Identifying the protein interactions of the cytosolic iron sulfur cluster

May 25, 2017 - ... Melissa D. Marquez , Eric J. Camire , Stephanie U. Esonwune , John D. Grossman , Kelly A. Gay , Jessica A. Cosman , Deborah L. Perl...
0 downloads 0 Views 965KB Size
Subscriber access provided by Binghamton University | Libraries

Article

Identifying the protein interactions of the cytosolic iron sulfur cluster targeting complex essential for its assembly and recognition of apo-targets Amanda Thuy Van Vo, Nicholas M. Fleischman, Mary J. Froehlich, Claudia Y. Lee, Jessica A. Cosman, Calina A. Glynn, Zanub O. Hassan, and Deborah L. Perlstein Biochemistry, Just Accepted Manuscript • Publication Date (Web): 25 May 2017 Downloaded from http://pubs.acs.org on May 30, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Identifying the protein interactions of the cytosolic iron sulfur cluster targeting complex essential for its assembly and recognition of apo-targets Amanda Vo, Nicholas M. Fleischman†, Mary J. Froehlich†, Claudia Y. Lee, Jessica A. Cosman**, Calina A. Glynn†, Zanub O. Hassan† and Deborah L. Perlstein* Department of Chemistry, Boston University, Boston, MA 02215

ABSTRACT: The cytosolic iron sulfur cluster assembly (CIA) system assembles iron sulfur (FeS) cluster cofactors and inserts them into >20 apo-protein targets residing in the cytosol and nucleus. Three CIA proteins, called Cia1, Cia2, and Met18 in yeast, form the targeting complex responsible for apo-target recognition. There is little information about the structure of this complex or its mechanism of CIA substrate recognition. Herein, we exploit affinity copurification and size exclusion chromatography to determine the subunit connectivity and stoichiometry of the CIA targeting complex. We conclude that Cia2 is the organizing center of the targeting complex, which contains one Met18, two Cia1, and four Cia2 polypeptides. To probe target recognition specificity, we utilize the CIA substrates Leu1 and Rad3 as well as the E. coli FeS-binding transcription factor FNR (fumerate nitrate reductase). We demonstrate that

1

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

both of the yeast CIA substrates are recognized, whereas the bacterial protein is not. Thus, while the targeting complex exhibits flexible target recognition in vitro, it cannot promiscuously recognize any FeS protein. Additionally, we demonstrate that the full CIA targeting complex is required to stably bind Leu1 in vitro whereas the Met18-Cia2 subcomplex is sufficient to recognize Rad3. Together these results allow us to propose a unifying model for the architecture of this highly conserved complex and demonstrate what component or subcomplexes are vital for target identification.

Iron sulfur (FeS) clusters are versatile and ubiquitous cofactors required for numerous fundamental biochemical transformations.1 The assembly and insertion of these metalloclusters requires essential biosynthetic pathways in which clusters are first assembled from their molecular building blocks and subsequently inserted into apo-targets, proteins that require FeS cofactors for their activity. Since targets differ in their sequences and structures, cluster biosynthesis machineries require a mechanism to provide specificity so that their apo-proteins can be identified.2, 3 However, the molecular details of target recognition are not well understood for any cluster biogenesis system. The cytosolic iron sulfur cluster assembly (CIA) system supplies FeS clusters to extramitochondrial targets, including nuclear enzymes essential for DNA replication and repair.2 Recent work from several laboratories has identified a multiprotein complex, termed the CIA targeting complex, which is proposed to perform CIA target recognition and FeS cluster insertion in the final step of the pathway.4-6 In yeast, this complex comprises Cia1, Cia2 and Met18.4, 5, 7, 8 Several high throughput proteomic studies have identified interactions between targeting complex components and apo-targets.6, 9-14 But the role(s) of the targeting complex subunits and the mechanism by which the complex identifies targets have not yet been delineated.

2

ACS Paragon Plus Environment

Page 2 of 30

Page 3 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

To begin answering these important questions, two teams of researchers have investigated the protein-protein interactions of the human targeting complex via coimmunoprecipitation (coIP).15, 16

However, these studies reached conflicting conclusions about the architecture of the complex

and the roles of its component subunits. Wietmarschen et al. found that human Cia1 (CIAO1, herein referred to as hCia1) binds both human Cia2 (FAM96B or MIP18, herein referred to as hCia2b) and human Met18 (MMS19, herein referred to as hMet18), but they did not detect a hCia2-hMet18 complex.15 Additionally, they reported that all three components of the human targeting complex can directly interact with some targets. In contrast, Seki et al. reported that hCia2 is the organizing center of the targeting complex, tethering hMet18 to hCia1, and targets can directly interact with both hMet18 and hCia2 but not with hCia1.16 Additional studies are essential to resolve these discrepancies, and advance our understanding of CIA cluster targeting. Herein, we investigate the protein interactions required for formation of the yeast targeting complex, and for its recognition of apo-proteins. We demonstrate that Cia2 is the central scaffold of the targeting complex, linking Met18 to Cia1. By determining the molecular weight of targeting complex and its stable subcomplexes, we propose the targeting complex comprises one Met18, four Cia2, and two Cia1 polypeptides. We also demonstrate that the Met18-Cia2 subcomplex is sufficient for recognition of Rad3 whereas the full Met18-Cia1-Cia2 complex is required to interact with Leu1. Together this work allows us to develop a preliminary model for the structure of the targeting complex, and for the interactions that drive apo-protein identification in the final step of cytosolic FeS cluster biogenesis.

EXPERIMENTAL PROCEDURES

3

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

Met18 expression and purification.. N-terminally His-tagged Met18 (HisMet18) was created by amplifying Met18 from genomic DNA using primers AV03 and AV04 (Table S1) and inserting Met18 between the EcoR1 and SalI sites of the pRSF-Duet vector via the Gibson DNA assembly method, creating pAV02.17 His-SUMO Met18 (SUMOMet18) was created by insertion of the SUMO coding sequence at the EcoR1 site of pAV02 using primers AV26 and AV27 (Table S1) and Gibson assembly to create pAV13. Double-tagged (DTMet18) was created by the addition of a C-terminal TEV protease and Strep-II tag to pAV13 via Q5 Mutagenesis Kit (NEB) using primers ZH01 and ZH02 (Table S1). These constructs, as well as all other constructs reported herein, were confirmed via DNA sequencing. His

Met18 expressed in E. coli Rosetta2(DE3) grown at 37˚C to an OD600 of 0.7-0.8. IPTG (1

mM) was added and the cells were collected 4 h later.

