Supporting InformationSuloway et al. 10.1073/pnas.0907522106SI TextCloning, Expression, and Purification. The A. fumigatus Get3 codingsequence was synthesized by PCR using primers designed withDNAWorks (1), and the S. cerevisiae GET3 gene was amplifiedby PCR from genomic DNA, both with NcoI and XhoI restric-tion sites added. Amplified DNA fragments were NcoI/XhoIdigested and ligated into pET33b (Novagen) to create C-terminally 6�His tagged constructs.

A. fumigatus and S. cerevisiae Get3 proteins were recombi-nantly expressed in E. coli BL21-Gold(DE3) cells grown in 2�YT medium for 3 h. at 37 °C after induction with 0.3 mMisopropyl-�-D-thiogalactopyranoside. Get3 was purified by Ni-NTA affinity chromatography and gel filtration on a Superdex200 16/60 column (GE Healthcare). Fractions from gel filtrationwere concentrated to 10–15 mg/mL of protein and dialyzed in abuffer of 5 mM Tris, pH 7.5, and 6 mM beta-mercaptoethanol(�ME) for crystallization.

Crystallization and Structure Determination. All crystals were grownby the sitting-drop vapor diffusion method with a 1:1 ratio ofprotein to precipitant solutions at 23 °C. ScGet3-apo crystalswere obtained in 0.1 M HEPES pH 8.0, 1.6 M ammonium sulfateand 6 mM �ME, AfGet3-apo crystals in 0.1 mM bis-Tris, pH 7.0,0.2 M NaCl, 1.5 M ammonium sulfate and 6 mM �ME, andAfGet3-ADP co-crystals in 0.2 M potassium citrate, 16% (wt/vol) polyethylene glycol 3350, and 6 mM �ME, with the proteinsolution supplemented with 2 mM ADP and 1 mM MgCl2.ScGet3-Apo were cryoprotected by transfer to 3.4 M sodiummalonate (pH 7), and AfGet3-Apo crystals and AfGet3-ADPco-crystals were serially transfered to artificial mother liquorsupplemented with 20% (wt/vol) sucrose and 20% (wt/vol)xylitol and with 20% ethylene glycol, respectively. All crystalswere flash-frozen to 100 K by direct transfer into liquid nitrogen.Selenomethionine (SeMet) derivatives of AfGet3 were handledin the same manner as the native protein.

All diffraction data were obtained on Stanford SynchrotronRadiation Lightsource beam line 12–2 at the SLAC NationalAccelerator Laboratory, at 100 K. (Table S1). Diffraction datawere integrated with MOSFLM and scaled with SCALA (2, 3).Multiple-wavelength anomalous dispersion (MAD) data from aSeMet derivative of AfGet3-ADP allowed the assignment offorty-nine selenium sites using SHELXD, and experimentalphases were calculated by SOLVE with an overall figure of meritof 0.63 (4, 5). BUCCANEER performed density modificationand built an initial model, and the complete model was manuallybuilt in COOT (6, 7).

Refinement against the 3.2-Å resolution native AfGet3-ADPdata used strict 6-fold NCS symmetry and consisted of cycles ofsimulated annealing and group B-factor refinement in CNSfollowed by manual rebuilding (8, 9). Unambiguous density wasobserved for residues 12–106, 125–189, 195–277, and 282–338 inall monomers. The overall topology is shown in Fig. S2 A. Densitywas observed for residues 190–194 in the SB2 loops but could notbe confidently modeled. Residues 195–200 were modeled indi-vidually into each monomer. TLS groups were determined usingthe TLSMD web server and NCS restraints were relaxed to allowvariation between subunits for final refinement in PHENIX toyield an R-factor of 21.2% and an R-free of 25.13% (8, 9).

Molecular replacement of the ScGet3-apo dataset was per-

formed with PHASER (10). The search model was preparedfrom the AfGet3 ADP structure by removing SB1 loop residues98 to 129, SB2 residues 177 to 214, and residues 282–288 of H9.A single copy was located in the asymmetric unit and the initialweighted 2 Fo - Fc maps showed density that allowed rebuildingof H9, an extension of the C terminus by 11 residues to form ahelix that packs against S8, and shifts in H5, H6, and H7 (Fig. 1Eand Fig. S2 A).

