Analysis of iron–sulfur protein maturation ineukaryotesAntonio J Pierik, Daili J A Netz & Roland Lill

Institut fur Zytobiologie und Zytopathologie, Philipps-Universitat Marburg, Marburg, Germany. Correspondence should be addressed to R.L. (lill@staff.uni-marburg.de).

Published online 30 April 2009; doi:10.1038/nprot.2009.39

Iron–sulfur (Fe/S) proteins play crucial roles in living cells by participating in enzyme catalysis, electron transfer and the regulation

of gene expression. The biosynthesis of the inorganic Fe/S centers and their insertion into apoproteins require complex cellular

machinery located in the mitochondria (Fe/S cluster (ISC) assembly machinery systems) and cytosol (cytosolic Fe/S protein assembly

(CIA) systems). Functional defects in Fe/S proteins or their biogenesis components lead to human diseases underscoring the

functional importance of these inorganic cofactors for life. In this protocol, we describe currently available methods to follow the

activity and de novo biogenesis of Fe/S proteins in eukaryotic cells. The assay systems are useful to follow Fe/S protein maturation in

different cellular compartments, identify novel Fe/S proteins and their biogenesis factors, investigate the molecular mechanisms

underlying the maturation process in vivo and analyze the effects of genetic mutations in Fe/S protein-related genes. Comprehensive

analysis of one biogenesis component or target Fe/S protein takes about 10 d.

INTRODUCTIONIron–sulfur (Fe/S) clusters are ubiquitous inorganic cofactors thatare present in all forms of life1. Despite the chemical simplicity ofFe/S clusters, their biosynthesis and insertion into apoproteinswithin cells require dedicated and complex machinery2,3. On thebasis of the identification of proteins responsible for the biosynth-esis of bacterial Fe/S enzymes,2,4 a related ISC assembly machineryhas been discovered in mitochondria of eukaryotes. In Baker’syeast, the best-studied eukaryote for this process, 15 proteins areknown as members of the mitochondrial ISC assembly machinery(for recent reviews, see refs. 3,5). The key players are the cysteinedesulfurase complex Nfs1/Isd11, which supplies sulfur to thescaffold protein Isu1 on which an Fe/S cluster is assembledde novo (see Fig. 1) (ref. 3). This biosynthetic step requires theredox chain NADH—ferredoxin reductase (Arh1)—ferredoxin(Yah1) and the putative iron donor frataxin (Yfh1). The preas-sembled Fe/S cluster is then transferred to apoproteins involving adedicated chaperone system6 and the monothiol glutaredoxinGrx5. Some mitochondrial Fe/S proteins further depend on Isa1,Isa2 and Iba57 proteins7 or GTP8 for functional assembly.

In eukaryotes, Fe/S proteins are localized in the mitochondria,cytosol and nucleus9. Strikingly, maturation of the extra-mitochon-drial Fe/S proteins depends on the function of the mitochondrial ISCassembly machinery10. A current working model (Fig. 1) suggeststhat the mitochondria export a (still unknown) compound synthe-sized by virtue of the ISC assembly machinery to the cytosol where itis utilized by the so-called CIA machinery11. The export reaction isfacilitated by members of the ISC export machinery, which encom-pass the mitochondrial inner membrane ABC transporter Atm1, theintermembrane space-located sulfhydryl oxidase Erv1 and glu-tathione. Over the past years, five CIA factors have been identified.Fe/S clusters are first assembled on the P-loop NTPases Cfd1 andNbp35, which thus serve as scaffolds in the eukaryotic cytosol(Fig. 1) (ref. 12). Later steps require the iron-only hydrogenase-like protein Nar1 (ref. 13) and the b-propeller protein Cia1 (ref. 14).Recently, the Fe/S protein Dre2 was found to be another crucialmember of the CIA machinery but its function is still unclear15.

The ISC and CIA proteins found in yeast are highly conserved ineukaryotes,16 suggesting that yeast provides a valuable model forFe/S protein maturation in eukaryotes. In fact, seven mammalianrelatives of yeast ISC and CIA proteins have been experimentallystudied to date and shown to serve a similar function in the cell astheir yeast counterparts17–23. The crucial importance of Fe/Sproteins and their assembly machineries has been recognized byvirtue of the association of lesions in ISC machinery genes withthree human diseases (for a review, see refs. 3,24). First, a GAAtriplet repeat expansion in an intronic region of frataxin is the mostfrequent cause of Friedreich ataxia, an autosomal recessive neuro-degenerative disorder25. Second, a splicing defect in GLRX5(encoding human Grx5) causes sideroblastic anemia26. Third, amyopathy of patients from northern Swedish descent has recentlybeen shown to derive from a splicing defect in the ISCU gene(encoding human Isu1)27,28.

The functional significance of the ISC and CIA machinerieshas gained progressive interest since a number of essential cytosolicand nuclear Fe/S proteins have been identified, the function ofwhich may critically depend on their Fe/S cofactors29–31. In fact, thepresence of essential Fe/S clusters in these proteins may explain theindispensable character of mitochondria and the ISC and CIAmachineries for cell viability29. Ribosome maturation andprotein translation require the 2� [4Fe–4S] cluster-containingprotein Rli1, which resides in both cytosol and nucleus29,32.Other essential Fe/S proteins are involved in the integrity, main-tenance and repair of DNA, but the presence of Fe/S clusters inthese proteins in vivo has not been verified yet30,31. Replication ofthe lagging strand critically depends on the primase complex,which includes the Fe/S cluster-containing large subunit Pri2(ref. 30). The DNA helicase Rad3 (XPD in humans) is one ofthe ten subunits of the RNA polymerase transcription factorIIH and is vital for nucleotide excision DNA repair33. A prokar-yotic homolog of Rad3 has been shown by biochemical andstructural studies to contain an essential [4Fe–4S] clusterinserted into its helicase fold31,34. This protocol describes various

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

NATURE PROTOCOLS | VOL.4 NO.5 | 2009 | 753

PROTOCOL

experimental approaches to verify that these proteins in factcontain Fe/S clusters in the living cell.

Currently, the majority of novel Fe/S proteins are identified byisolation after heterologous expression in Escherichia coli, chemicalreconstitution of Fe/S clusters, cysteine mutagenesis, bioinformaticanalysis and/or X-ray crystallography12,13,30,31. An important ques-tion is the relevance of the Fe/S cluster in the identified protein forthe physiology of the original organism. For example, on synthesisin E. coli metal centers different from those present in the originalhost can be inserted, for instance due to the lack of dedicatedmetallochaperones. Well-known examples are the single iron-con-taining Clostridium pasteurianum rubredoxin35 and the [2Fe–2S]-or [4Fe–4S]-containing Haemophilus influenzae IscU36, both ofwhich bind zinc on heterologous expression in E. coli. Thus,sensitive in vivo assays in the original organism are required tocomplement biochemical and structural approaches used to iden-tify novel Fe/S-containing proteins. Here, we provide severalmethods intended to verify the presence of Fe/S clusters in proteinsof interest in vivo35,36. We describe the use of radiolabeled iron(55Fe) and several enzymatic assays to determine the presence ofFe/S clusters in mitochondrial, cytosolic and nuclear proteins. Wefurther outline how the assays described here serve to unequivocallyidentify new biogenesis components, i.e., members of the ISC andCIA machineries. The use of genetic depletion methods to diminishthe cellular amounts of these usually essential proteins allows theinvestigation of their role in vivo. For instance, the stages at whichthese factors perform their function in the biosynthetic pathwaycan be determined12,37. In turn, this experimental approach opensthe way to address the phenotypes associated with the depletion ofthe ISC and CIA components. A number of additional assaysystems for testing the biogenesis of Fe/S proteins in whole cellswill not be described here because of their indirect character.The reader is referred to the summary of these approaches inrefs. 3,38,39. For the study of maturation of the ferredoxin Yah1

(ref. 40) and aconitase8, specialized protocols have been developedfor the use with isolated mitochondria. Aconitase can be measuredqualitatively in a gel-based assay (zymogram)21. A non-radioactivemethod, which also uses regulatable promoters, has recently beendeveloped for a bacterial system, Azotobacter vinelandii41,42. Thismethod allows determination of both Fe/S cluster amount andtype, but might not be applicable for eukaryotic systems due toinherently lower protein amounts.

Experimental designMeasurement of enzymatic activities of Fe/S proteins. There areseveral assays available to measure Fe/S enzyme activities at theirnatural abundance (Table 1) for testing the cellular Fe/S proteinbiogenesis machinery. As the activity of these enzymes depends onthe presence of a particular cluster type, i.e., [4Fe–4S] in aconi-tase43, activity measurements provide information on both thequantity and the Fe/S cluster integrity. It should be noted, however,that this strategy does not necessarily provide information aboutthe maturation efficiency of Fe/S proteins inside the cell, i.e., aboutde novo Fe/S protein biogenesis activity, as cluster lability, ironstatus43 and oxidative stress44 can lead to the damage or removal ofthe Fe/S cluster from the protein of interest.

Methods for the determination of two cytosolic Fe/S enzymes(isopropylmalate isomerase and sulfite reductase) and two mito-chondrial Fe/S enzymes (succinate dehydrogenase and aconitase)are given in Boxes 1 and 2, respectively. The activity of isopropyl-malate isomerase can be detected by the formation of the UV-absorbing double bond of isopropylmaleate on dehydration of 3-isopropylmalate10. In a similar manner, formation of cis-aconitatefrom isocitrate in concentrated aconitase preparations (mitochon-dria) can be followed10. For diluted aconitase samples (whole-cellextract), this assay cannot be used due to the high UV absorbancefrom DNA and other proteins. Instead, formation of NADPHat 340 nm can be recorded7, if the aconitase-catalyzed

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

Figure 1 | Current model for Fe/S protein

maturation in eukaryotes. After import of reduced

iron (red circle) into mitochondria by the carrier

proteins, Mrs3/4 (mitoferrin), the iron-binding

protein frataxin is believed to deliver Fe to the

scaffold protein Isu1 (yeast contains a second

highly related protein termed Isu2). Sulfur

(yellow circle) is released from cysteine by the

cysteine desulfurase complex Nfs1-Isd11

generating alanine. The sulfur is transferred to

Isu1 and possibly reduced by the electron transfer

chain NADH—ferredoxin reductase Arh1—

ferredoxin Yah1. These reactions lead to the de

novo assembly of an Fe/S cluster on Isu1. The

cluster is then transferred to apoproteins (Apo), a

reaction facilitated by the glutaredoxin Grx5 and

the chaperone system Ssq1-Jac1-Mge1. Functional

activation of aconitase- and biotin synthase-like

Fe/S proteins requires the additional function of

Isa1-Isa2-Iba57. Extra-mitochondrial Fe/S proteins

depend on most of the mitochondrial ISC assembly

components for maturation as well as on the ISC

export machinery with the central component

Atm1, which exports an unknown (?) compound to the cytosol for use by the CIA machinery. This leads to assembly of Fe/S clusters on the P-loop NTPases Cfd1-

Nbp35. The Fe/S cluster is then transferred to cytosolic and nuclear apoproteins involving the function of Nar1 and Cia1. The functional step of the essential CIA

protein Dre2 has not been identified yet. GSH, glutathione. For further details, see text and ref. 3.

