1

THE ROLE OF COMPLEX FORMATION BETWEEN THEESCHERICHIA COLI HYDROGENASE ACCESSORY FACTORS HYPB

AND SLYD †Michael R. Leach, Jie Wei Zhang, and Deborah B. Zamble

From the Department of Chemistry, University of Toronto, Toronto, Ontario, Canada M5S 3H6

Running Title: SlyD stimulates metal release from HypBAddress Correspondence to: Deborah B. Zamble, Department of Chemistry, University of Toronto, 80St. George St., Toronto, ON M5S 3H6, Phone (416) 978-3568. E-mail: dzamble@chem.utoronto.ca

The Escherichia coli protein SlyD is amember of the FK-506-binding protein (FKBP)family of peptidyl-prolyl isomerases (PPIases).In addition to its PPIase domain, SlyD iscomposed of a molecular chaperone domainand a C-terminal tail rich in potential metal-binding residues. SlyD interacts with the[NiFe]-hydrogenase accessory protein HypBand contributes to nickel insertion duringbiosynthesis of the hydrogenase metallocenter.This report examines the HypB-SlyD complexand its significance in hydrogenase activation.Protein variants were prepared to delineate theinterface between HypB and SlyD. Complexformation requires the HypB linker regionlocated between the high-affinity N-terminalNi(II) site and the GTPase domain. In the caseof SlyD, the deletion of a short loop in thechaperone domain abrogates the interactionwith HypB. Mutations in either protein thatdisrupt complex formation in vitro also resultin deficient hydrogenase production in vivo,indicating that the contact between HypB andSlyD is important for hydrogenase maturation.Surprisingly, SlyD stimulates release of nickelfrom the high-affinity Ni(II)-binding site ofHypB, an activity that is also disrupted bymutations that affect complex formation.Furthermore, a SlyD truncation lacking the C-terminal metal-binding tail still interacts withHypB but is deficient in stimulating metalrelease and is not functional in vivo. Theseresults suggest that SlyD could activate metalrelease from HypB during metallation of the[NiFe] hydrogenase.

The assembly of the [NiFe] metallocenterof Escherichia coli (E. coli) hydrogenase 3requires the participation of proteins encoded

by the hydrogenase pleiotropy (hyp)1 geneshypABCDEF (reviewed in (1-3)). HypA andHypC are replaced by the homologous HybFand HybG proteins, respectively, for theassembly of hydrogenases 1 and 2 (1,2).HypC, HypD, HypE and HypF participate inthe biosynthesis and delivery of theFe(CN)2(CO) cluster to the hydrogenaseprecursor protein (4-6). The subsequentincorporation of nickel (7,8) requires theGTPase HypB and HypA. These proteinswere initially implicated in the nickel insertionstep by genetic studies in which thehydrogenase deficiency resulting fromchromosomal mutations was at least partiallyrestored by growing the bacteria in excessnickel (9-13). E. coli HypB binds one nickelion with a Kd in the picomolar range to thecysteines in the N-terminal CxxCGC motif(referred to as the ‘high-affinity site’, seeFigure 1 for domain architecture) (14). Inaddition, both HypB and HypA bind a nickelion with micromolar affinity (14-17), HypA ata site that includes the conserved secondresidue His2 (15,16), and HypB to severalconserved amino acids in the GTPase domain(referred to as the ‘low-affinity site’) (14,18).Whether one or a combination of these metalsites serves as a source of nickel for thehydrogenase enzyme has not yet beendetermined.

Upon searching for addit ionalhydrogenase biosynthetic factors in E. coli, aprotein called SlyD was identified in acomplex with HypB and shown to play a rolein hydrogenase production (19). SlyD is a

http://www.jbc.org/cgi/doi/10.1074/jbc.M610834200The latest version is at JBC Papers in Press. Published on April 10, 2007 as Manuscript M610834200

Copyright 2007 by The American Society for Biochemistry and Molecular Biology, Inc.

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

2

member of the FK-506-binding protein(FKBP) family of peptidyl-prolyl isomerases(PPIases) (20,21). In general, PPIases arecapable of assisting in various protein foldingprocesses by catalyzing the cis-transisomerization of a prolyl amide bond, but inmany cases it is not clear that this activity isrequired for the physiological function of theprotein (22,23). SlyD, like many FKBPPPIases including the E. coli ribosome-associated trigger factor (24) and periplasmicFkpA (25), also exhibits general molecularchaperone activity (26). SlyD shares with asubset of FKBPs an additional domain, termedIF (insert in the flap), which has beenstructurally characterized for an archaealFKBP, MtFKBP17 from Methanococcusthermolithotrophicus, and may allow theseproteins to bind unfolded or extendedpolypeptide chains (27,28). Although theypossess many attributes suitable for ahousekeeping chaperone function, somePPIases bind only to a limited set of clientproteins and appear to play a regulatory role inspecific biochemical pathways (22,23).

One unusual feature of SlyD is that it has aC-terminal 50-residue metal-binding domain.This domain is rich in the metal-bindingamino acids, containing 15 histidines, 6cysteines, and 7 aspartate/glutamate residues(see Figure 1 for domain architecture), and isrequired for metal-dependent inhibition of thePPIase activity (29). An E. coli strain with adeletion in the slyD gene exhibits reducednickel accumulation as well as lowhydrogenase activity that can be fully restoredby the addition of excess nickel to the growthmedia (19), suggesting that SlyD contributesto the insertion of nickel into apo-hydrogenase. Unlike many other bacterialHypB proteins, E. coli HypB lacks apolyhistidine stretch that can bind multiplenickel ions and serve in nickel storage (30-33). Given that SlyD can bind multiple metalions (20,29), it is possible that SlyD hasassumed this storage function (19).

To investigate the role of SlyD in themetallation of apo-hydrogenase we examinedthe SlyD-HypB complex in more detail. Thesite of interaction was localized in bothproteins, and mutants of either HypB or SlyDthat fail to form the complex in vitro aredefective in activating hydrogenase in vivo.Furthermore, when SlyD binds to HypB itactivates metal release from the N-terminalhigh-affinity nickel-binding site. A SlyDvariant lacking the C-terminal tail is capableof interacting with HypB but does notcontribute to hydrogenase biosynthesis in vivoor stimulate metal release from HypB in vitro.These data support a role beyond nickelstorage for SlyD in triggering metal releasefrom HypB and promoting nickel insertioninto the hydrogenase precursor protein.

Experimental Procedures

Materials - Restriction endonucleases, T4DNA ligase and calf intestine phosphatase(CIP) were obtained from New EnglandBiolabs. Pfu DNA polymerase was purchasedfrom Stratagene. Chromatography mediawere from GE Healthcare Bio-Sciences.Isopropyl-β-D-thiogalactoside (IPTG), Tris-(2-carboxyethyl)phosphine (TCEP), ampicillinand kanamycin were purchased from BioShop(Toronto). All other reagents were analyticalgrade from Sigma. Primers (Table 1) werepurchased from Sigma Genosys. Solutionswere prepared with Milli-Q water, 18.2 MΩ-cm resistance (Millipore), and the pH valuesof the buffers were adjusted with HCl orNaOH.HypB and SlyD Expression Vectors andMutants - The generation of HypB-pET,Gdomain-pET, and SlyD-pET in the pET24bvector (Novagen) and pBAD-SlyD in thepBAD24 vector (American Type CultureCollection, ATCC) was described previously(14,19). To generate pET-SlyD(1-146), afragment of slyD coding for residues 1-146was amplified from DH5α E. coli by using the

