Mitochondria function in cytoplasmic FeS protein biogenesis

Andrew Dancis; Ashutosh K Pandey; Debkumar Pain

doi:10.1016/j.bbamcr.2024.119733

. Author manuscript; available in PMC: 2026 Jul 9.

Published in final edited form as: Biochim Biophys Acta Mol Cell Res. 2024 Apr 18;1871(5):119733. doi: 10.1016/j.bbamcr.2024.119733

Mitochondria function in cytoplasmic FeS protein biogenesis

Andrew Dancis ^1,^*, Ashutosh K Pandey ¹, Debkumar Pain ¹

PMCID: PMC13344148 NIHMSID: NIHMS2189715 PMID: 38641180

Abstract

Iron-sulfur (FeS) clusters are cofactors of numerous proteins involved in various essential functions including cellular respiration, protein translation, DNA synthesis and repair, ribosome maturation, anti-viral responses, and isopropylmalate isomerase activity. Novel FeS cluster proteins are still being discovered due to the widespread use of cryogenic electron microscopy (cryo-EM) and elegant genetic screens targeted at protein discovery. A complex sequence of biochemical reactions mediated by a conserved machinery controls biosynthesis of FeS clusters. In eukaryotes, a remarkable epistasis has been observed: the mitochondrial machinery, termed ISC (Iron-Sulfur Cluster), lies upstream of the cytoplasmic machinery, termed CIA (Cytoplasmic Iron-sulfur protein Assembly). The basis for this arrangement is the production of a hitherto uncharacterized intermediate, termed X-S or (Fe-S)_int, produced in mitochondria by the ISC machinery, exported by the mitochondrial ABC transporter Atm1 (ABC7 in humans), and then utilized by the CIA machinery for the cytoplasmic/nuclear FeS cluster assembly. Genetic and biochemical findings supporting this sequence of events are herein presented. New structural views of the Atm1 transport phases are reviewed. The key compartmental roles of glutathione in cellular FeS cluster biogenesis are highlighted. Finally, data are presented showing that every one of the ten core components of the mitochondrial ISC machinery and Atm1, when mutated or depleted, displays similar phenotypes: mitochondrial and cytoplasmic FeS clusters are both rendered deficient, consistent with the epistasis noted above.

Keywords: Mitochondria, FeS cluster trafficking, glutaredoxin, glutathione, Atm1, X-S, (Fe-S)_int, cytoplasm, FeS proteins

1. Introduction

Iron-sulfur (FeS) clusters are inorganic complexes of iron and sulfur with varying stoichiometries and redox states. In biology, the most commonly occurring are [2Fe-2S] and [4Fe-4S] clusters [1], and these are most often liganded to a protein backbone via cysteine residues that coordinate iron atoms in the complex. However, alternative ligations, such as histidine ligation of the iron in the Rieske protein [2], have also been described. In a proteomic study of human cells [3], fifty seven FeS proteins were identified, primarily in mitochondria and nuclear compartments. FeS clusters have also been found in the cytoplasm and along the external face of the endoplasmic reticulum. In terms of function, FeS proteins are essential for cellular respiration i.e., the reduction of oxygen to water by the electron transport chain in mitochondria, as respiratory complexes I, II, and III contain critical FeS cluster components. There are myriad biochemical steps that make use of FeS clusters, including the steps of the tricarboxylic acid cycle (e.g. aconitase), various biosynthetic processes (e.g. isopropylmalate isomerase in leucine biosynthesis in yeast), cytoplasmic protein translation and ribosomal maturation (e.g. Rli1), anti-viral response (e.g. the radical SAM enzyme viperin), and DNA synthesis, maintenance and repair (e.g. DNA polymerases, primases, and helicases) [2, 4, 5].

The number of cellular FeS proteins seems to be in flux. With more and more structural studies utilizing crystallography and cryo-EM, unanticipated FeS proteins have been discovered with high frequency. Recently in a cryo-EM study of mitochondrial ribosomes, three FeS proteins were discovered in association with the human mitochondrial small ribosome subunits [6]. These may serve a regulatory function, turning on and off mitochondrial translation in response to oxidative stress signals (off in response to greater oxidative stress) [7]. In recent work, a CRISPR/Cas9 knockout of the disease gene frataxin (Yfh1 in yeast) was performed. The knockout exhibited impaired mitochondrial protein translation. This phenotype was suppressed by overexpression of the [4Fe-4S] cluster protein METTL17 (Rsm22 in yeast). This newly discovered [4Fe-4S] cluster protein METTL17 was found to be associated with the small subunit of the mitochondrial ribosome and to function as an Fe-S cluster checkpoint for mitochondrial protein translation [8]. The compendium of cellular FeS clusters has been further enlarged by novel genetic methods designed to exhaustively identify all the FeS proteins in an organism [9]. According to one such method, E. coli was exposed to an iron chelator, leading to binding up the iron in all cellular FeS clusters, thereby exposing free cysteine residues that could be alkylated with fluorescein-linked indicators and later fished out and identified. This chemoproteomic strategy in E. coli identified two previously unannotated FeS proteins, TrhP and DppD [9]. In line with the theme of discovery of novel FeS clusters, Rouault and coworkers discovered FeS cofactors in the SARS-CoV-2 RNA dependent RNA polymerase that may serve as potential antiviral drug targets [10].

FeS clusters can be formed spontaneously in vitro by mixing high concentrations of ferrous iron with a chemical reductant (e.g. DTT) and sulfide under anaerobic conditions [11], but this is not what occurs in living cells. The high concentrations of ferrous iron and sulfide generally do not occur in cells, probably because of inherent redox toxicity to other components such as existing FeS clusters. In 1994, seminal work by Dennis Dean revealed that sulfur for nitrogenase originates from the amino acid cysteine and is mobilized by a cysteine desulfurase utilizing PLP (pyridoxal phosphate) chemistry to generate a persulfide at the active site of the enzyme. The persulfide sulfur is then transferred to a scaffold protein as part of a biological process of FeS cluster assembly [12]. This enzymatic process described by Dennis Dean was the first glimpse of the larger theme of biologically catalyzed FeS cluster assembly. The process is evolutionarily ancient and pervasive in all kingdoms, and the biogenesis utilizes similar conserved components and processes [13]. In a nutshell, a cysteine desulfurase acts on the amino acid cysteine to generate persulfide sulfur, which is transferred to an Isu type scaffold. Iron and biological reductants (e.g., ferredoxin/ferredoxin reductase redox couple) are needed as well as specialized chaperones and targeting/transfer factors, ultimately delivering FeS clusters to apoproteins in need of their cofactors [2].

In eukaryotic cells, FeS clusters exist in mitochondria and various other locations including cytoplasm, nucleus, and the outer face of the endoplasmic reticulum. The FeS cluster biosynthesis pathway is likewise partitioned into mitochondrial and extramitochondrial portions (Fig. 1). A curious epistasis has been described according to which mitochondrial FeS cluster biogenesis lies upstream of cytoplasmic/nuclear FeS cluster synthesis [14]. Mitochondria by themselves possess all the components needed to make FeS clusters. For example, isolated mitochondria fed with iron (as ferrous ascorbate) and sulfur (as cysteine) can synthesize new FeS clusters on apo-aconitase or imported apo-ferredoxin, for example [15]. The machinery for this process, termed ISC for Iron-Sulfur Cluster biogenesis machinery, is complete and functional. However, a similar experiment performed with isolated cytoplasm does not work. Something is missing. What is missing is a mitochondrial contribution, which has been called X-S or (Fe-S)_int. Thus, isolated cytoplasm supplemented with isolated mitochondria now permits FeS cluster synthesis on apoproteins such as bacterial expressed and purified apo-isopropylmalate isomerase (Leu1^R) or a cytosolic form of apo-ferredoxin (ΔNYah1) [16] [17]. The machinery that mediates this process is termed CIA for Cytoplasmic Iron-sulfur protein Assembly machinery (Fig. 1). The precise chemical nature of the mitochondrial synthesized and exported component/intermediate required for extramitochondrial FeS cluster biogenesis has not been solved to date, although this is a subject of active research. However, using various procedures described below, the pathways for production, export, and cytoplasmic utilization of the (Fe-S)_int intermediate have been roughly mapped out. Similarly, for S_int (the intermediate for tRNA thiolation), the overlapping but distinct pathways have been defined and will be discussed in a later section (see section 12).

2. Mitochondrial FeS cluster assembly

This biosynthetic process can be viewed as taking place in steps (Fig. 1) [1]. First, the Nfs1 cysteine desulfurase uses PLP (pyridoxal phosphate) chemistry to act on the amino acid substrate cysteine to generate a persulfide on the active site cysteine (Cys⁴²¹, numbering according to the precursor form of the yeast enzyme). The persulfide sulfur is transferred to a critical cysteine residue on the Isu scaffold (Cys¹³⁹, numbering according to the precursor form of the yeast protein) concurrent with iron insertion, sulfur reduction, and formation of a [2Fe-2S] cluster intermediate. Subsequently, Hsp70/DnaJ chaperones mediate dislocation of the [2Fe-2S] cluster intermediate, transferring it to Grx5. Grx5 may then function as a distributor of FeS cluster intermediates, and here the pathway branches. The Grx5 [2Fe-2S]-2GSH intermediate might provide [2Fe-2S] clusters directly to other recipients. Alternatively, the Grx5 intermediate can donate precursors to the downstream mitochondrial ISA machinery that forms [4Fe-4S] clusters by reductive coupling. The [4Fe-4S] clusters may then be inserted directly into some apoproteins like aconitase (Aco1). For other proteins such as respiratory complexes I and II, and lipoate synthase, the [4Fe-4S] insertion may occur via additional late-acting targeting factors [1]. Also, at the Grx5 branch point, the (Fe-S)_int intermediate (see below) may be transferred to the Atm1 transporter for export to the intermembrane space of mitochondria (IMS) and the cytoplasm (Fig. 1). In the cytoplasm, the (Fe-S)_int intermediate may contact the Grx3/Grx4/Bol2 complex, thereby entering the CIA pathway, which includes its own reductase, scaffold, and CTC (CIA Targeting Complex) steps before reaching the destination in recipient apoproteins in the cytoplasm and nucleus [2]. In summary, the role of mitochondria in cytoplasmic FeS cluster assembly involves ISC steps (up to Grx5) for formation of the (Fe-S)_int intermediate, followed by export steps (Atm1, GSH). The involvement of each component in these processes has been demonstrated by several lines of evidence: knockout or depletion of the component or combination of components leads to failure to make cytoplasmic FeS clusters. Additionally, the knockout or depletion strain often exhibits iron homeostatic abnormalities consisting of constitutive cellular iron uptake and mitochondrial iron accumulation [18]. In the following section, we will describe mainly the yeast components, although the homologous mammalian components and pathways may also be touched on.

2.1. Synthesis of new [2Fe-2S] clusters on the Isu scaffold

Structural information on a mitochondrial FeS cluster assembly complex has been gathered from various types of studies, including X-ray crystallography, cryo-EM, small angle X-ray scattering (SAXS), nuclear magnetic resonance (NMR), and cross linking analyses [19–22]. The complex consists of core elements with the Nfs1/Nfs1 dimer and Acp1/Isd11 at its center. The accessory subunits Yfh1, Yah1, and Isu1/2 bind to the Nfs1 portion of the dimer and interact with the complex and with each other. Notably, it is only the reduced form of Yah1 (Yah1_red) that interacts with the complex; it binds, delivers electrons, and then comes off. The Nfs1 enzyme binds the substrate cysteine in a substrate binding site containing pyridoxal phosphate (PLP). Using a Schiff base intermediate [12], the Nfs1 enzyme forms a persulfide (-SSH) at the catalytic Cys⁴²¹ on a specialized protein loop. The loop then moves, like a construction crane, from the catalytic site of Nfs1 to the surface of the complex and then towards the cysteine rich core of Isu1/2. This movement is necessary for delivery of the persulfide sulfur from Nfs1 to Isu1/2. Iron (Fe²⁺) from an unclear source enters the Isu1/2 core, perhaps displacing a Zn atom that resides there as a structural stabilizer [23]. The persulfide sulfur (-SSH) bound on the Nfs1 active site loop is moved into position to be transferred to Cys¹³⁹ on Isu. During this process frataxin functions as a guiding chaperone, binding close to the loop and the target Isu cysteine, and stabilizing the interactions. Frataxin also enhances the persulfide sulfur transfer directly as demonstrated by elegant alkylation assays using reconstituted component mixtures [23]. The reduced mitochondrial ferredoxin (Yah1_red) bound to the tips of the complex delivers electrons to the persulfide sulfur on Isu (Cys¹³⁹ of Isu1), reducing it to S^2-, the requisite valence for new FeS cluster formation. At this point the bound FeS cluster on Isu, must be converted to a [2Fe-2S] cluster, perhaps via an Isu dimerization process. Yah1, now oxidized after delivering its electrons, leaves the complex and the Isu1/2, with [2Fe-2S] cluster segregated on one subunit, also leaves the complex [1].

2.2. Nfs1 accessory proteins - Isd11 and Acp1

Isd11 and Acp1 chiefly contribute a structural stabilization function to the Nfs1 FeS cluster assembly complex [24], although some data indicate an additional role in activating the enzymatic function of Nfs1 [25]. The structural stabilization function of Acp1 is enhanced in response to environmental signals. Under conditions of high energy supply, the mitochondrial FASII (fatty acid synthase II) machinery synthesizes longer lipid side chains, including the phosphopantetheinyl (acyl-PP) cofactor covalently bound to Ser³⁶ of Acp1 and penetrating the three-helix bundle of Isd11. The longer chain fatty acids on Isd11 stabilize the interaction of Isd11 with the more negatively charged Acp1 via the LYRM motif, thereby enhancing FeS cluster assembly activity of the complex. The C14, C16 and C18 3-ketoacyl chains on Acp1 are implicated in binding to Isd11 and stabilization of the interaction [26]. We presume that under conditions of low energy supply, shorter acyl-PP chains are made, and the complex is destabilized, leading to decreased FeS cluster assembly activity. The FeS cluster assembly activity is thus coupled to cellular energy availability [1, 27]

The various steps in the synthesis of [2Fe-2S] clusters on the Isu scaffold have been reproduced in reconstitution reactions with purified components from various species including yeast [28], human [23], or microsporidia [29]. Hypomorphic or loss-of-function mutations have been studied for Nfs1, Isd11, Yfh1, Yah1, and Isu1/2. These mutations interfere with formation of the FeS cluster intermediate on Isu1/2 and the intermediate on Grx5. In these mutants, the (Fe-S)_int presumably is not formed or is formed in decreased amounts, and downstream FeS clusters in the cytoplasm are deficient. These mutants all exhibit an iron regulatory phenotype [30], characterized by activated cellular iron uptake and mitochondrial iron accumulation. The iron regulatory phenotype correlates with lack of formation of the (Fe-S)_int intermediate.

In terms of Acp1, its role in mitochondrial FeS cluster assembly and iron regulation indicates that it is a bona fide core component of the ISC complex [27]. In Acp1-depleted cells, aconitase protein, activity, and FeS cluster loading are deficient. All lipoic acid-containing proteins are markedly deficient. Nfs1 and Isd11 protein levels are decreased. Nfs1 is destabilized, which is the probable mechanism by which Acp1 depletion causes FeS cluster deficiencies. In terms of cytoplasmic FeS cluster assembly, Acp1-depleted cells show defective ΔNYah1 (a cytosolic form of ferredoxin) loading and loss of Leu1 activity in the cytoplasm, thus confirming the prediction that core mitochondrial ISC components are required for cytoplasmic FeS clusters (Pandey, A. et al., manuscript in preparation).

2.3. Chaperone cycle

The action of the core FeS cluster synthesis machine including Nfs1 dimers is followed by the action of dedicated chaperones, which serve to dislodge the [2Fe-2S] cluster intermediate from its binding site on the Isu scaffold and transfer it to the Grx5-GSH complex [31]. The first step is to free up the Isu [2Fe-2S] cluster containing subunit from entanglements in the ISC assembly complex. Jac1, the specialized Hsp40/DnaJ related chaperone, binds to a hydrophobic segment on the holo-Isu protein consisting of Leu⁶³, Val⁷², and Phe⁹⁴ residues (precursor form of yeast Isu1 numbering), subsequently transferring the cargo to Ssq1. Ssq1 is a specialized Hsp70 homolog, adapted for FeS cluster assembly in mitochondria, which binds the Isu motif ¹³²LPPVK¹³⁶ in the Hsp70 substrate binding cleft. The binding of Jac1 and Ssq1 activates a switch because the binding sites on Isu holoprotein are initially occluded by Nfs1-Yfh1 binding within the Nfs1 dodecameric machine [32]. The binding springs the Isu1-[2Fe-2S] dimer free from the complex, at the same time initiating the chaperone cycle. Jac1 delivers the holo-Isu1 to Ssq1, and binding of ¹³²LPPVK¹³⁶ in the peptide binding cleft triggers ATP hydrolysis and a major conformational change of the chaperone, locking in the Isu1 peptide segment binding but loosening the binding of Isu1 for its [2Fe-2S] cargo [33]. Next, Grx5 binds to Ssq1 of the chaperone complex, at a site remote from bound Isu1, and Jac1 is removed, creating a ternary complex of Isu1-[2Fe-2S]-Ssq1-Grx5. In the final step in the chaperone cycle, ADP bound to Ssq1 is recycled and replaced with fresh ATP, setting up the chaperone for another cycle of binding and release. The recycling factor that performs this step is Mge1, the same recycling factor that acts on Ssc1, the major Hsp70 involved in mitochondrial protein import and folding [34]. The [2Fe-2S] cluster intermediate is then bound to Grx5, where it is coordinated by 2 cysteine residues from the CGFS motif on Grx5 and 2 molecules of glutathione [35]. The chaperone components (Jac1, Ssq1, and Mge1) and the monothiol glutaredoxin Grx5 constitute core elements of the ISC machinery, and as such are required for synthesis of the (Fe-S)_int, which is transported by Atm1. Interference with their functions leads to lack of (Fe-S)_int and FeS cluster deficiency in the cytoplasm [36]. Concomitant with this defect, the iron regulatory phenotype is engendered, i.e., constitutive cellular iron uptake and iron accumulation in mitochondria.

2.4. Mitochondrial generation of [4Fe-4S] clusters and their target specific insertion

A dedicated machinery exists, branching off from Grx5, and passing through Isa1/Isa2/Iba57 for synthesis of mitochondrial [4Fe-4S] clusters. As drawn, this pathway branch (see Fig. 1 arrows emanating from Grx5) is necessary for biosynthesis of [4Fe-4S] clusters in mitochondria such as aconitase, complex I of the respiratory chain (not in S. cerevisiae but in other yeasts such as Yarrowia lipolytica, and in humans), lipoic acid synthase (LIAS) and dependent lipoic acid-containing proteins. Two major steps make up this branch of the pathway: formation of the [4Fe-4S] clusters in mitochondria and dedicated targeting of the clusters. In the first step, Grx5 delivers a [2Fe-2S] cluster intermediate to a protein complex of Isa1, Isa2, and Iba57. These components physically interact with each other and with Grx5. The human homologs, GLRX5 and ISCA1, have also been shown to interact by NMR and to accommodate transfer of two [2Fe-2S] clusters and formation of a [4Fe-4S] cluster by reductive coupling in the presence of high concentrations of a chemical reductant DTT [37]. The physiologic situation may be similar but not identical. A reconstituted system has been established for the human proteins [38], with GLRX5 as the [2Fe-2S] cluster donor and mitochondrial aconitase as the target [4Fe-4S] acceptor. ISCA1-ISCA2-IBA57 components forming a protein complex were necessary for aconitase activation. Notably, in this reconstituted system, electrons were not provided by a chemical reductant such as DTT. Rather, a physiologic electron transport chain consisting of NADPH-FDXR-FDX2 (equivalent to NADPH-Arh1-Yah1 in yeast) was needed for reductive coupling of 2 × [2Fe-2S] clusters on the ISCA1-ISCA2-IBA57 complex [38]. Mechanistic details of this transfer and coupling process remain to be worked out. In the second major step of the branch pathway, the [4Fe-4S] cluster intermediate must reach its final destination in target proteins in mitochondria, and for this purpose there exist dedicated targeting factors - these “touch” the donor ISCA (Isa in yeast) complex and recipient [4Fe-4S] cluster apoproteins in mitochondria, thereby transferring the [4Fe-4S] cargo. The IND1 factor is dedicated to synthesis of the respiratory complex I [4Fe-4S] clusters (no complex I exists in Saccharomyces cerevisiae). NFU1 (Nfu1 in yeast) has a minor role in [4Fe-4S] cluster loading of aconitase and SDH2, but a major role in conjunction with BOLA3 (Bol3 in yeast) in formation of the 2 × [4Fe-4S] clusters in LIAS [2]. The mechanistic details of these [4Fe-4S] cluster transfers remain to be worked out.

3. Cytoplasmic and nuclear FeS cluster assembly

Grx5 with its [2Fe-2S]-glutathione intermediate plays a vital role for FeS cluster synthesis not only in mitochondria but also outside mitochondria (Fig. 1). The Grx5 liganded cluster already has glutathione bound to it, and the Atm1 substrate also likely contains a GSH ligand(s) (see below). If the Atm1 transport substrate is a [2Fe-2S](GS)₄ compound as contended by Cowan et al. [39], perhaps direct interaction with the Atm1 substrate binding cavity will be sufficient for transferring this intermediate. Alternatively, new intermediates and targeting components that link Grx5 and Atm1 may remain to be discovered and defined. The (Fe-S)_int intermediate is transported out from mitochondria via the activity of the Atm1 transporter and its ATP-dependent transport cycle [39]. As (Fe-S)_int exits mitochondria and enters the cytoplasm, which CIA component(s) interact with it first? Unfortunately, no experiments and no data address this point. In prior review models [2], the intermediate exiting from mitochondria via ABCB7 (Atm1) was proposed to initially interact with GLRX3/BOLA2 (Grx3/Grx4/Bol2) or CFD1/NBP35 (Cfd1/Nbp35) complex in the cytoplasm. The role of Grx3/Grx4/Bol2 as the initial (Fe-S)_int target seems more likely to us, given the overlap of the (Fe-S)_int transport deficient phenotypes and iron homeostatic abnormalities. Grx3/Grx4 and/or Bol2 mutants confer strong iron homeostatic abnormalities, whereas Cfd1/Nbp35 mutants or depletion experiments have little effect on iron homeostasis [40]. In any case, initial cytoplasmic interactors of mitochondria-exported (Fe-S)_int in the cytoplasm remain to be experimentally determined.

The CIA FeS cluster assembly pathway consists of multiple components, which are distinct from the ISC counterparts but bear overall similarity in terms of functions [41]. Like the mitochondrial ISC machinery, there are scaffold, reductase and targeting components. The scaffold consists of the Cfd1/Nbp35 heterodimer, which resembles P-loop NTPases [42]. The heterodimers exhibit ATPase and GTPase activity, although the role of these activities in FeS cluster assembly remains incompletely defined [43]. The new FeS clusters are formed as bridging [4Fe-4S] cluster complexes bound by exposed Cys residues on Cfd1 and Nbp35. Key components needed for this step to occur are the Grx3/Grx4/Bol2 complex, which might grab and/or process exported (Fe-S)_int and then transfer FeS constituents to the nascent cluster on the scaffold (see above) [18]. Also, electrons are needed, perhaps for reductive coupling of the nascent clusters on Cfd1/Nbp35. A specialized electron transport chain serves this function, transferring electrons from cytoplasmic NADPH to the Tah18 flavin cofactor and from there to FeS clusters on Dre2 [44]. Dre2 contains one [2Fe-2S] and one [4Fe-4S] cluster [45] and may be reduced by Tah18. Reduced Dre2 then likely reduces FeS cluster intermediates on Cfd1/Nbp35. These newly formed [4Fe-4S] cluster intermediates of the CIA are subsequently transferred/dislocated, delivering them to specific targets via a Nar1 intermediate and the CIA targeting complex (CTC). Nar1 is related to hydrogenases but has lost the critical residues for this function and in fact does not have a clearly defined function [30, 46]. It can be placed in an intermediate position in the CIA pathway as judged by protein-protein interactions with both early and late components. Likewise, experiments with ⁵⁵Fe labeling of Nar1 show dependence on the early and not the late CIA components [30, 46]. The final CTC components (Cia1, Cia2, Mms19) are defined by their ability to bind FeS cluster intermediates and also to physically interact with various specific target proteins. Only Cia2 of the CTC may bind and transfer FeS cluster intermediates due to its conserved reactive cysteine; Cia1 and Mms19 rather fulfill tasks in protein-protein interactions. In this manner, the CTC functions in biogenesis of cytoplasmic FeS proteins such as Leu1, nuclear factors such as RNA primase, and translation factors such as Rli1, and methionine synthesis components [2]. In some sense, the entire CIA and its client proteins are subservient and downstream of the mitochondria and mitochondrial export of the (Fe-S)_int. A clever set of experiments was performed in which Lill and co-workers redirected ISC components to the cytosol by removing the mitochondrial targeting signals. These proteins remained in the cytoplasm, but cytoplasmic FeS cluster deficiencies were not rescued (in the absence of mitochondrial localization of the ISC components) [47]. Thus, the ISC components must reside in the mitochondria in order to function upstream of the CIA. Perhaps relocalization of ISC components to a cytoplasmic localization results in failure to form the vital (Fe-S)_int intermediate (Fig. 1).