Cfd1

Nbp35

CIA machinery

Mitochondrial Fe/S proteins

CytosolMitochondrion

Iron (Fe2+)

Mrs3/4

Ssq1-ATPJac1/Mge1

Iba57Isa1-Isa2

Yah1-Arh1-NADH

Ala

Grx5

GTP?Nfu1?

ISC exportmachinery

Cytosolic and nuclearFe/S proteins

Nar1 - Cia1

GSH

?

Atm1

Erv1

Frataxin

Nfs1-Isd11

Cfd1

Nbp35

Dre2

Apo

Apo

Isu1 Isu1

ISC assemblymachinery

Cys- SH

Aconitasebiotin synthase

754 | VOL.4 NO.5 | 2009 | NATURE PROTOCOLS

PROTOCOL

isocitrate formation from cis-aconitate is coupled with isocitratedehydrogenase and NADP+. Succinate dehydrogenase activity inintact mitochondria harboring intrinsic quinones and the bc1

complex is detected by cytochrome c reduction7. If the bc1 complexis lacking or the mitochondria are damaged, reduction of theartificial electron acceptor DCPIP can be used instead. The assayfor sulfite reductase employs detection of the acid-labile sulfideformed from NADPH-dependent SO3

2� reduction. NADPH is(re)generated in situ from NADP+ with glucose-6-phosphatedehydrogenase and glucose-6-phosphate14. In a noncontinuousassay, the enzymatically formed sulfide condenses in acidic envir-onment in an Fe3+-dependent manner with two N,N-dimethyl-p-phenylene diamine molecules to the conveniently detectablemethylene blue45.

Ideally, Fe/S proteins used as biosynthesis markers should beabundant enzymes that are expressed in a largely constitutivemanner and the activity of which is not influenced by growthconditions. However, as the majority of the Fe/S proteins aremetabolic enzymes or may be subject to regulation similar to anynon-Fe/S protein involved in biosynthetic processes, inherentalterations of the protein levels may occur that have to be accountedor corrected for. For instance, decreased levels of sulfur (methio-nine/cysteine) influence sulfite reductase expression in yeast46.Furthermore, the type of carbon source utilized determines theextent of fermentation and thereby influences the activities offermentative cytosolic and mitochondrial enzymes, includingthose of the respiratory chain and aconitase47. The yeast strainused should also be considered. High activity of isopropylmalateisomerase (Leu1) is particularly observed in the yeast W303 geneticbackground10. Effects by strain and medium differences are notconfined to the yeast system. In human cell lines, complicationsmay arise from significant metabolic responses. Finally, cell densityaffects the cellular iron status, which in turn can complicate Fe/Sprotein activity measurements, in particular those of cytosolicaconitase/IRP1 (ref. 23).

Caution should be taken with the use of enzymes that requiremultiple cofactors (Table 1). The presence of flavin and molybde-num cofactor (MoCo) is a serious drawback for the use of xanthinedehydrogenase (human and plant cells) and that of flavin andsiroheme for sulfite reductase (yeast) as Fe/S protein markerenzymes. Biosynthesis of MoCo requires other Fe/S proteins and

the lack of insertion of other cofactors could hinder Fe/S clusterassembly48. Certain enzymes, which in some organisms serve asconvenient indicators for Fe/S protein maturation, are absent inothers, and vice versa. For example, Baker’s yeast lacks a respiratorycomplex I and xanthine dehydrogenase but contains isopropylma-late isomerase and dihydroxyacid dehydratase (Ilv3), whereashuman cells have the opposite configuration3. An even morecomplex situation is found for DNA glycosylases, adenosine-5¢-phosphosulfate reductase, ferrochelatase and the AC40 subunit ofRNA polymerase II,49 which are present in plants, fungiand humans but, depending on the organism, possess or lackFe/S clusters.

The sensitivity of Fe/S enzyme detection can usually be increasedby cell fractionation and isolation of the respective compartmentsharboring the Fe/S enzyme of interest. This procedure substantiallydiminishes background values and in many cases is crucial fordetection of low levels of activity (e.g., for cytosolic aconitase inhuman cells19). If necessary, marker enzymes can be artificiallyintroduced into an organism or cellular compartment by ectopicexpression. This methodology allows convenient determination ofsoluble enzyme activities. Examples include cytosolic humanaconitase (IRP1) in a yeast deletion mutant lacking mitochondrialaconitase50, cytosolic15 or mitochondria-targeted51 isopropylma-late isomerase (Leu1) in a yeast cell lacking or with low cytosolicLeu1. In addition to the introduction of suitable marker Fe/Sproteins, such experiments show that central parts of both theISC and CIA machineries do not exhibit pronounced targetspecificity for Fe/S proteins of one given compartment.50,51

Incorporation of radioactive iron into Fe/S proteins in vivo. Thepulse radiolabeling of yeast cells with 55Fe provides a faithful assayfor the estimation of the de novo biogenesis of Fe/S cluster-contain-ing proteins in vivo10. A similar approach has been employed tostudy Fe/S protein assembly in isolated or lysed mitochondriain vitro52. Even though the high-energy isotope 59Fe (b-emitter,1.6 MeV, t1/2 ¼ 45 d) can be utilized in this assay, 55Fe (electroncapture, 5.9 keV, t1/2; ¼ 1,005 d) is preferable because of its lowradiation energy53. In our hands, Fe/S proteins such as Leu1 (ref.10) and Aco1 (ref. 7) with an abundance of 410,000 molecules peryeast cell54 can be detected at their natural abundance. Thosepresent at lower abundance (see refs. 7,12–14,55) require

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

TABLE 1 | Fe/S cluster-containing enzymes suitable for monitoring Fe/S protein maturation.

Protein Cluster type Other cofactors Description Reference

Aconitase 4Fe/3Fea None NADPH formation from cis-aconitate withisocitrate dehydrogenase

23

cis-Aconitate formationb 10Homoaconitase (Lys4) 4Fe None Homoaconitate formation from homoisocitrate 69Sulfite reductase 4Fe Siroheme, FAD, FMN Methylene blue formation from HS� produced by

sulfite reductionc14

Isopropylmalate isomerase (Leu1) 4Fe None Formation of isopropylmaleate from3-isopropylmalate

10

Complex II (succinate dehydrogenase) 4Fe, 3Fe, 2Fe FAD, cytochrome b Succinate-dependent DCPIP reduction 10Complex III (cytochrome c reductase)d 2Fe Cytochromes b and c1 Cytochrome c reduction from succinate 10Xanthine dehydrogenase 2 � 2Fe MoCo, FAD Production of superoxide as detected by the

Amplex Red method48

aThe enzymatically active [4Fe-4S]2+ (4Fe) form of aconitase can be converted to the apo- or [3Fe-4S]+ (3Fe) form depending on the conditions used for cell growth, preparation of mitochondria or cell extract. 2Fe,[2Fe-2S] cluster. bThe high UV absorbance of nucleotides prevents the use of this assay for cell extracts. cDiscontinuous assay. dNote that the assay measures both complexes II and III.

NATURE PROTOCOLS | VOL.4 NO.5 | 2009 | 755

PROTOCOL

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

BOX 1 | ASSAYS FOR MEASURING Fe/S ENZYME ACTIVITY IN WHOLE-CELL EXTRACTS

MATERIALSReagents� Saccharomyces cerevisiae strain W303-1A (MATa ura3-1 ade2-1 trp1-1 his3-11,15 leu2-3,112 can1-100) (strain 208352 from ATCC)� Tris-HCl (AppliChem, cat. no. A2264)� Disodium EDTA (Acros Organics, cat. no. 147850010)� NaCl (Roth, cat. no. 9652)� Glycerol (Sigma-Aldrich, cat. No. G7757)� Triton X-100 (Sigma-Aldrich, cat. no. T9284)� Phenylmethylsulfonyl fluoride (PMSF) (AppliChem, cat. no. A0999) ! CAUTION PMSF is toxic; wear suitable protective clothing, glovesand eye/face protection.� Ethanol (AppliChem, cat. no. A3693)� Triethanolamine hydrochloride (Fluka, cat. no. 90290)� NaOH (AppliChem, cat. no. A0991)� NADP (Roth, cat. no. AE13.3)� cis-Aconitic acid (Sigma-Aldrich, cat. no. A-3412)� Isocitrate dehydrogenase from porcine heart (IDH; Sigma-Aldrich, cat. no. I-2002)� DL-threo-3-Isopropylmalic acid (Wako, cat. no. 096-02681)� K2HPO4 (Roth, cat. no. 6878.1)� KH2PO4 (AppliChem, cat. no. A1364)� Glucose-6-phosphate dipotassium salt (Sigma-Aldrich, cat. no. G7375)� Glucose-6-phosphate dehydrogenase (Calbiochem, cat. no. 346774)� Na2SO3 (Sigma-Aldrich, cat. no. S-6547)� N,N-Dimethyl-p-phenylene diamine hydrochloride (DMPD; Sigma-Aldrich, cat. no. G-6378) ! CAUTION DMPD is neurotoxic; wear suitableprotective clothing, gloves and eye/face protection.� 25% HCl (7.7 M, Roth, cat. no. 6331-4)� FeCl3 (Sigma-Aldrich, cat. no. F-2877)� Li2S (Sigma-Aldrich, cat. no. 213241) ! CAUTION Li2S is toxic; wear suitable protective clothing, gloves and eye/face protection.Double-distilled deionized water (ddH2O) produced with a Elix 5 UV Millipore systemEquipment� Vortex (Genie-2 vortex; Scientific Instruments)� Centrifuges: Eppendorf 5810-R (Eppendorf), Beckman Coulter Avanti J-20XP and Optima LE-80K (Beckman Inc.)� Spectrophotometer Jasco V-550 (Jasco Inc.) or any suitable double-beam spectrophotometer� Glass (type 104B-OS) and Suprasil (type 104B-QS) quartz cuvettes from Hellma� 1.5-ml microfuge tubesReagent setupUnless indicated otherwise, all reagents are prepared on the day of use. Frozen solutions can be stored up to 1 month.� TNETG buffer: 10 mM Tris/Cl pH 7.4, 2.5 mM EDTA, 150 mM NaCl, 10% (vol/vol) glycerol, 0.5% (vol/vol) Triton X-100: per 500 ml use 5 mlof 1 M Tris-HCl pH 7.4, 2.5 ml of 0.5 M disodium EDTA (pH 7.4), 37.5 ml of 2 M NaCl, 50 ml of glycerol and 25 ml of 10% (wt/vol) Triton X-100� 200 mM PMSF in 100% ethanol. m CRITICAL STEP Prepare just before use; PMSF rapidly hydrolyses. ! CAUTION PMSF is toxic; wearsuitable protective clothing, gloves and eye/face protection.� 0.1 M NADP+ in ddH2O� 20 mM cis-aconitic acid in ddH2O� Triethanolamine buffer: 100 mM triethanolamine adjusted to pH 8 with 1 M NaOH� IDH 40 U ml�1 in triethanolamine buffer with 10% (wt/vol) glycerol, store in small aliquots at �80 1C� Isopropylmalate isomerase (Leu1) buffer: 20 mM Tris-Cl pH 7.4, 50 mM NaCl� 10 mM DL-threo-3-isopropylmalic acid in ddH2O, dilute freshly from a 100 mM stock solution in ddH2O stored frozen at �20 1C� Phosphate buffer: mix 1 M K2HPO4 and 1 M KH2PO4 in the right proportion to obtain pH 7.5� 100 mM glucose-6-phosphate dipotassium salt in 20 mM phosphate buffer (pH 7.5)� 100 U ml�1 glucose-6-phosphate dehydrogenase, dissolve 500 U in 5 ml of 0.1 M phosphate buffer (pH 7.5), store in smallaliquots at �80 1C� 10 mM Na2SO3 in ddH2O, prepare just before use� 20 mM DMPD in 7.2 M HCl. ! CAUTION DMPD is neurotoxic; wear suitable protective clothing, gloves and eye/face protection.� 30 mM FeCl3 in 1.2 M HCl� 1 mM Li2S calibration standard solution in ddH2O, prepared by dilution from a 100 mM solution of Li2S in ddH2O.m CRITICAL STEP Prepareboth solutions just before use; Li2S oxidizes. ! CAUTION Li2S is toxic; wear suitable protective clothing, gloves and eye/face protection.

