Author response:
The following is the authors’ response to the original reviews.
Reviewer #1 (Public review):
Summary:
The authors report the structure of the human CTF18-RFC complex bound to PCNA. Similar structures (and more) have been reported by the O'Donnell and Li labs. This study should add to our understanding of CTF18-RFC in DNA replication and clamp loaders in general. However, there are numerous major issues that I recommend the authors fix.
Strengths:
The structures reported are strong and useful for comparison with other clamp loader structures that have been reported lately.
Weaknesses:
The structures don't show how CTF18-RFC opens or loads PCNA. There are recent structures from other groups that do examine these steps in more detail, although this does not really dampen this reviewer's enthusiasm. It does mean that the authors should spend their time investigating aspects of CTF18-RFC function that were overlooked or not explored in detail in the competing papers. The paper poorly describes the interactions of CTF18-RFC with PCNA and the ATPase active sites, which are the main interest points. The nomenclature choices made by the authors make the manuscript very difficult to read.
Reviewer #2 (Public review):
Summary
Briola and co-authors have performed a structural analysis of the human CTF18 clamp loader bound to PCNA. The authors purified the complexes and formed a complex in solution. They used cryo-EM to determine the structure to high resolution. The complex assumed an auto-inhibited conformation, where DNA binding is blocked, which is of regulatory importance and suggests that additional factors could be required to support PCNA loading on DNA. The authors carefully analysed the structure and compared it to RFC and related structures.
Strength & Weakness
Their overall analysis is of high quality, and they identified, among other things, a human-specific beta-hairpin in Ctf18 that flexibly tethers Ctf18 to Rfc2-5. Indeed, deletion of the beta-hairpin resulted in reduced complex stability and a reduction in a primer extension assay with Pol ε. This is potentially very interesting, although some more work is needed on the quantification. Moreover, the authors argue that the Ctf18 ATP-binding domain assumes a more flexible organisation, but their visual representation could be improved.
The data are discussed accurately and relevantly, which provides an important framework for rationalising the results.
All in all, this is a high-quality manuscript that identifies a key intermediate in CTF18dependent clamp loading.
Reviewer #3 (Public review):
Summary:
CTF18-RFC is an alternative eukaryotic PCNA sliding clamp loader that is thought to specialize in loading PCNA on the leading strand. Eukaryotic clamp loaders (RFC complexes) have an interchangeable large subunit that is responsible for their specialized functions. The authors show that the CTF18 large subunit has several features responsible for its weaker PCNA loading activity and that the resulting weakened stability of the complex is compensated by a novel beta hairpin backside hook. The authors show this hook is required for the optimal stability and activity of the complex.
Relevance:
The structural findings are important for understanding RFC enzymology and novel ways that the widespread class of AAA ATPases can be adapted to specialized functions. A better understanding of CTF18-RFC function will also provide clarity into aspects of DNA replication, cohesion establishment, and the DNA damage response.
Strengths:
The cryo-EM structures are of high quality enabling accurate modelling of the complex and providing a strong basis for analyzing differences and similarities with other RFC complexes.
Weaknesses:
The manuscript would have benefitted from more detailed biochemical analysis to tease apart the differences with the canonical RFC complex.
I'm not aware of using Mg depletion to trap active states of AAA ATPases. Perhaps the authors could provide a reference to successful examples of this and explain why they chose not to use the more standard practice in the field of using ATP analogues to increase the lifespan of reaction intermediates.
Overall appraisal:
Overall the work presented here is solid and important. The data is sufficient to support the stated conclusions and so I do not suggest any additional experiments.
Reviewer #1 (Recommendations for the authors):
We thank the reviewer for their positive comments and for their thorough review. All raised points have been addressed below.
Major points
(1) The nomenclature used in the paper is very confusing and sometimes incorrect. The authors refer to CTF18 protein as "Ctf18", and the entire CTF18-RFC complex as "CTF18". This results in massive confusion because it is hard to ascertain whether the authors are discussing the individual subunits or the entire complex. Because these are human proteins, each protein name should be fully capitalized (i.e. CTF18, RFC4 etc). The full complex should be referred to more clearly with the designation CTF18-RFC or CTF18-RLC (RFC-like complex). Also, because the yeast and human clamp loader complexes use the same nomenclature for different subunits, it would be best for the authors to use the "A, B, C, D, E subunit" nomenclature that has been standard in the field for the past 20 years. Finally, the authors try to distinguish PCNA subunits by labeling them "PCNA2" or "PCNA1" (see Page 8 lines 180,181 for an example). This is confusing because the names of the RFC subunits have similar formats (RFC2, RFC3, RFC4, etc). In the case of RFC this denotes unique genes, whereas PCNA is a homotrimer. Could the authors think of another way to denote the different subunits, such as super/subscript? PCNA-I, PCNA-II, PCNA-III?