4. Mitochondrial exporter Atm1

In 1995, Leighton and Schatz utilized degenerate primers to amplify regions from the ATP binding segments of proteins in the entire yeast genome [48]. They cloned five such proteins, and following gene disruption, they found that one of the disruptants grew very poorly, even on rich medium. They called the gene ATM1 for ATP-dependent Transporter of Mitochondria. The N-terminus of the full length Atm1 protein was found to contain a mitochondrial targeting signal. A myc epitope tag was placed at the C-terminus of the Atm1 coding sequence, and the tagged protein was found to be enriched in purified mitochondria, again confirming that Atm1 is a mitochondrial protein. Careful topology mapping was performed by gauging the protease sensitivity of the C-terminal myc tag. Mitochondria isolated from an Atm1-myc expressing strain were exposed to proteinase K under conditions that rupture the outer membrane, leaving the inner membrane intact. The C-terminal myc tag remained protected in the matrix compartment. This C-terminal domain carries the ABC homology/ATPase region. A putative substrate binding domain was found adjacent to the ABC domain in the linear sequence. Thus, this substrate binding domain of the protein likely also lies to the matrix side of the membrane insertion segments of the protein. In sum, the data implied that the substrate transport direction for the Atm1 transporter was from inside (matrix) to outside (IMS, cytoplasm) [48]. The transport substrate could not be discerned. The slow growth of the atm1 mutant suggested that the substrate was something important for growth. We now believe that the Atm1 substrate is necessary for cytoplasmic FeS cluster formation, an essential cellular process. The atm1 mutants were also noted to rapidly convert to a rho minus state, suggesting that they were sustaining mtDNA damage. These mutants accumulate huge quantities of iron in mitochondria in the form of toxic phosphate nanoparticles [49], which may convert rho plus to rho minus strains over time due to free radical iron and oxygen induced DNA toxicity. The mechanism of mitochondrial iron accumulation in atm1 mutants is still not entirely clear but the phenotypes ensue when Atm1 function is inactivated [50] (Fig. 2A).

Fig. 2. — **A) Atm1 transport direction**. Atm1 (ABCB7 in humans) is an ATP binding cassette transporter in the mitochondrial inner membrane, with ATP binding sites and substrate (star) binding site oriented toward the mitochondrial matrix. Thus, the presumed transport direction for the substrate is from “inside” in the mitochondrial matrix to “outside” in the intermembrane space, which communicates with the cytoplasm [48]. **B) Compartmentalization of leucine biosynthesis**. The leucine biosynthetic pathway is subdivided into mitochondrial and cytoplasmic arms, connected by 3 transporters: Leu5 for CoA transport, OAC (oxaloacetate carrier) for alpha-IPM transport, and Atm1 for unknown intermediate transport. All 3 transporters are necessary for the Leu1 FeS cluster protein activity in the cytoplasm [51]. **C) Gal-Atm1**↓ **phenotype**. The depletion of Atm1 via a promoter swap strain generates a compartmental deficiency of FeS clusters. Aconitase a mitochondrial FeS cluster protein, is active, whereas Leu1, a cytoplasmic FeS cluster protein is inactive [14]. The standard error of the determination of enzyme activities was between 5 and 15% [14]. Hypothesis regarding mitochondrial transport intermediate. Mitochondrial ISC synthesizes an intermediate, called X-S or (Fe-S)_int, which is exported by Atm1 from mitochondria to cytoplasm, and there functions to mediate synthesis of cytoplasmic and nuclear FeS clusters [54].

4.1. Leu1 biosynthesis and dependence on Atm1

The next step in the definition of Atm1’s export function came with the fleshing out of the leucine biosynthetic pathway by Gunter Kohlhaw [51]. He not only defined the enzymes involved in leucine biosynthesis but also identified their cellular localizations and compartmental phenotypes. Studying various yeast mutants with leucine auxotrophy (i.e., dependence on exogenous leucine for growth), he was able to order the enzymes and define their biochemical correlates [51]. He found that the initial leucine biosynthetic steps by Leu4 and Leu9 take place inside mitochondria, whereas downstream steps, including the isomerase Leu1 and dehydratase Leu2, localize to the cytoplasm. This compartmentalization implied the existence of transporters moving functional units in and out of mitochondria.

The most interesting features of the leucine biosynthetic pathway are the mitochondrial transporters involved. There are three such transporters: Leu5, Oac1, and Atm1. Each one when mutated generates leucine auxotrophy. Leu5 encodes the CoA importer for mitochondria, and thus CoA levels in leu5 mutant mitochondria are decreased by 15-fold compared with cytoplasm. CoA is needed for making acetyl CoA, a cofactor for the enzymes Leu4 and Leu9 that mediate synthesis of alpha isopropylmalate (IPM), and the leucine auxotrophy of the leu5 mutant results from CoA deficiency within mitochondria [52]. Oac1 is a member of the mitochondrial carrier protein family, with multiple substrate specificities. Kohlhaw depicted its identity as an “unknown” in his published scheme for leucine biosynthesis [51], but subsequent work has identified this “unknown” as the oxaloacetate carrier [53]. The reconstituted oxaloacetate carrier, Oac1, efficiently transports alpha IPM in addition to its known substrates, oxaloacetate, sulfate, and malonate. Transport is saturable with half saturation constant of 75 μM for alpha IPM and 0.31 mM for beta IPM. In contrast, leucine, alpha ketoisovalerate, valine, or isoleucine were not transported. Cells lacking Oac1 were relatively growth inhibited in the absence of leucine, but they grew optimally on fermentable carbon sources in the presence of leucine. Oac1 is important for leucine biosynthesis on fermentable carbon sources, catalyzing the export of alpha IPM from mitochondria to cytoplasm, probably in exchange for oxaloacetate [53].

The third mitochondrial transporter that generates leucine auxotrophy when mutated is Atm1. Since the directionality of Atm1’s transport function was known (export occurs from mitochondria towards the cytoplasm), it was speculated that Atm1 could be exporting a substrate necessary for the cytoplasmic portions of the leucine biosynthetic pathway. Kohlhaw found that Leu1 was particularly labile when removed from its native context but could be stabilized by high glycerol, low oxygen, and high ammonium sulfate concentrations [51]. It is now appreciated that these properties were largely due to the lability of its [4Fe-4S] cluster cofactor. In sum, Atm1-mediated export from mitochondria was speculated to play a role in stabilizing the Leu1 cofactor, which was subsequently found to be a cytoplasmic FeS cluster (Fig. 2B).

4.2. Atm1 and X-S or (Fe-S)_int hypothesis

The key steps in the development of the X-S or (Fe-S)_int hypothesis came from experiments published by Lill and coworkers in 1999. Atm1-depleted (Gal-Atm1↓) cells were found to exhibit a compartmental FeS cluster deficiency phenotype, with deficient levels of cytoplasmic FeS proteins such as Leu1, and normal levels of mitochondrial FeS proteins including aconitase and succinate dehydrogenase [14]. In addition, not only was the Leu1 enzyme activity in the cytoplasm markedly decreased to undetectable levels, but ⁵⁵Fe labeling of the newly formed FeS clusters, and recovery of the radiolabeled protein by immunoprecipitation, revealed that new Leu1 clusters were not being synthesized in the Atm1-depleted cells (Fig. 2C).

By contrast, Nfs1-depleted (Gal-Nfs1↓) cells behaved differently. Nfs1 depletion induced deficiencies of both mitochondrial and cytoplasmic FeS protein activities, thus a global deficit was engendered. Nfs1 is the sole cysteine desulfurase for the yeast cell and is necessary for biogenesis of all FeS clusters. However, the picture is confounded by the localization of Nfs1. The enzyme is encoded by a nuclear gene NFS1 with a mitochondrial targeting signal (MTS), and as such it is imported into mitochondria and processed to its mature form. How can the mitochondrial Nfs1 be required for cytoplasmic FeS clusters? A hypothesis taking account of the Atm1 and Nfs1 phenotypes was formed. The idea was proposed that Nfs1 together with other ISC components synthesize an intermediate X-S that is exported via Atm1 to the cytoplasm, where it mediates FeS cluster assembly. The X-S could be an FeS cluster component or a signaling component permissive for making FeS cluster proteins [54].

Some objections to this idea were raised. Nfs1 protein has been shown by various evidence to be predominantly mitochondrial: immunoblotting with anti-Nfs1 antisera convincingly gives a major signal in the mitochondrial matrix [55]. The Nfs1 precursor protein has a mitochondrial targeting sequence and the radiolabeled Nfs1 precursor can be imported efficiently into isolated mitochondria, confirming a predominantly mitochondrial localization [25]. However, evidence for the existence of a small cytoplasmic and/or nuclear pool of Nfs1 has also been presented [56, 57]. The nuclear localization signal for Nfs1 is apparently essential for viability [58]. In mammalian cells, an alternative translational start site eliminating the MTS leads to cytoplasmic localization of some Nfs1 in those cells [59]. Perhaps the extramitochondrial Nfs1 could be responsible for FeS cluster synthesis in the cytoplasm, essentially bypassing the need for Atm1 transport. However, interesting experiments refuting this possibility were performed by Lill and co-workers in which the MTS of Nfs1 was removed, thereby deliberately targeting the protein to the cytoplasm in yeast. These constructs failed to rescue the Δatm1 phenotype for compartmental/cytoplasmic FeS cluster deficiency [56]. Another potential objection to the above model is that Δatm1 or Gal-Atm1↓ mutants in some experiments exhibited deficiencies in mitochondrial FeS cluster activities. We ourselves have observed decreased aconitase activity in the mitochondria of some Δatm1 strains grown aerobically. The explanation for these effects, however, is likely to be that Δatm1 accumulates huge quantities of iron in mitochondria, which are toxic and elicit secondary oxidative phenotypes. The aconitase deficiency in the Δatm1 mutant strain is likely a secondary effect of FeS cluster destruction due to toxic iron accumulation. An experiment that supports this notion was performed by Lindahl and co-workers [49]. They grew Gal-Atm1↓ (Atm1 depleted) cells strictly anaerobically and found that the mutant cells do not accumulate mitochondrial iron under these conditions. Furthermore, aconitase activity in these cells was normal under anaerobic conditions, confirming the compartmental nature of FeS cluster proteins in these Atm1 depleted cells - mitochondria normal, cytoplasm deficient.

5. Mitochondria requirement for cytoplasmic FeS cluster assembly – Biochemical assays

Both mitochondria and cytosol together, but neither individually, were sufficient for supporting FeS cluster formation on a cytoplasmic apoprotein substrate. Assays were developed to directly monitor insertion of newly formed and radiolabeled Fe-³⁵S clusters into apoproteins. In this assay, isolated mitochondria (“mito”) isolated from wild-type (WT) cells were mixed with [³⁵S]cysteine (sulfur source) and unlabeled ferrous ascorbate (iron source), and nucleotides (ATP, GTP, and NADH). Isolated cytoplasm (“cyto”) was added to the mixture and incubated with the addition of a purified apoprotein substrate. The apoprotein substrate was made by expressing an N-terminally truncated form of yeast mitochondrial ferredoxin in E. coli and purifying the protein in the apo form (apo-ΔNYah1). After incubation, the reaction mixture was centrifuged to separate the pellet (P) containing mitochondria and supernatant (S) containing cytoplasm. Samples were analyzed by native polyacrylamide gel electrophoresis followed by autoradiography to directly visualize radiolabeled products. As shown in Fig. 3A, WT mito alone (no cyto) gave a signal for the [4Fe-4³⁵S] cluster of aconitase (lane 1). This signal represents the endogenous aconitase protein that was radiolabeled during the assay. It seems a small portion (~3–4%) of endogenous aconitase is present in apo-form in WT cells under ordinary growth conditions. The labeling depends on the mitochondrial ISC machinery [15].

WT mito (no cyto), however, cannot make an FeS cluster on apo-ferredoxin outside mitochondria (Fig. 3A, lanes 9 and 10). Similarly, cyto alone mixed with the apo-ferredoxin substrate and the assay constituents yielded a blank lane (lane 7), because cyto alone cannot synthesize FeS clusters without additional help. Only when mito and cyto were mixed together, did we see labeling of the ferredoxin (ΔNYah1) cluster (Fig. 3A, lane 6). In summary, both mitochondria and cytoplasm were required for cytoplasmic FeS cluster synthesis on ΔNYah1 apo-ferredoxin substrate. Note that in the lanes with cyto and mito together, an additional signal appeared for radiolabeled thiolated tRNA (arrowhead in Fig. 3A) [17, 60]. Two distinct intermediates are likely generated by mitochondria and subsequently exported to the cytoplasm: one for tRNA thiolation (called S_int) and another for Fe-S cluster assembly (called (Fe-S)_int) in the cytosol [17]. The S_int and (Fe-S)_int intermediates are two distinct species in terms of their productions and activities, and yet both are exported by Atm1. Once exported, S_int is utilized for tRNA thiolation by a sulfur relay in the cytosol. On the other hand, exported (Fe-S)_int is used by the CIA machinery for cytosolic Fe-S cluster assembly. Most likely, the (Fe-S)_int intermediate exported from mitochondria provides both iron and sulfur species required for cytosolic Fe-S cluster assembly [61]. These issues will be further discussed in section 6 below.

More recently, a different biochemical assay was developed to further study mitochondria-cytoplasm interactions in terms of the intermediate (Fe-S)_int, using the Leu1 apoprotein (apo-Leu1^R) as an indicator for [4Fe-4S] cluster assembly in the cytoplasm. This time, the cytoplasmic FeS cluster assembly was monitored through reconstitution of Leu1 enzyme activity, and the results were similar. As indicated above, in S. cerevisiae, Leu1 [4Fe-4S] is the cytoplasmic isopropylmalate isomerase involved in leucine biosynthesis. Cells lacking the gene for Leu1 (Δleu1) were permeabilized and briefly centrifuged, separating crude mitochondria in the pellet (“PC pellet”) from the crude cytoplasm in the supernatant (“PC sup”) fractions. Both these fractions were needed together for [4Fe-4S] cluster assembly on added apo-Leu1^R with concomitant enzyme activation (Fig. 3B). The components, incubated separately with the substrate, gave virtually no activity. In this case, the apo-Leu1^R substrate was produced in E. coli with a His₆ tag and purified on a Ni-NTA matrix. As purified, it exhibited negligible activity in a spectrophotometric assay for the isomerization intermediate, which absorbs at 235 nm. If mixed with permeabilized cell supernatant (“PC sup”), a surrogate for crude cytoplasm, no activity was generated (Fig. 3B, lane 6). If mixed with permeabilized cell pellet (“PC pellet”), a surrogate for crude mitochondria, a small amount of basal activity was detected (lane 5). This might be because the mitochondria contain a complete FeS cluster assembly system, and so a small amount of mitochondrial rupture could bring this ISC system into contact with the extramitochondrial apo-Leu1^R substrate, leading to some background reconstituted enzyme activity. However, this was only 1.5 units of the activity, compared with 8 units of activity generated by mixing mito (“PC pellet”) and cyto (“PC sup”). The conclusion was reached that both mitochondria and cytoplasm together are needed for new FeS cluster synthesis on Leu1^R with concomitant activation of its enzyme activity. Most likely mitochondrial ISC components are needed to synthesize the (Fe-S)_int, which is transported by Atm1 to the cytoplasm, where the CIA mediates FeS cluster assembly on Leu1. Thus, a prediction is that mutations in critical mitochondrial ISC components will interfere with new Leu1 activity in the cytoplasm, and this has been substantiated by assays using the nfs1-14 mutant, a hypomorphic allele of Nfs1 [16]. Similarly, mutations in the mitochondrial export machinery (e.g., atm1 mutants) interfere with reconstitution of Leu1^R FeS cluster assembly/enzyme activity in the mixing assay, and this has been substantiated by assays with Δatm1 deletion strains [16]. Finding a validated atm1 mutant strain to perform these assays was tricky, due to the propensity of the Δatm1 strain to form extragenic suppressors that obscure the mutant phenotype [48]. We used various methods to get around this problem including backcrossing the mutant and working with fresh tetrads to remove confounding suppressor phenotypes. As a check on the phenotype, a plasmid carrying a genomic copy of ATM1 was reintroduced into the Δatm1 strain by transformation. Following transformation, the ability of the corrected Δatm1/ATM1 strain to reconstitute new Leu1^R enzyme activity in the mixing assay was completely restored as predicted. Finally, a CIA mutant (e.g., Dre2-depleted cells) was tested and found to be deficient in the formation of new FeS clusters on Leu1^R in the permeabilized cell mixing assay [16]. Thus, in these newly developed mitochondria-cytoplasm mixing assays, efficient FeS cluster assembly on cytoplasmic ΔNYah1 or Leu1^R requires mitochondrial ISC components, mitochondrial export machinery, and the CIA machinery.

5.1. Mitochondria-mediated activation of latent FeS clusters in the cytoplasm

One of the advantages of the “mixing” assays for FeS cluster assembly is that we can mix and match mitochondria and cytoplasm from different sources. During the course of experiments with nfs1-14 components and added apo-Leu1^R protein, we encountered the following results (Fig. 4A). i) Mixing of Δleu1 mitochondria (i.e., mitochondria isolated from Δleu1 cells that are otherwise wild-type) with Δleu1 cytoplasm gave strong Leu1^R activity reconstitution (12 units). ii) Mixing of nfs1-14 mitochondria (hypomorphic Nfs1 allele) with nfs1-14 cytoplasm gave virtually no activity (<1 unit). iii) The surprising result came when Δleu1 mitochondria were mixed with nfs1-14 cytoplasm. FeS cluster loading activity was almost quantitatively restored (11 units) (Fig. 4A, lane 5). This was a surprise because nfs1-14 cyto would be expected to be deficient in FeS cluster assembly activity. Many CIA components in the cytoplasm carry critical FeS cluster cofactors e.g., Dre2, Cfd1, Nbp35, and Nar1. Since nfs1-14 cells are deficient in apparently all their FeS clusters, it follows that the CIA pathway components should also be deficient, and CIA is needed for the cytoplasmic activity in the wild-type mito/nfs1-14 cyto mixing assay. In subsequent experiments, we have shown that during the time course of the assay, nfs1-14 cytoplasm starts off deficient at the time of the mixing, but becomes progressively normal for Leu1^R enzyme activation, suggesting that new FeS clusters were initially formed on deficient CIA components themselves and subsequently on Leu1^R. Time course experiments comparing WT (wild-type) mitochondria/WT cytoplasm mixture versus WT mitochondria/nfs1-14 cytoplasm mixture show kinetic delay in the latter. This is likely due to the WT mitochondria/nfs1-14 cytoplasm undergoing a process of restoration of latent FeS clusters in cytoplasmic CIA components [16]. The term “latent” denotes an FeS protein e.g. Dre2 that is primarily but not entirely in the apo form, and that can then be converted to the holo form in part by its own agency. The biology of the effect is confounded by a “chicken/egg conundrum” because the conversion of apo-Dre2 to holo-Dre2 is presumed to require some holo-Dre2 to be present at the outset.

Fig. 4. — **A) nfs1 mutant latent clusters in the cytoplasm**. Mitochondria and cytoplasm from various sources were mixed, apo-Leu1^R was added, and reconstitution of allpha IPM isomerase (Leu1) enzyme activity was measured. Δleu1 mito + Δleu1 cyto had normal activity (Bar 1); nfs1 mito + nfs1-14 cyto had negligible activity (Bar 2); nfs1 mito alone had negligible activity (Bar 3); nfs1 cyto alone had negligible activity (Bar 4). Surprisingly, Δleu1 mito plus nfs1 cyto had high activity equivalent to the normal (Bar 5), but nfs1 mito plus Δleu1 cyto had no activity (Bar 6). The symbol “+” represents mixing of fractions. nfs1 refers to the strain carrying the nfs1-14 hypomorphic allele. **B) 33C (ssq1 mutant) latent clusters in the cytoplasm**. Mitochondria and cytoplasm from various sources were mixed, apo-Leu1^R protein was added, and reconstitution of Leu1 activity was measured. Δleu1 mito + Δleu1 cyto had normal activity (Bar 1); 33C mito + 33C cyto had negligible activity (Bar 2); 33C mito alone had negligible activity (Bar 3); 33C cyto alone had negligible activity (Bar 4). Surprisingly, Δleu1 mito plus 33C cyto had significant activity equivalent to 4.5 units, less than normal but highly significant (Bar 5), but 33C mito plus Δleu1 cyto had no activity (Bar 6). The symbol “+” represents mixing of fractions. **C) Time course of latent FeS cluster recovery**. Top panel. WT mito were incubated with Δleu1 cyto (or 33C cyto) and apo-Leu1^R. Aliquots were removed at different time points as indicated, and the reconstituted Leu1 enzyme activity was measured. Bottom panel. WT mito were incubated with 33C cyto and apo-Leu1^R, aliquots were removed at indicated time points, and Leu1 activity was measured. The accelerating rate of recovery of Leu1 activity probably represents reconstitution of FeS clusters on CIA components Cfd1, Nbp35, Dre2, and/or Nar1, although this remains to be directly shown. Best fit values for the recovery curve were second order Y=A + B*X + C*X^2, with A = −0.1752, B=0.2451, and C=−0.002404.

Similar mixing experiments have been performed with isolated and enriched fractions of mitochondria (mito) and cytoplasm (cyto) using other combinations such as WT mito/Δatm1 cyto and WT mito/Ssq1↓ cyto [16], WT mito/Yfh1↓ cyto (our unpublished data) and WT mito/33C cyto (hypomorphic Ssq1allele; this work see below). The results were very similar in all cases. Immediately following the mixing of mitochondria and cytoplasm, the mixture showed low activity for FeS cluster assembly on Leu1^R. Over time, the activity increased, presumably as CIA activity was restored. Control mixing with nfs1 mito/nfs1 cyto, atm1 mito/atm1 cyto, yfh1 mito/yfh1 cyto, ssq1 mito/ssq1cyto all exhibited negligible activity that did not change over time. Thus, the reconstitution depended on a contribution from the WT mitochondria to the mutant cytoplasm. As an example, here we present data from a typical mixing assay for WT mito/33C cyto (hypomorphic ssq1 mutant). WT mito were mixed with Δleu1 cyto or 33C cyto and recombinant apo-Leu1^R. The mixture was incubated, and samples were removed at various time points to assess reconstituted Leu1^R enzyme activity. The activity curve for Δleu1 mixing increased rapidly whereas the 33C mixture increased much more slowly (Fig. 4B). A more detailed time course of the WT mito/33C cyto mixture showed a slow increase from negligible levels and then with increasing slope up to 2.8 units Leu1 activity (Fig. 4C). The controls such as 33C mito plus Leu1^R (no cyto), 33C cyto plus Leu1^R (no mito), and 33C mito plus 33C cyto plus Leu1^R showed virtually no activity.

In summary, a likely scenario is that following mixing of WT mito and 33C cyto, the WT mito contributed the (Fe-S)_int intermediate to the mutant cytosol via Atm1, and the CIA components containing latent FeS clusters in the 33C cytoplasm were rapidly restored/activated. This restoration process involved new FeS cluster assembly on the critical CIA components themselves containing FeS clusters. There is thus a chicken and egg conundrum here, since the FeS clusters on the CIA components are needed for synthesis of the FeS clusters on the same deficient CIA components. These details have not been worked out.

A fascinating phenotype initially noted by Lill and co-workers [14], and subsequently confirmed by us [16] is that mitochondrial membrane potential is needed for cytoplasmic FeS cluster assembly. Cells were grown with ⁵⁵Fe radionuclide in the presence of CCCP (carbonyl cyanide chlorophenylhydrazone) uncoupler. Virtually no Leu1 ⁵⁵FeS clusters were detected in the absence of mitochondrial membrane potential [14]. Subsequently in our “mixing” assay [16], a mixture of mitochondria plus cytoplasm plus bacterial expressed apo-Leu1^R showed excellent reconstitution of Leu1^R enzyme activity but virtually none (~3%) in the presence of CCCP [16]. Why should this be? The key mitochondrial membrane potential dependent step for cytoplasmic FeS cluster assembly has not been determined but is a subject of ongoing research. It could be related to mitochondrial iron import, mitochondrial iron delivery to a special intramitochondrial utilization pool for FeS cluster assembly, iron insertion into the Isu1/2 scaffold protein, synthesis of the (Fe-S)_int intermediate and/or export of the intermediate from mitochondria into cytoplasm.

6. Dependence of mitochondrial FeS cluster assembly and cytoplasmic FeS cluster assembly on nucleotides and iron

Assays using isolated mitochondria and isolated cytoplasm mixtures have the advantage that energetics can be dissected i.e. dependence of FeS cluster assembly on ATP, GTP, and NADH, all of which occur early in the mitochondrial ISC pathway [15, 62]. In mitochondria, ATP is required for Ssq1 chaperone activity and Atm1 transport activity [1], thus acting early in the FeS cluster biosynthetic process. In an assay with isolated mitochondria, formation of holo-aconitase depends on ATP hydrolysis and is blocked by ATPγS [60]. Cytoplasmic FeS cluster synthesis on ΔNYah1 likewise depends on ATP although the role of Ssq1 and Atm1 ATP requirements have not been separated [62]. In similar manner, GTP is required for mitochondrial FeS cluster assembly. In yeast, the cytosolic GTP enters mitochondria via the GTP/GDP exchanger, Ggc1, and within mitochondria acts to enhance formation of the scaffold intermediate on Isu1/2. The effect is likely mediated by a matrix GTPase but this has yet to be identified [15]. NAD(P)H also play an important role in FeS cluster assembly. Pos5 encodes a NADH kinase that provides mitochondrial NADPH [63] and enhances mitochondrial FeS cluster assembly [64], and NADPH in mitochondria probably acts by enhancing Yah1 reductase activity; Yah1 reductase mediates sulfur reduction (S⁰ to S²⁻) involved in formation of the Isu FeS cluster intermediate. As all these nucleotide requirements occur early in the mitochondrial ISC pathway, they would be predicted to be required for cytoplasmic FeS cluster assembly, although the latter has not been directly demonstrated.

