA1 and apo ISCA2 shift changes on N-labeled ISCA1–ISCA2 upon inter- via an associative process that involves a weak transient action with [2Fe–2S] GLRX5, it was found that the two 2+ protein–protein intermediate; and (3) [2Fe–2S] GLRX5 is [2Fe–2S] clusters received by GLRX5 were reductively 2+ more reactive than [2Fe–2S] GLRX5 to donate the cluster coupled to form a [4Fe–4S] cluster (Fig. 5). The same 2+ to apo ISCA1 and apo ISCA2. In conclusion, the equilibrium final [4Fe–4S] species was obtained after transferring one 2+ between [2Fe–2S] GLRX5 and [2Fe–2S] GLRX5 species [2Fe–2S] from GLRX5 to apo ISCA2 homo-dimer, and a b 2+ is the trigging factor specifically driving cluster transfer to then adding [2Fe–2S] ISCA1 homo-dimer to the mixture. 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 677 1 15 1 2+ Standard H– N HSQC and 1D H NMR experiments also form a [4Fe–4S] cluster bound to the hetero-complex. The provided detailed information on the mechanism of the proposed molecular model agrees with in vivo data on yeast, 2+ formation of the [4Fe–4S] cluster in the ISCA1–ISCA2 which showed that all three conserved cysteines of Isa1 and hetero-dimeric complex. By monitoring cluster transfer Isa2, the yeast homologues of ISCA1 and ISCA2, are essen- 2+ from [2Fe–2S] GLRX5 to Cys-to-Ser single mutants of tial for the maturation of [4Fe–4S] proteins [161, 162]. each conserved cysteine of ISCA2 (i.e., C79S, C144S, and C146S mutants), we were able to define the different roles Probing the human CIA machinery with solution of the cysteines in the cluster transfer process [129]. The NMR spectroscopy NMR data supported a model in which the two C-terminal cysteines, located in the unstructured and flexible C-terminal As it is for the mitochondrial ISC assembly machinery, the 2+ tail of the ISCAs proteins, extract the [2Fe–2S] cluster proteins of the human CIA machinery responsible for the from GLRX5 by forming a transient, low-populated, clus- synthesis, trafficking, and insertion of clusters into the ter-mediated GLRX5-ISCAs intermediate, where the two cytosolic and nuclear Fe–S protein targets are all soluble. In cluster-binding GSH molecules of GLRX5 are substituted by cytoplasm, the ratio of human proteins containing [2Fe–2S] the ISCA cysteines. This cluster-extraction mechanism from vs. [4Fe–4S] is 18:10 [163]. GLRX5 results in the formation of an ISCA1–ISCA2 species The current working model for cytosolic/nuclear that binds the cluster via the four C-terminal cysteines. The [4Fe–4S] protein maturation envisages that a [4Fe–4S] clus- latter species is, however, transient and no [2Fe–2S] cluster ter is assembled on a specific scaffold complex, formed by bound species can be isolated once Cys 79 is absent, i.e., two cytosolic Fe–S cluster assembly factors, NUBP1 and upon chemical reconstitution of the C79S ISCA2 mutant NUBP2 [164–166]. The [4Fe–4S] cluster is then transferred [129]. Cys 79 is, however, not involved in the cluster trans- to a high molecular weight complex composed by three pro- fer step, as the C79S ISCA2 mutant is still able to extract teins, named CIA targeting complex, that mediates its final 2+ the [2Fe–2S] cluster from GLRX5 [129], via the forma- incorporation into the cytosolic/nuclear targets [167, 168]. tion of the transient, low-populated intermediate with no Other CIA accessory proteins are often required to assist cluster release in solution. However, the transfer is not as in the incorporation of [4Fe–4S] clusters into specific pro- efficient as that observed for the wild-type protein. This sug- tein targets [169, 170]. The origin of iron and sulfur ions gests that cluster-binding affinity is lower once the cluster is used by the NUBP1 and NUBP2 scaffold complex to build coordinated via the C-terminal cysteines only (C79S ISCA2 the [4Fe–4S] cluster is still not identified. It has been pro- mutant case) than once Cys 79 participates to cluster binding posed that cytosolic monothiol glutaredoxins work as cyto- (wild-type ISCA2 case). According to this model, the C144S solic iron donors to cytosolic proteins and to Fe–S and to and C146S ISCA2 mutants, at variance with what happens heme binding proteins [171–173]. This proposal was based in the C79S ISCA2 mutant, can be isolated upon chemical on the fact that the cytosolic monothiol glutaredoxins play reconstitution [129]. Therefore, we propose that the species a role in intracellular iron trafficking and sensing, in iron coordinating the cluster with the four C-terminal cysteines homeostasis and hemoglobin maturation. Recently, it also can evolve into a more thermodynamically favored species, emerged that cytosolic monothiol glutaredoxins can assist 2+ which binds the [2Fe–2S] cluster in the ISCA1–ISCA2 Fe–S protein maturation in the cytosol by acting as [2Fe–2S] heterodimer by Cys 79 and Cys 144 of ISCA2, and Cys 57 cluster donors. The first work proposing a role of cytosolic and Cys 123 of ISCA1. 1D H NMR data showed, indeed, monothiol glutaredoxins in cytosolic [2Fe–2S] cluster traf- that this is the preferential coordination mode in wild-type ficking appeared two years ago, and it represents a very nice ISCA2 for binding either [2Fe–2S] or [4Fe–4S] clusters, example of how in vitro solution NMR data predicted this and presumably also in the heterodimeric ISCA1–ISCA2 function for the cytosolic monothiol glutaredoxins [127]. complex. This mechanism would also make two of the C-ter- One year after that study, human monothiol glutaredoxin minal cysteines (Cys 146 of ISCA2 and Cys 121 of ISCA1) GLRX3 was shown, indeed, to work as a Fe–S cluster chap- available for the coordination of a second cluster which can erone in human cells [174]. be extracted from GLRX5, upon the formation of another The human proteome contains only one monothiol glutar- GLRX5-ISCA intermediate. This