thereplicationfactorcclamploaderrequiresarginine ...odonnell.rockefeller.edu/assets/pdfs/174.pdf ·...

The Replication Factor C Clamp Loader Requires ArginineFinger Sensors to Drive DNA Binding and ProliferatingCell Nuclear Antigen Loading*�

Received for publication, June 26, 2006, and in revised form, September 14, 2006 Published, JBC Papers in Press, September 15, 2006, DOI 10.1074/jbc.M606090200

Aaron Johnson‡, Nina Y. Yao‡, Gregory D. Bowman§, John Kuriyan¶�, and Mike O’Donnell‡**From the **Laboratory of DNA Replication, Howard Hughes Medical Institute and ‡Rockefeller University, New York, New York10021, the ¶Howard Hughes Medical Institute, Department of Molecular and Cell Biology and Department of Chemistry, Universityof California, Berkeley, California 94720, the �Physical Biosciences Division, Lawrence Berkeley National Laboratory, Berkeley,California 94720, and the §Department of Biophysics, Johns Hopkins University, Baltimore, Maryland 21218

Replication factor C (RFC) is an AAA� heteropentamer thatcouples the energy of ATP binding and hydrolysis to the loadingof the DNA polymerase processivity clamp, proliferating cellnuclear antigen (PCNA), onto DNA. RFC consists of five sub-units in a spiral arrangement (RFC-A, -B, -C, -D, and -E, corre-sponding to subunits RFC1, RFC4, RFC3, RFC2, and RFC5,respectively). The RFC subunits are AAA� family proteins andthe complex contains four ATP sites (sites A, B, C, and D)located at subunit interfaces. In each ATP site, an arginine res-idue from one subunit is located near the �-phosphate of ATPbound in the adjacent subunit. These arginines act as “argininefingers” that can potentially perform two functions: sensing thatATP is bound and catalyzing ATP hydrolysis. In this study, thearginine fingers inRFCweremutated to examine the steps in thePCNA loadingmechanism that occur afterRFCbindsATP.Thisreport finds that the ATP sites of RFC function in distinct stepsduring loading of PCNAontoDNA.ATPbinding toRFCpowersrecruitment and opening of PCNA and activates a �-phosphatesensor in ATP site C that promotes DNA association. ATPhydrolysis in site D is uniquely stimulated by PCNA, and wepropose that this event is coupled to PCNA closure aroundDNA,which starts an orderedhydrolysis around the ring. PCNAclosure severs contact to RFC subunits D and E (RFC2 andRFC5), and the�-phosphate sensor ofATP siteC is switched off,resulting in low affinity of RFC for DNA and ejection of RFCfrom the site of PCNA loading.

All cellular organisms utilize a multiprotein ATP-drivenDNA replicase, which functions with other proteins to dupli-cate chromosomal DNA prior to cell division (1). DNA repli-cases are composed of a DNA polymerase, a ring-shaped pro-cessivity clamp, and a clamp loaderATPase (reviewed in Ref. 2).

The DNA polymerase binds to the processivity clamp protein,which completely encircles DNA. The clamp slides on DNAand is pulled along by the polymerase during chain extension,continuously holding polymerase to DNA and constraining itto act in a highly processive manner.The eukaryotic ring-shaped processivity factor, proliferating

cell nuclear antigen (PCNA),2 is a trimer of three identical sub-units arranged head-to-tail to generate a ring with a large cen-tral cavity for encircling DNA (3, 4). PCNA confers processivityon DNA polymerase � (5–7) and also interacts with other fac-tors involved in DNA metabolism such as DNA polymerase �,flap endonuclease-1, DNA ligase, mismatch repair proteins,andmany others (8, 9). An interface between two PCNAmono-mers must be disrupted and opened to place DNA into thecenter of the ring. The ringmust then be re-closed for the clampto remain bound to theDNAand functionwith the polymerase.The eukaryotic clamp loader complex, replication factor C(RFC), uses the energy of ATP binding and hydrolysis to recruitthe processivity clamp to DNA, break one clamp interface, andtopologically link the clamp to primed template DNA (10–15).Studies in the Escherichia coli system have shown that theclamp loader and DNA polymerase compete for the clamp, andtherefore the clamp loader must eject from the clamp for theDNA polymerase to use it (16). Likewise, RFC and Pol � alsocompete for binding to PCNA (17, 18). Therefore, after PCNAis linked to DNA, RFC ejects from the clamp to allow the poly-merase access to the clamp (19) (summarized in Fig. 1A).A crystal structure of Saccharomyces cerevisiaeRFC-ATP�S-

PCNA reveals RFC to consist of a circular collar fromwhich thefive ATP-binding modules are suspended (Fig. 1B) (20). Thissubunit arrangement is similar to the E. coli � complex clamploader (21), which loads the ring-shaped � clamp onto DNA(22). The current study refers to the five RFC subunits based ontheir sequential order in the pentamer: RFC-A for RFC1, RFC-Bfor RFC4, RFC-C for RFC3, RFC-D for RFC2, and RFC-E forRFC5 (see Fig. 1C). This alphabetical nomenclature facilitatesthe comparison of subunits in analogous positions in RFC and�complex.

* This work was supported by National Institutes of Health Grant GM38839(to M. O. D.), an institutional training grant (to A. J.), the Ruth L. KirschsteinNational Research Service Award through the National Institute of GeneralMedical Sciences (to G. D. B.), and by National Institutes of Health GrantGM45547 (to J. K.). The costs of publication of this article were defrayed inpart by the payment of page charges. This article must therefore be herebymarked “advertisement” in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

� This article was selected as a Paper of the Week.1 To whom correspondence should be addressed: 1230 York Ave., Box 228,

New York, NY 10021-6399. Tel.: 212-327-7255; Fax: 212-327-7253; E-mail:[email protected].

2 The abbreviations used are: PCNA, proliferating cell nuclear antigen; RFC,replication factor C; ATP�S, adenosine 5�-O-(thiotriphosphate); ssDNA, sin-gle-stranded DNA; SRC, Ser-Arg-Cys; SSB, single-stranded DNA-bindingprotein; BSA, bovine serum albumin; DTT, dithiothreitol; CHAPS, 3-[(3-chol-amidopropyl)dimethylammonio]-1-propanesulfonic acid.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 281, NO. 46, pp. 35531–35543, November 17, 2006Printed in the U.S.A.

NOVEMBER 17, 2006 • VOLUME 281 • NUMBER 46 JOURNAL OF BIOLOGICAL CHEMISTRY 35531

at Rockefeller U

niversity Library on July 30, 2015

http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

The structural similarity between prokaryotic and eukaryoticclamps and clamp loaders suggests that the clamp loadingmechanisms may be similar, in general terms, and biochemicalstudies in both systems support this suggestion (23–30). Theclamp loader binds ATP to promote clamp binding and open-ing. The clamp loader-ATP-open clamp complex then associ-ateswith primed templateDNA.The association of clamp load-er-ATP-clamp complex with DNA does not appear to requireATP hydrolysis, and it is thought that DNA is positioned insidethe central channel of the clamp during this step (20, 31–33).DNA binding triggers ATP hydrolysis in the clamp loader,which then closes the clamp and ejects from DNA, leaving the

closed ring onDNA. This final stageof the clamp loading mechanismrequires hydrolysis of ATP and iscurrently viewed as a single step,although multiple ATP moleculesare hydrolyzed, and thus this stageof the reaction is likely composed ofmultiple steps.The five RFC subunits each con-

tain three domains (domains I–III),and the primary sequence of thesedomains is homologous among thefive subunits; only the RFC-A(RFC1) subunit has additional N-and C-terminal regions (34, 35).The RFC pentamer complex ismainly held together by a tightlypacked “collar” region consisting ofdomain III of each RFC subunit (seeFig. 1, A and B). The N-terminalATPase modules, composed ofdomains I and II, are arranged in aright-handed spiral with an overallpitch that closely matches double-stranded DNA (20). A cavity existsin the center of this protein spiralthat is sufficiently large to accom-modate double-strandedDNA. Fur-thermore, conserved polar and pos-itively charged residues line thiscavity and are proposed to interactwith DNA (20, 31). There is a gapbetween two subunits, RFC-A andRFC-E (illustrated in Fig. 1, A andC), which is presumed to provide anexit path for single-stranded DNAfrom the central cavity (20). Muta-tion of the conserved residues thatline the central cavity in � complexcauses a decrease in DNA binding,which supports the hypothesis thatDNA binds in the central cavity ofthese clamp loaders (32). A similarstudy in RFC shows that residueswithin the central cavity arerequired for DNA binding (36).