Iron is required for FeS cluster assembly in both mitochondria and cytoplasm. The precise trafficking pathways and chemical forms of iron required for mitochondrial versus cytoplasmic FeS cluster assembly are still not clear. In yeast, iron enters cells via a high affinity transport system consisting of a plasma membrane protein complex of a copper ferroxidase (Fet3) and permease (Ftr1). The Fet3 ferroxidase activity converts ferrous iron to ferric iron, depositing it into a transport channel [65]. After cell entry, the ferric iron is re-reduced to the ferrous form, likely by cytoplasmic glutathione and then enters the cytoplasmic labile iron pool [66]. The iron traffics to mitochondria and gains entry to the mitochondrial matrix via the activity of the paralogous mitochondrial carrier proteins, Mrs3 and Mrs4 [67]. In the mitochondrial matrix, iron participates in FeS cluster assembly, providing the substrate for formation of the FeS cluster intermediate on the Isu1/2 scaffold protein. Exactly how this occurs has been a matter of great controversy. The role of frataxin as an iron chaperone has been debated. Frataxin can bind 2 ferrous iron atoms with modest 3 μM affinity on an acidic ridge created by the alpha helices overlying the beta sheet platform. Frataxin also shows strong specific interaction with Isu1 and in principle could deliver iron into the ISC pathway via this route [68]. However, new structural information has shown that the acidic binding site on frataxin is covered by Nfs1 in the quaternary structure of the FeS cluster assembly complex, decreasing the attractiveness of this hypothesis [1]. No other iron chaperone has come to light after years of work. An alternative “iron first” hypothesis has been proposed [23]. According to biochemical data, upon removal of a bound zinc ion in the Isu cluster binding site (C69, D71, C96, and H138 in yeast Isu), Isu binds an iron atom in the site and can carry out FeS cluster assembly with high efficiency. The steps involved are as follows. A persulfide is formed by Nfs1 on the catalytic cysteine and subsequently transferred to Isu C139 in a process facilitated by the persulfide sulfur chaperoning activity of frataxin. The sulfur Fe-Isu is then reduced to sulfide by electrons from Yah1, and a [2Fe-2S] cluster is formed. Note formation of a dimer of Isu may be needed at this step so that a second atom of iron and sulfide can be contributed to the complex. Subsequently, the [2Fe-2S] cluster is transferred to Grx5 in an Hsp70 chaperone mediated step [35]. The source of the iron in mitochondria for FeS cluster assembly is not entirely clear, but this may originate from the mitochondrial labile iron pool. Treatment of isolated mitochondria with o-phenanthroline blocks the process of FeS cluster formation by chelating iron in the mitochondrial labile iron pool [15, 69].

The next place in which iron is needed in mitochondrial/cytoplasmic FeS cluster assembly is for formation of the (Fe-S)_int intermediate. A key issue regarding the nature of the (Fe-S)_int is whether it contains iron. There is evidence that it contains glutathione, based on the shape of the binding site pocket in the Atm1 structure and structure-function analysis of the key glutathione binding site residues [70]. This intermediate almost certainly contains sulfur, because we use [³⁵S]cysteine as the source of sulfur and are able to detect cytoplasmic proteins radiolabeled with Fe³⁵S clusters [17]. What about iron? The following experiment suggests that the “(Fe-S)_int” intermediate exported from mitochondria contains iron (Fig. 5). The assay takes place in two steps. In the first step, mitochondria were incubated with [³⁵S]cysteine, plus or minus iron as ferrous ascorbate. The mitochondria were then washed and recovered. In the second step, ΔNYah1 and cytoplasm were added, again with or without iron. No further radioactivity was added in the second step so the Fe³⁵S cluster label must derive from the mitochondrial labeling during the first step. Looking at the results, we see that the ΔNYah1 radioactive signal from Fe-³⁵S loading (Fig. 5, lanes 2 and 6) appears only when the mitochondria were supplemented with iron in the first step. Iron supplementation of cytoplasm during the second step made no difference, and iron supplementation during the second step could not overcome the lack of iron loading in the first step (Fig. 5, lane 8) [17]. These data suggest that iron for FeS cluster assembly on the cytoplasm-localized apo-ferredoxin originated in the mitochondria and was exported to the cytoplasm, likely as a component of the Atm1 transport intermediate, (Fe-S)_int. Note that the tRNA signal, reflecting cytoplasmic tRNA thiolation, was basically unchanged throughout the various experimental conditions, as thiolation occurs by an overlapping but separate pathway that is cysteine desulfurase dependent but iron independent (see section 12 on thiolated tRNAs below).

Fig. 5. — A two-step assay was performed. In the first step, WT mito were incubated with [³⁵S]cysteine, with or without added iron (10 μM of ferrous ascorbate). Mito were recovered and then incubated with WT cyto with or without apo-ΔNYah1 substrate and added iron as indicated (second step). Samples were centrifuged, and the resulting supernatant fractions (“S”) containing the cytoplasm were analyzed by native PAGE, followed by autoradiography. The ³⁵S-labeled tRNA signal was present in all samples. The ΔNYah1 [2Fe-2³⁵S] signal was present only when iron was included during the first step of the assay (lanes 2 and 6), and the effect could not be achieved by adding iron during the second step (lane 8). The implication is that iron for new Fe³⁵S cluster on ΔNYah1 originates in mitochondria [17].

7. Atm1 structure and Atm1 transport cycle

Atm1 is positioned as a gate between the mitochondrial matrix and intermembrane space (IMS)/cytoplasm. The existence of a homodimeric Atm1 as the functional unit was surmised based on the half transporter sequence characteristics. A presumed substrate binding domain was found to reside in the predicted matrix domain, adjacent to the ATPase domain. Subsequently, structural studies have added more detailed information to this picture [70]. Bacterial expressed and purified yeast Atm1 was crystallized. Two sets of crystals were obtained, both apparently representing an inward open conformation without any nucleotides present. Surprisingly, although no substrate was added during the crystallization process, one set of crystals showed glutathione bound in a large positively charged hydrophilic cavity, apparently the substrate binding site that accommodates GSH through interaction with residues R280 and R284 (TM4), N343 (TM5), and N390, S394, R397 and D398 (TM6). The other crystal without glutathione had similar coordinates, thus glutathione binding did not induce any major conformation in the Atm1 backbone. Since no nucleotide was detected, the transport cycle for Atm1 likely involves substrate/glutathione binding prior to nucleotide binding followed by secondary conformational changes. The other transport states (eclipsed substrate, nucleotide bound, outward open) were not captured in this study [70]. The transport substrate was also not captured. Most likely, glutathione plays a key role, both as a component of the transport substrate and as a modulator of the transport/export activity (see below). In this regard, it is interesting that glutathione-depleted cells (Δgsh1 grown in defined media without added glutathione) phenocopy Atm1-depleted cells, manifesting high and dysregulated iron uptake to the cell and mitochondria, and manifesting specific cytoplasmic FeS cluster deficiencies [71, 72]. Glutathione binding residues, mutated according to their locations in the Atm1’s primary structure conveyed a loss-of- function phenotype [70]. The Atm1 structure forms a dimer consisting of 6 × 2 transmembrane domains with the C-terminal domains showing an unusual overlapping configuration. Curiously, the TM6 and TM7 swap positions and thus stabilize the C-terminal dimer interactions. The ATPase domain also at the C-terminus is contributed by both halves of the dimer and is stable without nucleotides.

What is the substrate of Atm1 transport? The short answer is that we don’t know, but there are some hints about its composition. The direction of transport from mitochondrial matrix to IMS/cytoplasm has been well established [48]. The characteristics of the substrate should explain the phenotype of the atm1 mutant i.e., specific lack of cytoplasmic FeS clusters. Perhaps the substrate is itself an FeS cluster, a complex thereof, or a signal that confers competence for FeS cluster assembly to the CIA machinery. The early ISC machinery within the matrix is dedicated to synthesizing [2Fe-2S] cluster intermediates [1]. Perhaps these are transported out via Atm1 and utilized directly or indirectly in the cytoplasm by the CIA for FeS clusters [73]. What is missing from the picture is biochemical evidence. There should be some biochemical confirmation of the role of the substrate, for example rescue of the biochemical phenotype of the atm1 mutant in a reconstitution assay such as the mitochondria-cytoplasm mixing assay with added substrate [16]. To date, no candidate has met this threshold of confirmation. Glutathione clearly has something to do with the Atm1 substrate. The role of glutathione while not entirely defined is well established by multiple avenues of evidence. Glutathione fits in the substrate binding cavity of the Atm1 structure (also of functional homologs like ABCB7 of humans and Atm3 of Arabidopsis), but the binding site is much larger than needed to accommodate glutathione alone, suggesting that glutathione itself is not the substrate, rather it might be a constituent of the complete substrate [70]. Mutations in the glutathione binding site in Atm1 confer a loss-of-function phenotype [74]. Again, glutathione alone is probably not the Atm1 export substrate, because glutathione addition cannot rescue the mutant phenotype of Δatm1 cells [72].

Two candidate molecules for the Atm1 substrate deserve further mention. Both are related to glutathione as a base compound. Balk and colleagues [75] found that GSSG or GSSSG stimulated the ATPase activity of purified Atm1 or Atm3 (Arabidopsis homolog). They utilized L. lactis inside out vesicles to show GSSSG transport by Atm1 into the vesicles in an ATP-dependent manner. A “transportomics” approach showed intra-vesicle accumulation of GSSSG at the expense of GSSG with specific accumulation of 3.1 pmol per mg of protein, indicating that this derivative could also be a substrate [75]. However, in vivo or in organelle studies were not performed, and the critical correction experiment was not attempted. In this experiment, the substrate would be added post-transport to the cytoplasm and compensate for lack of Atm1 transporter function.

Cowan and co-workers have provided evidence for the possible role of glutathione-bound [2Fe-2S] clusters [76]. Synthesis of the clusters was performed chemically by mixing ferric chloride and sodium sulfide with glutathione in ethanol and collecting the precipitate, which was dried under vacuum. The resulting [2Fe-2S](GS)₄ compound was apparently stable in aqueous solution. This chemically synthesized compound was shown to be transported into ABCB7 vesicles [73]. A cryo-EM study of CtAtm1 (homolog from Chaetomium thermophilum) showed a density binding specifically to an inward open complex in nanodiscs after addition of the compound, but once bound the FeS cluster could no longer be clearly discerned in the structural images [39]. The more definitive biological experiment of adding the putative substrate to the cytoplasm of transport-blocked cells to show reconstitution of cytoplasmic FeS cluster assembly in a transport mutant was not performed.

Atm1 may have more than one substrate. The compartmental phenotypes of the atm1 mutant cells include not only cytoplasmic defects in FeS clusters but also cytoplasmic defects in tRNA thiolation [60, 77]. Nfs1 in mitochondria is apparently able to generate a persulfide on the critical cysteine residue C421 and to transfer this persulfide sulfur to the Isu scaffold as part of the ISC FeS cluster assembly process in mitochondria. The persulfide sulfur though may also be donated, directly or indirectly, to certain tRNAs in the mitochondria or cytoplasm. Thiolation occurs in the 2-position on the U34 anticodon of certain cytoplasmic tRNAs (Glu, Lys, Gln) mediating enhanced and more accurate protein translation. A sulfide relay process involves transport of the persulfide sulfur intermediate from Nfs1 to Atm1 to cytoplasmic relay components and thence to cytoplasmic tRNAs (Glu, Lys, Gln) on the U34 anticodon [78]. The atm1 mutants lack the ability to transfer Nfs1-generated persulfide sulfur to the cytoplasm. Note 2-thiouridine formation for tRNA modification is an essential process, apparently necessary for translational fidelity of certain proteins [78]. The persulfide sulfur intermediate for cytosolic tRNA thiolation, called S_int, is not the same as the (Fe-S)_int needed for cytoplasmic FeS cluster assembly (see section on thiolation) [60, 79]. Thus far, alleles of Atm1 that separate the two transport functions have not been isolated, but these would be of great utility in understanding and defining the transport processes.

Progress has been made in defining the structural elements of the Atm1 transport cycle involving a “switch model” in which the nucleotide binding domains (NBDs) switch between a dimeric conformation closed around two molecules of ATP and a nucleotide-free dimeric open conformation [80]. CtAtm1, the transporter from a model fungal eukaryote, was expressed in E. coli and reconstituted into lipid nanodiscs. Using clever combinations of mutagenesis and drug/substrate additions, various transport intermediates were captured by cryo-EM, conveying a picture of the phases of the transport cycle [39] (Fig. 6). The different phases include: inw-opn (inward open) modeled by the E603Q mutant since this does not bind ATP; inw-opn/cluster (inward open/cluster) with substrate bound in the large positively charged cavity; inw-opn-occl (inward open occluded) as “legs” of the structure move towards each other and shrink the substrate binding cavity; occl/ATP (occluded ATP) as ATP binds the ATP duplicated ATP binding sites producing a major conformational change; out-open (out open) as the transporter closes to the matrix side and opens to the IMS side, releasing the substrate on the outside (Fig. 6). Some aspects of the transport cycle were not defined in this study, most importantly the inward-openoccluded, which showed no evidence of the purported cluster substrate and had to be supplemented with GSSG to generate the structural information. The out-open phase was not captured at all and had to be modeled, mostly based on a structure of Atm3 (Arabidopsis homolog) [81]. The putative [2Fe-2S](GS)₄ intermediate was not viewed bound to Atm1 during the transport cycle, perhaps due to lability or modification during transport or due to limitations in visualization.

Fig. 6. — **A) Schematic representation of changes in conformation during the CtAtm1 transport cycle as assessed by cryo-EM [**39]. CtAtm1^inw-opn. The resting state for the transporter is the inw-opn (inward open) state, with matrix oriented “legs” separated from each other, and maximum distance between with NBDs (nucleotide binding domains). The substrate binding site is lined by positive charges (+ in diagram). No substrate is present. The F396 “gate” residue is undisturbed and entry into the membrane domain is unperturbed. CtAtm1^{inw-opn/cluster}. The [2Fe-2S](GS)4 hypothetical and chemically synthesized substrate (star) binds in the substrate binding domain, generating the CtAtm1^{inw-opn/cluster} form. CtAtm1^inw-opn-occl (inward open occluded). A conformational change ensues, as the “legs” move towards each other, partially occluding the opening in the substrate binding cavity. Atm1^occl/ATP (occluded ATP). The occl/ATP conformation is generated by ATP binding to the NBDs, and occlusion of the substrate binding domain is completed. CtAtm1^out-opn. The F396 gate is disrupted to open a channel to the M-domain. CtAtm1^out-opn (out open). Note that this structure has not been observed but only modeled (according to the dashed line enclosing the CtAtm1^out-opn in Fig. 6A). In the next stage, the transporter reorients with the opening towards the M-domain and the IMS side. The substrate moves through to the other side of the membrane, and the gate is reestablished on the other side of the substrate, while interactions are formed between the substrate bound Atm1 R138 and target proteins in the IMS. Bound ADP is then released, allowing rebinding of ATP, concurrent with substrate release and reorienting of the transporter to resume its original inw-open configuration, ready for another cycle of binding and transport (not shown). **B) Cryo-EM images from Electron Microscopy Data Bank, depicting various transport stages for CtAtm1 and ScAtm1**. The inw-opn state (PDB-ID 7pqx, EMB-13606) is mimicked by the E603QCtAtm1 (PDB-ID 7psd, EMB-13612), because the ATP hydrolysis activity is abrogated by this mutation. The inw-opn/cluster state (PDB-ID 7pro, EMB-13609) depicts the [2Fe-2S](GS)4 substrate as a “pod” hanging from the substrate binding site. The inw-opn-occl state (PDB-ID 7pru, EMD-13610) occurs after substrate binding. The occl/ATP state (PDB-ID 7pr1, EMD-13607) occurs after ATP binding to the NBDs. Also shown are images of the *Saccharomyces cerevisiae* (Sc) Atm1 transporter, nucleotide free, with separation of the NBDs (PDB-ID 7psl, EMD-13613), and occluded after binding of the non-hydrolyzable analogue AMP-PNP, Mg²⁺ (PDB-ID 7psm, EMB-13614).

8. Glutathione and FeS cluster assembly

Glutathione is an abundant tripeptide (L-γ-glutamyl-L-cysteinyl-glycine), found at levels of up to 10 mM in eukaryotic cells. The reduced (GSH) and oxidized (GSSH) forms of glutathione act with various reductases to maintain overall cellular and cell compartment redox status. In most tissues, the estimated redox potential for the GSH/GSSG couple ranges from −210 mV to −150 mV, with significant differences between cellular compartments. Mitochondria are more oxidizing than cytoplasm, with implications for mitochondrial metabolism. GSH is synthesized in a two-step process catalyzed by L-glutamate L-cysteine γ-ligase (γ-GLCL, EC 6.3.2.2; also called γ-glutamyl-L-cysteine ligase) and glutathione synthase (GLS, EC 6.3.2.3), both localized in the cytoplasm (Fig. 7A). However, 15% of cellular glutathione resides in mitochondria [82], implying that there must be transporter(s) mediating glutathione import into mitochondria. A human protein SLC25A39, a member of the mitochondrial carrier family, was recently identified as a mitochondrial glutathione transporter [83]. The definitive experiment of reconstituting the protein into lipid vesicles and measuring the substrate exchange and affinities has not yet been done. However, uptake experiments with isolated mitochondria and radionuclide ([¹³C2, ¹⁵N]-GSH) showed marked reduction in transport for the SLC25A39-knockout HeLa cells compared with the same cells with reintroduced cDNA, strongly suggesting that this protein is a mitochondrial glutathione transporter. The key residues constituting the substrate binding site were predicted based on earlier informatics work on the entire family of mitochondrial carrier protein substrate binding sites by Kunji and co-worker [84], and mutation of two of these residues D226A or K329A abolished GSH transport by this transporter. A SCL25A39 paralog was discovered by a metabolism focused single guide RNA (sgRNA) screen for genes essential for cell proliferation in the absence of SCL25A39. This synthetic lethal study led to identification of SCL25A40 as responsible for redundant glutathione transport function into the mitochondrion [83]. The double knockout not only exhibited depletion of mitochondrial glutathione but also showed a marked reduction in proliferation and viability, thus demonstrating that an adequate level of mitochondrial glutathione is essential for these processes. The question arises of why this should be the case. Two excellent proteomic studies pointed to the same culprit - deficient mitochondrial FeS proteins [83]. In one study, HeLa cells were treated with buthionine sulfoximine (BSO) to chemically deplete glutathione, and glutathione levels were observed to progressively decrease with similar kinetics in both cytoplasm and mitochondria. Proteomic comparison of the mitochondria from untreated and treated cells revealed 269 protein differences including mitochondrial and cytoplasmic proteins. Many of these proteins exhibited greatly decreased abundance, representing mostly FeS proteins, including a subunit of succinate dehydrogenase of the mitochondrial electron transport chain (SDHB) and a mitochondrial redox sensor with unusual histidine ligation (CISD3). There were also many mitochondrial translation components that were decreased. In recent work, FeS clusters in mitochondrial ribosomal components have been discovered that act as sensors controlling mitochondrial protein translation, thus placing this entire process of mitochondrial protein translation downstream of global FeS cluster biogenesis[7, 85]. Interestingly, the proteomic pattern of BSO-treated cells was recapitulated by NFS1 knockdown cells that exhibited impaired FeS cluster synthesis. The second proteomic study compared SLC25A39/40 double knockout Jurkat cells to those re-expressing SLC25A39 cDNA [83]. Similarly, the double knockout was deficient in mitochondrial glutathione and showed lower levels of many FeS proteins and mtDNA encoded components. A biochemical feature of this phenotype was that the ratio of αKG (alpha-ketoglutarate) to succinate went up 4-fold in the double mutant [83]. The reason for this is that lipoic acid synthase (LIAS), a protein containing critical FeS clusters, was deficient, leading to deficient lipoic acid in the oxoglutarate dehydrogenase complex, thereby blocking conversion of αKG to succinate. In summary, mitochondrial glutathione is essential for cell viability because it is needed for FeS cluster biogenesis. This makes sense because critical components of the mitochondrial FeS cluster biogenesis machinery e.g., Grx5, (Fe-S)_int, and Atm1 may utilize glutathione as cofactor and/or constituent (see below) (Fig. 7A).

8.1. Bypassing mitochondrial glutathione transport

Wang et al. devised a clever bypass approach that was able to rescue the mitochondrial glutathione depletion in SLC25A39 knockout cells [83] (Fig. 7B). The usual eukaryotic pathway uses two genes/proteins to assemble glutamate, cysteine, and glycine in the synthesis of glutathione in the cytoplasm. The cytoplasmic glutathione then enters mitochondria via the SLC25A39 mitochondrial carrier. The SLC25A39 mutant is deficient in mitochondrial glutathione levels, and consequently exhibits FeS cluster deficiency. The bacterial Streptoccocus thermophilus gene, GshF, performs glutathione synthesis in a single step. A mitochondrial targeting signal was fused to the N-terminus of GshF, and this chimeric protein was expressed in SLC25A39 knockout cells. The heterologous expression of this GSH-generating system targeted to mitochondria bypassed the effect of the transport block and restored normal glutathione levels to the mitochondria and normal FeS cluster assembly activity. The expression of this GshF construct in mitochondria also rescued the proliferative defect in the SLC25A39/40 double mutant. In terms of FeS clusters, mitochondrial FeS proteins (e.g., Aco2) were deficient in the SLC25A39/40 double mutant and restored with GshF mitochondrial expression. Interestingly, cytoplasmic and nuclear FeS proteins (including PPAT, phosphoribosyl pyrophosphate amidotransferase) were also deficient in the SLC25A39/40 double mutant and also restored with GshF expression in mitochondria [83]. This finding illustrates the role of mitochondria in cytoplasmic FeS cluster assembly. Deficient mitochondrial glutathione in the SLC25A39/40 mutant perhaps led to deficiency of the Atm1/ABCB7 transport intermediate (Fe-S)_int, which in turn resulted in deficiency of cytoplasmic FeS clusters. The mitochondrial-expressed GshF by correcting the mitochondrial deficiency of GSH also corrected the cytoplasmic phenotype, restoring the functional status of cytoplasmic and nuclear FeS proteins (Fig. 7B).

8.2. Yeast and glutathione

Early work on glutathione and FeS cluster assembly in yeast demonstrated a puzzling phenotype associated with glutathione depletion [71, 72]. Yeast cells deleted for the enzyme Gsh1, involved in the first step of glutathione biosynthesis, cannot synthesize glutathione but they can take it up from the growth medium. In glutathione-free medium, a Δgsh1 yeast strain undergoes about 8 doublings and then undergoes growth arrest [71]. Before growth arrest, at about 6 doublings in the absence of added glutathione, these cells exhibit a compartmentalized FeS cluster defect: cytoplasmic FeS clusters are deficient and not made properly while mitochondrial FeS clusters are still made and are present at normal levels. The conclusion was made that glutathione is specifically needed for cytoplasmic FeS cluster assembly (and not mitochondrial FeS cluster assembly). However, mitochondrial glutathione may be simply more difficult to deplete during growth in glutathione-free medium. The yeast ortholog for SCL25A39/40 is encoded by the Mtm1 gene [83]. Studies can now be performed in yeast with Mtm1 depleted/deleted cells, perhaps with a Δgsh1 deletion to confirm the role of mitochondrial glutathione in cellular FeS cluster assembly.

A current consensus is that mitochondrial glutathione plays a vital role in the biogenesis of most, if not all, FeS proteins in the cell. In mitochondria, GSH is utilized in key components for mitochondrial FeS cluster assembly such as Grx5, where an FeS cluster intermediate is transiently coordinated by two molecules of Grx5 and two molecules of GSH (Fig. 7A). This intermediate then goes on to participate in later stages of mitochondrial FeS cluster assembly including [2Fe-2S] trafficking, late [4Fe-4S] cluster synthesis, and generation of (Fe-S)_int. Recall that GSH might also be a constituent of (Fe-S)_int, the intermediate that is made in mitochondria in an ISC dependent manner and then binds in the inward open configuration to Atm1/ABCB7, triggering conformational change and ATP-dependent export to the cytoplasm. In the cytoplasm, the (Fe-S)_int intermediate with its glutathione component interacts with one or more constituents involved in cytoplasmic FeS cluster assembly. The most likely recipient to “touch” the (Fe-S)_int is a complex of Grx3/4 (GLRX3/GLRX3 in humans) with glutathione, mediating formation of the first [2Fe-2S] cluster intermediate outside mitochondria [86]. An alternative model has the (Fe-S)_int directly interacting with the Grx3/Grx4/Bol2 heterocomplex and data from yeast show that iron regulation depends on this complex [87]. Subsequent activity of the CIA components, reductase Tah18/Dre2, scaffold Nbp35/Cfd1, hydrogenase like Nar1, and CTC (CIA targeting complex) mediate formation and insertion of FeS cluster intermediates into cytoplasmic and nuclear apoprotein recipients, but these steps are not mediated by glutathione as far as we know. In sum, the glutathione-dependent steps in FeS cluster assembly involve mitochondrial Grx5-GS, (Fe-S)_int formation, (Fe-S)_int transport, and cytoplasmic Grx3/Grx4/GS/Bol3 (Fig. 7A).

CGFS (Cys Gly Phe Ser) glutaredoxins (Grxs) and BolA proteins are found co-occurring in most sequenced genomes. In bacteria, they may co-occur in the same operon, suggesting a functional alignment [18]. The Grxs belong to the thioredoxin (Trx) superfamily and harbor a Trx fold consisting of four stranded anti-parallel beta sheets flanked by three alpha helices (Fig. 7C). Unlike the class I type Grx which has enzymatic activity and catalyzes thio-disulfide exchange reactions with GSH substrates, the class II Grxs are enzymatically inactive, but play roles in iron metabolism due to their ability to bind FeS clusters via their CGFS active sites and two GSH molecules. BolA proteins are also highly conserved and exhibit a characteristic fold with four alpha helices and three beta sheets [18]. BolA proteins interact with CGFS glutaredoxins, either in the apo or holo state, with micromolar affinity. BolA proteins also independently interact with FeS clusters making use of critical Cys and His ligands (e.g., Cys66 and His103 in Fra2). The conserved interactions of BolA and Grx orthologs in some cases destabilizes the bound cluster as part of an FeS cluster transfer process; in other cases, BolA and Grx interaction may stabilize FeS liganding. The BolA-Grx complex may sometimes act as an FeS cluster chaperone and storage pool [88]. Distinct Grx complexes, Grx5 in mitochondria or Grx3/Grx4 in cytoplasm and nucleus, play important roles in FeS cluster assembly. In conjunction with GSH-GS they can bind and release FeS clusters, thus participating in FeS cluster trafficking. The crystal structure of the bacterial 2Grx4-2GS-[2Fe-2S] cluster is shown as an example of a glutaredoxin coordinating a [2Fe-2S] cluster (Fig. 7C). The cysteine of the CGFS motif in the glutaredoxin supplies 2 of 4 ligands to the iron of the [2Fe-2S] cluster. The cysteine of the GSH (glutathione) supplies the other 2 of 4 ligands. The coordinated cluster is shown as a yellow rectangle with 2 iron atoms and 2 inorganic sulfurs at the apices. The cluster may be bound stably or dislocated and transferred as part of the assembly or trafficking process.