transient intermediate, edoxin in the cytosol, i.e., GLRX3, which consists of three 2+ which contains two [2Fe–2S] clusters, might be the spe- domains: one N-terminal thioredoxin (Trx) domain with no cies that, by accepting two electrons from a still unknown Trx enzymatic activity, but functionally indispensable [171, physiological electron donor (Fig. 5), evolves to the final 175], and two monothiol glutaredoxin (Grx) domains, each 2+ [4Fe–4S] cluster-bound ISCA1–ISCA2 complex. A reduc- able to bind a glutathione-coordinated [2Fe–2S] cluster via 2+ tive coupling of two [2Fe–2S] clusters, which is a general protein dimerization (Fig. 6) [176, 177]. Yeast-two-hybrid 2+ mechanism for generating a [4Fe–4S] cluster [159, 160], and affinity capture-MS screens showed that in vivo GLRX3 would therefore, occur on the latter transient intermediate to binds anamorsin [178]. Anamorsin contains two domains: a 1 3 678 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 Fig. 6 The NMR contribution to the elucidation of the GLRX3- through complementary charged residues, with the FMN-binding 1 15 dependent anamorsin maturation pathway. Standard H– N-HSQC domain of NDOR1 to perform the electron transfer reaction. Stand- 1 15 and H– N IR-HSQC-AP NMR experiments, combined with UV– ard NMR experiments showed that the [2Fe–2S] GLRX3–BOLA2 2 2 2+ Vis and EPR spectroscopy, showed that GLRX3 forms a 1:1 het- hetero-complex transfers in vitro both [2Fe–2S] clusters to apo ero-dimeric complex with anamorsin, in which both clusters from anamorsin, producing its mature holo state, and that this process goes [2Fe–2S] GLRX3 are transferred to anamorsin. In its mature holo via the same protein–protein recognition mechanism operating in the 2 2 state anamorsin interacts with NDOR1, forming a specific protein GLRX3-anamorsin interaction, i.e., specifically occurring between complex, where the anamorsin unstructured linker tightly interact the N-terminal domains of the two proteins. The BOLA2–GLRX3 with NDOR1, while the C-terminal CIAPIN1 domain of anamor- complex might be released in solution upon the interaction of holo sin, containing the [2Fe–2S] redox center, only transiently interacts, anamorsin with NDOR1 N-terminal well-folded domain (N-domain, hereafter) of 172 homologues of GLRX3 [171, 173]; (4) Fe–S cluster loading residues and a largely unstructured C-terminal domain of on Dre2 is independent of the cytosolic iron–sulfur protein 90 residues, named Cytokine-Induced Apoptosis INhibitor assembly machinery [180]. 2+ 1 (CIAPIN1, hereafter), essential for the viability of yeast The monomeric apo and dimeric [2Fe–2S] clus- [179, 180], and containing two highly conserved cysteine- ter-bound forms of GLRX3 (hereafter apo GLRX3 and rich motifs, easily able to independently bind a [2Fe–2S] [2Fe–2S] GLRX3 , respectively) [184] were first char - 2 2 cluster (Fig. 6) [99, 181]. These two domains are connected acterized by NMR, showing that the Trx domain does not by a long flexible and unstructured linker of 50 residues have intra- and inter-subunit interactions with the two Grx [181] (Fig. 6). domains, nor with the Trx domain of the other monomer in Anamorsin is the appropriate protein partner to investi- [2Fe–2S] GLRX3 , being therefore, fully available to be 2 2 gate the functional role of GLRX3 in transferring [2Fe–2S] potentially involved in protein–protein interactions [127]. clusters for these reasons: (1) it interacts with GLRX3 This structural aspect prompted us to investigate whether in vivo [178]; (2) it binds [2Fe–2S] and [4Fe–4S] clusters the Trx domain of GLRX3 drives a specific protein–protein [121, 182, 183]; (3) the insertion of the Fe–S cluster into interaction with anamorsin. The analysis of NMR titration the yeast homologue of anamorsin, Dre2, depends on Grx3 data, acquired after mixing the apo proteins, showed that and Grx4, the two cytosolic, functionally redundant yeast apo-GLRX3 and apo-anamorsin form a 1:1 heterodimeric 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 679 complex through their N-terminal domains and that the CIA- strictly linked to all anamorsin-dependent cellular processes. PIN1 domain of anamorsin and the Grx domains of GLRX3 Anamorsin is a crucial early step component of the CIA are not involved in any permanent interaction in this com- machinery, being essential for the maturation of cytosolic/ plex [127]. Molecular recognition between the N-terminal nuclear [4Fe–4S] proteins. Therefore, the decreased activi- domains is, therefore, the crucial factor determining com- ties of the cytosolic [4Fe–4S] proteins IRP1 and GPAT plex formation between the two proteins. A docking model observed by silencing human GLRX3 in HeLa cells [172] of the complex, based on the NMR titration data, showed can be due to impairment of the GLRX3-dependent anamor- that the interaction occurs among the α-helical regions of the sin maturation process. This makes, indeed, the CIA machin- two domains, but also involves a negatively charged, Glu- ery unable to function, i.e., to assemble the [4Fe–4S] clus- rich, region (Glu 71, Glu 75, Glu 78, and Glu 81) of GLRX3 ters of IRP1 and GPAT. and a specific region of the unstructured linker of anamorsin, The matured holo form of anamorsin forms a stable rich with positively charged Lys residues (Lys 175, Lys 180, complex with the cytosolic NADPH-dependent diflavin Lys 181, Lys 187) [127]. These two regions stabilize pro- reductase NDOR1 in the cell [180]. This complex, which tein–protein interactions through electrostatic recognition. receives two electrons from the NADPH cofactor, has been Indeed, the exchange regime between the free and the bound proposed to act as a source of reducing equivalents for the proteins switched from fast/intermediate on the NMR time assembly of target, but not to be a scaffold for Fe–S