A recent electron microscopic reconstruction of Pyrococcusfuriosus RFC reveals that PCNA adopts a spiral conformationwhenheld open in a complex of RFC-ATP�S-PCNA-DNA (37).Consistent with this observation, molecular dynamics simula-tions of yeast PCNA indicate that PCNA alone may readilyadopt a right-handed spiral when open (38). The spiral confor-mation of open PCNA may allow a more intimate contactbetween the clamp and the five-subunit spiral of RFC.The ATPase modules of the five RFC subunits are similar in

structure to the AAA� family of ATPases (39). This familyincludes a wide variety of factors that couple ATP binding andhydrolysis to remodel protein substrates (40). A common fea-

FIGURE 1. ATP-coupled mechanism of RFC. A, schematic model of the ATP binding- and hydrolysis-coupledsteps of PCNA loading. After PCNA is loaded, RFC must disengage from PCNA to allow the polymerase accessto the clamp. B, crystal structure of RFC-ATP�S-PCNA (20). RFC is partially engaged with the closed PCNA ring.The C-terminal collar of RFC (top) is composed of domain III of each RFC subunit and domains I and II form theAAA� modules that bind ATP and PCNA. C, right: schematic of the arrangement of ATP sites in the AAA�modules of the RFC heteropentamer. Each ATP site is at a subunit interface. The neighboring subunit containsan arginine finger in a conserved SRC motif that interacts with the �-phosphate of the ATP bound to theadjacent subunit. Left: schematic depicting ATP site C from the crystal structure of RFC-PCNA (left). ATP (black)bound to RFC-C (RFC3) interacts with the RFC-C P-loop (yellow). The SRC arginine finger (blue) of RFC-D isproximal to the �-phosphate of the ATP.

The ATP Cycle of RFC

35532 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 281 • NUMBER 46 • NOVEMBER 17, 2006

at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

ture of oligomeric AAA� complexes is the location of ATPsites at the interface of two subunits. Interactions betweenAAA� modules of RFC subunits create four interfacial ATPsites that are competent for hydrolysis (20, 41). The four ATPsites of RFC have previously been referred to numerically, withATP site 1 at the RFC-D/E interface. In this study we adopt analphabetical nomenclature for theATP sites to correspondwiththe RFC subunit nomenclature. Thus, the four ATPase sites arereferred to as ATP sites A–D (Fig. 1C), with site A at the RFC-A/B interface and site D at the RFC-D/E interface. RFC-E(RFC5) also binds nucleotide (20), but the nucleotide-bindingpocket of RFC-E is not competent for hydrolysis nor is it neededfor RFC activity (25).Each of the four interfacial ATP sites in RFC consist of one

subunit that binds the ATP, plus side chains from the adjacentsubunit (Fig. 1C). One of the most prominent residues emanat-ing from the adjacent subunit is a highly conserved argininethat is part of a Ser-Arg-Cys (SRC) motif found in RFC Box VII(34), a region of sequence similarity found in all clamp loaders.This arginine side chain is referred to as an “arginine finger” byanalogy to an arginine residue inserted at the active site of theRas protein by the GTPase-activating proteins (42). The RFC-Barginine finger functions in ATP site A, and the arginine fingerof RFC-C functions in ATP site B and so on (see Fig. 1C). In theE. coli � complex clamp loader these arginine fingers have beenshown to have two functions: they sense bound ATP and alsoplay a catalytic role inATP hydrolysis (43, 44). The sensor func-tions of the arginine fingers in the � complex clamp loader arethought to be coupled to conformational changes that promotebinding to the clamp and DNA (44). The arginine fingers areneeded for ATP hydrolysis (43) and are presumed to act bystabilizing the growing negative charge as the �-phosphate iscleaved from ATP (21, 42, 45). In both RFC and � complex,ATP�S binding promotes ring opening and DNA binding, butthe clamp does not re-close (12, 27, 33, 46, 47); DNA binding isstabilized instead by direct contacts between the clamp loaderand DNA (27, 32, 33, 36, 48).Previous studies of prokaryotic and eukaryotic clamp loaders

suggest that individual ATP sites play specific roles in clamploading (29, 44, 49).Mutational analysis in yeast RFChas shownthat ATP binding to sites C and D of RFC is essential for DNAbinding (29), but it is not clear how ATP binding is coupled toDNA binding. In E. coli � complex, the arginine finger in ATPsiteD functions allosterically in promoting� clamp association,and the arginine fingers in sites B and/or C are involved inpromoting DNA binding (44). These findings support the sug-gestion that there is a division of activities among ATP sites inclamp loaders. Perhaps specific ATP sites control other impor-tant clamp loader actions such as clamp closing and clamploader ejection from the clamp and DNA.This report examines different RFC complexes that have one

or more arginine finger residues mutated to alanine. We showthat none of the arginine fingers of RFC are needed for the stepsof PCNA interaction and ring opening. However, our resultsdemonstrate that certain ATP sites of RFC play distinct rolesdownstream of PCNA opening. The arginine finger in ATP siteC is needed for RFC to bind DNA. ATP hydrolysis in site D isspecifically triggered by PCNA, and we propose that hydrolysis

in this site leads to closure of PCNA around DNA. The crystalstructure of RFC-PCNA shows that closed PCNA no longerinteracts with RFC-D and RFC-E, and thus ring closure is pre-sumed to lead to loss of RFC affinity for PCNA and dissociationof RFC from the PCNA-DNA complex.

EXPERIMENTAL PROCEDURES

Materials—Unlabeled ADP and ATP were purchased fromSigma, unlabeled ATP�S and eosin-5-maleimide were fromRoche Diagnostics, and radioactive nucleoside triphosphateswere from PerkinElmer Life Sciences, Inc. The following pro-teins were purified as described: PCNA and PCNA (Cys81-Only) (50) and E. coli SSB (51). PCNA and RFC concentrationswere determined by absorbance at 280 nm in 6 M guanidiniumhydrochloride using extinction coefficients calculated from theTrp and Tyr content (PCNA, �280 � 5120 M�1 cm�1; RFC,�280 � 158,880 M�1 cm�1). All other protein concentrationswere determined by Bradford assay reagent (Bio-Rad) usingBSA as a standard. PCNAHK is a PCNAconstruct with aHis-tagand a N-terminal 6-residue kinase recognition site (52).PCNAHK was purified in one step by nickel chelate chromatog-raphy. PCNAHK was labeled to a specific activity of 100 dpm/fmol with [�-32P]ATP using the recombinant catalytic subunitof cAMP-dependent protein kinase produced in E. coli (a giftfrom Dr. Susan Taylor, University of California at San Diego).The following oligonucleotides were synthesized and gel-puri-fied by Integrated DNA Technologies: 30-mer, 5�-GCA ATAACT GGC CGT CGT TTG AAG ATT TCG-3�; 66-mer,5�-CCATTCTGTAACGCCAGGGTTTTCGCAGTCAACATT CGA AAT CTT CAA ACG ACG GCC AGT TAT TGC-3�. A 102-mer 3�-biotinylated oligonucleotide was synthesizedand gel-purified by the W. M. Keck Facility at Yale University:5�-CCATTCTGTAACGCCAGGGTTTTCGCAGTCAACATT CGA AAT CTT CAA ACG ACG GCC AGT TAT TGCTCT TCT TGA GTT TGA TAG CCA AAA CGA CCA TTATAG-3� (Biotin). To form synthetic primed template, the30-mer primer oligonucleotide was mixed with either the66-mer or 102-mer template strand in a 1:1.2 ratio of primer totemplate in 50 �l of 5 mM Tris-HCl, 150 mM NaCl, 15 mMsodium citrate (final pH 8.5), then incubated in a 95 °C waterbath, which was allowed to cool to room temperature over a30-min interval. For reactions using primedM13mp18 ssDNA,the M13mp18 ssDNA was purified (53) and primed with a30-mer DNA oligonucleotide as described (54). The Bio-GelA-15m resin was purchased from Bio-Rad. Nitrocellulosemembrane circles were purchased fromSchleicher and Schuell,and polyethyleneimine-cellulose TLC plates were purchasedfrom EM Science. Streptavidin-coated Dynabead M-280 mag-netic beads were purchased from Dynal Biotech.Buffers—Buffer A is 30 mM Hepes-NaOH (pH 7.5), 0.5 mM

EDTA, 2 mM DTT, and 10% glycerol (v/v). Buffer B is 20 mMTris-HCl (pH 7.5), 0.5 mM EDTA, 2mMDTT, and 10% glycerol(v/v). Membrane Wash Buffer is: 20 mM Tris-HCl (pH 7.5), 10mM MgCl2, and 150 mM NaCl. Reaction buffer is 20 mM Tris-HCl (pH 7.5), 0.1 mM EDTA, 5 mM DTT, 4% glycerol (v/v), and40 �g/ml BSA. Gel filtration buffer is 20mMTris-HCl (pH 7.5),0.5 mM EDTA, 2 mM DTT, 10 mM MgCl2, 100 �g/ml BSA, and4% glycerol (v/v). Clamp loading buffer is 30 mMHepes-NaOH