Cell extract preparation � TIMING B45 min1. Grow a 100 ml yeast culture in minimal or complete medium overnight by shaking at 150 r.p.m., 30 1Cm CRITICAL STEP For sulfite reductase activity measurements, use minimal medium lacking methionine, cysteine and sulfide.

756 | VOL.4 NO.5 | 2009 | NATURE PROTOCOLS

PROTOCOL

overproduction by substitution of the chromosomal promoter orexpression from high copy plasmids utilizing suitable strongpromoters (TDH3, GAL1-10 or MET25)56. The high naturalconcentration of iron in growth media and inside cells requiresthe depletion of ambient iron by cell cultivation at submicromolariron concentrations (‘iron poor’ conditions) before addition of theradioisotope10. This requirement does not pose a limitation inexperiments with yeast cells, as their growth is only mildly affectedat low iron concentrations and the frequent use of plasmids urgesthe use of defined (minimal) media. For radiolabeling of humancell lines, however, the depletion of iron severely decreases cellgrowth23, and an efficient radiolabeling procedure has not yetbeen described.

Our standard radiolabeling procedure is described in this pro-tocol. Radiolabeling of iron-starved yeast cells with 55Fe is per-formed for 1–4 h in the presence of the reducing agent ascorbate.The presence of ascorbate prevents the formation of ferric pre-cipitates in the growth medium. Following radiolabeling, cells arelysed with glass beads, a soluble cell extract is prepared and the55Fe/S proteins are immunoprecipitated with specific antibodies

(see Fig. 2). Alternatively, N- or C-terminal epitope tags can beused to affinity purify the Fe/S proteins of interest. In our hands,the tandem affinity purification (TAP), triple hemagglutinin (3HA)or Myc tags work well, provided the proteins are not functionallycompromised. The amount of radioactive iron associated with theFe/S protein of interest is quantified by liquid scintillation count-ing. The specificity of the 55Fe association as part of an Fe/S clustermay be verified by testing the dependence on members of the ISCand CIA machineries (see below). On depletion of particular ISCand CIA components by growth on glucose of yeast strainsharbouring a GAL promotor in front of the respective genes, the55Fe incorporation into Fe/S proteins usually drops at least five-fold10,13,14,29,51,55. As a control, the cellular uptake of 55Fe duringthe labeling period is measured, which reflects the extent ofcontamination with nonradioactive Fe in the medium. Vital 55Feuptake is also a viability and fitness parameter for the cells as activetransport of Fe against a concentration gradient can only be carriedout, if the cellular energy status is not compromised. In highereukaryotes, ferrochelatase is a mitochondrial Fe/S protein3, and theformation of radioactive heme, therefore, can be used to indirectly

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

2. Harvest cells by centrifugation for 5 min at 2,000g, room temperature (20–24 1C) remove the supernatant and resuspend the cell pellet with10 ml of ddH2O, transfer to a 15-ml Falcon tube and repeat the centrifugation.m CRITICAL STEP All subsequent steps should be carried out at 4 1C.3. Resuspend the pellet in 500 ml of ice-cold TNETG buffer; add 10 ml of 0.2 M PMSF and 1 ml of glass beads (with a porcelain spoon). Close thelid tightly. Vortex at maximum speed with the lid side of the tubes contacting the rotating plate for 1 min, repeat the vortexing three timeswith intermittent 1 min cooling periods on ice.m CRITICAL STEP Ensure that there are no glass beads trapped in the lid.4. Centrifuge for 5 min at 2,000g at 4 1C. Transfer the supernatant to 1.5-ml microfuge tubes and centrifuge for 10 min at 13,000g. Transfer thesupernatant to a new microfuge tube. Save 50 ml for protein determination73.5. The supernatant (typically 3–6 mg ml�1) can be used to measure Fe/S enzyme activity using the assays of options A–C. These enzymemeasurements can be performed under aerobic conditions but should be carried out immediately after isolation of the cell extracts to preventdeterioration of the enzymes.(A) Aconitase (coupled assay) � TIMING B1 h (including cell extract preparation)

(i) Pipette 950 ml of triethanolamine buffer, 12 ml of 20 mM cis-aconitic acid, 10 ml of 40 U ml�1 IDH, 12 ml of 0.1 M NADP+ and cellextract (50 mg protein) into the sample cuvette. Omit cis-aconitic acid and IDH in the reference cuvette. Mix well.(ii) Measure the increase of absorbance at 340 nm for 2–4 min; a short lag phase is usually observed. De340 nm ¼ 6,200 M�1 cm�1.

(B) Sulfite reductase � TIMING B1.5 h (including cell extract preparation)(i) During the cell extract preparation step, prepare a master mix for 10 reactions: 500 ml of 1 M phosphate buffer, pH 7.5, 10 ml of 0.1 MNADP+, 500 ml of 0.1 M glucose-6-phosphate solution, 25 ml of glucose-6-phosphate dehydrogenase (0.1 U ml�1) and 2.965 ml of ddH2O.(ii) Aliquot 400 ml of the master mix into 1.5-ml microfuge tubes, rapidly add sample (50 ml buffer for the reagent blank, 10–50 ml Li2Scalibration standard solution adjusted to 50 ml with buffer or 50 ml of cell extract (B200 mg of protein)), mix and incubate for 10 min at 30 1C.! CAUTION Li2S is toxic; wear suitable protective clothing, gloves and eye/face protection.(iii) Add 50 ml of 10 mM Na2SO3 and then immediately add 100 ml of DMPD solution, 100 ml of ferric reagent and mix. This sample servesas t¼0 min control.! CAUTION DMPD is neurotoxic; wear suitable protective clothing, gloves and eye/face protection.(iv) Incubate the remaining tubes for 10, 20 or 40 min in a thermoblock at 37 1C. At each time point, add 100 ml of DMPD solution and100 ml of ferric reagent and mix.(v) Incubate for 20 min at room temperature (i.e., 20–24 1C) to allow the color to develop. If sulfide has been formed from sulfite, a clearblue color is visible.(vi) Centrifuge for 2 min at 13,000g at room temperature and measure the absorbance at 670 nm. Typically, one finds e670 nm ¼ 14,000–25,000 M�1 cm�1 for the Li2S calibration standard under the conditions described.(vii) For the cell extract, calculate the sulfide formation using the calibration results and determine the sulfite reductase activity fromthe steepest linear part of the time course.

(C) Isopropylmalate isomerase (Leu1) � TIMING B1 h (including cell extract preparation)(i) Mix 970 ml of Leu1 buffer, 20 ml of 10 mM DL-threo-3-isopropylmalic acid and 5–10 ml of cell extract in a quartz cuvette.(ii) Measure the increase of absorbance at 235 nm for 90 s. De235 nm ¼ 4,530 M�1 cm�1.

BOX 1 | CONTINUED

NATURE PROTOCOLS | VOL.4 NO.5 | 2009 | 757

PROTOCOL

sample Fe/S cluster biogenesis (Fig. 2). Owing to its high solubilityin organic solvents, heme can be quantitatively extracted from thecell lysates into the organic phase after acidification57. The degree of55Fe incorporation into heme is quantified by liquid scintillation

counting. In a limited number of cases, the inherent lability ofprotein-bound Fe/S clusters may require specialized experimentalconditions such as anaerobic immuno- or affinity purification ofthe Fe/S protein, shorter and/or less stringent washing conditions

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

BOX 2 | ASSAYS FOR MEASURING Fe/S ACTIVITY IN ISOLATED MITOCHONDRIA

MATERIALSReagents� Tris-HCl (AppliChem, cat. no. A2264)� NaCl (Roth, cat. no. 9652)� DL-Isocitric acid trisodium salt (Sigma-Aldrich, cat. no. I-1252)� Dodecylmaltoside (Calbiochem, cat. no. 324356)� MgCl2 (Sigma-Aldrich, cat. no. M-0250)� K2HPO4 (Roth, cat. no. 6878.1)� KH2PO4 (AppliChem, cat. no. A1364)� NADP (Roth, cat. no. AE13.3)� KCN (Sigma-Aldrich, cat. no. 20.781-0) ! CAUTION KCN is toxic; wear gloves, safety glasses and ensure good ventilation.� Sodium succinate (Sigma-Aldrich, cat. no. S-2378)� Sodium malonate (Sigma-Aldrich, cat. no. M-1875)� Bovine heart cytochrome c (Sigma-Aldrich, cat. no. C-2037)� n-Decylubiquinone (Sigma-Aldrich, cat. no. D-7911)� Dichlorophenol indophenol (DCPIP; Fluka, cat. no. 33125)Equipment� Spectrophotometer Jasco V-550 (Jasco Inc.) or any suitable double-beam spectrophotometer.� Glass (type 104B-OS) and Suprasil (type 104B-QS) quartz cuvettes from Hellma� 1.5-ml microfuge tubesReagent setupUnless indicated otherwise, all reagents are prepared on the day of use.� Buffer A: 50 mM Tris-HCl pH 8.0, 50 mM NaCl� Aconitase buffer: dissolve 516 mg of DL-isocitric acid trisodium salt in 100 ml of buffer A� Mitochondria lysis buffer: dissolve 4.8 mg of dodecylmaltoside in 2 ml of buffer A� Triethanolamine buffer: 50 mM triethanolamine-Cl pH 8.5, 50 mM NaCl, 1.5 mM MgCl2� 100 mM KCN in ddH2O. ! CAUTION KCN is toxic; wear gloves, safety glasses and ensure good ventilation.� 20% (wt/vol) sodium succinate in buffer A� 20% (wt/vol) sodium malonate in buffer A� Bovine heart cytochrome c, 20 mg ml�1 in ddH2O� 10 mM n-decylubiquinone in 100% ethanol� 10 mM DCPIP in ddH2O

1. Isolate yeast mitochondria as described in detail elsewhere74.2. Determine the protein concentration with the Bradford method73.3. The enzymatic activity of the mitochondrial Fe/S protein can be measured using the assays described in options A–C.(A) Aconitase (direct assay) � TIMING B5 min

(i) Add 950 ml of aconitase buffer to a quartz cuvette.(ii) Dissolve mitochondria (10–20 mg of protein) in 60 ml of mitochondria lysis buffer, mix well and directly proceed with the activitymeasurement.(iii) Add 50 ml of the lysed mitochondria, mix well and measure the absorbance increase at 235 nm for 2 min. De235 nm ¼ 4,950 M�1 cm�1.

