The Atad5 RFC-like complex is the major unloader of proliferating cell nuclear antigen in Xenopus egg extracts

Proliferating cell nuclear antigen (PCNA) is a homo-trimeric clamp complex that serves as the molecular hub for various DNA transactions, including DNA synthesis and post-replicative mismatch repair. Its timely loading and unloading are critical for genome stability. PCNA loading is catalyzed by Replication factor C (RFC) and the Ctf18 RFC-like complex (Ctf18-RLC), and its unloading is catalyzed by Atad5/Elg1-RLC. However, RFC, Ctf18-RLC, and even some subcomplexes of their shared subunits are capable of unloading PCNA in vitro, leaving an ambiguity in the division of labor in eukaryotic clamp dynamics. By using a system that specifically detects PCNA unloading, we show here that Atad5-RLC, which accounts for only approximately 3% of RFC/RLCs, nevertheless provides the major PCNA unloading activity in Xenopus egg extracts. RFC and Ctf18-RLC each account for approximately 40% of RFC/RLCs, while immunodepletion of neither Rfc1 nor Ctf18 detectably affects the rate of PCNA unloading in our system. PCNA unloading is dependent on the ATP-binding motif of Atad5, independent of nicks on DNA and chromatin assembly, and inhibited effectively by PCNA-interacting peptides. These results support a model in which Atad5-RLC preferentially unloads DNA-bound PCNA molecules that are free from their interactors.

Proliferating cell nuclear antigen (PCNA) is a homo-trimeric clamp complex that serves as the molecular hub for various DNA transactions, including DNA synthesis and postreplicative mismatch repair.Its timely loading and unloading are critical for genome stability.PCNA loading is catalyzed by Replication factor C (RFC) and the Ctf18 RFC-like complex (Ctf18-RLC), and its unloading is catalyzed by Atad5/Elg1-RLC.However, RFC, Ctf18-RLC, and even some subcomplexes of their shared subunits are capable of unloading PCNA in vitro, leaving an ambiguity in the division of labor in eukaryotic clamp dynamics.By using a system that specifically detects PCNA unloading, we show here that Atad5-RLC, which accounts for only approximately 3% of RFC/RLCs, nevertheless provides the major PCNA unloading activity in Xenopus egg extracts.RFC and Ctf18-RLC each account for approximately 40% of RFC/RLCs, while immunodepletion of neither Rfc1 nor Ctf18 detectably affects the rate of PCNA unloading in our system.PCNA unloading is dependent on the ATP-binding motif of Atad5, independent of nicks on DNA and chromatin assembly, and inhibited effectively by PCNA-interacting peptides.These results support a model in which Atad5-RLC preferentially unloads DNA-bound PCNA molecules that are free from their interactors.Eukaryotic chromosome replication is a complex process involving DNA synthesis, Okazaki fragment maturation, chromatin assembly, and post-replicative mismatch repair (MMR) (1,2).These processes are tightly regulated to occur in the right order at the right place.The replication clamp, a ringshaped protein complex structurally conserved among three domains of life, encircles DNA, translocates along DNA, and interacts with numerous factors related to replication and repair (3,4).Its loading and unloading are critical steps for the sequential and temporal regulation of the assembly and disassembly of multiprotein complexes required for various DNA transactions.
Since free PCNA forms a closed ring structure, its DNA loading must involve the opening of the ring.This process is catalyzed by the evolutionarily conserved clamp loader complexes (19)(20)(21)(22).The major eukaryotic clamp loader, Replication Factor C (RFC), is a hetero-pentameric complex of Rfc1-5, all of which share the AAA+ ATPase domain and are arranged in a spiral conformation (23,24).RFC has four active ATP-binding sites, and a series of studies demonstrate that sequential ATP binding to RFC induces RFC-PCNA complex formation, the spring-washer-like opening of the PCNA ring, and the binding of the RFC-PCNA complex onto the 3 0 -terminus of a primer-template junction (24)(25)(26)(27).Subsequent hydrolysis of ATP molecules facilitates the closure of the PCNA ring and the dissociation of RFC from PCNA (28).
