Cryo-EM structures reveal that RFC recognizes both the 3′- and 5′-DNA ends to load PCNA onto gaps for DNA repair

  1. Fengwei Zheng
  2. Roxana Georgescu
  3. Nina Y Yao
  4. Huilin Li
  5. Michael E O'Donnell

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    The role of Replication Factor C (RFC) in DNA replication and repair has been known for many years. RFC/PCNA binds to a double strand-single strand DNA junction with a 3'-recessed end, with the DNA passing through a central chamber in the five-subunit protein. The current paper reports structures of RFC/PCNA with two separate DNA molecules, one containing the well characterized 3'-recessed DNA and surprisingly, a second 5'-recessed DNA outside the central chamber.The paper is an important addition to understanding RFC function, particularly in DNA repair, but it could be improved with some clarifications. The work is of interest to all studying DNA replication.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1, Reviewer #2 and Reviewer #3 agreed to share their name with the authors.)

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

RFC uses ATP to assemble PCNA onto primed sites for replicative DNA polymerases δ and ε. The RFC pentamer forms a central chamber that binds 3′ ss/ds DNA junctions to load PCNA onto DNA during replication. We show here five structures that identify a second DNA binding site in RFC that binds a 5′ duplex. This 5′ DNA site is located between the N-terminal BRCT domain and AAA+ module of the large Rfc1 subunit. Our structures reveal ideal binding to a 7-nt gap, which includes 2 bp unwound by the clamp loader. Biochemical studies show enhanced binding to 5 and 10 nt gaps, consistent with the structural results. Because both 3′ and 5′ ends are present at a ssDNA gap, we propose that the 5′ site facilitates RFC’s PCNA loading activity at a DNA damage-induced gap to recruit gap-filling polymerases. These findings are consistent with genetic studies showing that base excision repair of gaps greater than 1 base requires PCNA and involves the 5′ DNA binding domain of Rfc1. We further observe that a 5′ end facilitates PCNA loading at an RPA coated 30-nt gap, suggesting a potential role of the RFC 5′-DNA site in lagging strand DNA synthesis.

Article activity feed

  1. Version published to 10.7554/elife.77469 on eLife
  2. Version published to 10.1101/2022.02.04.479194v4 on bioRxiv
  3. eLife

    Evaluation Summary:

    The role of Replication Factor C (RFC) in DNA replication and repair has been known for many years. RFC/PCNA binds to a double strand-single strand DNA junction with a 3'-recessed end, with the DNA passing through a central chamber in the five-subunit protein. The current paper reports structures of RFC/PCNA with two separate DNA molecules, one containing the well characterized 3'-recessed DNA and surprisingly, a second 5'-recessed DNA outside the central chamber.The paper is an important addition to understanding RFC function, particularly in DNA repair, but it could be improved with some clarifications. The work is of interest to all studying DNA replication.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1, Reviewer #2 and Reviewer #3 agreed to share their name with the authors.)

  4. eLife

    Reviewer #1 (Public Review):

    Small, circular protein clamps bind to DNA either at a 3'-recessed ssDNA/dsDNA junction (PCNA clamp) or a 5'-recessed ssDNA/dsDNA junction (9:1:1 clamp). Replication Factor C (RFC) has been known for many years to play a role in DNA replication by loading PCNA onto 3'-recessed ssDNA/dsDNA junction DNAs. But RFC is also known to function in the repair of DNA with single-strand gaps or even single DNA nicks by loading PCNA, which then recruits other proteins to repair the gap or nick.
    The current paper reports the structures of a number of RFC-PCNA-DNA complexes using DNAs with either a 3'-recessed ssDNA/dsDNA junction or a 5'-recessed ssDNA/dsDNA junction. The data show clearly that RFC can bind both DNAs, loading PCNA onto the 3'-recessed ssDNA/dsDNA junction and binding at a second site the DNA with the 5'-recessed ssDNA/dsDNA junction. These data support previous studies that show that RFC can play a role in the repair of DNA lesions on gapped DNA. The studies, while substantially confirmatory of previous results, nevertheless show a definitive structure of the interaction of RFC/PCNA with gapped or nicked DNAs and as such add significantly to understanding mechanisms of DNA repair.

  5. eLife

    Reviewer #2 (Public Review):

    This review is carried out with the caveat that I am not a structural biologist and therefore, cannot judge the correctness of the structures presented in this paper. The paper presents structures of yeast RFC-ATPgammaS with DNA and PCNA in both the open and closed forms. The RFC clamp loader aspects of the study, i.e. complexes with PCNA and 3'-junction DNA and with ATPgammaS, build on previous clamp-clamp loader studies, and it closely resembles a cryoEM study by the Kelch group.

