Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2003 Feb 10;100(4):1633–1638. doi: 10.1073/pnas.0437927100

Loading of the human 9-1-1 checkpoint complex onto DNA by the checkpoint clamp loader hRad17-replication factor C complex in vitro

Vladimir P Bermudez *,†, Laura A Lindsey-Boltz ‡,†, Anthony J Cesare ‡, Yoshimasa Maniwa *, Jack D Griffith ‡, Jerard Hurwitz *, Aziz Sancar ‡,§
PMCID: PMC149884  PMID: 12578958

Abstract

The human DNA damage sensors, Rad17-replication factor C (Rad17-RFC) and the Rad9-Rad1-Hus1 (9-1-1) checkpoint complex, are thought to be involved in the early steps of the DNA damage checkpoint response. Rad17-RFC and the 9-1-1 complex have been shown to be structurally similar to the replication factors, RFC clamp loader and proliferating cell nuclear antigen polymerase clamp, respectively. Here, we demonstrate functional similarities between the replication and checkpoint clamp loader/DNA clamp pairs. When all eight subunits of the two checkpoint complexes are coexpressed in insect cells, a stable Rad17-RFC/9-1-1 checkpoint supercomplex forms in vivo and is readily purified. The two individually purified checkpoint complexes also form a supercomplex in vitro, which depends on ATP and is mediated by interactions between Rad17 and Rad9. Rad17-RFC binds to nicked circular, gapped, and primed DNA and recruits the 9-1-1 complex in an ATP-dependent manner. Electron microscopic analyses of the reaction products indicate that the 9-1-1 ring is clamped around the DNA.

Eukaryotic cells exposed to genotoxic agents activate the DNA damage checkpoint signaling pathway, which arrests cell-cycle progression and in so doing prevents cell death or mutations. Recent work has revealed that in mammalian cells, the ATM and ATR proteins, which belong to the phosphatidylinositide kinase-like kinase family, and the Rad17-replication factor C (Rad17-RFC) and the Rad9-Rad1-Hus1 (9-1-1) checkpoint complexes, which have structural similarities to the replication clamp loader and replication clamp RFC and proliferating cell nuclear antigen (PCNA), respectively, are involved in damage recognition, which activates the checkpoint response (reviewed in refs. 1–4). Studies with budding and fission yeasts have shown that the orthologs of these proteins perform similar functions. However, biochemical data on the specific roles of the phosphatidylinositide kinase-like kinase family members and the Rad17-RFC and 9-1-1 complexes are scarce, and hence the damage sensing step of the checkpoint response remains ill-defined. We previously reported that ATR directly recognizes and is activated by damaged DNA (5). In this article, we investigate the interactions of Rad17-RFC and the 9-1-1 checkpoint complexes with DNA to gain some insight into their roles as damage sensors.

Rad17-RFC is one of the three known RFC-like complexes in mammalian cells. In this form of RFC, the p140 subunit is replaced by the 75-kDa Rad17 protein, which has homology to all RFC subunits (6). Yeast genetic studies indicate that the orthologs of human Rad17 function exclusively in the DNA damage checkpoint response (7, 8). The 9-1-1 checkpoint complex is a heterotrimer of Rad9, Rad1, and Hus1 proteins, which were predicted to have structural homology to PCNA (9–13). Previously, we showed that Rad17 associates with the four small RFC subunits to make an RFC-like complex, which by electron microscopy exhibits an RFC-like structure (14). Similarly, we found that Rad9, Rad1, and Hus1 form a heterotrimeric complex with a PCNA-like ring structure (14). During replication, RFC binds to primed templates and recruits PCNA to the site of replication (reviewed in ref. 15). RFC opens the PCNA ring and closes it around the DNA at the primer/template junction, where PCNA acts as a polymerase clamp and a processivity factor for DNA polymerases δ and ɛ. Despite the structural similarities between the RFC/PCNA and Rad17-RFC/9-1-1 complex pairs, at present there is no direct biochemical evidence that Rad17-RFC and the 9-1-1 ring function in a manner analogous to their replicative counterparts. In this study, we investigated the interactions of Rad17-RFC with the 9-1-1 complex and the binding of these proteins to DNA. We find that Rad17-RFC binds to the 9-1-1 complex in an ATP-dependent manner and that the interaction is mediated mainly by binding of Rad9 to Rad17. By using nicked and gapped plasmid DNAs, we find that the binding of Rad17-RFC to DNA is ATP-independent but that the recruitment of the 9-1-1 complex depends on ATP but does not require ATP hydrolysis. Electron microscopic analysis of the DNA–protein complexes suggests that Rad17-RFC does in fact load the 9-1-1 complex on DNA.

Materials and Methods

Expression and Purification of Recombinant Proteins.

Baculoviruses used for expression of Flag-Rad17, His-Rad17, His-p140, p40, His-p38, p37, p36, Flag-Rad9, Rad9, Flag-Hus1, Hus1, and Rad1 were described (6). Infection of monolayer High Five (H5) insect cells (Invitrogen) and isolation of 9-1-1, Rad17-RFC, and the 9-1-1/Rad17-RFC complexes were as described (6, 14).