SUMO

Met18 and

DT

Met18 were grown by

autoinduction.18 For purification, the cell paste was resuspended in Buffer A (50 mM Tris (pH 8.0), 100 mM NaCl, 10% glycerol and 5 mM β-mercapoethanol) supplemented with protease inhibitors and DNase nuclease. Cells were lysed by sonication and the soluble extract was added to TALON affinity resin (Clontech). The column was washed with 50 column volumes (CV) Buffer A supplemented with 5 mM imidazole and eluted with Buffer A supplemented with 300 mM imidazole. Met18 was dialyzed overnight against Buffer A, concentrated to 1.0 mg/mL, and stored at -80˚C. Purification of

DT

Met18 proceeded as described above except the TALON

elution was loaded onto a streptactin column which was washed with 10 CV of Buffer A and eluted with Buffer A supplemented with 2.5 mM desthibiotin. Untagged Met18 was accessed by cleavage of the SUMO tag from

SUMO

Met18. The SUMO-

tag was removed with the addition of SUMO protease (0.6 mg/mL). After an overnight

4

ACS Paragon Plus Environment

Page 5 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

incubation at 4 ̊C, the mixture was passed over a HisBind column and Met18 was recovered from the flow-through. Purification of Cia1 and Cia2. Purification of untagged Cia2 and Cia2 with an N-terminal Strep-tag and C-terminal His-tag (DTCia2, for double tagged Cia2) as well as removal of the Histag from DTCia2 via Tev protease were carried out as described.19 Purification of Leu1. Leu1 was amplified from genomic DNA by primers MP01 and MP02 (Table S1) and inserted between the NdeI and XhoI sites of a modified pET15b vector where the thrombin protease site was replaced by a TEV protease site.17 The resulting plasmid was used to express Leu1 with an N-terminal His-TEV tag. Leu1 was expressed as described above for His

Met18 except the cells were induced for 16h at 15˚C.

To purify Leu1, cell paste was resuspended and lysed as described for

His

Met18. The soluble

lysate was treated with streptomycin sulfate (1% w/v). The supernatant was then batch adsorbed to nickel affinity resin. The resin was washed with ≥100 CV Buffer A containing 5-30 mM imidazole and eluted with Buffer A supplemented with 300 mM imidazole. Leu1 was dialyzed and concentrated as described for

His

Met18. UV/Vis spectra of purified Leu1 and ferrozine iron

assays did not reveal a significant amount of iron associated with Leu1, suggesting it is purified predominantly in the apo-form. To remove the His-tag,

His

Leu1 was incubated with His-tagged

TEV protease then the mixture was passed over IMAC resin. Untagged Leu1 was recovered from the flow-through. Purification of Rad3. Rad3 was amplified from plasmid ScCD00012711 obtained from DNASU (the DNA repository at Arizona State University) with primers AV13 and AV14 (Table S1) and inserted between NdeI and XhoI sites of pRSF-Duet via Gibson ligation.17 Expression

5

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

for Rad3 was as described for

His

Met18 except the temperature was lowered to 25˚C after IPTG

addition. To refold Rad3, the cell paste was resuspended, lysed, and centrifuged as described for His

Met18. The pellet was washed with Buffer B (50 mM Tris (pH 8), 100 mM NaCl, 1 mM

EDTA, 1 mM DTT, and 5% glycerol) supplemented with 0.5% Triton X100 and subsequently with Buffer B. The pellet was resuspended with 50 mM Tris (pH 8), 200 mM NaCl, 2 mM EDTA, 7 M guanidinium•HCl, centrifuged, and passed through a 0.2µ filter. Solubilized inclusions (1 mL) were added drop-wise to 49 mL of 50 mM Tris (pH 7.5), 10 µM betamercaptoethanol, 800 mM arginine, and 100 mM KCl. The solution was incubated with gentle agitation for 2h, centrifuged, exchanged into Buffer B, concentrated, and stored at -80˚C. FNR purification. FNR was amplified from E. coli genomic DNA with primers JDG03 and JDG04 (Table S1) and inserted via Gibson assembly between the NdeI and BamHI sites of modified pET15b plasmid (see Leu1 cloning) to create pJDG04. The resulting His-TEV-FNR was subcloned between the NdeI and HindIII sites of pSNAP-tag(T7)-2 (New England Biolabs) by amplification with primers JDG16 and JDG17. This inserted His-TEV-FNR in frame with the C-terminal SNAP tag. Expression of FNR was achieved as described

His

Met18. The cell paste was resuspended in

Buffer C (50 mM potassium phosphate, 0.1 M KCl, 10% glycerol, pH 6.8) supplemented with 1mM PMSF, lysozyme (1mg/mL) and DNase nuclease. The cells were disrupted by sonication. The soluble lysate was batch absorbed to His-Bind resin. Resin was washed with Buffer C containing 5-20 mM imidazole over 40 CV, and eluted with Buffer C with 350 mM imidazole. FNR was concentrated, and dialyzed against Buffer C supplemented with 5mM DTT. Labeling

6

ACS Paragon Plus Environment

Page 6 of 30

Page 7 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

of FNR with a fluorescent SNAP substrate was carried out according to manufacturer’s instructions (NEB). Co-purification assay. All experiments were carried out at 4 ˚C. Bait-only and prey-only controls were performed in parallel as indicated. An affinity tagged bait protein (His or Strep tag; ~100-250 µg) was mixed with an equimolar amount of one or more prey proteins for 1h. The mixture was passed though 100-200 µL of affinity resin. Nickel IMAC columns were washed with ≥ 15 CV of Buffer A and eluted with Buffer A with 300 mM imidazole. Streptactin columns were washed with ≥15 CV of Buffer B and eluted with Buffer B supplemented with 2.5 mM desthiobiotin. Elution fractions were analyzed by SDS-PAGE. Proteins were identified by their migration in SDS-PAGE as compared to pure standards and by Western blotting for the His- or Strep-tags as required. Size Exclusion Chromatography. Each component (~10 µM) was mixed in a final volume of 0.5 mL and incubated at room temperature for 1 h. The sample was injected onto a Superdex 200 Increase 10/300 GL column (GE healthcare) and eluted with 20 mM Tris, 100 mM NaCl, and 5% (v/v) glycerol at a flow rate of 0.5 mL min-1. The molecular weight of each peak was determined by comparison of the elution volume to a standard curve generated with ferritin (443kDa), alcohol dehydrogenase (150 kDa) ovalbumin (66 kDa), albumin (45 kDa), and cytochrome C (12.5 kDa). Standards were analyzed immediately before experimental samples. At least three independent measurements of molecular weight were determined and reported as the average ± the standard deviation. Multi-angle light scattering. Size exclusion chromatography coupled with multi-angle light scattering (SEC-MALS) was performed using an Agilent AdvanceBio 300 column attached to a Dawn Heleos MALS instrument (Wyatt Technology) and an Optilab rEX detector (Wyatt

7

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Technology).

His

Cia1, Cia2 and Met18 were mixed and purified on IMAC before injection into

SEC-MALS. The column was eluted with a flow rate of 0.5 mL/min in 25 mM HEPES pH 7.5, 100 mM NaCl, and 1% glycerol. Molar-masses were calculated using the Zimm model with Astra7 software (Wyatt Technology). For analysis, a dn/dc value of 0.186 mL g-1 was used. Both peaks presented were monodisperse (Mw/ Mn < 1.01). The errors reported are fitting error generated by the Astra7 software. BMOE crosslinking. A 20 mM solution of bismaleimidoethane (BMOE) in DMSO was prepared. A 10-fold excess of crosslinker was added to each protein sample and the proteins were cross-linked in sulfhydryl free buffer (50 mM Tris pH 8, 100 mM Nacl, 5% glycerol, 5 mM TCEP). The BMOE was incubated with the protein for 15 minutes at room temperature and the reaction was quenched by the addition of DTT (10 mM final concentration). Circular Dichroism. Spectra were acquired on an Applied Photophysics CS/2 Chirascan. Spectra were acquired in a 1 mm path length quartz cuvette at 1.2 s/nm with a 1 nm spectral bandwidth. The samples (0.3-0.5 mg/mL protein) were prepared in 10 mM potassium phosphate (pH 8.0), 100 mM potassium chloride, and 0.5 mM DTT. Data was collected from 200-260 nm. Buffer interference below 200 nm prevented data collection in this region. For each protein sample, at least three spectra were collected, averaged, smoothed and background subtracted using the Chirascan software.

