Dual Functions, Clamp Opening and Primer-Template Recognition

doi:10.1016/j.jmb.2004.07.097 J. Mol. Biol. (2004) 342, 1457–1469

Dual Functions, Clamp Opening and Primer-TemplateRecognition, Define a Key Clamp Loader Subunit

Maria Magdalena Coman1, Mi Jin2, Razvan Ceapa1, Jeff Finkelstein2,3

Michael O’Donnell2,3, Brian T. Chait2 and Manju M. Hingorani1*

1Molecular Biology andBiochemistry DepartmentWesleyan UniversityMiddletown, CT 06459, USA

2Rockefeller UniversityNew York, NY 10021, USA

3Howard Hughes MedicalInstitute, New York, NY 10021USA

0022-2836/$ - see front matter q 2004 E

Abbreviations used: dsDNA, doussDNA, single-stranded DNA; BrdUdeoxyuridine; MS, mass spectromeserum albumin; DHB, 2,5-dihydroxtrifluoroacetic acid.E-mail address of the correspond

[email protected]

Clamp loader proteins catalyze assembly of circular sliding clamps onDNA to enable processive DNA replication. During the reaction, the clamploader binds primer-template DNA and positions it in the center of a clampto form a topological link between the two. Clamp loaders are multi-protein complexes, such as the five protein Escherichia coli, Saccharomycescerevisiae, and human clamp loaders, and the two protein Pyrococcusfuriosus andMethanobacterium thermoautotrophicum clamp loaders, and thusfar the site(s) responsible for binding and selecting primer-template DNAas the target for clamp assembly remain unknown. To address this issue,we analyzed the interaction between the E. coli g complex clamp loader andDNA using UV-induced protein–DNA cross-linking and mass spec-trometry. The results show that the d subunit in the g complex makesclose contact with the primer-template junction. Tryptophan 279 in the d C-terminal domain lies near the 3 0-OH primer end and may play a key role inprimer-template recognition. Previous studies have shown that d alsobinds and opens the b clamp (hydrophobic residues in the N-terminaldomain of d contact b. The clamp-binding and DNA-binding sites on dappear positioned for facile entry of primer-template into the center of theclamp and exit of the template strand from the complex. A similar analysisof the S. cerevisiae RFC complex suggests that the dual functionalityobserved for d in the g complex may be true also for clamp loaders fromother organisms.

q 2004 Elsevier Ltd. All rights reserved.

Keywords:DNA replication; circular sliding clamp; clamp loader; UV-cross-linking; mass spectrometry
*Corresponding author
Introduction

Sliding clamp proteins are used by DNA poly-merases as mobile tethers for highly processivereplication of DNA. By themselves, the poly-merases can incorporate a few nucleotides into thegrowing DNA polymer in a single binding event; inassociation with clamps, however, their processivitycan increase to several thousand nucleotides. DNApolymerases in viruses, bacteria and archaebacteria,as well as eukaryotes use such clamps for rapid andefficient replication of chromosomal DNA.1,2

lsevier Ltd. All rights reserve

ble-stranded DNA;, bromo-

try; BSA, bovineybenzoic acid; TFA,

ing author:

Sliding clamps are composed of two (Escherichiacoli b clamp) or three subunits (bacteriophage T4gp45, Sacharomyces cerevisiae PCNA, human PCNA),arranged in the form of a ring with a central cavitywide enough to accommodate double-strandedDNA (dsDNA).3–8 Upon encircling the duplex,clamps are linked topologically to DNA, and yetfree to move on it; therefore, they can serveeffectively as mobile tethers for polymerases duringDNA synthesis. Several recent reports indicate thatcircular sliding clamps also play important roles inother cellular processes, including DNA repair andrecombination, DNA methylation, chromatinremodeling, and cell-cycle control, perhaps byhelping target key proteins in these processes totheir sites of action on DNA.9,10

Circular sliding clamps must be loaded ontoprimed sites on template DNA by multi-proteincomplexes known as clamp loaders.1 These proteinsuse ATP to fuel their actions, which include binding

d.

1458 DNA-binding Site on the Clamp Loader

the clamp, opening it, binding the DNA, andfacilitating closure of the clamp around the duplexportion of the primer-template.11–15 Consistent withtheir essential role in DNA metabolism andpossibly other cellular processes, clamp loaderproteins appear to be conserved across evol-ution.16–18 Numerous studies of clamp loaders,including the E. coli g complex, S. cerevisiae, Pyr-ococcus furiosus, and human RFC complexes, haverevealed detailed information about their structureand mechanisms of action. For example, g complex,the clamp loader of E. coli DNA polymerase IIIholoenzyme, is composed of five different proteins,g/t, d, d 0, c, and j, with three copies of g/t and oneeach of d and d 0 forming the minimal functionalbody of the loader.15,19,20 (c and j serve accessoryfunctions, such as coordinating clamp assemblywith primase and single-stranded DNA (ssDNA)binding protein activity at the replication fork.)21–23

The g, d, and d 0 subunits are arranged in apentameric ring in the shape of a claw, with the bclamp binding sites at the tips of the fingers (seemodel in Discussion).19,24 The g/t subunits bindand hydrolyze ATP and serve as the motors of theclamp-loading machine (g/t belong to the AAAC

ATPase family).16,25,26 The d subunit is the maincontact between g complex and the b clamp, andcan open the clamp by itself.11 The d 0 and g subunitsmodulate interaction between d and b.11,27,28 ATPbinding to the g/t subunits triggers conformationalchanges in g complex that allow d to bind b withhigh affinity and open the ring.12,24,29 The ATP-bound g complex-b complex binds primer-templateDNA with high affinity, presumably positioning itwithin the central cavity of the opened ring.12,30 TheDNA-binding event triggers rapid ATP hydrolysisat the g subunits, which is accompanied by areduction in g complex affinity for both b andDNA.13,14,31–33 Release of g complex from b andDNA, and closure of the clamp around DNAcomplete the assembly process, following whichDNA polymerase (or other proteins) can bind theclamp and commence work on DNA.

Clamp loaders from other organisms are com-posed of multiple subunits: the bacteriophage T4clamp loader has four copies of the gp44 subunitand one copy of gp62;34 the S. cerevisiae and humanRFC clamp loaders contain one copy each of fivedifferent proteins, RFC1, RFC2, RFC3, RFC4 andRFC5;35–38 archaebacterial clamp loaders containtwo proteins, RFC-l and RFC-s.39–41 The gp44 andthe RFC proteins share sequence similarities with gand d 0, and are members of the AAAC family; thus,like g complex, these clamp loaders utilize multipleATPase-active subunits for clamp assembly.16,17 Anew report on S. cerevisiae RFC structure from theKuriyan research group shows that the five RFCsubunits adopt a claw-like arrangement, reminis-cent of g complex.42 Electron microscopy images ofhuman RFC and P. furiosus RFC43 show the fivesubunits in a pentameric ring arrangement, andindicate ATP-dependent changes in clamp loaderconformation.44 The conservation of many elements

of clamp loader structure and biochemical activityacross different species suggests a common overallmechanism of clamp assembly.