Although evidence that Elg1/ATAD5-RLC is a primary PCNA unloader in vivo has been accumulated, PCNA unloading activity is clearly not specific to this complex.RFC is a pentameric complex of Rfc1, Rfc4, Rfc3, Rfc2, and Rfc5 arranged in this order, and Rfc1 is not essential for the opening of PCNA (51).Consistently, the human RFC4-RFC3-RFC2, the yeast Rfc4-Rfc3-Rfc2-Rfc5, and even the yeast Rfc2-Rfc5 subcomplexes are capable of unloading PCNA from nicked circular DNA, suggesting that the intrinsic PCNA unloading activity resides in the smaller subunits shared by all the clamploader/unloader complexes (51,52).In agreement with this notion, PCNA unloading activity has been detected in all RFC and RLCs.In yeast, both RFC and Rad24-RLC (the yeast counterpart of Rad17-RLC) unload PCNA from singly-nicked circular DNA in vitro (51), and human RFC unloads PCNA from nicked circular DNA in a purified system and covalently closed DNA after replication by the SV40 system (53,54).Likewise, yeast Ctf18-RLC unloads PCNA in vitro in the presence of replication protein A (RPA), a eukaryotic singlestranded DNA binding protein (55).These biochemical data raise the possibility that not only Elg1/ATAD5-RLC but also RFC, Ctf18-RLC, and Rad17-RLC contribute to PCNA unloading in vivo.However, the division of labor of RFC and other RLCs in PCNA unloading has not been fully determined, partly because the inactivation of clamp loaders masks their possible contribution to PCNA unloading by inhibiting the PCNA loading reaction.
To clarify the contribution of individual clamp loaders to PCNA unloading, we herein used an assay that specifically detects PCNA unloading in extracts prepared from the eggs of the frog Xenopus laevis, a physiological cell-free in vitro model system that recapitulates various DNA transactions, including PCNA loading/unloading and mismatch repair.Our data pinpoint Atad5-RLC as the major PCNA unloader in Xenopus egg extracts and support a model in which PCNA molecules that are free from their binding partners are preferentially targeted to be unloaded by Atad5-RLC.

Results
The nucleoplasmic extract (NPE) of Xenopus eggs is a nearly undiluted extract of nuclear proteins and functionally recapitulates various DNA transactions, including DNA replication, PCNA loading and unloading, and mismatch repair (13,56,57).To gain insight into the division of labor of clamp loaders, we first quantified the concentration of RFC and RLCs in NPE.Immunoprecipitation (IP) of a shared subunit Rfc3 removed most of the Rfc1, Ctf18, Atad5, and Rad17 proteins from NPE, suggesting that the majority of the large subunits are included in Rfc3-containing complexes (Fig. 1A, lane 2).Consistently, simultaneous IP of Rfc1, Ctf18, Atad5, and Rad17 co-depleted more than 90% of Rfc3 and Rfc5 from NPE, suggesting that most Rfc3 and Rfc5 are included in the RFC/RLC complexes (Fig. S1A).Therefore, we reasoned that the large and small RFC/RLC subunits form stoichiometric complexes in NPE.Using recombinant His-tagged Rfc3 purified from bacteria, we quantified the amount of Rfc3 in the Rfc1-IP fraction, and based on this number, we then estimated the concentration of RFC in NPE at approximately 1.1 μM.Similarly, we estimated the concentrations of Ctf18-RLC, Atad5-RLC, and Rad17-RLC at 1.3, 0.10, and 0.46 μM, respectively (Figs. 1 and S1B).