An anomalous difference map contained a strong peak be-tween C285 and C288 and a second peak at the proximal 3-foldaxis special position. X-ray fluorescence spectroscopy indicatedthat the crystal contained zinc. Density between the two peakscould be modeled as the C-terminal hexahistidine tag used inpurification as an extended strand, resulting in square planercoordination of a metal between the two cysteines and twohistidines, and octahedral coordination at the special position bysymmetry-related histidines.

After a manual rebuild, the helices and the �-sheet wererefined as rigid bodies and one isotropic B-factor was refined perresidue. Hydrogen bond and �/� angle restraints allowed fortorsion angle simulated annealing in CNS and PHENIX whilepreserving secondary structure geometry. The refined modelconsisted of residues 8–89, 137–190, 217–277, 284–316, and320–356 with an R-factor of 28.1% and an R-free of 33.5%.

The truncated AfGet3-ADP search model was also used formolecular replacement of the AfGet3-apo data. Density wasobserved in the weighted 2 Fo - Fc maps corresponding to theSB2 region (Fig. S2D). Two copies were located in the asym-metric unit in a relative orientation similar to the NHD dimeralthough rotated so that the SB1 and SB2 regions of the twomonomers are slightly closer (Fig. S2E). Symmetry relatedcopies form dimers similar to the arm dimer (Fig. S2F). Nofurther refinement of the apo structure was performed.

Figs. 1, 2, 4, and 5A, Figs. S2, and S4 were prepared usingPymol (11). Figs. 3 and 5 B and C were prepared using UCSFChimera (12, 13).

Growth Assays. The promoter region of GET3 was amplified fromgenomic DNA by PCR with XbaI and NcoI restriction sites andligated 5� to A. fumigatus and S. cerevisiae GET3 in the pET33bconstructs. The promoter region, promoter region with S. cer-evisiae GET3 and promoter region with A. fumigatus GET3 wereeach amplified by PCR with XbaI and NcoI restriction sitesadded and cloned into YEp352 vector (ATCC). GET3 mutantswere generated by site-directed mutagenesis. YEp352 constructswere transformed into BY4741 and BY4741 YDL100c::kanMX4cells (ATCC) for use in growth assays.

Growth defects of GET3 knockouts complemented withGET3 mutants in YEp352 on drop plates were scored were givenone of three scores: weak, moderate and strong. Growth defectsweaker than the knockout were categorized as weak if they werecloser to the WT and moderate if they were closer to theknockout. Growth defects similar to or greater than the knock-out were classified as strong. We determined the consensus scoreby taking the strongest growth defect among the different growthconditions and averaging it between the duplicates from separateexperiments.

1. Hoover DM, Lubkowski J (2002) DNAWorks: An automated method for designingoligonucleotides for PCR-based gene synthesis. Nucleic Acids Res 30:e43.

2. Leslie AG (1992) Recent changes to the MOSFLM package for processing film and imageplate data. Joint CCP4 � ESF-EAMCB Newsletter on Protein Crystallography 26.

Suloway et al. www.pnas.org/cgi/content/short/0907522106 1 of 10

3. Evans P (2006) Scaling and assessment of data quality. Acta Crystallogr D62:72–82.4. Sheldrick GM (2008) A short history of SHELX. Acta Crystallogr A64:112–122.5. Terwilliger TC, Berendzen J (1999) Automated MAD and MIR structure solution. Acta

Crystallogr D55:849–861.6. Crystallography Computing Project No 4 (1994) The CCP4 suite: Programs for protein

crystallography. Acta Crystallogr D50:760–763.7. Emsley P, Cowtan K (2004) Coot: Model-building tools for molecular graphics. Acta

Crystallogr D60:2126–2132.8. Painter J, Merritt EA (2006) Optimal description of a protein structure in terms of

multiple groups undergoing TLS motion. Acta Crystallogr D62:439–450.9. Adams PD, et al. (2002) PHENIX: Building new software for automated crystallographic

structure determination. Acta Crystallogr D58:1948–1954.

10. McCoy AJ (2007) Solving structures of protein complexes by molecular replacementwith Phaser. Acta Crystallogr D63:32–41.