RESULTS Met18 Purification and Characterization. Biochemical and biophysical studies of the CIA targeting complex require milligram quantities of protein, but no methods are currently available to access Met18.4, 5 Therefore, we developed a method to purify recombinant Met18. Without

8

ACS Paragon Plus Environment

Page 8 of 30

Page 9 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

any optimization, the expression level of Met18 in the codon enriched E. coli Rosetta(DE3) was low leading to a low yield of ≤500 µg HisMet18 from 50 g wet cell paste. The yield was improved using an N-terminal SUMO tag along with autoinduction expression conditions (Figure 1A). The low molecular weight contaminants of

SUMO

Met18 could be removed by the use of a double

tagged construct with an N-terminal His-tag and C-terminal StrepII tag and tandem affinity purification (Figure 1A, Lane 3).

Figure 1. Purification of Met18. A) SDS-PAGE of HisMet18 (Lane 1), SUMOMet18 (Lane 2), and DT

Met18 (Lane 3). Migration of MW standards (in kDa, left) and Met18 are indicated. B) CD

spectrum of

SUMO

Met18 (2 µM) in 10 mM KPO4 and 100 mM NaCl buffer demonstrating the

protein’s alpha helical character. C) SEC analysis of

SUMO

Met18 (130 kDa) which elutes as two

overlapping peaks with corresponding to molecular weights of 558 and 298 kDa for Peaks A and B, respectively. Peak C (25 kDa) contains low molecular weight contaminants seen in Lane 2 of Panel A. D) A representative standard curve used to determine molecular weight from SEC. standards used were apo-ferritin (1, 443 kDa), alcohol dehydrogenase (2, 150 kDa), albumin (3, 66 kDa), ovalbumin (4, 45 kDa), carbonic anhydrase (5, 29kDa), and cytochrome c (6, 12.4kDa).

To ensure that the tags added did not impact the native fold of Met18, we characterized the recombinant protein via circular dichroism (CD) and size exclusion chromatography (SEC). Met18 is predicted to be an all alpha helical HEAT (huntingtin, elongation factor 3, subunit A of

9

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

protein phosphatase 2A, and target of rapamycin) repeat protein.20 Consistent with this domain structure,

SUMO

Met18 has a CD spectrum that one would expect for an alpha-helical protein

(Figure 1B). When analyzed via SEC,

SUMO

Met18 eluted as two overlapping peaks with

molecular weights of 558 kDa and 298 kDa (Figure 1C). Since the

SUMO

Met18 polypeptide is

130 kDa, our results are most consistent with a Met18 forming a mixture of dimers and tetramers. However, it is likely Met18 has an elongated, nonspherical shape similar to other HEAT-repeat proteins, it is possible our SEC analysis overestimates the size of the Met18 homocomplexes.20, 21 Since our overall goal was to obtain quality recombinant protein for in vitro assays, we concluded from these biochemical experiments that we had optimized the expression and purification sufficiently to exploit this reagent for protein-protein interaction analysis. Targeting complex subunit connectivity and stoichiometry. Previously, two studies investigated the binary protein-protein interactions between subunits of the human targeting complex and they arrived at conflicting conclusions about the architecture of the complex.15, 16 Additionally, there is no information about the complex’s size or subunit stoichiometry. To define the yeast system’s subunit connectivity and stoichiometry, we probed all possible binary interactions between Met18, Cia1 and Cia2 via affinity copurification and determined the molecular weight of all stable complexes via SEC. We began by examining the interaction between Cia1 and Cia2. Yeast two hybrid and coIP studies have identified a Cia1-Cia2 interaction in vivo in several eukaryotic organisms and three different groups have reported a direct interaction between hCia1 and hCia2 in vitro.6, 9-16, 22 To determine if the yeast proteins also directly interact, we mixed Cia1 with Cia2 and looked for their association via affinity copurification. Consistent with the proteomic data and reports with

10

ACS Paragon Plus Environment

Page 11 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

the human homologs, we observe an interaction between

His

Cia1 and Cia2 (Figure 2A). The

similar intensities of the Cia1 and Cia2 bands in the elution fraction suggested that Cia1-Cia2 was a stable, stoichiometric complex.

Figure 2. SDS-PAGE analysis of affinity copurification experiments to determine subunit connectivity. A)

His

Cia1 and Cia2 were mixed (input) and separated via IMAC.

specifically retain Cia2 (elution). B)

DT

Cia2 and

DT

Cia1 bait. D)

Strep

Cia1,

Cia1 can

His

Met18 were mixed (input) and separated via

streptactin resin. DTCia2 can specifically retain HisMet18. C) SUMOMet18 and (input) and separated via streptactin resin. No

His

DT

Cia1 were mixed

SUMO

Met18 can be detected copurifying with the

His

Met18 and Cia2 were mixed and chromatographed on streptactin

resin. Both HisMet18 and Cia2 are retained by Cia1 (Lane 3) whereas no bands are detected in the control in which

DT

Cia1 was omitted (Lane 4). Molecular weight standards in kDa are shown to

the right of all the gels and the CIA targeting complex subunits are labeled to the left.

We estimated the molecular weight of the Cia1-Cia2 complex via SEC to determine the subunit stoichiometry. As expected from the reported crystal structure, Cia1 on its own eluted as

11

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

a monomer (Figure 3B, dashed line).23 Cia2 could not be analyzed due to its low extinction coefficient and its propensity to precipitate at concentrations above 0.5 mg/mL. Since a human homolog of Cia2 was recently reported to form a dimer,22 we examined whether the homobifunctional crosslinking reagent bismaleimidoethane (BMOE) could crosslink Cia2 via its absolutely conserved cysteine, Cys161. Incubation of Cia2 with BMOE resulted in appearance of a dimeric product whereas the C161A-Cia2 mutation significantly diminished BMOE-dependent crosslinking (Figure S1). We concluded that Cia2 can form a dimeric species. When a mixture of Cia1 and Cia2 was analyzed via SEC, the peak for the Cia1 monomer disappeared and a new peak with a molecular weight of 90±15 kDa appeared (Figure 3A, solid line). Based on the sizes of

His

Cia1 (40 kDa) and Cia2 (25.6 kDa), we conclude that the Cia1-

Cia2 complex contains one Cia1 polypeptide and two Cia2 polypeptides. Interestingly, it was previously reported that hCia1 and hCia2a can form a 1:1 and 1:2 complex with low nanomolar affinity.22 Thus, both the yeast and the human homologs form a Cia1•Cia22 complex. This observation is consistent with the high sequence and functional conservation of CIA from yeast to humans.24, 25

Figure 3. SEC of the Targeting Complex. In all panels, the SDS-PAGE gel insets are labeled with a, b, and c to indicate migration of Met18, Cia1, and Cia2, respectfully. A) The mixture of His

Cia1,

SUMO

Met18, and Cia2 eluted predominantly as a single, 344 kDa peak (Peak 1, Lane 1). 12

ACS Paragon Plus Environment

Page 13 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Peak 2, eluting at 40 kDa, contains Cia1 and low molecular weight contaminants (Lane 2). B) The HisCia1 (40 kDa) and Cia2 (25.5 kDa) mixture (solid line) eluted predominantly as single, 90 kDa peak (Peak 1) containing both Cia1 and Cia2 (Lane 1).