Rad1 baculovirus expressing the Rad1 with a Flag2-cAMP kinase site motif (LRRASV) at the N terminus was constructed by inserting a linker containing Flag-cAMP kinase motif sequence at the 5′ terminus of pFastBac1-Rad1 (6) and generating the baculovirus by using the Bac-to-Bac System (Invitrogen). Monolayer H5 insect cells (1 × 109 cells, Invitrogen) were infected with a multiplicity of infection of five for baculoviruses expressing Rad9, Hus1, and Flag-cAMP kinase motif-tagged Rad1 and were harvested 48 h after infection. The packed cells (7.5 ml) were washed with ice-cold PBS, lysed in 15 ml of hypotonic buffer (50 mM Tris⋅HCl, pH 8.0/10 mM sodium phosphate, pH 8.0/1.5 mM MgCl2/1 mM DTT/2 μg/ml aprotinin/2 μg/ml leupeptin/2 μg/ml antipain/0.1 mM benzamidine/0.5 mM PMSF) supplemented with 10 mM KCl by Dounce homogenization (10 strokes with a B pestle) and centrifuged at 2,400 × g for 30 min at 4°C. The supernatant (cytosolic fraction) was adjusted to 0.5 M NaCl. The pellet was resuspended in 15 ml of hypotonic buffer containing 0.5 M NaCl and centrifuged at 43,500 × g for 30 min at 4°C. The supernatant (nuclear fraction) and the cytosolic fraction were combined and centrifuged for 15 min at 43,500 × g at 4°C. Flag-M2 beads (Sigma, 0.75 ml), equilibrated with 50 mM Tris⋅HCl, pH 8.0/20 mM sodium phosphate, pH 8.0/0.5 M NaCl/1 mM DTT/2 μg/ml aprotinin/2 μg/ml leupeptin/2 μg/ml antipain/0.1 mM benzamidine/0.5 mM PMSF were added to the supernatant (41 ml, 360 mg), and the suspension was rocked overnight at 4°C. The Flag M2 beads were washed five times with 10 ml of Flag buffer (20 mM sodium phosphate, pH 8.0/10% glycerol/0.5 M NaCl/1 mM DTT/2 μg/ml aprotinin/2 μg/ml leupeptin/2 μg/ml antipain/0.1 mM benzamidine/0.5 mM PMSF). The bound Flag-tagged 9-1-1 was eluted two times with 0.75 ml of Flag buffer containing 0.2 mg/ml Flag peptide for 45 min at 4°C, yielding 0.53 mg of protein. The Flag-cAMP kinase-tagged 9-1-1 was labeled with [γ-32P]ATP to a specific activity of 1,800 cpm/fmol, as described (16).

In Vitro Interaction of the Rad17-RFC and Checkpoint 9-1-1 Complexes.

Two assays were used to analyze the Rad17-RFC and 9-1-1 interaction. In one assay, reaction mixtures (1 ml) containing the purified 9-1-1 complex (1 pmol) and the purified Rad17-RFC complex (1 pmol) were incubated in buffer containing 25 mM Tris⋅HCl, pH 7.5/0.15 M NaCl/10 mM MgCl2/1 mM DTT/0.1 mg/ml BSA in the presence or absence of 1 mM ATP for 1 h at 30°C. Protein was then immunoprecipitated with 1 μg of anti-Rad9 antibodies (Santa Cruz Biotechnology) and 20 μl of protein-A agarose, followed by rotation for 2 h at 4°C; the resin was washed three times with reaction buffer + 0.05% Nonidet P-40 (1 ml each time), and the bound and unbound proteins were analyzed by Western blotting with anti-Flag antibodies (Sigma). In the second assay, reaction mixtures (25 μl) contained Rad17-RFC (or RFC, 5 pmol), 32P-labeled 9-1-1 (or PCNA, 0.5 pmol), binding buffer (25 mM Hepes-KOH, pH 7.5/0.1 mM EDTA/1 mM DTT/0.15 M NaCl/0.5% Nonidet P-40/10 mM MgCl2/1 mg/ml BSA) and 1 mM ATP, as indicated. After incubation for 10 min at 37°C, an aliquot (10 μl) was diluted 10-fold with binding buffer and incubated with 1 μg of anti-RFCp37 or preimmune serum and 10 μl of protein-A agarose (Upstate Biochemicals) for 1 h at 4°C with constant agitation. The beads were washed three times with 0.5 ml of binding buffer and twice with 0.5 ml of binding buffer lacking BSA. Bound proteins were eluted in 20 μl of SDS loading buffer and subjected to 12% SDS/PAGE, and 32P-labeled 9-1-1 and PCNA were analyzed by autoradiography.

Formation of Rad17-RFC and 9-1-1 complexes was also analyzed by glycerol gradient sedimentation. Rad17-RFC (5 pmol) and 32P-labeled 9-1-1 (1 pmol) were incubated in 100 μl of binding buffer in the presence or absence of 10 μM nucleotide (as indicated) for 10 min at 37°C, and the mixture was layered onto a 5-ml 15–35% glycerol gradient containing 25 mM Tris⋅HCl, pH 7.5, 1 mM EDTA, 0.01% Nonidet P-40, 1 mM DTT, 10 mM MgCl2, and 50 μg/ml BSA. The gradient was centrifuged at 4°C for 20 h at 260,000 × g, and samples were collected from the bottom of the gradient. The distribution of 32P-labeled 9-1-1 was measured by Cerenkov counting, whereas distribution of the Rad17-RFC in gradient fractions was detected by slot blotting on nitrocellulose membrane.