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

3

D(1-146) primer set (Table 1). Afterpurification with the QIAquick PCRpurification kit (Qiagen) the PCR product wasdigested with the restriction enzymes NdeIand XhoI. The digested PCR product wasisolated from agarose gel pieces by using aQIAquick kit (Qiagen) and ligated into thepET24b vector digested with NdeI and XhoIand dephosphorylated with CIP. Ligationproducts were transformed into XL-2 Blue E.coli (Stratagene). For in vivo experimentshypB was amplified from pET24b-HypBusing the BpBAD primer set and a fragmentof s l yD coding for residues 1-146 wasamplified from DH5α E. coli using the D(1-146)pBAD primer set. PCR products weretreated as described above, except they weredigested with the restriction enzymes NheIand XbaI and ligated into the pBAD24 vectordigested with the same enzymes, to generatepBAD-HypB and pBAD-SlyD(1-146),respectively. To generate a construct codingfor a fusion between the maltose-bindingprotein (MBP) and the C-terminal tail ofSlyD, a fragment of slyD coding for residues140-196 was amplified with the primer setD(140-196). The purified PCR product wastreated as described above, digested with NdeIand XhoI, and ligated into the vector pIADL16((34); generously provided by Prof. C. T.Walsh, Harvard Medical School) digestedwith the same restriction enzymes and treatedwith CIP. Plasmids coding for amino acidvariants and deletion variants were generatedby Quick-Change PCR mutagenesis(Stratagene) using Pfu polymerase and pET-HypB, pBAD-HypB, pET-SlyD and pBAD-SlyD, as templates. The primers used toprepare HypB(PP29,32SS), HypB(28d36),HypB(11d76), HypB(19d76), and SlyD(Δflap)(producing a deletion of residues 107-111) areshown in Table 1. Parent plasmids weredigested with DpnI before transforming thereaction mix into competent cells. For routinehandling, plasmids were transformed into XL-2 Blue E. coli and isolated by using the

Qiagen plasmid mini- or midi-prep kits. Allplasmids were sequenced in the forward andreverse directions to verify the clonedsequences and mutations (ACGT, Toronto).Protein purification - HypB and HypBmutants were expressed in BL21(DE3) E. colicells grown in media containing 1 mM Ni(II)(14). The proteins were purified by sequentialDEAE, HiTrapQ, and Superdex S-200columns as previously described (14), andstored at -80°C in the Superdex buffer (25mM Hepes, pH 7.6, 200 mM NaCl, 1 mMTCEP). To determine the oligomeric state ofHypB, the purified protein was run on an S-200 gel filtration column calibrated with β-amylase (200 kDa), BSA (66.5 kDa), chickenovalbumin (45 kDa) and carbonic anhydrase(29 kDa) as standards. HypB (31.6 kDa) elutesat a volume consistent with a MW of 65 kDa,demonstrating that it is a dimer under theseconditions, in contrast to our previouslyreported experiments using a S-75 gelfiltration column (14), but in agreement withother reports (15,18,35). HypB and HypBvariants contained ≈ 0.95 equivalents Ni(II) aspurified.

SlyD variants were expressed in a∆slyD BL21(DE3) strain of E. coli prepared asdescribed for ∆ s l y D in other geneticbackgrounds (19). SlyD and SlyD variantswere purified by using Ni-NTAchromatography followed by anion exchangeon a MonoQ column and gel filtration on aSuperdex S-200 essentially as previouslydescribed (19). As SlyD(1-146) lacks thepolyhistidine tail the initial Ni-NTA columnwas replaced with a DEAE column.

Protein concentrations were estimatedusing their ε280 values as predicted by theExPasy Protparam program for the fullyreduced proteins (36). Protein molecularweights were determined by ESI-MS andcompared to the predicted values. The datawere consistent with the loss of the N-terminalMet for all HypB variants, except the GTPasedomain: HypB(PP29,32SS) 31,412.0 Da

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

4

(predicted 31,413.7 Da), HypB(28d36)30,554.0 Da (predicted 30,554.8 Da),HypB(19d76) 25,186.0 Da (predicted25,186.8) and HypB(11d76) 24,250.0 Da(predicted 24,252.9 Da.). ESI-MS data forHypB and the GTPase domain were asreported (14). Data for SlyD and its variantswere consistent with retention of the N-terminal methionine: SlyD 20,853.0 Da(predicted 20,852.8 Da), SlyD(1-146)15,833.2 Da (predicted 15, 833.6 Da) andSlyD(Δflap) 20,338.0 Da (predicted 20,339.2Da). The observed molecular weight of theMBP-SlyD(140-196) fusion protein wasconsistent with removal of the N-terminalMet, 48,796.0 Da (predicted 48,797.4 Da).Iodoacetamide modification of SlyD - PurifiedSlyD (1 mL of 500 µM) was treated with 2mM TCEP for 20 min then with 50 mMiodoacetamide for 30 min with stirring at RT.The sample was then dialyzed against twochanges of buffer (3 L of 10 mM Hepes, 100mM NaCl, pH 7.6) to remove the modificationreagent and analyzed by ESI-MS before use.Modified SlyD has a mass of 21,196.0 Daconsistent with all six cysteines having acarbamidomethyl modification (predicted21,195.2 Da).Circular Dichroism - CD spectra wererecorded on a Jasco J-710 spectropolarimeterwith an optical path length of 1 mm. Theprotein concentrations were 10-30 µM in 100mM potassium phosphate buffer, pH 7.0.GTPase Activity - To assay GTPase activity,released phosphate was detected by usingMalachite Green in the presence ofammonium molybdate based on an adaptationof a published method as previously described(14,37).PPIase Activity – A protease-free assay wasused to measure the PPIase activity of SlyDand the variants (38). The substrate, succinyl-Ala-Ala-Pro-Phe-nitroanilide (BachemBioscience), was dissolved in trifluoroethanolthat had been dried over sieves and 0.47 MLiCl. The reactions contained 35 mM Hepes,

pH 7.6, and 1 µM protein and were incubatedat 10 °C prior to the addition of 71 µMsubstrate. Isomerization was monitored at 330nm on a Cintra 40 spectrophotometer and fitto a single exponential decay. Theuncatalyzed rate was also measured and thesecond-order rate constant was calculated:kcat/KM = (kobs-kuncat)/[enzyme].Metal binding and EGTA competition - For allHypB variants we initially investigatedwhether the purified proteins had an electronicabsorption band at 320 nm, which isdiagnostic for bound nickel, and then preparedapo-protein from each of these proteins to testwhether they bind stoichiometric nickel withan affinity comparable to wild-type HypB(Table 2; (14)). Purified HypB variants wereincubated with 20 mM EDTA and 2 mMTCEP in an anaerobic glove box for 72 hours.The proteins were gel filtered twice throughPD-10 columns equilibrated with 25 mMHepes, pH 7.5, 100 mM NaCl. To test theirreduction state, all apo-proteins were treatedwith DTNB after dilution into 6 Mguanidinium hydrochloride and theabsorbance at 412 nm was compared to a β-mercaptoethanol standard curve. In all casesthe reduction state was > 95%. The apo-proteins were incubated with Ni(II) in theabsence or presence of 5 mM EGTA overnightat 4 ºC and samples were removed from theglove box to measure the UV/visibleabsorbance spectra. The data were treated aspreviously described (14).Metal release assay - HypB, as purified, wasdiluted to a final protein concentration of 5µM into 25 mM Hepes, pH 7.5, 200 mM NaCl(buffer A) containing 100 µM PAR. Therelease of metal was monitored every fiveminutes by an increase in the absorbance at500 nm due to the metal-PAR2 complex (39).To determine total metal in the sample (Amax),an aliquot of the same sample was treated with100 µM PMB and the data were converted to% metal bound (100*[1-(A/Amax)]) at a giventime point. Data of koff vs. SlyD were fit to a

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

5

saturation equation to estimate the maximalmetal release rate. The concentration of SlyDwas then adjusted for the amount bound toHypB, assuming 1:1 binding, to yield the freeSlyD concentration. These data were then fitto a saturation equation: koff =koff

Max([SlyD])/(Kd + [SlyD]).Molecular chaperone assays - The ability tosuppress aggregation of citrate synthase (CS)was used as an assay of general molecularchaperone activity (40). Chemicallydenatured, reduced CS (CSden/red) wasprepared by desalting an ammonium sulfatepreparation (Sigma) that was first diluted 4-fold with 50 mM Tris, pH 8.0 by using a PD-10 column pre-equilibrated with 50 mM Tris,pH 8.0. CS-containing fractions were pooled,concentrated, and adjusted to 40 µM based onthe monomeric MW of 48,969 Da and thepublished extinction coefficient at 280 nm of1.78 AU for a 1 mg/mL solution (40). The CSstock was chemically denatured and reducedby diluting the protein 4-fold in 6 Mguanidinium hydrochloride and 25 mM DTTand incubating at RT for 2 hours. For eachaggregation assay 5 µl of CSden/red was dilutedwith 500 µ l of buffer A (to a finalconcentration of 0.1 µM) in the presence orabsence of SlyD or a SlyD variant at a 20:1ratio (2 µM). Aggregation was monitored on afluorimeter (Jobin Yvon) with λex and λ em

both set at 500 nm and slit widths set at 2 nm.To assay for CS reactivation, 2 µl of