His

Cia1 (dashed line) eluted as a

single, 39 kDa peak (Peak 2, Lane 2). C) A SUMOMet18 (130 kDa) and Cia2 mixture eluted as a single, 245 kDa peak (Peak 1; Lane 1). The second, 25 kDa peak contains low molecular weight contaminants (Lane 2). Molecular weight standards in kDa are shown on to the right of all the gels.

Next, we investigated the Met18-Cia2 and Met18-Cia1 interactions. While previous studies with the human system were in agreement that hCia1 could interact with hCia2, these studies did not agree on whether hMet18 could form a binary complexes with hCia1 or hCia2.15, 16 When we mixed DTCia2 with HisMet18 and chromatographed the mixture on streptactin resin, we observed a large amount of the Met18 prey in the elution fraction (Figure 2B). We concluded that Met18 and Cia2 form a stable complex. Using SEC, we determined that the

SUMO

Met18-Cia2 complex

has a molecular weight of 245±6 kDa (Figure 3B). This molecular weight is consistent with a complex containing one

SUMO

Met18 (130 kDa) polypeptide and four Cia2 (25.6 kDa)

polypeptides. This Met18•Cia24 complex has a calculated molecular weight of 232.4 kDa, which is in good agreement with the observed molecular weight of 245 kDa. We used a similar approach to look for a Met18-Cia1 interaction. When the bait and

DT

Cia1 was used as

His

Met18 as the prey, no Met18-Cia1 complex was observed (Figure 2C). We

obtained the same result if Met18 was used as the bait and Cia1 as the prey (not shown). We concluded that Met18 and Cia1 cannot form a binary complex or, if they do interact, their affinity is too weak for the complex’s detection by copurification.

13

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

The results from our binary complex study indicate that Cia2 could be the bridge linking Met18 to Cia1 in the targeting complex. To probe the bridging function of Cia2 directly, we examined whether Cia2 could simultaneously interact with both Met18 and Cia1 or if Cia2 forms mutually exclusive binary complexes with these two interaction partners. When we mixed DT

Cia1,

His

Met18, and Cia2 and chromatographed the mixture on streptactin resin, we observed

both Cia2 and

His

Met18 eluting with the

DT

Cia1 bait (Figure 2D). Since Met18 does not form a

stable, binary complex with Cia1 (Figure 2C), its presence in the elution fraction when all three targeting complex subunits are present demonstrates that Cia2 is the central scaffold of the targeting complex. When we analyzed this mixture via SEC, we observed a 344±4 kDa peak containing all three polypeptides (Figure 3C). To assign the quaternary structure, we reasoned that the Met18-Cia1-Cia2 complex is comprised of the stable Cia1•Cia22 and Met18•Cia24 subcomplexes. Therefore, we would expect the ratio of subunits observed in the binary complexes to be maintained in the full targeting complex. Based on this hypothesis, we propose that the 344 kDa complex contains one Met18, two Cia1, and four Cia2 polypeptides. This SUMO

Met18•[HisCia1•Cia22]2 complex has a calculated molecular weight of 312 kDa which is in

good agreement with the observed molecular weight in the SEC experiment. Given the ~30 kDa difference between the expected and observed molecular weight, we additionally analyzed the molecular weight of the Met18•Cia1•Cia2 complex via a second approach, SEC coupled to multi-angle light scattering (MALS). For SEC-MALS, we mixed His

Cia1 with untagged Cia2 and untagged Met18 and purified the Cia1-containing complexes via

IMAC immediately before SEC-MALS analysis. We observed two major peaks with sufficient intensity that we could determine the molecular weight of the complex via MALS (Figure S2). The most abundant peak elutes with a molecular weight of 94.2±1.0 kDa, which corresponds to

14

ACS Paragon Plus Environment

Page 15 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

the HisCia1•Cia22 complex, which has an expected molecular weight of 91 kDa. The next largest peak elutes with a molecular weight of 285.4 ± 1.1 kDa, which is in good agreement with the predicted molecular weight of 300 kDa for the Met18•[HisCia1•Cia22]2 complex. In total, the SEC-MALS data is consistent with our assigned quaternary structure for the full targeting complex in which two Cia1•Cia22 complexes dock onto a single Met18 polypeptide. The specificity of CIA substrate recognition by the targeting complex. After determining the targeting complex’s subunit connectivity and stoichiometry, we next investigated its target recognition specificity. Co-immunoprecipitations studies have determined that the targeting complex interacts with numerous cytosolic and nuclear FeS proteins.4-6, 15, 16, 26, 27 One study also reported that the targeting can also associate with mitochondrial FeS proteins.15 This observation suggests that the targeting complex might be able to promiscuously identify any FeS protein, not just CIA substrates residing in the cytosol and the nucleus. To begin deciphering the rules governing target recognition, we examined the ability of the targeting complex to directly bind two validated CIA targets, Rad3 and Leu1, in addition to an FeS protein derived from E. coli, the fumerate-nitrate transcriptional activator FNR. Rad3 is a helicase required for nucleotide excision repair and transcription.28 The human homolog of Rad3 (ERCC2 or XPD; here referred to as hRad3) has been shown to associate with the CIA targeting complex in vivo and in vitro.4-6, 15, 16, 26, 27 Therefore, we chose Rad3 as the first target to validate our in vitro assay for target recognition. Since no methods to access recombinant Rad3 have been reported, we expressed Rad3 in E. coli and developed a method to refold the protein from solubilized inclusion bodies. CD confirmed that the refolded Rad3 has a large amount of secondary structure, consistent with successful refolding (Figure S3). Inspection of the UV-Vis

15

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

spectrum of refolded Rad3 did not reveal any features consistent with an FeS cluster, confirming Rad3 was isolated in its apo-form (not shown). To validate both that the refolding procedure was successful and that our reconstituted targeting complex can execute target recognition in vitro, we mixed

DT

Cia2 (bait) with

His

Cia1,

SUMO

Met18, and Rad3 and passed the mixture through a streptactin column. As shown in Figure

4A, a significant amount of Rad3 elutes with the Cia2 bait and the other targeting complex subunits. Importantly, no bands corresponding to Rad3, Cia1 or Met18 were observed in the elution when the DTCia2 bait was omitted. The intensity of the Rad3 band relative to those of the targeting complex subunits suggests that most of the sample contains targeting complex associated with Rad3.

Figure 4. Interaction of the Met18-Cia1-Cia2 complex with targets. target were mixed with the

SUMO

Met18,

His

Cia1 and

DT

Cia2 bait (input) and chromatographed on steptactin resin. The

bound proteins were eluted and analyzed by SDS-PAGE. Both Rad3 (Panel A) and Leu1 (Panel B) are enriched in the elution fraction specifically in the presence of the

DT

Cia2-bait, but not in

experiments where the bait was omitted. FNR (Panel C) is not enriched in a

DT

Cia2-dependent

manner as both Coomassie blue staining and in-gel fluorescence imaging of SNAP-FNR fusion

16

ACS Paragon Plus Environment

Page 17 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

contain similar amounts of FNR in elution fractions both in the presence (+) and absence (-) of DT

Cia2. Molecular weight standards in kDa are shown to the right of all the gels.