9. Iron homeostasis and FeS cluster biogenesis

Distinct Grx/Bol complexes are found in the cytoplasm of yeast or human cells [18]. There they function in FeS cluster biogenesis, FeS cluster storage, FeS cluster trafficking, and iron regulation. In yeast, the Grx3/Grx4 paralogs and Fra2 (also called Bol2) were identified by mutants with defective iron homeostasis. Grx3/Grx4 double mutants or Fra2 mutants were found to exhibit a similar iron starvation regulatory phenotype: cellular iron uptake was induced and the Aft1/2 iron sensing transcription factors were activated, leading to constitutive iron uptake to the cells and repressed iron utilization [89] (Fig. 8). These mutants have provided us with a glimpse of the physiologic iron regulatory machinery. Briefly, an iron replete state leads to generation of a regulatory FeS cluster on Aft1/2, leading to nuclear cytoplasmic shuttling and repression of iron-related genes [18]. These include iron uptake genes such as Fet3, Ftr1, Ccc2, Atx1, and iron regulatory genes such as and Cth1/2. The latter are expressed during iron deficiency and promote degradation of a battery of mRNAs leading to reprogramming of iron-dependent metabolism and storage [90].

Fig. 8. — **Iron regulation in yeast Saccharomyces cerevisiae** [18]. **Low iron conditions:** Under conditions of relative cellular iron starvation, the ISC responds to the iron starvation signal by making less (Fe-S)_int and consequently Aft1/2 remains in the apo form. Aft1/2 is recruited and binds to target sequences in DNA. The iron regulon is turned on i..e. iron aquisition genes are expressed while iron utilization genes are repressed. **High iron conditions:** Under conditions of iron availability, more active (Fe-S)_int is produced and exported by mitochondria. The (Fe-S)_int interacts more readily with the Grx3/4-Bol2 complex, facilitating formation of a [2Fe-2S] cluster coordinated by critical cysteine (of Grx3/4), critical histidine (of Bol2), and glutathione (“G”) cysteine residues. This complex recruits Aft1/2 while still bound to the DNA in the apo state. The combination complex consisting of Grx3/4, Bol2 and Aft1/2 recruits a second Aft1/2 molecule, and transfers the [2Fe-2S] cluster, thus making an Aft1/2 holo-homodimer. The remainder of the complex components, including apo-Grx3/4, glutathione and Bol2 are recycled. Note that there is no direct experimental evidence that the cellular iron starvation signal is communicated by mitochondrial (Fe-S)_int, although extensive genetic data supports this hypothesis. Mutations that interfere with the generation or export of (Fe-S)_int confer an “up” signal to the downstream iron acquisition system.

In terms of genetics, the formation of the regulatory FeS cluster on Aft1/2 requires a large swath of the FeS cluster biosynthetic apparatus for both mitochondria and cytoplasm, but not the entire pathway. The iron starvation phenotype results from a block in Aft1/2 FeS cluster formation. This can occur as a consequence of disruption of early ISC function. Later mitochondrial ISC components such as Isa1/Isa2/Ind1/Nfu1 are not involved, and mutations in those components do not generate an iron regulatory phenotype. Anything that interferes with (Fe-S)_int formation or transport generates an iron regulatory phenotype, including early ISC components up to Grx5, and (Fe-S)_int transport function, that is Atm1 itself. In the cytoplasm, Grx3/Grx4 and Fra2 are required, but downstream CIA components are not, as mutations of Cfd1/Nbp35 and others do not generate the iron regulatory phenotype [30]. On the other hand, reversal of the iron starvation signal by iron repleteness involves a series of events (Fig. 8): Early components of the ISC machinery mediate production of (Fe-S)_int. (Fe-S)_int, is exported to the cytoplasm by Atm1. (Fe-S)_int in the cytoplasm likely interacts with Grx3/Grx4. Grx3/Grx4 forms a complex with Fra2 (Bol2) generating a [2Fe-2S] bridged heterodimer that specifically transfers [2Fe-2S] clusters to Aft1/2, altering its multimeric state from monomer to dimer. The Aft1/2 dimer exhibits decreased affinity for the cognate DNA binding sites and becomes a substrate for nuclear export, in effect turning off expression of iron uptake genes (and turning on expression of iron utilization genes). The identification of the Fra2-Grx3-Bol2-glutathioe interactions and FeS cluster transfers was accomplished in lovely work by C. Outten and co-workers [18, 87]. The formation of the FeS clusters that mediate the switch from low iron to high iron signaling likely depends on the (Fe-S)_int intermediate, but this has not been directly demonstrated. Furthermore, exactly where the (Fe-S)_int intermediate interacts with the cytoplasmic/nuclear FeS cluster assembly system has not been shown, but the Grx/Bol component might directly interact with the (Fe-S)_int intermediate.

In human or animal cells, like in yeast, there is a linkage between FeS cluster biogenesis and iron homeostasis, although the specifics are quite different. In mammalian cells, iron regulation is mediated by two homologous proteins IRP1 and IRP2, with distinct mechanisms of iron sensing [91]. Notably, both involve mitochondrial functions for the production of cytoplasmic FeS cluster intermediates. In the best known case, IRP1 senses iron by interconverting between a holoprotein, the cytosolic aconitase, present in high iron conditions, and apoprotein, an IRE binding protein, present in low iron conditions [91]. The IRE is an RNA stem loop structure which binds the apo-IRP1. If the IRE is in the 5’ untranslated region of certain mRNAs, translation of the corresponding protein (e.g. ferritin) is thereby attenuated. By contrast if an IRE is present in the 3’ untranslated region of an mRNA (e.g. TFR1), binding by the apo-IRP1 stabilizes the message and induces the level of expression. A link to mitochondrial function exists here, because the IRE binding protein/cytosolic aconitase is an FeS cluster protein. As such it requires mitochondrial function for the formation of its FeS cluster: ISC is needed for synthesis of the (Fe-S)_int which must be exported via Atm1/ABCB7 in order to stimulate cytoplasmic FeS cluster assembly of IRP1, generating the holo-protein. Wingert et al [89] studied a Grx5 mutant zebrafish termed sir that was constitutively shifted towards the apo form of IRP1, leading to IRE binding protein repression of various mRNAs, including aminolevulinic acid (ALA) synthase. Grx5 mutated zebrafish, were found to be pale and anemic due to repression of ALA synthase, the rate limiting step for heme synthesis. Confirmatory experiments were performed by injecting zebrafish mutant embryos with various forms of IRE mRNA. When sir mutant was injected with an altered form of ALA synthase mRNA, lacking the IRE (CAGUGC) or with a mutated form of IRE that could not bind the IRP1, ALA synthase production was restored (Figs. 9A, 9B). The injection, in effect, bypassed the constitutive repression of ala synthase that occurred in the sir mutant, restoring heme synthesis and leading to red non-anemic fish. This lovely genetic experiment demonstrated the role of mitochondrial function in general and Grx5 in particular in iron regulation. Interestingly, a human disease (OMIM 616860) caused by loss of function of GLRX5 (Grx5 ortholog), is characterized by anemia and iron misregulation [95], similar to what occurs in the sir mutant zebrafish.

Fig. 9 — **A) Iron regulation in vertebrates and perturbation in Grx5 mutants [**97]. In an organism with mutated Grx5 such as the *sir* zebrafish, the mitochondrial contribution to cytoplasmic FeS cluster assembly is deficient: (Fe-S)_int is made at minimal levels, in effect generating an iron starvation signal, and cytoplasmic FeS clusters are deficient. IRP1 which toggles between the cytoplasmic aconitase and an IRE binding protein is shifted into the latter conformation, binding to cognate IREs in target proteins. One of these targets is ALAS2, the erythroid ALA synthase, has a 5’ IRE. Binding of the IRP1 to ALAS 5’ IRE represses expression, leading to deficiency of heme, low hemoglobin, and anemia in the fish [97]. A parallel regulatory pathway exists mediated by IRP2. In this case, under iron starvation conditions, or in the setting of mutated Grx5, the IRP2 binding protein FBXL5 is turned over, IRP2 is stabilized and binds to cognate IREs, such as the 3’ IREs present in TFR1. TFR1 goes up enhancing cellular iron uptake in response to iron deficiency. Note that the iron starvation signal is mediated through FBXL5 destabilization. This occurs in a two tier manner. The Hr motif in the apo form leads to FBXL5 destabilization. Also, loss of the the C-terminal FeS cluster in the apo form leads to FBXL5 destabilization. Defective mitochondrial function and defective FeS cluster biogenesis, like iron starvation, lead to FBXL5 destabilization and IRP2 stabilization [93]. **B) Bypass of IRP1/IRE mediated repression in the zebrafish *sir* (Grx5) mutant**. The wild-type zebrafish embryo stains for heme with o-dianisidine at 40 h.p.f. (panel 1). The *sir* mutant, mutated in Grx5, has no such staining indicating failure to make hemoglobin (panel 2). The hemoglobin staining was restored after injection of eALAS cDNA constructs that interfere with IRP1 binding to the eALAS stem loop: for example, injection of eALAS cDNA with deletion of the six base stem loop CAGUGC (panel 3). Morpholino knockdown of FCH (ferrochelatase), the mitochondrial enzyme that inserts iron into heme, resulted in accumulation of fluorescent porphyrins (panel 4, top half). *sir* mutation in Grx5, on the other hand, did not lead to fluorescent fish, because the synthesis of porphyrins was blocked upstream at the rate limiting step of ALA synthase (panel 4, bottom half) [97].

Both IRP1 and IRP2 may convey iron signals to downstream iron regulated proteins. IRP2 cannot bind an FeS cluster directly and thus has no aconitase activity, even under high iron conditions. The way in which IRP2 senses iron is via a specialized interacting protein FBXL5 [92]. FBXL5 contains an N-terminal hemerythrin (Hr) motif with a diiron center and a C-terminal [2Fe-2S] cluster that is redox sensitive. Under iron starvation conditions, the Hr motif at the N-terminus of FBXL5 cannot bind iron, and as a consequence, the protein undergoes a conformational change leading to destabilization, ubiquitination, and degradation of the entire protein. In this situation IRP2 is stabilized, and cognate mRNAs with target 3’ stem loops are also stabilized e.g. TFR1, whereas cognate mRNAs with target 5’ stem loops are repressed, e.g. ALA synthase. As iron levels rise in the cell, the Hr motif is stabilized and the [2Fe-2S] cluster in the C terminus of FBXL5 is stabilized. In the presence of adequate oxygen tension, the FBXL5 cluster becomes oxidized [2Fe-2S]²⁺. The oxidized cluster-containing FBXL5, in effect acting as a ubiquitin ligase, binds to IRP2 and stimulates ubiquitination and degradation. The turnover of IRP2 enables translation of IREs at the 5’ end and leads to destabilization TFR1 mRNA [93]. It should be noted that the high iron signal is conveyed by FBLX5 and IRP2 turnover, which depend on the presence of the oxidized cluster on FBXL5; thus conditions impairing cytoplasmic [2Fe-2S] cluster formation such as mitochondrial dysfunction would be expected to stabilize IRP2 [93].

10. Human diseases associated with FeS cluster deficiency

An X-linked inherited human disease, XLSA-A (X linked sideroblastic anemia and spinocerebellar ataxia), was mapped to the locus of the human homolog of Atm1, termed ABCB7 [94]. This disease is characterized by the presence of ringed sideroblasts in the red cell precursors of the bone marrow. In these cells, iron carried in mitochondrial ferritin [95] or as iron phosphate nanoparticles [96], accumulates within mitochondria of the red cell precursors leading to premature turnover and causing anemia. On the other hand, porphyrin synthesis in these cells is deficient due to a cytoplasmic block in erythroid ALA synthase production secondary to activation of the IRE binding protein, which in this situation lacks its FeS cluster cofactor and functions as a repressor of protein translation [97]. A particularly instructive mutation, E433K, causes inherited sideroblastic anemia and when mimicked in yeast Atm1 (i.e., D398K) leads to impairment in the maturation of cytosolic FeS cluster proteins [74]. Note this same mutation corresponds to a glutathione coordinating residue in the Atm1 crystal structure, thus reemphasizing the functional importance of GSH binding for Atm1 function [70].

Mutations in the human genes for GRLX5 or BOLA3 have also been linked to diseases. The inherited GLRX5 associated diseases exhibit allele dependence. One such disease, SIDBA-3 (sideroblastic anemia, pyridoxine refractory, OMIM 616860) is caused by loss-of-function of the protein. In one case, a splicing defect led to undetectable GLRX5 protein levels [98]. In two other cases of compound heterozygotes (K101Q/L148S in one patient and C69Y/M128K in another), FeS cluster binding to GLRX5 was compromised [99]. The ensuing disease, called sideroblastic anemia, is characterized by anemia and the presence of ringed sideroblasts in the bone marrow (see above). The other allelic disease of GLRX5 is called SPAHGC or childhood onset spasticity with hyperglycinemia (OMIM 616859) [100]. The mutant allele in these individuals carries a deletion of a highly conserved Lys51 located on the surface of the GLRX5 protein and interfering with FeS cluster insertion into LIAS. The disease is secondary to accumulated glycine, due to defective glycine cleavage enzyme, which is dependent on LIAS for production of its lipoic acid cofactor. The high glycine level leads to neurologic toxicity, seizures, spasticity, arrested development and is most often lethal. Note that the different GLRX5 alleles give rise to completely different disease phenotypes due to the ability of GLRX5 to act in multiple pathways and interact with multiple partners relevant to mitochondrial and cytoplasmic FeS cluster assembly. BOLA3, a partner protein for GLRX5, is also associated with a human disease. Homozygous mutation in BOLA3 leads to a disease called MMDS2, multiple mitochondrial dysfunction syndrome 2 with hyperglycinemia (OMIM 613183). The phenotype/disease is like the GLRX5 Lys51 mutant allele phenotype, in that LIAS is defective and high blood glycine level causes defective neurologic functions. The LIAS defects lead to deficiency of PDHc (pyruvate dehydrogenase) and αKGDHc (alpha ketoglutarate dehydrogenase). Also decreased are complex I and II activities in the respiratory transport chain, which require FeS cluster cofactors [101].

11. Evolutionary Tinkering

Several FeS cluster assembly systems i.e., NIF, ISC, SUF, and CIA have been discovered since Dennis Dean identified Nifs in Azotobacter vinelandii [2]. They share certain features and general outlines: a cysteine desulfurase interacts with the amino acid cysteine and generates a persulfide using PLP chemistry, then transfers the persulfide sulfur to a scaffold protein. This is the sulfur source. Ferredoxin or ferredoxin like cofactors provide electrons from NADPH that are needed to reduce sulfane sulfur (S⁰) to sulfide (S²⁻) as part of nascent FeS clusters. Iron joins at this stage, although a specific iron chaperone has not been found. Also, the “iron first” or “sulfur first” conundrum has never been answered definitively, although data in support of an “iron first” scenario has been presented [23]. Specialized components then mediate dislocation and transfer of a cluster intermediate from the scaffold protein to intermediate recipients e.g., Grx5, which coordinates the intermediate between dimers and glutathione moieties. Next, a [4Fe-4S] assembly system e.g., Isa1/Isa2 mediates formation of [4Fe-4S] clusters from [2Fe-2S] cluster intermediates by reductive fusion. Finally, targeting factors such as Cia1, Cia2 and Mms19 receive the FeS cluster intermediates and interact with specific target apoproteins, inserting nascent FeS clusters into them [2].

NIF (nitrogen fixation) is a specialized system adapted for high volume, high output synthesis of FeS clusters in nitrogenases of heterotropic organisms such as A. vinelandii [102]. The ISC (iron sulfur cluster) operon, similar to the NIF system, was found to encode a housekeeping system for basal production of FeS clusters in many organisms including A. vinelandii and E. coli [13]. A highly homologous ISC system was found to reside in mitochondria, consistent with the origins of mitochondria by an endosymbiotic event in which a purple proteobacterium was engulfed by an archaea host [103]. Similarly, the SUF (sulfur fixation) system originated within unicellular algae that was engulfed in a distinct endosymbiotic event, forming plastid organelles. This system also contains cysteine desulfurase (SufS), scaffold (SufBD), trafficking component (SufF), and targeting factor (Nfu) [104]. The SUF and ISC systems are both functional in E. coli, but SUF activity predominates under conditions of oxidative stress or iron starvation, whereas the ISC is repressed under those conditions. The simultaneous interruption of SUF and ISC in E. coli is synthetically lethal [105]. Finally, an entirely separate system was found in the eukaryotic cytoplasm called CIA (cytosolic iron-sulfur cluster assembly). This system seems to have originated in the LECA (last eukaryotic common ancestor), and it appears throughout most eukaryotic branches of life [106]. It also exhibits some of the same basic elements as the other systems. The source of the sulfide is likely to be indirect from the mitochondrial ISC, rather than an independent cysteine desulfurase in the cytoplasm, but CIA has its own unique scaffold proteins (Cfd1/Nbp35, or sometimes just a duplicated Nbp35), electron donor (Dre2/Tah18), trafficking protein (Ciao3 or Nar1), and targeting factors (Ciao1/Ciao2A and B/Mms19) [106].

We can learn a great deal about the biological functioning of the various FeS assembly systems by examining how they interact in different evolutionary settings (Fig. 10). In many eukaryotic cells such as yeast, the ISC makes the (Fe-S)_int intermediate, which is trafficked via Grx5-GS to Atm1 and exported perhaps as a glutathione-containing intermediate to the cytoplasm, where it interacts with Grx/Bol components and possibly other CIA components [30]. During evolution, there are many variations on this theme, however. Evolutionary reduction of mitochondria is a process that has been observed to occur in many evolutionary branches, whereby mitochondrial components are lost, leaving minimal and highly reduced organelles [106]. Reduced mitochondria or mitochondria-related organelles (MROs) are found in distant branches of the evolutionary tree of eukaryotes. Consider mitosomes, the MRO of microsporidia (Fig. 10). Microsporidia are related to fungi and live as intracellular pathogens in animal cells. Most mitochondrial functions have been lost, including the tricarboxylic acid cycle, electron transport chain, ATP synthesis, and heme synthesis. However, these organelles still retain a core ISC machinery consisting of Nfs1, Isu1, Hsp70/Jac1, and Arh1/Yah1 redox couple. An Atm1 homolog is present, making it likely that the (Fe-S)_int intermediate is made in this mitosome and exported to cytosol just as in yeast and other eukaryotes. The microsporidia cytoplasm expresses a reduced CIA system, consisting of Tah18/Dre2, Nbp35, Cfd1, Nar1, Cia1/Cia2, and target FeS proteins. Thus, mitosomes have maintained a core ISC system and mitochondrial export system (Atm1 and glutathione) in the setting of a reduced CIA pathway, despite having lost most other mitochondrial functions [29] (Fig. 10).

Fig. 10. — A) *Trypanosoma brucei* *T. brucei* is a flagellate belonging to the supergroup Excavata. It is the causative agent of African sleeping sickness, existing in two states, the procyclic with an active mitochondrion and the pathogenic bloodstream stage with a metabolically repressed organelle. The mitochondrial FeS cluster biogenesis machinery is overall similar to that present in yeast and humans. Mitochondria possess a cysteine desulfurase IscS, scaffold IscU, frataxin homolog, Grx5 equivalent, and Atm1 transporter. Direct evidence for (FeS)_int does not exist, but knockdown of IscS and IscU led to reduced activity of aconitase in the cytoplasm, suggesting a functional link. In the cytoplasm, the Cia components resemble those of yeast and humans with reductase (Tah18/Dre2), scaffold (Nbp35/Cfd1), intermediate (Nar1) and CTC (Cia1, Cia2, Mms19) [106]. **B) Microsporidia (mitosome)**. Microsporidia are related to fungi and live as intracellular pathogens in animal cells. They have undergone an evolutionary process called reduction, according to which mitochondria have lost many functions such as ETC, TCA, and ATP synthesis. The FeS cluster assembly functions are reduced (compared with yeast and humans) with ISC containing only early but not late components. The early components include Nfs1, Isu1, Hsp70/Jac1, and Arh1/Yah1 redox comple; Isa1/2 and Grx5 are not there. An Atm1 homolog is present as well as a CIA pathway. The CIA pathway includes reductase (Tah18/Dre2), scaffold (Nbp35/Cfd1), intermediate (Nar1) and part of the CTC (Cia1/Cia2 but not Mms19). The Atm1 homolog can be presumed to link the ISC and CIA as in yeast and humans, but there is no direct experimental evidence for this [29]. **C) Pygsuia (MRO)**. This is an anaerobe belonging to the Breviatae classification. It has an extremely reduced MRO (mitochondrion-related organelle) content with only a few FeS cluster transfer components such as Nfu1 and Ind1. Apparently in parallel there exists a SUF machinery, present in both mitochondrial and cytoplasmic compartments. There is no Atm1 homolog, and presumably there is no FeS intermediate exported from mitochondria to the cytoplasm. Exactly how the FeS cluster assembly pathway works is unclear as there are no experimental data. We can speculate tht the SUF components in mitochondria generate an FeS intermediate used by Nfu1 and Ind1 transfer components within the MRO, whereas SUF components in the cytoplasm interact with the rudimentary CIA, consisting of Nbp35/Cfd1, Nar1, and CTC [2]. **D) Oxymonads**. Oxymonads are eukaryotes but they completely lack mitochondria or MROs. This has been confirmed by genome sequence analysis which reveals no mitochondrial protein import machinery and no ISC components. SUF and CIA elements are predicted to be present, however, in the cytoplasm of this organism [2]. How the FeS cluster assembly works in this organism is unclear as there is no experimental work. Presumably SUF generates the FeS intermediate on its own scaffold, and this is transferred to the rudimentary CIA (NBP35 and CIAO3), and from there to recipient apoproteins [2].

Entamoeba are quite different. Entamoeba are parasites of animals living within the intestines and found in the stool of infected hosts [2]. No Atm1 homolog has been identified for these organisms. The mitochondrial ISC system has been supplanted by a NIF system, containing NifS, and tripartite NifU, consisting of scaffold protein, ferredoxin, and targeting components. Curiously, the NIF components are duplicated and expressed in both the mitochondria and cytoplasm. We speculate that in this organism, mitochondria and cytoplasmic FeS cluster assembly systems are segregated, without the epistasis of ISC and CIA that appears in yeast and human cells. The mitochondrial NIF components may generate intermediates for mitochondrial FeS clusters, and the cytoplasmic components may generate intermediates for cytoplasmic FeS clusters. Perhaps cytoplasmic NIF components make a compound or FeS cluster precursor analogous to (Fe-S)_int of yeast and then deliver it directly to downstream CIA components in the cytoplasm. An Atm1 transport step would be unnecessary here. More biochemical and genetic work will be needed to define the mechanisms/pathways of FeS cluster assembly in Entamoeba.

Another solution to the compartment problem for eukaryotic FeS cluster assembly occurs in Blastocystis, a parasite that can live in the gastrointestinal tracts of many types of animals including humans [2, 106]. The MROs of this organism express an Fe-Fe hydrogenase, which is necessary for hydrogen and energy production. To synthesize FeS clusters for this protein, MRO and cytoplasmic machineries exist, linked by an Atm1 homolog. In the MRO, a core ISC machinery is present, including components from Nfs1 to Grx5, and also a late ISC machinery from Isa to Ind1 (insertase). In the cytoplasm, Blastocystis species express a minimal SUF machinery consisting of SufCB. There is also a reduced CIA machinery including Nbp35 and Cfd1, Nar1, and Cia1/Cia2. In terms of how this works, primary experimental data are lacking. No genetics exists for these species, and one wonders if Atm1 is really necessary for delivering the (Fe-S)_int from ISC to cytoplasm, because the cytoplasmic SUF machinery might directly generate the (Fe-S)_int. in the cytoplasm, obviating the need for mitochondrial synthesis and Atm1 dependent transport.

Another protist organism, Pygsuia biforma, is an anaerobe belonging to the Breviatae classification. It has an extremely reduced MRO content, with only a few SUF components, SufCB, and transfer components Nfu1 and Ind1. No Atm1 or homologous transporter has been identified. Pygsuia expresses a cytosolic SufCB, and a rudimentary CIA consisting of Nbp35/Cfd1, Nar1, and CTC (CIA targeting complex) components. In this case, the MRO, which has its own SUF, and cytosol-localized SUF systems likely function completely independently, as no Atm1 exists to mediate cooperation. The cytosolic SUF components likely perform the function of the mitochondrial ISC generating an (Fe-S)_int intermediate or equivalent, but in this case the generator and target are in the same compartment and cooperate directly. The origins of iron, sulfur and electrons are unknown [2, 106] (Fig. 10).