    The novelty of the paper lies in the analysis of the Rfc1 N-terminal BRCT domain, and its binding to 5'-junction DNA. And herein also lies the major concern that I have with the study. Biochemical studies from several labs have shown that the BRCT domain binds with a very high preference to 5'-junctions with a 5'-phosphate. Yet, this study was carried out with DNA substrates that lacked the 5'-phosphate. Two structures are obtained (RFC-3'DNA1-5'DNA2 and PCNA-RFC-3'DNA1-5'DNA2) that show binding of the 5'-junction DNA to the BRCT domain, yet the 5'-junction itself appears not to be engaged in the structure. Without an evaluation of this issue, the importance of this study remains undefined.

    The authors provide additional experiments showing that RFC has a preference for gapped DNA structures. They propose hypothetical models for processes in which gap binding might regulate the execution of these processes, but they do not test these models in yeast. A repair function for the BRCT should be readily testable in yeast, and so should be the phenotype of BRCT point mutations that abrogate DNA interactions.

  6. eLife

    Reviewer #3 (Public Review):

    Sliding clamps are closed protein rings (dimeric or trimeric) that encircle DNA and enable/mediate/stimulate DNA synthesis as well as a variety of other DNA modifying enzymatic reactions. Loading of sliding clamps onto DNA requires the activity of the sliding clamp loader, a heteropentameric AAA+ ensemble that uses the energy of ATP binding and hydrolysis to open the clamp, escort DNA into the opened clamp, and release the clamp for self-closure on DNA. The structure and function of sliding clamps and clamp loaders are conserved throughout evolution, albeit with significant differences in molecular mechanisms.

    Five distinct subunits (RFC1, RFC2, RFC3, RFC4, and RFC5) assemble into the eukaryotic RFC clamp loader. The RFC subunits are structurally related; each feature at least three domains; two of these domains fold into the AAA module (termed AAA-ATP and AAA-lid). The third RFC domain is a helical bundle that combines with related domains from other RFC subunits to assemble into the so-termed RFC collar. The eukaryotic sliding clamp is a homotrimer of PCNA. The RFC clamp loader loads the PCNA ring onto 3' recessed primer-template junctions. Unique to eukaryotes is a series of RFC-like complexes (RLCs) that substitute RFC1 for another subunit to create a new entity with distinct properties. One RLC is the Rad24 clamp loader for RFC1 to create an entity that loses its ability to bind to 3' recessed DNA and PCNA but gains the property of binding to 5' recessed DNA and the 911 clamp.

    Our understanding of the structure and function of clamp/clamp loaders, and especially the eukaryotic loaders, owes much to a series of thoughtful structural, biochemical, and genetic studies over the past three decades. Recently, Kelch and coworkers produced a series of structures of yeast RFC bound to PCNA and primer-template DNA (eLife, 2022). These structures show, inter alia, the details of the primer-template DNA within the loader and with the dsDNA portion of the DNA within the PCNA ring. In addition, Remus, Hite and co-workers and O'Donnell/Li have also submitted Bioarxiv preprints that describe analyses of the Rad24 clamp loader bound to the 911 clamp. Notably, the Rad24 loader appears to exhibit a novel binding mode. The 5' recessed DNA in the Rad24 loader is not found within the loader and the clamp, but 'above' (in a specific reference pose) the 911 ring.

    The finding that the Rad24 loader acquires the capacity to bind DNA above the ring by swapping a single subunit (Rad24 for RFC1) prompted Zheng et al to ask whether this property is unique to the Rad24 loader or does the RFC loader also harbors this property. Astonishingly, the answer is likely: yes.

    The authors note prior biochemical evidence suggesting that the picture provided by prior structural biology of RFC was incomplete. Susan Hardin and coworkers (1998) and Gregg Siegel and coworkers (2006/2010) had investigated the BRCT domain found at the N-terminus of the RFC1 subunit (Drosophila and Human) and had found that, in isolation, this domain bound to one DNA molecule with a preference for a recessed 5' DNA structure. The finding that RFC had binding sites for 3' and 5' recessed DNA was not accommodated in the extant crystal and EM structures.