Interactions Between Rad17 and Rad9.

H5 insect cells (5 × 106) were coinfected with baculoviruses expressing either His-Rad17 or His-RFCp140, together with Flag-Rad9, Flag-Rad1, Flag-Hus1, or Flag-Hus1, together with untagged Rad9 and Rad1 at a multiplicity of infection of five. After lysing the cells in 1.5 ml of 50 mM Tris⋅HCl, pH 7.5/0.3 M NaCl/0.05% Nonidet P-40, the protein was then immunoaffinity-purified by using 10 μl of anti-Flag agarose. After washing the resin three times in the same buffer (1 ml), protein was eluted with 0.2 mg/ml Flag peptide in 0.15 M NaCl, 50 mM Tris⋅HCl, pH 7.5 and 10% glycerol. Aliquots of the load and the eluted material were subjected to SDS/12% PAGE and Western blotting.

The Rad17-interacting domain of Rad9 was mapped by using baculoviruses that expressed five fragments of Rad9 containing a C-terminal Flag-tag (described in Fig. 4B) that were generated by PCR amplification by using pFast-Flag-Rad9 (6) as the template. These fragments were expressed in H5 cells and bound to anti-Flag agarose. After washing the resin three times, it was incubated with extract made from H5 cells infected with either His-Rad17 or His-p140. After washing the resin, the bound protein was eluted with Flag peptide and analyzed by Western analysis with anti-His (Santa Cruz Biotechnology) and anti-Flag antibodies.

Figure 4.

Figure 4

Open in a new tab

Rad9 mediates the Rad17-RFC/9-1-1 interaction. (A) Rad17 interacts with Rad9. H5 cells were coinfected with baculoviruses expressing either His-Rad17 alone (lane 1) or His-Rad17 together with Flag-Rad1 (lane 2), Flag-Rad9 (lane 3), Flag-Hus1 (lane 4), or Flag-Hus1, and untagged Rad1 and Rad9 (lane 5). The proteins were immunoaffinity purified with anti-Flag agarose, and proteins were analyzed by Western blotting with anti-Rad17 or anti-Flag antibodies. (B) Binding of the PCNA-like domain of Rad9 to Rad17. The following fragments of Rad9 were produced in H5 cells and bound to anti-Flag agarose resin: amino acids 1–130, lane A; amino acids 130–270, lane B; amino acids 1–270, lane C; amino acids 260–391, lane D; amino acids 130–391, lane E; and full length, lane F. The resin was then incubated with extracts made from H5 cells infected with either His-Rad17 or His-p140. After washing the resin, bound protein was eluted with Flag peptide and analyzed by Western blotting with anti-His and anti-Flag antibodies.

Recruitment of Checkpoint 9-1-1 Complex to DNA by Rad17-RFC.

Reaction mixtures (50 μl) containing Rad17-RFC (2 pmol), 32P-labeled 9-1-1 (0.6 pmol), 0.15 pmol of nicked pBluescript (17), 40 mM Tris⋅HCl, pH 7.8, 10 mM magnesium acetate, 1 mM DTT, 0.2 mg/ml BSA, 2 mM [γ-thio]ATP (ATPγS) (or ATP), and 0.15 M NaCl were incubated for 10 min at 37°C. Reactions were loaded onto 5 ml of Bio-Gel A-15M (0.7 × 15 cm) column. The columns were eluted with 40 mM Tris⋅HCl, pH 7.8, 0.1 mM EDTA, 0.2 mg/ml BSA, 10 mM magnesium acetate, 4% glycerol, 0.15 M NaCl, 1 mM DTT, and 0.01% Nonidet P-40. The amount of labeled 9-1-1 in the fractions (150 μl each) was measured by Cerenkov counting, and Rad17-RFC was detected by slot blotting of the fractions on nitrocellulose membranes.

Electron Microscopy.

Reaction mixtures (20 μl) containing purified 9-1-1 (1 pmol) and Rad17-RFC (0.2 pmol) were incubated alone or together with 100 ng of a 6.9-kb nicked plasmid (pFastBac1-Rad17) or a 718-bp circular DNA with a 40-nt ssDNA gap (to be described elsewhere) with 30 mM Hepes, pH 7.5/8 mM magnesium acetate/1 mM DTT/2 mM ATP/5% PEG at 37°C for 10 min. Reactions were then diluted 10-fold with addition of NaCl to 0.5 M and EDTA to 2 mM and incubated for 5 min at 20°C to release loosely bound proteins. The samples were treated with 0.6% glutaraldehyde at 20°C for 5 min and chromatographed over a 2.5-ml column of Bio-Gel A-5m, and fractions containing the DNA–protein complexes were collected and prepared for electron microscopy (18). Analysis was done on a Phillips CM12 instrument. Images on negatives were scanned with a Nikon LS4500 film scanner, and the figures were prepared by using PHOTOSHOP (Adobe Systems, Mountain View, CA).