Having demonstrated recognition of a nuclear target, we next examined if targeting complex can recognize a very different target, the cytosolic FeS-dependent isomerase Leu1. Genetic studies have revealed that Leu1 is dependent on Cia1, Cia2 and Met18 for cofactor acquisition in vivo.4, 5, 7, 8 Yet, to our knowledge, no coIP, yeast two hybrid, nor protein complementation study has reported interaction between Leu1 and any component of the targeting complex. Thus, it is unknown if Leu1 is identified by the targeting complex directly, or if an adaptor is required to mediate Leu1 recognitions similar to what was recently proposed for recognition of Rli1.29 When we mixed apo-Leu1 with the other components of the targeting complex in our copurification assay, Leu1 is observed in the elution fraction along with the targeting complex subunits (Figure 4B). However, unlike the result with Rad3, the intensity of the Leu1 band was low relative to the intensities of the Met18, Cia1, and Cia2 bands. However, the Leu1 band intensity was reproducibly above the background observed in the control where the

DT

Cia2 bait

was omitted. These observations demonstrate that both Rad3 and Leu1 can directly bind the targeting complex in vitro. They additionally suggest that the affinity for Leu1 is less than that for Rad3. This result could explain why there are numerous reports of Rad3 associating with Met18, Cia1, and/or Cia2 subunits but none reporting interaction with Leu1. Our observation that the targeting complex can bind both Leu1 and Rad3 demonstrates that the targeting complex can flexibly recognize very different apo-targets in vitro. Importantly, these two targets do not share any easily detectable sequence or structural similarities. In fact, the only similarity we could find was the four cysteine residues required to bind their [Fe4S4]-cofactors.

17

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 30

Therefore, we tested whether the targeting complex was able to promiscuously recognize any FeS-protein via their cluster-ligating cysteine motifs. For this purpose, we utilized the E. coli fumarate and nitrate reductase regulatory (FNR) transcription factor which exploits an [Fe4S4] cluster sensor to regulate gene expression in response to oxygen.30 We cloned FNR fused to a both His- and SNAP-tags. The SNAP fusion shifted the molecular weight of FNR away from that of the Cia1 and Cia2 proteins so we could continue to use SDSPAGE for analysis. The SNAP-tag also enabled specific detection of FNR by exploiting commercially available fluorescent labels for the tag. As expected, the UV-Vis spectrum of the protein confirmed little, if any, of FNR’s oxygen labile [Fe4S4] cluster was associated with the aerobically purified protein (not shown). However, if we introduced a cluster-stabilizing mutation, FNR purified with some FeS cluster associated demonstrating the SNAP fusion did not perturb the FNR tertiary structure (not shown).31 When we examined the ability of FNR to bind the targeting complex, little FNR was observed in the elution fraction though bands for Met18, Cia1, and Cia2 were all clearly present (Figure 4C). Since the elution fractions from both the experimental and control samples contained a faint band close to the molecular weight of FNR, we labeled FNR with a fluorescent SNAP substrate and monitored the in-gel fluorescence (Figure 4C, right panel). The specific detection of SNAPFNR revealed that there is no enrichment of the target in the interaction experiment relative to the control where the bait protein was omitted. Therefore, while the CIA targeting complex can directly bind two of validated apo-substrates of CIA, it cannot promiscuously recognize any FeS enzyme in vitro. The Met18-Cia2 subcomplex is sufficient for Rad3 recognition. Having established that the targeting complex can specifically identify multiple apo-targets, we wanted to pinpoint which

18

ACS Paragon Plus Environment

Page 19 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

subunit or subunits executes target recognition. In vitro coIP studies with the human system have reported that both hMet18 and hCia2 can form binary complexes with hRad3; however these previous studies disagreed on the ability of hCia1 to recognize hRad3.15, 16 Since Rad3 bound the targeting complex with the highest stoichiometry, yet there was conflicting data as to which subunits directly interact with its human homolog, we used Rad3 to define the minimal requirements for target recognition in vitro. We began by investigating the binary interactions between Rad3 and the individual targeting complex subunits. The results showed that Cia1 did not interact with Rad3 (Figure 5A, middle panel), but the data were less clear-cut for binding to Met18 or Cia2. When either SUMOMet18 or DT

Cia2 was used as the bait, the elution fractions contained small amounts of Rad3 (Figure 5A,

upper and lower panels). This result contrasts with the strong association seen for Rad3 to the full targeting complex (Figure 4A). Additionally, the observation of a small amount of Rad3 binding to Met18 or Cia2 was variable and was sensitive to the composition of the wash buffer and amount of wash buffer used in the copurification experiment. We concluded that Cia1 does not form a stable binary complex with Rad3, and that Met18 and Cia2 likely do bind Rad3 target but only weakly. Thus, the full targeting complex appears to bind Rad3 with higher affinity than any of its individual subunits.

19

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

Figure 5. SDS-PAGE analysis of Rad3’s interaction with targeting complex components. Each panel shows results of the interaction tests both in the presence (+) and absence of the bait, which is marked with an asterisk to the left of each gel. A) Binary interaction tests with Rad3 (prey) and the targeting complex subunits Met18 (top), Cia1 (middle), and Cia2 (bottom). Both Met18 and Cia2 baits resulted in some enrichment of Rad3 in the elution, whereas the Cia1 bait did not. B) Rad3 and SUMOMet18 were mixed with the DT Cia2 bait (input). The streptactin column elution contains Met18, Rad3 and Cia2 demonstrating the Met18-Cia2 complex can bind Rad3. C) Rad3 and

His

Cia1 were mixed in the presence and absence of the

DT

Cia2 bait (input). The streptactin

column elution contains Cia1 and Cia2 whereas little, if any, Rad3 is present. D) Leu1 and SUMO

Met18 were mixed in the presence and absence of the

DT

Cia2 bait (input). The streptactin

column elution shows little, if any, Leu1 is present in the elution fraction. E) Leu1 and were mixed with

DT

His

Cia1

Cia2 (input). The streptactin column elution shows little, if any, Leu1 is

present. Molecular weight standards in kDa are shown to the right of all the gels.

20

ACS Paragon Plus Environment

Page 21 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Next, we investigated whether the Cia1-Cia2 or Met18-Cia2 subcomplexes could stably bind Rad3. When we mixed co-purifying with

DT

Cia2 with Rad3 and either

DT

Cia2 only in the presence of

SUMO

Met18 or

His

Cia1, we observed Rad3

SUMO

Met18 (Figure 5B). Little, if any, Rad3

was observed in the elution fraction in the presence of

DT

Cia2 and

His

Cia1 (Figure 5C). These

results indicate that the Met18-Cia2 subcomplex is sufficient to stably associate with Rad3, whereas the Cia1-Cia2 subcomplex is not. To determine if this result was generalizable to other apo-targets, we tested the interaction of Leu1 with both the Met18-Cia2 and Cia1-Cia2 subcomplexes. However, unlike what we observed with Rad3, we detected little, if any, Leu1 in the elution fraction above the background observed with the no bait control (Figure 5D and 5E). We observed a similar result when we looked for a binary complex between Leu1 and each of the targeting complex subunits (not shown). Therefore, neither of the binary subcomplexes nor the individual subunits can stably associate with Leu1 in vitro. Since the Met18-Cia2 complex was sufficient to bind Rad3 but not Leu1, our results reveal differences in how different targets are recognized by the targeting complex in vitro. Together, our work with Leu1 suggests that its affinity the targeting complex is lower than of Rad3 and that none of the individual targeting complex subunits or its subcomplexes are sufficient to recognize Leu1.