A key piece of information missing from ourknowledge of the clamp loader mechanism is howthis multi-protein complex binds DNA and recog-nizes primer-template as the target site for clampassembly. Thus far, no discrete DNA-binding site(s),or even subunit(s) in the clamp loader has beenidentified as having DNA-binding activity that isessential for clamp assembly. Prior reports indicatethat an N-terminal domain in RFC1 that shareshomology with DNA ligase can bind DNA;45–47

however, removal of this domain from RFC1appears to have no effect on RFC clamp-loadingactivity.48–50 The g complex and S. cerevisiae RFCcan interact with ssDNA and primer-templateDNA, and RFC can even bind dsDNA with highaffinity.12,51 Nonetheless, g complex ATPase kineticsreveal that only primer-template DNA triggersrapid ATP hydrolysis, which in turn results inrelease of the DNA and clamp.30,33 In the case ofRFC as well, primer-template DNA stimulates thegreatest increase in steady-state ATPase activity.52

These data indicate that clamp loaders have arefined DNA-binding activity that is capable ofdistinguishing primer-template DNA from otherDNA structures, and is coupled to both clamp-binding/opening and ATPase activities to facilitateefficient assembly of clamps on primed DNAduring DNA replication.

In this study, we probed the identity of theDNA-binding subunit(s) on E. coli g complex andS. cerevisiae RFC clamp loaders. Using UV-inducedprotein–DNA cross-linking at specific locations onprimer-template DNA, we found that d, the clamp-binding/opening subunit, also contacts DNA at the3 0-OH primer-template junction. In the case of RFC,the RFC1 subunit appears to be the major contactbetween the clamp loader and DNA. Furthercharacterization of the cross-linked complexes bymass spectrometry revealed the DNA-binding siteon d, and these findings, in conjunction with theknown crystal structure of g complex, suggest apossible mechanism by which clamp loaders canrecognize primed DNA as the substrate for clampassembly.

Results and Discussion

The following sections describe: (a) preparationof DNA substrates with photo-reactive and radio-active nucleotides at desired locations, necessaryfor identification of proteins bound at the primer-template junction; (b) SDS-PAGE used to identifythe DNA-binding subunit in the g complex andS. cerevisiae RFC; (c) mass spectrometric analysis ofthe protein–DNA complex to identify the DNA-binding site; and (d) a model for primer-templateDNA binding and selection by the clamp loader forclamp assembly.

DNA-binding Site on the Clamp Loader 1459

Primer–template DNA with 5-bromodeoxyuridine(BrdU) and 32P inserted at specific locations

We have used site-specific photo-affinity labelingto map interactions between protein(s) in the multi-subunit clamp loader and primed DNA.53 Theschematic in Figure 1 shows polymerase-catalyzedsynthesis and purification of a DNA substratecontaining BrdU and 32P-labeled deoxyadenosinemonophosphate (dA). During biosynthesis, thepolymerase incorporates BrdU, [32P]dA, and thenthe other nucleotides into a DNA primer as definedby the template strand sequence. The duplex DNAproduct is bound to streptavidin beads (via thebiotin-tagged template) and the extended primerstrand is separated by alkali denaturation andpurified. This strand now serves as the templatefor primers of different lengths that position BrdUat various distances from the primer-templatejunction (primer-template DNA length is greaterthan the minimal length required for interactionwith g complex and RFC).54,55 BrdU allows zero-length cross-linking between protein and DNA, andis therefore a good probe for close protein–DNAcontacts.56 The 32P-label marks the protein linkedcovalently to DNA and, most importantly, proxi-mity of the 32P label to BrdU permits nearlycomplete degradation of DNA in the protein–DNA complex without loss of the radioactivemarker, allowing identification of the cross-linkedprotein simply by size.

A C-terminal domain of the d subunit in gcomplex cross-links DNA at the primer-templatejunction

Binding of g complex to p/tC1 primer-templateDNAwas examined first; p/tC1 contains two BrdUnucleotides on the single-stranded region immedi-ately flanking the 3 0-OH primer end (Figure 1). Thereactions were performed in the presence of ATPgS,as ATP binding to g complex facilitates its

interactions with both DNA and the b clamp and,in this regard, ATPgS is a good non-hydrolyzablemimic of ATP.12 Exposure of the protein and DNAto UV light resulted in a single predominantcomplex that was separated from free DNA bySDS-PAGE (Figure 2(A), lane 3; 3% yield); a controlexperiment showed no such complex formationwith DNA alone (Figure 2(A), lane 1). The protein–DNA complex was treated with DNaseI and S1nuclease, but the 32P label adjacent to BrdU wasprotected by the protein and the complex remainedradiolabeled (Figure 2(A), lane 4; in the correspond-ing control in lane 2 no free DNA remains aftertreatment with nuclease). Since the mass of theprotein–DNA complex is reduced to nearly that ofthe free protein following treatment with nuclease,comparison of the radioactive image and Coom-massie brilliant blue stain of the same gel revealsthe identity of the cross-linked protein/s (Figure2(A) and (B), lane 4). Because of their similar sizes,both d (39 kDa) and d 0 (37 kDa) subunits of gcomplex were implicated in DNA binding by thisanalysis. Further investigation of the interactionshowed that the extent of DNA cross-linking to dand/or d 0 increased from 3% to 10% when b wasadded to the reaction (Figure 2(A), compare lanes3/4 and 7/8), even though b itself does not formany covalent links with DNA (Figure 2(A), lanes 5and 6). The stimulatory effect of the clamp impliesthat the protein–DNA complex trapped by UVcross-linking is likely a relevant intermediate in theclamp assembly process. This effect is consistentwith prior reports that g complex-b has higheraffinity for primed DNA, and rapidly assemblesclamps on it, as compared with free g complex.12,31,32

Next we attempted to resolve whether d or d 0, orboth proteins, bind primer-template DNA. Thecross-linking experiment was performed withindividual g, d, and d 0 subunits, as well as acomplex of c/j (j alone is insoluble). Only dexhibits strong cross-linking to p/tC1 DNA (Figure

Figure 1. Design and synthesis ofprimed DNA templates containing5-bromodeoxyuridine and 32P atspecific locations. BrdU and[32P]dATP were incorporated intoindicated positions within a 54nucleotide DNA strand by T7DNA polymerase. The 54-mer waspurified and a 33 nucleotide primerannealed to it to form p/tC1, aprimed DNA substrate with two[32P]dA in the double-strandedregion and two BrdU in the single-stranded region immediately flank-ing the primer-template junction.Primers composed of 28 nucleo-tides and 23 nucleotides were usedto generate p/tC6 and p/tC11DNA substrates, respectively.

Figure 2. The d subunit in g complex cross-links primer-template DNA. Protein–DNA complexes formed by UV-induced cross-linking were analyzed by SDS-PAGE. (A) Phosphorimage and (B) Coommassie brilliant blue-stained gelshowing cross-linking of g complex to p/tC1 DNA: lanes 1 and 2 contain p/tC1 alone, loaded onto the gel before andafter treatment with DNaseICS1 nuclease, respectively; lanes 3 and 4 show similar analysis of p/tC1 DNA with gcomplex in the reaction; lanes 5 and 6 show p/tC1 with b clamp; lanes 7 and 8 show p/tC1 DNAwith g complex and b.(C) Phosphorimage and (D) Coommassie brilliant blue-stained gel showing similar analysis of p/tC1 cross-linking to gcomplex subunits: lanes 1 and 2 show g, lanes 3 and 4 show gCb, lanes 5 and 6 show d, and lanes 7 and 8 show dCbloaded onto the gel before (lanes 1, 3, 5 and 7) and after (lanes 2, 4, 6 and 8) treatment with nucleases. (E) Phosphorimageand (F) Coommassie brilliant blue-stained gel of a similar analysis as in (C and D), except with d 0 and c/j (K/Cb).


