CIP2A Mediates the Recruitment of the SLX4-MUS81-XPF Tri-Nuclease Complex in Mitosis and Protects Against Replication Stress

  1. Alice Meroni
  2. Annica Pellizzari
  3. Andrea Hänel
  4. Nathalie Varisco
  5. Pascale Brasier-Lutz
  6. Isabell Witzel
  7. Alessandro A Sartori
  8. Manuel Stucki

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

DNA replication stress frequently elicits problems in mitosis because incompletely replicated chromosomes or replication intermediates physically link sister chromatids together and prevent their proper segregation during anaphase. We and others recently discovered a mitotic role for the CIP2A oncoprotein, which is critically implicated in chromosome stability maintenance and chromosome fragment clustering during mitosis. In addition, in homologous recombination deficient (HRD) cells, CIP2A is essential and may thus constitute a new drug target in HRD cancers. Yet, the precise mechanisms by which CIP2A suppresses chromosomal instability during mitosis and thus allows for the survival of HRD cancer cells remain largely elusive. Here we characterize CIP2A’s role in DNA replication stress responses. We show that upon replication stress, wild-type cells show an elevated accumulation of CIP2A foci during mitosis, indicating its involvement in the mitotic response to replication stress. Defective DNA replication leads to the accumulation of under-replicated DNA, which can be carried into mitosis. We demonstrate that in the absence of CIP2A, cells fail to recruit the SLX4-MUS81-XPF (SMX) tri-nuclease complex to sites of under-replicated DNA in mitosis, resulting in a high incidence of lagging chromosomes during anaphase and subsequent micronuclei formation. In a subset of cell lines, we also observed CIP2A-dependent mitotic DNA synthesis (MiDAS) upon replication stress. However, our data suggest that MiDAS and SMX recruitment are not functionally linked. This novel role of CIP2A in managing under-replicated DNA may provide insights into the molecular mechanisms underlying therapeutic vulnerabilities in cancer cells.

Article activity feed

  1. Review Commons

    Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Reply to the Reviewers

    I thank the Referees for their...

    Referee #1

    1. The authors should provide more information when...

    Responses

    • The typical domed appearance of a hydrocephalus-harboring skull is apparent as early as P4, as shown in a new side-by-side comparison of pups at that age (Fig. 1A).
    • Though this is not stated in the MS
    1. Figure 6: Why has only...

    Response: We expanded the comparison

    Minor comments:

    1. The text contains several...

    Response: We added...

    Referee #2

  2. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    Meroni and colleagues present evidence that CIP2A is required to recruit the SMX complex to sites of replication stress in mitotic cells. Whilst the data generated when using U2OS cells seems to support a role for CIP2A in recruiting the SMX complex to sites of replication stress to facilitate MiDAS, as the authors point out, this pathway is not conserved in DLD1 cells. Although the authors suggest that this discrepancy in the data may relate to the fact that U2OS cells are ALT positive and the DLD1 cells are not, there is no experimental evidence to support this hypothesis. It would have been nice if the authors had backed up this hypothesis with data relating to how CIP2A regulates the SMX-MiDAS pathway in other ALT positive and negative cell lines. Taken together, after reading this manuscript, I am left wondering whether CIP2A is really important for SMX-dependent MiDAS or whether it is phenomenon that is found in some commonly used cancer cell lines and not others. Whilst it is important to publish conflicting results as they can explain why some research labs can reproduce published data and others can't, I think this manuscript would benefit from assessment of the role of CIP2A in mediating the recruitment of the SMX complex to carry out MiDAS in a variety of additional cancer cell lines and also non-cancer cell lines, such as RPE1-hTERT cells to obtain some sort of consensus about the importance of CIP2A in dealing with mitotic replication stress.