CSden/red was diluted in 100 µl of buffer A (to afinal concentration of 0.2 µM) in the presenceor absence of SlyD or a SlyD variant at a 20:1ratio (4 µM). After two hours, a 2 µl aliquotwas removed and assayed for CS activity bydiluting into 98 µl of reaction mix (93 µl 50mM Tris, 2 mM EDTA, pH 8.0, 3 µl of 5 mMacetyl-CoA, 1 µl of 10 mM oxaloacetic acid,and 1 µl of 10 mM DTNB). The reaction wasmonitored at 412 nm due to the reaction ofCoASH with DTNB and the initial rate wascompared to that of a control reaction

performed with an equivalent amount ofuntreated CS, defined as 100% activity.Chemical crosslinking - Purified HypB (8 – 10µM), SlyD (10 - 40 µM) or the appropriatevariant proteins were incubated overnight at 4ºC in 10 µl of buffer A. 1-Ethyl-3-[3-dimethylamino-propyl]carbodiimidehydrochloride (EDC) was then added to aconcentration of 5 mM and the samplesfurther incubated at room temperature for 1hour. Samples were then subjected to SDS-PAGE on 12.5% polyacrylamide gels andstained with Coomassie Blue.Cell Strains – The ∆slyD mutation in a DY330background (41) was described previously(19). The ∆s l yD cells in the M C 4 1 0 0background (42), used for the experimentsshown in Figure 4C, were prepared in thesame manner. The ∆hypB strain (DHP-B) inthe MC4100 background was generouslyprovided by Prof. A. Böck (University ofMunich, Germany (43)). Protein expressionfrom the pBAD plasmids was induced with 1µM and 100 µM arabinose in the MC4100 andDY330 strains, respectively, unless otherwisenoted.Growth Condition and Preparation of CrudeCell Extracts - Cells were grown aerobicallyin LB media overnight prior to anaerobicgrowth in TGYEP (44) supplemented with 1µM sodium molybdate, 1 µM sodium selenite,0.8% glycerol, and 15 mM sodium fumarateas well as arabinose and NiSO4 at theindicated concentrations. Cell extracts wereprepared as previously described and testedfor hydrogenase activity under a H2(g)-containing a tmosphere by usingbenzylviologen as an electron acceptoraccording to the method of Ballantine andBoxer (19,45).Western Blotting - Crude cell extracts wereseparated by SDS-PAGE, transferred onto aPVDF membrane and incubated with either a1:1,000 dilution of anti-HypB (from Prof. A.Böck) or anti-SlyD polyclonal antibodies(prepared by immunization of rabbits with

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

6

purified SlyD, Division of ComparativeMedicine, University of Toronto), followed byincubation with a goat anti-rabbit-HRPsecondary antibody (BioRad). Enhancedchemiluminescence (Pierce) was used fordetection.

RESULTS

SlyD, but not SlyD(Δflap), suppresses theaggregation of citrate synthase (CS). ManyPPIases exhibit molecular chaperone activitythat is distinct from the PPIase activity, butsome are chaperones of only a limited set ofclient proteins (reviewed in (23)). Todetermine if SlyD acts as a general molecularchaperone, experiments were performed toexamine if full-length SlyD is capable ofpreventing the aggregation of chemicallydenatured, reduced CS. As shown in Figure2A, SlyD almost completely suppressesaggregation at a ratio of 20:1. The same invitro assay was recently used to demonstratethat a truncated SlyD lacking part of the C-terminal tail, SlyD(1-165), acts as a molecularchaperone (26). The SlyD variant lacking thecomplete C-terminal tail, SlyD(1-146), has asimilar activity as that reported for SlyD(1-165) (26), and is partially impaired incomparison with full-length SlyD (Figure 2B).This result indicates that the tail does play alimited role in the molecular chaperoneactivity.

SlyD, like some of the other FKBPs,possesses an additional IF domain that maybind to substrates with an extendedconformation (27,28). To examine whether theIF domain has a role in the ability of SlyD toact as a molecular chaperone, we engineered avariant lacking residues 107-111 (referred toas SlyD(∆flap)). This deletion, based onsequence similarity with the structurallycharacterized MtFKBP17 (27), is predicted toremove a short loop near the putativepolypeptide-binding site in the IF domain.SlyD(∆flap) retains 50% of wild-type PPIase

activity and has a CD spectrum similar to thatof wild-type SlyD (data not shown), ruling outthe possibility that this deletion results in lossof structure. However, a test of SlyD(∆flap)in the aggregation assay with chemicallydenatured CS revealed that this mutant isseverely impaired in its ability to preventaggregation compared to wild-type SlyD(Figure 2B).

SlyD, but not SlyD(Δflap), can promote CSreactivation. Upon dilution from denaturant,CS can refold to form active enzyme, but thecompeting aggregation pathway limits theyield of active CS. In the absence of addedchaperone, unfolded CS undergoes a time-dependent refolding to a maximum of ≈ 30%initial activity (40). To confirm SlyD’schaperone capabilities, experiments wereperformed to determine if the protein wouldinfluence productive folding of CS. Therefolding yield of CS is significantly enhancedin the presence of a 20-fold excess SlyD,increasing to 68% reactivation from the 32%spontaneous refolding (Figure 2C), a level ofenhancement equivalent to that observed inthe presence of a 1:1 ratio ofGroEL/GroES+ATP (40). In correspondencewith the aggregation suppression experiments,SlyD(1-146) promotes refolding of CS to alesser degree than full-length SlyD, andSlyD(Δflap) has an effect that is only slightlyabove basal levels of spontaneousreactivation.

SlyD(Δflap) does not crosslink with HypBin vitro or activate hydrogenase in vivo. Asthe IF domain is a potential mediator ofprotein-protein interactions, mutations in thisdomain that result in compromised molecularchaperone activity might also affectinteractions with specific partner proteins. Thefact that SlyD(Δ flap) failed to suppressaggregation or reactivate CS suggested thatthis variant did not interact with the foldingintermediates. To test if this deletion alsodisrupts the complex between SlyD and thehydrogenase accessory protein HypB, a

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

7

chemical crosslinking assay was used. Asshown previously, the addition of EDC, awater soluble, zero-length crosslinker, to pre-incubated HypB and SlyD results in a robustcrosslink that migrates at the MW of theheterodimer on a denaturing polyacrylamidegel (Figure 3A and (19)). A crosslinked dimeris, however, not observed between HypB andSlyD(Δflap), indicating that this loop from theSlyD IF domain is required for complexformation with HypB. In agreement with thisconclusion, removal of the C-terminal tail ofSlyD (SlyD(1-146)) did not prevent theinteraction with HypB (Figure 3B).

To determine whether one or both of theproperties of SlyD disrupted in theSlyD(Δ flap) variant, folding chaperoneactivity and HypB binding, are required for itsfunction in hydrogenase biosynthesis, ΔslyDcells were transformed with pBAD24, pBAD-SlyD or pBAD-SlyD(Δflap). Thehydrogenase activity in the cell lysates of thetransformed cells was measured by usingbenzylviologen as a chromophoric electronacceptor and compared to the activity inextracts from wild-type cells. The ΔslyD cellextracts had an activity that was only about25% of the parent cells as previously reported(19), demonstrating that SlyD plays a role inthe activation of hydrogenase but is notabsolutely essential (Figure 4A). Wild-typelevels of activity can be restored by thetransformation of ΔslyD cells with pBAD-SlyD, but no restoration is achieved bytransformation with pBAD-SlyD(Δflap),indicating that either the molecular chaperoneactivity of SlyD or its ability to bind HypB isrequired for its function in the hydrogenaseactivation pathway. Western analysis of thecell extracts probed with an anti-SlyDpolyclonal antibody confirmed that theexpression of both SlyD and SlyD(Δflap) fromthe pBAD vector was at levels comparable toSlyD expression in DY330 control cells (datanot shown). The results for SlyD(1-146) in thehydrogenase assay will be discussed below.