2(C) and (D), lanes 5 and 6; 11% cross-linking yield).A trace amount of cross-linked product detectedwith the g subunit, although not reproducible in allexperiments, may indicate contact between g andDNA aswell (Figure 2(C) and (D), lanes 1/2 and 3/4;0.4% yieldK/Cb). In the presence of b, the yield ofd-DNA increases from 11% to 20%, suggesting thatb binding to d (and b opening?) may stabilize theinteraction between d and DNA and/or helpposition d to favor cross-linking BrdU in the ptC1DNA (Figure 2C, compare lanes 5/6 and 7/8). Thus,the d subunit in g complex appears to play adominant role in DNA binding during clampassembly.

We questioned whether the interaction detected

between d and DNA is specific to a primed DNAtemplate. The BrdU and 32P-containing templatewas used alone (ssDNA), annealed to its com-plement (dsDNA), annealed to a 33 nt primer(p/tC1), a 28 nt primer (p/tC6), or a 23 nt primer(p/tC11), in cross-linking reactions with g complexin the presence of ATPgS (K/Cb) and with d(K/Cb). The g complex does cross-link to ssDNA(Figure 3(A), lane 1) but, unlike the single pre-dominant species observed with p/tC1 DNA(Figure 3(A), lane 5), ssDNA yields multiple cross-linked species (the origin of non-specific species isnot entirely clear). Following treatment with nucle-ase, however, the 32P label again marks d as thesubunit cross-linked to BrdU in ssDNA (Figure 3(A)

Figure 3. The d-DNA cross-link is specific to the primer-template junction. (A) Lanes 1 and 2 show cross-linking of gcomplex to single-stranded DNA (ssDNA), loaded onto the gel before and after treatment with nuclease, respectively;lanes 3 and 4 show similar analysis with double-stranded DNA; lanes 5 and 6 show p/tC1 DNA; lanes 7 and 8 showp/tC6 DNA and lanes 9 and 10 show p/tC11 DNA (containing BrdU 5 and 10 bases from the junction, respectively).(B) Similar analyses with g complexCb, (C) with the d subunit, (D) with dCb.


and (C), lanes 1 and 2). Interestingly, both p/tC6and p/tC11 DNA, which have BrdU positioned inthe single-stranded region at six and 11 nucleotidesfrom the primer-template junction, respectively,show reduced levels of cross-linking to g complexcompared with p/tC1 DNA (Figure 3(A), lanes5/6, 7/8, and 9/10). The effect is more striking inthe presence of b (p/tC1: 10%; p/tC6: 2%; p/tC11:2%; Figure 3(B), lanes 5/6, 7/8, and 9/10); accord-ing to nitrocellulose membrane filtration assays, gcomplex can bind all these DNAs (data not shown).Barely any cross-linking is detectable with thedsDNA substrate, which is consistent with priorreports of low or no interaction between g complexand duplex DNA12 but may reflect inaccessibility ofBrdU within the double helix. Cross-linking experi-ments with d indicate that it can distinguish theprimer-template junction by itself (Figure 3(C)(Kb) and D (Cb), compare lanes 5/6, 7/8, and9/10). Thus, it appears that the amino acidresidue(s) that can cross-link DNA are located ina region of d that binds the primer-templatejunction specifically.

Further investigation of the DNA binding/recog-nition properties of d and g complex was facilitatedby knowledge of the crystal structure of d, whichshows three distinct domains: N-terminal domain I(1K140), domain II (141K210), and C-terminal

domain III (211K343) (Figure 7(A)).19 The ddomains I (dI) and III (dIII) can be expressed athigh levels and in soluble form,24,57 and dI has infact been co-crystallized with the b clamp and issimilar in structure to domain I within d in the gcomplex (except for elements that change uponcontact with b).24 Experiments performedwith the ddomains show that dIII forms a cross-linkwith p/tC1DNA, but dI does not (Figure 4(A) and (B), lanes 3/4and 5/6, respectively). These data suggest that bothclamp-binding and DNA-binding activities of theclamp loader are located on the same protein, inseparate domains, I and III, respectively. Moreover,dIII exhibits some preference for p/tC1 DNA overthe other DNAs, although the difference is not asstriking as that seen for full-length d (Figure 4(C),compare lanes 1, 3, 5, and 7). However, dIII incomplex with g and d 0 (g3dIIId

0) can apparentlydistinguish a primer-template junction as it cross-links BrdU in p/tC1 DNA but not in p/tC6,p/tC11 DNAs (c and j were not included in thecomplex because they are not absolutely necessaryfor clamp assembly and because similarity in theirsizes with dIII would have complicated analysis).These results indicate that domain III is necessaryand sufficient for interaction between d and DNA,although specific placement of d at the primer-template junction may be most effective when

Figure 4. A C-terminal domain of d (dIII) is responsible for interaction with the primer-template junction. (A)Phosphorimage and (B) Coommassie brilliant blue-stained gel with p/tC1 and full-length d in lanes 1 and 2 (before andafter treatment with nuclease, respectively), N-terminal dI domain in lanes 3 and 4, and C-terminal dIII domain in lanes 5and 6. (C) Analysis of dIII and (D) g3dIIId

0 cross-linking to single-stranded DNA, p/tC1, p/tC6, and p/tC11 DNAs.


full-length d protein or d domain III bind DNAwithin the context of the g complex.

Identification of the cross-linked amino acidresidue in dIII and d by mass spectrometry

The UV cross-linking experiments yielded a fairlysubstantial amount of dIII-DNA complex (10–15%yield), enabling further analysis by mass spec-trometry (MS) and identification of the amino acidresidue(s) in close proximity to the 3 0-OH primerend. p/tC1 DNAwas synthesized on a large scale,and cross-linking of 1 nmol of DNA to dIII followedby SDS-PAGE and electro-elution from the gelyielded 30–50 pmol of pure dIII-DNA complex(Figure 5(A)). The complex was subjected toproteolysis by immobilized trypsin. Because of thehigh molecular mass of the DNA attached to theprotein, it proved convenient to treat the resultingtryptic peptides with nuclease to digest the majorityof the DNA.58 This treatment reduces them/z of thecross-linked peptide ions to a convenient range foranalysis (m/z!4000) and improves their MSresponse. Direct comparison of the MS profilebefore and after treatment with nuclease enabledus to distinguish and identify peptides modified byDNA cross-links among the other unmodified dIIItryptic peptides (Figure 5(B)). Six peaks emergedreproducibly in the MS spectrum after exhaustivetreatment with nuclease (Figure 5(B) and Table 1).These six peaks were analyzed by multi-stage MS inorder to identify the modified peptide, the modified

residue, and the nature of the modification (Figure5(C)). For example, the peak at m/z 1301.6 waspresent in the MS spectrum only after exhaustivetreatment with both DNaseI and S1 nuclease, butnot in the MS spectrum of a sample treated withonly DNaseI, suggesting it as a potential candidatecontaining the DNA cross-linked site. MS2 analysison 1301.6 gave characteristic neutral losses of 98 Daand 196 Da, which correspond to H3PO4 andphosphodeoxyribose, respectively. MS3 on thepeak at m/z 1105.5 yielded a dominant production at m/z 949.5, corresponding to loss of theC-terminal arginine residue. Further MS4 analysison m/z 949.5 gave a series of b and y ions thatindicated the original peptide sequence wasHRVWQNR, and that the tryptophan residue inthis peptide was modified with a uracilyl group.These results agreed well with the photo-cross-linking chemistry of the BrdU reagent used in theexperiment.59 Similar analysis of the other fivemodified peptide peaks revealed that they allcontained the same cross-linked tryptophan 279residue, and originated from partial cleavage bytrypsin or nuclease (Table 1). A similar approachwas applied to d-p/tC1 DNA complex, but becauseMS peaks from the cross-linked tryptic peptides offull-length dwere masked by background noise, weresorted to analysis by hypothesis-driven multi-stage MS as described.60 As shown in Figure 5(D),we found that even in the full-length d protein, thesame tryptophan 279 residue cross-links with theuracil base at the primer-template junction.