(B) Succinate dehydrogenase (DCPIP reduction assay) � TIMING B5 min(i) Pipette in the following order (reference cuvette): 950 ml of buffer A, 7 ml of 10 mM n-decylubiquinone, 10 ml of 10 mM DCPIP, 12 ml of20% (wt/vol) malonate, 12 ml of 20% (wt/vol) succinate and isolated mitochondria (10–20 mg protein). For the sample cuvette, omit themalonate solution. Mix well.(ii) Measure the increase of absorbance at 600 nm of the sample cuvette versus the reference cuvette for 2 min.De600 nm ¼ 21,000 M�1 cm�1.

(C) Succinate dehydrogenase (cytochrome c reduction assay) � TIMING B5 min(i) Immediately before beginning the assay, mix 200 ml of 100 mM KCN solution with 19.8 ml of buffer A.! CAUTION KCN is toxic; wear gloves, safety glasses and ensure good ventilation.(ii) Pipette in the following order (reference cuvette): 920 ml of buffer A (1 mM KCN), 12 ml of 20% (wt/vol) malonate, 12 ml of 20% (wt/vol)succinate, 50 ml of cytochrome c and isolated mitochondria (10–20 mg protein). For the sample cuvette, omit the malonate solution. Mix well.m CRITICAL STEP Ensure that the reaction is measured for a sufficient amount of time: after an initial lag phase (up to 2 min) the linearphase with maximal activity occurs between 2 and 4 min.(iii) Measure the increase of absorbance at 550 nm of the sample cuvette versus the reference cuvette for 5 min.De550 nm ¼ 20,000 M�1 cm�1.

758 | VOL.4 NO.5 | 2009 | NATURE PROTOCOLS

PROTOCOL

than those used for stable Fe/S proteins and optimized times forradiolabeling. As 55Fe incorporation does not give any informationon the cluster type, ideally radiolabeling assays and methods thatdetect the cluster type (i.e., enzymatic activity measurement) areperformed in parallel12–14,55.

The purchase, storage, use and disposal of the 55Fe isotoperequire an appropriate license, training of staff, planning or con-struction of facilities (isotope lab) and monitoring of safety andcontamination.

Depletion of ISC or CIA components. The specificity of theassays described above can best be verified by analyzing thedependence of the Fe/S enzyme activities or 55Fe/S cluster incor-poration on the function of known components of the ISC and CIAmachineries10,29. Mitochondrial target Fe/S proteins specificallydepend on the mitochondrial ISC assembly machinery for matura-tion, whereas cytosolic and nuclear Fe/S proteins require theactivities of the two ISC systems and the CIA machinery3

(Fig. 1). Potential novel members of the three machineries identi-fied by other methods (genetic screens and bioinformatics) can beinvestigated. Deletion of the majority of the ISC and CIA genes islethal necessitating genetic means to downregulate synthesis of theencoded proteins by regulated gene expression. A convenientpromoter is that of the GAL1-10 gene, which is induced in thepresence of galactose and repressed by glucose. To create aGal-YFG1 (your favorite gene 1) strain, the endogenous YFG1promoter is exchanged for that of GAL1-10 by generating a suitablePCR product that can be inserted in front of the YFG1 promoterlocus by homologous recombination58. Attenuated versions of thispromoter termed GALL have been developed avoiding overexpres-sion with galactose and allowing faster depletion in the presence of

glucose. It is crucial to test for possible sugar effects that may arise,e.g., from the switch from respiratory to fermentative growth. Asgeneral glucose repression is particularly weak in the W303 yeastgenetic background, this strain is well suited for the use of theGAL1-10 promoter during (mitochondrial) Fe/S protein biogenesisinvestigations. Other suitable depletion techniques are the tetra-cycline-regulated promoter system59, the heat-induced proteolysiswith the degron system60, temperature-sensitive strains61 and thedecreased abundance by mRNA perturbation approach62. Down-regulation of the target genes must be optimized individually toassure the occurrence of phenotypical consequences, yet no viabi-lity loss. The specificity of downregulation should be verified bycomplementation with a plasmid-borne YFG1 gene. For theGal-YFG1 strains, depletion to physiologically critical levels isachieved by several passages in liquid media under repressiveconditions (see Fig. 2b in ref. 29). Once the behavior of theGal-YFG1 strain has been determined, the yeast cells can be usedfor the assays described above. If necessary, the yeast strain can betransformed with a plasmid encoding a potential Fe/S protein usinga non-sugar-dependent promoter.

Depletion of the ISC and CIA proteins in human cells such asHeLa cell lines is achieved by the RNA interference technology (seee.g., ref. 19). Depletion of the proteins under study is followed bywestern blotting. To avoid off-target effects of the RNA interferencetreatment, it is recommended to validate the specificity of thephysiological depletion effects by complementing the cells bysynthesis of the native protein using genes with silent mutationsthat are resistant to RNA interference. In many cases, one round ofRNA interference treatment is not sufficient to reach critical levelsof depletion. Therefore, it is useful to repeat the treatment until areasonable depletion is achieved without losing cell viability17.

MATERIALSREAGENTS.Yeast nitrogen base without amino acids (ForMedium, cat. no. CYN0510).Yeast nitrogen base without amino acids, iron-free (ForMedium, cat. no.

CYN1202).(NH4)2SO4 (Roth, cat. no. 3746.1).Galactose (Sigma-Aldrich, cat. no. G0625).Glucose (Sigma-Aldrich, cat. no. G7528).Agar-agar (Roth, cat no. 5210.2).Yeast extract (Roth, cat. no. 2363.2).Casein peptone (MP Biomedicals, cat. no. 3066-132).Phenylmethylsulfonyl fluoride (PMSF) (AppliChem, cat. no. A0999)! CAUTION PMSF is toxic; wear suitable protective clothing, gloves andeye/face protection.

.Ethanol (AppliChem, cat. no. A3693)

.NaOH (AppliChem, cat. no. A0991)

.2-Mercaptoethanol (Fluka, cat. no. 63689)

.Trichloroacetic acid (TCA) (Roth, cat. no. 8789.2) ! CAUTION TCA iscorrosive, causes severe burns; wear suitable protective clothing, gloves andeye/face protection.

.Acetone (Roth, cat. no. 9372.5)

. 55FeCl3 (NEN/Perkin Elmer, cat. no. NEZ043-110711B) ! CAUTION Causescancer, designate area for handling 55Fe and clearly label all containers. StoremCi quantities of 55Fe behind thin lead shielding. Wear disposable lab coats,wrist guards and gloves for secondary protection. For more detailedinformation, see instructions of occupational limits in the NRC regulations(10 CFR) part 20, Standards for protection against radiation (http://www.nrc.gov/reading-rm/doc-collections/cfr/part020/).

.25% HCl (7.7 M; Roth, cat. no. 6331.4)

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

ISC, exportand CIA

machineries

Plasmid

TAP tag

Targetprotein

[55Fe-S] Quantitation of 55Fe by scintillation

counting

Preparationof cell extract

Cellular iron uptake

Heme extraction

Immunoblot

Immunoprecipitation

Bead

[55Fe-S]

Growth in the presence of 55Fe

Yeast cell

Figure 2 | Experimental strategy of the in vivo 55Fe incorporation into Fe/S

proteins. In this example, a plasmid drives constitutive expression of a target

Fe/S protein (blue) with a C-terminally fused TAP tag (red) in a yeast cell.

After radiolabeling with 55Fe and the preparation of a cell extract by glass

beads under nondenaturing conditions, the target protein is affinity-isolated

through binding of the protein A domain of the TAP tag to the Fc part of IgG

(black Y) covalently coupled to Sepharose beads (blue circle). The beads are

washed and the 55Fe radioactivity indicative of Fe/S cluster insertion into the

apoprotein is quantified by liquid scintillation counting. In parallel controls,

the cellular 55Fe uptake may be followed by measuring the radioactivity

present in the cell extract. Heme formation can be estimated by organic

extraction and scintillation counting, and the amount of protein is evaluated

by western blotting.

NATURE PROTOCOLS | VOL.4 NO.5 | 2009 | 759

PROTOCOL

.Sodium ascorbate (Fluka, cat. no. 11140)

.Trisodium citrate dihydrate (Roth, cat. no. 3580.3)

.Disodium EDTA (AcrosOrganics, cat. no. 147850010)

.HEPES (Sigma-Aldrich, cat. no. H3375)

.KOH (Roth, cat. no. 6751.1)

.Tris-HCl (AppliChem, cat. no. A2264)

.NaCl (Roth, cat. no. 9652)

.Glycerol (Sigma-Aldrich, cat. no. G7757)

.Triton X-100 (Sigma-Aldrich, cat. no. T9284)

.Glass beads (+ 0.7-1.0 mm; Roth, cat. no. A554.1)

.IgG Sepharose 6 Fast Flow beads (GE Healthcare, cat. no. 17-0969-01)

.Protein A-Sepharose CL-4B (GE Healthcare, cat. no. 17-0780-01)

.Anti-hemagglutinin (F-7) agarose beads (Santa Cruz, cat. no.Sc-7392AC)

.Anti-Myc (A-14) agarose beads (Santa Cruz, cat. no. Sc-789AC)

.Antibodies specifically raised against target proteins

.Ultima gold liquid scintillation fluid (Perkin Elmer, cat. no. 6013329)! CAUTION irritant, wear suitable protective clothing, gloves and eye/faceprotection.

.Reagents for SDS–polyacrylamide gel electrophoresis (PAGE) electrophoresisand western blotting

.Double-distilled deionized water (ddH2O) produced with an Elix 5 UVMillipore system

EQUIPMENT.Simple spectrophotometer suitable for the measurement of optical density

(OD) at 600 nm (Genesys 20, Thermo Scientific) and double-beamspectrophotometer for accurate absorbance measurements (Jasco V-550spectrophotometer, Jasco Inc.)