    Comments:

    1. Fig.2A-E: Can the authors comment on the difference in number of APH-induced FANCD2, SLX4, Mus81 and XPF foci in mitotic U2OS cells? Given that SLX4 should be recruiting both XPF and Mus81, there is a disparity between the numbers of mitotic foci given that there are approximately 30 FANCD3 foci per mitotic cell following APH treatment. Additionally, why do the XPF foci not increase after APH exposure?
    2. Fig.2G: I would say that the 'full rescue' of Mus81 foci in the CIP2A KO cells complemented with WT CIP2A is not hugely convincing since there is only a difference of 1-2 foci between the WT and CIP2A KO cells treated with APH.
    3. Fig.3A: I am not really sure how biologically meaningful a difference of 0.03-0.04 EdU foci per chromosome is when comparing BRCA2 KO DLD1 cells treated with control siRNA versus CIP2A siRNA. Would it not have been better to treat the BRCA2 KO DLD1 cells with APH?
    4. Fig.3H-I: Given that the reduction in MiDAS in the CIP2A KO cl.7 cell line is likely a clonal artifact not related to the loss of CIP2A, it is unclear how to interpret the data about the EdU foci pattern on chromosomes presented in Fig.3H-I and its relevance to CIP2A. Therefore, I am not sure this data really adds anything to the manuscript.
    5. Fig.4H: The difference in Mus81 foci per mitotic cell with/without the expression of B6L is only one focus per mitotic cell. Based on this, it is difficult to make any real conclusions about whether the TOPBP1-CIP2A interaction is really required for the recruitment of Mus81 to sites of mitotic replication stress.

    Significance

    As mentioned above, it is clear that the role of CIP2A in regulating the mitotic replication stress response by promoting recruitment of the SMX complex to sites of mitotic replication stress to promote MiDAS is complicated and may be specific to some cancer cell lines and not others. Whilst it is not clear what the underlying reason for this is, this manuscript would definitely benefit from additional analysis of this pathway in other cancer and non-cancer cell lines to obtain a consensus about the role of CIP2A.

    This manuscript would appeal to fundamental research scientists interested in understanding the mechanisms underlying DNA damage repair, the replication stress response and mitotic regulation.

  3. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    Summary:

    In the manuscript entitled "CIP2A Mediates the Recruitment of the SLX4-MUS81-XPF Tri-Nuclease Complex in Mitosis and Protects Against Replication Stress" by Meroni et al the authors have characterized localization of the CIP2A-TopBP1 complex as well as some aspects of its function in U2OS and DLD1 cell lines exposed to different types of stress. They find that replication stress due to BRCA2 KO, APH or ATRi results in increased focus formation of the CIP2A-TopBP1 complex in mitotic cells. Moreover, the authors find significant decrease in EdU incorporartion in mitotic cells when disrupting CIP2A in (i) U2OS exposed to ATRi or Aph; (ii) in DLD1 BRCA2 KO; (iii) in one clone of DLD1 with Cip2A KO, and a non significant decrease the other DLD1 with Cip2A KO that they tested. Thus, under most of the tested conditions CIP2A is facilitating MiDAS. However, the authors find that expression of a previously characterised fragment of TopBP1 called B6L, which disrupts CIP2A-TopBP1 interaction, does not inhibit MiDAS in DLD1 cells.

    Major comments:

    It is convincing but not surprising that CIP2A-TopBP1 form more foci in mitotic cells after replication stress. The authors statement in the abstract: "We demonstrate that in the absence of CIP2A, cells fail to recruit the SLX4-MUS81-XPF (SMX) tri-nuclease complex to sites of under-replicated DNA in mitosis, resulting in a high incidence of lagging chromosomes during anaphase and subsequent micronuclei formation" is not supported by experiments. The authors indeed show that absence of CIP2A leads to lagging chromosomes during anaphase and subsequent micronuclei formation (which has previously been shown) but they have not shown that it is the failure to recruit the SMX complex that results in the phenotypes they mention. The authors should rephrase or remove this claim.