Figure 5. Differential peptide mass mapping and multiple stage mass spectrometry analysis of dIII$p/tC1 DNA andd$p/tC1 DNA complexes reveal tryptophan-279 as the site of cross-linking to DNA. (A) Phosphorimage andCoommassie brilliant blue-stained gel showing separation of cross-linked protein–DNA complex from impurities,including unreacted protein and DNA; lanes 1, 2 and 3 have dIII alone, dIIICp/tC1 DNA, and protein size marker,respectively. (B) Comparison of MS spectra of dIIIKp/tC1 DNA complex tryptic digest treated with (top) only DNaseI



RFC1 subunit in the S. cerevisiae clamp loadercross-links primer-template DNA

The five subunits of the S. cerevisiae clamp loader,RFC (RFC1-5), are homologous to the g and d 0

subunits of g complex; however, as there is no clearsequence similarity between any of these proteinsand d, it is not immediately obvious whichsubunit(s) in RFC binds and opens the PCNAclamp. RFC1 is proposed to be the functionalequivalent of d due to certain common structuralfeatures, including a pair of conserved hydrophobicresidues that contact PCNA.24,61 However, theRFC3 subunit also appears to bind PCNA, and itis not known which subunit actually opens theclamp for assembly on DNA.61,62 In the recentlysolved structure of S. cerevisiae RFC-PCNA, RFC1and RFC3 (and RFC 4 to some extent) contactPCNA.42 Given our finding that d possesses bothclamp-binding/opening and DNA-binding activi-ties, we wondered if RFC1 and/or RFC3 might alsobind primer-template DNA.

In UV cross-linking experiments, the large RFC1subunit was found linked to p/tC1 DNA (Figure 6,lanes 1 and 2). Less prominent products of the sizeof RFC2-5 were visible, but these were furtherreduced in the presence of PCNA (Figure 6(A),compare lanes 2 and 4). Since reports in theliterature indicate that an N-terminal domain inRFC1, which is unnecessary for clamp loading,binds DNA, we tested an RFC complex with RFC1replaced by RFC1S, which has the 282 amino acidresidue domain deleted from the N terminus.37

RFC1S also cross-links p/tC1 DNA and weakercross-links between the small RFC subunits andDNA are lost when PCNA is included in thereaction, highlighting the relevance of the contactbetween RFC1 and DNA for clamp assembly(Figure 6(A), lanes 5/6 and 7/8).

Implications for the mechanism of primer-template recognition and clamp assembly byclamp loader proteins

Of the five subunits in the g complex, d is the onlyone capable of opening the b clamp by itself. Whileessential, this action of d is not sufficient for clampassembly, as g and d 0 are required for placement ofthe clamp around a primed DNA template.11 Sincethe identity of the DNA-binding subunit wasunknown, it was considered possible that the gand/or d 0 subunits were required for interactionbetween g complex and DNA, and therefore d alonecould not function as a clamp loader. Here, we showthat the d protein has both clamp-binding/openingand primer-template binding/recognition activi-ties, located on two separate domains.

and (bottom) both DNaseI and S1 nuclease. Labeled peaks emecontained the DNA cross-linked site. (C) Identification ofm/zof cross-linking using multiple stage MS. (D) Identification ofsite of cross-linking using multiple stage MS.

Tryptophan 279 found cross-linked at the primer-template junction is in a cationic amino acid-richregion in the C-terminal domain of d (dIII; Figure7).42 The tryptophan flanks a narrow pass in d, withtwo lysine residues located on the other side of thepass (lysine 113 and lysine 116 in domain I). Theaverage width of the pass appears to be about 12 A,which is wide enough to accommodate ssDNA, butnot dsDNA. The correspondence between the cross-linking data and the protein structure in proximityto the cross-link suggests a pathway for entry ofDNA into the g complex as well as a means forselection of primer-template as the substrate forclamp assembly (Figure 7(A)). Once the clamploader-open clamp complex is formed, primedDNA can be positioned in it with part of thedouble-stranded region in the central cavity of theclamp, and the rest, including the primer-templatejunction and single-stranded region, located withinthe dome-shaped chamber of the g complex. As thetemplate strand goes through the pass, the lysineresidues may help position the DNA backbone,42

while the tryptophan residue stacks against a baseat or near the primer-template junction, andperhaps helps lock the DNA substrate into placefor clamp assembly (Figure 7(B)). Viewed from thetop of the g complex, the template appears to exitfrom the gap between the d and d 0 subunits(consistent with proposed paths for DNA instructural analyses of the S. cerevisiae and P. furiosusRFC clamp loaders).42,43 In this model pathway forDNA, the tryptophan residue is in the right locationto form a covalent link with a UV-induced uracilylradical in the template strand flanking the primer-template junction. Tryptophan 279 function may beanalogous to the “bracketing” action of tryptophanresidues in the structure of p19 protein bound to itssiRNA substrate, a 19-bp duplex with 3 0-dinucleo-tide overhangs at both ends.63,64 In this case,tryptophan residues stack over the terminal base-pairs of the substrate, facilitating sequence-inde-pendent selection of duplex RNAs of specificlengths.

The defined position of BrdU in our DNAsubstrates allowed us to identify an amino acidpresent near that location. However, BrdU tends tocouple most easily to tryptophan residues, particu-larly when excited at shorter wavelengths (254 nmin this case),56 so it is highly likely that there areother interactions between g complex/RFC alongthe length of the DNA substrate that are notdetected by this particular analysis. For example,the lysine residues in d domain I, which also appearto be in close proximity to the DNA, were notpicked up in the cross-linking analysis with DNA.In fact, the RFC structure reported by the Kuriyanresearch group shows a track of positive

rged after treatment with S1 nuclease, indicating that they1301.6 cross-linked peptide from dIII-p/tC1 DNA and sitem/z 1301.6 cross-linked peptide from d-DNA p/tC1 and

Figure 6. The RFC1 subunit in S. cerevisiae RFC clamp loader, which is analogous to the d subunit, cross-links primedDNA. (A) Phosphorimage and (B) Coommassie brilliant blue-stained gel showing cross-linking of RFC with p/tC1DNA in the absence (lanes 1 and 2) and in the presence (lanes 3 and 4) of PCNA; lanes 5–8 show similar analysis of RFCcontaining an N-terminal-truncated RFC1 subunit (RFC1s).


electrostatic potential along all the RFC subunitsthat matches the minor groove of a double helixplaced within the central cavity in the complex.42