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

(pH 7.5), 7mMMgCl2, 130mMNaCl, 1mMDTT, 1mMCHAPS.ATPase buffer is 20mMTris-HCl (pH 7.5), 0.1mMEDTA, 5mMDTT, and 10% glycerol (v/v).Construction and Purification of RFC(SAC) Mutants—Indi-

vidual RFC-B (RFC4), RFC-C (RFC3), RFC-D (RFC2), andRFC-E (RFC5) genes were mutated using the QuikChangemethod (Stratagene). The following mutations were made:RFC-B (RFC4) Arg157 to Ala, RFC-C (RFC3) Arg160 to Ala,RFC-D (RFC2) Arg183 to Ala, RFC-E (RFC5) Arg184 to Ala. AllS. cerevisiae five-subunit RFC complexes used in this studywere expressed in E. coli using a two-plasmid system describedpreviously (55). In all RFC constructs, the first 283 residues ofthe RFC-A (RFC1) subunit were deleted and RFC-A containedan N-terminal His-kinase tag (52). All RFC complexes werepurified as follows. Plasmids were transformed into E. colistrain BL21(DE3) and grown in 12 liters of Luria brothmediumcontaining 100 �g/ml ampicillin and 50 �g/ml kanamycin at37 °C. Upon reaching an OD of 0.6, cells were brought to 15 °C,and isopropyl �-D-thiogalactopyranoside was added to 0.5 mM,followed by a 12-h incubation with shaking at 15 °C. Cells wereharvested by centrifugation, resuspended in 100 ml of Buffer Acontaining 150 mM NaCl, and lysed using a French press at22,000 p.s.i. 2 mM PMSF was added to the cell lysate, and thelysate was clarified by centrifugation. The supernatant wasapplied to a 100 ml of SP-Sepharose (Amersham Biosciences)column equilibrated in Buffer A containing 150 mMNaCl. Pro-tein was eluted in a 750-ml gradient from 150–600 mM NaCl.The five-subunit RFC complex eluted at �510 mM NaCl, sepa-rating from RFC subcomplexes that eluted in earlier fractions.Fractions from the SP-Sepharose column containing five-sub-unit RFC were pooled, diluted in Buffer B to a conductivityequal to 100 mM NaCl, and loaded onto a 15-ml Fast FlowQ-Sepharose column (Amersham Biosciences) equilibrated inBuffer B containing 90 mM NaCl. Protein was eluted using an80-ml gradient of Buffer B from 90–500 mM NaCl. Fractionscontaining RFC eluted at �250 mM NaCl and were pooled,aliquoted, and stored at �80 °C. The protein yield was �50 mgfrom 12 liters of E. coli culture.ATP�S Binding Assays—Nitrocellulose membrane circles

(25 mm) were washed with 0.5 M NaOH, rinsed immediatelywith water, and equilibrated in membrane wash buffer beforeuse. Reactions contained 1�MRFC and [35S]ATP�S (0–25�M)in Buffer B � 160mMNaCl and 10mMMgCl2 in a total volumeof 15�l. After 1-min incubation on ice, 10�l of the reactionwasfiltered through a pretreated nitrocellulose membrane on aglass microanalysis filter assembly (Fisher) at a rate of 60�l/min. Themembranewas immediately washedwith 150�l ofice-cold membrane wash buffer at �1 ml/min. Membrane cir-cles were dried, and radioactivity was measured by liquid scin-tillation counting. An apparent Kd was determined using thesimple bindingmodel, RFC�ATP�S7RFC-ATP�S, to fit thedata using KaleidaGraph (Synergy Software).Clamp Loading Assays Using Primed M13mp18 ssDNA—

Clamp loading was measured using 32P-PCNAHK (29.4 nM astrimer), which was incubated for 10 min at 30 °C with RFCBCDE(SAC) or wild-type RFC (16.7 nM), in 60 �l of ReactionBuffer containing primed M13mp18 ssDNA (17 nM), SSB (4.2�M as tetramer), 0.5 mM ATP, and 10 mM MgCl2. The reaction

was applied to a 5-ml Bio-Gel A-15m column (Bio-Rad) equil-ibrated in gel filtration buffer containing 100mMNaCl at 23 °C.Thirty-five fractions of 180 �l each were collected, and 100 �lwas analyzed by liquid scintillation. 32P-PCNAHK-DNA com-plex elutes in the early fractions (fractions 9–15), while the free32P-PCNAHK elutes later (fractions 18–31). The amount ofPCNA in each fraction was determined from its known specificactivity.Magnetic Bead PCNA Clamp Loading Assay—The 30-mer/

102-mer biotinylated DNA was conjugated to DynabeadsM-280 Streptavidin (Dynal Biotech), and the DNA-bead conju-gate was incubated with 5 mg/ml BSA in 10 mM Tris-HCl (pH7.5), 1mMEDTA, 150mMNaCl at room temperature for 30minwith agitation and washed three times in the same buffer untilno BSA remained in the supernatant. Typical yield was �100pmol of DNA/mg of Dynabeads, as analyzed by comparingDNAethidiumbromide staining intensities in a polyacrylamidegel using known amounts of primed DNA as a standard. Clamploading reactions were performed at 16 °C in loading buffer.Mixtures containing final concentrations of 0.5 mM ATP, 170nM DNA, 510 nM E. coli SSB (tetramer), 700 nM 32P-PCNAHK

(trimer), but lacking RFC, were agitated by pipette and incu-bated for 1 min at 16 °C, then clamp loading was initiated uponadding RFC to a final concentration of 100 nM. Reactions werequenchedwith a final concentration of 21mMEDTA (pH7.5) atthe indicated time points. Quenched reactions were placed onice for 1min, then theDNAbeadswere isolatedwith amagneticconcentrator (1min) at 4 °C. Beads were washed twice in clamploading buffer containing 300 mM NaCl at 4 °C. Protein wasstripped from the beads using 1% SDS and counted by liquidscintillation.PCNA Ring Opening Assay—The following residues in the

S. cerevisiae PCNAgeneweremutated by theQuikChange pro-cedure (Stratagene): Cys22 to Ser, Cys30 to Ser, Cys62 to Ser.Oneburied native cysteine residue (Cys81) at each trimer interfaceremained in this PCNA construct, referred to as PCNA (Cys81-Only). PCNA ring opening was tested by treatment with eosin-5-maleimide, a thiol-reactive dye. Ring opening reactions con-tained 20 mM Tris-HCl (pH 7.1), 200 mM NaCl, 0.5 mM EDTA,10 mMMgCl2, 100 �MATP�S, 3 �M PCNA (Cys81-Only), 3 �MRFC, and 30 �M eosin-5-maleimide (final concentrations in 15�l). Reactions were incubated without eosin-5-maleimide for 1min on ice, initiated with eosin-5-maleimide, and quenchedafter 30 s with 5 �l of 1 M DTT. Control assays without RFCcontained 1.1 mg/ml BSA. Quenched reactions were analyzedin the dark on a 10% SDS-polyacrylamide gel, and labeledPCNA was visualized in a UV scan (305 nm) on a Fluor-SMultiImager (Bio-Rad).Fluorescent DNA Binding Assays—RFC binding to DNA was

measured using a fluorescently labeled primed template DNA.A synthetic 30-mer oligonucleotide, modified at the 3�-hy-droxyl with aC6-spacer primary amino group, was reactedwitha Rhodamine Red-X-NHS-ester (Integrated DNA Technolo-gies) to form a Rhodamine Red-X-conjugated 30-mer. “Rh-P/T” primed templateDNAwas prepared bymixing 1.0 nmol ofRhodamine Red-X-conjugated 30-mer (Integrated DNATech-nologies) with 1.2 nmol of unlabeled 66-mer in 100 �l of 5 mMTris-HCl, 150 mM NaCl, 15 mM sodium citrate (final pH 8.5)



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

and annealed at 37 °C for 1 h. RFCcomplexeswere titrated from0–420 nM into reactions of 60 �l final volume containing 20mMTris-HCl (pH7.5), 175mMNaCl, 10mMMgCl2, 5mMDTT,0.5 mM EDTA, 100 �M ATP�S or ADP, and 35 nM Rh-P/TDNA. When present, PCNA trimer was at a concentration of810 nM. The fluorophorewas excited at 570 nm, and a 580–680nm scan was taken. Relative intensity of the peak at 588 nmwasplotted versus RFC concentration. Apparent Kdmeasurementswere determined using a simple model (RFC � DNA7 RFC-DNA) to fit the data using the KaleidaGraph program (SynergySoftware). Attempts to measure primer-template associationby a change in fluorescence anisotropy were complicated by alow signal of RFC that bound nonspecifically to regions otherthan the primer-template junction. Measurement of fluores-cence intensity change of the probe at the primer-templatejunction avoids the influence of this nonspecific binding activ-ity.Mutation ofDNA-interacting residues of RFC-B, -C, and -Dwas performed as described previously (36).Steady-state ATPHydrolysis of RFC(SAC)Mutants—ATPase

assays contained 500 nM RFC, 1 mM [�-32P]ATP, 2 �M PCNAtrimer (when present), and 1 �M synthetic 30/66 DNA (whenpresent) in a final volume of 30 �l of ATPase buffer containing150 mM NaCl. The synthetic primer/template DNA is linearand allows PCNA to slide off the ends after it is loaded. Hence,PCNA is continuously recycled during these assays. Reactionswere brought to 30 °C and initiated upon addition of RFC. Ali-quots of 5�l were removed at 1min intervals from0–3min (for“�DNA” and “�PCNA�DNA” samples) or at 5 min intervalsfrom 0–20 min (for “alone” and “�PCNA” samples) andquenched with an equal volume of 0.5 M EDTA (pH 7.5). 1 �l ofeach quenched aliquot was spotted on a polyethyleneimine cel-lulose TLC sheet (EM Science) and developed in 0.6 M potas-sium phosphate buffer (pH 3.4). The TLC sheet was dried, and[�-32P]ATP and [�-32P]ADP were quantitated using a Phos-phorImager (GE Healthcare). Three independent experimentswere performed and the standard deviation is indicated in thefigure chart. A similar procedure was used when PCNA wastitrated into a fixed concentration of either wild-type RFC orRFC E(SAC), except that each reactionwas analyzed at a fixed 2min time point.