Until 2016, it was thought that mitochondria or some forms of MRO were essential for viability of all eukaryotic organisms. Then the Oxymonad, Monocercomonoides exilis, was discovered. M. exilis is a symbiont inhabitant of the intestines of small mammals, belonging to the Excavata supergroup. The sequencing of the genome gave support to the assertion that it lacks mtDNA and mitochondrial proteins. The functions of even the most rudimentary MROs were not present, including mitochondrial protein import components and mitochondria localized FeS cluster assembly factors. The genome, however, was not devoid of FeS cluster assembly components. A curious assortment of components was deduced based on genome sequence analysis. The oxymonads were found to express a SUF machinery consisting of SufDSU and SufB/SufC scaffold, and a rudimentary CIA pathway containing Nbp35 (no Cfd1), Ciao3 and Ciao1/Cio2AB insertases. Here the bacterial SUF system, likely acquired by horizontal transfer, performs the function of the mitochondrial ISC plus Atm1, in generating precursor or intermediate, that then proceeds through the steps of cytoplasmic FeS cluster synthesis. The SUF and CIA machineries, which coexist in the same cellular cytoplasmic compartment, must directly cooperate. SUF acts upstream of CIA, perhaps generating a persulfide sulfur or iron-sulfur intermediate. No biochemical or reconstitution experiments exist to explain in detail how this happens. One thing we do not see in the scrambling of FeS cluster assembly components that occurs in evolution is the relocalization of the ISC to the cytoplasm. In humans and yeast, a small population of ISC components may exist outside of mitochondria, although the functional significance remains uncertain [107] (Fig. 10).

Trypanosoma brucei is an important protozoan pathogen belonging to the Excavata supergroup, with world-wide distribution [108]. It causes sleeping sickness and is transmitted by the bite of the tsetse fly, thus there are fly specific developmental and reproductive stages. It expresses prominent down-regulated mitochondria in the blood-stream stage and active procyclic mitochondria in the insect phase of reproduction in the digestive tract of the tsetse fly. As distinct from many of the other organisms mentioned above, T. brucei has been developed as an experimental organism, in that both phases can be cultured in vitro, and effective and apparently simple RNAi knockdown technology exists. FeS clusters are abundant, especially in the procyclic phase, and informative knockdown experiments have been performed [109]. The organism has a well-developed and well-characterized mitochondrial ISC machinery, Atm1 homolog, and cytoplasmic CIA machinery. Ablation of TbAtm1 affects the downstream CIA machinery with a disrupted FeS cluster incorporation into cytoplasmic proteins [110]. In one set of experiments [109], knockdowns of FeS cluster assembly proteins IscS and IscU downregulated the active mitochondrion of the procyclic form. Knockdown of the Nfs1 homolog was associated with reduced activity of the FeS enzyme aconitase in both mitochondria and cytoplasm [110]. These data from T. brucei suggest that the yeast model applies well here, according to which mitochondrial ISC makes the (Fe-S)_int intermediate that is exported to the cytoplasm and contributes to cytoplasmic FeS cluster assembly (Fig. 10).

General rules are hard to discern for combinations of FeS cluster assembly systems in eukaryotes. It seems that two systems are needed. An upstream system is required that makes an intermediate and feeds it to the downstream system, almost always a CIA pathway system. The intermediate may be a nascent cluster or some component thereof. However, if the upstream system is ISC, it most often segregates within a separate mitochondrial compartment, and these organisms usually also carry an Atm1 homolog, suggesting that ISC cannot work with CIA directly in the same compartment. Generally speaking, ISC [111] must be in mitochondria and CIA must be in cytoplasm. Atm1 is needed to move the ISC product (Fe-S)_int from mitochondria to the cytoplasm. It is not clear why this arrangement is selected for during evolution. If the upstream system is SUF or NIF, this prohibition does not apply. The SUF or NIF machinery can work directly with the CIA in the same compartment (see Blastocystis and Entamoeba for example). In those cases, the critical intermediates may be made by the upstream components and directly handed off to the CIA component (usually Nbp35). Missing from the picture are details and direct biochemical experiments demonstrating these relationships.

12. tRNA thiolation in mitochondria and cytoplasm

tRNAs are essential for genetic code decoding, a fundamental process for all living organisms. All tRNAs feature chemical modifications that stabilize their structures and fine tune the decoding process [78]. In yeast, sulfur modification of tRNAs has been shown to function as a sensor for sulfur availability, providing a link between protein translation and metabolism [112]. Formation of 2-thiouridine (S²U) in position 34 of the anticodon is particularly important, as sulfur modification of uridines in the wobble position of tRNA anticodons influences ribosomal A-site binding, thus enhancing translational efficiency and accuracy for certain proteins [78]. In eukaryotes, sulfur modification of tRNAs is important in both mitochondria and cytoplasm for optimal protein translation to occur. In the cytoplasm, the codon triplet in messenger RNA (codon positions 1, 2, and 3) base pairs with anticodons in the tRNA (positions 36, 35, and 34, respectively, in the anticodon of tRNAs for amino acids Lys, Glu, and Gln). The lack of the 2-thio uridine34 modification alters the tRNA structure and creates steric hindrance, interfering with translational efficiency [78]. Thiolation of tRNAs, both mitochondrial and cytoplasmic, is critically important and mutations that interfere with tRNA thiolation confer nonviability (Fig. 11A).

Fig. 11. — **A) tRNA thiolation site**. In the cytoplasm of eukaryotes, the wobble uridine (U34) of three tRNAs specific for lysine, glutamate, and glutamine contain a methoxycarbonylmethyl (mcm5) functional group and a thio-modification (S²) at positions 5 and 2 respectively. Loss of these modifications reduces the rate of translation and causes protein aggregates, because nascent proteins fail to adopt their proper conformations [78]. **B) Pathway for cytoplasmic tRNA thiolation**. There are two distinct intermediates, (S_int) and (Fe-S)_int, exported by Atm1, mediating cytosolic tRNA thiolation or FeS cluster assembly, respectively. The mitochondrial cysteine desulfurase complex generates persulfide from cysteine. Bifurcating from the FeS cluster biogenesis machinery, there is a sulfur pathway that is Isu1/2 but not Ssq1 dependent. The mitochondrial Nfs1 derived persulfide forms the S_int intermediate, which is exported from the mitochondrion via Atm1, enters a sulfur relay pathway consisting of Tum1, Uba4 and Urm1, ending with S²U₃₄ thiolation of cytosolic tRNAs for Glu, Gln and Lys. A separate Nfs1 persulfide pathway is chaperone (Ssq1) dependent, generating a distinct intermediate, (Fe-S)_int, which also exits the mitochondrion via Atm1, then enters the CIA pathway for generation of [4Fe-4S] clusters. The picture is further complicated by the fact that there are FeS cluster proteins such as Elp3 and Ncs6 (Ncs2) that mediate tRNA thiolation in the cytosol. Note this diagram is essentially reproduced from the article by Braymer and Winge [79].

Recall that Nfs1 is a pyridoxal phosphate dependent enzyme that interacts with the amino acid cysteine, generating a protein bound persulfide that can be delivered for various uses. These uses include tRNA thiolation in mitochondria and cytoplasm, as well as FeS cluster assembly in mitochondria and cytoplasm. In a simple but telling experiment, Nakai et al. [77] used a promoter swap Gal-Nfs1 strain that was induced or repressed for Nfs1 expression, and then briefly radiolabeled the cells with [³⁵S]cysteine. Total tRNA was extracted, and the ³⁵S-labeled tRNAs were separated by PAGE. Only the cells expressing Nfs1 exhibited tRNA radiolabeling, indicating that sulfur modification of tRNA labeling was dependent on Nfs1. Importantly, Nfs1 was required for thiolation of both mitochondrial and cytoplasmic tRNAs. This was shown by the following experiment. Nfs1 expression was induced or repressed, then total tRNA was extracted and analyzed by APM [(N-Acryloylamino)phenyl] mercuric chloride) gels. APM binds specifically to the thiolated forms of tRNA, and retards migration in 8M urea gels. The gel separated/retarded thiolated tRNA was then probed with specific DNA labeled probes for mitochondrial or cytoplasmic tRNAs, and both forms hybridized with their targets: mt-tRNA^LYS_UUU, mt-tRNA^GLN_UUG, cy-tRNA^LYS2_UUU, and cy-RNA^GLU3_UUC. In all these cases, the signal for the shifted/modified tRNA was present in the Gal-Nfs1↓ up condition and attenuated in the Gal-Nfs1↓ repressed condition [77].

The eukaryotic Nfs1 is primarily located within mitochondria, although low levels of Nfs1 were reported in the mammalian cytoplasm [59], and a small population of extramitochondrial Nfs1 was reported to reside in the nucleus [77]. This raises the question of which population of Nfs1 is responsible for the thiolation of tRNAs. The APM experiments show that Nfs1 is involved in modifications of both mitochondrial and cytoplasmic tRNAs. A surprising result was that the mitochondrial Nfs1 was responsible for the cytoplasmic tRNA thiolation. The following three-step experiment makes use of isolated mitochondria to demonstrate this point [60]. Isolated mitochondria were supplemented with [³⁵S]cysteine, in the presence of ATPγS, a non-hydrolyzable ATP analog. This blocks function of Atm1 since the latter depends on ATP hydrolysis for its reaction cycle, and thus no intermediates can be exported from mitochondria to cytoplasm. In the second step, the isolated mitochondria were recovered and washed to remove excess/free [³⁵S]cysteine and the ATPγS. ATP was then added back to reverse the Atm1 transport block. In the third step, the exported material was added to the cytoplasm and ³⁵S-tRNA thiolation was observed, without any further requirement of mitochondria. In summary, mitochondrial Nfs1 generated a sulfur intermediate (called S_int) that was exported by Atm1 and subsequently utilized for tRNA thiolation in the cytoplasm [60]. Note that isolated cytoplasm by itself could not thiolate tRNAs because no ³⁵S-tRNAs were detected when isolated cytoplasm alone (no mitochondria) was directly incubated with [³⁵S]cysteine.

12.1. Overlap with FeS cluster biogenesis pathways

Nfs1 is the cysteine desulfurase that contributes sulfur as a persulfide sulfur intermediate for tRNA thiolation occurring both inside and outside of mitochondria. Nfs1 also contributes sulfur for FeS cluster assembly. The requirements for tRNA thiolation and FeS cluster assembly overlap but are not identical. The nfs1-14 strain, expressing a hypomorphic NFS1 allele, was deficient in cytoplasmic tRNA thiolation and cytoplasmic FeS cluster assembly (Fig. 11B). Likewise, the Gal-Isu1/Δisu2 strain was deficient in both cytoplasmic tRNA thiolation and FeS cluster assembly. However, surprisingly, a ssq1 mutant with strong defects in FeS cluster assembly was completely normal for cytoplasmic tRNA thiolation [60]. We believe therefore that there are two distinct intermediates, one for tRNA thiolation (S_int) and one for FeS cluster assembly ((Fe-S)_int). Formation of both these intermediates are dependent on Nfs1 and Isu1/2, but formation of only the (Fe-S)_int intermediate additionally requires Ssq1 function. Each intermediate is separately exported from mitochondria to cytoplasm via Atm1 [79] (Fig. 11B). This notion is supported by the observation that atm1 mutant mitochondria, when added to isolated WT cytoplasm, cannot efficiently promote cytoplasmic tRNA thiolation or Fe-S cluster assembly [16, 17]. Further evidence for the existence of distinct intermediates was provided by the effects of iron deprivation. Iron deprivation (exposure of the isolated mitochondria to o-phenanthroline) abrogated the FeS intermediate ((Fe-S)_int) but had no effect on the tRNA thiolation intermediate (S_int) [17]. The pathways for FeS cluster assembly and tRNA thiolation are closely intertwined. An added layer of complexity is the discovery of the elongator complex, which is required for 2-thiolation of uridine in cytoplasmic tRNAs, and yet itself contains cytosolic FeS proteins such as Elp3 and Ncs6 [112] (Fig. 11B).

13. Evidence for involvement of core mitochondrial ISC components and Atm1 in cytoplasmic FeS cluster assembly and iron homeostasis

A prediction of the mitochondria/cytoplasm epistasis noted above is that depletion or deletion of core components of the mitochondrial ISC would lead to deficiencies of cytoplasmic FeS cluster assembly. We reviewed published evidence for this assertion and found that Nfs1, Isd11, Acp1, Yah1, Yfh1, Isu1/2, Ssq1, Grx5, Atm1 had been tested and did indeed exhibit this phenotype. However surprisingly, Jac1, and Mge1, were not previously tested. Therefore, we performed the requisite experiments (see below), confirming that all 10 core mitochondrial ISC components and Atm1 were necessary for both mitochondrial and cytoplasmic FeS cluster assembly.

Nfs1: 1) Nfs1-depleted cells were grown with ⁵⁵Fe nuclide added. After immunoprecipitation, virtually no ⁵⁵Fe signal associated with cytoplasmic Leu1 protein was detected in these mutant cells [14]. Similarly, permeabilized cells from a hypomorphic Nfs1 allele were fractionated and a mito/cyto mixture had negligible loading activity for apo-Leu1^R [16]. 2) Nfs1-depleted cells or the Nfs1 hypomorphic allele nfs1-14 showed the characteristic iron homeostatic abnormality, high cellular iron uptake, and mitochondrial accumulation of an insoluble form of iron [55].

Isd11: 1) Cellular depletion of Isd11 also led to a strong decrease in ⁵⁵Fe incorporation into Leu1 in vivo [24]. The activity of Leu1 was two to three-fold decreased in extracts of isd11–1 cells (temperature sensitive mutant of Isd11), whereas the activity of the control enzyme alcohol dehydrogenase did not change significantly. The Leu1 protein levels were comparable [113]. 2) After promoter swap of Isd11, the protein was depleted, and iron homeostasis was examined. The depleted strain showed high cellular iron uptake and mitochondrial iron accumulation [25].

Acp1: 1) Promoter swap was used to deplete cells of Acp1, and to further assess perturbation of cytosolic FeS function, Van Vranken et al. quantified the cytosolic FeS-containing enzyme sulfite reductase and observed a diminution [27]. We performed a similar promoter swap and mitochondria were isolated from the Acp1 depleted condition. In the mixing assay of Gal-Acp1↓ mito and Δleu1 cyto, only minimal activity for apo-Leu1^R loading was noted (2.5 units compared with 12.5 units for mixing with WT mito) (Pandey A. et al., manuscript in preparation). 2) Acp1-depleted cells exhibited the characteristic iron homeostatic abnormality: activation of the iron regulon and mitochondrial iron accumulation in an insoluble form [27].

Isu1/Isu2: 1) Isu (Isu1 and Isu2) were depleted in mitochondria of a Gal-Isu1/Δisu2 strain. The depleted strain showed deficient Leu1 FeS cluster loading in the cytoplasm as assessed by ⁵⁵Fe labeling and immunoprecipitation. Furthermore, the depleted strain accumulated iron mitochondria as observed for other FeS cluster assembly mutants [114]. 2) We used an alternative approach. We analyzed the effects of an M141E hypomorphic allele of Isu1 in the context of Δisu2 deletion. The mitochondria from this strain mixed with WT cytoplasm failed to load apo-Leu1^R efficiently in a biochemical assay (3.7 units versus 12.5 units reconstituted Leu1 activity for the WT) (our unpublished observations). The M141E Isu1 allele conferred characteristic iron homeostatic abnormalities of high cellular iron uptake and mitochondrial iron accumulation [115].

Yah1: 1) Depletion of Yah1 led to a marked reduction, up to a 10-fold decrease, in the isopropylmalate isomerase activity of Leu1 [116]. 2) Yah1 depletion by regulated gene expression resulted in a 30-fold accumulation of iron within mitochondria, similar to what has been reported for other components involved in FeS cluster synthesis [116].

Yfh1: 1) Gal-Yfh1 cells were grown under repressing condition for Yfh1 expression, and incorporation of radioactive ⁵⁵Fe into the cytoplasmic FeS proteins Leu1 and Rli1 were followed. The amount of ⁵⁵Fe incorporated into these proteins was reduced to 40% of wild-type levels. Immunoblotting analyses established that Leu1 and Rli1 proteins were present at wild-type levels. Moreover, the cells were impaired in the biosynthesis of amino acids that involved cytoplasmic FeS proteins, indicating a general defect in FeS protein assembly in the cytoplasm [117]. 2) Gal-Yfh1 cells showed accumulation of iron in mitochondria up to 24 nmol Fe/mg protein of mitochondria [117].

Ssq1: 1) Mitochondria were isolated from Ssq1-depleted cells (Gal-Ssq1) and mixed with Δleu1 cytoplasm and recombinant apo-Leu1^R apoprotein. Essentially no Leu1 activity was reconstituted. By comparison, WT mitochondria mixed with Δleu1 cytoplasm and apo-Leu1^R yielded 7 units of Leu1 activity [16]. In another experiment, mitochondria from 33C, carrying a hypomorphic allele of Ssq1, were mixed with Δleu1 cytoplasm and apo-Leu1^R. Little or no (0.3 units) reconstituted activity was measured. Controls with mixtures of Δleu1 mitochondria plus Δleu1 cytoplasm plus apo-Leu1^R yielded high activity for isopropylmalate dehydrogenase (8.5 units) (this work, Fig. 4B). 2) A loss-of-function mutant allele of Ssq1 was identified using the FRE1-HIS3 selection. This mutant grown in different concentrations of media iron (0.9, 1.8 or 5 μM) showed increasingly dramatic (>100x) accumulation of mitochondrial iron compared with wild-type indicating loss of iron homeostasis [118].

Jac1: 1) Mitochondria were isolated from a loss-of-function mutant allele of Jac1, called jac1-6.3 [119]. When these mutant mitochondria were mixed with Δleu1 cytoplasm and apo-Leu1^R, markedly decreased Leu1 activity was reconstituted (2 units compared with 12.5 units for WT mito)(our unpublished observations). 2) A mutant allele of Jac1 was isolated after selection of FRE1-His3 activity, indicating that it conferred loss of iron homeostasis. As expected, growth of the mutant in various concentrations of iron resulted in dramatic mitochondrial iron accumulation [119].

Mge1: Mge1, nucleotide exchange factor, was able to form a specific complex with Ssq1. Perturbation of Mge1 function or amount resulted in direct effects on Ssq1 activity [34]. Therefore, a prediction could be made that Mge1 lack would phenocopy Ssq1 loss-of-function. In accordance with this idea, a Gal-Mge1 promoter swap strain was constructed. When Mge1 expression was repressed, the corresponding mitochondria showed very low aconitase activity and inefficient reconstitution of apo-Leu1^R activity in the mixing assay with Δleu1 cytoplasm (7.5 units compared with 13 units for the WT mito) (our unpublished data).

Grx5: 1) The Δgrx5 deleted cells exhibit auxotrophy for the amino acids leucine, lysine and glutamic acid. Each of these amino acids requires FeS enzymes for their biosynthesis [120]. 2) Absence of Grx5 promotes iron accumulation in the cell and loss of iron homeostasis [120].

Atm1: 1) The Gal-Atm1 strain was used to deplete Atm1 protein, and ⁵⁵Fe labeling of Leu1 was performed. In the depleted cells, the recovered label was decreased to negligible levels (equivalent to cpm recovered with preimmune serum) [14]. A mixture of mitochondria from a Δatm1 strain and Δleu1 cytoplasm plus apo-Leu1^R failed to reconstitute 3-isopropylmalate isomerase activity (<1 unit) whereas Δleu1 mitochondria and Δleu1 cytoplasm plus apo-Leu1^R conferred high level activity (8 units) [16]. 2) The ABC transporter Atm1 is required for mitochondrial iron homeostasis [50].

14. Outlook

The field of the biochemistry of FeS protein biogenesis has advanced dramatically, since discovery of the catalytic formation of nitrogenase by Dennis Dean [12]. The compendium of FeS proteins has expanded rapidly. Frequent discovery of new clusters occurs, in the setting of novel regulatory mechanisms (e.g. FBXL5) [93], new protein structures (e.g. ribosomal structures) [6], and targeted genetic approaches to FeS cluster discovery in whole genomes [9]. This process of discovery of novel FeS proteins promises to continue in the years ahead. The components and pathways mediating FeS cluster biogenesis in various organisms have been mostly discovered, but many details and many aspects remain to be fleshed out. For example, the epistasis of mitochondria in cytoplasmic FeS cluster assembly and tRNA thiolation points to the existence of intermediates that function in eukaryotic FeS cluster assembly and tRNA thiolation [17]. These need to be isolated, identified, and chemically characterized. Hopefully this will happen in the next few years.

Most of the pieces of FeS protein biogenesis pathways are now known, but some may be missing. It is astonishing that new components continue to be discovered after all these years (see discovery of Acp1 as a core ISC component, after it was observed to be present in Nfs1 crystal structures [19, 21]). There may be additional genes/components mediating FeS cluster formation and trafficking that remain to be discovered. The story of iron delivery into the FeS cluster assembly pathway via Isu1/Isu2 is incomplete. The role of frataxin/Yfh1 as an iron chaperone [121] may or may not hold up, and it is possible that some other iron chaperone will be found. We uncovered a role for GTP in formation of the scaffold intermediate on Isu, but the target GTPase(s) or the intervening step(s) remain to be identified [15]. Perhaps in the future this novel FeS cluster assembly component will be more clearly identified and characterized. What happens downstream after export of the (Fe-S)_int intermediate to the cytoplasm is not entirely clear and needs more work and definition. If the chemistry of the intermediates is better defined, perhaps it will be possible to synthesize active compounds that bypass mutations occurring in some of the iron-sulfur cluster diseases. Perhaps these compounds could be developed as therapeutics.

A picture of a generic FeS cluster assembly machinery has been developed, but there are still many unknown variations and peculiarities in different organisms. Very interesting permutations on the theme appear in various protist organisms, many of which do not have experimentally pliable genetic systems. It will be interesting in the future to define how FeS cluster synthesis occurs in eukaryotic organisms such as Pygsuia species, which completely lack an Atm1 homolog [106]. Likewise, in differentiated human tissues (e.g., nerve and brain), specialized components for FeS cluster assembly are likely to exist, and specialized regulatory signals are likely to function. New work will continue to define these specializations [122].

The structural aspects of the FeS cluster assembly process have come into view much better than before. However, much remains to be done. With new cryo-EM methods, we can expect definition of new protein complexes and FeS cluster assembly “machines”. The Nfs1/Isu dodecamer has been studied by crystallography, cryo-EM, SAXS, and protein cross linking [19]. We expect more structures of the chaperone complexes to be solved. The Grx5 complexes and various interacting subcomplexes need more structural definition. The Isa [4Fe-4S] cluster biosynthesis machine needs structural definition. Bol/Grx heterocomplexes exist in various iterations and in various compartments and play roles in regulation and FeS cluster trafficking. These need better structural characterization. No structure yet exists for Grx/Bol complexes to our knowledge. The CIA pathway for FeS cluster assembly is made up of subcomplexes [123] that associate and dissociate as part of the cytoplasmic FeS cluster biosynthesis process. These complexes were glimpsed in proteomic studies, but they could benefit from better structural characterization (e.g., Dre2/Tah18 reductase, Npb35/Cfd1/Nar1/CTC complexes). A CTC crystal structure bound to the DNA replication factor primase or to the DNA helicase DNA2 were recently reported [124], but the structures lack the client FeS cluster. Perhaps in future the actual FeS cluster transfer from CTC to client protein can be captured in a cryo-EM.

Regulatory controls for FeS cluster assembly still need to be discovered. Iron-sulfur clusters probably are made in response to biological need, and thus feedback from apoproteins to the biosynthetic machinery can be expected. Some controls have been discovered. For example, iron dependence of Isu expression via Aft1/2 has been described and it makes perfect sense, since iron is needed for FeS cluster assembly [125]. Some level of metabolic regulatory control is provided by fatty acid stabilization of the Acp1-Isd11 interaction [1, 26, 27]. In mammalian cells, microRNA-210 may impact on the expression of ISCU [126]. Additional controls are likely to exist. These have not yet been found and still need to be discovered and characterized. New methods for tracking new FeS cluster formation [127] may help to delve into these regulatory networks.

Finally, we can expect progress in understanding and treatment of FeS cluster related human diseases. Most likely, the scope of human pathophysiology subservient to FeS proteins will expand as knowledge of new diseases develops. We know that some forms of neurodegeneration arise from frataxin deficiency that leads to defects in FeS cluster biogenesis [128]. The solution for Friedreich’s ataxia may be to correct the defective gene or to replace the deficient protein in mitochondria. But we might also imagine alternative therapeutic interventions - a downstream frataxin bypass mutant, the so-called Isu suppressor [115], might be deployed as a therapeutic solution in diseased tissues. Alternatively, if biologically active and deliverable FeS cluster modules can be designed, FeS cluster replacement therapy might cure FeS cluster deficiency diseases. Sideroblastic anemia is characterized by bone marrow failure and stem cell depletion in the bone marrow [129]. Sideroblastic anemia may be due to defective Atm1 transport of the (Fe-S)_int intermediate [74]. Perhaps downstream delivery of the Atm1 substrate into the cytoplasm of blood cell progenitors will be able to correct the phenotypic defect and restore bone marrow function.

Acknowledgements

We thank our laboratory colleagues Jayashree Pain and Brindha Jaya Chandran for their help and collaborations in various phases of this work. The laboratory is supported by National Institutes of Health Grants R01 GM107542 and R01 GM107542-06S1.