11. Delano WL (1998) The PyMOL Molecular Graphics System (DeLano Scientific, Palo Alto,CA).

12. Pettersen E, et al. (2004) UCSF Chimera—a visualization system for exploratory researchand analysis. J Comput Chem 25:1605–1612.

13. Sanner MF, Olson AJ, Spehner JC (1996) Reduced surface: An efficient way to computemolecular surfaces. Biopolymers 38:305–320.

14. Larkin MA, et al. (2007) Clustal W and Clustal X version 2.0. Bioinformatics 23:2947–2948.

Suloway et al. www.pnas.org/cgi/content/short/0907522106 2 of 10

* * * * ** ** * * * * **** * * **

** ** ** *** * **** * * *** # * * ** * *** ** ** ** * *

* ** ** * ** *

Afumi -----------MSSTAVVHGDDLMEPTLQSILSQKTLRWIFVGGKGGVGKTTTSCSLAIQLAKVR--KSVLLISTDPAHNLSDAFGQKFGKEARLVDGYSNLSAMEIDPN--------GSIQDLLASGDSQG-DDPLAGLGMGNMMQDLA 128Spomb -----------MS-------FDPLPGTLENLLEQTSLKWIFVGGKGGVGKTTTSCSLAIQMSKVR--SSVLLISTDPAHNLSDAFGTKFGKDARKVPGFDNLSAMEIDPN--------LSIQEMTEQADQQNPNNPLSG-----MMQDLA 117Dreri MAASVE------DEFEDAPDVEPLEPTLKNIIEQKSLKWIFVGGKGGVGKTTCSCSLAVQLAAVR--ESVLIISTDPAHNISDAFDQKFSKVPTKVKGYDNLFAMEIDPS-------LG-VAELPDEFFE---EDNMLSMGK-KMMQEAM 130Hsapi MAAGVAGWGVEAEEFEDAPDVEPLEPTLSNIIEQRSLKWIFVGGKGGVGKTTCSCSLAVQLSKGR--ESVLIISTDPAHNISDAFDQKFSKVPTKVKGYDNLFAMEIDPS-------LG-VAELPDEFFE---EDNMLSMGK-KMMQEAM 136Xlaev MAAPVD------DEFEDAPDVEPLEPTLSNVIDQRSLRWIFVGGKGGVGKTTCSCSLAVQLSLVR--DSVLIISTDPAHNISDAFDQKFSKVPTKVRGYDNLFAMEIDPS-------LG-VAELPDEIFE---EDNMLSMGK-KMMQEAM 130Dmela ----------------MADNLEPLEPSLQNLVEQDSLKWIFVGGKGGVGKTTCSSSLAVQLSKVR--ESVLIISTDPAHNISDAFDQKFTKVPTKVNGFDNLFAMEIDPN-------AG-LNELPEEYFDG--ENEALRVSK-GVMQEMI 121Celeg -------------------MSDQLEASIKNILEQKTLKWIFVGGKGGVGKTTCSCSLAAQLSKVR--ERVLLISTDPAHNISDAFSQKFTKTPTLVEGFKNLFAMEIDSNPNGEGVEMGNIEEMLQNAAQNEGGSGGFSMGK-DFLQSFA 128Athal ------------------MAADLPEATVQNILDQESLKWVFVGGKGGVGKTTCSSILAICLASVR--SSVLIISTDPAHNLSDAFQQRFTKSPTLVQGFSNLFAMEVDPT------------VETDDMAG----TDGMDG----LFSDLA 110

Scere -------------------MDLTVEPNLHSLITSTTHKWIFVGGKGGVGKTTSSCSIAIQMALSQPNKQFLLISTDPAHNLSDAFGEKFGKDARKVTGMNNLSCMEIDPSAALKDMNDMAVSRANNNGSDGQ-GDDLGSLLQGGALADLT 130

1F48N -----------------------------MQFLQNIPPYLFFTGKGGVGKTSISCATAIRLAEQG--KRVLLVSTDPASNVGQVFSQTIGNTIQAIASVPGLSALEIDPQ---------AAAQQYRARIVDPIKGVLPDDVVSSINEQLS 1101F48C ------------------RPDIPSLSALVDDIARNEHGLIMLMGKGGVGKTTMAAAIAVRLADMG--FDVHLTTSDPAAHLSMTLNGSL N---------- NLQVSRIDPH---------EETERYRQHVLETKGKELDEAGKRLLEEDLR 419

1. ......10........20........30........40........50... .....60........70........80........90...... . 10........11 .......12......... .