    There is a discrepancy between the B6L-mediated disruption of TopBP1-CIP2A interaction having no effect on MiDAS in DLD1 cells (fig. 4F) whereas knockout of CIP2A in DLD1 cells seem to have an effect (fig 3E). The most obvious explanation for this observation is that the B6L peptide does not fully abolish TopBP1-CIP2A interaction and can still allow for some SLX4-MUS81 recruitment that is not visible as foci but still sufficient to induce MiDAS. To understand whether MiDAS in DLD1 expressing B6L is dependent on the fraction of TopBP1 that can still form foci (according to Fig 4D) the authors must co-stain for TopBP1 together with EdU detection to address whether they observe any colocalization of TopBP1 with MiDAS.

    Many of the experiments are only performed with two independent replicates. The authors must perform 3 independent replicates. Also, it is not clear how many cells were analysed for each replicate. This should be clearly stated and the mean of each replicate should always be shown. Statistical analyses should be carried out using the means of the replicates. The authors must provide data showing the efficiency of CIP2A knockdown and CIP2A expression in the complementation assay (Fig. 2G)

    Minor comments:

    The authors should change "U-2 OS" in the figures to "U2OS" for consistency.

    In figure 4D - is the increase with APH and S1 significant compared to S1 alone?

    Figure 3 B and C. It is worrying that there is a huge difference in the EdU foci/mitotic cell in untreated condition from panel B to pabel C.

    Fig 3F - is the increase in EdU incorporation after complementation significant?

    For figure 3I representative images should be added

    Significance

    The data presented in the manuscript is of high quality but unfortunately does not present a big advance compared to current knowledge. Nevertheless, it is useful to have side-by-side comparison of different cell lines and conditions and IF localization studies. Given the therapeutic interest in the CIP2A-TopBP1 pathway it is important to get all the details right and researches with interest in DNA repair during mitosis will have interest in this work.

    Moreover, in this manuscript the authors demonstrate that the impact of CIP2A disruption on MiDAS is variable across different cell lines-and even between individual clones. The concept of MiDAS is still clouded by considerable ambiguity, possibly due to earlier studies overstating the consequences of knockdown or knockout. It is therefore great that this manuscript presents clear, unbiased observations, highlighting both inter-cell line differences and the partial nature of the effects. This kind of nuanced reporting is valuable for the field.

  4. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    Summary and Significance

    This is a timely and exciting study that provides us with some new molecular insights into mitotic DNA repair. It builds on previous studies that identified the CIP2A-TOPBP1 complex as a molecular tether that connects broken DNA ends that get transmitted from interphase into mitosis (PMID: 30898438, 35842428, 35842428). The results are also largely complementary with those of Martin et al. (BioRxiv preprint at https://doi.org/10.1101/2024.11.12.621593) and de Haan et al. (BioRxiv preprint at https://www.biorxiv.org/content/10.1101/2025.04.03.647079v1).

    The authors report three main findings, as summarized below.

    1. The CIP2A oncoprotein is involved in the cellular response to replication stress in mitosis.

    2. CIP2A is required for the recruitment of SLX4, MUS81, and XPF into foci during mitosis. SLX4 is a well-established protein scaffold for multiple DNA repair factors, including three structure-selective endonucleases called SLX1, MUS81-EME1, and XPF-ERCC1 that together, form the SMX tri-nuclease that removes DNA repair intermediates and chromosome entanglements during mitosis. In some cell lines, the SMX complex is required for mitotic DNA synthesis at sites of under-replicated DNA, thus ensuring complete DNA replication prior to cell division.

    3. The role(s) of CIP2A in MiDAS are cell line-dependent/context-dependent.

    In general, this is a solid body of microscopy-based work that includes appropriate cell models and experimental controls. The manuscript is well-written, and the data is presented coherently. The main findings will have important implications for researchers interested in mitotic DNA damage, genome stability, and cancer biology. After addressing the points below, I believe this manuscript will be suitable for publication.