Results

Purification of Rad17-RFC/9-1-1 Checkpoint Supercomplex.

We previously showed that the Rad17-RFC and the 9-1-1 checkpoint complexes could be expressed in baculovirus-infected insect cells and purified in quantities sufficient for biochemical analysis (6). We wished to determine whether these two complexes associate when free in solution or whether they assemble on DNA exclusively. Infection of insect cells with baculoviruses expressing the three subunits of the 9-1-1 complex or the five subunits of the Rad17-RFC complex resulted in the formation of these complexes in vivo, and the complexes were readily purified by affinity chromatography (Fig. 1) as reported (6). Significantly, when the insect cells were coinfected with eight different viruses, each expressing one of the subunits, we were also able to purify the checkpoint supercomplex Rad17-RFC/Rad9-Rad1-Hus1, containing all eight subunits in essentially stoichiometric ratios, by using affinity resins for tags on both Rad17 and Hus1 (Fig. 1, lane 3). Thus, it seems that Rad17-RFC and the 9-1-1 checkpoint complex can associate off DNA. Furthermore, the yield of the eight-subunit complex assembled in vitro was unaffected by the presence of DNA (data not presented).

Figure 1.

Figure 1

Open in a new tab

Purification of Rad17-RFC/9-1-1 checkpoint supercomplex. The checkpoint complexes were reconstituted in H5 cells by coinfection with: lane 1 (9-1-1 complex), three baculoviruses expressing Flag-Rad9, Hus1, and Rad1; lane 2 (Rad17-RFC), five baculoviruses expressing Flag-Rad17, p40, His-p38, p37, and p36; and lane 3 (supercomplex), eight baculoviruses expressing His-Rad17, p40, His-p38, p37, p36, Rad9, Flag-Hus1, and Rad1. Complexes were purified by chromatography with Ni-NTA and/or anti-Flag agarose as described in Materials and Methods, and proteins were visualized after SDS/PAGE by silver staining.

Effect of ATP on the Formation of the Rad17-RFC/9-1-1 Supercomplex.

Because replicative RFC binds PCNA in an ATP-dependent manner (19), we investigated the effect of ATP on Rad17-RFC/9-1-1 complex formation. Binding was analyzed by both protein pull-down assays and by glycerol gradient sedimentation. The pull-down assay was performed by immunoprecipitating Rad9 or RFCp37 in a mixture that contained Rad17-RFC and 9-1-1, and then probing the immunoprecipitates for either the Rad17 or Rad1 subunits of the supercomplex. Fig. 2A shows that Rad9 antibodies coimmunoprecipitate Rad17 and presumably the entire Rad17-RFC complex. Similarly, Fig. 2B shows that anti-RFCp37 antibodies pull-down Rad1 and, presumably, the entire 9-1-1 complex. Although some background association is seen in Fig. 2A, the binding of the two complexes depends strongly on adenosine nucleotides. Quantitative analysis of Fig. 2B shows that binding is strongest in the presence of ATPγS, followed by ATP, dATP, and ADP, and that GTP, UTP, and CTP are ineffective (Fig. 2C). In these experiments, the level of nucleotide used was 10 μM (the Km found for ATP in this reaction). In Fig. 2D, complex formation was analyzed by glycerol gradient sedimentation. The results confirm the data from the other panels, and moreover, the sedimentation coefficient of the complex reveals that in the presence of ATP or ADP, but not AMP or adenosine 5′-[β,γ-imido]triphosphate (AMPPNP), a stable complex assembles that is consistent with the predicted mass of Rad17-RFC/9-1-1 complex. Thus, as in the case of RFC and PCNA, ATP binding but not hydrolysis is required for complex formation between Rad17-RFC and the 9-1-1 ring.

Figure 2.

Figure 2

Open in a new tab

Nucleotide cofactor requirement for binding of Rad17-RFC to the 9-1-1 complex. (A) Coimmunoprecipitation with anti-Rad9 antibodies. One picomole each of purified Rad17-RFC and 9-1-1 complexes were incubated with or without ATP and immunoprecipitated with anti-Rad9 antibodies; the bound and unbound proteins were analyzed by Western blotting with anti-Flag antibodies. (B) Coimmunoprecipitation with anti-RFCp37 antibodies. Reaction mixtures (25 μl) containing 5 pmol of purified RFC or Rad17-RFC and 0.5 pmol of 32P-labeled 9-1-1 were incubated with or without 10 μM of the indicated nucleotide as described in Materials and Methods. Lo denotes 10% input, Pr denotes preimmune serum, and 37 denotes anti-RFCp37 serum used for immunoprecipitations. The bound proteins were eluted with SDS loading buffer, subjected to SDS/12% PAGE, and autoradiographed. (C) The 32P-labeled 9-1-1 bound to Rad-17RFC shown in B was quantitated by phosphorimaging analyses. (D) Analysis by glycerol gradient velocity sedimentation. Reaction mixtures (100 μl) containing purified Rad17-RFC (5 pmol) and 32P-labeled 9-1-1 (1 pmol) were incubated in the presence or absence of 10 μM indicated nucleotides. After incubation, mixtures were subjected to glycerol gradient centrifugation as described in Materials and Methods. Nucleotide additions were as follows: ⧫, no ATP; ◊, ATP; ●, ADP; ■, AMP; and ▴, adenosine 5′-[β,γ-imido]triphosphate (AMPPNP).