cyto- scale upon interaction of the N-terminal domains of GLRX3 solic proteins [99, 180]. This means that, once the cluster and anamorsin, to slow when the N-domain of GLRX3 transfer from GLRX3 to anamorsin has occurred in the cell, interacted with full-length anamorsin, in agreement with the complex between the two proteins needs to be termi- a significant increase in the protein–protein affinity when nated, so that the functional process(es) performed by the the linker is present. In conclusion, the NMR/biomolecular mature form of anamorsin can proceed. Recently, we found docking data defined, for the first time, the function of the that the N-terminal domain of anamorsin is not involved in N-domains of the two proteins and identified a role of the protein–protein recognition, and that the C-terminal CIA- linker of anamorsin in stabilizing the protein–protein inter- PIN1 domain of anamorsin, containing the [2Fe–2S] redox action [127]. center, only transiently interacts, through complementary 1 15 1 15 Standard H– N-HSQC and H– N-IR-HSQC-AP NMR charged residues, with the FMN-binding domain of NDOR1 experiments, combined with UV–Vis and EPR spectros- to perform the electron transfer reaction [99]. On the con- copy, showed that [2Fe–2S] GLRX3 forms a 1:1 hetero- trary, the unstructured linker of anamorsin tightly interacts 2 2 dimeric complex with anamorsin, the same formed by the with NDOR1, inducing the formation of a specific and stable apo proteins, in which both clusters of GLRX3 have been protein complex [99]. On this basis, we suggested that upon unidirectionally transferred to the two cluster binding sites interaction of the GLRX3-anamorsin complex with NDOR1, of the CIAPIN1 domain of anamorsin (Fig. 6) [127]. This the linker of anamorsin weakens its interaction with GLRX3, means that the GLRX3 molecule in the 1:1 heterodimeric while favoring the interaction with NDOR1 (Fig. 6). As the complex is in the apo state, that the N-terminal domains (Trx stabilizing effect of the linker on GLRX3-anamorsin interac- of GLRX3 and N-domain of anamorsin) interact in the com- tion is lost, the binding affinity of the N-terminal domain of plex, and that the C-terminal, cluster-binding domains (Grxs GLRX3 with that of anamorsin is decreased, and, as a conse- and CIAPIN1 domains) are not involved in a stable pro- quence, the complex between GLRX3 and anamorsin might tein–protein interaction (Fig. 6). The interaction between the switch to the complex between anamorsin and NDOR1. The N-domains is a fundamental requisite in the cluster transfer linker interaction is thus able to modulate the formation and mechanism to drive [2Fe–2S] cluster transfer from GLRX3 release of the various protein–protein complexes, enabling to the CIAPIN1 domain of anamorsin. We suggested that the the redox-competent state of anamorsin to receive electrons protein–protein interaction between the N-terminal domains from NDOR1. The high flexibility and intrinsic disorder of make the cluster binding domains, i.e., the Grx donors and the linker fits well with the interaction with multiple part- the CAPIN1 acceptor, in the optimal reciprocal orienta- ners, as commonly observed for intrinsically disordered tion for the cluster transfer to efficiently occur. Therefore, it proteins/regions. appears that the transfer process from GLRX3 to anamorsin Several lines of evidence, including affinity purification, is a thermodynamically favored process under kinetic con- yeast two-hybrid studies and gene co-occurrence analysis trol. This mechanism guarantees that two [2Fe–2S] clusters indicated that the monothiol Grxs functionally and physi- are concomitantly transferred in a single molecular event cally interact with another widely conserved protein family, to the target protein requiring two [2Fe–2S] clusters. This the BolA-like proteins [176, 185–187]. Eukaryotic organ- NMR-based study opens new perspectives on the cellu- isms contain BolA-like proteins in both mitochondria and lar function of GLRX3 in humans, showing that GLRX3 cytoplasm. In yeast, mitochondrial BolA1 and BolA3 pro- function, by playing a key role in maturing anamorsin, is teins are involved in the ISC assembly machinery, working 1 3 680 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 as specific mitochondrial ISC assembly factors that facilitate GLRX3 [126, 184]. However, the existence and a functional [4Fe–4S] cluster insertion into a subset of mitochondrial role of the apo complex at the cellular level cannot defini- 2+ proteins, such as lipoate synthase and succinate dehydro- tively be excluded. Which is the system donating [2Fe–2S] genase [81, 147]. On the contrary, cytosolic yeast BolA2 clusters to the apo complex is still under investigation. A protein plays a role in iron homeostasis [188]. In humans, possible pathway, that has been recently proposed, involves the functional role of the mitochondrial BOLA1 and BOLA3 the mitochondrial ISC assembly machinery: the de novo 2+ proteins is still not clearly defined, but it was found that they biosynthesis of the [2Fe–2S] cluster on mitochondrial form a hetero-dimeric complex with GLRX5 in both apo and ISCU2 is followed by cluster export, as [2Fe–2S](GS) [2Fe–2S]-cluster bound states [81]. NMR data, combined complex, via a membrane transporter; then, the [2Fe–2S] with other spectroscopic information, allowed us to obtain cluster is believed to be uptaken by the cytosolic form an experimentally driven docking model of [2Fe–2S] clus- of ISCU2, which delivers it to cytosolic form of NFU1, ter-bridged dimeric BOLA1–GLRX5 and BOLA3–GLRX5 which finally transfers the cluster to GLRX3 [189– 192]. complexes, showing that the BOLA1–GLRX5 complex This molecular process is, however, only based on in vitro coordinates a reduced, Rieske-type [2Fe–2S] cluster, while information, and