.Vortex (Genie-2 vortex, Scientific Industries)

.pH meter (C6840, Schott Instruments)

.Rotary shaker (type Reax2, Heidolph)

.Incubator (Thermo Forma Scientific Incubator Stericult 3035, ThermoForma Scientific) and shaking incubator (Multitron, Infors HT) for growthof yeast cells

.Centrifuges: for 1.5-ml microfuge tubes (Eppendorf 5810-R; Eppendorf),swing out centrifuge for harvest of yeast cells and separation of glass beads in50- and 15-ml Falcon tubes (type 5810R, Eppendorf) and fix-angle high-speed centrifuge for fractionation of cell extracts (Beckman Coulter OptimaLE-80K, Beckman Inc.)

.Scintillation counter (LS 6500, Beckman Coulter Inc.)

.Disposable plastic tubes (15 and 50 ml, Falcon) and 1.5-ml microfuge tubes

.Electrophoresis and western blotting supplies (see ref. 63)REAGENT SETUPAll reagents are stored at room temperature (i.e., 20–24 1C) and are stable for 1month, unless indicated otherwise.Minimum essential medium (SC) (1 l) Dissolve 1.7 g of yeast nitrogenbase without amino acids and 5 g of (NH4)2SO4 in 900 ml of ddH2O (add 20 gof agar-agar for solid agar plates). After autoclaving, add 100 ml of sterile20% (wt/vol) galactose or glucose stock solution and auxotrophic markersappropriate for the yeast strains and plasmids used64. For SC medium withoutiron, use iron-free yeast nitrogen base.Galactose and glucose stock solutions Galactose and glucose stock solutionsof 20% (wt/vol) in ddH2O, dissolve and autoclave. m CRITICAL The galactosestock solution should be slightly warmed (40 1C) in case galactose does notdissolve or has formed a crust after cooling. For iron-free SC medium, confirmthat the iron concentration in the SC medium is below 0.1 mM by colorimetry65.Galactose from some suppliers is contaminated with iron.

YP medium (1 l) Dissolve 10 g of yeast extract and 20 g of casein peptone in900 ml of ddH2O (add 20 g of agar-agar for solid agar plates). After autoclaving,

add 100 ml of sterile 20% (wt/vol) galactose or glucose stock solution andauxotrophic markers appropriate for the yeast strains and plasmids used64.PMSF PMSF (200 mM) in 100% ethanol was used. m CRITICAL Prepare justbefore use, PMSF rapidly hydrolyses. ! CAUTION PMSF is toxic; wear suitableprotective clothing, gloves and eye/face protection.

Alkaline lysis mix Mix on the day of use 6.91 ml of ddH2O, 1.85 ml of 10 MNaOH (40% (wt/vol) in ddH2O) and 0.74 ml of 2-mercaptoethanol. Add 0.5 mlPMSF solution just before use. ! CAUTION The alkaline lysis mix is caustic,irritant and toxic; wear suitable protective clothing, gloves and eye/faceprotection. ! CAUTION PMSF is toxic; wear suitable protective clothing, glovesand eye/face protection.

TCA solution (50% wt/vol) Per 100 ml use 50 g of TCA.

TCA solution (30% wt/vol) Per 100 ml use 30 g of TCA. ! CAUTION TCAis corrosive, causes severe burns; wear suitable protective clothing, gloves andeye/face protection55FeCl3 with a specific radioactivity of 155–190 Gbq mg�1 (76–94 mCi mg�1)Aliquots of 1 mCi in 15–25 ml 0.5 M HCl are delivered by NEN/Perkin Elmerin individual containers, to which 0.1 M of HCl is added to obtain1 mCi ml�1 (B250 mM Fe). ! CAUTION Causes cancer, designate area forhandling 55Fe and clearly label all containers. Store mCi quantities of 55Febehind thin lead shielding. Wear disposable lab coats, wrist guards and gloves forsecondary protection. For more detailed information, see instructions ofoccupational limits in the NRC regulations (10 CFR) part 20, Standards forprotection against radiation (http://www.nrc.gov/reading-rm/doc-collections/cfr/part020/).Ascorbate solution (0.1 M) Dissolve 0.198 g in 10 ml of ddH2O on theday of use.Citrate buffer (50 mM citrate, 1 mM EDTA, pH 7.0) Dissolve 14.7 g oftrisodium citrate and 0.372 g of disodium EDTA in 800 ml of ddH2O, adjust thepH to 7.0 with 1 M HCl, bring up to a final volume of 1 l with ddH2O andautoclave.HEPES buffer (20 mM HEPES/KOH pH 7.4) Per 100 ml, use 4 ml 0.5 MHEPES adjusted to pH 7.4 with 2 M KOH, and autoclave.TNETG buffer (10 mM Tris/Cl pH 7.4, 2.5 mM EDTA, 150 mM NaCl,10% (vol/vol) glycerol, 0.5% (vol/vol) Triton X-100) Per 500 ml, use 5 mlof 1 M Tris-HCl pH 7.4, 2.5 ml of 0.5 M disodium EDTA (pH 7.4), 37.5 ml of2 M NaCl, 50 ml of glycerol, and autoclave. On the day of use, add Triton X-100to a final concentration of 0.5% (vol/vol).

Beads for immunoprecipitation All steps should be carried out at 4 1C.Beads for immunoprecipitation of TAP-tagged proteins: Suspend the com-mercial IgG Sepharose 6 Fast Flow beads by inversion, transfer 500 ml to a1.5-ml microfuge tube and centrifuge for 5 min at 800g at 4 1C to removethe supernatant. Wash three times with 500 ml of TNETG buffer and finallyadjust the volume to 500 ml with TNETG buffer.

Beads for immunoprecipitation with serum raised against a protein ofinterest: Swell 50 mg of Protein A-Sepharose CL-4B in 0.5 ml cold TNETGbuffer in a 1.5-ml microfuge tube by incubation for at least 30 min in a rotaryshaker. Centrifuge for 5 min at 800g at 4 1C, remove the supernatant and add500 ml of antibody serum. After incubation for at least 1 h in a rotary shaker,centrifuge for 5 min at 800g at 4 1C and remove the supernatant. Wash fivetimes with 500 ml of TNETG and centrifuge as before. Adjust the volume to500 ml with TNETG buffer.

Beads for immunoprecipitation with antibodies against hemagglutinin orMyc epitopes coupled to agarose: The 25% (wt/vol) commercialsuspension can directly be used after repeated inversion. m CRITICAL Resus-pend the beads by gentle swirling or inverting. Do not vortex, use amagnetic stirrer or spatula. Use blue or yellow pipette tips where 4 mmof the end have been cutoff with scissors, otherwise beads fracture anddistribute unevenly.

PROCEDURETransform yeast cells with plasmid � TIMING 5 h1| Grow yeast cells overnight in 50 ml of YP medium; on the next day, make competent yeast cells66, transform with 0.5–5 mgof plasmid of choice and plate onto SC agar containing 2% (wt/vol) galactose and the appropriate auxotrophy selectionmarkers64. Incubate at 30 1C until colonies of the transformants of 1–2 mm have grown, this usually takes 3–5 d.m CRITICAL STEP Also, transform cells with a control plasmid lacking the gene of interest.? TROUBLESHOOTING

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

760 | VOL.4 NO.5 | 2009 | NATURE PROTOCOLS

PROTOCOL

Grow yeast transformants on plate � TIMING 30 min2| Plate individual transformants from Step 1 onto a quadrant of an SC agar plate with 2% (wt/vol) galactose and appropriateauxotrophy selection markers64. Incubate at 30 1C until a rather continuous but not overgrown lawn has developed, whichusually takes 2–3 d.m CRITICAL STEP For stronger protein depletion, some galactose-regulatable mutants have to be downregulated by additionalgrowth on glucose-containing SC agar plates before growth in liquid medium. Cell densities should always remain lower than anOD of 2.

Confirm the presence of the protein of choice in the transformants (optional) � TIMING B8 h including SDS–PAGE andwestern blot3| Take a few milligrams of cells from each quadrant (Step 2) using a yellow tip, transfer the cells to a 1.7-ml microfuge tubefilled with 1 ml of ddH2O, centrifuge for 3 min at 13,000g, 4 1C. Add 75 ml of alkaline lysis mix and incubate for 10 min on ice.Add 575 ml of 50% (wt/vol) TCA, vortex and incubate for 10 min on ice and centrifuge for 3 min at 13,000g, 4 1C. Remove thesupernatant and wash the pellet twice with 1 ml of ice-cold acetone, dissolve in 50 ml sample buffer63 and perform western blotafter SDS–PAGE63.! CAUTION TCA is corrosive, causes severe burns, wear suitable protective clothing, gloves and eye/face protection.

Pre-culture of yeast cells � TIMING 15 min4| Inoculate 50 ml of SC medium supplemented with a final concentration of 2% (wt/vol) galactose (or 2% (wt/vol) glucose)and the appropriate auxotrophy selection markers64 with a small scoop of yeast cells from the plate (Step 2). Incubate at 30 1Cfor 24 h.m CRITICAL STEP Some galactose-regulatable mutants have to be downregulated by repeating this pre-culturing step in SC glucosemedium. Cell densities should always remain lower than an OD of 2 (see Step 5).? TROUBLESHOOTING

Culture of yeast cells in iron-free medium � TIMING B1 h5| Determine the OD of the pre-culture from Step 4 at 600 nm.

6| Remove an appropriate volume of the pre-culture from Step 4 for the inoculation of 100 ml of iron-free SC medium at anOD of 0.2 (i.e., 20 divided by the OD measured in Step 5 (in milliliters)).

7| Transfer the culture to 50-ml Falcon tubes and centrifuge for 5 min at 2,000g at room temperature. Remove the supernatantand add 10 ml of sterile ddH2O, vortex the cell pellet and repeat the centrifugation as before.

8| Prepare sterile 300 ml Erlenmeyer flasks with 100 ml of iron-free SC medium with 2% (wt/vol) galactose (or 2% (wt/vol)glucose) and the appropriate auxotrophy selection markers64. Use 10 ml of this medium to resuspend the pellets obtained inStep 7 and transfer the cells to the same flask. Grow the cells overnight (16 h) in a shaking incubator at 150 r.p.m., 30 1C.

Radiolabeling with 55FeCl3 � TIMING B3 h9| Harvest the cells from the 100 ml cultures into a single 50-ml Falcon tube by two successive centrifugations for 5 min at2,000g at room temperature.

10| Add 10 ml of ddH2O and resuspend the cell pellet, transfer to a 15-ml Falcon tube and centrifuge for 5 min at 1,500g atroom temperature.

11| Determine the wet weight of the cell pellet (0.5–0.75 g). Resuspend the cells in 10 ml of iron-free SC medium with 2%(wt/vol) galactose (or 2% (wt/vol) glucose).? TROUBLESHOOTING

12| Remove the appropriate volume that corresponds to 0.5 g of cells and place in a 15-ml Falcon tube. Adjust the volume to10 ml with iron-free SC medium with 2% (wt/vol) galactose (or 2% (wt/vol) glucose).

13| Incubate the cell suspension for 10 min in a shaking incubator at 150 r.p.m., 30 1C.