References

[1].Lill R, Freibert SA, Mechanisms of Mitochondrial Iron-Sulfur Protein Biogenesis, Annu Rev Biochem 89 (2020) 471–499. [DOI] [PubMed] [Google Scholar]
[2].Braymer JJ, Freibert SA, Rakwalska-Bange M, Lill R, Mechanistic concepts of iron-sulfur protein biogenesis in Biology, Biochim Biophys Acta Mol Cell Res 1868(1) (2021) 118863. [DOI] [PubMed] [Google Scholar]
[3].Andreini C, Putignano V, Rosato A, Banci L, The human iron-proteome, Metallomics 10(9) (2018) 1223–1231. [DOI] [PubMed] [Google Scholar]
[4].Makins C, Ghosh S, Roman-Melendez GD, Malec PA, Kennedy RT, Marsh EN, Does Viperin Function as a Radical S-Adenosyl-l-methionine-dependent Enzyme in Regulating Farnesylpyrophosphate Synthase Expression and Activity?, J Biol Chem 291(52) (2016) 26806–26815. [DOI] [PMC free article] [PubMed] [Google Scholar]
[5].Petronek MS, Allen BG, Maintenance of genome integrity by the late-acting cytoplasmic iron-sulfur assembly (CIA) complex, Front Genet 14 (2023) 1152398. [DOI] [PMC free article] [PubMed] [Google Scholar]
[6].Itoh Y, Khawaja A, Laptev I, Cipullo M, Atanassov I, Sergiev P, Rorbach J, Amunts A, Mechanism of mitoribosomal small subunit biogenesis and preinitiation, Nature 606(7914) (2022) 603–608. [DOI] [PMC free article] [PubMed] [Google Scholar]
[7].Zhong H, Janer A, Khalimonchuk O, Antonicka H, Shoubridge EA, Barrientos A, BOLA3 and NFU1 link mitoribosome iron-sulfur cluster assembly to multiple mitochondrial dysfunctions syndrome, Nucleic Acids Res (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
[8].Ast T, Itoh Y, Sadre S, McCoy JG, Namkoong G, Wengrod JC, Chicherin I, Joshi PR, Kamenski P, Suess DLM, Amunts A, Mootha VK, METTL17 is an Fe-S cluster checkpoint for mitochondrial translation, Mol Cell 84(2) (2024) 359–374 e8. [DOI] [PMC free article] [PubMed] [Google Scholar]
[9].Bak DW, Weerapana E, Monitoring Fe-S cluster occupancy across the E. coli proteome using chemoproteomics, Nat Chem Biol 19(3) (2023) 356–366. [DOI] [PMC free article] [PubMed] [Google Scholar]
[10].Maio N, Lafont BAP, Sil D, Li Y, Bollinger JM Jr., Krebs C, Pierson TC, Linehan WM, Rouault TA, Fe-S cofactors in the SARS-CoV-2 RNA-dependent RNA polymerase are potential antiviral targets, Science 373(6551) (2021) 236–241. [DOI] [PMC free article] [PubMed] [Google Scholar]
[11].Kennedy MC, Emptage MH, Dreyer JL, Beinert H, The role of iron in the activation-inactivation of aconitase, J Biol Chem 258(18) (1983) 11098–105. [PubMed] [Google Scholar]
[12].Zheng L, Dean DR, Catalytic formation of a nitrogenase iron-sulfur cluster, J Biol Chem 269(29) (1994) 18723–6. [PubMed] [Google Scholar]
[13].Zheng L, Cash VL, Flint DH, Dean DR, Assembly of iron-sulfur clusters. Identification of an iscSUA-hscBA-fdx gene cluster from Azotobacter vinelandii, J Biol Chem 273(21) (1998) 13264–72. [DOI] [PubMed] [Google Scholar]
[14].Kispal G, Csere P, Prohl C, Lill R, The mitochondrial proteins Atm1p and Nfs1p are essential for biogenesis of cytosolic Fe/S proteins, EMBO J 18(14) (1999) 3981–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
[15].Amutha B, Gordon DM, Gu Y, Lyver ER, Dancis A, Pain D, GTP is required for iron-sulfur cluster biogenesis in mitochondria, J Biol Chem 283(3) (2008) 1362–1371. [DOI] [PubMed] [Google Scholar]
[16].Pandey AK, Pain J, A. Dancis BJ, Pain D, Essential mitochondrial role in iron-sulfur cluster assembly of the cytoplasmic isopropylmalate isomerase Leu1 in Saccharomyces cerevisiae, Mitochondrion 69 (2023) 104–115. [DOI] [PMC free article] [PubMed] [Google Scholar]
[17].Pandey AK, Pain J, Dancis A, Pain D, Mitochondria export iron-sulfur and sulfur intermediates to the cytoplasm for iron-sulfur cluster assembly and tRNA thiolation in yeast, J Biol Chem 294(24) (2019) 9489–9502. [DOI] [PMC free article] [PubMed] [Google Scholar]
[18].Talib EA, Outten CE, Iron-sulfur cluster biogenesis, trafficking, and signaling: Roles for CGFS glutaredoxins and BolA proteins, Biochim Biophys Acta Mol Cell Res 1868(1) (2021) 118847. [DOI] [PMC free article] [PubMed] [Google Scholar]
[19].Boniecki MT, Freibert SA, Muhlenhoff U, Lill R, Cygler M, Structure and functional dynamics of the mitochondrial Fe/S cluster synthesis complex, Nat Commun 8(1) (2017) 1287. [DOI] [PMC free article] [PubMed] [Google Scholar]
[20].Cai K, Frederick RO, Dashti H, Markley JL, Architectural Features of Human Mitochondrial Cysteine Desulfurase Complexes from Crosslinking Mass Spectrometry and Small-Angle X-Ray Scattering, Structure 26(8) (2018) 1127–1136 e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
[21].Cory SA, Van Vranken JG, Brignole EJ, Patra S, Winge DR, Drennan CL, Rutter J, Barondeau DP, Structure of human Fe-S assembly subcomplex reveals unexpected cysteine desulfurase architecture and acyl-ACP-ISD11 interactions, Proc Natl Acad Sci U S A 114(27) (2017) E5325–E5334. [DOI] [PMC free article] [PubMed] [Google Scholar]
[22].Fox NG, Yu X, Feng X, Bailey HJ, Martelli A, Nabhan JF, Strain-Damerell C, Bulawa C, Yue WW, Han S, Structure of the human frataxin-bound iron-sulfur cluster assembly complex provides insight into its activation mechanism, Nat Commun 10(1) (2019) 2210. [DOI] [PMC free article] [PubMed] [Google Scholar]
[23].Gervason S, Larkem D, Mansour AB, Botzanowski T, Muller CS, Pecqueur L, Le Pavec G, Delaunay-Moisan A, Brun O, Agramunt J, Grandas A, Fontecave M, Schunemann V, Cianferani S, Sizun C, Toledano MB, D’Autreaux B, Physiologically relevant reconstitution of iron-sulfur cluster biosynthesis uncovers persulfide-processing functions of ferredoxin-2 and frataxin, Nat Commun 10(1) (2019) 3566. [DOI] [PMC free article] [PubMed] [Google Scholar]
[24].Adam AC, Bornhovd C, Prokisch H, Neupert W, Hell K, The Nfs1 interacting protein Isd11 has an essential role in Fe/S cluster biogenesis in mitochondria, EMBO J 25(1) (2006) 174–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
[25].Pandey A, Golla R, Yoon H, Dancis A, Pain D, Persulfide formation on mitochondrial cysteine desulfurase: enzyme activation by a eukaryote-specific interacting protein and Fe-S cluster synthesis, Biochem J 448(2) (2012) 171–87. [DOI] [PubMed] [Google Scholar]
[26].Majmudar JD, Feng X, Fox NG, Nabhan JF, Towle T, Ma T, Gooch R, Bulawa C, Yue WW, Martelli A, 4’-Phosphopantetheine and long acyl chain-dependent interactions are integral to human mitochondrial acyl carrier protein function, Medchemcomm 10(2) (2019) 209–220. [DOI] [PMC free article] [PubMed] [Google Scholar]
[27].Van Vranken JG, Jeong MY, Wei P, Chen YC, Gygi SP, Winge DR, Rutter J, The mitochondrial acyl carrier protein (ACP) coordinates mitochondrial fatty acid synthesis with iron sulfur cluster biogenesis, Elife 5 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
[28].Webert H, Freibert SA, Gallo A, Heidenreich T, Linne U, Amlacher S, Hurt E, Muhlenhoff U, Banci L, Lill R, Functional reconstitution of mitochondrial Fe/S cluster synthesis on Isu1 reveals the involvement of ferredoxin, Nat Commun 5 (2014) 5013. [DOI] [PubMed] [Google Scholar]
[29].Freibert SA, Goldberg AV, Hacker C, Molik S, Dean P, Williams TA, Nakjang S, Long S, Sendra K, Bill E, Heinz E, Hirt RP, Lucocq JM, Embley TM, Lill R, Evolutionary conservation and in vitro reconstitution of microsporidian iron-sulfur cluster biosynthesis, Nat Commun 8 (2017) 13932. [DOI] [PMC free article] [PubMed] [Google Scholar]
[30].Lill R, Dutkiewicz R, Freibert SA, Heidenreich T, Mascarenhas J, Netz DJ, Paul VD, Pierik AJ, Richter N, Stumpfig M, Srinivasan V, Stehling O, Muhlenhoff U, The role of mitochondria and the CIA machinery in the maturation of cytosolic and nuclear iron-sulfur proteins, Eur J Cell Biol 94(7–9) (2015) 280–91. [DOI] [PubMed] [Google Scholar]
[31].Delewski W, Paterkiewicz B, Manicki M, Schilke B, Tomiczek B, Ciesielski SJ, Nierzwicki L, Czub J, Dutkiewicz R, Craig EA, Marszalek J, Iron-Sulfur Cluster Biogenesis Chaperones: Evidence for Emergence of Mutational Robustness of a Highly Specific Protein-Protein Interaction, Mol Biol Evol 33(3) (2016) 643–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
[32].Majewska J, Ciesielski SJ, Schilke B, Kominek J, Blenska A, Delewski W, Song JY, Marszalek J, Craig EA, Dutkiewicz R, Binding of the chaperone Jac1 protein and cysteine desulfurase Nfs1 to the iron-sulfur cluster scaffold Isu protein is mutually exclusive, J Biol Chem 288(40) (2013) 29134–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
[33].Dutkiewicz R, Nowak M, Craig EA, Marszalek J, Fe-S Cluster Hsp70 Chaperones: The ATPase Cycle and Protein Interactions, Methods Enzymol 595 (2017) 161–184. [DOI] [PMC free article] [PubMed] [Google Scholar]
[34].Schmidt S, Strub A, Rottgers K, Zufall N, Voos W, The two mitochondrial heat shock proteins 70, Ssc1 and Ssq1, compete for the cochaperone Mge1, J Mol Biol 313(1) (2001) 13–26. [DOI] [PubMed] [Google Scholar]
[35].Uzarska MA, Dutkiewicz R, Freibert SA, Lill R, Muhlenhoff U, The mitochondrial Hsp70 chaperone Ssq1 facilitates Fe/S cluster transfer from Isu1 to Grx5 by complex formation, Mol Biol Cell 24(12) (2013) 1830–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
[36].Dutkiewicz R, Nowak M, Molecular chaperones involved in mitochondrial iron-sulfur protein biogenesis, J Biol Inorg Chem 23(4) (2018) 569–579. [DOI] [PMC free article] [PubMed] [Google Scholar]
[37].Camponeschi F, Ciofi-Baffoni S, Calderone V, Banci L, Molecular Basis of Rare Diseases Associated to the Maturation of Mitochondrial [4Fe-4S]-Containing Proteins, Biomolecules 12(7) (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
[38].Weiler BD, Bruck MC, Kothe I, Bill E, Lill R, Muhlenhoff U, Mitochondrial [4Fe-4S] protein assembly involves reductive [2Fe-2S] cluster fusion on ISCA1-ISCA2 by electron flow from ferredoxin FDX2, Proc Natl Acad Sci U S A 117(34) (2020) 20555–20565. [DOI] [PMC free article] [PubMed] [Google Scholar]
[39].Li P, Hendricks AL, Wang Y, Villones RLE, Lindkvist-Petersson K, Meloni G, Cowan JA, Wang K, Gourdon P, Structures of Atm1 provide insight into [2Fe-2S] cluster export from mitochondria, Nat Commun 13(1) (2022) 4339. [DOI] [PMC free article] [PubMed] [Google Scholar]
[40].Rutherford JC, Ojeda L, Balk J, Muhlenhoff U, Lill R, Winge DR, Activation of the iron regulon by the yeast Aft1/Aft2 transcription factors depends on mitochondrial but not cytosolic iron-sulfur protein biogenesis, J Biol Chem 280(11) (2005) 10135–40. [DOI] [PubMed] [Google Scholar]
[41].Dos Santos PC, Dean DR, Bioinorganic chemistry: electrons in Fe-S protein assembly, Nat Chem Biol 6(10) (2010) 700–1. [DOI] [PubMed] [Google Scholar]
[42].Netz DJ, Pierik AJ, Stumpfig M, Bill E, Sharma AK, Pallesen LJ, Walden WE, Lill R, A bridging [4Fe-4S] cluster and nucleotide binding are essential for function of the Cfd1-Nbp35 complex as a scaffold in iron-sulfur protein maturation, J Biol Chem 287(15) (2012) 12365–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
[43].Grossman JD, Gay KA, Camire EJ, Walden WE, Perlstein DL, Coupling Nucleotide Binding and Hydrolysis to Iron-Sulfur Cluster Acquisition and Transfer Revealed through Genetic Dissection of the Nbp35 ATPase Site, Biochemistry 58(15) (2019) 2017–2027. [DOI] [PubMed] [Google Scholar]
[44].Netz DJ, Stumpfig M, Dore C, Muhlenhoff U, Pierik AJ, Lill R, Tah18 transfers electrons to Dre2 in cytosolic iron-sulfur protein biogenesis, Nat Chem Biol 6(10) (2010) 758–65. [DOI] [PubMed] [Google Scholar]
[45].Zhang Y, Yang C, Dancis A, Nakamaru-Ogiso E, EPR studies of wild type and mutant Dre2 identify essential [2Fe−-2S] and [4Fe−-4S] clusters and their cysteine ligands, J Biochem 161(1) (2017) 67–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
[46].Urzica E, Pierik AJ, Muhlenhoff U, Lill R, Crucial role of conserved cysteine residues in the assembly of two iron-sulfur clusters on the CIA protein Nar1, Biochemistry 48(22) (2009) 4946–58. [DOI] [PubMed] [Google Scholar]
[47].Biederbick A, Stehling O, Rosser R, Niggemeyer B, Nakai Y, Elsasser HP, Lill R, Role of human mitochondrial Nfs1 in cytosolic iron-sulfur protein biogenesis and iron regulation, Mol Cell Biol 26(15) (2006) 5675–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
[48].Leighton J, Schatz G, An ABC transporter in the mitochondrial inner membrane is required for normal growth of yeast, EMBO J 14(1) (1995) 188–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
[49].Miao R, Kim H, Koppolu UM, Ellis EA, Scott RA, Lindahl PA, Biophysical characterization of the iron in mitochondria from Atm1p-depleted Saccharomyces cerevisiae, Biochemistry 48(40) (2009) 9556–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
[50].Kispal G, Csere P, Guiard B, Lill R, The ABC transporter Atm1p is required for mitochondrial iron homeostasis, FEBS Lett 418(3) (1997) 346–50. [DOI] [PubMed] [Google Scholar]
[51].Kohlhaw GB, Leucine biosynthesis in fungi: entering metabolism through the back door, Microbiol Mol Biol Rev 67(1) (2003) 1–15, table of contents. [DOI] [PMC free article] [PubMed] [Google Scholar]
[52].Prohl C, Pelzer W, Diekert K, Kmita H, Bedekovics T, Kispal G, Lill R, The yeast mitochondrial carrier Leu5p and its human homologue Graves’ disease protein are required for accumulation of coenzyme A in the matrix, Mol Cell Biol 21(4) (2001) 1089–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
[53].Marobbio CM, Giannuzzi G, Paradies E, Pierri CL, Palmieri F, alpha-Isopropylmalate, a leucine biosynthesis intermediate in yeast, is transported by the mitochondrial oxalacetate carrier, J Biol Chem 283(42) (2008) 28445–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
[54].Lill R, Kispal G, Maturation of cellular Fe-S proteins: an essential function of mitochondria, Trends Biochem Sci 25(8) (2000) 352–6. [DOI] [PubMed] [Google Scholar]
[55].Li J, Kogan M, Knight SA, Pain D, Dancis A, Yeast mitochondrial protein, Nfs1p, coordinately regulates iron-sulfur cluster proteins, cellular iron uptake, and iron distribution, J Biol Chem 274(46) (1999) 33025–34. [DOI] [PubMed] [Google Scholar]
[56].Muhlenhoff U, Balk J, Richhardt N, Kaiser JT, Sipos K, Kispal G, Lill R, Functional characterization of the eukaryotic cysteine desulfurase Nfs1p from Saccharomyces cerevisiae, J Biol Chem 279(35) (2004) 36906–15. [DOI] [PubMed] [Google Scholar]
[57].Naamati A, Regev-Rudzki N, Galperin S, Lill R, Pines O, Dual targeting of Nfs1 and discovery of its novel processing enzyme, Icp55, J Biol Chem 284(44) (2009) 30200–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
[58].Nakai Y, Nakai M, Hayashi H, Kagamiyama H, Nuclear localization of yeast Nfs1p is required for cell survival, J Biol Chem 276(11) (2001) 8314–20. [DOI] [PubMed] [Google Scholar]
[59].Land T, Rouault TA, Targeting of a human iron-sulfur cluster assembly enzyme, nifs, to different subcellular compartments is regulated through alternative AUG utilization, Mol Cell 2(6) (1998) 807–15. [DOI] [PubMed] [Google Scholar]
[60].Pandey A, Pain J, Dziuba N, Pandey AK, Dancis A, Lindahl PA, Pain D, Mitochondria Export Sulfur Species Required for Cytosolic tRNA Thiolation, Cell Chem Biol 25(6) (2018) 738–748 e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
[61].Pandey AK, Pain J, Dancis A, Pain D, Mitochondria export iron-sulfur and sulfur intermediates to the cytoplasm for iron-sulfur cluster assembly and tRNA thiolation in yeast, The Journal of biological chemistry 294(24) (2019) 9489–9502. [DOI] [PMC free article] [PubMed] [Google Scholar]
[62].Muhlenhoff U, Richhardt N, Gerber J, Lill R, Characterization of iron-sulfur protein assembly in isolated mitochondria. A requirement for ATP, NADH, and reduced iron, J Biol Chem 277(33) (2002) 29810–6. [DOI] [PubMed] [Google Scholar]
[63].Outten CE, Culotta VC, A novel NADH kinase is the mitochondrial source of NADPH in Saccharomyces cerevisiae, EMBO J 22(9) (2003) 2015–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
[64].Pain J, Balamurali MM, Dancis A, Pain D, Mitochondrial NADH kinase, Pos5p, is required for efficient iron-sulfur cluster biogenesis in Saccharomyces cerevisiae, J Biol Chem 285(50) (2010) 39409–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
[65].Stearman R, Yuan DS, Yamaguchi-Iwai Y, Klausner RD, Dancis A, A permease-oxidase complex involved in high-affinity iron uptake in yeast, Science 271(5255) (1996) 1552–7. [DOI] [PubMed] [Google Scholar]
[66].Kakhlon O, Cabantchik ZI, The labile iron pool: characterization, measurement, and participation in cellular processes(1), Free Radic Biol Med 33(8) (2002) 1037–46. [DOI] [PubMed] [Google Scholar]
[67].Zhang Y, Lyver ER, Knight SA, Pain D, Lesuisse E, Dancis A, Mrs3p, Mrs4p, and frataxin provide iron for Fe-S cluster synthesis in mitochondria, J Biol Chem 281(32) (2006) 22493–502. [DOI] [PubMed] [Google Scholar]
[68].Cook JD, Kondapalli KC, Rawat S, Childs WC, Murugesan Y, Dancis A, Stemmler TL, Molecular details of the yeast frataxin-Isu1 interaction during mitochondrial Fe-S cluster assembly, Biochemistry 49(40) (2010) 8756–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
[69].Duby G, Foury F, Ramazzotti A, Herrmann J, Lutz T, A non-essential function for yeast frataxin in iron-sulfur cluster assembly, Hum Mol Genet 11(21) (2002) 2635–43. [DOI] [PubMed] [Google Scholar]
[70].Srinivasan V, Pierik AJ, Lill R, Crystal structures of nucleotide-free and glutathione-bound mitochondrial ABC transporter Atm1, Science 343(6175) (2014) 1137–40. [DOI] [PubMed] [Google Scholar]
[71].Kumar C, Igbaria A, D’Autreaux B, Planson AG, Junot C, Godat E, Bachhawat AK, Delaunay-Moisan A, Toledano MB, Glutathione revisited: a vital function in iron metabolism and ancillary role in thiol-redox control, EMBO J 30(10) (2011) 2044–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
[72].Sipos K, Lange H, Fekete Z, Ullmann P, Lill R, Kispal G, Maturation of cytosolic iron-sulfur proteins requires glutathione, J Biol Chem 277(30) (2002) 26944–9. [DOI] [PubMed] [Google Scholar]
[73].Pearson SA, Wachnowsky C, Cowan JA, Defining the mechanism of the mitochondrial Atm1p [2Fe-2S] cluster exporter, Metallomics 12(6) (2020) 902–915. [DOI] [PMC free article] [PubMed] [Google Scholar]
[74].Bekri S, Kispal G, Lange H, Fitzsimons E, Tolmie J, Lill R, Bishop DF, Human ABC7 transporter: gene structure and mutation causing X-linked sideroblastic anemia with ataxia with disruption of cytosolic iron-sulfur protein maturation, Blood 96(9) (2000) 3256–64. [PubMed] [Google Scholar]
[75].Schaedler TA, Thornton JD, Kruse I, Schwarzlander M, Meyer AJ, van Veen HW, Balk J, A conserved mitochondrial ATP-binding cassette transporter exports glutathione polysulfide for cytosolic metal cofactor assembly, J Biol Chem 289(34) (2014) 23264–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
[76].Pearson SA, Cowan JA, Glutathione-coordinated metal complexes as substrates for cellular transporters, Metallomics 13(5) (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
[77].Nakai Y, Umeda N, Suzuki T, Nakai M, Hayashi H, Watanabe K, Kagamiyama H, Yeast Nfs1p is involved in thio-modification of both mitochondrial and cytoplasmic tRNAs, J Biol Chem 279(13) (2004) 12363–8. [DOI] [PubMed] [Google Scholar]
[78].Shigi N, Biosynthesis and Degradation of Sulfur Modifications in tRNAs, Int J Mol Sci 22(21) (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
[79].Braymer JJ, Winge DR, Sulfur from Within: Cytosolic tRNA Thiouridinylation, Cell Chem Biol 25(6) (2018) 645–647. [DOI] [PubMed] [Google Scholar]
[80].Linton KJ, Higgins CF, Structure and function of ABC transporters: the ATP switch provides flexible control, Pflugers Arch 453(5) (2007) 555–67. [DOI] [PubMed] [Google Scholar]
[81].Fan C, Rees DC, Glutathione binding to the plant AtAtm3 transporter and implications for the conformational coupling of ABC transporters, Elife 11 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
[82].Lushchak VI, Glutathione homeostasis and functions: potential targets for medical interventions, J Amino Acids 2012 (2012) 736837. [DOI] [PMC free article] [PubMed] [Google Scholar]
[83].Wang Y, Yen FS, Zhu XG, Timson RC, Weber R, Xing C, Liu Y, Allwein B, Luo H, Yeh HW, Heissel S, Unlu G, Gamazon ER, Kharas MG, Hite R, Birsoy K, SLC25A39 is necessary for mitochondrial glutathione import in mammalian cells, Nature 599(7883) (2021) 136–140. [DOI] [PMC free article] [PubMed] [Google Scholar]
[84].Ruprecht JJ, Kunji ERS, Structural Mechanism of Transport of Mitochondrial Carriers, Annu Rev Biochem 90 (2021) 535–558. [DOI] [PubMed] [Google Scholar]
[85].Ast TI, Y; Sadre S; McCoy JG; Namkoog G; Chicherin I; Joshi PR; Kamenski P; Suess DLM; Amunts A; Mootha VK, METTL17 is an Fe-S cluster checkpoint for mitochondrial translation, bioRxiv 10.110/2022.11.24.517765 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
[86].Camponeschi F, Prusty NR, Heider SAE, Ciofi-Baffoni S, Banci L, GLRX3 Acts as a [2Fe-2S] Cluster Chaperone in the Cytosolic Iron-Sulfur Assembly Machinery Transferring [2Fe-2S] Clusters to NUBP1, J Am Chem Soc 142(24) (2020) 10794–10805. [DOI] [PMC free article] [PubMed] [Google Scholar]
[87].Li H, Outten CE, The conserved CDC motif in the yeast iron regulator Aft2 mediates iron-sulfur cluster exchange and protein-protein interactions with Grx3 and Bol2, J Biol Inorg Chem 24(6) (2019) 809–815. [DOI] [PMC free article] [PubMed] [Google Scholar]
[88].Frey AG, Palenchar DJ, Wildemann JD, Philpott CC, A Glutaredoxin.BolA Complex Serves as an Iron-Sulfur Cluster Chaperone for the Cytosolic Cluster Assembly Machinery, J Biol Chem 291(43) (2016) 22344–22356. [DOI] [PMC free article] [PubMed] [Google Scholar]
[89].Ojeda L, Keller G, Muhlenhoff U, Rutherford JC, Lill R, Winge DR, Role of glutaredoxin-3 and glutaredoxin-4 in the iron regulation of the Aft1 transcriptional activator in Saccharomyces cerevisiae, J Biol Chem 281(26) (2006) 17661–9. [DOI] [PubMed] [Google Scholar]
[90].Puig S, Vergara SV, Thiele DJ, Cooperation of two mRNA-binding proteins drives metabolic adaptation to iron deficiency, Cell Metab 7(6) (2008) 555–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
[91].Zhang DL, Ghosh MC, Rouault TA, The physiological functions of iron regulatory proteins in iron homeostasis - an update, Front Pharmacol 5 (2014) 124. [DOI] [PMC free article] [PubMed] [Google Scholar]
[92].Wang H, Shi H, Rajan M, Canarie ER, Hong S, Simoneschi D, Pagano M, Bush MF, Stoll S, Leibold EA, Zheng N, FBXL5 Regulates IRP2 Stability in Iron Homeostasis via an Oxygen-Responsive [2Fe2S] Cluster, Mol Cell 78(1) (2020) 31–41 e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
[93].Rouault TA, Maio N, How Oxidation of a Unique Iron-Sulfur Cluster in FBXL5 Regulates IRP2 Levels and Promotes Regulation of Iron Metabolism Proteins, Mol Cell 78(1) (2020) 1–3. [DOI] [PMC free article] [PubMed] [Google Scholar]
[94].Allikmets R, Raskind WH, Hutchinson A, Schueck ND, Dean M, Koeller DM, Mutation of a putative mitochondrial iron transporter gene (ABC7) in X-linked sideroblastic anemia and ataxia (XLSA/A), Hum Mol Genet 8(5) (1999) 743–9. [DOI] [PubMed] [Google Scholar]
[95].Levi S, Ripamonti M, Dardi M, Cozzi A, Santambrogio P, Mitochondrial Ferritin: Its Role in Physiological and Pathological Conditions, Cells 10(8) (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
[96].Jhurry ND, Chakrabarti M, McCormick SP, Gohil VM, Lindahl PA, Mossbauer study and modeling of iron import and trafficking in human jurkat cells, Biochemistry 52(45) (2013) 7926–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
[97].Wingert RA, Galloway JL, Barut B, Foott H, Fraenkel P, Axe JL, Weber GJ, Dooley K, Davidson AJ, Schmid B, Paw BH, Shaw GC, Kingsley P, Palis J, Schubert H, Chen O, Kaplan J, Zon LI, C. Tubingen Screen, Deficiency of glutaredoxin 5 reveals Fe-S clusters are required for vertebrate haem synthesis, Nature 436(7053) (2005) 1035–39. [DOI] [PubMed] [Google Scholar]
[98].Ye H, Jeong SY, Ghosh MC, Kovtunovych G, Silvestri L, Ortillo D, Uchida N, Tisdale J, Camaschella C, Rouault TA, Glutaredoxin 5 deficiency causes sideroblastic anemia by specifically impairing heme biosynthesis and depleting cytosolic iron in human erythroblasts, J Clin Invest 120(5) (2010) 1749–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
[99].Liu G, Guo S, Anderson GJ, Camaschella C, Han B, Nie G, Heterozygous missense mutations in the GLRX5 gene cause sideroblastic anemia in a Chinese patient, Blood 124(17) (2014) 2750–1. [DOI] [PubMed] [Google Scholar]
[100].Liu G, Wang Y, Anderson GJ, Camaschella C, Chang Y, Nie G, Functional Analysis of GLRX5 Mutants Reveals Distinct Functionalities of GLRX5 Protein, J Cell Biochem 117(1) (2016) 207–17. [DOI] [PubMed] [Google Scholar]
[101].Baker PR 2nd, Friederich MW, Swanson MA, Shaikh T, Bhattacharya K, Scharer GH, Aicher J, Creadon-Swindell G, Geiger E, MacLean KN, Lee WT, Deshpande C, Freckmann ML, Shih LY, Wasserstein M, Rasmussen MB, Lund AM, Procopis P, Cameron JM, Robinson BH, Brown GK, Brown RM, Compton AG, Dieckmann CL, Collard R, Coughlin CR 2nd, Spector E, Wempe MF, Van Hove JL, Variant non ketotic hyperglycinemia is caused by mutations in LIAS, BOLA3 and the novel gene GLRX5, Brain 137(Pt 2) (2014) 366–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
[102].Dos Santos PC, Johnson DC, Ragle BE, Unciuleac MC, Dean DR, Controlled expression of nif and isc iron-sulfur protein maturation components reveals target specificity and limited functional replacement between the two systems, J Bacteriol 189(7) (2007) 2854–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
[103].Kleczewska M, Grabinska A, Jelen M, Stolarska M, Schilke B, Marszalek J, Craig EA, Dutkiewicz R, Biochemical Convergence of Mitochondrial Hsp70 System Specialized in Iron-Sulfur Cluster Biogenesis, Int J Mol Sci 21(9) (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
[104].Blahut M, Sanchez E, Fisher CE, Outten FW, Fe-S cluster biogenesis by the bacterial Suf pathway, Biochim Biophys Acta Mol Cell Res 1867(11) (2020) 118829. [DOI] [PMC free article] [PubMed] [Google Scholar]
[105].Tanaka N, Kanazawa M, Tonosaki K, Yokoyama N, Kuzuyama T, Takahashi Y, Novel features of the ISC machinery revealed by characterization of Escherichia coli mutants that survive without iron-sulfur clusters, Mol Microbiol 99(5) (2016) 835–48. [DOI] [PubMed] [Google Scholar]
[106].Tsaousis AD, On the Origin of Iron/Sulfur Cluster Biosynthesis in Eukaryotes, Front Microbiol 10 (2019) 2478. [DOI] [PMC free article] [PubMed] [Google Scholar]
[107].Karnkowska A, Treitli SC, Brzon O, Novak L, Vacek V, Soukal P, Barlow LD, Herman EK, Pipaliya SV, Panek T, Zihala D, Petrzelkova R, Butenko A, Eme L, Stairs CW, Roger AJ, Elias M, Dacks JB, Hampl V, The Oxymonad Genome Displays Canonical Eukaryotic Complexity in the Absence of a Mitochondrion, Mol Biol Evol 36(10) (2019) 2292–2312. [DOI] [PMC free article] [PubMed] [Google Scholar]
[108].Pena-Diaz P, Lukes J, Fe-S cluster assembly in the supergroup Excavata, J Biol Inorg Chem 23(4) (2018) 521–541. [DOI] [PMC free article] [PubMed] [Google Scholar]
[109].Smid O, Horakova E, Vilimova V, Hrdy I, Cammack R, Horvath A, Lukes J, Tachezy J, Knock-downs of iron-sulfur cluster assembly proteins IscS and IscU down-regulate the active mitochondrion of procyclic Trypanosoma brucei, J Biol Chem 281(39) (2006) 28679–86. [DOI] [PubMed] [Google Scholar]
[110].Lukes J, Basu S, Fe/S protein biogenesis in trypanosomes - A review, Biochim Biophys Acta 1853(6) (2015) 1481–92. [DOI] [PubMed] [Google Scholar]
[111].Maio N, Rouault TA, Outlining the Complex Pathway of Mammalian Fe-S Cluster Biogenesis, Trends Biochem Sci 45(5) (2020) 411–426. [DOI] [PMC free article] [PubMed] [Google Scholar]
[112].Glatt S, Zabel R, Vonkova I, Kumar A, Netz DJ, Pierik AJ, Rybin V, Lill R, Gavin AC, Balbach J, Breunig KD, Muller CW, Structure of the Kti11/Kti13 heterodimer and its double role in modifications of tRNA and eukaryotic elongation factor 2, Structure 23(1) (2015) 149–160. [DOI] [PubMed] [Google Scholar]
[113].Wiedemann N, Urzica E, Guiard B, Muller H, Lohaus C, Meyer HE, Ryan MT, Meisinger C, Muhlenhoff U, Lill R, Pfanner N, Essential role of Isd11 in mitochondrial iron-sulfur cluster synthesis on Isu scaffold proteins, EMBO J 25(1) (2006) 184–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
[114].Gerber J, Neumann K, Prohl C, Muhlenhoff U, Lill R, The yeast scaffold proteins Isu1p and Isu2p are required inside mitochondria for maturation of cytosolic Fe/S proteins, Mol Cell Biol 24(11) (2004) 4848–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
[115].Yoon H, Knight SA, Pandey A, Pain J, Turkarslan S, Pain D, Dancis A, Turning Saccharomyces cerevisiae into a Frataxin-Independent Organism, PLoS Genet 11(5) (2015) e1005135. [DOI] [PMC free article] [PubMed] [Google Scholar]
[116].Lange H, Kaut A, Kispal G, Lill R, A mitochondrial ferredoxin is essential for biogenesis of cellular iron-sulfur proteins, Proc Natl Acad Sci U S A 97(3) (2000) 1050–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
[117].Muhlenhoff U, Richhardt N, Ristow M, Kispal G, Lill R, The yeast frataxin homolog Yfh1p plays a specific role in the maturation of cellular Fe/S proteins, Hum Mol Genet 11(17) (2002) 2025–36. [DOI] [PubMed] [Google Scholar]
[118].Knight SA, Sepuri NB, Pain D, Dancis A, Mt-Hsp70 homolog, Ssc2p, required for maturation of yeast frataxin and mitochondrial iron homeostasis, J Biol Chem 273(29) (1998) 18389–93. [DOI] [PubMed] [Google Scholar]
[119].Kim R, Saxena S, Gordon DM, Pain D, Dancis A, J-domain protein, Jac1p, of yeast mitochondria required for iron homeostasis and activity of Fe-S cluster proteins, J Biol Chem 276(20) (2001) 17524–32. [DOI] [PubMed] [Google Scholar]
[120].Rodriguez-Manzaneque MT, Tamarit J, Belli G, Ros J, Herrero E, Grx5 is a mitochondrial glutaredoxin required for the activity of iron/sulfur enzymes, Mol Biol Cell 13(4) (2002) 1109–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
[121].Campbell CJ, Pall AE, Naik AR, Thompson LN, Stemmler TL, Molecular Details of the Frataxin-Scaffold Interaction during Mitochondrial Fe-S Cluster Assembly, Int J Mol Sci 22(11) (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
[122].Porras CA, Rouault TA, Iron Homeostasis in the CNS: An Overview of the Pathological Consequences of Iron Metabolism Disruption, Int J Mol Sci 23(9) (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
[123].Fan X, Barshop WD, Vashisht AA, Pandey V, Leal S, Rayatpisheh S, Jami-Alahmadi Y, Sha J, Wohlschlegel JA, Iron-regulated assembly of the cytosolic iron-sulfur cluster biogenesis machinery, J Biol Chem 298(7) (2022) 102094. [DOI] [PMC free article] [PubMed] [Google Scholar]
[124].Kassube SA, Thoma NH, Structural insights into Fe-S protein biogenesis by the CIA targeting complex, Nat Struct Mol Biol 27(8) (2020) 735–742. [DOI] [PubMed] [Google Scholar]
[125].Garland SA, Hoff K, Vickery LE, Culotta VC, Saccharomyces cerevisiae ISU1 and ISU2: members of a well-conserved gene family for iron-sulfur cluster assembly, J Mol Biol 294(4) (1999) 897–907. [DOI] [PubMed] [Google Scholar]
[126].Chan SY, Zhang YY, Hemann C, Mahoney CE, Zweier JL, Loscalzo J, MicroRNA-210 controls mitochondrial metabolism during hypoxia by repressing the iron-sulfur cluster assembly proteins ISCU1/2, Cell Metab 10(4) (2009) 273–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
[127].Hoff KG, Culler SJ, Nguyen PQ, McGuire RM, Silberg JJ, Smolke CD, In vivo fluorescent detection of Fe-S clusters coordinated by human GRX2, Chem Biol 16(12) (2009) 1299–308. [DOI] [PubMed] [Google Scholar]
[128].Monfort B, Want K, Gervason S, D’Autreaux B, Recent Advances in the Elucidation of Frataxin Biochemical Function Open Novel Perspectives for the Treatment of Friedreich’s Ataxia, Front Neurosci 16 (2022) 838335. [DOI] [PMC free article] [PubMed] [Google Scholar]
[129].Ochi T, Fujiwara T, Ono K, Suzuki C, Nikaido M, Inoue D, Kato H, Onodera K, Ichikawa S, Fukuhara N, Onishi Y, Yokoyama H, Nakamura Y, Harigae H, Exploring the mechanistic link between SF3B1 mutation and ring sideroblast formation in myelodysplastic syndrome, Sci Rep 12(1) (2022) 14562. [DOI] [PMC free article] [PubMed] [Google Scholar]
[130].Iwema T, Picciocchi A, Traore DA, Ferrer JL, Chauvat F, Jacquamet L, Structural basis for delivery of the intact [Fe2S2] cluster by monothiol glutaredoxin, Biochemistry 48(26) (2009) 6041–3. [DOI] [PubMed] [Google Scholar]