experimental evidences from in vivo data 2+ an oxidized, ferredoxin-like [2Fe–2S] cluster is present in are required to definitively validate it. Regardless of how the BOLA3–GLRX5 complex [116]. It also appeared that GLRX3 acquires the [2Fe–2S] clusters, the functional the [2Fe–2S] BOLA1–GLRX5 complex is preferentially role of the [2Fe–2S] GLRX3–BOLA2 have been defined. formed over the [2Fe–2S] BOLA3–GLRX5 complex, as a NMR data showed that the [2Fe–2S] GLRX3–BOLA2 2 2 2+ result of a higher cluster binding affinity. The different struc- complex in vitro transfers both its [2Fe–2S] clusters to tural and redox properties observed for the two [2Fe–2S] apo anamorsin producing its mature holo state, through BOLAs–GRX5 complexes as well as their different stabil- the same protein–protein recognition mechanism opera- ity suggested that they can have a diverse molecular func- tive in the GLRX3-anamorsin interaction, that specifically tion. Possibly, [2Fe–2S] BOLA1–GLRX5 complex might be occurs between the N-terminal domains of the two proteins involved in electron transfer processes, while the [2Fe–2S] (Fig. 6) [126]. The [2Fe–2S] GLRX3–BOLA2 complex 2 2 BOLA3–GLRX5 might be involved in cluster transfer ver- maturing anamorsin cannot be formed via the interaction sus client proteins along the ISC assembly pathway. How- between BOLA2 and [2Fe–2S] GLRX3 , since NMR data 2 2 ever, functional data are required to verify such proposed showed that the formed heterotrimeric GLRX3–BOLA2 2+ molecular function of the mitochondrial BOLAs–GLRX5 complex contains only one [2Fe–2S] cluster per com- complexes. As it is for mitochondrial BOLAs and GLRX5 plex. These in vitro data support a model where the apo proteins, cytosolic BOLA2 and GLRX3 proteins form a heterotrimeric GLRX3–BOLA2 complex is the species hetero-complex in both the apo and the [2Fe–2S] cluster that, being able to acquire two [2Fe–2S] clusters, matures bound forms [177, 184]. In both cases, this hetero-complex anamorsin [126]. UV–Vis CD data showing incomplete con- is composed by two BOLA2 molecules and one GLRX3 version from [2Fe–2S] GLRX3 homodimer to [2Fe–2S] 2 2 2 molecule. Recently, solution NMR contributed to unravel GLRX3–BOLA2 heterotrimer are in agreement with this the functional role of this complex. NMR titration data model [184]. These NMR data represent the first experi- showed that apo BOLA2 interacts simultaneously with mental evidence that the heterotrimeric GLRX3–BOLA2 2+ both Grx domains of GLRX3 with an apparent dissociation complex might work as a [2Fe–2S] cluster transfer com- constant of 25 μM, without showing a preferential inter- ponent in CIA machinery pathways. Accordingly with this action toward one of the two Grx domains (Fig. 6) [126]. proposal, a work in human cells reproduced these in vitro On the contrary, the Trx domain is not involved in any findings, showing that GLRX3–BOLA2 complex delivers interaction with BOLA2. Chemical shift mapping iden- [2Fe–2S] clusters to anamorsin via a direct protein–protein tified well-defined interacting regions on both proteins, interaction [174]. Collectively, in vitro and in vivo data comprising the conserved His ligand of BOLA2 and the showed that the GLRX3–BOLA2 complex in mammalian conserved Cys ligand of GLRX3 [126]. This apo complex cells functions as a [2Fe–2S] cluster chaperone, storing and is thus assembled in the proper structural arrangement to delivering [2Fe–2S] clusters. It has been found that GLRX3 2+ bind/receive two bridging [2Fe–2S] clusters (Fig. 6). In homodimers represent a rare species of GLRX3 in cells with human cells, a stable complex between GLRX3 and BOLA2 respect to B OLA2 –GLRX3 homotrimers [174], possibly is observed only when they coordinate bridging [2Fe–2S] because cluster binding is more labile and oxygen-sensitive clusters [174]. The apo complex is not detected, possibly as in GLRX3 than in GLRX3–BOLA2 [184, 193]. There- 2 2 a consequence of its lower stability with respect to the holo fore, it cannot be definitively excluded that also [2Fe–2S] complex, which stabilizes the BOLA2–GLRX3 interaction GLRX3 works as [2Fe–2S] cluster chaperone in human 2+ by bridging two [2Fe–2S] clusters between two BOLA2 cells. molecules and each monothiol glutaredoxin domain of 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 681 In vivo data also showed that the iron bound to anamor- First, these interactions are expected to be transient and not sin did not completely disappear in cells lacking GLRX3 or permanent. Indeed, they need to be formed once the Fe–S BOLA2 [174]. This indicates that anamorsin may be capa- cluster is delivered from the donor(s) to the final target, and ble of acquiring Fe–S clusters from an alternative source. then to be disrupted once the cluster has been transferred, It has been proposed that the mitoNEET/miner1 family of thus producing a matured final target. Second, solution NMR [2Fe–2S] proteins, which in vitro transfer their two [2Fe–2S] is the optimal technique to investigate, at near-physiological clusters to anamorsin [194], might be the alternative source conditions and at an atomic level, protein–protein interaction of [2Fe–2S] clusters [195]. However, whether mitoNEET/ events that are transient and weak. Solution NMR will be miner1 proteins transfer [2Fe–2S] clusters to anamorsin able to clearly define the molecular function of these inter - in vivo is still unknown. So far, it has been only shown that actions, and thus open up new horizons on how to deal with mitoNEET can repair oxidatively damaged [4Fe–4S] clusters the human diseases related to defects in Fe–S protein bio- of iron regulatory protein 1 (IRP1) [195], a critical regu- genesis processes, which have been growing rapidly. In this lator of genes important for iron homeostasis and oxygen respect, we believe that a crucial