14| Mix 10 ml of 55FeCl3 solution (10 mCi) with 100 ml of 0.1 M sodium ascorbate and add to the cells. Incubate for 2 h at150 r.p.m., 30 1C.! CAUTION Causes cancer; designate area for handling 55Fe and clearly label all containers. Store mCi quantities of 55Fe behindthin lead shielding. Wear disposable lab coats, wrist guards and gloves for secondary protection. For more detailed information,

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

NATURE PROTOCOLS | VOL.4 NO.5 | 2009 | 761

PROTOCOL

see instructions of occupational limits in the NRC regulations (10 CFR) part 20, Standards for protection against radiation(http://www.nrc.gov/reading-rm/doc-collections/cfr/part020/).! CAUTION From this point onward, all liquid and solids have to be disposed of according to radiation safety regulations ordecontaminated and sterilized for repeated use by washing with citrate buffer, water and ethanol.

Cell extract preparation � TIMING B1 h15| Transfer the 55Fe-labeled cells to a 15-ml Falcon tube and harvest by centrifugation for 5 min at 2,000g at roomtemperature. Remove the supernatant and wash the cell pellet with 10 ml of citrate buffer to remove residual 55Fe from themedium and the outside of cells. Centrifuge for 5 min at 2,000g at room temperature, remove the supernatant and wash the cellpellet with 2 ml of 20 mM HEPES-KOH, pH 7.4 to remove the citrate buffer.

16| Centrifuge the cells for 5 min at 2,000g at room temperature, remove the supernatant by pipetting and determine the wetweight of the cells (typically 0.45–0.55 g).

17| Resuspend the cell pellet in a volume of TNETG buffer equal to the mass of the cell pellet and place the tubes on ice. Add10 ml of 0.2 M PMSF and 1 ml of glass beads. Close the lid tightly.

18| Invert the tube by flicking the contents to the lid side of the tube in one motion. Vortex at maximum speed with the lidside of the tubes contacting the rotating plate for 1 min, repeat a total of three times with intermittent 3 min coolingperiods on ice.m CRITICAL STEP Ensure that there are no glass beads trapped in the lid. All subsequent Steps (19–23) should be carried outat 4 1C.

19| Centrifuge the cells for 5 min at 2,000g at 4 1C to pellet the glass beads and unlysed cells. Transfer the supernatant to 1.5-mlmicrofuge tubes, centrifuge for 10 min at 13,000g at 4 1C and carefully transfer 450–500 ml of the supernatant to a fresh tube.m CRITICAL STEP Avoid the transfer of the lipid-containing surface layer.

20| Remove 5 ml of the extract for the determination of the cellular iron uptake: mix with 45 ml of ddH2O in a 1.5-ml microfugetube, add 1 ml of scintillation fluid, vortex for 30 s and determine the radioactivity by scintillation counting. For western blot-ting of Fe/S and non-Fe/S marker proteins, take a 25-ml sample of the extract and add 175 ml of ice-cold 30% (wt/vol) TCA,vortex and incubate on ice for 10 min.! CAUTION TCA is corrosive, causes severe burns, wear suitable protective clothing, gloves and eye/face protection

21| Centrifuge the TCA-treated extract from Step 20 for 10 min at 13,000g, 4 1C and remove the supernatant. Add 500 ml ofice-cold acetone, repeat the centrifugation step and wash again with 500 ml of ice-cold acetone. Remove the supernatant andair dry the protein pellet. Dissolve the pellet in 50 ml of sample buffer.63

’ PAUSE POINT The sample can be stored up to 1 month at –20 1C before SDS–PAGE and western blotting (see Step 25).

Immunoprecipitation � TIMING B1.5–2 h22| Add either 20–50 ml of self-made antibody Protein A-Sepharose beads, 20–40 ml of commercially available IgG Sepharosebeads, or 10–15 ml of anti-hemagglutinin or anti-Myc (A-14) beads to 200–250 ml of cleared cell extract (from Step 19) andincubate the mixtures in 1.5-ml microfuge tubes for 1 h on a rotary shaker at 4 1C.

23| Collect the beads by centrifuging for 5 min at 800g at 4 1C and carefully remove the supernatant by repeated pipettingwith a 200 ml pipette. Wash the beads three times in 500 ml of ice-cold TNETG and collect the beads by centrifugation.m CRITICAL STEP The first two removals of supernatant are critical for a successful experiment (i.e., low background). Use a blackbackground to improve visibility of the beads during pipetting.

24| Add 50 ml of ddH2O and 1 ml of scintillation cocktail to the beads, vortex for 30 s, place the microfuge tube in a plasticcounting vial and measure the 55Fe radioactivity associated with the beads in a scintillation counter with settings appropriatefor 3H (ref. 67).? TROUBLESHOOTING

Protein analysis by western blotting � TIMING 1 d25| Analyze protein extracts from Step 21 by SDS–PAGE and western blotting63.

� TIMINGStep 1, transform yeast cells with plasmid: 5 h plus 3–5 d for cell growthStep 2, grow yeast transformants on plate: 30 min plus 2–3 d for cell growth

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

762 | VOL.4 NO.5 | 2009 | NATURE PROTOCOLS

PROTOCOL

Step 3, confirm the presence of protein of choice in transformants (optional): B8 h including SDS–PAGE and western blotStep 4, pre-culture of yeast cells: 15 min plus 24 h for cell growthSteps 5–8, culture of yeast cells in iron-free medium: B1 h plus 16 h for cell growthSteps 9–14, radiolabeling with 55FeCl3: B3 hSteps 15–21, cell extract preparation: B1 hSteps 22–24, immunoprecipitation: B1.5–2 hStep 25, protein analysis by western blotting: 1 dBox 1, culture of cells: 1–10 d (depends on the availability of pre-culture and necessity of transformation); cell extractpreparation: B45 min; protein determination: 30 min; activity measurements: 5 min for aconitase, 5 min for isopropylmalateisomerase, 45 min for sulfite reductaseBox 2, culture of cells: 1–10 d (depends on the availability of pre-culture and necessity of transformation); preparation ofmitochondria: B6 h; protein determination: 30 min; activity measurements: 5 min for aconitase and 5 min for succinatedehydrogenase

? TROUBLESHOOTINGTroubleshooting advice can be found in Table 2.

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

TABLE 2 | Troubleshooting table.

Step Problem Possible reason Solution

1 No transformants Wrong auxotrophic marker(s) Check medium, yeast genotype and plasmid

Toxic plasmid Use a weaker promotor

Few or too many transformants Plasmid quantity inappropriate Vary the length of heat shock in transformationprotocol and modify the plasmid quantity

4 Incomplete protein depletion Insufficient downregulationon glucose

Screen downregulation: try depletion on glucose agarplate and/or vary culture time in liquid medium (16, 40and 64 h) until target protein cannot be detected onwestern blot

11 Insufficient cell mass Slow growth Grow a larger volume of pre-culture, use a fresher pre-culture, inoculate with slightly higher OD

Too high expression level Use moderate expression level by use of low-copyplasmid and/or weaker promoter

Tag interferes with biologicalfunction

Change tag or tag position (N-/C-terminal), useuntagged protein and antibodies against the protein

24 Low radioactivity in theimmunoprecipitate

Weak binding of proteins to beads Try other tag, change tag to N- or C-terminal position,raise new antibodies, buy fresh beads, check binding ofantibody to Protein A

Oxygen lability of Fe/S Perform cell extract preparation and further steps in ananaerobic chamber; all buffers should be anaerobic

Protein is degraded Add protease inhibitors. PMSF should be preparedfreshly. If the tag is cleaved off, try another tag,change to N- or C-terminal position or co-synthesizeinteraction partner that might stabilize the target12

Cells lysed inefficiently Determine protein content of whole-cell extracts(should be 3–6 mg/ml), optimize glass bead/vortexprocedure with nonradioactive cells

(continued)

NATURE PROTOCOLS | VOL.4 NO.5 | 2009 | 763

PROTOCOL

ANTICIPATED RESULTS55Fe incorporation into Fe/S proteins in yeast is a powerful method to determine (1) the presence of an Fe/S cluster in theanalyzed protein in vivo, (2) new members of the ISC assembly, ISC export or CIA machineries, (3) the target specificity ofISC and CIA components by the effect of their downregulation on various target Fe/S proteins, (4) the presence of Fe/Sclusters in scaffold and other proteins of the ISC and CIA machineries, (5) the order of in vivo events during de novo assembly ofFe/S clusters by downregulation of the ISC and CIA components. Using this approach, we previously were able to define thescaffold function of Isu1 in vivo and distinguish early and late components of the ISC assembly machinery37. Further, weidentified the P-loop NTPases Cfd1 and Nbp35 as cytosolic scaffold proteins and thus were able to stage the requirement of the

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

TABLE 2 | Troubleshooting table (continued).

Step Problem Possible reason Solution

Protein not expressed Check several transformants by western blot, change tohigh-copy plasmid and/or use stronger promoter

Iron contamination of medium Check iron content of medium components, changesupplier

Target is not an Fe/S protein Confirm that the protocol functions for a known Fe/Sprotein (Table 3)

High or variable background Inefficient washing of beads Remove all liquid, especially in the first wash step

Centrifuge cell extracts for longer, remove supernatantmore carefully. If yeast cells loose viability, the back-ground may vary. Try shorter downregulation times

Triton or chelator concentrationtoo low

Try other detergents, use EDTA or citrate

TABLE 3 | Summary of expected results for yeast Fe/S proteins studied by 55Fe/S cluster incorporation and enzymatic activity.

Protein Gene

55Fe (pmolper g cells)a

Activity (U per mgof protein)a Biological function Reference

Cytosolic and nuclear Fe/S proteinsIsopropylmalate isomerase Leu1 15–23 0.1–0.3 Biosynthesis of leucine 14,55Sulfite reductase Ecm17 NDb 0.0035 Biosynthesis of methionine

and cysteine14

ABC protein Rli1 Rli1 27–35 ND Biogenesis of ribosomes,translation initiation

29,55

P-loop NTPase Nbp35 Nbp35 11–13 ND Maturation of cytosolic andnuclear Fe/S proteins

12,55

P-loop NTPase Cfd1 Cfd1 8–12 ND Maturation of cytosolic andnuclear Fe/S proteins

12

Hydrogenase-like protein Nar1 6–10 ND Maturation of cytosolic andnuclear Fe/S proteins

12,13

DNA glycosylase Ntg2 9–11 ND DNA repair 55

Mitochondrial Fe/S proteinsAconitase Aco1 13 0.5c Citric acid cycle 7

2.4–3.5d 10,55Homoaconitase Lys4 12–21 ND Biosynthesis of lysine 7Glutamate dehydrogenase Glt1 ND 0.021–0.071e Biosynthesis of glutamate 70Lipoate synthase Lip5 10 ND Biosynthesis of lipoate 7Biotin synthase Bio2 80–300 ND Biosynthesis of biotin 14,71Complex II Sdh2 ND 0.15–0.3f Succinate dehydrogenase 10,55

Sdh2-Rip1 0.15–0.3g 10Ferredoxin Yah1 6–11 ND Maturation of Fe/S proteins,

heme A synthesis37,51,72

Scaffold protein Isu1 Isu1 13 ND Maturation of Fe/S proteins 37aValues refer to yeast cells grown in SC medium supplemented with 2% (wt/vol) galactose. Here 1.07 � 103 c.p.m. correspond to 1 pmol 55Fe under the conditions used in the procedure. All radiolabeling data referto overproduction of the respective Fe/S proteins from plasmids except for Leu1 and Aco1. Enzyme activities correspond to wild-type levels. bNot determined. cSpecific activity in whole-cell extract with the coupledassay (Box 1). dSpecific activity in mitochondria with the direct assay (Box 2). eValues refer to yeast cells grown in medium supplemented with 2% (wt/vol) glucose and various nitrogen sources. fSpecific activityexpressed as DCPIP reduction (Box 2). gSpecific activity expressed as cytochrome c reduction (Box 2), which requires functional complex III.