DISCUSSION A major unsolved problem in cytosolic FeS cluster biogenesis is pinpointing how apo-targets that differ in their sequence, structure, and function are identified as CIA substrates. To begin defining the targeting mechanism, two groups previously investigated the structural organization of the human targeting complex, but arrived at conflicting models regarding its connectivity.15, 16

21

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 30

Herein, we investigated all of the binary interactions between the subunits of the yeast targeting complex to create a unifying model for its architecture. Importantly, the previous studies depended on IP coupled to Western blotting to detect the interactions while our approach is based on affinity copurification and direct visualization of the proteins via SDS-PAGE. Therefore, we can qualitatively assess the degree of complex formation by comparison of band intensities. We detected both Met18-Cia2 and Cia1-Cia2 complexes but we did not observe a stable Met18-Cia1 complex (Figure 2). Therefore, our results are consistent with those of Seki et al. and establish that Cia2 is at the center of the targeting complex. Based on our SEC analysis (Figure 3), we propose that the Met18-Cia1-Cia2 complex is formed by the docking of two Cia1•Cia22 complexes onto Met18 (Figure 6).

Figure 6. A working model for structure of the targeting complex and its interaction with targets. The Cia2 dimer can interact with either Met18 or Cia1 to form the observed Cia1•Cia22 and Met18•(Cia22)2 complexes. Two Cia1•Cia22 complexes bind Met18 to form the full targeting complex. The Met18•(Cia22)2 and the Met18•(Cia1•Cia22)2 complex are the minimal complexes required to bind Rad3 and Leu1, respectfully.

22

ACS Paragon Plus Environment

Page 23 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

The result that Cia2 is the central subunit of the targeting complex was surprising since before completing this work, we had hypothesized that Cia1 would be the organizing hub. Cia1 is made up of WD40 domains which commonly serve as platforms for scaffolding multiprotein complexes.32 Although we cannot rule out the possibility that Cia1 binds to Met18 with low affinity, our work suggests that Cia1’s versatile protein interaction platform is utilized for another purpose, such as binding the CIA factor Nar1 or for target recognition.7, 15, 16 In fact, hCia1 has been reported to directly associate with targets including DNA polymerase ε, hRad3, and viperin.16, 33, 34 WD40 proteins are well known to mediate specific interactions with multiple interaction partners.32 While we found Cia1 is required to bind Leu1 to the targeting complex, we could not detect a stable binary complex between these two proteins. Thus, if Cia1 and Leu1 do directly bind to one another, they must do so via a low affinity interaction that is strengthened in the context of the full targeting complex. Our results are most consistent model that Leu1, and possibly additional targets, bind the targeting complex via a direct interaction with Cia1. Future quantitative studies of the protein-protein interaction affinities will be required to further refine the model for Leu1 identification and clarify the role of Cia1 plays in helping to tether Leu1 and possibly additional targets to the targeting complex. To begin defining the targeting mechanism, we demonstrate that two very different targets can both interact with the targeting complex in vitro (Figure 4). The simplest model to explain this result would be if Leu1 and Rad3 share a motif sufficient for their recognition, similar to the mechanism recently proposed for the mitochondrial FeS biogenesis system.3,

35, 36

However,

analysis of the Leu1 and Rad3 sequences failed to identify short regions of high homology. The only similarity we could identify is a series of cluster-binding cysteines. Therefore, we considered the possibility that any FeS protein could bind the CIA targeting complex via its

23

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 30

cysteine-rich FeS binding motif. To test this hypothesis, we investigated whether the targeting complex could identify FNR, an FeS protein derived from bacteria that does not have any eukaryotic homologs. As shown in Figure 4, we could not detect an interaction between FNR and the Met18-Cia1-Cia2 complex demonstrating that the cluster binding cysteine motif is not sufficient for target recognition. This finding is consistent with reports aimed at pinpointing the region of hRad3 responsible for association hMet18.16, 26, 27 While the published studies disagree as to what region of hRad3 is essential for binding hMet18, as described below, they do agree that the hMet18 binding site of Rad3 lies outside of its FeS-binding domain. Finally, we investigated what component or components of the targeting complex execute target recognition (Figure 5). This question is especially important because recent proteomic studies have suggested that Cia1, Cia2, and Met18 can form mixtures of complexes, each capable of interacting with specific subsets of targets.6 This model predicts that individual subunits or subcomplexes will interact with distinct subsets of targets and that each subunit of the targeting complex contributes to target recognition. When we investigated which subunits or subcomplexes were sufficient to bind Rad3 and Leu1, we discovered that the Met18-Cia2 subcomplex, but none of the isolated subunits nor the Cia1-Cia2 complex, was able to tightly bind Rad3 (Figure 5). Since only the full targeting complex is sufficient to bind Leu1, our work reveals target-specific differences in CIA substrate recognition that were predicted by high throughput IP-MS analyses (Figure 6).6 In total, our in vitro studies support the model that CIA substrate recognition can be executed by the full targeting complex or by its stable subcomplexes. Our observation that Rad3 stably binds the Met18-Cia2 subcomplex, but not to either of the individual proteins, can be explained by two possible models. First, the association of Met18 and

24

ACS Paragon Plus Environment

Page 25 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Cia2 stabilizes the apo-protein binding component in a conformation competent to interact with target. Alternatively, Met18 and Cia2 could each bind Rad3 relatively weakly and these interactions synergize in the ternary complex to stably bind Rad3. Since we and others have observed both Met18 and Cia2 or their human homologs weakly bind Rad3 (Figure 5),15, 16 we favor the second model. This in turn predicts that two regions of Rad3 will be required to bind the targeting complex, one region to bind Met18 and one region to bind Cia2. Indeed, Lansdorp, Uringa, and coworkers demonstrated the FeS-binding domain of hRad3 (residues 110-196) binds to hCia2 while Tanaka and coworkers have pinpointed the hMet18 binding site as residing in the C-terminal region of hRad3 between residues 438-637.15, 26 Although an investigation into how the hRad3 ortholog RTEL1 binds to hMet18 supported the proposal of Tanaka and coworkers that the hMet18 binding site lies within the C-terminal region of hRad3,15 Wohlshlegel et al. recently proposed that residues 277-286 within hRad3’s Arch domain mediates the interaction with hMet18.27 Together, these studies support two important aspects of our model for Rad3 identification by the CIA targeting complex; that the motif(s) for CIA targeting complex binding lie outside of the FeS-binding domain and that there are multiple points of contact between Rad3 and the targeting complex. However, the conflicting data in the literature demonstrate that much more work remains to be done to uncover the cryptic code driving CIA target identification. Access to a fully reconstituted system as described herein will undoubtedly enable further development of quantitative assays crucial for bringing the mechanism of CIA substrate recognition into molecular focus.

FOR TABLE OF CONTENTS USE ONLY

25

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ASSOCIATED CONTENT Supporting Information. Supporting information include supplementary Figures S1-S3 and Table S1.