breakthrough in the Fe–S sensing [196]. This potential cluster transfer function of protein field might be determined by the development and mitoNEET is linked to the redox state of the two [2Fe–2S] the application of paramagnetic NMR methods to character- clusters bound to the protein, since, in their reduced state, ize protein–protein complexes directly in mammalian cells. the clusters are not released, while, in their oxidized state, Indeed, it will be possible to study at atomic resolution the the clusters can be transferred to apo proteins [197, 198]. interactions responsible for maturing Fe–S proteins in living A possible redox system regulating the cluster redox state cells and to investigate physiological cluster-binding proper- of mitoNEET is composed by the cytosolic electron-donor ties and cluster redox states. NADPH/Ndor1/anamorsin complex, the component of the Acknowledgements This article is based upon work from COST CIA machinery discussed above. NMR data showed that Action CA15133, supported by COST (European Cooperation in Sci- the [2Fe–2S] clusters of mitoNEET are reduced by anamor- ence and Technology). sin via the formation of a transient complex that brings the [2Fe–2S] clusters of mitoNEET close to the redox-active Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://creat iveco [2Fe–2S] cluster of anamorsin [199]. These data provide an mmons .org/licen ses/by/4.0/), which permits use, duplication, adapta- in vitro evidence of a possible direct link between the CIA tion, distribution and reproduction in any medium or format, as long machinery and the mitoNEET-dependent repair pathway of as you give appropriate credit to the original author(s) and the source, IRP1: once oxidative stress is not occurring anymore in the provide a link to the Creative Commons license and indicate if changes were made. cell, the Ndor1/anamorsin complex of the CIA machinery is functionally active in the cytoplasm [200] and can reduce the clusters of mitoNEET. In this way, the CIA machinery would stop the mitoNEET cluster transfer pathway repairing References IRP1. The repair pathway is, indeed, no longer needed, once cellular oxidative stress is no more effective. In conclusion, 1. Kowalsky A (1965) Biochemistry 4:2382–2388 the in vitro NMR data provided valuable input for testing, 2. Wüthrich K (1969) Proc Natl Acad Sci USA 63:1071–1078 via cellular studies, whether a direct link between the CIA 3. McDonald CC, Phillips WD, Vinogradov SN (1969) Biochem Biophys Res Commun 36:442–449 pathway and the mitoNEET-cluster transfer pathway exist 4. Phillips WD, Poe M, Weiher JF, McDonald CC, Lovenberg W in human cells. (1970) Nature 227:574–577 5. Poe M, Phillips WD, McDonald CC, Lovenberg W (1970) Proc Natl Acad Sci USA 65:797–804 6. Phillips WD, Poe M, McDonald CC, Bartsch RG (1970) Proc Conclusions Natl Acad Sci USA 67:682 7. Dunham WR, Palmer G, Sands RH, Bearden AJ (1971) Biochim We believe that, in the coming years, solution NMR will Biophys Acta 253:373–384 be fundamental to describe the mechanisms through which 8. Saalmen I, Palmer G (1972) Arch Biochem Biophys 150:767 9. Poe M, Phillips WD, Glickson JD, McDonald CC, San Pietro A [2Fe–2S] and [4Fe–4S] clusters are specifically transferred (1971) Proc Natl Acad Sci USA 68:68 to the mitochondrial and cytosolic final targets. These pro- 10. Anderson RE, Dunham WR, Sands RH, Bearden AJ, Crespi HL cesses are based on a concerted action of several accessory (1975) Biochim Biophys Acta 408:306 proteins, which have been clearly identified. However, the 11. Packer EL, Rabinowitz JC, Sternlicht H (1978) J Biol Chem 253:7722–7730 interaction networks among these accessory proteins are 12. Werth MT, Kurtz DM Jr, Moura I, LeGall J (1987) J Am Chem still quite elusive, and their molecular action is not clearly Soc 109:273–275 defined. We believe that solution NMR can contribute sig- 13. Xia B, Westler WM, Cheng H, Meyer J, Moulis J-M, Markley JL nificantly to answer these questions for two main reasons. (1995) J Am Chem Soc 117:5347–5350 1 3 682 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 14. Goodfellow BJ, Tavares P, Romão MJ, Czaja C, Rusnak F, Le 48. Perrin BS Jr, Niu S, Ichiye T (2013) J Comput Chem 34:576–582 Gall J, Moura I, Moura JJG (1996) JBIC 1:341–354 49. Agarwal A, Li D, Cowan JA (1995) Proc Natl Acad Sci USA 15. Banci L, Bertini I, Luchinat C (1990) Struct Bond 72:113–135 92:9440–9444 16. Beinert H, Holm RH, Munck E (1997) Science 277:653–659 50. Bertini I, Borsari M, Bosi M, Eltis LD, Felli IC, Luchinat C, 17. Noodleman L (1988) Inorg Chem 27:3677–3679 Piccioli M (1996) J Biol Inorg Chem 1:257–263 18. Blondin G, Girerd J-J (1990) Chem Rev 90:1359–1376 51. Mathews R, Charlton S, Sands RH, Palmer G (1974) J Biol Chem 19. Machonkin TE, Westler WM, Markley JL (2005) Inorg Chem 249:4326–4328 44:779–797 52. Bertini I, Briganti F, Luchinat C, Messori L, Monnanni R, Scoz- 20. Banci L, Bertini I, Briganti F, Luchinat C (1991) N J Chem zafava A, Vallini G (1991) FEBS Lett 289:253–256 15:467–477 53. Mouesca J-M, Noodleman L, Case DA, Lamotte B (1995) Inorg 21. Dunham WR, Bearden AJ, Salmeen I, Palmer G, Sands RH, Chem 34:4347–4359 Orme-Johnson WH, Beinert H (1971) Biochim Biophys Acta 54. Le Pape L, Lamotte B, Mouesca J-M, Rius GJ (1997) J Am Chem 253:134–152 Soc 119:9757–9770 22. Banci L, Bertini I, Luchinat C (1991) Nuclear and electron 55. Antanaitis BC, Moss TH (1975) Biochim Biophys Acta 405:262 relaxation. The magnetic nucleus-unpaired electron coupling in 56. Middleton P, Dickson DPE, Johnson CE, Rush JD (1980) Eur J solution. VCH, Weinheim Biochem 104:289–296 23. Cheng H, Xia B, Reed GH, Markley JL (1994) Biochemistry 57. Nettesheim DG, Harder SR, Feinberg BA, Otvos JD (1992) Bio- 33:3155–3164 chemistry 31:1234–1244 24. Orio M, Mouesca JM (2008) Inorg Chem 47:5394–5416 58. Banci L, Bertini