[R1] [1].Lill R, Freibert SA, Mechanisms of Mitochondrial Iron-Sulfur Protein Biogenesis, Annu Rev Biochem 89 (2020) 471–499. [DOI] [PubMed] [Google Scholar]

[R2] [2].Braymer JJ, Freibert SA, Rakwalska-Bange M, Lill R, Mechanistic concepts of iron-sulfur protein biogenesis in Biology, Biochim Biophys Acta Mol Cell Res 1868(1) (2021) 118863. [DOI] [PubMed] [Google Scholar]

[R3] [3].Andreini C, Putignano V, Rosato A, Banci L, The human iron-proteome, Metallomics 10(9) (2018) 1223–1231. [DOI] [PubMed] [Google Scholar]

[R4] [4].Makins C, Ghosh S, Roman-Melendez GD, Malec PA, Kennedy RT, Marsh EN, Does Viperin Function as a Radical S-Adenosyl-l-methionine-dependent Enzyme in Regulating Farnesylpyrophosphate Synthase Expression and Activity?, J Biol Chem 291(52) (2016) 26806–26815. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R5] [5].Petronek MS, Allen BG, Maintenance of genome integrity by the late-acting cytoplasmic iron-sulfur assembly (CIA) complex, Front Genet 14 (2023) 1152398. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] [6].Itoh Y, Khawaja A, Laptev I, Cipullo M, Atanassov I, Sergiev P, Rorbach J, Amunts A, Mechanism of mitoribosomal small subunit biogenesis and preinitiation, Nature 606(7914) (2022) 603–608. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R7] [7].Zhong H, Janer A, Khalimonchuk O, Antonicka H, Shoubridge EA, Barrientos A, BOLA3 and NFU1 link mitoribosome iron-sulfur cluster assembly to multiple mitochondrial dysfunctions syndrome, Nucleic Acids Res (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] [8].Ast T, Itoh Y, Sadre S, McCoy JG, Namkoong G, Wengrod JC, Chicherin I, Joshi PR, Kamenski P, Suess DLM, Amunts A, Mootha VK, METTL17 is an Fe-S cluster checkpoint for mitochondrial translation, Mol Cell 84(2) (2024) 359–374 e8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R9] [9].Bak DW, Weerapana E, Monitoring Fe-S cluster occupancy across the E. coli proteome using chemoproteomics, Nat Chem Biol 19(3) (2023) 356–366. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] [10].Maio N, Lafont BAP, Sil D, Li Y, Bollinger JM Jr., Krebs C, Pierson TC, Linehan WM, Rouault TA, Fe-S cofactors in the SARS-CoV-2 RNA-dependent RNA polymerase are potential antiviral targets, Science 373(6551) (2021) 236–241. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R11] [11].Kennedy MC, Emptage MH, Dreyer JL, Beinert H, The role of iron in the activation-inactivation of aconitase, J Biol Chem 258(18) (1983) 11098–105. [PubMed] [Google Scholar]

[R12] [12].Zheng L, Dean DR, Catalytic formation of a nitrogenase iron-sulfur cluster, J Biol Chem 269(29) (1994) 18723–6. [PubMed] [Google Scholar]

[R13] [13].Zheng L, Cash VL, Flint DH, Dean DR, Assembly of iron-sulfur clusters. Identification of an iscSUA-hscBA-fdx gene cluster from Azotobacter vinelandii, J Biol Chem 273(21) (1998) 13264–72. [DOI] [PubMed] [Google Scholar]

[R14] [14].Kispal G, Csere P, Prohl C, Lill R, The mitochondrial proteins Atm1p and Nfs1p are essential for biogenesis of cytosolic Fe/S proteins, EMBO J 18(14) (1999) 3981–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] [15].Amutha B, Gordon DM, Gu Y, Lyver ER, Dancis A, Pain D, GTP is required for iron-sulfur cluster biogenesis in mitochondria, J Biol Chem 283(3) (2008) 1362–1371. [DOI] [PubMed] [Google Scholar]

[R16] [16].Pandey AK, Pain J, A. Dancis BJ, Pain D, Essential mitochondrial role in iron-sulfur cluster assembly of the cytoplasmic isopropylmalate isomerase Leu1 in Saccharomyces cerevisiae, Mitochondrion 69 (2023) 104–115. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] [17].Pandey AK, Pain J, Dancis A, Pain D, Mitochondria export iron-sulfur and sulfur intermediates to the cytoplasm for iron-sulfur cluster assembly and tRNA thiolation in yeast, J Biol Chem 294(24) (2019) 9489–9502. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] [18].Talib EA, Outten CE, Iron-sulfur cluster biogenesis, trafficking, and signaling: Roles for CGFS glutaredoxins and BolA proteins, Biochim Biophys Acta Mol Cell Res 1868(1) (2021) 118847. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] [19].Boniecki MT, Freibert SA, Muhlenhoff U, Lill R, Cygler M, Structure and functional dynamics of the mitochondrial Fe/S cluster synthesis complex, Nat Commun 8(1) (2017) 1287. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] [20].Cai K, Frederick RO, Dashti H, Markley JL, Architectural Features of Human Mitochondrial Cysteine Desulfurase Complexes from Crosslinking Mass Spectrometry and Small-Angle X-Ray Scattering, Structure 26(8) (2018) 1127–1136 e4. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] [21].Cory SA, Van Vranken JG, Brignole EJ, Patra S, Winge DR, Drennan CL, Rutter J, Barondeau DP, Structure of human Fe-S assembly subcomplex reveals unexpected cysteine desulfurase architecture and acyl-ACP-ISD11 interactions, Proc Natl Acad Sci U S A 114(27) (2017) E5325–E5334. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R22] [22].Fox NG, Yu X, Feng X, Bailey HJ, Martelli A, Nabhan JF, Strain-Damerell C, Bulawa C, Yue WW, Han S, Structure of the human frataxin-bound iron-sulfur cluster assembly complex provides insight into its activation mechanism, Nat Commun 10(1) (2019) 2210. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R23] [23].Gervason S, Larkem D, Mansour AB, Botzanowski T, Muller CS, Pecqueur L, Le Pavec G, Delaunay-Moisan A, Brun O, Agramunt J, Grandas A, Fontecave M, Schunemann V, Cianferani S, Sizun C, Toledano MB, D’Autreaux B, Physiologically relevant reconstitution of iron-sulfur cluster biosynthesis uncovers persulfide-processing functions of ferredoxin-2 and frataxin, Nat Commun 10(1) (2019) 3566. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] [24].Adam AC, Bornhovd C, Prokisch H, Neupert W, Hell K, The Nfs1 interacting protein Isd11 has an essential role in Fe/S cluster biogenesis in mitochondria, EMBO J 25(1) (2006) 174–83. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R25] [25].Pandey A, Golla R, Yoon H, Dancis A, Pain D, Persulfide formation on mitochondrial cysteine desulfurase: enzyme activation by a eukaryote-specific interacting protein and Fe-S cluster synthesis, Biochem J 448(2) (2012) 171–87. [DOI] [PubMed] [Google Scholar]

[R26] [26].Majmudar JD, Feng X, Fox NG, Nabhan JF, Towle T, Ma T, Gooch R, Bulawa C, Yue WW, Martelli A, 4’-Phosphopantetheine and long acyl chain-dependent interactions are integral to human mitochondrial acyl carrier protein function, Medchemcomm 10(2) (2019) 209–220. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R27] [27].Van Vranken JG, Jeong MY, Wei P, Chen YC, Gygi SP, Winge DR, Rutter J, The mitochondrial acyl carrier protein (ACP) coordinates mitochondrial fatty acid synthesis with iron sulfur cluster biogenesis, Elife 5 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R28] [28].Webert H, Freibert SA, Gallo A, Heidenreich T, Linne U, Amlacher S, Hurt E, Muhlenhoff U, Banci L, Lill R, Functional reconstitution of mitochondrial Fe/S cluster synthesis on Isu1 reveals the involvement of ferredoxin, Nat Commun 5 (2014) 5013. [DOI] [PubMed] [Google Scholar]

[R29] [29].Freibert SA, Goldberg AV, Hacker C, Molik S, Dean P, Williams TA, Nakjang S, Long S, Sendra K, Bill E, Heinz E, Hirt RP, Lucocq JM, Embley TM, Lill R, Evolutionary conservation and in vitro reconstitution of microsporidian iron-sulfur cluster biosynthesis, Nat Commun 8 (2017) 13932. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] [30].Lill R, Dutkiewicz R, Freibert SA, Heidenreich T, Mascarenhas J, Netz DJ, Paul VD, Pierik AJ, Richter N, Stumpfig M, Srinivasan V, Stehling O, Muhlenhoff U, The role of mitochondria and the CIA machinery in the maturation of cytosolic and nuclear iron-sulfur proteins, Eur J Cell Biol 94(7–9) (2015) 280–91. [DOI] [PubMed] [Google Scholar]

[R31] [31].Delewski W, Paterkiewicz B, Manicki M, Schilke B, Tomiczek B, Ciesielski SJ, Nierzwicki L, Czub J, Dutkiewicz R, Craig EA, Marszalek J, Iron-Sulfur Cluster Biogenesis Chaperones: Evidence for Emergence of Mutational Robustness of a Highly Specific Protein-Protein Interaction, Mol Biol Evol 33(3) (2016) 643–56. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] [32].Majewska J, Ciesielski SJ, Schilke B, Kominek J, Blenska A, Delewski W, Song JY, Marszalek J, Craig EA, Dutkiewicz R, Binding of the chaperone Jac1 protein and cysteine desulfurase Nfs1 to the iron-sulfur cluster scaffold Isu protein is mutually exclusive, J Biol Chem 288(40) (2013) 29134–42. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R33] [33].Dutkiewicz R, Nowak M, Craig EA, Marszalek J, Fe-S Cluster Hsp70 Chaperones: The ATPase Cycle and Protein Interactions, Methods Enzymol 595 (2017) 161–184. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] [34].Schmidt S, Strub A, Rottgers K, Zufall N, Voos W, The two mitochondrial heat shock proteins 70, Ssc1 and Ssq1, compete for the cochaperone Mge1, J Mol Biol 313(1) (2001) 13–26. [DOI] [PubMed] [Google Scholar]

[R35] [35].Uzarska MA, Dutkiewicz R, Freibert SA, Lill R, Muhlenhoff U, The mitochondrial Hsp70 chaperone Ssq1 facilitates Fe/S cluster transfer from Isu1 to Grx5 by complex formation, Mol Biol Cell 24(12) (2013) 1830–41. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R36] [36].Dutkiewicz R, Nowak M, Molecular chaperones involved in mitochondrial iron-sulfur protein biogenesis, J Biol Inorg Chem 23(4) (2018) 569–579. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R37] [37].Camponeschi F, Ciofi-Baffoni S, Calderone V, Banci L, Molecular Basis of Rare Diseases Associated to the Maturation of Mitochondrial [4Fe-4S]-Containing Proteins, Biomolecules 12(7) (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R38] [38].Weiler BD, Bruck MC, Kothe I, Bill E, Lill R, Muhlenhoff U, Mitochondrial [4Fe-4S] protein assembly involves reductive [2Fe-2S] cluster fusion on ISCA1-ISCA2 by electron flow from ferredoxin FDX2, Proc Natl Acad Sci U S A 117(34) (2020) 20555–20565. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R39] [39].Li P, Hendricks AL, Wang Y, Villones RLE, Lindkvist-Petersson K, Meloni G, Cowan JA, Wang K, Gourdon P, Structures of Atm1 provide insight into [2Fe-2S] cluster export from mitochondria, Nat Commun 13(1) (2022) 4339. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R40] [40].Rutherford JC, Ojeda L, Balk J, Muhlenhoff U, Lill R, Winge DR, Activation of the iron regulon by the yeast Aft1/Aft2 transcription factors depends on mitochondrial but not cytosolic iron-sulfur protein biogenesis, J Biol Chem 280(11) (2005) 10135–40. [DOI] [PubMed] [Google Scholar]

[R41] [41].Dos Santos PC, Dean DR, Bioinorganic chemistry: electrons in Fe-S protein assembly, Nat Chem Biol 6(10) (2010) 700–1. [DOI] [PubMed] [Google Scholar]

[R42] [42].Netz DJ, Pierik AJ, Stumpfig M, Bill E, Sharma AK, Pallesen LJ, Walden WE, Lill R, A bridging [4Fe-4S] cluster and nucleotide binding are essential for function of the Cfd1-Nbp35 complex as a scaffold in iron-sulfur protein maturation, J Biol Chem 287(15) (2012) 12365–78. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R43] [43].Grossman JD, Gay KA, Camire EJ, Walden WE, Perlstein DL, Coupling Nucleotide Binding and Hydrolysis to Iron-Sulfur Cluster Acquisition and Transfer Revealed through Genetic Dissection of the Nbp35 ATPase Site, Biochemistry 58(15) (2019) 2017–2027. [DOI] [PubMed] [Google Scholar]

[R44] [44].Netz DJ, Stumpfig M, Dore C, Muhlenhoff U, Pierik AJ, Lill R, Tah18 transfers electrons to Dre2 in cytosolic iron-sulfur protein biogenesis, Nat Chem Biol 6(10) (2010) 758–65. [DOI] [PubMed] [Google Scholar]

[R45] [45].Zhang Y, Yang C, Dancis A, Nakamaru-Ogiso E, EPR studies of wild type and mutant Dre2 identify essential [2Fe−-2S] and [4Fe−-4S] clusters and their cysteine ligands, J Biochem 161(1) (2017) 67–78. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R46] [46].Urzica E, Pierik AJ, Muhlenhoff U, Lill R, Crucial role of conserved cysteine residues in the assembly of two iron-sulfur clusters on the CIA protein Nar1, Biochemistry 48(22) (2009) 4946–58. [DOI] [PubMed] [Google Scholar]