AUTHOR INFORMATION

Corresponding Author * Department of Chemistry, Boston University, 590 Commonwealth Avenue, Boston, MA 02114. Tel.: 617-358-6180; E-mail: [email protected]. Present Addresses †Nicholas Fleischman: ZoomRx Inc, Cambridge, MA 02142. Calina Glynn: University of California, Los Angeles, Los Angeles CA, 90095. Mary Froehlich: University of Nevada, Reno, NV 89557. Zaynab Hassan: University of Illinois, Chicago, IL 60607 Footnote ** This manuscript is dedicated to the memory of Jessica Cosman. Her passion for science and remarkable talent for research will continue to inspire us all. Author Contributions ATV collected and analyzed results and wrote the paper. NMF collected and analyzed the CD data. MJF, CYL, ZOH, CAG, JAC and NMF created clones, purified proteins, and collected

26

ACS Paragon Plus Environment

Page 26 of 30

Page 27 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

data. DLP designed the study, analyzed results and wrote the paper. All authors edited the manuscript. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Funding Sources CAG, MJF, CYL, and JAC thank the Undergraduate Research Opportunities Program for support. We also thank the NSF for fellowship support for ATV (GK-12 Graduate STEM fellowship in K-12 Education, DGE-0947950 and CHE-1555295) and ZOH (NSF-REU, CHE1156666) and for access to the CD spectrometer (MRI, CHE-1126545). We also acknowledge Boston University for funding this work. ACKNOWLEDGMENT We thank Mihayl Petkov for construction of the Leu1 plasmid. We also thank Dr. Jeffrey Bacon and the staff in the Chemistry Instrumentation Center for their assistance with the CD data collection. We thank Dr. Kelly Arnett at the Center for Macromolecular Interactions in the Department of Biological Chemistry and Molecular Pharmacology at Harvard Medical School for her assistance with the collection and analysis of the SEC-MALS data.

REFERENCES [1] Johnson, D. C., Dean, D. R., Smith, A. D., and Johnson, M. K. (2005) Structure, function, and formation of biological iron-sulfur clusters, Annu. Rev. Biochem. 74, 247-281. [2] Paul, V. D., and Lill, R. (2015) Biogenesis of cytosolic and nuclear iron-sulfur proteins and their role in genome stability, Biochim. Biophys. Acta 1853, 1528-1539. [3] Maio, N., and Rouault, T. A. (2015) Iron-sulfur cluster biogenesis in mammalian cells: New insights into the molecular mechanisms of cluster delivery, Biochim. Biophys. Acta 1853, 1493-1512. [4] Gari, K., León Ortiz, A. M., Borel, V., Flynn, H., Skehel, J. M., and Boulton, S. J. (2012) MMS19 links cytoplasmic iron-sulfur cluster assembly to DNA metabolism., Science 337, 243-245.

27

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

[5] Stehling, O., Vashisht, A. A., Mascarenhas, J., Jonsson, Z. O., Sharma, T., Netz, D. J. A., Pierik, A. J., Wohlschlegel, J. A., and Lill, R. (2012) MMS19 Assembles Iron-Sulfur Proteins Required for DNA Metabolism and Genomic Integrity, Science 337, 195-199. [6] Stehling, O., Mascarenhas, J., Vashisht, A. A., Sheftel, A. D., Niggemeyer, B., Rösser, R., Pierik, A. J., Wohlschlegel, J. A., and Lill, R. (2013) Human CIA2A-FAM96A and CIA2B-FAM96B integrate iron homeostasis and maturation of different subsets of cytosolic-nuclear iron-sulfur proteins., Cell Metab. 18, 187-198. [7] Balk, J., Aguilar Netz, D. J., Tepper, K., Pierik, A. J., and Lill, R. (2005) The essential WD40 protein Cia1 is involved in a late step of cytosolic and nuclear iron-sulfur protein assembly, Mol. Cell. Biol. 25, 10833-10841. [8] Weerapana, E., Wang, C., Simon, G. M., Richter, F., Khare, S., Dillon, M. B., Bachovchin, D. A., Mowen, K., Baker, D., and Cravatt, B. F. (2010) Quantitative reactivity profiling predicts functional cysteines in proteomes, Nature 468, 790-795. [9] Uetz, P., Giot, L., Cagney, G., Mansfield, T. A., Judson, R. S., Knight, J. R., Lockshon, D., Narayan, V., Srinivasan, M., Pochart, P., Qureshi-Emili, A., Li, Y., Godwin, B., Conover, D., Kalbfleisch, T., Vijayadamodar, G., Yang, M., Johnston, M., Fields, S., and Rothberg, J. M. (2000) A comprehensive analysis of protein-protein interactions in Saccharomyces cerevisiae, Nature 403, 623-627. [10] Ito, T., Chiba, T., Ozawa, R., Yoshida, M., Hattori, M., and Sakaki, Y. (2001) A comprehensive two-hybrid analysis to explore the yeast protein interactome, Proc. Natl. Acad. Sci. U. S. A. 98, 4569-4574. [11] Krogan, N. J., Cagney, G., Yu, H., Zhong, G., Guo, X., Ignatchenko, A., Li, J., Pu, S., Datta, N., Tikuisis, A. P., Punna, T., Peregrin-Alvarez, J. M., Shales, M., Zhang, X., Davey, M., Robinson, M. D., Paccanaro, A., Bray, J. E., Sheung, A., Beattie, B., Richards, D. P., Canadien, V., Lalev, A., Mena, F., Wong, P., Starostine, A., Canete, M. M., Vlasblom, J., Wu, S., Orsi, C., Collins, S. R., Chandran, S., Haw, R., Rilstone, J. J., Gandi, K., Thompson, N. J., Musso, G., St Onge, P., Ghanny, S., Lam, M. H., Butland, G., Altaf-Ul, A. M., Kanaya, S., Shilatifard, A., O'Shea, E., Weissman, J. S., Ingles, C. J., Hughes, T. R., Parkinson, J., Gerstein, M., Wodak, S. J., Emili, A., and Greenblatt, J. F. (2006) Global landscape of protein complexes in the yeast Saccharomyces cerevisiae, Nature 440, 637-643. [12] Havugimana, P. C., Hart, G. T., Nepusz, T., Yang, H., Turinsky, A. L., Li, Z., Wang, P. I., Boutz, D. R., Fong, V., Phanse, S., Babu, M., Craig, S. A., Hu, P., Wan, C., Vlasblom, J., Dar, V. U., Bezginov, A., Clark, G. W., Wu, G. C., Wodak, S. J., Tillier, E. R., Paccanaro, A., Marcotte, E. M., and Emili, A. (2012) A census of human soluble protein complexes, Cell 150, 1068-1081. [13] Gavin, A. C., Bosche, M., Krause, R., Grandi, P., Marzioch, M., Bauer, A., Schultz, J., Rick, J. M., Michon, A. M., Cruciat, C. M., Remor, M., Hofert, C., Schelder, M., Brajenovic, M., Ruffner, H., Merino, A., Klein, K., Hudak, M., Dickson, D., Rudi, T., Gnau, V., Bauch, A., Bastuck, S., Huhse, B., Leutwein, C., Heurtier, M. A., Copley, R. R., Edelmann, A., Querfurth, E., Rybin, V., Drewes, G., Raida, M., Bouwmeester, T., Bork, P., Seraphin, B., Kuster, B., Neubauer, G., and Superti-Furga, G. (2002) Functional organization of the yeast proteome by systematic analysis of protein complexes, Nature 415, 141-147. [14] Ho, Y., Gruhler, A., Heilbut, A., Bader, G. D., Moore, L., Adams, S.-L., Millar, A., Taylor, P., Bennett, K., Boutilier, K., Yang, L., Wolting, C., Donaldson, I., Schandorff, S., Shewnarane, J., Vo, M., Taggart, J., Goudreault, M., Muskat, B., Alfarano, C., Dewar, D., Lin, Z., Michalickova, K., Willems, A. R., Sassi, H., Nielsen, P. A., Rasmussen, K. J., 28