HypB linker region participates incomplex formation with SlyD. To localize thesite of interaction with SlyD on HypB, achemical crosslinking experiment was firstperformed with SlyD and the isolated GTPasedomain of HypB (residues 77-290). Acrosslink between the HypB GTPase domainand SlyD was not observed, and the additionof an excess of the GTPase domain did notresult in a reduction in the amount of crosslinkbetween full-length HypB and SlyD (data notshown). This result suggested that SlyD bindsto the N-terminal region of HypB.

As a PPIase, SlyD catalyzes isomerizationin vitro on model substrates with ahydrophobic residue preceding the targetproline (29) and in the sequence linking the N-terminal CxxCGC motif and the GTPasedomain there is a proline-containing sequence,AP29FAP32AARP (see Figure 1). To assesswhether this sequence is involved in the SlyD-HypB interaction, serine was substituted forthe first two of the three prolines to yieldHypB(PP29,32SS). This mutation resulted ina reduced amount of crosslinked complex withSlyD (Figure 3C) supporting the hypothesisthat SlyD interacts with HypB at least partiallyvia this proline-containing motif. A similarresult was observed for a HypB(28d36)mutant, which has the sequence betweenresidues 28 and 36 deleted. Furthermore,proteins prepared with more substantial linkerdeletions, HypB(19d76) and HypB(11d76), donot produce any crosslink with SlyD under ourstandard conditions, indicating that there is anadditional site of interaction beyond theAP29FAP32AARP sequence within this linkerregion (Figure 3C and data not shown). All ofthe HypB variants bound stoichiometric nickelwith high affinity (Table 2), were as active aswild-type HypB in the GTPase assay withinexperimental error, and exhibited secondarystructure content similar to HypB(HypB(PP29,32SS) and HypB(28d36)) or theHypB GTPase domain (HypB(19d76) and

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

8

HypB(11d76)) as assessed by CDspectroscopy (data not shown).

Impaired SlyD-HypB complex formationin vitro correlates with decreasedhydrogenase activation in vivo. To assess theimportance of the HypB linker region in thehydrogenase metallocenter assembly pathway,arabinose-inducible plasmids expressingHypB(PP29,32SS), HypB(28d36), orHypB(11d76) were transformed into ∆hypBcells. Extracts from ∆hypB cells transformedwith pBAD-HypB and grown under the sameconditions had a relative hydrogenase activityof ≈ 80% compared to the control cells, while∆hypB cells had no hydrogenase activity asexpected (Figure 4B; (43)). Hydrogenaseactivities measured in extracts from cellsexpressing HypB with the Pro-Ser mutations(pBAD-HypB(PP29,32SS)) or theAP29FAP32AARP sequence removed (pBAD-HypB(28d36)) were lower than for pBAD-HypB-transformed cell extracts, suggestingthat hydrogenase activation is compromisedby disrupting the binding of SlyD to HypB.Under the standard conditions of inductionwith 1 µM arabinose, cell extracts frompBAD-HypB(11d76) cells had negligiblehydrogenase activity, but unlike the otherHypB variants, Western blotting indicatedthat the HypB(11d76) was poorly expressed(data not shown). Increasing the arabinoseconcentration from 1 µM to 10 µM resulted inprotein expression at a level similar to that ofwild-type HypB in the MC4100 cells. TheseHypB(11d76) cell extracts exhibited limitedactivity (Figure 4B), suggesting that theinteraction of SlyD with linker region ofHypB plays a role in full hydrogenaseactivation.

Nickel Complements HypB Mutant. Theweak hydrogenase activation of HypB(11d76)could result from this HypB variant blockingNi(II) insertion or failing to release its boundNi(II). As has been previously reported, theaddition of excess nickel to the growth mediacan complement to a small degree the

hydrogenase-deficient phenotype of theΔhypB cells (Figure 4C, (10)) and, to a greaterdegree, the ΔslyD cells (Figure 4C, (19)).While the hydrogenase activities of wild-typeand pBAD-HypB cells are not affected by theaddition of nickel, the activity of pBAD-HypB(11d76) cells increases significantly(Figure 4C), suggesting that the loss ofinteraction between HypB and SlyD can becompensated partially by the addition ofnickel to the growth medium.

SlyD stimulates metal release from HypB.To investigate the consequences of theinteraction between SlyD and HypB, a kineticmetal release assay was employed. In thisexperiment released metal is captured by 4-(2-pyridylazo)resorcinol (PAR), a chromophoricmetal chelator that is used to monitor metalrelease from proteins (39). We havepreviously used this assay to show that nickelis slowly released from the high-affinity siteof HypB (t1/2 ≈ 22 hours; Figure 5A) (14). Inthis study, the PAR assay was used to evaluatethe effect of SlyD on nickel binding to HypB.The addition of increasing amounts of SlyDresults in a dramatic increase in the rate ofmetal release from HypB, with the effectsaturating at a SlyD:HypB ratio of about 10:1and a t1/2 ≈ 20 min (Figure 5A). A plot of theobserved rates of metal release versus SlyDconcentration yields a saturation curve with aKd for the HypB-SlyD interaction of 9 µM,assuming a 1:1 binding stoichiometry (Figure5B). These results reveal that SlyD iscatalyzing metal release from HypB. To ruleout the possibility of protein degradationduring the assay, a sample of HypB wasincubated with SlyD for two hours at roomtemperature and subjected to SDS-PAGE(data not shown). To test the possibility thatSlyD was causing metal release by using aredox mechanism, a reaction was set up in ananaerobic glove box (96% nitrogen, 4%hydrogen) and SlyD-stimulated metal releasefrom HypB was still observed. Metal releaseunder the standard aerobic conditions was also

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

9

unaffected by the addition of a reducing agent(1 mM TCEP) to the reaction buffer. Pre-incubation of HypB and SlyD for two hoursbefore the addition of PAR did not result in animmediate significant signal at 500 nm,indicating that PAR acts to sequester the metalfrom the proteins once it is released from thehigh-affinity site of HypB.

Metal release depends on HypB-SlyDcomplex formation. To determine if thephysical interaction between HypB and SlyDis required for faster metal release from HypB,the HypB and SlyD variants described abovewith defective complex formation in thechemical crosslinking assay were examined.The addition of SlyD(Δflap) results in slowermetal release from HypB compared to thereaction with wild-type SlyD at the sameSlyD:HypB ratio of 10:1 (t1/2

> 5 hourscompared to t1/2 ≈ 20 min; Figure 6A).Similarly, an examination of the HypB linkerregion variants revealed a correlation betweencomplex formation with SlyD and SlyD-induced metal release (Figure 6B). Theaddition of SlyD to HypB(PP29,32SS)stimulated metal release, but not asdramatically as for wild-type HypB (t1/2 ≈ 1hour). Metal release was even slower fromHypB(28d36) (t1/2 ≈ 2 hours) and very weakfrom HypB(19d76) and HypB(11d76) (t1/2 ≈ 5and 6 hours, respectively). The metal releasereactions were repeated at a lower SlyD toHypB variant ratio (4:1 instead of 10:1) andSlyD barely stimulated metal release fromHypB(19d76) and HypB(11d76) (≈10%released after 2 h compared to the 90% releaseobserved for the same ratio of SlyD with wild-type HypB), confirming that SlyD-stimulatedrelease from these variants is greatly impaired(Figure 6A and data not shown). Anassessment of the nickel-binding activities ofthe four HypB variants demonstrated thatHypB(19d76) and HypB(11d76) exhibit anincrease in their ε 320 values (Table 2),although the profile of the entire spectraremains the same (data not shown), possibly

indicative of a change in the ligandenvironment. Clearly, however, all fourvariants retain high-affinity Ni(II)-bindingactivity (Table 2). These data are consistentwith the slow metal release kinetics from theseHypB variants observed in the absence ofSlyD (data not shown).