RESULTS

Arginine Finger Mutants of RFC Bind ATP Similarly to WildType—The N-terminal region of RFC-A, containing the first283 residues, is not required for cell viability, displays nonspe-cific DNA binding activity, and deletion of this region results ina complex with higher specific activity than RFC containingfull-length RFC-A (56, 57). All RFC complexes used in thisstudy carry a deletion of this region. The RFC complex contain-ing the truncated RFC-A with no additional point mutations toany subunit is referred to herein as “wild-type” RFC.Arginine to alanine mutations were constructed in each of

the four S. cerevisiae RFC subunits with SRC motifs. Five-sub-unit complexes of RFC-A/B/C/D/E containing the mutation ineither RFC-B, RFC-C, RFC-D, or RFC-E were expressed inE. coli using a two-plasmid system and purified as describedpreviously for wild-type RFC (55). These single mutant five-subunit complexes are referred to as: RFC B(SAC), RFC

C(SAC), RFC D(SAC), and RFC E(SAC). RFC-A does not con-tain an SRCmotif. Therefore, no point mutants of RFC-A wereused in this study.We also constructed a quadruple arginine finger mutant,

RFC BCDE(SAC) complex, which lacks arginine fingersentirely. Residual activities inherent to RFC BCDE(SAC) maybe presumed to be independent of the arginine fingers. All ofthe single RFC(SAC) mutants should therefore be able to pro-gress through the PCNA loading mechanism at least as far asthe RFC BCDE(SAC) mutant. The RFC BCDE(SAC) complexand the single RFC(SAC) complexes were expressed and puri-fied similarly to wild-type RFC and have comparable appear-ance in an SDS-polyacrylamide gel (Fig. 2A), suggesting that theRFC(SAC) mutants fold properly.Previous studies of the arginine fingers in the E. coli � com-

plex clamp loader concluded that these residues do not contrib-ute toATP binding (43). To test whether this is true for the RFCarginine fingers, we examined the ability of the RFCBCDE(SAC) mutant to bind nucleoside triphosphate. Using anitrocellulose filter binding assaywe determined theKd value ofwild-type RFC for ATP�S to be�3.3�M. The RFCBCDE(SAC)quadruple mutant displayed a Kd for ATP�S of 5.5 �M, compa-rable withwild-type RFC (Fig. 2B). This result indicates that thearginine fingers of RFC do not significantly contribute to ATPbinding.The RFC BCDE(SAC) Complex Cannot Load PCNA—To test

whether ATP binding to RFC can promote PCNA loading inthe absence of the arginine fingers, we measured the ability ofthe RFC BCDE(SAC) mutant to recruit PCNA to a primer-template DNA. A circular single-stranded M13mp18 phageDNA, uniquely primed with a DNA oligonucleotide and coatedwith E. coli SSB, was used as a substrate for PCNA loading. Tofollow clamp loading on DNA we used a PCNA construct con-taining a six-residue N-terminal kinase recognition sequencethat enables it to be labeled with [32P]phosphate. The largeDNA template, and 32P-PCNA attached to it, elutes in the earlyfractions (fractions 9–15) of a Bio-Gel A-15m gel filtration col-umn, while the comparatively small 32P-PCNA that is notattached toDNAelutes in later fractions (fractions 18–31) (Fig.2C). Incubation of 32P-PCNA with the primed DNA, ATP, andwild-type RFC generates a peak of 32P-PCNA that co-eluteswith DNA in the early fractions (Fig. 2C). However, the RFCBCDE(SAC) mutant does not load PCNA onto DNA. A smallamount of 32P-PCNA in the DNA-containing fractions wasobserved for the RFC BCDE(SAC) sample. This profile isobserved when 32P-PCNA is applied to the BioGel column inthe absence of clamp loader and DNA (data not shown) andtherefore is likely due to a small amount of aggregated 32P-PCNA in the protein preparation. Therefore, one or more argi-nine fingers are needed to load PCNA onto DNA.The Arginine Fingers of ATP Sites C and D Are Required for

Efficient PCNA Loading—We next addressed which argininefingers are required for RFC to load PCNA by testing singleRFC(SAC) mutants for PCNA loading activity. Since wild-typeRFC can load PCNA onto DNA in a very short time frame, wedesigned a PCNA loading assay that allows rate measurementsover a 25-s time course (see scheme in Fig. 3). In this assay, asynthetic primed template is first conjugated to a magnetic



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

bead. To prevent PCNA sliding off the free end of the linearDNA, the template was designed with sufficient single-strandedDNAon both sides of the primer to facilitate the bind-

ing of E. coli SSB, which blocks PCNA from sliding off (51).32P-Labeled PCNA was added to the magnetic bead-coupledprimedDNA, and clamp loading was initiated upon addition ofRFC. Reactions were rapidly quenched at timed intervals uponadding EDTA and then placed at 4 °C. PCNA remains stablyassociatedwith the SSB-coatedDNAwith a half-life of�50minat 4 °C (data not shown). The magnetic bead-conjugated DNAtemplate enabled rapid separation of PCNA-DNA complexesfrom free PCNA within 1 min. Western analysis showed thatRFC does not remain bound to the PCNA-DNA complex (datanot shown).The single (SAC) mutants displayed a spectrum of PCNA

loading rates (Fig. 3). The RFC B(SAC)mutant was comparablewith wild-type RFC. The RFC C(SAC) mutant was moderatelydeficient, displaying �35% the rate of clamp loading by wild-type RFC. In contrast, the RFC D(SAC) and RFC E(SAC)mutants were less than 5% the rate of wild-type RFC, suggestingthat ATP sites C (location of the RFC-D(SAC) mutation) andATP site D (location of the RFC-E(SAC)mutation) play impor-tant roles in driving PCNA loading.The Arginine Finger Mutants of RFC Can Bind and Open

PCNA—Which steps in clamp loading are the arginine fingermutants of RFC deficient in? Do any of the steps occur prior toATP hydrolysis?We used the non-hydrolyzable ATP analogue,ATP�S, to test the single RFC(SAC) mutants for a role prior toATP hydrolysis. PCNA ring opening and DNA binding do notrequire ATP hydrolysis by RFC, and therefore we developedassays to examine these two steps.First, we developed a PCNA opening assay based on a similar

assay for ring opening that we used previously for the E. colisystem (27, 58).We prepared a construct of PCNA inwhich thethree surface-exposed cysteine residuesweremutated to serine,leaving only one cysteine, Cys81, that is buried at each trimerinterface (Fig. 4A). This “PCNA (Cys81-Only)” clamp is resist-ant to labeling by the thiol-reactive dye, eosin-5-maleimide

FIGURE 2. The RFC arginine fingers are required for PCNA loading, but notfor ATP binding. A, analysis of 6 �g each of wild-type and mutant RFC com-plexes in a 10% SDS-PAGE stained with Coomassie Brilliant Blue (G250).Migration positions of the five RFC subunits are indicated to the left. B, wild-type RFC (black diamonds) and RFC BCDE(SAC) (red circles) were incubatedwith the indicated amounts of [35S]ATP�S followed by analysis of boundnucleotide on nitrocellulose filters as described under “Experimental Proce-dures.” C, wild-type RFC (black diamonds) and RFC BCDE(SAC) (red circles)were tested for ability to load 32P-PCNA onto singly primed circular M13mp18ssDNA coated with E. coli SSB (see scheme at top). DNA-bound 32P-PCNAmigrates early in the elution profile (fractions 9 –15) while unbound 32P-PCNAmigrates later (fractions 18 –31).

FIGURE 3. RFC D(SAC) and RFC E(SAC) are severely deficient in PCNA load-ing. The scheme at the top illustrates the magnetic bead-conjugated primedDNA assay to measure the kinetics of 32P-PCNA loading. E. coli SSB blocks theend of the DNA to prevent 32P-PCNA sliding off the linear DNA. The time-course of 32P-PCNA loading by the indicated RFC complexes was performedas described under “Experimental Procedures.”