I, Capozzi F, Carloni P, Ciurli S, Luchinat C, 25. Dugad LB, La Mar GN, Banci L, Bertini I (1990) Biochemistry Piccioli M (1993) J Am Chem Soc 115:3431–3440 29:2263–2271 59. Bertini I, Gaudemer A, Luchinat C, Piccioli M (1993) Biochem- 26. Sheridan RP, Allen LC, Carter CWJ (1981) J Biol Chem istry 32:12887–12893 256:5052–5057 60. Banci L, Bertini I, Ciurli S, Ferretti S, Luchinat C, Piccioli M 27. Xia B, Pikus JD, McClay K, Steffan RJ, Chae YK, Westler WM, (1993) Biochemistry 32:9387–9397 Markley JL, Fox DJ (1999) Biochemistry 38:727–739 61. Bominaar EL, Borshch SA, Girerd J-J (1994) J Am Chem Soc 28. Skjeldal L, Westler WM, Oh B-H, Krezel AM, Holden HM, 116:5362–5372 Jacobson BL, Rayment I, Markley JL (1991) Biochemistry 62. Banci L, Bertini I, Eltis LD, Felli IC, Kastrau DHW, Luchinat 30:7363–7368 C, Piccioli M, Pierattelli R, Smith M (1994) Eur J Biochem 29. Huynh BH, Moura JJG, Moura I, Kent TA, LeGall J, Xavier AV, 225:715–725 Münck E (1980) J Biol Chem 255:3242–3244 63. Bertini I, Luchinat C, Piccioli M (2001) Methods Enzymol 30. Kent TA, Huynh BH, Munk E (1980) Proc Natl Acad Sci USA 339:314–340 77:6574–6576 64. Bertini I, Eltis LD, Felli IC, Kastrau DHW, Luchinat C, Piccioli 31. Bertini I, Ciurli S, Luchinat C (1995) Struct Bond 83:1–54 M (1995) Chem Eur J 1:598–607 32. Cheng H, Grohmann K, Sweeney WV (1990) J Biol Chem 65. Bertini I, Couture MMJ, Donaire A, Eltis LD, Felli IC, Luchinat 265:12388–12392 C, Piccioli M, Rosato A (1996) Eur J Biochem 241:440–452 33. Busse SC, La Mar GN, Yu LP, Howard JB, Smith ET, Zhou ZH, 66. Banci L, Bertini I, Dikiy A, Kastrau DHW, Luchinat C, Som- Adams MWW (1992) Biochemistry 31:11952–11962 pornpisut P (1995) Biochemistry 34:206–219 34. Macedo AL, Moura I, Moura JJG, LeGall J, Huynh BH (1993) 67. Bertini I, Dikiy A, Kastrau DHW, Luchinat C, Sompornpisut P Inorg Chem 32:1101–1105 (1995) Biochemistry 34:9851–9858 35. Bentrop D, Bertini I, Luchinat C, Mendes J, Piccioli M, Teixeira 68. Bertini I, Donaire A, Luchinat C, Rosato A (1997) Proteins Struct M (1996) Eur J Biochem 236:92–99 Funct Genet 29:348–358 36. Papaefthymiou V, Girerd J-J, Moura I, Moura JJG, Münck E 69. Aono S, Bentrop D, Bertini I, Donaire A, Luchinat C, Niikura Y, (1987) J Am Chem Soc 109:4703–4710 Rosato A (1998) Biochemistry 37:9812–9826 37. Thomson AJ, Robinson AE, Johnson MK, Moura JJG, Moura I, 70. Davy SL, Osborne MJ, Moore GR (1998) J Mol Biol Xavier AV, LeGall J (1981) Biochim Biophys Acta 670:93 277:683–706 38. Thompson CL, Johnson CE, Dickson DPE, Cammack R, Hall 71. Im S-C, Liu G, Luchinat C, Sykes AG, Bertini I (1998) Eur J DO, Weser U, Rao KK (1974) Biochem J 139:97 Biochem 258:465–477 39. Bertini I, Briganti F, Luchinat C, Scozzafava A (1990) Inorg 72. Goodfellow BJ, Macedo AL (1999) Ann Rep NMR Spectrosc Chem 29:1874–1880 37:119–177 40. Bertini I, Capozzi F, Ciurli S, Luchinat C, Messori L, Piccioli M 73. Goodfellow BJ, Macedo AL, Rodrigues P, Moura I, Wray V, (1992) J Am Chem Soc 114:3332–3340 Moura JJG (1999) J Biol Inorg Chem 4:421–430 41. Bertini I, Capozzi F, Luchinat C, Piccioli M, Vicens Oliver M 74. Webert H, Freibert SA, Gallo A, Heidenreich T, Linne U, (1992) Inorg Chim Acta 198–200:483–491 Amlacher S, Hurt E, Muhlenhoff U, Banci L, Lill R (2014) Nat 42. Bertini I, Capozzi F, Luchinat C, Piccioli M (1993) Eur J Bio- Commun 5:5013 chem 212:69–78 75. Kudhair BK, Hounslow AM, Rolfe MD, Crack JC, Hunt DM, 43. Bertini I, Capozzi F, Luchinat C, Piccioli M, Vila AJ (1994) J Buxton RS, Smith LJ, Le Brun NE, Williamson MP, Green J Am Chem Soc 116:651–660 (2017) Nat Commun 8:2280 44. Bertini I, Donaire A, Feinberg BA, Luchinat C, Piccioli M, Yuan 76. Ramelot TA, Cort JR, Goldsmith-Fischman S, Kornhaber GJ, H (1995) Eur J Biochem 232:192–205 Xiao R, Shastry R, Acton TB, Honig B, Montelione GT, Ken- 45. Backes G, Mino Y, Loehr TM, Meyer TE, Cusanovich MA, nedy MA (2004) J Mol Biol 344:567–583 Sweeney WV, Adman ET, Sanders-Loehr J (1991) J Am Chem 77. Kim JH, Tonelli M, Kim T, Markley JL (2012) Biochemistry Soc 113:2055–2064 51:5557–5563 46. Langen R, Jensen GM, Jacob U, Stephen PJ, Warshel A (1992) J 78. Pastore C, Adinolfi S, Huynen MA, Rybin V, Martin S, Mayer Biol Chem 267:25625–25627 M, Bukau B, Pastore A (2006) Structure 14:857–867 79. Xu X, Scanu S, Chung JS, Hirasawa M, Knaff DB, Ubbink M 47. Jensen GM, Warshel A, Stephen PJ (1994) Biochemistry (2010) Biochemistry 49:7790–7797 33:10911–10924 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 683 80. Goodfellow BJ, Duarte IC, Macedo AL, Volkman BF, Nunes 111. Perkins JR, Diboun I, Dessailly BH, Lees JG, Orengo C (2010) SG, Moura I, Markley JL, Moura JJ (2010) J Biol Inorg Chem Structure 18:1233–1243 15:409–420 112. Lill R (2009) Nature 460:831–838 81. Uzarska MA, Nasta V, Weiler BD, Spantgar F, Ciofi-Baffoni S, 113. Maio N, Rouault TA (2015) Biochim Biophys Acta Saviello MR, Gonnelli L, Muhlenhoff U, Banci L, Lill R (2016) 1853:1493–1512 Elife 5:e16673 114. Ciofi-Baffoni S, Nasta V, Banci L (2018) Metallomics 10:49–72 82. Brancaccio D, Gallo A, Mikolajczyk M, Zovo K, Palumaa P, 115. Dominguez C, Boelens R, Bonvin AM (2003) J Am Chem Soc Novellino E, Piccioli M, Ciofi-Baffoni S, Banci L (2014) J Am 125:1731–1737 Chem Soc 136:16240–16250 116. Nasta V, Giachetti A, Ciofi-Baffoni S, Banci L (2017) Biochim 83. Pochapsky TC, Mei Ye X, Ratnaswamy G, Lyons TA (1994) Biophys Acta 1861:2119–2131 Biochemistry 33:6424–6432 117. Yan R, Adinolfi S, Iannuzzi C, Kelly G, Oregioni A, Martin S, 84. Lelong C, Sétif P, Bottin H, André F, Neumann J-M (1995) Bio- Pastore A (2013) PLoS One 8:e78948 chemistry 34:14462–14473 118. Fuzery AK, Tonelli M, Ta DT, Cornilescu G, Vickery LE, Mar- 85. Feng Y, Zhong N, Rouhier N, Hase T, Kusunoki M, Jacquot JP, kley JL (2008) Biochemistry 47:9394–9404 Jin C, Xia B (2006) Biochemistry 45:7998–8008 119. Kim JH, Bothe JR, Alderson TR, Markley JL (2015) Biochim 