1. ......10........20........30........40........50........60........70........80........90....... 10........11.. ......12........ . 1

3

.

1 .......10........20........30..... ...40........50...... 60........70........ 80........90...... .10........1

1

..3 ........32........33........34........35.... ....36........37...... ..38....... .39........40..... ..41.........1.

Afumi NQLLFPKEG----SGCEQCNARRKMQKKYLEQIEELYEDFNVVRMPLLVEEVRGKEKLEKFSEMLVHPYVPPQ-------------------------------- 340Spomb NQLLLDPN-----TTCPQCMARRKMQQKYLAQIEELYEDFHVVKVPQVPAEVRGTEALKSFSEMLVKPYVYPTSGKE---------------------------- 329Dreri NQLVFPD----NERPCKMCEARHKIQSKYLDQMEDLYEDFHIVKLPLLPHEVRGADKVNTFSKQLLEPYSPPKK------------------------------- 341Hsapi NQLVFPD----PEKPCKMCEARHKIQAKYLDQMEDLYEDFHIVKLPLLPHEVRGADKVNTFSALLLEPYKPPSAQ------------------------------ 348Xlaev NQLVFPD----PEKPCRMCEARHKIQSKYLDQMEDLYEDFHIAKLPLLPHEVRGVENVNTFSKLLLEPYKPPSGK------------------------------ 342Dmela NQLLFLQN---SHDSCSMCASRFKIQEKYLDQIADLYEDFHVTKLPLLEKEVRGPESIRSFSENLMKPYNPKGEPKE---------------------------- 336Celeg NQLLFPDTDANGTVSCRKCASRQAIQSKYLTDIDELYEDFHVVKLPLLEAEVRGGPAILQFSERMVDPEANKN-------------------------------- 342Athal NQVLYDD----EDVESKLLRARMRMQQKYLDQFYMLYDDFNITKLPLLPEEVTGVEALKAFSHKFLTPYHPTTSRSNVEELERKVHTLRLQLKTAEEELERVKSG 353

Scere NQLLFAEND--QEHNCKRCQARWKMQKKYLDQIDELYEDFHVVKMPLCAGEIRGLNNLTKFSQFLNKEYNPITDGKVIYELEDKE-------------------- 354

1F48N NGVLPKTEA----ANDTLAAAIWEREQEALANLPADLAGLPTDTLFLQPVNMVGVSALSRLLSTQPVASPSSDEYLQQ--------------------------- 3081F48C NNSLSIADTR-----SPLLRMRAQQELPQIESVKRQHASR-VALVPVLASEPTGIDKLKQLAG------------------------------------------ 583

........2 8 .......29........30........31........32.........33........34.

........2 8 .......29........30........31........32........33........34........35....

.....24.. ......25........26........27........28........29........30.......

...53..... ...54........55........56 ........57........58..