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

(Fig. 4B). RFC and ATP�S stimulate labeling of Cys81 in PCNA(Cys81-Only), which is likely due to PCNA ring opening, thusincreasing exposure of the buried Cys81 for reaction with thefluorescent maleimide. As a control we tested an RFC complexthat lacks domains I and II of RFC-E (RFC5), which binds butdoes not open or unload PCNA from DNA (36). This mutantRFC did not stimulate labeling of the PCNA (Cys81-only) in thepresence of ATP�S (Fig. 4B, top). We were unable to perform acontrol reaction in the absence of ATP�S, because in theabsence of nucleotide, RFC and PCNA form an insolubleaggregate.3

Next, we tested various RFC(SAC) mutants for ability toopen PCNA in the presence of ATP�S. The single RFC(SAC)mutants were all active for PCNA opening (Fig. 4B,middle). Infact, even the RFC BCDE(SAC) mutant was capable of openingPCNA (Fig. 4B, bottom). Therefore the clamp loading defi-

ciency of the RFC D(SAC) and RFC E(SAC) mutants (e.g. seeFig. 3) must be blocked at a step in the PCNA loading mecha-nism that occurs after PCNA opening.A �-Phosphate Sensor in ATP Site C of RFC Regulates DNA

Binding—We next examined the RFC mutants for ability tobind DNA. ATP binding is known to promote RFC associationwith DNA, even in the absence of PCNA (28). To analyze DNAbinding by RFC, we used a synthetic DNA with a fluorescentrhodamine derivative attached to the 3� terminus of the primerstrand (see scheme in Fig. 5). We refer to this template as Rh-P/T. A change in fluorescence intensity indicates association ofRFC at the primer-template junction and we have shown pre-viously that RFCbinds thisDNA substratewith an affinity com-parable with that of unmodified DNA (28, 36).In Fig. 5A, wemeasured the affinity forDNAofwild-type and

arginine finger mutant RFC complexes by titrating RFC into afixed concentration of Rh-P/T DNA in the presence of ATP�S.Wild-type RFC produces an increase in fluorescence intensityin this assay. Fig. 5B shows that ADP does not suffice to powerthe conformational change in wild-type RFC needed for DNAbinding. The RFC C(SAC) and RFC B(SAC) mutants had affin-ities for the primed DNA that were equal or 2-fold reduced,respectively, compared with wild-type RFC; the RFC E(SAC)mutant displayed a 6-fold reduced affinity for the primedDNA.However, RFCD(SAC) lost essentially all detectableDNAbind-ing in this assay (Kd �5 �M). Therefore, the arginine finger ofRFC-D is likely needed to recognize ATP in site C to promotethe conformational change in RFC required for DNA binding.The RFC BCDE(SAC) complex did not display significant

DNA binding activity, presumably due to the D(SAC)mutation(Fig. 5C). Surprisingly, when the RFC BCDE(SAC) mutant wastitrated into a reactionwith the Rh-P/TDNA in the presence ofsaturating PCNA, a significant amount of DNAbinding activitywas observed (Fig. 5C). ADP could not substitute for ATP�S topromote this PCNA-dependent DNA binding (Fig. 5C), whichmay reflect the need for RFC to bind ATP to interact withPCNA. This result demonstrates that PCNA partially rescuesthe DNA binding deficiency of the RFC D(SAC) mutant.Removing DNA Interaction Residues in RFC-B, -C, or -D

Causes a Deficiency in DNA Binding—The result that RFCD(SAC) is deficient in association with primed template DNAindicates thatATP siteC is important inmodulating RFC-DNAinteraction. Three conserved positively charged residues ofRFC-D are predicted to interact with the phosphate backboneof double-stranded DNA that is bound in the central chamberof RFC (Fig. 6A) (20, 31, 32). Previously, we demonstrated thatremoval of these residues from RFC-D, in combination withremoval of the corresponding residues in RFC-B and RFC-C,abolishedDNAbinding byRFC (36).Weused theDNAbindingassay described above to test an RFC complex in which only thethree RFC-D residues were mutated. The result demonstratesthat the RFC D(R101A,R107A,R175A) mutant does not pro-duce a change in fluorescence intensity like wild-type RFC (Fig.6B), suggesting that removal of these residues is sufficient tobreak the strong interaction between RFC and the primed tem-plate. Interestingly, RFC complexes with the correspondingthree mutations in either RFC-B or RFC-C were similarly defi-cient in DNA binding (Fig. 6B). This disruption of the interac-3 G. D. Bowman and J. Kuriyan, unpublished observation.

FIGURE 4. PCNA ring opening assays. A, space-filling representation of thePCNA clamp structure. The three PCNA monomers are colored blue, yellow,and cyan. Cysteine 81 (sulfur atom in orange) of PCNA is buried at each trimerinterface of the ring (left diagram). In the diagram to the right, one monomeris removed to expose Cys81 at the trimer interface. B, scheme of eosin-5-maleimide labeling assays using PCNA (Cys81 only). The top panel is a controlreaction comparing eosin-labeling reactions containing no RFC, wild-typeRFC, or RFC lacking domains I and II of RFC-E (RFC E(�I,II)). The middle panelcompares single RFC(SAC) mutants and wild-type RFC in the PCNA openingassay. The bottom panel compares PCNA ring opening reactions with no RFC,wild-type RFC, or RFC BCDE(SAC). Proteins were separated by 10% SDS-PAGEand visualized by UV scanning at 305 nm on a Fluor-S Multi-Imager (Bio-Rad).



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

tion between any oneRFC subunit and the phosphate backboneof DNA appears to be sufficient to prevent RFC-DNAinteraction.The Arginine Fingers of RFC-D and RFC-E Are Most Impor-

tant for the ATPase Cycle of RFC—The experiments describedabove show that the arginine fingers of RFC are not needed forATP binding or PCNA opening but are needed for DNA bind-ing (i.e. the RFC-D arginine finger). The arginine fingers arealso needed for ATP hydrolysis which drives the steps of PCNA

loading after formation of the RFC-PCNA-DNA complex, resulting inclosure of the clamp and ejection ofRFC. Next we examine the effect ofindividual arginine fingermutationson the steady-state rate of ATPhydrolysis of RFC using a syntheticprimed/template in the presence ofPCNA. The short linear primedDNA allows PCNA to slide off theDNAafter it is loaded, and thus RFCcan perform repeated clamp loadingevents without running out ofPCNA and DNA substrate. Wild-type RFC produced a steady-staterate of �123 ATP hydrolyzed/RFC/min (Fig. 7A). The RFCBCDE(SAC)mutant gave no ATP hydrolysiswithin the time frame of the assay.Next, we examined the rate of ATPhydrolysis of the single RFC(SAC)mutants. If all four ATP sites con-tribute equally to the ATPase cycleof RFC, each single RFC(SAC)mutant may be expected to hydro-lyze ATP at �75% the rate of wild-type RFC. However, the resultsshow that two of the RFC(SAC)mutants, RFC D(SAC) and RFCE(SAC), hydrolyze ATP at less than20% the rate of wild-type RFC. Wehave shown earlier in this reportthat the RFCD(SAC)mutant is defi-cient in DNA binding. If the RFCD(SAC) mutant does not bindDNA, ATP hydrolysis will be com-promised since DNA stimulatesATP hydrolysis by RFC (Fig. 7B).However, the ATPase reaction isperformed in the presence of suffi-cient PCNA to enable the RFCD(SAC) mutant to bind to DNA.Therefore the RFC D(SAC) andRFC E(SAC) mutants are likelybound to PCNA and DNA, but theATP hydrolysis cycle is blocked,preventing clamp loading. One pos-sible explanation for this is thatATPbound at the mutant site must

hydrolyze before the remaining ATP sites can fire. The RFCC(SAC) andRFCB(SAC)mutants each hydrolyzeATP at abouthalf the rate of wild-type RFC. Therefore, inability to hydrolyzeATP in sites A or B also hinders hydrolysis in other sites, but theblock is less severe than when sites C or D are mutated. Theseassays were performed in the presence of PCNA and DNA.Addition of either PCNA or DNA alone to RFC also stimulatesthe rate of ATP hydrolysis by RFC but to a lesser extent thanwhen they are both present with RFC simultaneously.

FIGURE 5. The RFC-D arginine finger is required for DNA binding by RFC. The fluorescent primed templatecontains a 3�-terminal rhodamine moiety attached to the primer strand. Addition of RFC in the presence ofATP�S results in an increase in fluorescence intensity (see scheme at top). A, wild-type RFC and RFC(SAC)mutants were titrated into a reaction containing the fluorescent primed template and the relative fluorescenceintensity change (I/I0) was measured. The Kd values for RFC interaction with the primed template are indicatedto the right. B, wild-type RFC was titrated into reactions containing the fluorescent primed template in eitherthe presence of ATP�S (black diamonds) or ADP (red open circles). C, RFC BCDE(SAC) was titrated into reactionscontaining the fluorescent primed template and ATP�S in the presence (closed circles) or absence (closeddiamonds) of a saturating concentration of PCNA (810 nM). RFC BCDE(SAC) was also titrated into the reaction inthe presence of ADP and PCNA (crosses).