The C-terminal tail of SlyD is required invivo for hydrogenase activation. Given thatthe C-terminal tail of SlyD is rich in metal-binding residues and that the protein can bindnickel ions (29), it is reasonable to suggestthat this region of the protein is an importantcomponent of SlyD function in hydrogenaseproduction, with a possible role in nickelstorage. In support of this hypothesis, thehydrogenase activity of extracts from ΔslyDcells expressing SlyD(1-146) was the same asthe activity in the ΔslyD cells, indicating thatthe C-terminal tail is required for SlyD tofunction in the hydrogenase activationpathway (Figure 4A).

The C-terminal tail of SlyD is necessarybut not sufficient to stimulate metal releasefrom HypB. The C-terminal tail of SlyD isinvolved with metal-dependent inhibition ofPPIase activity (29) and it influences themolecular chaperone activity of SlyD (Figure2), so even though this domain of SlyD is notrequired for complex formation with HypB(Figure 3B) it remained possible that it couldalso modulate metal release from HypB. Forthis reason, SlyD(1-146) was also tested in themetal release assay and only a very slowrelease of metal from HypB was observed (t1/2

> 4 hours; Figure 6A). SlyD(1-146) retains60% of wild-type SlyD PPIase activity ((29)and data not shown), but has a significantlyaltered CD spectrum compared to SlyD. Thissuggests that the C-terminal tail may interactand stabilize the PPIase domain to somedegree. To investigate whether the tail on itsown was sufficient to activate metal releasewe constructed MBP-SlyD(140-196), and thisfusion protein had no effect on metal releasefrom HypB (t1/2 ≈ 24 hours, Figure 6A). To

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

10

assess whether the cysteine residues of the C-terminal tail were required to stimulate metalrelease, SlyD was treated with an excess ofiodoacetamide followed by dialysis. Thistreatment fully modified the six cysteineresidues of the C-terminal tail as monitored bymass spectrometry, but the protein was stillcapable of stimulating metal release fromHypB (data not shown).

DISCUSSION

SlyD derives its name from theobservation that it is required for E. colisensitivity to lysis mediated by the phageφX174 protein E (21), possibly by stabilizingthe E protein through direct interactions(46,47). However, in consideration of itsphysiological function the C-terminal metal-binding domain and its ability to bind metalions led to the hypothesis that SlyD wasinvolved in E. coli metal homeostasis (21,29).Subsequent studies revealed that SlyD doesplay such a role by contributing to nickelaccumulation and nickel delivery to the apo-hydrogenases (19). These experimentssuggested that SlyD does direct nickel to thehydrogenase biosynthetic pathway as a resultof its interaction with HypB. In vitroexperiments were also used to demonstratethat SlyD is a PPIase (29) and a molecularchaperone (26), but the roles of these activitiesin the biosynthesis of hydrogenase were notassessed.

The SlyD variant SlyD(Δflap), whichlacks a five-residue loop imbedded in aputative molecular chaperone domain, isimpaired in its ability to suppress theaggregation and promote the reactivation ofCS, a model protein folding substrate.Furthermore, in a screen of spontaneousmutations the same deletion in SlyD producedresistance to the phage φX174 lysis (48). Inthe homologous MtFKBP17 this loop is partof the IF domain that is required for thechaperone activity but not the PPIase activity

of the protein (28). The structure ofMtFKBP17 revealed a hydrophobic surface onthe IF domain (27), proposed to interact withsubstrates and prevent aggregation. Thisfunction of the IF domain is supported by thebehavior of SlyD(Δflap) in the CS assays aswell as by the fact that this variant fails tointeract with HypB in vitro and is unable tosupport the activation of hydrogenase in vivo.

On the HypB side of the SlyD-HypBcomplex the protein-protein contacts lie in thelinker region between the NH2-CxxCGC motifand the GTPase domain. The decrease in theinteraction observed between SlyD andHypB(PP29,32SS) and the lack of a detectableinteraction with HypB(11d76) suggests aproline-directed interaction with additionalbinding determinants in the linker region. Thedeficiency in complex formation in vitrocorrelates with reduced hydrogenasebiosynthesis in vivo, which can becounteracted by excess nickel in the growthmedia. These results support the hypothesisthat a physical interaction between SlyD andHypB is required for SlyD’s role in Ni(II)insertion into apo-hydrogenase.

HypB has two Ni(II)-binding sites ofvastly different affinities, one in the GTPasedomain that has a micromolar Kd and oneinvolving the N-terminal CxxCGC motif thathas a sub-picomolar Kd (14). Given that theoff-rate of the latter site is likely to be manytimes slower than the doubling time of E. coli,and that this site is critical for hydrogenasebiosynthesis (C. Mulvihill, M.R.L. andD.B.Z., manuscript in preparation), anoutstanding question is how metal releasefrom this high-affinity site could occur in aphysiologically relevant time frame. In thisreport SlyD is shown to stimulate the in vitrorelease of Ni(II) from the high-affinity site ofHypB to a chromophoric indicator, perhaps bystabilizing a conformation of HypB that ismore open, while SlyD(Δflap) is unable to doso.

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

11

The C-terminal tail of SlyD is notabsolutely required for the molecularchaperone activity of SlyD but it does enhanceit, perhaps by acting as a hydrophilic lid to thepolypeptide-binding site or merely increasingthe solubility of complexes with denaturedsubstrates. This domain is also not essentialfor the interaction between SlyD and HypB invitro, but it is critical for SlyD’s role inhydrogenase biosynthesis. At a first glance,this result implies that the putative metalstorage capacity of SlyD, localized to the tailwith its numerous histidine, cysteine andcarboxylate residues, is just as critical forhydrogenase activation as the ability to bind toHypB. However, the SlyD variant lacking theC-terminal tail is also deficient in the in vitrostimulation of metal release from HypB.Thus it is likely that the in vivo role of theSlyD tail is more than just metal storage, andone possible explanation is that it mediates amore accessible conformation of the Ni(II)high-affinity site of HypB. In analogy, just asSlyD catalyzes in vitro transfer to the indicatorPAR, this SlyD-stabilized open conformationof HypB could also allow nickel collection bythe coordinating cysteines of the hydrogenaseenzymes in vivo, thus facilitating nickeldelivery. Several questions that remain to beresolved include whether a metal-binding siteon SlyD itself or its PPIase activity areinvolved in the metal transfer from HypB.

Although none of the other membersof the PPIase superfamily have a metal-binding domain like SlyD, some have beenimplicated in metal homeostasis. Forexample, FKBP52 binds to Atox1, ametallochaperone that plays a role indelivering copper for export by the Wilsonand Menkes proteins, and overexpression ofFKBP52 enhances copper release in a cellculture model system (49). There are otherproteins that are involved in metalhomeostasis and that have polyhistidineregions. In E. coli, RcnA (formerly YohM)functions as a nickel/cobalt efflux transporter

and contains a polyhistidine stretch as part ofa predicted cytoplasmic loop (50). Somehomologs of the accessory protein UreE,required for the metallocenter assembly ofnickel-containing urease, also possess apolyhistidine region, although this region isnot essential for enzyme production (3).Unlike the C-terminal tail of SlyD, however,these regions in RcnA and UreE are not rich inacidic residues or cysteine. In contrast, the twoH. pylori proteins HspA and Hpn havesequences rich in a mix of metal-bindingresidues that can bind multiple nickel ions(51-53). It is interesting to note that in thisorganism, which depends on bothhydrogenase and urease for efficientcolonization (54), the C-terminus of the SlyDhomolog has 3-fold fewer histidine residuesthan the E. coli protein, but whether the othernickel-binding proteins can functionallyreplace SlyD is not known.