    This manuscript encompasses the following studies, experiments, and findings:

    1. 5 cryo-EM derived models:

    a. RFC with DNA in the loader (3'-DNA1) at 3.3 Å;

    b. RFC with two DNA molecules (5'DNA2) at 3.3 Å;

    c. RFC bound to the closed planar PCNA ring, 3' DNA (RFC−closed PCNA−DNA1) at 3.3 Å;

    d. RFC bound to the open PCNA (with a 14 Å opening) ring, with 3' DNA (RFC−open PCNA−DNA1) at 3.4 Å;

    e. RFC bound to the closed PCNA ring with the 3'DNA and the 5' DNA and the 5' DNA (3.1 Å) featuring a highly ordered model of the BRCT domain at the amino terminus of RFC1 RFC−3′-DNA1−5′-DNA2−PCNA (the BRCT domain had never been visualized in any RFC structure until now).

    The cryo-EM analyses that included DNA were carried out with a DNA oligo with both 3' and 5' recessed ends. However, in none of the above cryo-EM analyses was the DNA visualized in its entirety.

    The cryo-EM analyses in this work are of excellent quality as reflected in Supplementary Table 1 and the PDB validation reports.

    The main novel finding of the above structures is the presence in the RFC loader of a second binding site for DNA. This second DNA site is located above the PCNA ring, with the dsDNA segment resting on the AAA+ ATPase domain and packed again against the AAA-lid domain. In one structure, the dsDNA is also packed against the ordered BRCT domain of RFC1 (RFC-3'DNA1-5'DNA2-PCNA). This work represents the first description of RFC structures bound to two DNA molecules. The finding of a second DNA binding site above the PCNA ring links this work to efforts to understand the structure and function of the Rad24 loader.

    From the series of structures, the authors conclude that the 3' DNA binds first in the RFC clamp loader; this binding event leads to a structural change in RFC1 that enables the formation of the second DNA binding site (for the 5' DNA structure). This structural change is like that seen in the Kelch 2022 series of structures.

    Left unexplained is why the BRCT domain becomes ordered in the presence of PCNA and the second DNA molecule (RFC−3′-DNA1−5′-DNA2−PCNA) but is not ordered in the structure without PCNA (5'DNA2). Perhaps, the authors could address this.

    Zheng et al suggest that the 3' and 5' DNA molecule binds in a manner reminiscent of how a single DNA molecule with a gap might bind. Inspection of the structure suggests that a gap of 7 nucleotides (or larger with extrusion) might be the optimal size. This is to say that the two input DNA molecules bind with the chain directions arranged just so as to enable (virtual) linkage. Ample evidence exists in structural and biochemical studies to conclude how the 3' DNA is situated in the RFC loader chamber. However, no such evidence exists for the second DNA molecule. And, while the proposed virtual linkage scheme makes sense (and other schemes do not), could the authors comment on whether the chain direction for the second DNA molecule could be ascertained directly from the EM maps?

    1. Biochemical analysis RFC-PCNA on DNA with nicks and varying length gaps.

    a. The authors conclude that RFC binds tightly to all the substrates tested but binds with a two-fold (~15 nM vs ~30 nM) higher affinity for DNA molecules with gaps that have five or more bases.

    b. The 5 nucleotide or large gaps compares well with the structural finding.

    c. The presence of PCNA in the measurement reveals tighter binding than in its absence, but with the same trend as seen when PCNA was omitted.

    d. Analysis of an RFC ensemble whose RFC1 lacks the BRCT domain revealed 2-fold tighter binding (16 nM vs. 32 nM) when BRCT was present than when not.

    In view of the extensive set of contacts to DNA mediated by the BRCT domain, it is surprising that there is not a greater difference in Kd between primer-template and gapped DNA structures. Likewise, why does deletion of BRCT hardly change the affinity for DNA?

    Supplementary Figure 7 provides the fits to the fluorescence binding data, which were fit to a simple one-site model corrected for non-specific binding. Could the authors provide the equation used? Also, are two copies of the primed-DNA template expected to bind to RFC as seen in the cryo-EM structure. If so, would it be appropriate for the binding data to be analyzed with a two-site model?

    1. PCNA loading experiments on DNA structures harboring 10 or 50 nucleotide gaps (without and with RPA coating).

    a. Measurements of loading efficiencies on the above DNA substrates showed a higher efficiency of loading when both primer #1 and primer #2 are included, in comparison to measurements with primer #1 alone.

    Taken together, the structural and biochemical data point to the possibility that two DNA molecules, or gapped DNA structures, could be involved in the RFC loading reaction.

    The authors then discuss their findings in the broader context of DNA repair and DNA replication.

  7. Version published to 10.1101/2022.02.04.479194v3 on bioRxiv
  8. Version published to 10.1101/2022.02.04.479194v2 on bioRxiv
  9. Version published to 10.1101/2022.02.04.479194v1 on bioRxiv