The cross-link formed between tryptophan 279 andBrdU at the 3 0-OH end of the primer does, however,highlight a special role for d versus the other clamploader subunits in primer-template selection. We do

Figure 7. Possible positioning of primed DNAwithin the cl(A) A model of the open b clamp docked onto g complex, illusduplex portion passing through the b ring and the single-stdomain III of d. The structure of d is shown with the DNA-lysine 116 and lysine 119 colored purple (the alphabet nomencthe S. cerevisiae RFC crystal structure).11 (B) Another view of thagainst the primer-template junction.

not know whether the observed cross-link betweenRFC1 and DNA also reflects interaction between atryptophan residue (or another base-stacking resi-due) and the primer-template junction. However,previous studies have noted several similaritiesbetween RFC1 and d, including the fact that bothRFC1 and d contact their clamps; both lack the SRC

amp loader-open clamp complex during clamp assembly.trating a possible path for primer-template DNAwith theranded portion exiting the complex near the C-terminalcross-linking residue, tryptophan 279, colored green andlature for g complex subunits corresponds to that used fore complex, illustrating possible stacking of tryptophan 279


amino acid motif essential for ATPase activity; bothdo not contribute ATPase activity towards clampassembly; both are most divergent in sequence fromother subunits in their respective complexes; bothappear to be positioned similarly within theirrespective complexes, relative to the other subunits,and these observations fuel the hypothesis thatRFC1 functions like d in clamp assembly. Here, weadd to this list of similarities by demonstrating thatboth proteins bind primer-template DNA. Furtheranalysis is underway to identify the amino acid(s)in RFC1 that contacts DNA (note: the RFC1 DNA-binding site detected in this study is not located inthe N-terminal ligase homology domain, which isknown to bind DNA but is unnecessary for clampassembly).48–50

The d sequence is poorly conserved, making itdifficult to predict even its functional homologue inother clamp loaders, let alone the location of theDNA-binding site in the said d homologue.Recently, however, the McHenry research groupused an iterative j-blast approach to identify dsequences in known bacterial genomes and notedsix conserved regions in these proteins.57 Of theamino acid residues we have identified as potentialcontributors to the DNA-binding site in d, lysine 116and lysine 119 are conserved in Region 3. Trypto-phan 279 in d does not lie in a conserved region;however, most of the proteins contain tryptophan,tyrosine, or phenylalanine residues at or within twoto three residues of that location.57 Thus, it ispossible that the DNA-binding site and mechanismfor recognition of primer-template DNA might beconserved among bacterial clamp loaders.

An interesting corollary to the above discussion isthe growing evidence that alternate RFC complexesfunction in DNA metabolic processes other thangenomic DNA replication. These eukaryotic RFC-like complexes are pentamers of RFC2, RFC3, RFC4,and RFC5 subunits plus an RFC1 homologue, suchas Rad17 (S. cerevisiae Rad24), Ctf18/Chl12, or Egl1,and are known to clamp proteins other than PCNAonto DNA (e.g. human Rad1-Hus1$Rad9 andS. cerevisiae Ddc1-Mec3-Rad17).65,66 A recent reporton the human Rad17-RFC2-5 complex revealed thatit catalyzes assembly of Rad1-Hus1-Rad9 preferen-tially on primer-template DNA with a recessed 5 0

primer end, in marked contrast to the RFC complex,which loads PCNA onto primer-template DNAwith a recessed 3 0 primer end.67 The protein-loadingactivity of alternate RFC complexes is proposed toplay a role in recognition/resolution of DNAstructures containing recessed 5 0 ends, such as areformed during stalled DNA replication, nucleotideexcision repair, recombination-coupled DNArepair, and telomere maintenance.68 These obser-vations are strikingly consistent with our findingthat the RFC1 subunit in RFC complex bindsprimer-template DNA with a recessed 3 0 end.Perhaps RFC1-like proteins perform analogousfunctions with other DNA structures.

An exact understanding of the mechanism ofclamp loader action awaits further structural and

biochemical studies; in the meantime, however, it istempting to speculate the existence of a primordialclamp loader comprising a single protein that couldbind DNA, bind and open the clamp, and load theclamp on DNA, and perhaps even had ATPaseactivity to drive needed conformational changes.Multiplication of this primordial clamp loaderpossibly yielded better clamp-loading efficiencyand regulation of activity; however, the keyactivities of clamp binding/opening and DNAsubstrate recognition were retained on one subunit.

Experimental Procedures

Preparation of protein and nucleotide reagents

The g, d, d 0, c, and j proteins were purified and the gcomplex reconstituted,11 S. cerevisiae RFC (containingRFC1 or RFC1S) was over-expressed and purified fromE. coli,37 dI was cloned and purified,24 and E. coli b3 andS. cerevisiae PCNA51 were purified as described in earlierreports. dIII, a 212–343 amino acid residue fragment of d,was cloned into pET11a, over-expressed in E. coli, andpurified over an SP Sepharose column (AmershamBiosciences; 100–700 mM NaCl gradient). BacteriophageT7 DNA polymerase (exoK mutant) was a gift from DrSmita Patel (Robert Wood Johnson Medical School),thioredoxin was purchased from Sigma-Aldrich, andDeoxyribonuclease I and S1 nuclease were purchasedfrom Life Technologies.The 5-BrdUTP was purchased from Sigma-Aldrich,

other dNTPs were purchased from Amersham Bio-sciences, and [a-32P]dATP was purchased from PerkinElmer. Short synthetic DNAs were purchased fromIntegrated DNA Technologies:ST-5 0 biotin-TAGCAGGGTGGAGGTTGTCAGGGTA

GGGGGCTTAACGGCGCGGTCTTAGTCAGC 3 0

SP-5 0 GCTGACTAAGACCGCGCCG 3 0

PC1-5 0 TAGCAGGGTGGAGGTTGTCAGGGTAGGGGGCTT 3 0

PC6-50 TAGCAGGGTGGAGGTTGTCAGGGTAGGG30

PC11-5 0 TAGCAGGGTGGAGGTTGTCAGGG 3 0

For synthesis of BrdU and 32P-containing templateDNA, first the ST and SP DNAs were annealed at aconcentration of 5–25 mM (90 8C for five minutes and slowcooling to 25 8C in 20 mM Tris–HCl (pH 7.5), 150 mMNaCl). SP/ST at a concentration of 0.2 mM (or 2.5 mM forlarge-scale synthesis) was extended by 0.2 mM phage T7DNA polymerase (pre-incubated for 20 minutes with1 mM thioredoxin, 3 mMDTT) at 37 8C in 40 mMTris–HCl(pH 7.5), 12.5 mM MgCl2, 50 mM NaCl, 0.1 mg/ml ofbovine serum albumin (BSA), 1 mM DTT, 1 mM EDTA.Fifty micromolar BrdUTP and trace amounts of[a-32P]dATP were added to the reaction and incubatedfor ten minutes (25 minutes for large scale), followed byall four dNTPs at 1 mM each and further incubation for30 minutes. The reaction was mixed with MagnaBindStreptavidin beads (Pierce) for 30 minutes at 37 8C to bindthe DNA. The beads were collected and washed threetimes with 10 mM Tris–HCl (pH 7.5), 2 M NaCl, 1 mMEDTA, then incubated with 0.08 M NaOH for one minuteto denature the duplex, and then centrifuged to separateST (bound to the beads via biotin–streptavidin inter-action) from the extended SP DNA (released into thesupernatant). The supernatant was collected and dialyzedagainst 20 mM Hepes–NaOH (pH 7.5), 30 mM NaCl.0.2 mM (or 2 mM for large scale) of the DNAwas annealed


with PC1, PC6, or PC11 DNAs to prepare the primer-templates.