The hydrogenase activity observed forcel ls express ing pBAD-HypB(11d76)resembles that observed for the slyD knock-out cells and is significantly above thatobserved for ∆hypB cell extracts. Given thedisruption of the HypB-SlyD interaction, thisphenotype suggests that SlyD only contributesto hydrogenase biosynthesis through itsinteraction with HypB. SlyD is still present in∆hypB cells expressing HypB(11d76), so thediminished hydrogenase activity cannot bedue to just a loss of SlyD’s Ni(II) storagefunction, but rather to a loss of nickel deliveryand/or to a loss of SlyD-stimulated metalrelease from HypB. The fact that the lack ofan interaction between HypB(11d76) andSlyD does not result in a complete inability toactivate hydrogenase, in contrast to the ΔhypBcells, indicates that SlyD-stimulated metalrelease cannot be the only route of nickeladdition through HypB, but may be theprivileged pathway under normal growthconditions. This model is supported by theobservation that the hydrogenase activity inthe cells expressing HypB(11d76) can be

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

12

significantly restored by excess nickel in thegrowth media. Since HypB(11d76) can stillbind Ni(II) with high affinity, this resultsuggests that the addition of Ni(II) to themedia bypasses the need for SlyD-stimulatedNi(II) release. This extra nickel may be ableto reach the apo-hydrogenase from a pool ofnickel separate from the hydrogenase-directedsupply of nickel employed in the absence ofexcess metal (55). Another possibility is thatin addition to compromising the interactionwith SlyD, HypB(11d76) fails to interactproperly with the hydrogenase large subunit,resulting in impaired nickel insertion. Thisissue will have to be resolved in future studies.

In contrast, the minimal amount ofactivity restored in the ∆hypB cells grown inexcess nickel, unlike the linker deletionmutants, suggests that HypB has a critical rolefor nickel insertion that is independent ofSlyD. This function may be related to theessential GTPase activity of HypB (10), whichis not disturbed in the variants analyzed. Onecould propose a model in which GTPhydrolysis, rather than playing a role in Ni(II)insertion per se, is linked to a subsequentquality control step that drives aconformational change releasing the boundaccessory proteins HypA, HypB and HypCfrom the large subunit of hydrogenase. Such arole would not require a tight or specificmetal-binding site in the GTPase domain,since protein-protein interactions couldmaintain HypB in proximity with the

metallocenter of hydrogenase. Indeed, themetal-binding site in the GTPase domain ofHypB has a Kd only in the micromolar rangeand is not selective for Ni(II) over Zn(II) (14).A recent X-ray crystal structure of HypB fromMethanocaldococcus jannaschii shows thatthe metal-binding residues are intimatelylinked to the GTPase active site, suggesting apossible triggering mechanism whereby metalbinding could stimulate GTP hydrolysis (18).Supporting this hypothesis is the fact that thelow-affinity metal-binding site of HypB isrequired for hydrogenase activation (C.Mulvihill, M.R.L. and D.B.Z., manuscript inpreparation). HypB also dimerizes in solution(this work and (19,35), but the role of HypBdimerization in hydrogenase maturation, asopposed to complex formation with SlyD, hasnot been assessed. Now that the playersinvolved have been identified andcharacterized, the intriguing molecular detailsof the metal transfer pathway for nickeladdition to hydrogenase can be examined.

ACKNOWLEDGEMENTSWe thank Prof. A. Böck for the generousdonation of the anti-HypB antibodies as wellas for the E. coli strains MC4100 and DHP-B.We also thank Dr. G. Butland and Prof. A.Emili for the ∆slyD MC4100 and BL21(DE3)strains, and members of the Zamble lab forhelpful discussions.

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

13

REFERENCES

1. Casalot, L., and Rousset, M. (2001) Trends Microbiol. 9, 228-2372. Blokesch, M., Paschos, A., Theodoratou, E., Bauer, A., Hube, M., Huth, S., and Böck, A.

(2002) Biochem. Soc. Trans. 30, 674-6803. Kuchar, J., and Hausinger, R. P. (2004) Chem. Rev. 104, 509-5254. Reissmann, S., Hochleitner, E., Wang, H., Paschos, A., Lottspeich, F., Glass, R. S., and

Böck, A. (2003) Science 299, 1067-10705. Blokesch, M., Albracht, S. P., Matzanke, B. F., Drapal, N. M., Jacobi, A., and Böck, A.

(2004) J. Mol. Biol. 344, 155-1676. Blokesch, M., Paschos, A., Bauer, A., Reissmann, S., Drapal, N., and Böck, A. (2004)

Eur. J. Biochem. 271, 3428-34367. Loscher, S., Zebger, I., Andersen, L. K., Hildebrandt, P., Meyer-Klaucke, W., and

Haumann, M. (2005) FEBS Lett. 579, 4287-42918. Winter, G., Buhrke, T., Lenz, O., Jones, A. K., Forgber, M., and Friedrich, B. (2005)

FEBS Lett. 579, 4292-42969. Waugh, R., and Boxer, D. H. (1986) Biochimie 68, 157-16610. Maier, T., Lottspeich, F., and Böck, A. (1995) Eur. J. Biochem. 230, 133-13811. Olson, J. W., Mehta, N. S., and Maier, R. J. (2001) Mol. Microbiol. 39, 176-18212. Lutz, S., Jacobi, A., Schlensog, V., Böhm, R., Sawers, G., and Böck, A. (1991) Mol.

Microbiol. 5, 123-13513. Hube, M., Blokesch, M., and Böck, A. (2002) J. Bacteriol. 184, 3879-388514. Leach, M. R., Sandal, S., Sun, H., and Zamble, D. B. (2005) Biochemistry 44, 12229-

1223815. Mehta, N., Olson, J. W., and Maier, R. J. (2003) J. Bacteriol. 185, 726-73416. Blokesch, M., Rohrmoser, M., Rode, S., and Böck, A. (2004) J. Bacteriol. 186, 2603-

261117. Atanassova, A., and Zamble, D. B. (2005) J. Bacteriol. 187, 4689-469718. Gasper, R., Scrima, A., and Wittinghofer, A. (2006) J. Biol. Chem. 281, 27492-2750219. Zhang, J. W., Butland, G., Greenblatt, J. F., Emili, A., and Zamble, D. B. (2005) J. Biol.

Chem. 280, 4360-436620. Wülfing, C., Lombardero, J., and Plückthun, A. (1994) J. Biol. Chem. 269, 2895-290121. Roof, W. D., Horne, S. M., Young, K. D., and Young, R. (1994) J. Biol. Chem. 269,

2902-291022. Göthel, S. F., and Marahiel, M. A. (1999) CMLS 55, 423-43623. Barik, S. (2006) Cell. Mol. Life Sci. 63, 2889-290024. Scholz, C., Stoller, G., Zarnt, T., Fischer, G., and Schmid, F. X. (1997) EMBO J. 16, 54-

5825. Arié, J.-P., Sassoon, N., and Betton, J.-M. (2001) Mol. Microbiol. 39, 199-21026. Scholz, C., Eckert, B., Hagn, F., Schaarschmidt, P., Balbach, J., and Schmid, F. X. (2006)

Biochemistry 45, 20-3327. Suzuki, R., Nagata, K., Yumoto, F., Kawakami, M., Nemoto, N., Furutani, M., Adachi,

K., Maruyama, T., and Tanokura, M. (2003) J. Mol. Biol. 328, 1149-116028. Furutani, M., Ideno, A., Iida, T., and Maruyama, T. (2000) Biochemistry 39, 453-462

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

14

29. Hottenrott, S., Schumann, T., Plückthun, A., Fischer, G., and Rahfeld, J.-U. (1997) J.Biol. Chem. 272, 15697-15701

30. Rey, L., Imperial, J., Palacios, J.-M., and Ruiz-Argüeso, T. (1994) J. Bacteriol. 176,6066-6073

31. Fu, C., Olson, J. W., and Maier, R. J. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 2333-2337

32. Olson, J. W., Fu, C., and Maier, R. J. (1997) Mol. Microbiol. 24, 119-12833. Olson, J. W., and Maier, R. J. (2000) J. Bacteriol. 182, 1702-170534. McCafferty, D. G., Lessard, I. A. D., and Walsh, C. T. (1997) Biochemistry 36, 10498-

1050535. Maier, T., Jacobi, A., Sauter, M., and Böck, A. (1993) J. Bacteriol. 175, 630-63536. Gasteiger, E., Hoogland, C., Gattiker, A., Duvaud, S., Wilkins, M. R., Appel, R. D., and

Bairoch, A. (2005) Protein Identification and Analysis Tools on the ExPASy Servier. TheProteomics Protocols Handbook (Walker, J. M., Ed.), Humana Press