Biochemical studies with purified proteins have shown that not only Elg1/ATAD5-RLC but also RFC, Rad17-RLC, Ctf18-RLC, and several Rfc2-5 subcomplexes can unload PCNA from DNA (51)(52)(53)(54)(55).Given that Atad5-RLC represents only a minor fraction of total RFC/RLCs in Xenopus egg extracts, we next compared the PCNA unloading capability of all four RFC/ RLC complexes.We previously established an assay that specifically detects PCNA unloading in Xenopus egg extracts (13).Briefly, we load human PCNA (hPCNA) with human RFC, 2 to 10 trimers per DNA on average, onto nicked-circular DNA immobilized on Sepharose beads, ligate the nick, wash out unbound proteins including RFC and DNA ligase, and incubate the hPCNA-DNA complex in the extracts (Fig. 2A).Since hPCNA can functionally replace xPCNA in NPE but is specifically detected with a monoclonal antibody that does not recognize Xenopus PCNA (xPCNA), this system can discriminate the unloading events of hPCNA from the possible loading events of xPCNA.In a mock-treated NPE, hPCNA was quickly unloaded from covalently closed circular DNA within 20 min (Fig. 2, B and C).Rfc3 was recovered in the DNA-bead fraction, likely reflecting the DNA-binding affinity of RFC/ RLCs.In contrast, in the Rfc3-depleted NPE, where no Rfc3 Atad5-RLC-dependent PCNA unloading binding was observed, the rate of the unloading of hPCNA was significantly retarded, suggesting that most of the unloading events in NPE depend on the Rfc3-containing complexes.Although we observed some hPCNA dissociation in Rfc3depleted NPE, this level of PCNA dissociation likely reflects the intrinsic instability of hPCNA under our experimental condition since a similar level of hPCNA dissociation was observed in heat-inactivated Xenopus egg extracts (Fig. S2, A  and B).Importantly, Atad5-depletion slowed down the rate of hPCNA unloading to a similar level to that seen in Rfc3depleted NPE (Figs. 2, B-D, and S2C, see "ΔAtad5").In contrast, Rfc1-, Ctf18-, and Rad17-depletion did not detectably reduce the rate of hPCNA unloading compared to the mocktreated NPE (Figs. 2, B-D, and S2C).Rfc3 loading was significantly reduced in Rfc1-depleted NPE, suggesting that RFC has the strongest binding affinity to closed-circular DNA among all RFC/RLCs.Although these experiments do not exclude the possibility that RFC, Ctf18-RLC, and Rad17-RLC contribute to a minor fraction of hPCNA unloading, they suggest that Atad5-RLC is the major PCNA unloader in Xenopus egg extracts.
To study further the regulation of the PCNA unloading activity of Atad5-RLC, we expressed full-length human ATAD5 with N-terminal monomeric Azami-Green (mAG) and C-terminal FLAG tags in HEK293T cells, either wild-type or a mutant form of the Walker A motif (K1138E), along with smaller RFC2-5 subunits (46), and purified the respective complexes by FLAG affinity chromatography (Fig. 3, A and B).We found that the high-speed supernatant (HSS) of Xenopus eggs containing both cytosolic and nuclear proteins supports efficient PCNA unloading depending on Atad5 (Fig. 3C, lanes 1-5) and decided to use HSS as an easy-to-make alternative to NPE.Consistent with the prediction that Atad5-RLC is the major PCNA unloader in Xenopus egg extracts, the wild-type ATAD5-RLC restored PCNA unloading not only in Atad5depleted HSS but also in Rfc3-depleted HSS (Fig. 3, C and  D).In contrast, ATAD5-RLC carrying the K1138E mutation was not capable of restoring PCNA unloading in Atad5- depleted HSS (Fig. 3, E and F).These results suggest that PCNA unloading by Atad5-RLC depends on its ATP-binding or hydrolysis.Kubota et al. have shown that PCNA unloading occurs only after the completion of Okazaki fragment ligation in yeast (58).To gain further insight into the regulation of PCNA unloading by Atad5-RLC, we next tested whether single-strand nicks or chromatin assembly affects PCNA unloading in Xenopus egg extracts.First, we repeated our PCNA unloading assay in the presence of the Nb.BtsI nicking enzyme, which cleaves our DNA substrate at two loci.This enzyme efficiently converted closed-circular DNA into the open-circular form, suggesting continuous nicking in HSS, which supports efficient nick ligation (Fig. 4A, top panel).Despite nearly complete conversion of the plasmids into the open form, however, the rate of PCNA unloading was not detectably affected by the presence of the nicking enzyme (Fig. 4A, lanes 2-8, and B).Also, the observed unloading was largely inhibited by Atad5-depletion (Fig. 4A, lanes 9-14, and B).These results suggest that Atad5-RLC can unload PCNA regardless of the presence of single-strand nicks in Xenopus egg extracts.