    Major comments

    1. Figure 1C: The CIP2A-TOPBP1 PLA experiments are lacking critical controls, namely cells lacking or depleted of CIP2A and TOPBP1. These controls are necessary to provide confidence for the results presented in Figure 1C. If these controls are too expensive or time-demanding for the manuscript, then I recommend removing the PLA data from Figure 1C.

    2. In Figure 2, the authors conclude that the loss of SLX4, XPF, and MUS81 foci in CIP2A depleted cells is synonymous with the loss of recruitment to DNA lesions. However, I can think of many other reasons that could explain the loss of foci. For example, do the authors know that the proteins are expressed to similar levels in cells with and without CIP2A (this should be tested by a simple western blot). Along the same vein, a biochemical fractionation and western blot of the soluble vs chromatin-bound fraction would complement and substantiate their microscopy-based assays in Figure 2. If the fractionation is not possible, then the text should be adjusted accordingly.

    3. The experimental set-up in Figure 2 probes whether CIP2A mediates the recruitment of SMX subunits - SLX4, XPF, MUS81 - but not the SMX complex per se, which would require the study of SLX4 point mutants that selectively ablate the interactions with XPF or MUS81 (but not CIP2A). As such, I suggest that they rephrase their wording appropriately.

    4. Western blots must be provided to substantiate the experiments performed with siRNA (Figure 1G-J, Figure 2A-E and 2H, Figure 3A-D, Figure 5B-D). Similarly, the authors should provide western blots to confirm the BRCA2 and CIP2A statuses in their KO cell lines, as well as the complementation cell lines. In the absence of this information, it is difficult for someone to make an independent and meaningful interpretation of their data.

    5. Most of the data presented in this manuscript is derived from n = 2 biological replicates. All of the experiments reported in the study should be repeated for n = 3 biological replicates.

    6. Since the authors report the median of their data, they should also report the interquartile range or confidence interval to display the uncertainty.

    Minor comments

    1. The references can be improved by acknowledging some of the foundational papers on SLX4 and the SMX tri-nuclease.

    1.a) Page 3: Neither Minocherhomji et al. 2015 nor Pedersen et al. 2015 were the first to describe SLX4 as a scaffold for structure-selective endonucleases. The founding papers were published in 2009 (Svendsen et al. 2009, Munoz et al. 2009, Fekairi et al. 2009, Andersen et al. 2009) with important mechanistic studies on nuclease activation reported in 2013 (Wyatt et al. 2013, Castor et al. 2013) and 2017 (Wyatt et al. 2017).

    1.b) Page 6: The authors should cite Wyatt et al. 2013, alongside Castor et al. 2013 and Garner et al. 2013 since these 3 articles were published at similar times. They may also want to acknowledge previous work from the Hickson and Rosselli labs showing that XPF-ERCC1 and MUS81-EME1 are recruited to fragile sites in mitosis.

    1. To improve broad readability, the authors should remove the following abbreviations: Aph and WT.

    2. In several figures, the authors show that a given treatment causes a very small change in the number of foci observed per mitotic cell. Although the values may be statistically different, it is important that they discuss the biological significance of these small effects - for example, I am not convinced that a difference of 2-3 foci per cell is sufficient to induce a robust cellular response.

    3. The methods could be expanded to ensure reproducibility, particularly with respect to the drug treatments (e.g., timing, washes, etc.).

    Significance

    This is a timely and exciting study that provides us with some new molecular insights into mitotic DNA repair. It builds on previous studies that identified the CIP2A-TOPBP1 complex as a molecular tether that connects broken DNA ends that get transmitted from interphase into mitosis (PMID: 30898438, 35842428, 35842428). The results are also largely complementary with those of Martin et al. (BioRxiv preprint at https://doi.org/10.1101/2024.11.12.621593) and de Haan et al. (BioRxiv preprint at https://www.biorxiv.org/content/10.1101/2025.04.03.647079v1).

  5. Version published to 10.1101/2025.09.19.677284 on bioRxiv