37. Lanzetta, P. A., Alvarez, L. J., Reinach, P. S., and Candia, O. A. (1979) Anal. Biochem.100, 95-97

38. Janowski, B., Wollner, S., Schutkowski, M., and Fischer, G. (1997) Anal. Biochem. 252,299-307

39. Hunt, J. B., Neece, S. H., and Ginsburg, A. (1985) Anal. Biochem. 146, 150-15740. Buchner, J., Grallert, H., and Jakob, U. (1998) Methods Enzymol. 290, 323-33841. Yu, D., Ellis, H. M., Lee, E.-C., Jenkins, N. A., Copeland, N. G., and Court, D. L. (2000)

Proc. Natl. Acad. Sci. U. S. A. 97, 5978-598342. Casadaban, M. J., and Cohen, S. N. (1979) Proc. Natl. Acad. Sci. U. S. A. 76, 4530-453343. Jacobi, A., Rossmann, R., and Böck, A. (1992) Arch. Microbiol. 158, 444-45144. Begg, Y. A., Whyte, J. N., and Haddock, B. A. (1997) FEMS Microbiol. Lett. 2, 47-5045. Ballantine, S. P., and Boxer, D. H. (1985) J. Bacteriol. 163, 454-45946. Mendel, S., Holbourn, J. M., Schouten, J. A., and Bugg, T. D. H. (2006) Microbiol. 152,

2959-296747. Bernhardt, T. G., Roof, W. D., and Young, R. (2002) Mol. Microbiol. 45, 99-10848. Roof, W. D., Fang, H. Q., Young, K. D., Sun, J., and Young, R. (1997) Mol. Microbiol.

25, 1031-104649. Sanokawa-Akakura, R., Dai, H., Akakura, S., Weinstien, D., Fajardo, J. E., Lang, S. E.,

Wadsworth, S., Siekierka, J., and Birge, R. B. (2004) JBC 279, 27845-2784850. Rodrigue, A., Effantin, G., and Mandrand-Berthelot, M. A. (2005) J. Bacteriol. 187,

2912-291651. Gilbert, J. V., Ramakrishna, J., Sunderman, F. W., Wright, A., and Plaut, A. G. (1995)

Infect. Immun. 63, 2682-268852. Kansau, I., Guillain, F., Thiberge, J.-M., and Labigne, A. (1996) Mol. Microbiol. 22,

1013-102353. Ge, R., Watt, R. M., Sun, X., Tanner, J. A., He, Q.-Y., Huang, J.-D., and Sun, H. (2006)

Biochem. J. 393, 285-29354. Maier, R. J., Benoit, S., and Seshadri, S. (2007) Biometals, 10.1007/s10534-10006-

19061-1053855. Rowe, J. L., Starnes, G. L., and Chivers, P. T. (2005) J. Bacteriol. 187, 6317-6323

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

15

FOOTNOTES†This work was supported in part by funding from the Canadian Institutes of Health Researchand the Petroleum Research Fund (ACS). M.R.L. was funded by a postdoctoral fellowship fromthe Natural Sciences and Engineering Research Council of Canada, and D.B.Z. is funded by aCanada Research Chair.

1Abbreviations: CD, circular dichroism; DEAE, diethylaminoethyl; DTNB, 5,5'-dithiobis(2-nitrobenzoic acid); DTT, dithiothreitol; E. coli, Escherichia coli; EDC, (1-Ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride; EDTA, ethylenediaminetetraacetic acid;FKBP, FK506-binding protein; G domain, the GTPase domain of HypB consisting of residues77-290; Hepes, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid; H. pylori, Helicobacterpylori; hyp, hydrogenase pleiotropy; IPTG; isopropyl-β-D-thiogalactoside; PAR, 4-(2-pyridylazo)resorcinol; PMB, parahydroxymercuribenzoic acid; PMSF, phenylmethysulfonylfluoride; PPIase, peptidyl-prolyl isomerase; SDS-PAGE, sodium dodecyl sulfate-polyacrylamidegel electrophoresis; SlyD(Δflap), SlyD with residues 107-111 deleted; TCEP, Tris-(2-carboxyethyl)phosphine; Tris, Tris-(hydroxymethyl)aminomethane

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

16

Table 1: PCR primers used for cloning and mutagenesis.a

Name SequenceB11d76forward 5’-GTTGCGGTGAAGGC/ATGCTGGAAGTCGAAATTG-3’B11d76reverse 5’-CGACTTCCAGCAT/GCCTTCACCGCAACCG-3’B19d76forward 5’-CGAGGGTGATGAA/ATGCTGGAAGTCGAAATTG-3’B19d76reverse 5’-CGACTTCCAGCATTTCAT/CACCCTCGATATAC-3’B28d36forward 5’-CATTCCGCGTTTCGTAGC/AAGATGAAAATCACCGGC-3’B28d36reverse 5’-CGGTGATTTTCATCTT/GCTACGAAACGCGGAATGAG-3’BPP29,32SSforward 5’-GTTTCGTAGCGCGtCATTTGCCtCGGCGGCACG-3’BPP29,32SSreverse 5’-CGTGCCGCCGaGGCAAATGaCGCGCTACGAAAC-3’D(1-146)forward 5’-CAGGAGATCATATGAAAGTAGCAAAAGACCTGGTG-3’D(1-146)reverse 5’-CGTGACCTCGAGCTTATTCTTCTTCAGTCGCTTC-3’D(140-196)forward 5’-GAAGTTGTGCATATGCGCGAAGCGACTGAAG-3’D(140-196)reverse 5’-GTCACTTCTCGAGTATTAGTGGCAACCGCAAC-3’DΔflapforward 5’-CAGGGTCCGGTACCG/GTTGAAGACGATCACG-3’DΔflap reverse 5’-GTGATCGTCTTCAAC/CGGTACCGGACCCTGG-3’BpBADforward 5’-GTTTAACGCTAGCAAGGAGATATACATAATGTGTACAAC-3’BpBADreverse 5’-GGTGGTTCTAGAGAACGCCTATGCACATCG-3’D(1-146)pBADforward 5’-CTATTCTTCGCTAGCTCAGGAGATATCATGAAAG-3’D(1-146)pBADreverse 5’-GAACGTGATCTAGAGCttaTTCTTCTTCAGTCGCTTC-3’aRestriction enzyme sites are shown in bold. Mutations are shown in lower case. For loop-outmutagenesis, the junction is indicated with a forward slash. Names starting with “B” or “D”indicate primers used for the cloning or mutagenesis of hypB or slyD, respectively.

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

17

Table 2: Metal-binding characteristics of HypB variants.

Name ε320 value a Kd

HypB (7.3 ± 0.1) × 103 M-1cm-1 (1.3 ± 0.2) × 10-13 MHypB(PP29,32SS) (7.5 ± 0.3) × 103 M-1cm-1 (1.2 ± 0.2) × 10-13 MHypB(28d36) (7.4 ± 0.2) × 103 M-1cm-1 (1.0 ± 0.4) × 10-13 MHypB(19d76) (8.5 ± 0.5) × 103 M-1cm-1 (1.9 ± 0.3) × 10-13 MHypB(11d76) (10.2 ± 0.3) × 103 M-1cm-1 (1.8 ± 0.4) × 10-13 MaMeasured by using direct titration with Ni(II). In some cases metal analysis of Ni(II)-HypB wasalso performed on protein incubated with nickel followed by gel filtration, and the extinctioncoefficients at 320 nm were within 5% of the listed values.

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

18

FIGURE LEGENDS

FIG. 1. Domain organization of E. coli HypB and SlyD. A, E. coli HypB consists of an NH2-CxxCGC motif involved in high-affinity Ni(II) binding (14), a linker region, and a GTPasedomain with a lower-affinity metal-binding site. HypB proteins from some bacteria include apolyhistidine sequence in the linker, but E. coli HypB has only a total of four histidine residuesin the linker. A proline-containing sequence in the linker of E. coli HypB was mutated in thisstudy. B, E. coli SlyD consists of an FKBP-type PPIase domain that has an IF (insert in the flap)chaperone domain, and a metal-binding C-terminal tail rich in histidines, cysteines, glutamatesand aspartates (20,21,27,29).