Since the completion of DNA synthesis also involves nucleosome assembly, we next tested whether nucleosome assembly affects PCNA unloading.Asf1 is the histone chaperone critical for histone deposition in both replicationcoupled (CAF-1-dependent) and replication-independent (HIRA-dependent) pathways, and the depletion of Asf1 effectively inhibits nucleosome assembly in Xenopus egg extracts (59).Immunodepletion of Asf1 from HSS indeed prevented the supercoiling of the plasmid substrate, a readout of nucleosome deposition (Fig. 4, C-E, see "ΔAsf1").However, it did not significantly affect the rate of PCNA unloading, suggesting that PCNA unloading occurs regardless of chromatin assembly in Xenopus egg extracts.
We have previously shown that the MutSα mismatch sensor complex prevents PCNA unloading to maintain MMR capability (13).Since this reaction largely depends on the PIP motif located on the N-terminus of the Msh6 subunit of MutSα, plausible scenario is that the binding of the PIP peptide to the hydrophobic pocket on PCNA inhibits PCNA unloading.To test this possibility, we repeated the PCNA unloading assay in the presence of various PCNA-binding peptides.The PIPpeptide of p21 is one of the strongest PCNA-binding peptides, and it indeed effectively prevented PCNA unloading in HSS (Fig. 5, A and B).Both alanine mutations in the PIP consensus motif and jumbling of the peptide sequence completely inactivated the inhibitory effect of the peptide on PCNA unloading.Likewise, the PIP-peptide from MutSα impeded PCNA unloading, an effect that was neutralized by the PIP mutation and the jumbling of the peptide sequence (Fig. 5, C and D).These results suggest that PCNA unloading depends on the access of Atad5-RLC to the PIP-binding pocket on PCNA.

Discussion
By using a method that explicitly distinguishes PCNA unloading from its loading, we show here that Atad5-RLC provides the major PCNA unloading activity in Xenopus egg extracts.Thus, Atad5-depletion attenuated the rate of PCNA unloading to the level comparable to Rfc3-depletion, purified ATAD5-RLC restored PCNA unloading in not only Atad5-but also Rfc3-depleted extracts, and Rfc1-, Ctf18-, or Rad17depletion did not detectably affect the PCNA unloading kinetics (Figs. 2 and 3).Given the fact that Atad5-RLC constitutes only approximately 3% of RFC/RLCs (Fig. 1), it is conceivable that Atad5-RLC is the RFC-family clamp-loader/ unloader complex that is specialized for PCNA unloading.Recent structural studies show that the Rfc1 subunit dominates the DNA binding interface over the Rfc2-5 smaller subunits (60-62), and the Rad24 subunit of Rad24-RLC directs its binding to the double-stranded region at a 5 0 junction of single-stranded and double-stranded DNA in an arrangement opposite to RFC (36,37).Therefore, it is highly likely that the largest subunits critically regulate the DNA binding specificity of RFC/RLCs, thereby restricting the situation in which each complex functions.Since almost all Rfc3 and Rfc5 are included in the RFC/RLC complexes (Figs. 1 and S1A), the large subunits may limit the PCNA loading/unloading activity residing in the smaller subunits.