764 | VOL.4 NO.5 | 2009 | NATURE PROTOCOLS

PROTOCOL

CIA proteins in the pathway of cytosolic Fe/S cluster assembly12. Typical results for 55Fe incorporation into reporter Fe/Sproteins in wild-type cells are given in Table 3. Except for Aco1 and Leu1, the Fe/S proteins were overproduced from expressionplasmids. Likewise, Table 3 provides representative specific enzyme activities in galactose-grown wild-type cells. Appropriatenon-Fe/S enzymes should be measured in parallel as controls10,19,68.

Figure 3 shows typical results for an in vivo 55Fe incorporation experiment in yeast. In Figure 3a, the mitochondrialcysteine desulfurase Nfs1 was synthesized or depleted in Gal-NFS1 cells by growth in galactose- or glucose-containing media,respectively. These cells carry the NFS1 gene under the control of the GAL1-10 promoter, which is induced by galactose andrepressed by glucose. The incorporation of 55Fe into the scaffold Cfd1 (carrying a TAP tag) and the cytosolic Fe/S target proteinLeu1 was determined as described above. Incorporation of 55Fe into both Cfd1 and Leu1 strongly decreased on depletion of Nfs1verifying that the associated 55Fe is part of an Fe/S cluster. Detection of 55Fe on Cfd1 depended on its overproduction, whereaswild-type levels of Leu1 were sufficient for the measurement of 55Fe/S cluster assembly. The background levels of 55Fe assay canbe determined, e.g., by omitting production of Cfd1 (Fig. 3a) or by the use of pre-immune serum instead of specificantiserum. Typically, the background values vary between 2 and 15% of the total signal depending on the Fe/S protein andantiserum. In Figure 3b, we used the yeast mutant Gal-CFD1 with the CFD1 gene under the control of the GAL1-10 promoter todeplete Cfd1 (see western blot in Fig. 3b). After 55Fe radiolabeling, cell extracts were prepared as described under Protocols.Cytosolic (Leu1) and mitochondrial (Bio2) Fe/S proteins were immunoprecipitated with specific antibodies coupled to ProteinA-Sepharose beads, whereas the HA-tagged Fe/S proteins Rli1 (cytosol/nucleus) and Ntg2 (nucleus) were immunoprecipitatedusing commercially available HA beads. The latter three Fe/S proteins require overproduction from appropriate plasmids due totheir low cellular abundance. The radioactivity associated with mitochondrial Bio2 remained almost unchanged on depletion ofcytosolic Cfd1 (Fig. 3b). In contrast, the nuclear and cytosolic Fe/S proteins had only 15–25% of the radioactivity associatedcompared with galactose-grown cells. These data, in conjunction with western blotting experiments assessing the presence orabsence of the relevant proteins, show the specific function of Cfd1 for Fe/S cluster assembly on extra-mitochondrial proteins.

ACKNOWLEDGMENTS We acknowledge generous support from DeutscheForschungsgemeinschaft (SFB 593 and TR1, Gottfried-Wilhelm Leibniz program, andGRK 1216), the German–Israeli Foundation (GIF), Rhon Klinikum AG, von Behring-Rontgen Stiftung, Max-Planck Gesellschaft, Alexander-von-Humboldt Stiftung andFonds der chemischen Industrie.

Published online at http://www.natureprotocols.com/Reprints and permissions information is available online at http://npg.nature.com/reprintsandpermissions/

1. Beinert, H., Holm, R.H. & Munck, E. Iron–sulfur clusters: nature’s modular,multipurpose structures. Science 277, 653–659 (1997).

2. Johnson, D.C., Dean, D.R., Smith, A.D. & Johnson, M.K. Structure, function andformation of biological iron–sulfur clusters. Ann. Rev. Biochem. 74, 247–281(2005).

3. Lill, R. & Muhlenhoff, U. Maturation of iron–sulfur proteins in eukaryotes:mechanisms, connected processes, and diseases. Annu. Rev. Biochem. 77,669–700 (2008).

4. Ayala-Castro, C., Saini, A. & Outten, F.W. Fe–S cluster assembly pathways inbacteria. Microbiol. Mol. Biol. Rev. 72, 110–125 (2008).

5. Lill, R. & Muhlenhoff, U. Iron–sulfur protein biogenesis in eukaryotes:components and mechanisms. Annu. Rev. Cell Dev. Biol. 22, 457–486 (2006).

6. Vickery, L.E. & Cupp-Vickery, J.R. Molecular chaperones HscA/Ssq1 andHscB/Jac1 and their roles in iron–sulfur protein maturation. Crit. Rev. Biochem.Mol. Biol. 42, 95–111 (2007).

7. Gelling, C., Dawes, I.W., Richhardt, N., Lill, R. & Muhlenhoff, U. MitochondrialIba57p is required for Fe/S cluster formation on aconitase and activation ofradical SAM enzymes. Mol. Cell Biol. 28, 1851–1861 (2008).

8. Amutha, B. et al. GTP is required for iron–sulfur cluster biogenesis inmitochondria. J. Biol. Chem. 283, 1362–1371 (2008).

9. Lill, R. et al. Mechanisms of iron–sulfur protein maturation in mitochondria,cytosol and nucleus of eukaryotes. Biochim. Biophys. Acta 1763, 652–667(2006).

10. Kispal, G., Csere, P., Prohl, C. & Lill, R. The mitochondrial proteins Atm1p andNfs1p are required for biogenesis of cytosolic Fe/S proteins. EMBO J. 18,3981–3989 (1999).

11. Balk, J. & Lill, R. The cell’s cookbook for iron–sulfur clusters: recipes for fool’sgold? Chem. Biol. Chem. 5, 1044–1049 (2004).

12. Netz, D.J., Pierik, A.J., Stumpfig, M., Muhlenhoff, U. & Lill, R. The Cfd1-Nbp35complex acts as a scaffold for iron–sulfur protein assembly in the yeast cytosol.Nat. Chem. Biol. 3, 278–286 (2007).

13. Balk, J., Pierik, A.J., Aguilar Netz, D., Muhlenhoff, U. & Lill, R. The hydrogenase-like Nar1p is essential for maturation of cytosolic and nuclear iron–sulphurproteins. EMBO J. 23, 2105–2115 (2004).

14. Balk, J., Aguilar Netz, D.J., Tepper, K., Pierik, A.J. & Lill, R. The essential WD40protein Cia1 is involved in a late step of cytosolic and nuclear iron–sulfur proteinassembly. Mol. Cell Biol. 25, 10833–10841 (2005).

15. Zhang, Y. et al. Dre2, a conserved eukaryotic Fe/S cluster protein,functions in cytosolic Fe/S protein biogenesis. Mol. Cell Biol. 28, 5569–5582(2008).

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

α-Cfd1α-Nfs1

0

4

8

12a

55F

e in

corp

orat

ion

into

Cfd

1 (1

03 c.

p.m

. per

g c

ells

)

0

5

10

15Leu1

Cfd1

Cfd1-TAP

α-Leu1

+ + + –

16 h 40 h40 hGal

Gal-NFS1

40 hGlc

55F

e in

corp

orat

ion

into

Leu1

(10

3 c.

p.m

. per

g c

ells

)

0

25

50

75

100b

55F

e in

corp

orat

ion

(%)

(Glc

/Gal

)

α-Bio2 α-Leu1 α-HAα-Cfd1

Bio2 Rli1-HA

Gal Glc

Ntg2-HA

α-HA

Gal-CFD1Gal

Glc

Gal Glc

Gal Glc

Leu1

Figure 3 | Incorporation of 55Fe into yeast Fe/S proteins in vivo. Gal-NFS1

(a) and Gal-CFD1 (b) cells were grown in galactose (Gal) or glucose (Glc) to

induce or repress the synthesis of Nfs1 and Cfd1, respectively. Cells were

grown for the indicated time periods (a) or 40 h (b), radiolabeled for 2 h with55Fe according to the Protocol and cell extracts were prepared. An aliquot of

the extract was immediately TCA-precipitated for detection of the proteins by

western blotting (lower panels). The remainder (B250 ml) was added to IgG

(Cfd1-TAP), anti-HA antibodies (Rli1-HA and Ntg2-HA) or home-made

antibodies (against yeast Leu1 and Bio2) coupled to beads. After 1 h of

incubation, the beads were extensively washed, and the amount of 55Fe

associated with the beads was determined by scintillation counting (upper

panels). Reproduced from ref. 12 previously published in Nature Chemical

Biology.

NATURE PROTOCOLS | VOL.4 NO.5 | 2009 | 765

PROTOCOL

16. Lill, R. & Muhlenhoff, U. Iron–sulfur protein biogenesis in eukaryotes. TrendsBiochem. Sci. 30, 133–141 (2005).

17. Stehling, O., Elsasser, H.P., Bruckel, B., Muhlenhoff, U. & Lill, R. Iron–sulfurprotein maturation in human cells: evidence for a function of frataxin. Hum. Mol.Genet. 13, 3007–3015 (2004).

18. Wingert, R.A. et al. Deficiency of glutaredoxin 5 reveals Fe–S clusters are requiredfor vertebrate haem synthesis. Nature 436, 1035–1039 (2005).

19. Biederbick, A. et al. Role of human mitochondrial Nfs1 in cytosolic iron–sulfurprotein biogenesis and iron regulation. Mol. Cell Biol. 26, 5675–5687 (2006).

20. Pondarre, C. et al. The mitochondrial ATP-binding cassette transporter Abcb7 isessential in mice and participates in cytosolic iron–sulphur cluster biogenesis.Hum. Mol. Genet. 15, 953–964 (2006).

21. Tong, W.H. & Rouault, T.A. Functions of mitochondrial ISCU and cytosolic ISCU inmammalian iron–sulfur cluster biogenesis and iron homeostasis. Cell Metab. 3,199–210 (2006).

22. Song, D. & Lee, F.S. A role for IOP1 in mammalian cytosolic iron–sulfur proteinbiogenesis. J. Biol. Chem. 283, 9231–9238 (2008).

23. Stehling, O. et al. Human Nbp35 is essential for both cytosolic iron–sulfur proteinassembly and iron homeostasis. Mol. Cell Biol. 28, 5517–5528 (2008).