Afumi -FSIPGVDEAMSFAEVLKQVKSLS------YEVIVFDTAPTGHTLRFLQFPTVLEKALAKLSQLSSQFGPMLNSILGARGGLPGGQNIDELLQKMESLRETISEVNTQFKNPDMTTFVCVCIAEFLSLYETERMIQELTSYGIDTHAIVV 271Spomb -FTIPGIDEALAFAEILKQIKSME------FDCVIFDTAPTGHTLRFLNFPTVLEKALGKLGGLSSRFGPMIN-QMGSIMGVN--ANEQDLFGKMESMRANISEVNKQFKNPDLTTFVCVCISEFLSLYETERMIQELTSYEIDTHNIVV 257Dreri -SAFPGIDEAMSYAEVMRLVKGMN------FSVVVFDTAPTGHTLRLLNFPTIVERGLGRLMQIKNQISPFIS-QMCNMLGLGDMN-ADQLASKLEETLPVIRSVSEQFKDPEQTTFICVCIAEFLSLYETERLIQELAKCRIDTHNIIV 271Hsapi -SAFPGIDEAMSYAEVMRLVKGMN------FSVVVFDTAPTGHTLRLLNFPTIVERGLGRLMQIKNQISPFIS-QMCNMLGLGDMN-ADQLASKLEETLPVIRSVSEQFKDPEQTTFICVCIAEFLSLYETERLIQELAKCKIDTHNIIV 277Xlaev -SAFPGIDEAMSYAEVMRLVKGMN------FSVVVFDTAPTGHTLRLLNFPTIVERGLGRLMQIKNQISPFIS-QMCNMLGLGDMN-ADQLASKLEETLPVIRSVSEQFKDPEQTTFICVCIAEFLSLYETERLIQELAKCSIDTHNIIV 271Dmela -NALPGIDEAMSYAEVMKLVKGMN------FSVVVFDTAPTGHTLRLIAFPQVVEKGLGKLLRLKMKVAPLLS-QFVSMLGMADVN-ADTLSQKLDDMLRVITQVNEQFKNPDQTTFVCVCIAEFFSLYETERLVQELTKCGIDVHNIIV 262Celeg -GGLPGIDEAMSFGEMIKLIDSLD------FDVVVFDTAPTGHTLRLLQFPTLLEKVFTKILSLQGMFGPMMN-QFGGMFGMGGGS-MNEMIEKMTTTLESVKKMNAQFKDPNCTTFVCVCIAEFLSLYETERLIQELSKQGIDTHNIIV 269Athal -NAIPGIDEAMSFAEMLKLVQTMD------YATIVFDTAPTGHTLRLLQFPATLEKGLSKLMSLKSRFGGLMT-QMSRMFGMEDEFGEDALLGRLEGLKDVIEQVNRQFKDPDMTTFVCVCIPEFLSLYETERLVQELAKFEIDTHNIII 252

Scere -GSIPGIDEALSFMEVMKHIKRQEQGEGETFDTVIFDTAPTGHTLRFLQLPNTLSKLLEKFGEITNKLGPMLN----SFMGAGNVD----ISGKLNELKANVETIRQQFTDPDLTTFVCVCISEFLSLYETERLIQELISYDMDVNSIIV 271

1F48N GACTTEIAAFDEFTGLLTDASLLTR-----FDHIIFDTAPTGHTIRLLQLPGAWSSFIDSNPEGASCLGPMAG---------------------LEKQREQYAYAVEALSDPKRTRLVLVARLQKSTLQEVARTHLELAAIGLKNQYLVI 2341F48C SPCTEEIAVFQAFSRVIREAGK RF--------- VVMDTAPTGHTLLLLDATGAYH REIAKKMG------ ----------------------------EKG HFTTPMMLLQDPERTKVLLVTLPETTPVLEAANLQADLERAGIHPWGWII 526

.......16........17........18........19........20........21........22........23........24........25........26........2713........14........15. .

...... 1 .. ......22........23........24........25........26........27.......14........15... .

... ....14....... 15........16........17....... .18........19........20........21........22........23.....3........45.....4 .46.... ....47.... ...48........49........50........51........52.......4

........12........142........43........

P-loop SwI

SwII

A-loop

Substrate Binding Loop 2

Substrate Binding Loop 1

..

.

16........17........18........19........20 .......2

*

H1 S1 H2 S2 H3 S3 S4 H4

H5 S5 S6 S7

S8

H6 H7 H8

H9 H10 H11

♦

♦♦

♦

♦ ♦

Fig. S1. Sequence alignment of Get3 homologues. Sequences were aligned using the program ClustalX (14). Residue coloring is based on the program output(colored based on amino acid type). The species in order are Scere (S. cerevisiae), Afumi (A. fumigatus), Spomb (Schizosaccharomyces pombe), Dreri (Danio rerio),Xlaev (Xenopus laevis), Dmela (Drosophila melongaster), Celeg (Caenorhabditis elegans), Athal (Arabidopsis thaliana), and 1F48N and C (the N and C-terminalsequence of E. coli Arsa). Numbering, from top to bottom, is based on Sc, Af, and EcArsA with disordered residues in ScGet3, AfGet3-ADP and PDBID 1f48 coloredin red. Secondary structure for AfGet3-ADP is shown on top, along with numbering as in Fig. S2A, and the N terminus of 1f48 is shown on the bottom, �-helicesas red rectangles and �-sheets as yellow arrows. Structural elements discussed in the text are in boxes above the alignment colored as in Fig. 1B. Below thealignment, gray bars show degree of conservation at a given position based on Get3 sequences. Mutations described in the text are indicated by asterisks forSc (*) and pound for Af (#) colored strong (red), moderate (orange) and weak (yellow) for LOF phenotypes and cyan for mutants that did not display a phenotypein our assays. Gray diamonds (filled diamonds) represent residues in ArsA that coordinate metal binding.