Specificity of the Rad17-RFC/9-1-1 Interaction.

It is remarkable that despite apparent structural similarities, the function of RFC/PCNA is restricted to replication and that of Rad17-RFC/9-1-1 complex is restricted to the checkpoint. To gain some insight into this specificity, we analyzed the interactions of RFC and Rad17-RFC with both PCNA and the 9-1-1 complex. Fig. 3 (Upper) shows that in the presence of ATP, Rad17-RFC, but not RFC, is capable of binding to the 9-1-1 complex. Surprisingly, both Rad17-RFC and RFC can bind PCNA (Fig. 3 Lower) even in the absence of ATP. The marginal ATP stimulation of the PCNA–RFC interaction is most likely due to the salt conditions used in this reaction (0.15 M NaCl). The findings that Rad17-RFC can bind both 9-1-1 and PCNA trimeric rings (as seen in Fig. 3) would suggest that although RFC cannot substitute for Rad17-RFC in the checkpoint response, Rad17-RFC may be capable of substituting for RFC in replication. However, we have determined that Rad17-RFC will not substitute for RFC in an in vitro DNA polymerase δ, PCNA-dependent replication reaction or in the loading of PCNA onto DNA (data not shown). The physiological significance of the Rad17-RFC/PCNA interaction remains to be determined.

Figure 3.

Figure 3

Open in a new tab

Interactions of RFC and Rad17-RFC with PCNA and the 9-1-1 checkpoint complex. Reaction mixtures (25 μl) containing purified Rad17-RFC or RFC (5 pmol) and 1 pmol of either 32P-labeled 9-1-1 (Upper) or PCNA (Lower) were incubated in the presence or absence of 1 mM ATP as described in Materials and Methods. The 9-1-1 (or PCNA) bound to Rad17-RFC or RFC was immunoprecipitated with anti-RFCp37 antibody, and the bound material was analyzed by SDS/PAGE followed by autoradiography. Lo, 10% input; Pr, preimmune serum; 37, anti-RFCp37 antibody.

Our data show that it is the Rad17 subunit of Rad17-RFC that enables this complex to bind to the 9-1-1 ring. Hence, we investigated the interaction of Rad17 with the subunits of the 9-1-1 complex. Fig. 4A shows that Rad9, and to a much lesser degree Rad1, interacts with Rad17, but Hus1 does not. Having thus found that Rad9 is the subunit mainly responsible for interaction with Rad17-RFC, to the exclusion of RFC, we examined whether the non-PCNA-like C-terminal extension of Rad9 was responsible for the Rad17-RFC/9-1-1 complex specificity. Deletion constructs missing either the amino (amino acids 1–130) or carboxyl (amino acids 260–391) terminus of Rad9 were made and tested for their interaction with Rad17. Fig. 4B (lane F) shows, as expected, that full-length Rad9 binds to Rad17 but not to RFCp140, which replaces Rad17 in replicative RFC. Importantly, we find that the PCNA-like domain of Rad9 (amino acids 1–270) is necessary and sufficient for binding to Rad17 (Fig. 4B, lane C). Thus, the C-terminal non-PCNA-like extension of Rad9 is not responsible for specific interactions with Rad17-RFC. This extension contains the Rad9 phosphorylation sites, as were reported (20). It is apparent from the anomalous (slow) migration of the Rad9 fragments, which contain this region, that it is heavily phosphorylated even when expressed in insect cells (Fig. 4B, lanes D–F). It is possible that the C-terminal extension is involved in the effector functions of the 9-1-1 checkpoint complex.

Recruitment of the 9-1-1 Ring to DNA by Rad17-RFC.

In vivo data in both budding yeast and humans indicate that Rad17-RFC together with the 9-1-1 complex associate with the damage site early on in the checkpoint response (21–23). Because Rad17-RFC binds to DNA in vitro and the 9-1-1 complex does not, we reasoned that the 9-1-1 ring may be recruited to the site of DNA damage in a manner analogous to the recruitment of PCNA by RFC to the primer terminus during replication. To test this model, a nicked plasmid DNA was incubated with Rad17-RFC, the 9-1-1 complex, or both, and the DNA was separated from unbound proteins by gel-exclusion chromatography (Fig. 5). Rad17-RFC binds DNA in the absence or presence of the 9-1-1 complex, and this binding is ATP-independent (data not shown). In contrast, the 9-1-1 complex binds to DNA only when Rad17-RFC is present, indicating that Rad17-RFC does recruit the 9-1-1 complex to DNA. Because the loading of PCNA onto DNA is ATP-dependent, we investigated the nucleotide cofactor requirement for recruitment. As apparent from Fig. 5, the recruitment of the 9-1-1 complex to DNA by Rad17-RFC (its elution in the excluded region) (i) requires a nicked or gapped DNA substrate (form II), (ii) requires a nucleotide cofactor, and (iii) is more efficient in the presence of ATPγS than ATP and is most efficient when incubated with ATPγS followed by incubation with ATP. Thus, it can be concluded that ATP binding but not ATP hydrolysis is needed for recruitment.