86. Pochapsky TC, Jain NU, Kuti M, Lyons TA, Heymont J (1999) Biophys Acta 1853:1416–1428 Biochemistry 38:4681 120. Prischi F, Pastore A (2017) Front Mol Biosci 4:12 87. Hatanaka H, Tanimura R, Katoh S, Inagaki F (1997) J Mol Biol 121. Banci L, Ciofi-Baffoni S, Mikolajczyk M, Winkelmann J, Bill E, 268:922–933 Eirini Pandelia M (2013) J Biol Inorg Chem 18:883–893 88. Mo H, Pochapsky SS, Pochapsky TC (1999) Biochemistry 122. Musco G, Stier G, Kolmerer B, Adinolfi S, Martin S, Frenkiel T, 38:5666 Gibson T, Pastore A (2000) Structure 8:695–707 89. Marg BL, Schweimer K, Sticht H, Oesterhelt D (2005) Biochem- 123. Cai K, Liu G, Frederick RO, Xiao R, Montelione GT, Markley istry 44:29–39 JL (2016) Structure 24:2080–2091 90. Miura R, Ichikawa Y (1991) J Biol Chem 266:6252–6258 124. Noguera ME, Aran M, Smal C, Vazquez DS, Herrera MG, 91. Xia B, Volkman BF, Markley JL (1998) Biochemistry Roman EA, Alaimo N, Gallo M, Santos J (2017) Arch Biochem 37:3965–3973 Biophys 636:123–137 92. Bentrop D, Bertini I, Iacoviello R, Luchinat C, Niikura Y, Picci- 125. Li J, Ding S, Cowan JA (2013) Biochemistry 52:4904–4913 oli M, Presenti C, Rosato A (1999) Biochemistry 38:4669–4680 126. Banci L, Camponeschi F, Ciofi-Baffoni S, Muzzioli R (2015) J 93. Bertini I, Luchinat C, Niikura Y, Presenti C (2000) Proteins Am Chem Soc 137:16133–16143 Struct Funct Genet 41:75–85 127. Banci L, Ciofi-Baffoni S, Gajda K, Muzzioli R, Peruzzini R, 94. Rodrigues PM, Macedo AL, Goodfellow BJ, Moura I, Moura JJ Winkelmann J (2015) Nat Chem Biol 11:772–778 (2006) J Biol Inorg Chem 11:307–315 128. Banci L, Brancaccio D, Ciofi-Baffoni S, Del Conte R, Gadepalli 95. Gray HB, Ellis WR Jr (1994) Electron transfer. In: Bertini I, R, Mikolajczyk M, Neri S, Piccioli M, Winkelmann J (2014) Gray HB, Lippard SJ, Valentine JS (eds) Bioinorganic chemistry. Proc Natl Acad Sci USA 111:6203–6208 University Science Books, Mill Valley, pp 315–363 129. Brancaccio D, Gallo A, Piccioli M, Novellino E, Ciofi-Baffoni 96. Jain NU, Tjioe E, Savidor A, Boulie J (2005) Biochemistry S, Banci L (2017) J Am Chem Soc 139:719–730 44:9067–9078 130. Keizers PH, Mersinli B, Reinle W, Donauer J, Hiruma Y, Hanne- 97. Bertini I, Cowan JA, Luchinat C, Natarajan K, Piccioli M (1997) mann F, Overhand M, Bernhardt R, Ubbink M (2010) Biochem- Biochemistry 36:9332–9339 istry 49:6846–6855 98. Cheng H, Westler WM, Xia B, Oh BH, Markley JL (1995) Arch 131. Cai K, Frederick RO, Kim JH, Reinen NM, Tonelli M, Markley Biochem Biophys 316:619–634 JL (2013) J Biol Chem 288:28755–28770 99. Banci L, Bertini I, Calderone V, Ciofi-Baffoni S, Giachetti A, 132. Cai K, Tonelli M, Frederick RO, Markley JL (2017) Biochemis- Jaiswal D, Mikolajczyk M, Piccioli M, Winkelmann J (2013) try 56:487–499 Proc Natl Acad Sci USA 110:7136–7141 133. Boniecki MT, Freibert SA, Muhlenhoff U, Lill R, Cygler M 100. Bermel W, Bertini I, Felli IC, Piccioli M, Pierattelli R (2006) (2017) Nat Commun 8:1287 Prog NMR Spectrosc 48:25–45 134. Gakh O, Ranatunga W, Smith DY, Ahlgren EC, Al-Karadaghi S, 101. Machonkin TE, Westler WM, Markley JL (2002) J Am Chem Thompson JR, Isaya G (2016) J Biol Chem 291:21296–21321 Soc 124:3204–3205 135. Cory SA, Van Vranken JG, Brignole EJ, Patra S, Winge DR, 102. Machonkin TE, Westler WM, Markley JL (2004) J Am Chem Drennan CL, Rutter J, Barondeau DP (2017) Proc Natl Acad Sci Soc 126:5413–5426 USA 114:E5325–E5334 103. Kostic M, Pochapsky SS, Pochapsky TC (2002) J Am Chem Soc 136. Muhlenhoff U, Gerber J, Richhardt N, Lill R (2003) EMBO J 124:9054–9055 22:4815–4825 104. Goodfellow BJ, Nunes SG, Rusnak F, Moura I, Ascenso 137. Uzarska MA, Dutkiewicz R, Freibert SA, Lill R, Muhlenhoff U C, Moura JJ, Volkman BF, Markley JL (2002) Protein Sci (2013) Mol Biol Cell 24:1830–1841 11:2464–2470 138. Bandyopadhyay S, Gama F, Molina-Navarro MM, Gualberto JM, 105. Zartler ER, Jenney FE Jr, Terrell M, Eidsness MK, Adams MW, Claxton R, Naik SG, Huynh BH, Herrero E, Jacquot JP, Johnson Prestegard JH (2001) Biochemistry 40:7279–7290 MK, Rouhier N (2008) EMBO J 27:1122–1133 106. Bertini I, Cavallaro G, Cosenza M, Kümmerle R, Luchinat C, 139. Rodriguez-Manzaneque MT, Tamarit J, Belli G, Ros J, Herrero Piccioli M, Poggi L (2002) J Biomol NMR 23:115–125 E (2002) Mol Biol Cell 13:1109–1121 107. Piccioli M, Turano P (2015) Coord Chem Rev 284:313–328 140. Muhlenho ff U, Richter N, Pines O, Pierik AJ, Lill R (2011) J Biol 108. Bertini I, Jiménez B, Piccioli M (2005) J Magn Reson Chem 286:41205–41216 174:125–132 141. Muhlenhoff U, Gerl MJ, Flauger B, Pirner HM, Balser S, Rich- 109. Ciofi-Baffoni S, Gallo A, Muzzioli R, Piccioli M (2014) J Biomol hardt N, Lill R, Stolz J (2007) Eucaryotic Cell 6:495–504 NMR 58:123–128 142. Sheftel AD, Wilbrecht C, Stehling O, Niggemeyer B, Elsasser 110. Zuiderweg ER (2002) Biochemistry 41:1–7 HP, Muhlenhoff U, Lill R (2012) Mol Biol Cell 23:1157–1166 1 3 684 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 143. Gelling C, Dawes IW, Richhardt N, Lill R, Muhlenhoff U 173. Ojeda L, Keller G, Muhlenhoff U, Rutherford JC, Lill R, Winge (2008) Mol Cell Biol 28:1851–1861 DR (2006) J Biol Chem 281:17661–17669 144. Song D, Tu Z, Lee FS (2009) J Biol Chem 284:35297–35307 174. Frey AG, Palenchar DJ, Wildemann JD, Philpott CC (2016) J 145. Tong WH, Jameson GN, Huynh BH, Rouault TA (2003) Proc Biol Chem 291:22344–22356 Natl Acad Sci USA 100:9762–9767 175. Hoffmann B, Uzarska MA, Berndt C, Godoy JR, Haunhorst P, 146. Sheftel AD, Stehling O, Pierik AJ, Netz DJ, Kerscher S, Lillig CH, Lill R, Muhlenhoff U (2011) Antioxid Redox Signal Elsasser HP, Wittig I, Balk J, Brandt U, Lill R (2009) Mol 15:19–30 Cell Biol 29:6059–6073 176. Li H, Outten CE (2012) Biochemistry 51:4377–4389 147. Melber A, Na U, Vashisht A, Weiler BD, Lill R, Wohlschlegel 177. Haunhorst P, Berndt C, Eitner S, Godoy JR, Lillig CH (2010) JA, Winge DR (2016) Elife 5:e15991 Biochem Biophys