24. Rouault, T.A. & Tong, W.H. Iron–sulfur cluster biogenesis and human disease.Trends Genet. 24, 398–407 (2008).

25. Campuzano, V. et al. Friedreich’s ataxia: autosomal recessive disease caused by anintronic GAA triplet repeat expansion. Science 271, 1423–1427 (1996).

26. Camaschella, C. et al. The human counterpart of zebrafish shiraz showssideroblastic-like microcytic anemia and iron overload. Blood 110, 1353–1358(2007).

27. Mochel, F. et al. Splice mutation in the iron–sulfur cluster scaffold protein ISCUcauses myopathy with exercise intolerance. Am. J. Hum. Genet. 82, 652–660(2008).

28. Olsson, A., Lind, L., Thornell, L.E. & Holmberg, M. Myopathy with lactic acidosis islinked to chromosome 12q23.3–24.11 and caused by an intron mutation in theISCU gene resulting in a splicing defect. Hum. Mol. Genet. 17, 1666–1672 (2008).

29. Kispal, G. et al. Biogenesis of cytosolic ribosomes requires the essential iron–sulphur protein Rli1p and mitochondria. EMBO J. 24, 589–598 (2005).

30. Klinge, S., Hirst, J., Maman, J.D., Krude, T. & Pellegrini, L. An iron–sulfur domainof the eukaryotic primase is essential for RNA primer synthesis. Nat. Struct. Mol.Biol. 14, 875–877 (2007).

31. Rudolf, J., Makrantoni, V., Ingledew, W.J., Stark, M.J. & White, M.F. The DNArepair helicases XPD and FancJ have essential iron–sulfur domains. Mol. Cell 23,801–808 (2006).

32. Yarunin, A., Panse, V., Petfalski, E., Tollervey, D. & Hurt, E. Functional linkbetween ribosome formation and biogenesis of iron–sulfur proteins. EMBO J. 24,580–588 (2005).

33. Sung, P. et al. Human xeroderma pigmentosum group D gene encodes a DNAhelicase. Nature 365, 852–855 (1993).

34. Fan, L. et al. XPD helicase structures and activities: insights into the cancer andaging phenotypes from XPD mutations. Cell 133, 789–800 (2008).

35. Petillot, Y., Forest, E., Mathieu, I., Meyer, J. & Moulis, J.-M. Analysis, byelectrospray ionization mass spectrometry, of several forms of Clostridiumpasteurianum rubredoxin. Biochem. J. 296, 657–661 (1993).

36. Ramelot, T.A. et al. Solution NMR structure of the iron–sulfur cluster assemblyprotein U (IscU) with zinc bound at the active site. J. Mol. Biol. 344, 567–583(2004).

37. Muhlenhoff, U., Gerber, J., Richhardt, N. & Lill, R. Components involved inassembly and dislocation of iron–sulfur clusters on the scaffold protein Isu1p.EMBO J. 22, 4815–4825 (2003).

38. Molik, S., Lill, R. & Muhlenhoff, U. Methods for studying iron metabolism in yeastmitochondria. Methods Cell Biol. 80, 261–280 (2007).

39. Stehling, O. et al. Investigation of iron-sulfur protein maturation in eukaryotes.In Methods in Molecular Biology (eds. Leister, D. & Herrmann, J.M.) 325–342(Humana Press Inc., Totowa, NJ, 2007).

40. Lutz, T., Westermann, B., Neupert, W. & Herrmann, J.M. The mitochondrialproteins Ssq1 and Jac1 are required for the assembly of iron sulfur clusters inmitochondria. J. Mol. Biol. 307, 815–825 (2001).

41. Dos Santos, P.C., Johnson, D.C., Ragle, B.E., Unciuleac, M.-C. & Dean, D.R.Controlled expression of nif and isc iron–sulfur protein maturation componentsreveals target specificity and limited functional replacement between the twosystems. J. Bacteriol. 189, 2854–2862 (2007).

42. Raulfs, E.C., O’Carroll, I.P., Dos Santos, P.C., Unciuleac, M.C. & Dean, D.R.In vivo iron–sulfur cluster formation. Proc. Natl. Acad. Sci. USA 105, 8591–8596(2008).

43. Haile, D.J. et al. Cellular regulation of the iron-responsive element bindingprotein: disassembly of the cubane iron–sulfur cluster results in high-affinity RNAbinding. Proc. Natl. Acad. Sci. USA 89, 11735–11739 (1992).

44. Strain, J. et al. Suppressors of superoxide dismutase (SOD1) deficiency inSaccharomyces cerevisiae. Identification of proteins predicted to mediate iron–sulfur cluster assembly. J. Biol. Chem. 273, 31138–31144 (1998).

45. Rabinowitz, J.C. Analysis of acid-labile sulfide and sulfhydryl groups. MethodsEnzymol. 53, 275–277 (1978).

46. Thomas, D., Barbey, R., Henry, D. & Surdin-Kerjan, Y. Physiological analysis ofmutants of Saccharomyces cerevisiae impaired in sulphate assimilation. J. Gen.Microbiol. 138, 2021–2028 (1992).

47. De Winde, J.H. & Grivell, L.A. Global regulation of mitochondrial biogenesisin Saccharomyces cerevisiae. Prog. Nucleic Acid Res. Mol. Biol. 46, 51–91(1993).

48. Mendel, R.R., Smith, A.G., Marquet, A. & Warren, M.J. Metal and cofactorinsertion. Nat. Prod. Rep. 24, 963–971 (2007).

49. Hirata, A., Klein, B.J. & Murakami, K.S. The X-ray crystal structure of RNApolymerase from Archaea. Nature 451, 851–854 (2008).

50. Roy, A., Solodovnikova, N., Nicholson, T., Antholine, W. & Walden, W.E. A noveleukaryotic factor for cytosolic Fe–S cluster assembly. EMBO J. 22, 4826–4835(2003).

51. Lange, H., Kaut, A., Kispal, G. & Lill, R. A mitochondrial ferredoxin is essentialfor biogenesis of cellular iron–sulfur proteins. Proc. Natl. Acad. Sci. USA 97,1050–1055 (2000).

52. Muhlenhoff, U., Richhardt, N., Gerber, J. & Lill, R. Characterization of iron–sulfurprotein assembly in isolated mitochondria: a requirement for ATP, NADH andreduced iron. J. Biol. Chem. 277, 29810–29816 (2002).

53. Van Ammel, R., Pomme, S. & Sibbens, G. Half-life measurement of 55Fe. Appl.Radiat. Isot. 64, 1412–1416 (2006).

54. Ghaemmaghami, S. et al. Global analysis of protein expression in yeast. Nature425, 737–741 (2003).

55. Hausmann, A. et al. The eukaryotic P-loop NTPase Nbp35: an essential componentof the cytosolic and nuclear iron–sulfur protein assembly machinery. Proc. Natl.Acad. Sci. USA 102, 3266–3271 (2005).

56. Funk, M. et al. Vector systems for heterologous expression of proteins inSaccharomyces cerevisiae. Methods Enzymol. 350, 248–257 (2002).

57. Lange, H., Kispal, G. & Lill, R. Mechanism of iron transport to the site ofheme synthesis inside yeast mitochondria. J. Biol. Chem. 274, 18989–18996(1999).

58. Janke, C. et al. A versatile toolbox for PCR-based tagging of yeast genes: newfluorescent proteins, more markers and promoter substitution cassettes. Yeast 21,947–962 (2004).

59. Belli, G., Gari, E., Piedrafita, L., Aldea, M. & Herrero, E. An activator/repressordual system allows tight tetracycline-regulated gene expression in budding yeast.Nucleic Acids Res. 26, 942–947 (1998).

60. Kanemaki, M., Sanchez-Diaz, A., Gambus, A. & Labib, K. Functional proteomicidentification of DNA replication proteins by induced proteolysis in vivo. Nature423, 720–724 (2003).

61. Ben-Aroya, S. et al. Toward a comprehensive temperature-sensitive mutantrepository of the essential genes of Saccharomyces cerevisiae. Mol. Cell 30,248–258 (2008).

62. Schuldiner, M. et al. Exploration of the function and organization of the yeastearly secretory pathway through an epistatic miniarray profile. Cell 123, 507–519(2005).

63. Siu, F.K., Lee, L.T. & Chow, B.K. Southwestern blotting in investigatingtranscriptional regulation. Nat. Protoc. 3, 51–58 (2008).

64. Sherman, F. Getting started with yeast. Methods Enzymol. 350, 3–41 (2002).65. Pierik, A.J., Wolbert, R.B.G., Mutsaers, P.H.A., Hagen, W.R. & Veeger, C.

Purification and biochemical characterization of a putative [6Fe–6S] prismane-cluster-containing protein from Desulfovibrio vulgaris (Hildenborough). Eur. J.Biochem. 206, 697–704 (1992).

66. Gietz, R.D. & Woods, R.A. Transformation of yeast by lithium acetate/single-stranded carrier DNA/polyethylene glycol method. Methods Enzymol. 350, 87–96(2002).

67. Gibson, J.A. & Marshall, M. The counting efficiency for 55Fe and other E.C.nuclides in liquid scintillator solutions. Int. J. Appl. Radiat. Isot. 23, 321–328(1972).

68. Muhlenhoff, U., Richhardt, N., Ristow, M., Kispal, G. & Lill, R. The yeast frataxinhomologue Yfh1p plays a specific role in the maturation of cellular Fe/S proteins.Hum. Mol. Genet. 11, 2025–2036 (2002).

69. Wallace, M.A. et al. Superoxide inhibits 4Fe–4S cluster enzymes involved in aminoacid biosynthesis. Cross-compartment protection by CuZn-superoxide dismutase.J. Biol. Chem. 279, 32055–32062 (2004).

70. Valenzuela, L., Ballario, P., Aranda, C., Filetici, P. & Gonzalez, A. Regulation ofexpression of GLT1, the gene encoding glutamate synthase in Saccharomycescerevisiae. J. Bacteriol. 180, 3533–3540 (1998).

71. Pelzer, W. et al. Mitochondrial Isa2p plays a crucial role in the maturation ofcellular iron–sulfur proteins. FEBS Lett. 476, 134–139 (2000).

72. Barros, M.H., Nobrega, F.G. & Tzagoloff, A. Mitochondrial ferredoxin is required forheme A synthesis in Saccharomyces cerevisiae. J. Biol. Chem. 277, 9997–10002(2002).

73. Bradford, M.M. A rapid and sensitive method for the quantification of microgramquantities of protein utilizing the principle of protein dye binding. Anal.Biochem. 72, 248–254 (1976).

74. Diekert, K., deKroon, A.I.P.M., Kispal, G. & Lill, R. Isolation and sub-fractionationof mitochondria from the yeast Saccharomyces cerevisiae. Methods Cell Biol. 65,37–51 (2001).

p

uor

G g

n ih si l

bu

P eru ta

N 900 2©

nat

ure

pro

toco

ls/

moc.er

ut an.

ww

w//:ptt

h

766 | VOL.4 NO.5 | 2009 | NATURE PROTOCOLS

PROTOCOL