ACS Paragon Plus Environment

Page 28 of 30

Page 29 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Andersen, J. R., Johansen, L. E., Hansen, L. H., Jespersen, H., Podtelejnikov, A., Nielsen, E., Crawford, J., Poulsen, V., Sørensen, B. D., Matthiesen, J., Hendrickson, R. C., Gleeson, F., Pawson, T., Moran, M. F., Durocher, D., Mann, M., Hogue, C. W. V., Figeys, D., and Tyers, M. (2002) Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry, Nature 415, 180-183. [15] van Wietmarschen, N., Moradian, A., Morin, G. B., Lansdorp, P. M., and Uringa, E.-J. (2012) The Mammalian Proteins MMS19, MIP18, and ANT2 Are Involved in Cytoplasmic Iron-Sulfur Cluster Protein Assembly., J. Biol. Chem. 287, 43351-43358. [16] Seki, M., Takeda, Y., Iwai, K., and Tanaka, K. (2013) IOP1 Protein Is an External Component of the Human Cytosolic Iron-Sulfur Cluster Assembly (CIA) Machinery and Functions in the MMS19 Protein-dependent CIA Pathway, J. Biol. Chem. 288, 16680-16689. [17] Gibson, D. G., Young, L., Chuang, R. Y., Venter, J. C., Hutchison, C. A., 3rd, and Smith, H. O. (2009) Enzymatic assembly of DNA molecules up to several hundred kilobases, Nat. Methods 6, 343-345. [18] Studier, F. W. (2014) Stable expression clones and auto-induction for protein production in E. coli, Methods Mol. Biol. 1091, 17-32. [19] Fleischman, N. M., Vo, A., Camire, E. J., Cosman, J., and Perlstein, D. L. (2017) Defining the domains of Cia2 required for its function in vivo and in vitro, in preparation. [20] Hatfield, M. D., Reis, A. M., Obeso, D., Cook, J. R., Thompson, D. M., Rao, M., Friedberg, E. C., and Queimado, L. (2006) Identification of MMS19 domains with distinct functions in NER and transcription, DNA Repair (Amst) 5, 914-924. [21] Fournier, D., Palidwor, G. A., Shcherbinin, S., Szengel, A., Schaefer, M. H., Perez-Iratxeta, C., and Andrade-Navarro, M. A. (2013) Functional and genomic analyses of alphasolenoid proteins, PLoS One 8, e79894. [22] Chen, K. E., Richards, A. A., Ariffin, J. K., Ross, I. L., Sweet, M. J., Kellie, S., Kobe, B., and Martin, J. L. (2012) The mammalian DUF59 protein Fam96a forms two distinct types of domain-swapped dimer, Acta Crystallogr. D Biol. Crystallogr. 68, 637-648. [23] Srinivasan, V., Netz, D. J. A., Webert, H., Mascarenhas, J., Pierik, A. J., Michel, H., and Lill, R. (2007) Structure of the yeast WD40 domain protein Cia1, a component acting late in iron-sulfur protein biogenesis, Structure 15, 1246-1257. [24] Tsaousis, A. D., Gentekaki, E., Eme, L., Gaston, D., and Roger, A. J. (2014) Evolution of the cytosolic iron-sulfur cluster assembly machinery in Blastocystis species and other microbial eukaryotes, Eukaryot. Cell 13, 143-153. [25] Kachroo, A. H., Laurent, J. M., Yellman, C. M., Meyer, A. G., Wilke, C. O., and Marcotte, E. M. (2015) Evolution. Systematic humanization of yeast genes reveals conserved functions and genetic modularity, Science 348, 921-925. [26] Ito, S., Tan, L. J., Andoh, D., Narita, T., Seki, M., Hirano, Y., Narita, K., Kuraoka, I., Hiraoka, Y., and Tanaka, K. (2010) MMXD, a TFIIH-independent XPD-MMS19 protein complex involved in chromosome segregation, Mol. Cell 39, 632-640. [27] Vashisht, A. A., Yu, C. C., Sharma, T., Ro, K., and Wohlschlegel, J. A. (2015) The Association of the Xeroderma Pigmentosum Group D DNA Helicase (XPD) with Transcription Factor IIH Is Regulated by the Cytosolic Iron-Sulfur Cluster Assembly Pathway, J. Biol. Chem. 290, 14218-14225. [28] Wu, Y., and Brosh, R. M. (2012) DNA helicase and helicase-nuclease enzymes with a conserved iron-sulfur cluster., Nucleic Acids Res. 40, 4247-4260. 29

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

[29] Paul, V. D., Muhlenhoff, U., Stumpfig, M., Seebacher, J., Kugler, K. G., Renicke, C., Taxis, C., Gavin, A. C., Pierik, A. J., and Lill, R. (2015) The deca-GX proteins Yae1-Lto1 function as adaptors recruiting the ABC protein Rli1 for iron-sulfur cluster insertion, Elife 4, e08231. [30] Crack, J., Green, J., and Thomson, A. J. (2004) Mechanism of oxygen sensing by the bacterial transcription factor fumarate-nitrate reduction (FNR), J. Biol. Chem. 279, 9278-9286. [31] Bates, D. M., Popescu, C. V., Khoroshilova, N., Vogt, K., Beinert, H., Munck, E., and Kiley, P. J. (2000) Substitution of leucine 28 with histidine in the Escherichia coli transcription factor FNR results in increased stability of the [4Fe-4S](2+) cluster to oxygen, J. Biol. Chem. 275, 6234-6240. [32] Stirnimann, C. U., Petsalaki, E., Russell, R. B., and Muller, C. W. (2010) WD40 proteins propel cellular networks, Trends Biochem. Sci. 35, 565-574. [33] Moiseeva, T. N., Gamper, A. M., Hood, B. L., Conrads, T. P., and Bakkenist, C. J. (2016) Human DNA polymerase ε is phosphorylated at serine-1940 after DNA damage and interacts with the iron-sulfur complex chaperones CIAO1 and MMS19, DNA Repair 43, 9-17. [34] Upadhyay, A. S., Vonderstein, K., Pichlmair, A., Stehling, O., Bennett, K. L., Dobler, G., Guo, J. T., Superti-Furga, G., Lill, R., Overby, A. K., and Weber, F. (2014) Viperin is an iron-sulfur protein that inhibits genome synthesis of tick-borne encephalitis virus via radical SAM domain activity, Cell. Microbiol. 16, 834-848. [35] Maio, N., Ghezzi, D., Verrigni, D., Rizza, T., Bertini, E., Martinelli, D., Zeviani, M., Singh, A., Carrozzo, R., and Rouault, T. A. (2016) Disease-Causing SDHAF1 Mutations Impair Transfer of Fe-S Clusters to SDHB, Cell Metab. 23, 292-302. [36] Maio, N., Singh, A., Uhrigshardt, H., Saxena, N., Tong, W. H., and Rouault, T. A. (2014) Cochaperone binding to LYR motifs confers specificity of iron sulfur cluster delivery, Cell Metab. 19, 445-457.

30

ACS Paragon Plus Environment

Page 30 of 30