A key regulation of PCNA unloading is to distinguish PCNA molecules that are no longer used from those in use.Since the ligation of Okazaki fragments is an essential prerequisite for PCNA unloading in yeast cells (58), the regulation of PCNA unloading is likely linked to Okazaki fragment maturation.The trimeric nature of PCNA enables it to function as a molecular toolbelt, which sequentially recruits Pol δ, Fen1, and Lig1 to coordinate flap processing and nick ligation during Okazaki fragment maturation (63)(64)(65)(66).The inactivation of Lig1 may compromise the disassembly of the PCNA toolbelt complex and prevent the access of the PCNA unloader to PCNA.Consistent with this prediction, Kang et al. showed that Okazaki fragment maturation factors inhibit PCNA unloading by ATAD5-RLC in a purified system (47), and we found that PCNA-interacting peptides, but not continuous nicking of DNA, strongly impede PCNA unloading in a physiological condition recapitulated in Xenopus egg extracts (Figs. 4 and 5).Therefore, PCNA interactors can indeed prevent PCNA unloading.Interestingly, a biochemical study showed that Lig1 remains bound onto DNA along with PCNA even after the ligation of nicks (66).It is to be determined whether PCNA must be completely free of its interactors to be unloaded or whether some PCNA interactors are compatible with PCNA unloading.
The above findings provide a framework as to how PCNAmediated reactions are regulated: PCNA interactors prevent its unloading to secure the PCNA clamps that are in use.The completion of DNA synthesis, repair, and Okazaki fragment maturation liberates PCNA from its interactors, allowing the access of the PCNA unloader to DNA-bound PCNA.In addition, PCNA unloading can be inhibited by supplying PIP peptides around DNA-bound PCNA.PCNA is a stranddiscrimination marker for eukaryotic MMR (12), and ongoing MMR inhibits PCNA unloading to maintain the strand-discrimination capability ( 13).The MMR system likely Atad5-RLC-dependent PCNA unloading  Figure 5. PCNA-binding peptides inhibit PCNA unloading.A, the PCNA unloading assay in HSS supplemented with the solvent (H 2 O, lanes 2 and 3), wildtype (p21 WT , lanes 4 and 5), PIP-mutant (p21 pip , lanes 6 and 7), or sequence-scrambled p21 peptides (p21 jumbled , lanes 8 and 9).A quantitative immunoblot of DNA-bound PCNA (top) and linearized DNA separated by agarose gel electrophoresis (bottom) are presented, along with the estimated numbers of PCNA molecules per DNA and the amounts of DNA.B, the average numbers of DNA-bound PCNA per plasmid at the 15-min time points were plotted into a graph.Filled circles represent individual data from two independent experiments, including the one shown in (A).C, the PCNA unloading assay in HSS supplemented with the solvent (H 2 O, lanes 2 and 3), wild-type p21 peptide (p21 WT , lanes 4 and 5), wild-type (Msh6 WT , lanes 6 and 7), PIP-mutant (Msh6 pip , lanes 8 and 9), or sequence-scrambled Msh6 peptides (Msh6 jumbled , lanes 10 and 11).A quantitative immunoblot of DNA-bound PCNA (top) and linearized DNA (bottom) are presented, along with the estimated numbers of PCNA molecules per DNA and the amounts of DNA.D, the average numbers of DNA-bound PCNA per plasmid at the 15-min time points were plotted into a graph.Filled circles represent individual data from two independent experiments, including the one shown in (C).
does so by loading multiple MutSα onto a mismatch site and providing a high concentration of the PIP peptides, each of which is located at the N-terminus of a long and flexible linker on the Msh6 subunit (67).

Experimental procedures
Preparation of Xenopus egg extracts X. laevis was purchased from Kato-S-Science and handled according to the animal experimental regulations at Kyushu University.NPE and HSS were prepared as described previously (68,69).
Cloning of the X. laevis rfc3 gene was performed as follows: The rfc3 gene was amplified from Xenopus egg cDNA by twostep PCR using primers listed in Supporting Table S1 and cloned into pDONR201 (Thermo Fisher Scientific) by the Gateway BP reaction, resulting in pDONR-xRFC3.For protein expression in Escherichia coli, the rfc3 gene on pDONR-xRFC3 was transferred into pET-HSD, an in-house Gateway destination vector carrying a T7 promoter and an N-terminal His 6tag, by the Gateway LR reaction.