Figure 5.

Figure 5

Open in a new tab

Recruitment of the 9-1-1 complex to DNA by Rad17-RFC. Reaction mixtures (50 μl) containing 32P-labeled 9-1-1 (0.6 pmol), Rad17-RFC (2 pmol), and nicked pBluescript plasmid (0.15 pmol), where indicated, were incubated with or without nucleotide. The mixtures were then loaded onto a 5-ml BioGel A-15M column as described in Materials and Methods. ■, form II (nicked plasmid)/Rad17-RFC/9-1-1/ATPγS-ATP (mixture was incubated with ATPγS for 5 min and then with 8 mM ATP for 5 min); ▴, form II (nicked plasmid)/Rad17-RFC/9-1-1/ATPγS; ●, form II (nicked plasmid)/Rad17-RFC/9-1-1/ATP; ♦, form II (nicked plasmid)/Rad17-RFC/9-1-1; ○, form II (nicked plasmid)/9-1-1/ATP; □, Rad17-RFC/9-1-1/ATPγS.

Loading of the 9-1-1 Ring onto DNA by Rad17-RFC.

During replication, RFC binds to PCNA, opens the ring, and clamps it onto DNA where PCNA can bind to Polδ and Polɛ to aid in processive DNA synthesis (15). The loading of PCNA by RFC is catalytic, and an RFC molecule is capable of loading multiple PCNAs onto DNA. A consequence of this mode of action is that when a nicked or gapped circular DNA is used for PCNA recruitment, multiple PCNAs are loaded onto the circle and slide away from the site of loading. If the plasmid is linearized with a restriction enzyme, the PCNA molecules not associated with RFC slide off the DNA with the consequence of having RFC and PCNA on the DNA at a molar stoichiometry of 1:1.

To find out whether the Rad17-RFC/9-1-1 complex behaved similarly, we performed loading experiments with these complexes (Fig. 6). First, in agreement with previous reports, we found that the amount of PCNA loaded by RFC on linearized DNA was 5-fold lower than that loaded onto nicked circular DNA (Fig. 6A). This is consistent with the RFC-free PCNA rings sliding off linear DNA but not off circular DNA. However, when similar experiments were performed with the Rad17-RFC/9-1-1 pair, only a marginal decrease in the amount of DNA-associated 9-1-1 complexes was observed with linear DNA compared with circular DNA (Fig. 6B). Identical results were obtained when PCNA and 9-1-1 were loaded onto the nicked plasmid, isolated by gel-exclusion chromatography, and then linearized by restriction. These results, at face value, imply that Rad17-RFC recruits the 9-1-1 ring to DNA, but it does not clamp it around the duplex in a manner similar to PCNA. Alternatively, the data are also consistent with a model in which the vast majority of the DNA–protein complexes in the Rad17-RFC/9-1-1 reaction mixture contain both the clamp loader and the clamp, and hence the clamp cannot escape from the loading site. A small fraction of the 9-1-1 ring placed around the duplex and no longer in contact with Rad17-RFC, and thus freely diffusible around the duplex, would not be detected by this method. We reasoned that such molecules might be more readily detected by electron microscopy and analyzed the Rad17-RFC/9-1-1/DNA complexes by electron microscopy.

Figure 6.

Figure 6

Open in a new tab

Sliding assay for PCNA and the 9-1-1 complex. (A) PCNA loading. RFC (2 pmol) and 32P-labeled PCNA (0.6 pmol) were assembled on nicked pBluescript (♦) or BamHI-linearized nicked pBluescript (■) (0.15 pmol) in a 50-μl reaction mixture as described in Materials and Methods. The mixture was incubated in 2 mM ATP for 5 min and then loaded onto a 5-ml Bio-Gel A-15M and eluted as described in Materials and Methods. (B) 9-1-1 loading. Rad17-RFC (2 pmol) and 32P-labeled 9-1-1 (0.6 pmol) were assembled on nicked pBluescript (●) or BamHI-linearized nicked pBluescript (○) (0.15 pmol) in a 50-μl reaction mixture as described in Materials and Methods. The mixture was incubated in 2 mM ATPγS for 5 min and then supplemented with 8 mM ATP and incubated for 5 min. The mixture was loaded onto a 5-ml Bio-Gel A-15M and eluted as described in Materials and Methods.