We thank the reviewer for pointing out the confusing nomenclature. Following the referee suggestion, we now refer to the CTF18 full complex as “CTF18-RFC”. We prefer keeping the nomenclature used for CTFC18 subunits as RFC2, RFC3 etc., as recently used in Yuan et al, Science, 2024. However, we followed the referee’s suggestion for PCNA subunits, now referred to as PCNA-I, PCNA-II and PCNA-III.
(2) I believe that the authors are over-interpreting their data in Figure 1. The claim that "less sharp definition" of the map corresponding to the AAA+ domain of Ctf18 supports a relatively high mobility of this subunit is largely unsubstantiated. There are several reasons why one could get varying resolution in a cryo-EM reconstruction, such as compositional heterogeneity, preferred orientation artifacts, or how the complex interacts with the air-water interface. If other data were presented that showed this subunit is flexible, this evidence would support that data but cannot alone as justification for subunit mobility. Along these lines, how was the buried surface area (2300 vs 1400 A2) calculated? Is this the total surface area or only the buried surface area involving the AAA+ domains? It is surprising that these numbers are so different considering that the subunits and complexes look so similar (Figures 1c and 2b).
We respectfully disagree with the suggestion that our interpretation of local flexibility in the AAA+ domain of Ctf18 is overreaching. Several lines of evidence support this interpretation. First, compositional heterogeneity is unlikely, as the A′ domain of Ctf18 is well-resolved and forms stable interactions with RFC3, indicating that Ctf18 is consistently incorporated into the complex. Second, preferred orientation artifacts are excluded, as the particle distribution shows excellent angular coverage (Fig. S9a). Third, we now include a 3D variability analysis (3DVA; Supplementary Video 1), which reveals local conformational heterogeneity centered around the AAA+ domain of Ctf18, consistent with intrinsic flexibility.
Regarding the buried surface area values, the reported numbers refer specifically to the interfaces between the AAA+ domain of Ctf18 and RFC2, and are derived from buried surface area calculations performed with PISA. The smaller interface (~1400 Ų) compared to RFC1–RFC2 (~2300 Ų) reflects low sequence identity (~26%) and divergent structural features, including the absence of conserved elements such as the canonical PIP-box in Ctf18. We have clarified and expanded this explanation in the revised manuscript (Page 7).
(3) The authors very briefly discuss interactions with PCNA and how the CTF18-RFC complex differs from the RFC complex. This is amongst the most interesting results from their work, but also not well-developed. Moreover, Figure 3D describing these interactions is extremely unclear. I feel like this observation had potential to be interesting, but is largely ignored by the authors.
We thank the referee for pointing this out. We have expanded the section describing the interactions of CTF18-RFC and PCNA (Page 9 in the new manuscript), and made a new panel figure with further details (Fig. 3D).
(4) The authors make the observation that key ATP-binding residues in RFC4 are displaced and incompatible with nucleotide binding in their CTF18-RFC structure compared to the hRFC structure. This should be a main-text figure showing these displacements and how it is incompatible with ATP binding. Again, this is likely an interesting finding that is largely glossed over by the authors.
We now discuss this feature in detail (Pag 11 in the new manuscript), and added two figure insets (Fig. 4c) describing the incompatibility of RFC4 with nucleotide binding.
(5) The authors claim that the work of another group (citation 50) "validate(s) our predictions regarding the significant similarities between CTF18-RFC and canonical RFC in loading PCNA onto a ss/dsDNA junction." However, as far as this reviewer can tell the work in citation 50 was posted online before the first draft of this manuscript appeared on biorxiv, so it is dubious to claim that these were "predictions."
We agree with the referee about this claim. We have now revised the text as follows:
“While our work was being finalized, several cryo-EM structures of human CTF18-RFC bound to PCNA and primer/template DNA were reported by another group (He et al, PNAS, 2024). These findings are consistent with the distinct features of CTF18-RFC observed in our structures and independently support the notion of significant mechanistic similarity between CTF18-RFC and canonical RFC in loading PCNA onto a ss/dsDNA junction”.
(6) The authors use a primer extension assay to test the effects of truncating the Nterminal beta hairpin of CTF18. However, this assay is only a proxy for loading efficiency and the observed effects of the mutation are rather subtle. The authors could test their hypothesis more clearly if they performed an ATPase assay or even better a clamp loading assay.
We thank the referee for this valuable suggestion. In response, we have performed clamp loading assays comparing the activities of human RFC, wild-type CTF18-RFC, and the β-hairpin–truncated CTF18-RFC mutant. The results, now presented in Fig. 6 and Table 1 of the revised manuscript, clearly show that truncation of the N-terminal βhairpin results in a slower rate of PCNA loading. We propose that this reduced loading rate likely contributes to the diminished Pol ε–mediated DNA synthesis observed in the primer extension assays.