Cloning of the human ATAD5 gene was performed as follows: The ATAD5 gene was amplified from the HeLa cDNA library with a FLAG-tag sequence fused to the C-terminus of ATAD5 by two-step PCR using primers listed in Supporting Table S1, digested with BamHI (New England Biolabs, #R3136) and SbfI (New England Biolabs, #R3642), and ligated with a BamHI/SbfI-digested vector fragment that had been amplified by PCR from pCSII-EF-mAG-TEV-6His-Claspin-3Flag (70) using primers listed in Supporting Table S1, resulting in pCSII-EF-mAG-hATAD5-FLAG.The Walker A motif mutant of hATAD5 (hATAD5-K1138E) was constructed by two-step PCR using primers listed in Supporting Table S1.The amplified fragment was digested with BamHI and HpaI (New England Biolabs, #R0105) and inserted between the BamHI and HpaI sites in pCSII-EF-mAG-hATAD5-FLAG, resulting in pCSII-EF-mAG-hATAD5-K1138E-FLAG.

Protein expression and purification
Expression and purification of wild-type and the Walker A mutant (K1138E) of ATAD5-RLC, human PCNA, and human RFC were carried out as described previously (13,71).
Expression and purification of the His 6 -tagged Xenopus Rfc3 protein were performed as follows: Recombinant protein expression was induced in E.coli BL21-CodonPlus (DE3)-RIPL cells transformed with pET-HSD-xRFC3 by the addition of 0.5 mM Isopropyl β-D-1-thiogalactopyranoside (IPTG) in Terrific broth for 2 h at 37 C. Cells were harvested, lysed with 1 mg/ml lysozyme, sonicated in buffer S (50 mM Naphosphate pH 8.0, 300 mM NaCl, 5% glycerol, 0.1% Triton X-100, 2.5 mM 2-mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride [PMSF]), and centrifuged at 20,400g for 10 min.Inclusion bodies containing the Rfc3 protein were resuspended in buffer S, sonicated, and centrifuged again at 20,400g for 5 min, and the procedure was repeated twice.The Rfc3 protein was extracted from purified inclusion bodies with Laemmli's SDS sample buffer.
For immunodepletion of Rfc1, Atad5, or Asf1, 3 vol of an antiserum was bound to 1 vol of recombinant Protein A Sepharose Fast Flow (PAS; Cytiva, #17127902) at 4 C overnight.For Ctf18, Rad17, or Rfc3 depletion, 5 μg of purified IgG was bound to 1 μl PAS.To deplete extracts, 0.2 vol of the antibody-coupled PAS beads were incubated with 1 vol of NPE or HSS at 4 C for 1 h, and the procedure was repeated twice.

Atad5-RLC-dependent PCNA unloading
In most cases, we depleted 2060 μl of extracts for an experiment.For quadruple depletion of Rfc1, Ctf18, Atad5, and Rad17 from NPE, antibody-coupled beads were prepared separately as described above, 2 μl each of IgG beads were combined, and 8 μl of the combined IgG beads were incubated in 10 μl of NPE that had been diluted to 50 μl (5-fold) with Egg lysis buffer salts (ELB-salts: 10 mM Hepes-KOH pH 7.7, 2.5 mM MgCl 2 , 50 mM KCl) at 4 C for 1 h, and the procedure was repeated twice.

Quantification of RFC/RLCs in Xenopus egg extracts
The concentration of RFC and other RLCs was estimated from the amount of Rfc3 co-precipitated with the largest subunits.For IP of Rfc1 or Atad5, 1.5 vol of an antiserum was bound to 1 vol of PAS, and for IP of Rfc3, Ctf18, or Rad17, 15 μg of affinity-purified antibodies were bound to 1 μl PAS at 4 C overnight.NPE was diluted 10-fold with ELB-salts and clarified by centrifugation at 20,400g for 10 min to remove insoluble debris.10 μl each of antibody-coupled beads were incubated with 100 μl of diluted NPE at 4 C for 1h, washed three times with ELB-salts, and bound proteins were eluted with 40 μl of Laemmli's SDS sample buffer.IP samples were separated by SDS-PAGE, alongside a dilution series of recombinant Rfc3, transferred onto polyvinylidene fluoride membranes, and probed with antibodies against Rfc3 and other subunits.The amounts of Rfc3 in the IP fractions were calculated using the dilution series of recombinant Rfc3 as a standard.