Suloway et al. www.pnas.org/cgi/content/short/0907522106 3 of 10

Fig. S2. The ScGet3 (A–C) and AfGet3 (D–F) crystal forms. (A) The ScGet3-apo monomer with labeling and color similar to Fig. 1A. The modeled zinc (slate) andputative Ni (green) are shown as spheres with coordinating side-chains from the monomer as sticks. (B) A composite omit map contoured at 1� with the proteinmodel in sticks colored as in A. (C) Opposing crystallographic ScGet3-apo dimer showing the oriented SB1/2 loops that interact in the crystal lattice. One monomeris colored by motifs and the other is in salmon. The SB1/2 loops are generally disordered and not clearly interpretable in our structure. (D) A monomer of theAfGet3-apo dimer found by molecular replacement. Only the portions of the AfGet3-ADP structure used as a search model are shown. Density is a 1.2� mapcalculated using phases from the molecular replacement solution. Additional density for the truncated SB2 can be seen. (E) The NHD dimer interface is slightlydifferent in the apo form. We have modeled this movement using the AfGet3-ADP form. That dimer is in gray, and the movement in the apo form is modeledin purple. (F) The arm dimer seen in the crystal packing for the apo form. Loops are poorly ordered and the ADP form is modeled as transparent helices for clarity.

Suloway et al. www.pnas.org/cgi/content/short/0907522106 4 of 10

Fig. S3. Mutant rescue experiments. The full panel of yeast mutants similar to Fig. 4A. Subpanel letters indicates a group of mutants plated together, one platefor each growth condition. Every plate contained the parent strain, knockout strain and plasmid complemented transformant as functional controls. Growthdefects were tested on SC-Ura supplemented with 2 mM CuSO4 at 30 °C and 37 °C, 200 mM hydroxyurea at 30 °C, and SC-Ura at 40 °C. SC-Ura at 30 °C was usedas a growth condition control. Experiments not performed are highlighted by hashed rectangles. Our interpretation of these results is indicated in Table S1.

Suloway et al. www.pnas.org/cgi/content/short/0907522106 5 of 10

Fig. 3 Continued

Suloway et al. www.pnas.org/cgi/content/short/0907522106 6 of 10

Fig. S4. Phenotype regions described in the text. (A–D) On the left are zoomed in views of the regions described in the text. Residues are shown as sticks andcolored as in 4B and some of the phenotype residues are labeled. Hydrogen-bonds are shown as gray dashes. On the left is a ribbon diagram of the full dimerin the same orientation with a few of the residues drawn as sticks for orientation. The dashed boxes indicated the approximate region shown on the right.

Suloway et al. www.pnas.org/cgi/content/short/0907522106 7 of 10

ADP or

apoATP

ADP and Pi

ATP and TA

(1) Hexamer resting state

?

(2) Open NHD dimer (3) Closed NHD dimer

Fig. S5. Model for tail-anchored recognition. The discussion suggests the following model. In a resting/open form Get3 may alternate between (1) a hexamerand (2) a NHD dimer. Binding of ATP would lead to a conformational change that would facilitate formation of a stable complex with a TA-protein. Binding ofboth would result in a complex (3) primed for ATP hydrolysis. Additional conformational changes or partner binding would stimulate ATP hydrolysis. Loss of theinorganic phosphate (Pi) would lead to release of the TA-protein and a return to the open form.