Minor points
(1) Page 3 line 53 the introduction suggests that ATP hydrolysis prompts clamp closure. While this may be the case, to my knowledge all recent structural work shows that closure can occur without ATP hydrolysis. It may be better to rephrase it to highlight that under normal loading conditions, ATP hydrolysis occurs before clamp closure.
The text now reads (Page 3):
“DNA binding prompts the closure of the clamp and hydrolysis of ATP induces the concurrent disassembly of the closed clamp loader from the sliding clamp-DNA complex, completing the cycle necessary for the engagement of the replicative polymerases to start DNA synthesis.”
(2) Page 3 line 60, I do not see how the employment of alternative loaders highlights the specificity of the loading mechanism - would it not be possible for multiple loaders to have promiscuous clamp loading?
We thank the referee for this comment. The text now reads (Page 3):
“However, eukaryotes also employ alternative loaders (20), including CTF18-RFC (6, 21-24), which likely use a conserved loading mechanism but are functionally specialized through specific protein interactions and context-dependent roles in DNA replication.”
(3) Page 4 line 75 could you please cite a study that shows Ctf8 and Dcc1 bind to the Ctf18 C-terminus and that a long linker is predicted to be flexible?
Two references have been added (Stokes et al, NAR, 2020 and Grabarczyk et al, Structure, 2018)
(4) Figure 2A has the N-terminal region of Ctf18 as bound to RFC3 but should likely be labeled as bound to RFC5. This caused significant confusion while trying to parse this figure. Further, the inclusion of "X" as a sequence - does this refer to a sequence that was not buildable in the cryo-EM map? I would be surprised that density immediately after the conserved DEXX box motif is unbuildable. If this is the case, it should be clearly stated in the figure legend that "X" denotes an unbuildable sequence. For the conserved beta-hairpin in the sequence, could the authors superimpose the AlphaFold prediction onto their structure? It would be more informative than just looking at the sequence.
We apologize for this confusion. The error in Figure 2A has been corrected. The figure caption now explicitely says that “X” refers to amino acid residues in the sequence which were not modelled. A superposition of the cryo-EM model of the N-terminal Beta hairpin in human Ctf18 and AlphaFold predictions for this feature in drosophila and yeast Ctf18 is now presented in Figure 2A.
(5) Page 8 line 168, the use of the term "RFC5" here feels improper, since the "C" subunit is not RFC5 in all lower eukaryotes (see comment above about nomenclature). For instance, in S cerevisiae, the C subunit is RFC3. I would expect this interaction to be maintained in all C subunits, not all RFC5 subunits.
The text now reads (Page 8):
“Therefore, lower eukaryotes may use a similar b-hairpin motif to bind the corresponding subunit of the RFC-module complex (RFC5 in human, Rfc3 in S. cerevisiae), emphasizing its importance.”
(6) Page 10 line 228, the authors claim that hydrolysis is dispensable at the Ctf18/RFC2 interface based on evidence from RFC1/RFC2 interface, by analogy that this is the "A/B" interface in both loaders. However, the wording makes it sound as if the cited data were collected while studying Ctf18 loaders. The authors should clarify this point.
The text has been modified as follows (Pag 11):
“Prior research has indicated that hydrolysis at the large subunit/RFC2 interface is not essential for clamp loading by various loaders (48-51), while the others are critical for the clamp-loading activity of eukaryotic RFCs. “
(7) Page 11 line 243/244 the authors introduce the separation pin. Could they clarify whether Ctf18 contains any aromatic residues in this structural motif that would suggest it serves the same functional purpose? Also, the authors highlight this is similar to yeast RFC, which makes it sound like this is not conserved in human RFC, but the structural motif is also conserved in human RFC.
We thank the reviewer for this helpful comment. We have clarified in the revised text (Page 12) that the separation pin is conserved not only in yeast RFC but also in human RFC, and now note that human Ctf18 also harbors aromatic residues at the corresponding positions. This observation is supported by the new panel in Figure 4e.
Minutia
(1) Page 2 line 37 please remove the word "and" before PCNA.
This has been corrected.
(2) Please define AAA+ and update the language to clarify that not all pentameric AAA+ ATPases are clamp loaders.
AAA+ has been now defined (Page 3).
(3) Page 4 line 86 Given the relatively weak interaction of Pol ε.
This has been corrected.
(4) Page 8 line 204 the authors likely mean "leucine" and not "lysine".
We thank the reviewer for catching this. The error has been corrected.
(5) Page 14 line 300, the authors claim that CTF18 utilizes three subunits but then list four.
We have corrected this.
Reviewer #2 (Recommendations for the authors):
We thank the reviewer for their positive comments and valuable suggestions. The points raised by the referee have been addressed below.