Preparation of substrates for the in vitro PCNA-loading assay
In vitro synthesis of plasmids carrying a site-specific biotin modification was performed as described previously (13).Briefly, an oligonucleotide carrying a biotin-dT modification was annealed on a 3-kb single-stranded phagemid DNA, and the complementary DNA strand was synthesized by in-housepurified T7 DNA polymerase.After ligation of nicks by T4 DNA ligase (Nippon gene, #311-00404), covalently closed DNA molecules were purified by cesium chloride density gradient ultra-centrifugation.To make a PCNA loading site, a locus-specific nick was introduced by Nt.BbvCI (New England Biolabs, #R0632).The DNA was bound onto Sepharose beads at a ratio of 100 ng per 1 μl, following the procedure described previously (72).
The PCNA unloading assay in Xenopus egg extracts Xenopus egg extracts (HSS and NPE) were supplemented with 2 mM ATP, 20 mM PC, 5 μg/ml CPK, and pre-incubated at 22 C for 5 min.The PCNA-DNA complex was incubated in the extract at the concentration of 20 ng/μl with respect to immobilized DNA at 22 C.At appropriate time points, the mixture was quickly diluted with 200 μl of ELB-salts containing 0.2% Triton X-100, overlayed onto 200 μl of ELB-salts containing 0.5 M Sucrose and centrifuged at 12,700g for 1 min at 4 C in a TMS-21 swinging bucket rotor (Tomy Seiko, #0621650333).The beads were washed once with 200 μl of ELB-salts and resuspended in an appropriate volume of ELBsalts buffer.DNA beads were then split into two aliquots to quantify the amount of bead-bound DNA and DNA-loaded PCNA.To quantify DNA, an aliquot of the beads was mixed with 100 μl of stop buffer (1% sodium dodecyl sulfate [SDS] and 20 mM EDTA) and treated with 50 μg/ml Proteinase K (Nacalai Tesque, #29442-14) at 37 C for 1 h.DNA was then purified by phenol/chloroform extraction and ethanol precipitation and resuspended in TE buffer (10 mM Tris-HCl pH 7.4, 1 mM EDTA) containing 10 μg/ml RNase A. DNA samples were digested with XmnI (New England Biolabs, #R0194) and analyzed by 0.8% agarose gel electrophoresis followed by staining with SYBR Gold nucleic acid gel stain (Thermo Fisher Scientific, #S11494).Fluorescent signals were detected using the Amersham Typhoon scanner five system (Cytiva) and processed using the ImageQuant TL software (Cytiva).To evaluate the PCNA amount, another aliquot of the beads was mixed with Laemmli's SDS sample buffer, and DNA-bound proteins were analyzed by SDS-PAGE followed by immunoblotting.The p21 and Msh6 peptides (p21 wild-type, NH 2 -KRRQTSMTDFYHSKRRLIFS-COOH; p21 pip, NH 2 -KRRATSATDAAHSKRRLIFS-COOH; p21 jumbled, NH 2 -QDKTRYFHRTMSRSKSIRLF-COOH; Msh6 wild-type, NH 2 -MSKQKTLFSFFTKSPPVSSS-COOH; Msh6 pip, NH 2 -MSKAKTLFSAATKSPPVSSS-COOH; Msh6 jumbled, NH 2 -SVSSFTPKLQSFKSPTSFKM-COOH) (73) were synthesized by Eurofins Genomics, and added to HSS at the final concentration of 1 mg/ml.To introduce nicks on plasmids, Nb.BtsI (New England Biolabs, #R0707) was added to HSS at the concentration of 0.5 units/μl.For heat-inactivation, HSS was boiled at 95 C for 5 min, centrifuged at 20,400g for 10 min at 4 C, and the supernatant was recovered as heatinactivated HSS.

Quantification of DNA-bound PCNA
The amount of bead-bound DNA and DNA-bound PCNA was quantified using known amounts of XmnI-digested DNA and recombinant PCNA as standards, respectively.To calculate DNA-bound PCNA per plasmid, the estimated number of Atad5-RLC-dependent PCNA unloading

Figure 1 .Figure 2 .Figure 3 .