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

DNA Triggers ATP Hydrolysis in Site C—At 1 �M DNA,which is saturating in the ATPase assay, ATP hydrolysis bywild-type RFC and all four single RFC(SAC) mutants reaches amaximum, but the RFC D(SAC) mutant was deficient in DNA-stimulated ATP hydrolysis in the presence or absence of PCNA(Fig. 7A and data not shown). This result suggests that DNAmay trigger ATP hydrolysis in the unmutated ATP site C ofwild-type RFC. To test this possibility we compared the rate ofDNA-stimulated ATP hydrolysis of the triple RFC(SAC)mutants. Triple RFC(SAC) mutants contain only one ATP sitethat is competent for hydrolysis. The RFC BCE(SAC) mutant,which only contains a competent ATP site C, is most stronglystimulated by DNA and in fact displays a rate of �23 ATPhydrolyzed/RFC/min, about 80% the rate ofwild-type RFC (Fig.7B). This result indicates that DNA triggers ATP hydrolysis insite C and suggests that it may drive hydrolysis in additionalATP sites.

PCNATriggers ATPHydrolysis in ATP Site D—PCNA syner-gistically stimulates ATP hydrolysis by wild-type RFC in thepresence of DNA, presumably reflecting repeated cycles ofPCNA loading. In contrast, the RFC E(SAC) mutant displayedmore steady-state ATP hydrolysis activity in the presence ofDNA alone than in the presence of PCNA and DNA (Fig. 7Aand data not shown). This PCNA effect is examined moreclosely in Fig. 8A by titrating PCNA into reactions containingRFC and primed template DNA. The steady-state rate of ATPhydrolysis bywild-type RFC in the presence of primer-templateDNA is stimulated by PCNA, while PCNA has the oppositeeffect on theATPase activity of the RFCE(SAC)mutant. PCNAinhibition of ATP hydrolysis by the RFC E(SAC) mutantimplies that when ATP in site D cannot be hydrolyzed, PCNAblocks the ATPase cycle.In Fig. 8B we tested the RFC E(SAC) mutant and the other

single RFC(SAC) mutants for ATPase activity in the absenceand presence of PCNA,withoutDNA.TheRFCE(SAC)mutantis not significantly stimulated by PCNA. This result suggeststhat association of PCNA with RFC-ATP normally stimulatesthe hydrolysis of ATP bound to site D. To test this proposal, weexamined the triple RFC(SAC) mutants for PCNA-stimulated

FIGURE 6. Residues in the central chamber of RFC are required for DNAbinding. Three residues from RFC-B, RFC-C, and RFC-D that are proposed tointeract with DNA were mutated to alanine (A, from RFC-PCNA crystal struc-ture with DNA modeled through the central cavity of RFC and PCNA (20)). B,DNA binding assays were performed with a primer-template DNA labeled atthe 3� end of the primer (see scheme at top). An increase in the relative fluo-rescence (I/I0) indicates association of RFC with the DNA. RFC complexes withmutations in DNA-interacting residues of either RFC-B (R84A,R90A,K149A),RFC-C (R88A,R94A,K152A), or RFC-D (R101A,R107A,R175A) were assayedalong with wild-type RFC.

FIGURE 7. DNA stimulates ATP hydrolysis in ATP site C of RFC. All ATPaseassays were performed as described under “Experimental Procedures.” Reac-tions were performed in triplicate and error bars indicate one standard devi-ation of the data set. A, steady-state ATPase activity of single RFC(SAC) andRFC BCDE(SAC) mutants in the presence of PCNA and the synthetic primedtemplate. B, ATPase activity of RFC(SAC) triple mutants (i.e. three mutant sub-units in each complex) in the absence (�) or presence (�) of DNA.



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

ATPase activity (Fig. 8C). The results show that PCNA has thegreatest stimulatory effect on ATP hydrolysis by the RFCBCD(SAC) triple mutant, which only has a competent ATP siteD (Fig. 8C). PCNA stimulates the basal rate of the triple RFCBCD(SAC)mutant 10-fold, which is significantlymore than the3–4 fold stimulation by PCNA on wild-type RFC ATPhydrolysis.

DISCUSSION

The spiral heteropentameric RFC clamp loader from S. cer-evisiae has four bipartite ATP sites, each located at subunit-subunit interfaces (see Fig. 1). At each site one subunit contrib-utes a catalytic arginine finger that is contained within a highlyconserved SRC motif in RFC box VII. The other subunit con-tains the P-loop andotherATPbinding elements. In this report,conserved arginine fingers are mutated individually and incombination and the resultingATP sitemutant RFC complexesare examined for ability to progress through the different steps

in clamp loading. The results reveal individual functions of dif-ferent ATP sites during the clamp loading cycle.PCNA Is Open Prior to ATP Hydrolysis—In the absence of

ATP, neitherE. coli � complex nor eukaryotic RFCbind to theirrespective clamp or to DNA (28, 48, 59, 60). ATP binding pow-ers a conformational change that enables the clamp loader tobind the clamp and DNA (12, 27, 33, 46, 47). Use of ATP�Spromotes ring opening and ring loading ontoDNA, but the ringstays open and the clamp loader also remains with the openclamp on DNA. This was demonstrated in the E. coli systemusing a labeled maleimide and a � clamp containing a cysteineresidue at a buried position within the clamp interface (27). Inthe presence of ATP�S, clamp opening is signaled by anincrease in reactivity of the buried cysteine residue at the inter-face. Several lines of evidence exist for RFCholding PCNAopenwhile it is bound to DNA with ATP�S: 1) an electron micro-graphic reconstruction of a P. furiosus RFC-ATP�S-PCNA-

FIGURE 8. PCNA triggers ATP hydrolysis in ATP site D of RFC. Reactions were performed as in Fig. 7 unless indicated. Error bars indicate one standarddeviation of the data set. A, PCNA was titrated into a reaction containing primed DNA and either wild-type RFC or RFC E(SAC); ADP production was analyzedafter 2-min incubation. B, ATPase activity of RFC(SAC) single site mutants was analyzed in either the absence (�) or presence (�) of PCNA. C, ATPase activity ofRFC(SAC) triple site mutants was analyzed in either the absence (�) or presence (�) of PCNA.



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

DNA complex shows an open PCNA ring in a right-handedhelix (37), 2) A FRET study of yeast RFC indicates that theclamp remains open uponDNA binding (33), similar to the EMstudy of the P. furiosus clamp loader, and 3) study of RFCmutants indicates that the ring is open in the RFC-ATP�S-PCNA-DNA complex and that RFC, not PCNA provides thegrip that holds RFC-ATP�S-PCNA to DNA (36).

The PCNA ring is closed in the RFC-ATP�S-PCNA struc-ture in which the four arginine fingers of RFC were mutated toglutamine to inhibit nucleotide hydrolysis (20). The reason forthis result is not understood at present. However, the studies ofthe current report show that the RFC with four SRC ¡ SACmutations (i.e. RFC BCDE(SAC)) is still active in binding andopening PCNA. The RFCBCDE(SAC)mutant is also capable ofassembling PCNA onto DNA with ATP�S, forming the RFC-ATP�S-PCNA-DNA complex (Fig. 5C),4 which is consistentwith the ability of RFC BCDE(SAC) to open PCNA (Fig. 4B,bottom).

The current study of arginine finger mutants of RFC demon-strates that the arginine finger of RFC-D is required for RFC tobind DNA. This ATP site corresponds to site C of E. coli �

complex (at the interface of two � subunits), the one site thatremains closed in � complex when it is cocrystallized withATP�S (61). Furthermore, the � complex structure with twobound ATP�S molecules is in essentially the same conforma-tion as inactive � complex with no bound nucleotide. There-fore, binding of ATP to site C is presumed to power the confor-mational change in � complex required to bind both � andDNA. The ATP site C arginine finger mutation of RFC is theonly one that blocks a step powered by ATP binding, and thusonemay infer thatATP siteCof RFCmay carry a similar impor-tance to the ATP-dependent conformation of RFC as ATP siteC in E. coli � complex. However, structures of the fully ATPbound state of � complex and of the unliganded state of RFCwill be needed to confirm whether this prediction is indeed thecase.The ATP Cycle of RFC—The scheme in Fig. 9 summarizes a

model of the ATP hydrolysis cycle of RFC inferred from thefindings of this report. ATP binding powers all the steps that

4 A. Johnson, N. Y. Yao, G. D. Bowman, J. Kuriyan, and M. O’Donnell, unpub-lished results.

FIGURE 9. Model of ATP site function by RFC during clamp loading. Schematic models depicting the findings of this study in the context of the PCNA loadingcycle. A, ATP binding powers binding and opening of PCNA by RFC. The arginine fingers of RFC are not involved in the formation of this RFC-ATP-(open)PCNAcomplex. B, �-phosphate recognition by the ATP site C arginine finger coordinates RFC-(open)PCNA binding to DNA through the central chamber. C, ATPhydrolysis in site D is coupled to closure of PCNA by breaking the RFC-D/E contact to the clamp. D, DNA triggers ATP hydrolysis first in site C, which proceedsto site B and then site A. RFC dissociates from the site of loading, aided by the lowered affinity for DNA due to the lack of triphosphate in site C.