Major point:
(1) Please quantify Figure 6 and S9 from 3 independent repeats and determine the standard deviation to show the variability of the Ctf18 beta hairpin deletion. The authors suggest that a suboptimal Ctf18 complex interaction with PCNA impacts the stability of the complex, but do not test this hypothesis. Could the suboptimal PIP motif in Ctf18 be changed to an improved motif and the impact tested in the primer extension assay? Although not essential, it would be a nice way to explore the mechanism.
We thank the reviewer for the suggestion. However, we note that Figure 6b (now 7b) already presents the quantification of the primer extension assay from three independent replicates, with error bars showing standard deviations, and includes the calculated rate of product accumulation. These data clearly indicate a 42% reduction in primer synthesis rate upon deletion of the Ctf18 β-hairpin.
We agree that we do not provide direct evidence of impaired complex stability upon deletion of the Ctf18 β-hairpin. However, the 2D classification of the cryo-EM dataset (Figure S9) shows a marked reduction in the number of particles corresponding to intact CTF18-RFC–PCNA complexes in the β-hairpin deletion sample, with the majority of particles corresponding to free PCNA. This contrasts with the wild-type dataset, where complex particles are predominant. These findings indirectly suggest that deletion of the β-hairpin compromises the stability or assembly of the clamp-loader–clamp complex.
We thank the reviewer for the valuable suggestion to mutate the weak PIP-box of Ctf18. While an interesting direction, we instead sought to directly test the mechanism by performing quantitative clamp loading assays. These assays revealed a significant reduction in the rate of PCNA loading by the CTF18Δ165–194-RFCmutant (Figure 6), supporting the conclusion that the β-hairpin contributes to productive PCNA loading. This loading delay likely underlies the reduced rate of primer extension observed in the Pol ε assay (Figure 7), consistent with impaired formation of processive polymerase– clamp complexes.
(2) I did not see the method describing how the 2D classes were quantified to evaluate the impact of the Ctf18 beta hairpin deletion on complex formation. Please add the relevant information.
The relevant information has been added to the Method section:
“For quantification of complex stability, the number of particles contributing to each 2D class was extracted from the classification metadata (Datasets 1 and 3). All classes showing isolated PCNA rings were summed and compared to the total number of particles in classes representing intact CTF18-RFC–PCNA complexes. This analysis was performed for both wild-type and β-hairpin deletion mutant datasets. Notably, no 2D classes corresponding to free PCNA were observed in the wild-type dataset, whereas in the mutant dataset, a substantial fraction of particles corresponded to isolated PCNA, suggesting reduced stability of the mutant complex.”
Minor point:
(1) Page 2, line 25. Detail what type of mobility is referred to. Do you mean flexibility in the EM-map?
We have clarified this. The text now reads:
“The unique RFC1 (Ctf18) large subunit of CTF18-RFC, which based on the cryo-EM map shows high relative flexibility, is anchored to PCNA through an atypical low-affinity PIP box”
(2) Page 4, line 82. Please introduce CMGE, or at least state what the abbreviation stands for.
This has been addressed.
(3) Page 4, line 89. Specify that the architecture of the HUMAN CTF18-RFC module is not known, as the yeast one has been published.
At the time our study was initiated, the architecture of the human CTF18-RFC module was unknown. A structure of the human complex was published by another group during the final stages of our work and is now properly acknowledged in the Discussion.
(4) Page 6. Is it possible to illustrate why the autoinhibited state cannot bind to DNA? A visual representation would be nice.
We thank the reviewer for this suggestion. Figure 4b in the original manuscript already illustrates why the autoinhibited, overtwisted conformation of the CTF18-RFC pentamer cannot accommodate DNA. In this state, the inner chamber of the loader is sterically occluded, precluding the binding of duplex DNA.
Reviewer #3 (Recommendations for the authors):
We thank Reviewer #3 for their constructive feedback and positive overall assessment of our work.
We also thank the reviewer for their remarks on the use of Mg depletion to halt hydrolysis. Magnesium is an essential cofactor for ATP hydrolysis, and its depletion is expected to effectively prevent catalysis by destabilizing the transition state, possibly more completely than the use of slowly hydrolysable analogues such as ATPγS. We have recently employed Mg²+ depletion to successfully trap a pre-hydrolytic intermediate in a replicative AAA+ helicase engaged in DNA unwinding (Shahid et al., Nature, 2025). This precedent supports the rationale for our choice, and the reference has now been included in the revised manuscript.
I think the authors deposited the FSC curve for the +Mg structure in the -Mg structure PDB/EMDB entry according to the validation report.
We thank the reviewer for their careful inspection of the deposition materials. The discrepancy in the deposited FSC curve has now been corrected, and the appropriate FSC curves have been assigned to the correct PDB/EMDB entries.