FIG. 2. Aggregation or reactivation of CS in the presence of SlyD and SlyD variants. A,chemically denatured, reduced CS was diluted to 0.1 µM in buffer, incubated on its own or withSlyD at the indicated ratios, and the extent of aggregation was determined by monitoring lightscattering at 500 nm. B, same as A except the CS was incubated with 20:1 SlyD, SlyD(1-146) orSlyD(∆flap). C, chemically denatured, reduced CS was diluted to 0.2 µM with buffer, incubatedwith 4 µM SlyD, SlyD(1-146) or SlyD(∆flap), and aliquots were tested for CS activity after twohours. Initial rates were normalized to the rate catalyzed by an equivalent concentration ofuntreated CS. The data are averages of four experiments and error bars indicate ± one standarddeviation (S.D.). Asterisks indicate data that are different at the 99.9% confidence level fromthose generated with wild-type SlyD as determined by using a Student’s t-test.

FIG. 3. Mapping the interface of the HypB-SlyD complex. A, HypB (8 µM) was incubated byitself or with SlyD or SlyD(∆flap) (40 µM) at 4°C overnight and then for one hour at roomtemperature in the presence of 5 mM EDC. B, SlyD(1-146) (10 µM) was incubated with orwithout HypB (10 µM) in the presence of EDC. C, HypB, HypB(11d76), or HypB(PP29,32SS)(8 µM) was incubated with or without SlyD (40 µM) in the presence of EDC. All reactions (A-C) were resolved on 12.5% SDS-polyacrylamide gels and the proteins were stained withCoomassie Blue dye. For all gels the heterodimeric HypB-SlyD crosslink is indicated with anasterisk.

FIG. 4. Hydrogenase activity of E. coli expressing HypB and SlyD variants. A, wild-type(DY330), ∆slyD (DY330), and ∆slyD cells transformed with pBAD, pBAD-SlyD, pBAD-SlyD(1-146) or pBAD-SlyD(∆flap) were grown anaerobically in TGYEP with 100 µM arabinose for sixhours. Cell extracts were prepared and tested for hydrogenase activity by using benzylviologenas a chromophoric electron acceptor in an anaerobic solution assay. The rates of benzylviologenreduction were normalized for total protein concentration and the data were averaged and thennormalized to the value for wild-type cells. Asterisks indicate data that are different at the 99%confidence level from those generated with pBAD-SlyD as determined by using a Student’s t-test. B, wild-type (MC4100), ∆hypB(MC4100), and ∆hypB cells transformed with pBAD,pBAD-HypB, pBAD-HypB(PP29,32SS), pBAD-HypB(28d36), or pBAD-HypB(11d76) weregrown anaerobically in TGYEP, 5-10 µM Ni(II), and 1µM or 10 µM (pBAD-HypB(11d76)arabinose, and tested for hydrogenase activity as above. Asterisks indicate data that are differentat the 98% confidence level from those generated with pBAD-HypB as determined by using aStudent’s t-test. C, wild-type (MC4100), ∆slyD and ∆hypB prepared from the MC4100 strain,

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

19

and ∆hypB cells transformed with pBAD-HypB or pBAD-HypB(11d76) were grown as above inthe absence (grey bars) or presence (black bars) of extra 300 µM Ni(II) in the growth media andcell extracts were tested for hydrogenase activity. Error bars indicate ± 1 S.D. Asterisks indicatedata from cells grown in extra nickel that are different at the 95% confidence level from thosegenerated with the same cells grown in the absence of nickel as determined by using a Student’st-test.

FIG. 5. SlyD stimulates metal release from HypB to PAR. A, HypB (5 µM) containing closeto stoichiometric amounts of nickel was incubated with 100 µM PAR (closed circles) and theindicated amounts of SlyD. Metal release was monitored by measuring the absorbance at 500 nmdue to the metal-PAR2 complex. The data were converted to percent metal bound bydetermining total metal content following treatment of a separate HypB sample with 100 µMPMB, and then fit to an exponential decay equation. B, the first-order rate constants of metalrelease from HypB in the presence of increasing concentrations of SlyD were fit to a saturationequation with a Kd value of 9 µM.

FIG. 6. Variants deficient in SlyD-dependent metal release from HypB to PAR. A, HypB(5 µM) was incubated with 100 µM PAR in the presence of 10:1 SlyD, SlyD(∆flap), MBP-SlyD(140-196), or SlyD(1-146). B, HypB, HypB(PP29,32SS), HypB(19d76) or HypB(11d76)(5 µM) were incubated with 100 µM PAR in the presence or absence of 10:1 SlyD. For bothplots metal release was monitored by measuring the absorbance at 500 nm due to the metal-PAR2 complex and the data treated as described for Figure 5A.

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

1961461

12076

Peptidyl-Prolyl Isomerase Domain Metal-Binding Domain

IF-Chaperone Domain

2 77

GTPase DomainLow-Affinity Metal Site

290

A28PFAPAARP36C2XXC5GC7

Linker RegionHigh-Affinity

Ni(II) Site

A

B

Figure 1Leach et al.

8

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

CS + SlyD

(1-14

6)

0

20

40

60

80

100

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4 5

CSCS + SlyDCS + SlyD(∆flap)CS + SlyD(1-146)

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4 5

CSCS + 2:1 SlyDCS + 10:1 SlyDCS + 20:1 SlyD

A

Time (min)

Rela

tive

Ligh

t Sca

tterin

g

CSCS +

SlyDCS +

SlyD(∆f

lap)%

Rel

ative

Citr

ate

Synt

hase

Act

ivity

Rela

tive

Ligh

t Sca

tterin

g

Time (min)

B C

Figure 2Leach et al.

**

*

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

A

SlyD

SlyD

+Hyp

BSl

yD(∆

flap)

+Hyp

B

SlyD

(∆fla

p)

*

(kDa)116

66

45

35

25

SlyD

HypB

HypB

+SlyD

HypB

(11d

76)

HypB

(11d

76)+

SlyD

HypB

(PP2

9,32S

S)Hy

pB(P

P29,3

2SS)

+SlyD

HypB

+ S

lyD(1

-146

)

HypB

SlyD

(1-1

46) CB

Figure 3Leach et al.

**

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

0

20

40

60

80

100

120

No Ni(II) Added300 uM Ni(II)

0

20

40

60

80

100

120

0

20

40

60

80

100

120

A

WT

pBAD

pBAD

-SlyD

pBAD

-SlyD

(1-14

6)

pBAD

-SlyD

(∆fla

p)

% R

elat

ive H

ydro

gena

se A

ctivi

ty

WTpB

AD-H

ypB

pBAD

-Hyp

B(PP

29,32

SS)

pBAD

-Hyp

B(28

d36)

pBAD

-Hyp

B(11

d76)

pBAD WT

ΔslyD

pBAD

HypB(

11d7

6)

pBAD

-Hyp

B

B C

∆slyD

∆hypB

∆hypB

Figure 4Leach et al.

* *

*

*

**

*

*

*

0

20

40

60

80

100

No Ni(II) Added300 uM Ni(II)

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

A

Time (min)

% M

etal

Bou

nd

Met

al R

elea

se C

onst

ant (

min

-1)

[SlyD] (µM)

B

Figure 5Leach et al.

SlyD/HypB

0

0.2

0.5

24100

20

40

60

80

100

0 20 40 60 80 100 120

HypBHypB + 0.2:1 SlyDHypB + 0.5:1 SlyDHypB + 2:1 SlyDHypB + 4:1 SlyDHypB + 10:1 SlyD

0

0.005

0.01

0.015

0.02

0.025

0.03

0.035

0.04

0 50 100 150

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

A

Time (min)

% M

etal

Bou

nd

B

Time (min)

Figure 6Leach et al.

0

20

40

60

80

100

0 20 40 60 80 100 120

HypBHypB + SlyDHypB + SlyD(∆flap)HypB + MBP-SlyD(140-196)HypB + SlyD(1-146)

0

20

40

60

80

100

0 20 40 60 80 100 120

HypBHypB + SlyDHypB(PP29,32SS) + SlyDHypB(28d36) + SlyDHypB(19d76) + SlyDHypB(11d76) + SlyD

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Michael R. Leach, Jie Wei Zhang and Deborah B. Zamblefactors HYPB and SLYD

The role of complex formation between the Escherichia coli hydrogenase accessory

published online April 10, 2007J. Biol. Chem.

10.1074/jbc.M610834200Access the most updated version of this article at doi:

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

by guest on March 14, 2018

http://ww

w.jbc.org/

Dow

nloaded from