Fig. 7A shows an EM picture of a nicked plasmid that was incubated with Rad17-RFC/9-1-1 complex. The size and shape of the complex, indicated by a large arrow, are consistent with that of a Rad17-RFC/9-1-1 supercomplex, whereas those indicated by small arrows are consistent with the shape and size of the 9-1-1 complex. Thus, a plausible explanation of the data in Fig. 7A is that the Rad17-RFC at the nick site has already loaded two 9-1-1 clamps around the DNA and that these two have diffused away from the loading site while the complex is in the process of loading a third 9-1-1 clamp. Indeed, such complexes were not observed when the reaction was performed in the absence of ATP (data not shown). Similar experiments were also performed with a 718-bp duplex with a 40-nt gap, and some representative photographs from those loading experiments are shown in Fig. 7 B–D. Fig. 7D shows an EM picture of Rad17-RFC bound to the gapped DNA. Based on EM analyses of numerous Rad17-RFC and 9-1-1 complexes, it is safe to assume that the DNA-bound proteins in Fig. 7 B and C are 9-1-1 rings (flat and hollow) around the DNA when compared with the compact and relatively dense Rad17-RFC shown in Fig. 7D. EM pictures of gel-excluded material from reactions lacking Rad17-RFC failed to detect 9-1-1 rings around the DNA (data not presented). Thus, whereas hydrodynamic experiments support only recruitment of 9-1-1 by Rad17-RFC, single molecule experiments by electron microscopy clearly support loading of 9-1-1 rings onto DNA by the Rad17-RFC clamp loader.

Figure 7.

Figure 7

Open in a new tab

Electron microscopic evidence for loading of the 9-1-1 complex by Rad17-RFC. (A) 9-1-1 loading on a 6.9-kb nicked plasmid. Large white arrow indicates presumed supercomplex of Rad17-RFC and 9-1-1; small white arrows indicate loaded 9-1-1 protein complexes. (B–D) Loaded 9-1-1 complexes (B and C) and Rad17-RFC (D) on a 718-bp circular template with a 40-nt single strand gap. Samples were directly mounted onto thin carbon-coated foils and rotary shadowed with tungsten. Shown in reverse contrast. [Bar = 1,000 bp (A); = 500 bp (B–D).]

Discussion

It seems that human cells have at least three RFC-like complexes with specialized functions: RFC for replication, Rad17-RFC for the DNA damage checkpoint, and Ctf18-RFC for sister chromatid separation (24, 25). The only known function of replicative RFC is to load PCNA onto DNA and thus form a clamp that enhances the processivity of replicative polymerases (15). It has been assumed that the other RFC analogs function in a similar manner. Indeed, analyses of Rad17-RFC and the 9-1-1 complex by electron microscopy revealed structures very similar to RFC and PCNA, respectively, suggesting functional similarity as well (14, 26). Despite the structural similarities, however, it has been experimentally challenging to demonstrate that Rad17-RFC/9-1-1 behaves exactly like the classic clamp loader/clamp complex. This may be due, in part, to the intrinsic differences between the ways these two complexes function. However, part of the problem undoubtedly stems from the unique physicochemical properties of the Rad17–RFC complex. This complex is prone to aggregate in ionic strength below 0.3 M and in the absence of a nucleotide cofactor (data not presented). The necessity to work within these constraints has made it difficult to study the complex biochemically. Despite these limitations, in this paper we have been able to show the formation of a Rad17-RFC/9-1-1 supercomplex and the recruitment of the 9-1-1 ring to DNA in a manner analogous to recruitment of PCNA to DNA by RFC. Although it was not possible to demonstrate unambiguously that Rad17-RFC loads the 9-1-1 complex onto DNA by using hydrodynamic methods, we did obtain evidence from electron microscopic analysis that Rad17-RFC does load the 9-1-1 complex onto DNA. We believe the difficulty in demonstrating 9-1-1 loading compared with PCNA loading may be due to stronger interactions between Rad17-RFC and the 9-1-1 complex, which could reduce the rate with which the loaded 9-1-1 complex moves away from Rad17-RFC. Nevertheless, the EM results suggest that with further optimization of the reaction conditions, we will be able to demonstrate 9-1-1 loading by hydrodynamic methods.

Assuming that the Rad17-RFC/9-1-1 pair functions similarly to the RFC/PCNA pair, then the question arises as to why Rad17-RFC/9-1-1 is required at all. Clearly, the replication machinery and the checkpoint signaling system involve quite distinct pathways with their unique protein components, and the utilization of two distinct clamp loader/DNA clamps for these two pathways may make mechanistic sense. It must be noted that we find that Rad17-RFC is capable of interacting with PCNA; however, we do not detect any loading of PCNA by Rad17-RFC.

Finally, how a clamp loader/polymerase clamp functions in the DNA damage checkpoint is an interesting question. In current models for the checkpoint, there are no suggestions of a DNA clamp for enhancing the processivity of a polymerase or any other DNA-traversing enzyme. Rather, these models suggest that the Rad17-RFC/9-1-1 complex and ATR/ATM accumulate at the site of damage (a single-strand DNA gap or a double-strand break) and recruit the signal transduction kinases to these sites to activate the checkpoint. There is no obvious role in these models for a clamp loader/clamp complex. We expect that further biochemical analysis will help refine the models and explain why such a complex is needed.

Acknowledgments

We thank Inger Tappin for technical assistance. This work was supported by National Institutes of Health Grants GM38559 (to J.H.), F32-GM20830 (to L.A.L.-B.), GM32833 (to A.S.), and GM31819 (to J.D.G.). J.H. is an American Cancer Society Professor.