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

lead to formation of the complex of RFC bound to an open,spiral PCNA clamp (diagramA). To take the next step, bindingto DNA, the arginine finger of RFC-D is needed to sense the�-phosphate of ATP bound to site C. This ATP sensing stepbrings the RFC-PCNA complex into the proper conformationto bind primed DNA (diagram B). This conformational changecould be either subtle, for example by bringing the DNA bind-ing residues of RFC-D into proper alignment for interactionwith the phosphate backbone, or could be amore global change,such as altering the spiral pitch of the AAA� domains of thefive subunits of RFC, thus bringing all the subunits into registerfor DNA interaction. The results presented in Fig. 6B indicatethat removal of DNA-interacting residues from either RFC-B,-C, or -D eliminates RFC-DNA contact, suggesting that localchanges in RFC-Dmay be sufficient to modulate DNA bindingof the RFC complex.RFC-PCNA binds to primed DNA through the slot between

RFC subunits A and E, and through the open gap in PCNA.Once the DNA and PCNA substrates are properly oriented, theATPhydrolysis phasemay begin.Wepropose in Fig. 9 thatATPsite D may fire first, allowing PCNA to close (diagram C), fol-lowed by hydrolysis at sites C, B, and A. We have shown previ-ously that RFC-E, possibly along with RFC-D, opens the PCNAclamp (36). We show here that hydrolysis in ATP site D is spe-cifically stimulated by PCNA, and we propose that hydrolysisleads to PCNA closing around DNA, thereby severing the con-tact to RFC-D/E (diagramC). ADP formation in the ATP bind-ing pocket of RFC-D (ATP site D) may then act by a local allo-steric mechanism to position the RFC-D arginine finger forcatalysis of ATP in site C. This wave of cause-and-effect maylead to hydrolysis proceeding around the pentamer to ATP siteB and then ATP site A, which is located across the subunit gapfrom ATP site D. The fact that RFC activity is least affected bymutation of ATP site A, compared with the other ATP sites, isconsistentwith hydrolysis of this site occurring at the end of thehydrolysis cycle.Hydrolysis of ATP in site C is sensed by the arginine finger of

RFC-D, thus lowering the affinity of RFC for DNA. A decreasein affinity of RFC for DNA is also promoted by the absence ofnucleotide in ATP sites C and D as indicated by a study ofP-loop mutants of RFC that no longer bind ATP (29). In themodel of Fig. 9, ATP hydrolysis continues to site B and then theATP site of RFC-A is last to fire. RFC-A forms a tight connec-tion to PCNA (20), and hydrolysis in site A may loosen thiscontact (diagram D). Hence, the ATP hydrolysis cycle of Fig. 9first causes release of two RFC subunits (D and E) from PCNA,then loosens the interaction of RFCwith DNA (site C) followedby severing the remaining contact between RFC-A and PCNA,thus ejecting the clamp loader.ComparisonwithOtherClampLoaders—Unlike RFC, E. coli

� complex contains only three ATP sites. These sites corre-spond to ATP sites B, C, and D of RFC. In � complex, thearginine finger of ATP site D is contained in �� (E subunit),and mutation of this arginine yields a clamp loader complexthat has lower affinity for the � clamp (44). Arginine fingersof the other ATP sites cannot be mutated individually sincethe � subunits are encoded by the same gene. However,mutation of the arginine fingers in both ATP sites B and C

yields a complex that no longer interacts tightly with DNA(44). Thus, there are striking parallels between arginine fin-ger mutants of RFC and � complex. The arginine finger ofRFC in ATP site C is needed for DNA binding and thus hassimilar behavior to the arginine finger mutation in ATP sitesB and C in � complex. If RFC and � complex act in a similarfashion, then onemay predict that the arginine finger in ATPsite C of � complex, and not the arginine in ATP site B, isresponsible for the ATP binding-induced DNA affinity of �complex. Mutation of the � complex arginine finger in ATPsite D (between the D and E positions), like the same muta-tion in RFC, prevents ATPase activity in response to theclamp (43). Interestingly, the RFC-D and RFC-E subunits ofRFC are needed to open the PCNA clamp (36). However, in asignificant departure between the activities of the eukaryoticRFC and � complex, the � wrench (A subunit) of � complexopens the � clamp, while RFC-A lacks this activity (36). It isinteresting to note that the proposed order of ATP hydroly-sis in � complex (43) is also opposite to that proposed herefor RFC.TheRFCofArchaeoglobus fulgidus is similar toE. coli� com-

plex and eukaryotic RFC in the ability to load a clamp ontoDNA prior to ATP hydrolysis, and in all cases the clamp loaderremains on DNA with the clamp. One may presume that theclamp remains open from the work in the E. coli system andfrom the electron micrographic reconstruction of the RFC-ATP�S-PCNA-DNA complex from the archaeon P. furiosus(37). Archael RFC complexes consist of two different subunits:RFC-s (s for small) and RFC-l (l for large), in a 4:1 ratio, respec-tively. Studies on A. fulgidus RFC suggest that four ATP mole-cules bind to archaeal RFC and are hydrolyzed during a fullclamp loading cycle (49). A. fulgidus RFC with arginine fingermutations in the four RFC-s subunits still binds ATP but is nolonger functional in formation of the RFC-PCNA-DNA ternarycomplex (62), suggesting that a step prior to hydrolysis is dis-rupted by this mutation. This step correlates with the findinghere that the arginine finger of RFC-D is needed to sense ATPbinding in site C for RFC to bind DNA.Clamp loader subunit sequences and clamp loader structures

are similar in all three domains of life: bacteria, archaea, andeukaryotes (20, 21, 34, 37, 63–66). ATP sites C and D in RFC,which are critical for efficient PCNA loading, also are present inbacterial and archaeal clamp loaders. These two ATP sites arealso present in the alternate RFC clamp loaders in which the Asubunit (RFC1) exchanges with another protein (e.g. Rad24(Rad17 in humans), Ctf18, or Elg1). These “alternate” RFCcom-plexes act in other cellular processes such as DNA damagerepair and sister chromatid cohesion (15). For example, theRad24 subunit substitutes for RFC-A to form Rad24-RFCwhich loads the 9-1-1 heterotrimer clamp onto DNA (67, 68).The Ctf18-RFC acts in sister chromatid cohesion and is effi-cient in unloading PCNA, leading to the suggestion that it mayserve in this capacity when replication forks deal with cohesins(69–71). Since ATP sites C and D are present in all alternateclamp loaders, these two sites may be expected to perform sim-ilar functions to those shown here for the replicative RFCcomplex.



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

REFERENCES1. Kornberg, A., and Baker, T. (1992) DNA replication, 2nd Ed., W.H. Free-

man, New York2. Johnson, A., and O’Donnell, M. (2005) Annu. Rev. Biochem. 74, 283–3153. Krishna, T. S., Kong, X. P., Gary, S., Burgers, P. M., and Kuriyan, J. (1994)

Cell 79, 1233–12434. Gulbis, J.M., Kelman, Z., Hurwitz, J., O’Donnell,M., andKuriyan, J. (1996)

Cell 87, 297–3065. Tan, C. K., Castillo, C., So, A. G., and Downey, K. M. (1986) J. Biol. Chem.

261, 12310–123166. Prelich, G., Tan, C. K., Kostura, M., Mathews, M. B., So, A. G., Downey,

K. M., and Stillman, B. (1987) Nature 326, 517–5207. Prelich, G., Kostura, M., Marshak, D. R., Mathews, M. B., and Stillman, B.

(1987) Nature 326, 471–4758. Warbrick, E. (2000) Bioessays 22, 997–10069. Maga, G., and Hubscher, U. (2003) J. Cell Sci. 116, 3051–306010. Tsurimoto, T., and Stillman, B. (1989) EMBO J. 8, 3883–388911. Tsurimoto, T., and Stillman, B. (1990) Proc. Natl. Acad. Sci. U. S. A. 87,

1023–102712. Tsurimoto, T., and Stillman, B. (1991) J. Biol. Chem. 266, 1950–196013. Lee, S. H., Kwong, A. D., Pan, Z. Q., and Hurwitz, J. (1991) J. Biol. Chem.

266, 594–60214. Fien, K., and Stillman, B. (1992)Mol. Cell. Biol. 12, 155–16315. Majka, J., and Burgers, P. M. (2004) Prog. Nucleic Acids Res. Mol. Biol. 78,

227–26016. Naktinis, V., Turner, J., and O’Donnell, M. (1996) Cell 84, 137–14517. Mossi, R., Jonsson, Z. O., Allen, B. L., Hardin, S. H., and Hubscher, U.

(1997) J. Biol. Chem. 272, 1769–177618. Oku, T., Ikeda, S., Sasaki, H., Fukuda, K., Morioka, H., Ohtsuka, E., Yo-

shikawa, H., and Tsurimoto, T. (1998) Genes Cells 3, 357–36919. Podust, V. N., Tiwari, N., Stephan, S., and Fanning, E. (1998) J. Biol. Chem.