Suloway et al. www.pnas.org/cgi/content/short/0907522106 8 of 10

Table S1. Crystallographic statistics

AfGet3-ADP AfGet3-ADP MAD ScGet3-apo AfGet3-apo

Data collectionSpace group P212121 P212121 H32 P4232Cell dimensionsa, b, c, Å 67.76, 154.78, 242.88 68.72, 155.51, 242.72 115.32, 115.32, 281.11 181.02, 181.02, 181.02�, �, �,° 90, 90, 90 90, 90, 90 90, 90, 120 90, 90, 90Resolution, Å 50–3.2(3.37–3.2) 50–4.5(4.74–4.5) 50–3.7(3.9–3.7) 50–7.5(7.91–7.5)

Peak Inflection RemoteWavelength 1.00000 0.97941 0.97954 0.91837 1.00000 1.00462Rmerge, % 12.4(66.7) 8.0(13.4) 7.7(12.0) 7.7(12.5) 9.9(62.7) 6.4(41.3)I/�I 10.4(2.5) 18.0(12.7) 19(13.5) 18.6(13.5) 8.9(2.7) 16.9(3.9)Completeness, % 100.0(100.0) 100.0(100.0) 100.0(100.0) 100.0(100.0) 99.9(100.0) 97.1(97.9)Redundancy 4.9(5.0) 7.3(7.5) 7.3(7.5) 7.3(7.5) 5.9(6.0) 5.0(5.2)RefinementResolution, Å 50–3.2 50–3.7No. reflections 43100 7936Rwork/Rfree, % 21.2/25.13 28.1/33.5No. atomsProtein 14082 2184Ligand/ion 156 2B-factorsProtein 86 182Ligands/Ions 71 120Bond RMSDLengths, Å 0.01 0.007Angles, ° 1.362 1.147

Values in parentheses are for the highest-resolution shell.

Suloway et al. www.pnas.org/cgi/content/short/0907522106 9 of 10

Table S2. Summary of mutants

Numbering Phenotypes

FigureS3

Numbering Phenotypes

Fig. S3ScGet3 AfGet32 mM

Cu 30°C2 mM

Cu 37°C200 mMHU 30°C 40°C overall ScGet3 AfGet3

2 mMCu 30°C

2 mMCu 37°C

200 mMHU 30°C 40°C overall

E6A E14 W W C and D K185A K177 B and DT17A T25 W W W E and I L186S A178 F and JK19A R27 E and I L187S L179 C and DK26A K34 S S S S S F and I K189A K181 B and DK26R K34 M M M S S F and I F190S L182 F and JG30R G38 S S S S S C and D I193S L185 F and JH60A H66 W � F and I M200S M192 B and DD64S D70 S S W M S E and I L201S L193 W W B and EK69A K75 � W M �/M C and D N202A N194 B and EK72A K78 W F and I G206P G202 F and JD73A D79 W F and I I212S L212 F and JR75A R81 W W W W W A and D K215A K215 B and EE87A E93 S M W W S E and I L216S M216 F and JD89A D95 E and I L219S L219 W M W G and JV102S I100 W A and D E245A R245 W W W E and JA105S L103 W W F and I F246A F246 S S S S S E and JL117S L115 W W W W F and I L247S L247 M M M S S G and JL120S L118 F and I S248A S248 W M W S M G and JG123P G121 F and I Y250A Y250 S S S S S G and JA125S M123 W W W W A and D E251A E251 S S S S S G and JL126S M124 C and D E253A E253 S S S S S G and JD128A D126 W W A and D R254A R254 W E and JL129S L127 F and I Q257A Q257 W W W E and JI133S I131 W F and I E258A E258 W M M M GJ4I136S V134 W W W W A and D D265A D265 S S M S S B and EI136D V134 W W W F and I C285T-C288T C283T-C286T S S S S S C and DD137A D135 W W W W A and D R291A R289 S S S S S B and EE138A E136 W W W W E and I M294A M292 W W W W E and JL140S M138 W W M W A and D K297A K295 W W W W W E and JS141A S139 W W W M M A and D Y298A Y296 M M S S S E and JM143S A141 W W W W W A and D D300A E298 G and HE144S E142 W C and D E320A E318 W M M M E and HV145S V143 F and I Y338A Y336 W W W W W B and EH172A H164 M M W W M F and IR175A R167 M W W W W E and I G30 G38R M S M M M C and DL183S L175 F and J - R200A M M M G and H

Phenotypes in bold represent mutants that had been tested in previous studies. Mutants that could rescue knockout are unmarked. Mutants that could notfully rescue are shown by minus signs are graded by severity of growth defect: weak (W), moderate (M) and strong (S). Mutants that showed apparent gain offunction are shown by plus signs (�).

Suloway et al. www.pnas.org/cgi/content/short/0907522106 10 of 10