[R47] [47].Biederbick A, Stehling O, Rosser R, Niggemeyer B, Nakai Y, Elsasser HP, Lill R, Role of human mitochondrial Nfs1 in cytosolic iron-sulfur protein biogenesis and iron regulation, Mol Cell Biol 26(15) (2006) 5675–87. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R48] [48].Leighton J, Schatz G, An ABC transporter in the mitochondrial inner membrane is required for normal growth of yeast, EMBO J 14(1) (1995) 188–95. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R49] [49].Miao R, Kim H, Koppolu UM, Ellis EA, Scott RA, Lindahl PA, Biophysical characterization of the iron in mitochondria from Atm1p-depleted Saccharomyces cerevisiae, Biochemistry 48(40) (2009) 9556–68. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R50] [50].Kispal G, Csere P, Guiard B, Lill R, The ABC transporter Atm1p is required for mitochondrial iron homeostasis, FEBS Lett 418(3) (1997) 346–50. [DOI] [PubMed] [Google Scholar]

[R51] [51].Kohlhaw GB, Leucine biosynthesis in fungi: entering metabolism through the back door, Microbiol Mol Biol Rev 67(1) (2003) 1–15, table of contents. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R52] [52].Prohl C, Pelzer W, Diekert K, Kmita H, Bedekovics T, Kispal G, Lill R, The yeast mitochondrial carrier Leu5p and its human homologue Graves’ disease protein are required for accumulation of coenzyme A in the matrix, Mol Cell Biol 21(4) (2001) 1089–97. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R53] [53].Marobbio CM, Giannuzzi G, Paradies E, Pierri CL, Palmieri F, alpha-Isopropylmalate, a leucine biosynthesis intermediate in yeast, is transported by the mitochondrial oxalacetate carrier, J Biol Chem 283(42) (2008) 28445–53. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R54] [54].Lill R, Kispal G, Maturation of cellular Fe-S proteins: an essential function of mitochondria, Trends Biochem Sci 25(8) (2000) 352–6. [DOI] [PubMed] [Google Scholar]

[R55] [55].Li J, Kogan M, Knight SA, Pain D, Dancis A, Yeast mitochondrial protein, Nfs1p, coordinately regulates iron-sulfur cluster proteins, cellular iron uptake, and iron distribution, J Biol Chem 274(46) (1999) 33025–34. [DOI] [PubMed] [Google Scholar]

[R56] [56].Muhlenhoff U, Balk J, Richhardt N, Kaiser JT, Sipos K, Kispal G, Lill R, Functional characterization of the eukaryotic cysteine desulfurase Nfs1p from Saccharomyces cerevisiae, J Biol Chem 279(35) (2004) 36906–15. [DOI] [PubMed] [Google Scholar]

[R57] [57].Naamati A, Regev-Rudzki N, Galperin S, Lill R, Pines O, Dual targeting of Nfs1 and discovery of its novel processing enzyme, Icp55, J Biol Chem 284(44) (2009) 30200–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R58] [58].Nakai Y, Nakai M, Hayashi H, Kagamiyama H, Nuclear localization of yeast Nfs1p is required for cell survival, J Biol Chem 276(11) (2001) 8314–20. [DOI] [PubMed] [Google Scholar]

[R59] [59].Land T, Rouault TA, Targeting of a human iron-sulfur cluster assembly enzyme, nifs, to different subcellular compartments is regulated through alternative AUG utilization, Mol Cell 2(6) (1998) 807–15. [DOI] [PubMed] [Google Scholar]

[R60] [60].Pandey A, Pain J, Dziuba N, Pandey AK, Dancis A, Lindahl PA, Pain D, Mitochondria Export Sulfur Species Required for Cytosolic tRNA Thiolation, Cell Chem Biol 25(6) (2018) 738–748 e3. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R61] [61].Pandey AK, Pain J, Dancis A, Pain D, Mitochondria export iron-sulfur and sulfur intermediates to the cytoplasm for iron-sulfur cluster assembly and tRNA thiolation in yeast, The Journal of biological chemistry 294(24) (2019) 9489–9502. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R62] [62].Muhlenhoff U, Richhardt N, Gerber J, Lill R, Characterization of iron-sulfur protein assembly in isolated mitochondria. A requirement for ATP, NADH, and reduced iron, J Biol Chem 277(33) (2002) 29810–6. [DOI] [PubMed] [Google Scholar]

[R63] [63].Outten CE, Culotta VC, A novel NADH kinase is the mitochondrial source of NADPH in Saccharomyces cerevisiae, EMBO J 22(9) (2003) 2015–24. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R64] [64].Pain J, Balamurali MM, Dancis A, Pain D, Mitochondrial NADH kinase, Pos5p, is required for efficient iron-sulfur cluster biogenesis in Saccharomyces cerevisiae, J Biol Chem 285(50) (2010) 39409–24. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R65] [65].Stearman R, Yuan DS, Yamaguchi-Iwai Y, Klausner RD, Dancis A, A permease-oxidase complex involved in high-affinity iron uptake in yeast, Science 271(5255) (1996) 1552–7. [DOI] [PubMed] [Google Scholar]

[R66] [66].Kakhlon O, Cabantchik ZI, The labile iron pool: characterization, measurement, and participation in cellular processes(1), Free Radic Biol Med 33(8) (2002) 1037–46. [DOI] [PubMed] [Google Scholar]

[R67] [67].Zhang Y, Lyver ER, Knight SA, Pain D, Lesuisse E, Dancis A, Mrs3p, Mrs4p, and frataxin provide iron for Fe-S cluster synthesis in mitochondria, J Biol Chem 281(32) (2006) 22493–502. [DOI] [PubMed] [Google Scholar]

[R68] [68].Cook JD, Kondapalli KC, Rawat S, Childs WC, Murugesan Y, Dancis A, Stemmler TL, Molecular details of the yeast frataxin-Isu1 interaction during mitochondrial Fe-S cluster assembly, Biochemistry 49(40) (2010) 8756–65. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R69] [69].Duby G, Foury F, Ramazzotti A, Herrmann J, Lutz T, A non-essential function for yeast frataxin in iron-sulfur cluster assembly, Hum Mol Genet 11(21) (2002) 2635–43. [DOI] [PubMed] [Google Scholar]

[R70] [70].Srinivasan V, Pierik AJ, Lill R, Crystal structures of nucleotide-free and glutathione-bound mitochondrial ABC transporter Atm1, Science 343(6175) (2014) 1137–40. [DOI] [PubMed] [Google Scholar]

[R71] [71].Kumar C, Igbaria A, D’Autreaux B, Planson AG, Junot C, Godat E, Bachhawat AK, Delaunay-Moisan A, Toledano MB, Glutathione revisited: a vital function in iron metabolism and ancillary role in thiol-redox control, EMBO J 30(10) (2011) 2044–56. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R72] [72].Sipos K, Lange H, Fekete Z, Ullmann P, Lill R, Kispal G, Maturation of cytosolic iron-sulfur proteins requires glutathione, J Biol Chem 277(30) (2002) 26944–9. [DOI] [PubMed] [Google Scholar]

[R73] [73].Pearson SA, Wachnowsky C, Cowan JA, Defining the mechanism of the mitochondrial Atm1p [2Fe-2S] cluster exporter, Metallomics 12(6) (2020) 902–915. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R74] [74].Bekri S, Kispal G, Lange H, Fitzsimons E, Tolmie J, Lill R, Bishop DF, Human ABC7 transporter: gene structure and mutation causing X-linked sideroblastic anemia with ataxia with disruption of cytosolic iron-sulfur protein maturation, Blood 96(9) (2000) 3256–64. [PubMed] [Google Scholar]

[R75] [75].Schaedler TA, Thornton JD, Kruse I, Schwarzlander M, Meyer AJ, van Veen HW, Balk J, A conserved mitochondrial ATP-binding cassette transporter exports glutathione polysulfide for cytosolic metal cofactor assembly, J Biol Chem 289(34) (2014) 23264–74. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R76] [76].Pearson SA, Cowan JA, Glutathione-coordinated metal complexes as substrates for cellular transporters, Metallomics 13(5) (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R77] [77].Nakai Y, Umeda N, Suzuki T, Nakai M, Hayashi H, Watanabe K, Kagamiyama H, Yeast Nfs1p is involved in thio-modification of both mitochondrial and cytoplasmic tRNAs, J Biol Chem 279(13) (2004) 12363–8. [DOI] [PubMed] [Google Scholar]

[R78] [78].Shigi N, Biosynthesis and Degradation of Sulfur Modifications in tRNAs, Int J Mol Sci 22(21) (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R79] [79].Braymer JJ, Winge DR, Sulfur from Within: Cytosolic tRNA Thiouridinylation, Cell Chem Biol 25(6) (2018) 645–647. [DOI] [PubMed] [Google Scholar]

[R80] [80].Linton KJ, Higgins CF, Structure and function of ABC transporters: the ATP switch provides flexible control, Pflugers Arch 453(5) (2007) 555–67. [DOI] [PubMed] [Google Scholar]

[R81] [81].Fan C, Rees DC, Glutathione binding to the plant AtAtm3 transporter and implications for the conformational coupling of ABC transporters, Elife 11 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R82] [82].Lushchak VI, Glutathione homeostasis and functions: potential targets for medical interventions, J Amino Acids 2012 (2012) 736837. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R83] [83].Wang Y, Yen FS, Zhu XG, Timson RC, Weber R, Xing C, Liu Y, Allwein B, Luo H, Yeh HW, Heissel S, Unlu G, Gamazon ER, Kharas MG, Hite R, Birsoy K, SLC25A39 is necessary for mitochondrial glutathione import in mammalian cells, Nature 599(7883) (2021) 136–140. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R84] [84].Ruprecht JJ, Kunji ERS, Structural Mechanism of Transport of Mitochondrial Carriers, Annu Rev Biochem 90 (2021) 535–558. [DOI] [PubMed] [Google Scholar]

[R85] [85].Ast TI, Y; Sadre S; McCoy JG; Namkoog G; Chicherin I; Joshi PR; Kamenski P; Suess DLM; Amunts A; Mootha VK, METTL17 is an Fe-S cluster checkpoint for mitochondrial translation, bioRxiv 10.110/2022.11.24.517765 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R86] [86].Camponeschi F, Prusty NR, Heider SAE, Ciofi-Baffoni S, Banci L, GLRX3 Acts as a [2Fe-2S] Cluster Chaperone in the Cytosolic Iron-Sulfur Assembly Machinery Transferring [2Fe-2S] Clusters to NUBP1, J Am Chem Soc 142(24) (2020) 10794–10805. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R87] [87].Li H, Outten CE, The conserved CDC motif in the yeast iron regulator Aft2 mediates iron-sulfur cluster exchange and protein-protein interactions with Grx3 and Bol2, J Biol Inorg Chem 24(6) (2019) 809–815. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R88] [88].Frey AG, Palenchar DJ, Wildemann JD, Philpott CC, A Glutaredoxin.BolA Complex Serves as an Iron-Sulfur Cluster Chaperone for the Cytosolic Cluster Assembly Machinery, J Biol Chem 291(43) (2016) 22344–22356. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R89] [89].Ojeda L, Keller G, Muhlenhoff U, Rutherford JC, Lill R, Winge DR, Role of glutaredoxin-3 and glutaredoxin-4 in the iron regulation of the Aft1 transcriptional activator in Saccharomyces cerevisiae, J Biol Chem 281(26) (2006) 17661–9. [DOI] [PubMed] [Google Scholar]

[R90] [90].Puig S, Vergara SV, Thiele DJ, Cooperation of two mRNA-binding proteins drives metabolic adaptation to iron deficiency, Cell Metab 7(6) (2008) 555–64. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R91] [91].Zhang DL, Ghosh MC, Rouault TA, The physiological functions of iron regulatory proteins in iron homeostasis - an update, Front Pharmacol 5 (2014) 124. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R92] [92].Wang H, Shi H, Rajan M, Canarie ER, Hong S, Simoneschi D, Pagano M, Bush MF, Stoll S, Leibold EA, Zheng N, FBXL5 Regulates IRP2 Stability in Iron Homeostasis via an Oxygen-Responsive [2Fe2S] Cluster, Mol Cell 78(1) (2020) 31–41 e5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R93] [93].Rouault TA, Maio N, How Oxidation of a Unique Iron-Sulfur Cluster in FBXL5 Regulates IRP2 Levels and Promotes Regulation of Iron Metabolism Proteins, Mol Cell 78(1) (2020) 1–3. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R94] [94].Allikmets R, Raskind WH, Hutchinson A, Schueck ND, Dean M, Koeller DM, Mutation of a putative mitochondrial iron transporter gene (ABC7) in X-linked sideroblastic anemia and ataxia (XLSA/A), Hum Mol Genet 8(5) (1999) 743–9. [DOI] [PubMed] [Google Scholar]

[R95] [95].Levi S, Ripamonti M, Dardi M, Cozzi A, Santambrogio P, Mitochondrial Ferritin: Its Role in Physiological and Pathological Conditions, Cells 10(8) (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R96] [96].Jhurry ND, Chakrabarti M, McCormick SP, Gohil VM, Lindahl PA, Mossbauer study and modeling of iron import and trafficking in human jurkat cells, Biochemistry 52(45) (2013) 7926–42. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R97] [97].Wingert RA, Galloway JL, Barut B, Foott H, Fraenkel P, Axe JL, Weber GJ, Dooley K, Davidson AJ, Schmid B, Paw BH, Shaw GC, Kingsley P, Palis J, Schubert H, Chen O, Kaplan J, Zon LI, C. Tubingen Screen, Deficiency of glutaredoxin 5 reveals Fe-S clusters are required for vertebrate haem synthesis, Nature 436(7053) (2005) 1035–39. [DOI] [PubMed] [Google Scholar]

[R98] [98].Ye H, Jeong SY, Ghosh MC, Kovtunovych G, Silvestri L, Ortillo D, Uchida N, Tisdale J, Camaschella C, Rouault TA, Glutaredoxin 5 deficiency causes sideroblastic anemia by specifically impairing heme biosynthesis and depleting cytosolic iron in human erythroblasts, J Clin Invest 120(5) (2010) 1749–61. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R99] [99].Liu G, Guo S, Anderson GJ, Camaschella C, Han B, Nie G, Heterozygous missense mutations in the GLRX5 gene cause sideroblastic anemia in a Chinese patient, Blood 124(17) (2014) 2750–1. [DOI] [PubMed] [Google Scholar]

[R100] [100].Liu G, Wang Y, Anderson GJ, Camaschella C, Chang Y, Nie G, Functional Analysis of GLRX5 Mutants Reveals Distinct Functionalities of GLRX5 Protein, J Cell Biochem 117(1) (2016) 207–17. [DOI] [PubMed] [Google Scholar]

[R101] [101].Baker PR 2nd, Friederich MW, Swanson MA, Shaikh T, Bhattacharya K, Scharer GH, Aicher J, Creadon-Swindell G, Geiger E, MacLean KN, Lee WT, Deshpande C, Freckmann ML, Shih LY, Wasserstein M, Rasmussen MB, Lund AM, Procopis P, Cameron JM, Robinson BH, Brown GK, Brown RM, Compton AG, Dieckmann CL, Collard R, Coughlin CR 2nd, Spector E, Wempe MF, Van Hove JL, Variant non ketotic hyperglycinemia is caused by mutations in LIAS, BOLA3 and the novel gene GLRX5, Brain 137(Pt 2) (2014) 366–79. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R102] [102].Dos Santos PC, Johnson DC, Ragle BE, Unciuleac MC, Dean DR, Controlled expression of nif and isc iron-sulfur protein maturation components reveals target specificity and limited functional replacement between the two systems, J Bacteriol 189(7) (2007) 2854–62. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R103] [103].Kleczewska M, Grabinska A, Jelen M, Stolarska M, Schilke B, Marszalek J, Craig EA, Dutkiewicz R, Biochemical Convergence of Mitochondrial Hsp70 System Specialized in Iron-Sulfur Cluster Biogenesis, Int J Mol Sci 21(9) (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R104] [104].Blahut M, Sanchez E, Fisher CE, Outten FW, Fe-S cluster biogenesis by the bacterial Suf pathway, Biochim Biophys Acta Mol Cell Res 1867(11) (2020) 118829. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R105] [105].Tanaka N, Kanazawa M, Tonosaki K, Yokoyama N, Kuzuyama T, Takahashi Y, Novel features of the ISC machinery revealed by characterization of Escherichia coli mutants that survive without iron-sulfur clusters, Mol Microbiol 99(5) (2016) 835–48. [DOI] [PubMed] [Google Scholar]

[R106] [106].Tsaousis AD, On the Origin of Iron/Sulfur Cluster Biosynthesis in Eukaryotes, Front Microbiol 10 (2019) 2478. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R107] [107].Karnkowska A, Treitli SC, Brzon O, Novak L, Vacek V, Soukal P, Barlow LD, Herman EK, Pipaliya SV, Panek T, Zihala D, Petrzelkova R, Butenko A, Eme L, Stairs CW, Roger AJ, Elias M, Dacks JB, Hampl V, The Oxymonad Genome Displays Canonical Eukaryotic Complexity in the Absence of a Mitochondrion, Mol Biol Evol 36(10) (2019) 2292–2312. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R108] [108].Pena-Diaz P, Lukes J, Fe-S cluster assembly in the supergroup Excavata, J Biol Inorg Chem 23(4) (2018) 521–541. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R109] [109].Smid O, Horakova E, Vilimova V, Hrdy I, Cammack R, Horvath A, Lukes J, Tachezy J, Knock-downs of iron-sulfur cluster assembly proteins IscS and IscU down-regulate the active mitochondrion of procyclic Trypanosoma brucei, J Biol Chem 281(39) (2006) 28679–86. [DOI] [PubMed] [Google Scholar]

[R110] [110].Lukes J, Basu S, Fe/S protein biogenesis in trypanosomes - A review, Biochim Biophys Acta 1853(6) (2015) 1481–92. [DOI] [PubMed] [Google Scholar]

[R111] [111].Maio N, Rouault TA, Outlining the Complex Pathway of Mammalian Fe-S Cluster Biogenesis, Trends Biochem Sci 45(5) (2020) 411–426. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R112] [112].Glatt S, Zabel R, Vonkova I, Kumar A, Netz DJ, Pierik AJ, Rybin V, Lill R, Gavin AC, Balbach J, Breunig KD, Muller CW, Structure of the Kti11/Kti13 heterodimer and its double role in modifications of tRNA and eukaryotic elongation factor 2, Structure 23(1) (2015) 149–160. [DOI] [PubMed] [Google Scholar]

[R113] [113].Wiedemann N, Urzica E, Guiard B, Muller H, Lohaus C, Meyer HE, Ryan MT, Meisinger C, Muhlenhoff U, Lill R, Pfanner N, Essential role of Isd11 in mitochondrial iron-sulfur cluster synthesis on Isu scaffold proteins, EMBO J 25(1) (2006) 184–95. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R114] [114].Gerber J, Neumann K, Prohl C, Muhlenhoff U, Lill R, The yeast scaffold proteins Isu1p and Isu2p are required inside mitochondria for maturation of cytosolic Fe/S proteins, Mol Cell Biol 24(11) (2004) 4848–57. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R115] [115].Yoon H, Knight SA, Pandey A, Pain J, Turkarslan S, Pain D, Dancis A, Turning Saccharomyces cerevisiae into a Frataxin-Independent Organism, PLoS Genet 11(5) (2015) e1005135. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R116] [116].Lange H, Kaut A, Kispal G, Lill R, A mitochondrial ferredoxin is essential for biogenesis of cellular iron-sulfur proteins, Proc Natl Acad Sci U S A 97(3) (2000) 1050–5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R117] [117].Muhlenhoff U, Richhardt N, Ristow M, Kispal G, Lill R, The yeast frataxin homolog Yfh1p plays a specific role in the maturation of cellular Fe/S proteins, Hum Mol Genet 11(17) (2002) 2025–36. [DOI] [PubMed] [Google Scholar]

[R118] [118].Knight SA, Sepuri NB, Pain D, Dancis A, Mt-Hsp70 homolog, Ssc2p, required for maturation of yeast frataxin and mitochondrial iron homeostasis, J Biol Chem 273(29) (1998) 18389–93. [DOI] [PubMed] [Google Scholar]

[R119] [119].Kim R, Saxena S, Gordon DM, Pain D, Dancis A, J-domain protein, Jac1p, of yeast mitochondria required for iron homeostasis and activity of Fe-S cluster proteins, J Biol Chem 276(20) (2001) 17524–32. [DOI] [PubMed] [Google Scholar]

[R120] [120].Rodriguez-Manzaneque MT, Tamarit J, Belli G, Ros J, Herrero E, Grx5 is a mitochondrial glutaredoxin required for the activity of iron/sulfur enzymes, Mol Biol Cell 13(4) (2002) 1109–21. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R121] [121].Campbell CJ, Pall AE, Naik AR, Thompson LN, Stemmler TL, Molecular Details of the Frataxin-Scaffold Interaction during Mitochondrial Fe-S Cluster Assembly, Int J Mol Sci 22(11) (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R122] [122].Porras CA, Rouault TA, Iron Homeostasis in the CNS: An Overview of the Pathological Consequences of Iron Metabolism Disruption, Int J Mol Sci 23(9) (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R123] [123].Fan X, Barshop WD, Vashisht AA, Pandey V, Leal S, Rayatpisheh S, Jami-Alahmadi Y, Sha J, Wohlschlegel JA, Iron-regulated assembly of the cytosolic iron-sulfur cluster biogenesis machinery, J Biol Chem 298(7) (2022) 102094. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R124] [124].Kassube SA, Thoma NH, Structural insights into Fe-S protein biogenesis by the CIA targeting complex, Nat Struct Mol Biol 27(8) (2020) 735–742. [DOI] [PubMed] [Google Scholar]

[R125] [125].Garland SA, Hoff K, Vickery LE, Culotta VC, Saccharomyces cerevisiae ISU1 and ISU2: members of a well-conserved gene family for iron-sulfur cluster assembly, J Mol Biol 294(4) (1999) 897–907. [DOI] [PubMed] [Google Scholar]

[R126] [126].Chan SY, Zhang YY, Hemann C, Mahoney CE, Zweier JL, Loscalzo J, MicroRNA-210 controls mitochondrial metabolism during hypoxia by repressing the iron-sulfur cluster assembly proteins ISCU1/2, Cell Metab 10(4) (2009) 273–84. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R127] [127].Hoff KG, Culler SJ, Nguyen PQ, McGuire RM, Silberg JJ, Smolke CD, In vivo fluorescent detection of Fe-S clusters coordinated by human GRX2, Chem Biol 16(12) (2009) 1299–308. [DOI] [PubMed] [Google Scholar]

[R128] [128].Monfort B, Want K, Gervason S, D’Autreaux B, Recent Advances in the Elucidation of Frataxin Biochemical Function Open Novel Perspectives for the Treatment of Friedreich’s Ataxia, Front Neurosci 16 (2022) 838335. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R129] [129].Ochi T, Fujiwara T, Ono K, Suzuki C, Nikaido M, Inoue D, Kato H, Onodera K, Ichikawa S, Fukuhara N, Onishi Y, Yokoyama H, Nakamura Y, Harigae H, Exploring the mechanistic link between SF3B1 mutation and ring sideroblast formation in myelodysplastic syndrome, Sci Rep 12(1) (2022) 14562. [DOI] [PMC free article] [PubMed] [Google Scholar]

[R130] [130].Iwema T, Picciocchi A, Traore DA, Ferrer JL, Chauvat F, Jacquamet L, Structural basis for delivery of the intact [Fe2S2] cluster by monothiol glutaredoxin, Biochemistry 48(26) (2009) 6041–3. [DOI] [PubMed] [Google Scholar]

PERMALINK

Mitochondria function in cytoplasmic FeS protein biogenesis

Andrew Dancis

Ashutosh K Pandey

Debkumar Pain

Abstract

1. Introduction

Fig. 1. Global scheme for cellular FeS cluster assembly in eukaryotes (yeast components are shown).

2. Mitochondrial FeS cluster assembly

2.1. Synthesis of new [2Fe-2S] clusters on the Isu scaffold

2.2. Nfs1 accessory proteins - Isd11 and Acp1

2.3. Chaperone cycle

2.4. Mitochondrial generation of [4Fe-4S] clusters and their target specific insertion

3. Cytoplasmic and nuclear FeS cluster assembly

4. Mitochondrial exporter Atm1

Fig. 2. Development of the X-S or (Fe-S)int hypothesis.

4.1. Leu1 biosynthesis and dependence on Atm1

4.2. Atm1 and X-S or (Fe-S)int hypothesis

5. Mitochondria requirement for cytoplasmic FeS cluster assembly – Biochemical assays

Fig. 3. Assays involving mixing of mitochondria and cytoplasm.

5.1. Mitochondria-mediated activation of latent FeS clusters in the cytoplasm

Fig. 4. Mitochondria-mediated activation of “latent” FeS clusters in the cytoplasm.

6. Dependence of mitochondrial FeS cluster assembly and cytoplasmic FeS cluster assembly on nucleotides and iron

Fig. 5. Iron for (Fe-S)int originates in mitochondria.

7. Atm1 structure and Atm1 transport cycle

Fig. 6. Chaetomium thermophilum Atm1 (CtAtm1) transport conformations.

8. Glutathione and FeS cluster assembly

Fig. 7. The role of glutathione in FeS cluster assembly.

8.1. Bypassing mitochondrial glutathione transport

8.2. Yeast and glutathione

9. Iron homeostasis and FeS cluster biogenesis

Fig. 8. Iron regulation and FeS cluster assembly.

Fig. 9. Iron regulation in vertebrates.

10. Human diseases associated with FeS cluster deficiency

11. Evolutionary Tinkering

Fig. 10. Evolutionary variations on the theme.

12. tRNA thiolation in mitochondria and cytoplasm

Fig. 11. tRNA thiolation in the cytoplasm.

12.1. Overlap with FeS cluster biogenesis pathways

13. Evidence for involvement of core mitochondrial ISC components and Atm1 in cytoplasmic FeS cluster assembly and iron homeostasis

14. Outlook

Acknowledgements

References

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

Fig. 2. Development of the X-S or (Fe-S)_int hypothesis.

4.2. Atm1 and X-S or (Fe-S)_int hypothesis

Fig. 5. Iron for (Fe-S)_int originates in mitochondria.