273, 31992–3199920. Bowman, G. D., O’Donnell, M., and Kuriyan, J. (2004) Nature 429,

724–73021. Jeruzalmi, D., O’Donnell, M., and Kuriyan, J. (2001) Cell 106, 429–44122. Stukenberg, P. T., Studwell-Vaughan, P. S., and O’Donnell, M. (1991)

J. Biol. Chem. 266, 11328–1133423. O’Donnell, M., Jeruzalmi, D., and Kuriyan, J. (2001) Curr. Biol. 11,

R935–R94624. Ellison, V., and Stillman, B. (1998) J. Biol. Chem. 273, 5979–598725. Cai, J., Yao, N., Gibbs, E., Finkelstein, J., Phillips, B., O’Donnell, M., and

Hurwitz, J. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 11607–1161226. Podust, V. N., Tiwari, N., Ott, R., and Fanning, E. (1998) J. Biol. Chem. 273,

12935–1294227. Hingorani, M. M., and O’Donnell, M. (1998) J. Biol. Chem. 273,

24550–2456328. Gomes, X. V., and Burgers, P. M. (2001) J. Biol. Chem. 276, 34768–3477529. Schmidt, S. L., Gomes, X. V., and Burgers, P. M. (2001) J. Biol. Chem. 276,

34784–3479130. Hingorani, M. M., and Coman, M. M. (2002) J. Biol. Chem. 277,

47213–4722431. Bowman, G. D., Goedken, E. R., Kazmirski, S. L., O’Donnell, M., and

Kuriyan, J. (2005) FEBS Lett. 579, 863–86732. Goedken, E. R., Kazmirski, S. L., Bowman, G. D., O’Donnell, M., and

Kuriyan, J. (2005) Nat Struct Mol Biol 12, 183–19033. Zhuang, Z., Yoder, B. L., Burgers, P. M., and Benkovic, S. J. (2006) Proc.

Natl. Acad. Sci. U. S. A.34. Cullmann, G., Fien, K., Kobayashi, R., and Stillman, B. (1995) Mol. Cell.

Biol. 15, 4661–467135. Uhlmann, F., Gibbs, E., Cai, J., O’Donnell,M., andHurwitz, J. (1997) J. Biol.

Chem. 272, 10065–1007136. Yao, N. Y., Johnson, A., Bowman, G., Kuriyan, J., andO’Donnell,M. (2006)

J. Biol. Chem. 281, 17528–1753937. Miyata, T., Suzuki, H., Oyama, T.,Mayanagi, K., Ishino, Y., andMorikawa,

K. (2005) Proc. Natl. Acad. Sci. U. S. A. 102, 13795–13800

38. Kazmirski, S. L., Zhao, Y., Bowman, G. D., O’Donnell, M., and Kuriyan, J.(2005) Proc. Natl. Acad. Sci. U. S. A. 102, 13801–13806

39. Neuwald, A. F., Aravind, L., Spouge, J. L., and Koonin, E. V. (1999) Ge-nome. Res. 9, 27–43

40. Erzberger, J. P., andBerger, J.M. (2006)Annu. Rev. Biophys. Biomol. Struct.35, 93–114

41. Yao, N., Coryell, L., Zhang, D., Georgescu, R. E., Finkelstein, J., Coman,M. M., Hingorani, M. M., and O’Donnell, M. (2003) J. Biol. Chem. 278,50744–50753

42. Ahmadian,M. R., Stege, P., Scheffzek, K., andWittinghofer, A. (1997)Nat.Struct. Biol. 4, 686–689

43. Johnson, A., and O’Donnell, M. (2003) J. Biol. Chem. 278, 14406–1441344. Snyder, A. K.,Williams, C. R., Johnson, A., O’Donnell,M., andBloom, L. B.

(2004) J. Biol. Chem. 279, 4386–439345. Ogura, T., Whiteheart, S. W., and Wilkinson, A. J. (2004) J. Struct. Biol.

146, 106–11246. Lee, S. H., and Hurwitz, J. (1990) Proc. Natl. Acad. Sci. U. S. A. 87,

5672–567647. Burgers, P. M. (1991) J. Biol. Chem. 266, 22698–2270648. Ason, B., Handayani, R., Williams, C. R., Bertram, J. G., Hingorani, M. M.,

O’Donnell, M., Goodman, M. F., and Bloom, L. B. (2003) J. Biol. Chem.278, 10033–10040

49. Seybert, A., and Wigley, D. B. (2004) EMBO J. 23, 1360–137150. Ayyagari, R., Impellizzeri, K. J., Yoder, B. L., Gary, S. L., and Burgers, P. M.

(1995)Mol. Cell. Biol. 15, 4420–442951. Yao, N., Hurwitz, J., and O’Donnell, M. (2000) J. Biol. Chem. 275,

1421–143252. Kelman, Z., Yao, N., and O’Donnell, M. (1995) Gene 166, 177–17853. Turner, J., and O’Donnell, M. (1995)Methods Enzymol. 262, 442–44954. Studwell, P. S., and O’Donnell, M. (1990) J. Biol. Chem. 265, 1171–117855. Finkelstein, J., Antony, E., Hingorani, M. M., and O’Donnell, M. (2003)

Anal. Biochem. 319, 78–8756. Uhlmann, F., Cai, J., Gibbs, E., O’Donnell,M., andHurwitz, J. (1997) J. Biol.

Chem. 272, 10058–1006457. Gomes, X. V., Gary, S. L., and Burgers, P. M. (2000) J. Biol. Chem. 275,

14541–1454958. Turner, J., Hingorani,M.M., Kelman, Z., andO’Donnell,M. (1999)EMBO

J. 18, 771–78359. Naktinis, V., Onrust, R., Fang, L., and O’Donnell, M. (1995) J. Biol. Chem.

270, 13358–1336560. Gerik, K. J., Gary, S. L., and Burgers, P. M. (1997) J. Biol. Chem. 272,

1256–126261. Kazmirski, S. L., Podobnik, M., Weitze, T. F., O’Donnell, M., and Kuriyan,

J. (2004) Proc. Natl. Acad. Sci. U. S. A. 101, 16750–1675562. Seybert, A., Singleton,M. R., Cook, N., Hall, D. R., andWigley, D. B. (2006)

EMBO J. 25, 2209–221863. O’Donnell, M., Onrust, R., Dean, F. B., Chen, M., and Hurwitz, J. (1993)

Nucleic Acids Res. 21, 1–364. Chen, M., Pan, Z. Q., and Hurwitz, J. (1992) Proc. Natl. Acad. Sci. U. S. A.

89, 2516–252065. Chen, M., Pan, Z. Q., and Hurwitz, J. (1992) Proc. Natl. Acad. Sci. U. S. A.

89, 5211–521566. Guenther, B., Onrust, R., Sali, A., O’Donnell, M., and Kuriyan, J. (1997)

Cell 91, 335–34567. Bermudez, V. P., Lindsey-Boltz, L. A., Cesare, A. J., Maniwa, Y., Griffith,

J. D., Hurwitz, J., and Sancar, A. (2003) Proc. Natl. Acad. Sci. U. S. A. 100,1633–1638

68. Majka, J., and Burgers, P. M. (2003) Proc. Natl. Acad. Sci. U. S. A. 100,2249–2254

69. Mayer, M. L., Gygi, S. P., Aebersold, R., and Hieter, P. (2001)Mol. Cell 7,959–970

70. Bermudez, V. P., Maniwa, Y., Tappin, I., Ozato, K., Yokomori, K., andHurwitz, J. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 10237–10242

71. Bylund, G. O., and Burgers, P. M. (2005)Mol. Cell. Biol. 25, 5445–5455



at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://www.jbc.org/

Bowman, John Kuriyan and Mike O'DonnellAaron Johnson, Nina Y. Yao, Gregory D. Nuclear Antigen LoadingDNA Binding and Proliferating CellRequires Arginine Finger Sensors to Drive The Replication Factor C Clamp LoaderRecombination:DNA: Replication, Repair, and

doi: 10.1074/jbc.M606090200 originally published online September 15, 20062006, 281:35531-35543.J. Biol. Chem.

10.1074/jbc.M606090200Access the most updated version of this article at doi:

.JBC Affinity SitesFind articles, minireviews, Reflections and Classics on similar topics on the

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

http://www.jbc.org/content/281/46/35531.full.html#ref-list-1

This article cites 69 references, 44 of which can be accessed free at

at Rockefeller U


http://ww

w.jbc.org/

Dow

nloaded from

http://affinity.jbc.org/

http://www.jbc.org/lookup/doi/10.1074/jbc.M606090200

http://affinity.jbc.org

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;281/46/35531&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/281/46/35531

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=281/46/35531&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/281/46/35531

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/281/46/35531.full.html#ref-list-1

http://www.jbc.org/

thereplicationfactorcclamploaderrequiresarginine ...odonnell.rockefeller.edu/assets/pdfs/174.pdf ·...

Documents