UV-induced protein–DNA cross-linking and SDS-PAGE analysis

The g complex, g3dd0cj, or g3dIIId

0 (1 mM), d (2.5 mM),dIII (5 mM), or RFC (1 mM)was incubated with 0.05 mMp/tDNA (or 1 mM DNA for large scale) and 200 mM ATPgS(in the case of g complexes) in 30 ml of 30 mM Hepes–NaOH (pH 7.5), 4 mMMgCl2 for tenminutes at 25 8C. Thereactions were placed on ice and exposed to UV light(254 nm) in a StrataLinker (Stratagene) for 25 minutes.Following cross-linking, each reaction was split into two:(a) 15 ml was quenched with 6 ml of SDS dye (1.5 M Trisbase, 3% (w/v) SDS, 35% (v/v) glycerol, 150 mM DTT,bromophenol blue) C 6 ml of 0.5 M EDTA; (b) 15 ml wastreated with 0.2 ml of DNaseI (200 units/ml) for tenminutes at 25 8C, then with 0.5 ml of 10% SDS and boilingfor threeminutes to destroy DNaseI, followed by additionof 2 ml of 10!S1 nuclease buffer and 1 ml of S1 nuclease(20 units/ml) and incubation at 37 8C for five minutes,after which the reaction was quenchedwith 0.5 ml of 0.5 MTris base and 15 ml of SDS/dye. The reactions were boiledand analyzed by SDS-PAGE. The gels were scanned byPhosphorimager (Molecular Dynamics) and then stainedby GelCode Blue stain (Pierce). For mass spectrometryanalysis, the protein–DNA band was cut out of the geland eluted using the BioRad Electro-Eluter (50 mMNH4HCO3 with 0.1% SDS for six hours and withoutSDS for four hoursmore). The eluate was dialyzed against20 mM Tris–HCl (pH 7.5), 30 mM NaCl and dried undervacuum. The pellet was washed twice with 50 ml of water,0.5 ml of 1 M HCl, 450 ml of acetone (12 hours at K20 8C)and twice with 50 ml of water/150 ml of acetone to removeSDS.

Protease digestion and mass spectrometry analysis

dIII-DNA and d-DNAwere dissolved in 20 ml of 50 mMNH4HCO3/acetonitrile (9/1, v/v), and digested with1.5 ml of immobilized trypsin (Pierce) at 37 8C, shaking at900 rpm for five hours. Next, 125 pg DN-EP (Sigma-Aldrich) was added to digest duplex DNA at roomtemperature for two hours; then 2 ml of the reactionproduct was removed for MS analysis. The rest of thesample was neutralized to pH 5 with 10% acetic acid andthen zinc acetate (to 1 mM) and 0.5 unit of S1 nucleaseadded to digest ssDNA at room temperature for eighthours. MS spectra were collected on an in-housemodified prototype Sciex QqTOF mass spectrometer.Tandem mass spectrometry experiments were performedon an in-house assembled matrix-assisted laser desorp-tion/ionization (MALDI)-ion trap mass spectrometer thatincorporates a Finnigan LCQ DecaXP (Thermo Electron)mass analyzer. 2,5-Dihydroxybenzoic acid (DHB) (Sigma-Aldrich) was prepared as a saturated solution inmethanol/acetonitrile/water/trifluoroacetic acid (TFA)(50/20/30/0.1, by vol.) and diluted threefold in thesame solution. Portions (2 ml) of the samples (before andafter treatment with S1 nuclease) were mixed with 6 ml ofdiluted DHB solution and deposited onto MALDI targetCD at 4 ml per spot, and air-dried before the massspectrometry analysis. For the multi-stage MS experi-ments, samples were further concentrated using reversedphase oligoR3 beads (Applied Biosystems) and elutedwith 4 ml of 1/4 saturated DHB solution.

Acknowledgements

This work was supported by grants GM64514-01(to M.M.H), RR00862 (to B.T.C), and GM38839 (toM.O’D). We thank Smita S. Patel for the generousgift of T7 DNA polymerase (exoK mutant), IshitaMukerji and Kelly Knee for tests with UV lasercross-linking, David Jeruzalmi and Gregory Bow-man for helpful ideas and discussions, and San-chaita Das for artistic input.

References

1. Jeruzalmi, D., O’Donnell, M. & Kuriyan, J. (2002).Clamp loaders and sliding clamps. Curr. Opin. Struct.Biol. 12, 217–224.

2. Kelman, Z. & O’Donnell, M. (1995). DNA polymeraseIII holoenzyme: structure and function of a chromo-somal replicating machine. Annu. Rev. Biochem. 64,171–200.

3. Kong, X. P., Onrust, R., O’Donnell, M. & Kuriyan, J.(1992). Three-dimensional structure of the b subunitof E. coli DNA polymerase III holoenzyme: a slidingDNA clamp. Cell, 69, 425–437.

4. Moarefi, I., Jeruzalmi, D., Turner, J., O’Donnell, M. &Kuriyan, J. (2000). Crystal structure of the DNApolymerase processivity factor of T4 bacteriophage.J. Mol. Biol. 296, 1215–1223.

5. Krishna, T. S., Kong, X. P., Gary, S., Burgers, P. M. &Kuriyan, J. (1994). Crystal structure of the eukaryoticDNA polymerase processivity factor PCNA. Cell, 79,1233–1243.

6. Gulbis, J. M., Kelman, Z., Hurwitz, J., O’Donnell, M. &Kuriyan, J. (1996). Structure of the C-terminal regionof p21(WAF1/CIP1) complexed with human PCNA.Cell, 87, 297–306.

7. Matsumiya, S., Ishino, Y. & Morikawa, K. (2001).Crystal structure of an archaeal DNA sliding clamp:proliferating cell nuclear antigen from Pyrococcusfuriosus. Protein Sci. 10, 17–23.

8. Hingorani, M. M. & O’Donnell, M. (2000). Slidingclamps: a (tail)ored fit. Curr. Biol. 10, R25–R29.

9. Maga, G. & Hubscher, U. (2003). Proliferating cellnuclear antigen (PCNA): a dancer with many part-ners. J. Cell. Sci. 116, 3051–3060.

10. Warbrick, E. (2000). The puzzle of PCNA’s manypartners. Bioessays, 22, 997–1006.

11. Turner, J., Hingorani, M. M., Kelman, Z. & O’Donnell,M. (1999). The internal workings of a DNA polymer-ase clamp-loading machine. EMBO J. 18, 771–783.

12. Hingorani, M. M. & O’Donnell, M. (1998). ATPbinding to the Escherichia coli clamp loader powersopening of the ring-shaped clamp of DNA polymer-ase III holoenzyme. J. Biol. Chem. 273, 24550–24563.

13. Williams, C. R., Snyder, A. K., Kuzmic, P., O’Donnell,M. & Bloom, L. B. (2004). Mechanism of loading theEscherichia coli DNA polymerase III sliding clamp: I.Two distinct activities for individual ATP sites in the gcomplex. J. Biol. Chem. 279, 4376–4385.

14. Snyder, A. K., Williams, C. R., Johnson, A., O’Donnell,M. & Bloom, L. B. (2004). Mechanism of loading theEscherichia coli DNA polymerase III sliding clamp: II.Uncoupling the b and DNA binding activities of the gcomplex. J. Biol. Chem. 279, 4386–4393.

15. Onrust, R., Finkelstein, J., Naktinis, V., Turner, J.,Fang, L. & O’Donnell, M. (1995). Assembly of achromosomal replication machine: two DNA


polymerases, a clamp loader, and sliding clamps inone holoenzyme particle. I. Organization of the clamploader. J. Biol. Chem. 270, 13348–13357.

16. Iyer, L. M., Leipe, D. D., Koonin, E. V. & Aravind, L.(2004). Evolutionary history and higher order classi-fication of AAAC ATPases. J. Struct. Biol. 146, 11–31.

17. Davey, M. J., Jeruzalmi, D., Kuriyan, J. & O’Donnell, M.(2002). Motors and switches: AAAC machines withinthe replisome. Nature Rev. Mol. Cell. Biol. 3, 826–835.

18. O’Donnell, M., Onrust, R., Dean, F. B., Chen, M. &Hurwitz, J. (1993). Homology in accessory proteins ofreplicative polymerases–E. coli to humans.Nucl. AcidsRes. 21, 1–3.

19. Jeruzalmi, D., O’Donnell, M. & Kuriyan, J. (2001).Crystal structure of the processivity clamp loader gcomplex of E. coli DNA polymerase III. Cell, 106,429–441.

20. Pritchard, A. E., Dallmann, H. G., Glover, B. P. &McHenry, C. S. (2000). A novel assembly mechanismfor the DNA polymerase III holoenzyme DnaXcomplex: association of dd 0 with DnaX(4) formsDnaX(3)dd 0. EMBO J. 19, 6536–6545.

21. Kelman, Z., Yuzhakov, A., Andjelkovic, J. &O’Donnell, M. (1998). Devoted to the laggingstrand-the subunit of DNA polymerase III holo-enzyme contacts SSB to promote processiveelongation and sliding clamp assembly. EMBO J. 17,2436–2449.

22. Glover, B. P. &McHenry, C. S. (1998). The c j subunitsof DNA polymerase III holoenzyme bind to single-stranded DNA-binding protein (SSB) and facilitatereplication of an SSB-coated template. J. Biol. Chem.273, 23476–23484.

23. Yuzhakov, A., Kelman, Z. & O’Donnell, M. (1999).Trading places on DNA—a three-point switchunderlies primer handoff from primase to thereplicative DNA polymerase. Cell, 96, 153–163.

24. Jeruzalmi, D., Yurieva, O., Zhao, Y., Young, M.,Stewart, J., Hingorani, M. et al. (2001). Mechanism ofprocessivity clamp opening by the d subunit wrenchof the clamp loader complex of E. coli DNA poly-merase III. Cell, 106, 417–428.

25. Johnson, A. & O’Donnell, M. (2003). Ordered ATPhydrolysis in the g complex clamp loader AAACmachine. J. Biol. Chem. 278, 14406–14413.

26. Podobnik, M., Weitze, T. F., O’Donnell, M. & Kuriyan,J. (2003). Nucleotide-induced conformational changesin an isolated Escherichia coli DNA polymerase IIIclamp loader subunit. Structure (Camb), 11, 253–263.

27. Indiani, C. & O’Donnell, M. (2003). Mechanism of thedwrench in opening the b sliding clamp. J. Biol. Chem.278, 40272–40281.

28. Leu, F. P. & O’Donnell, M. (2001). Interplay of clamploader subunits in opening the b sliding clamp ofEscherichia coli DNA polymerase III holoenzyme.J. Biol. Chem. 276, 47185–47194.

29. Goedken, E. R., Levitus, M., Johnson, A., Bustamante,C., O’Donnell, M. & Kuriyan, J. (2004). Fluorescencemeasurements on the E. coli DNA polymerase clamploader: implications for conformational changesduring ATP and clamp binding. J. Mol. Biol. 336,1047–1059.

30. Ason, B., Bertram, J. G., Hingorani, M. M., Beechem,J. M., O’Donnell, M., Goodman, M. F. & Bloom, L. B.(2000). A model for Escherichia coli DNA polymeraseIII holoenzyme assembly at primer/template ends.DNA triggers a change in binding specificity of the gcomplex clamp loader. J. Biol. Chem. 275, 3006–3015.

31. Bloom, L. B., Turner, J., Kelman, Z., Beechem, J. M.,

O’Donnell, M. & Goodman, M. F. (1996). Dynamics ofloading the b sliding clamp of DNA polymerase IIIonto DNA. J. Biol. Chem. 271, 30699–30708.

32. Bertram, J. G., Bloom, L. B., Hingorani, M. M.,Beechem, J. M., O’Donnell, M. & Goodman, M. F.(2000). Molecular mechanism and energetics of clampassembly in Escherichia coli. The role of ATP hydroly-sis when g complex loads b on DNA. J. Biol. Chem.275, 28413–28420.

33. Ason, B., Handayani, R., Williams, C. R., Bertram, J. G.,Hingorani, M. M., O’Donnell, M. et al. (2003). Mechan-ism of loading the Escherichia coliDNApolymerase III bsliding clamp on DNA. Bona fide primer/templatespreferentially trigger the g complex to hydrolyze ATPand load the clamp. J. Biol. Chem. 278, 10033–10040.

34. Benkovic, S. J., Valentine, A. M. & Salinas, F. (2001).Replisome-mediated DNA replication. Annu. Rev.Biochem. 70, 181–208.

35. Cullmann, G., Fien, K., Kobayashi, R. & Stillman, B.(1995). Characterization of the five replication factor Cgenes of Saccharomyces cerevisiae. Mol. Cell. Biol. 15,4661–4671.

36. Burgers, P. M. (1999). Overexpression of multisubunitreplication factors in yeast. Methods, 18, 349–355.

37. Finkelstein, J., Antony, E., Hingorani, M. M. &O’Donnell, M. (2003). Overproduction and analysis ofeukaryotic multiprotein complexes in Escherichia coliusing a dual-vector strategy. Anal. Biochem. 319, 78–87.

38. Cai, J., Uhlmann, F., Gibbs, E., Flores-Rozas, H., Lee,C. G., Phillips, B. et al. (1996). Reconstitution ofhuman replication factor C from its five subunits inbaculovirus-infected insect cells. Proc. Natl Acad. Sci.USA, 93, 12896–12901.

39. Kelman, Z. & Hurwitz, J. (2000). A unique organiz-ation of the protein subunits of the DNA polymeraseclamp loader in the archaeon Methanobacteriumthermoautotrophicum deltaH. J. Biol. Chem. 275,7327–7336.

40. Oyama, T., Ishino, Y., Cann, I. K., Ishino, S. &Morikawa, K. (2001). Atomic structure of the clamploader small subunit from Pyrococcus furiosus. Mol.Cell, 8, 455–463.

41. Seybert, A., Scott, D. J., Scaife, S., Singleton, M. R. &Wigley, D. B. (2002). Biochemical characterisation of theclamp/clamp loader proteins from the euryarchaeonArchaeoglobus fulgidus. Nucl. Acids Res. 30, 4329–4338.

42. Bowman, G. D., O’Donnell, M. & Kuriyan, J. (2004).Structural analysis of a eukaryotic sliding clamp-clamp loader complex. Nature, 429, 724–730.

43. Miyata, T., Oyama, T., Mayanagi, K., Ishino, S., Ishino,Y. &Morikawa, K. (2004). The clamp-loading complexfor processive DNA replication. Nature Struct. Mol.Biol. 11, 632–636.

44. Shiomi, Y., Usukura, J., Masamura, Y., Takeyasu, K.,Nakayama, Y., Obuse, C. et al. (2000). ATP-dependentstructural change of the eukaryotic clamp-loaderprotein, replication factor C. Proc. Natl Acad. Sci.USA, 97, 14127–14132.

45. Allen, B. L., Uhlmann, F., Gaur, L. K., Mulder, B. A.,Posey, K. L., Jones, L. B. & Hardin, S. H. (1998). DNArecognition properties of the N-terminal DNA bind-ing domain within the large subunit of replicationfactor C. Nucl. Acids Res. 26, 3877–3882.

46. Burbelo, P. D., Utani, A., Pan, Z. Q. & Yamada, Y.(1993). Cloning of the large subunit of activator 1(replication factor C) reveals homology with bacterialDNA ligases. Proc. Natl Acad. Sci. USA, 90,11543–11547.

47. Fotedar, R., Mossi, R., Fitzgerald, P., Rousselle, T.,


Maga, G., Brickner, H. et al. (1996). A conserveddomain of the large subunit of replication factor Cbinds PCNA and acts like a dominant negativeinhibitor of DNA replication in mammalian cells.EMBO J. 15, 4423–4433.

48. Uhlmann, F., Cai, J., Gibbs, E., O’Donnell, M. &Hurwitz, J. (1997). Deletion analysis of the largesubunit p140 in human replication factor C revealsregions required for complex formation and replica-tion activities. J. Biol. Chem. 272, 10058–10064.

49. Podust, V. N., Tiwari, N., Stephan, S. & Fanning, E.(1998). Replication factor C disengages from prolifer-ating cell nuclear antigen (PCNA) upon sliding clampformation, and PCNA itself tethers DNA polymerased to DNA. J. Biol. Chem. 273, 31992–31999.

50. Gomes, X. V., Gary, S. L. & Burgers, P. M. (2000).Overproduction in Escherichia coli and characteriz-ation of yeast replication factor C lacking the ligasehomology domain. J. Biol. Chem. 275, 14541–14549.

51. Hingorani, M. M. & Coman, M. M. (2002). On thespecificity of interaction between the Saccharomycescerevisiae clamp loader replication factor C andprimed DNA templates during DNA replication.J. Biol. Chem. 277, 47213–47224.

52. Gomes, X. V., Schmidt, S. L. & Burgers, P. M. (2001).ATP utilization by yeast replication factor C. II.Multiple stepwise ATP binding events are requiredto load proliferating cell nuclear antigen onto primedDNA. J. Biol. Chem. 276, 34776–34783.

53. Persinger, J. & Bartholomew, B. (2001) Site-directedDNA photoaffinity labeling of RNA polymerase IIItranscription complexes. In Methods in MolecularBiology, (Moss, T., ed.), vol. 148, DNA-Protein inter-actions: Principles and Protocols, 2nd edit., p. 363,Human Press, Totowa, NJ.

54. Yao, N., Leu, F. P., Anjelkovic, J., Turner, J. &O’Donnell, M. (2000). DNA structure requirementsfor the Escherichia coli g complex clamp loader andDNA polymerase III holoenzyme. J. Biol. Chem. 275,11440–11450.

55. Tsurimoto, T. & Stillman, B. (1991). Replication factorsrequired for SV40 DNA replication in vitro. I. DNAstructure-specific recognition of a primer-templatejunction by eukaryotic DNA polymerases and theiraccessory proteins. J. Biol. Chem. 266, 1950–1960.

56. Dietz, T. M. & Koch, T. H. (1987). Photochemicalcoupling of 5-bromouracil to tryptophan, tyrosineand histidine, peptide-like derivatives in aqueousfluid solution. Photochem. Photobiol. 46, 971–978.

57. Bullard, J. M., Pritchard, A. E., Song,M. S., Glover, B. P.,Wieczorek, A., Chen, J. et al. (2002). A three-domainstructure for the d subunit of the DNA polymerase IIIholoenzyme d domain III binds d 0 and assembles intothe DnaX complex. J. Biol. Chem. 277, 13246–13256.

58. Qin, J. & Chait, B. T. (1997). Identification andcharacterization of posttranslational modifications ofproteins by MALDI ion trap mass spectrometry. Anal.Chem. 69, 4002–4009.

59. Steen, H., Petersen, J., Mann, M. & Jensen, O. N.(2001). Mass spectrometric analysis of a UV-cross-linkedprotein–DNAcomplex: tryptophans 54and88ofE. coli SSB cross-link to DNA. Protein Sci. 10, 1989–2001.

60. Kalkum,M., Lyon, G. J. & Chait, B. T. (2003). Detectionof secreted peptides by using hypothesis-drivenmultistage mass spectrometry. Proc. Natl Acad. Sci.USA, 100, 2795–2800.

61. Yao, N., Coryell, L., Zhang, D., Georgescu, R. E.,Finkelstein, J., Coman, M. M. et al. (2003). Replicationfactor C clamp loader subunit arrangement within thecircular pentamer and its attachment points toproliferating cell nuclear antigen. J. Biol. Chem. 278,50744–50753.

62. Venclovas, C., Colvin, M. E. & Thelen, M. P. (2002).Molecular modeling-based analysis of interactions inthe RFC-dependent clamp-loading process. ProteinSci. 11, 2403–2416.

63. Ye, K., Malinina, L. & Patel, D. J. (2003). Recognition ofsmall interfering RNA by a viral suppressor of RNAsilencing. Nature, 426, 874–878.

64. Vargason, J. M., Szittya, G., Burgyan, J. & Tanaka Hall,T. M. (2003). Size selective recognition of siRNA by anRNA silencing suppressor. Cell, 115, 799–811.

65. Bermudez, V. P., Lindsey-Boltz, L. A., Cesare, A. J.,Maniwa, Y., Griffith, J. D., Hurwitz, J. & Sancar, A.(2003). Loading of the human 9-1-1 checkpointcomplex onto DNA by the checkpoint clamp loaderhRad17-replication factor C complex in vitro. Proc.Natl Acad. Sci. USA, 100, 1633–1638.

66. Majka, J. & Burgers, P. M. (2003). Yeast Rad17/Mec3/Ddc1: a sliding clamp for the DNA damagecheckpoint. Proc. Natl Acad. Sci. USA, 100, 2249–2254.

67. Ellison, V. & Stillman, B. (2003). Biochemical charac-terization of DNA damage checkpoint complexes:clamp loader and clamp complexes with specificityfor 5 0 recessed DNA. PLoS Biol. 1, E33.

68. Kim, J.&MacNeill, S.A. (2003).Genomestability: a newmember of the RFC family. Curr. Biol. 13, R873–R875.

Edited by J. Doudna

(Received 12 June 2004; received in revised form 27 July 2004; accepted 27 July 2004)

Documents

Dual Functions, Clamp Opening and Primer-Template Recognition