Figure 1.Quantification of the concentration of RFC and RLCs in NPE.A and B, immunoprecipitation was performed from NPE using preimmune-("mock", lane 1), Rfc3-(lane 2), Rfc1-(lane 3), Ctf18-(lane 4), Atad5-(lane 5), or Rad17-antibodies (lane 6), and 0.25 μl each of the supernatant (A) and the IP fractions corresponding to 2.5 μl each of NPE (B) were separated by SDS-PAGE alongside a serial dilution series of the supernatant of mock-IP NPE (A) or the indicated amounts of recombinant His 6 -Rfc3 (B), followed by immunoblotting with indicated antibodies.Orc2 serves as a loading control.(*) Cross-reacting band.C, the estimated concentrations of RFC and RLCs in NPE were plotted on a graph.Filled circles represent individual data from two independent experiments, including the one shown in (B).

Figure 4 .
Figure 4. Single-strand nicks and chromatin assembly do not detectably affect PCNA unloading in Xenopus egg extracts.A, the PCNA unloading assay in mock-treated HSS supplemented with buffer (lanes 3-5) or Nb.BtsI (lanes 6-8), or Atad5-depleted HSS supplemented with buffer (lanes 9-11) or Nb.BtsI (lanes 12-14), presented with the input samples before (lane 1) and after nick ligation (lane 2).Untreated DNA separated by agarose gel electrophoresis in the presence of ethidium bromide (top), a quantitative immunoblot of DNA-bound PCNA (middle), and linearized DNA separated by agarose gel electrophoresis (bottom) are shown, along with the estimated numbers of PCNA molecules per DNA and the amounts of DNA.B, the average numbers of DNA-bound PCNA per plasmid at the 10-min time points were plotted into a graph.Filled circles represent individual data from two independent experiments, including the one shown in (A).C, 2 μl each of mock-treated (lane 1) and Asf1-depleted HSS (lane 2) were analyzed by The mouse monoclonal antibody against human PCNA (MBL International Corporation, #MH-12-3) is commercially available.The rabbit antiserum against Xenopus Orc2 was a kind gift from Dr Johannes Walter.The rabbit Ctf18, Rad17, Rfc5, Ctf8, and Asf1 antibodies were raised against peptides NH 2 -CINEEFGENDSEILENDDNA-COOH, corresponding to residues 269 to 287 of Ctf18, NH 2 -CAAQAIMEDEELKIEEYDSD-COOH, corresponding to residues 269 to 287 of Rad17, NH 2 -CEHVVKEERVDISPDGMK-COOH, corresponding to residues 182 to 198 of Rfc5, NH 2 -CKIIFKTRPKPIITNVPKKV-COOH, corresponding to residues 103 to 121 of Ctf8, NH 2 -APSKGLAAALNTLPENSMDC-COOH, corresponding to residues 180 to 199 of Asf1, respectively.All antibodies except anti-Ctf8 and anti-Asf1 were affinity-purified using corresponding antigens.Polyclonal antibodies against Xenopus Atad5 and Rfc1 were raised in rabbits against the following polypeptides: Atad5: N-terminal 105 amino acids of Atad5 with a 6-histidine tag at the C-terminus (MVGILAMSASLEEYGCQPCKKSRKDEEAPIKTITNYFSPVS KNTEKVLSSPRSNNIADYFKQNSPINEKKQTSKAENAAIQ QDTPQAAVADSSAASGKPSKCRKR); Rfc1: N-terminal 100 amino acids of Rfc1 with a 6-histidine tag at the C-terminus (MDIRNFFGVKPVAKKHGTEKTDIKEKKKSPEAKKKPKD SKVKSPSS DDSLKGMNVKKKKRIIYDSDAEEESPPVKKA KKPSEKSPPLPRPHKIRKPDPVV).The specificity of all the antibodies has been validated by immunoprecipitation and immunoblotting using recombinant proteins and Xenopus egg extracts.