Res Commun 394:372–376 148. Blanc B, Gerez C, de Ollagnier CS (2015) Biochim Biophys 178. Saito Y, Shibayama H, Tanaka H, Tanimura A, Matsumura Acta 1853:1436–1447 I, Kanakura Y (2011) Biochem Biophys Res Commun 149. Shi Y, Ghosh M, Kovtunovych G, Crooks DR, Rouault TA 408:329–333 (2012) Biochim Biophys Acta 1823:484–492 179. Soler N, Delagoutte E, Miron S, Facca C, Baille D, d’Autreaux 150. Sheftel AD, Stehling O, Pierik AJ, Elsasser HP, Muhlenhoff U, B, Craescu G, Frapart YM, Mansuy D, Baldacci G, Huang ME, Webert H, Hobler A, Hannemann F, Bernhardt R, Lill R (2010) Vernis L (2011) Mol Microbiol 82:54–67 Proc Natl Acad Sci USA 107:11775–11780 180. Netz DJ, Stumpfig M, Dore C, Muhlenhoff U, Pierik AJ, Lill 151. Prudencio M, Ubbink M (2004) J Mol Recognit 17:524–539 R (2010) Nat Chem Biol 6:758–765 152. Gray HB, Winkler JR (2003) Q Rev Biophys 36:341–372 181. Banci L, Bertini I, Ciofi-Baffoni S, Boscaro F, Chatzi A, 153. Johansson C, Roos AK, Montano SJ, Sengupta R, Filippako- Mikolajczyk M, Tokatlidis K, Winkelmann J (2011) Chem poulos P, Guo K, von Delft F, Holmgren A, Oppermann U, Biol 18:794–804 Kavanagh KL (2011) Biochem J 433:303–311 182. Zhang Y, Yang C, Dancis A, Nakamaru-Ogiso E (2017) J Bio- 154. Bilder PW, Ding H, Newcomer ME (2004) Biochemistry chem 161:67–78 43:133–139 183. Netz DJ, Genau HM, Weiler BD, Bill E, Pierik AJ, Lill R 155. Cupp-Vickery JR, Silberg JJ, Ta DT, Vickery LE (2004) J Mol (2016) Biochem J 473:2073–2085 Biol 338:127–137 184. Li H, Mapolelo DT, Randeniya S, Johnson MK, Outten CE 156. Morimoto K, Yamashita E, Kondou Y, Lee SJ, Arisaka F, (2012) Biochemistry 51:1687–1696 Tsukihara T, Nakai M (2006) J Mol Biol 360:117–132 185. Vilella F, Alves R, Rodriguez-Manzaneque MT, Belli G, Swa- 157. Wada K, Hasegawa Y, Gong Z, Minami Y, Fukuyama K, Taka- minathan S, Sunnerhagen P, Herrero E (2004) Comp Funct hashi Y (2005) FEBS Lett 579:6543–6548 Genom 5:328–341 158. Beilschmidt LK, de Ollagnier CS, Fournier M, Sanakis I, 186. Huynen MA, Spronk CA, Gabaldon T, Snel B (2005) FEBS Hograindleur MA, Clemancey M, Blondin G, Schmucker S, Lett 579:591–596 Eisenmann A, Weiss A, Koebel P, Messaddeq N, Puccio H, 187. Zhou YB, Cao JB, Wan BB, Wang XR, Ding GH, Zhu H, Yang Martelli A (2017) Nat Commun 8:15124 HM, Wang KS, Zhang X, Han ZG (2008) Mol Cell Biochem 159. Agar JN, Krebs C, Frazzon J, Huynh BH, Dean DR, Johnson 317:61–68 MK (2000) Biochemistry 39:7856–7862 188. Kumanovics A, Chen OS, Li L, Bagley D, Adkins EM, Lin H, 160. Chandramouli K, Unciuleac MC, Naik S, Dean DR, Huynh BH, Dingra NN, Outten CE, Keller G, Winge D, Ward DM, Kaplan Johnson MK (2007) Biochemistry 46:6804–6811 J (2008) J Biol Chem 283:10276–10286 161. Jensen LT, Culotta VC (2000) Mol Cell Biol 20:3918–3927 189. Wachnowsky C, Fidai I, Cowan JA (2016) FEBS Lett 162. Kaut A, Lange H, Diekert K, Kispal G, Lill R (2000) J Biol 590:4531–4540 Chem 275:15955–15961 190. Li J, Cowan JA (2015) Chem Commun (Camb) 51:2253–2255 163. Andreini C, Banci L, Rosato A (2016) J Proteome Res 191. Qi W, Li J, Chain CY, Pasquevich GA, Pasquevich AF, Cowan 15:1308–1322 JA (2012) J Am Chem Soc 134:10745–10748 164. Stehling O, Netz DJ, Niggemeyer B, Rosser R, Eisenstein RS, 192. Qi W, Li J, Cowan JA (2014) Chem Commun (Camb) Puccio H, Pierik AJ, Lill R (2008) Mol Cell Biol 28:5517–5528 50:3795–3798 165. Roy A, Solodovnikova N, Nicholson T, Antholine W, Walden 193. Nuttle X, Giannuzzi G, Duyzend MH, Schraiber JG, Narvaiza WE (2003) EMBO J 22:4826–4835 I, Sudmant PH, Penn O, Chiatante G, Malig M, Huddleston 166. Netz DJ, Pierik AJ, Stumpfig M, Muhlenhoff U, Lill R (2007) J, Benner C, Camponeschi F, Ciofi-Baffoni S, Stessman HA, Nat Chem Biol 3:278–286 Marchetto MC, Denman L, Harshman L, Baker C, Raja A, 167. Stehling O, Vashisht AA, Mascarenhas J, Jonsson ZO, Sharma Penewit K, Janke N, Tang WJ, Ventura M, Banci L, Antonacci T, Netz DJ, Pierik AJ, Wohlschlegel JA, Lill R (2012) Science F, Akey JM, Amemiya CT, Gage FH, Reymond A, Eichler EE 337:195–199 (2016) Nature 536:205–209 168. Gari K, Leon Ortiz AM, Borel V, Flynn H, Skehel JM, Boulton 194. Lipper CH, Paddock ML, Onuchic JN, Mittler R, Nechushtai SJ (2012) Science 337:243–245 R, Jennings PA (2015) PLoS One 10:e0139699 169. Paul VD, Muhlenhoff U, Stumpfig M, Seebacher J, Kugler KG, 195. Ferecatu I, Goncalves S, Golinelli-Cohen MP, Clemancey M, Renicke C, Taxis C, Gavin AC, Pierik AJ, Lill R (2015) Elife Martelli A, Riquier S, Guittet E, Latour JM, Puccio H, Drapier 4:e08231 JC, Lescop E, Bouton C (2014) J Biol Chem 289:28070–28086 170. Stehling O, Mascarenhas J, Vashisht AA, Sheftel AD, Nigge- 196. Rouault TA, Klausner RD (1996) Trends Biochem Sci meyer B, Rosser R, Pierik AJ, Wohlschlegel JA, Lill R (2013) 21:174–177 Cell Metab 18:187–198 197. Zuris JA, Harir Y, Conlan AR, Shvartsman M, Michaeli D, 171. Muhlenhoff U, Molik S, Godoy JR, Uzarska MA, Richter N, Tamir S, Paddock ML, Onuchic JN, Mittler R, Cabantchik ZI, Seubert A, Zhang Y, Stubbe J, Pierrel F, Herrero E, Lillig CH, Jennings PA, Nechushtai R (2011) Proc Natl Acad Sci USA Lill R (2010) Cell Metab 12:373–385 108:13047–13052 172. Haunhorst P, Hanschmann EM, Brautigam L, Stehling O, Hoff- 198. Golinelli-Cohen MP, Lescop E, Mons C, Goncalves S, Clem- ancey M, Santolini J, Guittet E, Blondin G, Latour JM, Bouton mann B, Muhlenhoff U, Lill R, Berndt C, Lillig CH (2013) Mol C (2016) J Biol Chem 291:7583–7593 Biol Cell 24:1895–1903 1 3 JBIC Journal of Biological Inorganic Chemistry (2018) 23:665–685 685 199. Camponeschi F, Ciofi-Baffoni S, Banci L (2017) J Am Chem Soc 201. Xia B, Jenk D, LeMaster DM, Westler WM, Markley JL (2000) 139:9479–9482 Arch Biochem Biophys 373:328–334 200. Vernis L, Facca C, Delagoutte E, Soler N, Chanet R, Guiard B, Faye G, Baldacci G (2009) PLoS One 4:e4376 1 3
Journal
JBIC Journal of Biological Inorganic Chemistry – Springer Journals
Published: Mar 22, 2018