Abbreviations

RFC

replication factor C

PCNA

proliferating cell nuclear antigen

ATPγS

[γ-thio]ATP

References

  • 1.Abraham R T. Genes Dev. 2001;15:2177–2196. doi: 10.1101/gad.914401. [DOI] [PubMed] [Google Scholar]
  • 2.Nyberg K A, Michelson R J, Putnam C W, Weinert T A. Annu Rev Genet. 2002;36:617–656. doi: 10.1146/annurev.genet.36.060402.113540. [DOI] [PubMed] [Google Scholar]
  • 3.Osborn A J, Elledge S J, Zou L. Trends Cell Biol. 2002;12:509–516. doi: 10.1016/s0962-8924(02)02380-2. [DOI] [PubMed] [Google Scholar]
  • 4.Rouse J, Jackson S P. Science. 2002;297:547–551. doi: 10.1126/science.1074740. [DOI] [PubMed] [Google Scholar]
  • 5.Unsal-Kacmaz K, Makhov A M, Griffith J D, Sancar A. Proc Natl Acad Sci USA. 2002;99:6673–6678. doi: 10.1073/pnas.102167799. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Lindsey-Boltz L A, Bermudez V P, Hurwitz J, Sancar A. Proc Natl Acad Sci USA. 2001;98:11236–11241. doi: 10.1073/pnas.201373498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Griffiths D J, Barbet N C, McCready S, Lehmann A R, Carr A M. EMBO J. 1995;14:5812–5823. doi: 10.1002/j.1460-2075.1995.tb00269.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Lydall D, Weinert T. Mol Gen Genet. 1997;256:638–651. doi: 10.1007/s004380050612. [DOI] [PubMed] [Google Scholar]
  • 9.Burtelow M A, Roos-Mattjus P M, Rauen M, Babendure J R, Karnitz L M. J Biol Chem. 2001;276:25903–25909. doi: 10.1074/jbc.M102946200. [DOI] [PubMed] [Google Scholar]
  • 10.Volkmer E, Karnitz L M. J Biol Chem. 1999;274:567–570. doi: 10.1074/jbc.274.2.567. [DOI] [PubMed] [Google Scholar]
  • 11.Thelen M P, Venclovas C, Fidelis K. Cell. 1999;96:769–770. doi: 10.1016/s0092-8674(00)80587-5. [DOI] [PubMed] [Google Scholar]
  • 12.Caspari T, Dahlen M, Kanter-Smoler G, Lindsay H D, Hofmann K, Papadimitriou K, Sunnerhagen P, Carr A M. Mol Cell Biol. 2000;20:1254–1262. doi: 10.1128/mcb.20.4.1254-1262.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Venclovas C, Thelen M P. Nucleic Acids Res. 2000;28:2481–2493. doi: 10.1093/nar/28.13.2481. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Griffith J D, Lindsey-Boltz L A, Sancar A. J Biol Chem. 2002;277:15233–15236. doi: 10.1074/jbc.C200129200. [DOI] [PubMed] [Google Scholar]
  • 15.Waga S, Stillman B. Annu Rev Biochem. 1998;67:721–751. doi: 10.1146/annurev.biochem.67.1.721. [DOI] [PubMed] [Google Scholar]
  • 16.Kelman Z, Yao N, O'Donnell M. Gene. 1995;166:177–178. doi: 10.1016/0378-1119(95)00556-7. [DOI] [PubMed] [Google Scholar]
  • 17.Zhang G, Gibbs E, Kelman Z, O'Donnell M, Hurwitz J. Proc Natl Acad Sci USA. 1999;96:1869–1874. doi: 10.1073/pnas.96.5.1869. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Griffith J D, Christiansen G. Annu Rev Biophys Bioeng. 1978;7:19–35. doi: 10.1146/annurev.bb.07.060178.000315. [DOI] [PubMed] [Google Scholar]
  • 19.Gerik K J, Gary S L, Burgers P M. J Biol Chem. 1997;272:1256–1262. doi: 10.1074/jbc.272.2.1256. [DOI] [PubMed] [Google Scholar]
  • 20.St. Onge R P, Besley B D, Park M, Casselman R, Davey S. J Biol Chem. 2001;276:41898–41905. doi: 10.1074/jbc.M105152200. [DOI] [PubMed] [Google Scholar]
  • 21.Kondo T, Wakayama T, Naiki T, Matsumoto K, Sugimoto K. Science. 2001;294:867–870. doi: 10.1126/science.1063827. [DOI] [PubMed] [Google Scholar]
  • 22.Melo J A, Cohen J, Toczyski D P. Genes Dev. 2001;15:2809–2821. doi: 10.1101/gad.903501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Zou L, Cortez D, Elledge S J. Genes Dev. 2002;16:198–208. doi: 10.1101/gad.950302. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Hanna J S, Kroll E S, Lundblad V, Spencer F A. Mol Cell Biol. 2001;21:3144–3158. doi: 10.1128/MCB.21.9.3144-3158.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Mayer M L, Gygi S P, Aebersold R, Hieter P. Mol Cell. 2001;7:959–970. doi: 10.1016/s1097-2765(01)00254-4. [DOI] [PubMed] [Google Scholar]
  • 26.Shiomi Y, Shinozaki A, Nakada D, Sugimoto K, Usukura J, Obuse C, Tsurimoto T. Genes Cells. 2002;7:861–868. doi: 10.1046/j.1365-2443.2002.00566.x. [DOI] [PubMed] [Google Scholar]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES