The RepID–CRL4 ubiquitin ligase complex regulates metaphase to anaphase transition via BUB3 degradation

The spindle assembly checkpoint (SAC) prevents premature chromosome segregation by inactivating the anaphase promoting complex/cyclosome (APC/C) until all chromosomes are properly attached to mitotic spindles. Here we identify a role for Cullin–RING ubiquitin ligase complex 4 (CRL4), known for modulating DNA replication, as a crucial mitotic regulator that triggers the termination of the SAC and enables chromosome segregation. CRL4 is recruited to chromatin by the replication origin binding protein RepID/DCAF14/PHIP. During mitosis, CRL4 dissociates from RepID and replaces it with RB Binding Protein 7 (RBBP7), which ubiquitinates the SAC mediator BUB3 to enable mitotic exit. During interphase, BUB3 is protected from CRL4-mediated degradation by associating with promyelocytic leukemia (PML) nuclear bodies, ensuring its availability upon mitotic onset. Deficiencies in RepID, CRL4 or RBBP7 delay mitotic exit, increase genomic instability and enhance sensitivity to paclitaxel, a microtubule stabilizer and anti-tumor drug.

The authors supported each individual conclusion with multiple lines of evidence, and the quality of data was generally high. A relatively complete story could be seen from several key pieces of discoveries presented here. However, obvious gaps exist for the story, for which at least some discussions are warranted. It will also be helpful if they integrate the prior information of different ubiquitin ligases in mitosis progression into their discussions. Additionally, some data deviations have to be addressed.

Point by point response to reviewers' comments:
We thank all the reviewers for their thoughtful evaluation of our original submission and for their helpful suggestions. We have revised the manuscript based on the reviewers' comments. We believe that reviewers' suggestions have significantly improved the paper, and we appreciate the reviewers' time and help. Below is our detailed response to the reviewers' comments.

Reviewer #1 (Remarks to the Author):
In this manuscript, Jang et al first made an observation that RepID depleted cells exhibited delayed mitotic exit and then followed up with mechanistic studies. RepID is an adaptor protein that recruits  to chromatin prior to DNA replication. Their initial observation suggested that RepID/CRL4 also plays a role in mitosis progression. The authors found that RepID interacts with BUB3, a spindle assembly checkpoint protein, particularly during G2/M phase. The authors showed that BUB3 protein level dropped as HCT116 cells exit from mitosis, due to CRL4mediated ubiquitylation. RepID is required for maintaining CRL4 at condensed chromosomes during mitosis but does not affect BUB3 localization. Surprisingly, RepID was found to dissociate from CRL4 before BUB3 degradation, suggesting another substrate-binding adaptor protein is responsible for BUB3 degradation. The authors identified RBBP7 as the adaptor protein, and proposed CRL4 switched adaptor proteins at the metaphase-anaphase transition and CRL4-RBBP7 mediates BUB3 degradation for mitotic exit. According to the authors, interphase BUB3 was spared of degradation partially due to sequestration in the PML nuclear bodies. They also showed that RepID-CRL4-RBBP7 depletion sensitized cells to microtubule stabilizing drug paclitaxel.
The authors supported each individual conclusion with multiple lines of evidence, and the quality of data was generally high. A relatively complete story could be seen from several key pieces of discoveries presented here. However, obvious gaps exist for the story, for which at least some discussions are warranted. It will also be helpful if they integrate the prior information of different ubiquitin ligases in mitosis progression into their discussions. Additionally, some data deviations have to be addressed.
Response: We thank the reviewer for this evaluation of the manuscript and its significance, and for the assessment of the quality of the data. We have addressed all the gaps identified by the reviewer, either by including additional data or by expanding the discussion as suggested. Detailed responses to all the points raised by the reviewer are presented below. and BUB3 is S-phase specific and that it occurs only in PML-depleted cells. As the reviewer has suggested, beyond corroborating the hypothesis that the CDT2-BUB3 interaction is prevented by PML, BUB3 accumulation in G2 might, therefore, also reflect the absence of CDT2-mediated degradation of BUB3. Eventually, the accumulation of undegraded BUB3 molecules builds a pool of BUB3 molecules that can potentially mediate the SAC during mitosis. This point is discussed in the revision (page 14, last paragraph). Fig 4g summarizes Fig 4C) will be informative.

Comment 1 (cont): -Fig 7 and
Response: Thank you for this suggestion, we agree that it is important to characterize the association of CUL4 with chromatin directly. In the revised version, we have analyzed the localization of CUL4A during mitosis as suggested using immunofluorescence and super resolution microscopy (new Supplementary Fig. 3f, also shown to the left). These images confirm that CUL4A localizes to mitotic chromosomes only in RepID proficient cells. We agree with the reviewer's interpretation that its recruitment, but not retention, on metaphase chromosomes requires RepID. We conclude that the association of CUL4 with chromatin (as a consequence of recruitment by RepID) is required for degradation of BUB3, although the catalytic DCAF mediating this degradation is RBBP7. This specific point is discussed in the revision (page 13, last paragraph). Response: Thank you, we agree that more information about the localization of RBBP7 and BUB3 is needed. In the revision we have replaced Fig. 4d with a new imaging study in which we performed triple staining for RBBP7, BUB3, and CREST during metaphase (new Fig. 4d), and added Supplemental Fig. 3e with images of RBBP7 and tubulin to test for spindle-like localization of RBBP7 (both images also pasted below). This co-staining study suggests that RBBP7 colocalizes with the mitotic spindle and associates with kinetochores. Consistent with the role of CRL4 RBBP7 in the degradation of BUB3, the association between RBBP7 and BUB3 during metaphase was stronger in RepID-deficient cells, which do not recruit CRL4 to mitotic chromatin (see above), than in RepID-proficient cells, in which CRL4 is recruited to chromatin and facilitates BUB3 degradation. To further probe into these interactions, we also tested for RBBP7, BUB3, and CREST localization in mitotic cells after exposure to a p97 inhibitor, which prevented the degradation of BUB3, and observed an increased association between BUB3 and RBBP7. This observation is similar to what we observed in the absence of RepID, when CRL4 was depleted and BUB3 degradation was inhibited. Response: Yes, thank you, we agree that this is a key point and it is important to emphasize it. Our hypothesis, based on our observation that RepID is required to recruit CRL4 is recruited to chromatin (PMID: 30018425, Ref #14), suggests that RepID-deficient cells show lower levels of BUB3 ubiquitylation because of CRL4's absence from chromatin in those cells. As the reviewer has suggested, we tested this hypothesis directly in the revision. The revised submission includes new co-immunoprecipitation data measuring interactions between RBBP7 and CRL4 in chromatin fractions following release from nocodazole (new Supplementary Fig. 3b, also shown below). As expected, CRL4 chromatin-bound levels (input) were lower in RepID-KO cells, but during the first 30 min after release from a nocodozole block, both KO and WT cells exhibited a similar fraction of RBBP7-bound BUB3 and CRL4 (CUL4A/DDB1). We have also tested BUB3 ubiquitination with purified proteins to assess the catalytic role of both RBBP7 and RepID (Supplemental Fig. 3i,  below). The results suggest that RBBP7 is the catalytic DCAF, and the absence of BUB3

Comment 1 (cont): -Fig4c and
A super-resolution microcopy analysis using RBBP7, BUB3 and CREST antibodies with DAPI staining in RepID WT and KO cells with and without CB5083, a p97 inhibitor.
A super-resolution microcopy analysis using anti-RBBP7 and tubulin antibodies with DAPI staining in RepID WT and KO cells.

ubiquitylation in RepID deficient cells reflects the fact that CRL4 is not recruited to chromatin in those cells. RBBP7 binds CRL4 after RepID's dissociation, and plays a catalytic role in BUB3
ubiquitylation, supporting the adaptor handover model. Response: Thank you for pointing out this apparent discrepancy, we agree that this point should indeed be discussed. First, the discrepancy observed between the two modes of cell synchronization can be explained by the fact that the use of nocodazole can introduce delay in G2/M release. The live imaging analysis presented in Fig. 1d is following cells' release from CDK1 inhibitor while the FACS in Fig. 1b Fig. 1a, also shown to the left).

Comment
Supplementary Fig. 3: (b) HCT116 RepID WT and KO cells transfected with FLAG-tagged RBBP7 were synchronized by nocodazole and released into fresh medium. Chromatin fractions were immunoprecipitated by FLAG antibody (detecting RBBP7) and analyzed by immunoblotting with the indicated antibodies. Binding ratio was calculated by dividing input, followed by precipitated RBBP7 level. (i) In vitro ubiquitination assay using purified proteins, followed by immunoblotting with BUB3 antibody.
HCT116 RepID WT and KO cells were released in fresh media from the mitotic block and collected every 30 minutes, followed by flow cytometry analysis. (Fig. 5d)". Why?

Comment 2 (cont): -Fig 1 shows RepID KO cells released from nocodazole accumulated subG1 apoptotic cells. However, "growth assays showed that RepID WT and KO cells exhibit similar sensitivities to the microtubule polymerization inhibitor nocodazole
Response: Again we thank the reviewer for pointing this out, as we need to clarify that the extent of acute and chronic responses to nocodozole cannot be compared directly. Because acute exposure to nocodazole induces prolonged mitotic arrest that can induce apoptosis, we were avoiding a highly toxic dose for the colony formation assay and used a lower concentration of nocodazole for this assay (less than 30 nM vs. 100 nM when employing nocodozole to induce a complete cell cycle blockage). The acute apoptosis we have observed in RepID deficient cells using a high dose of nocodozole did not translate to a significant loss of viability in long-term assays using a lower dose. This issue is discussed in the revised paper (page 10, end of the first paragraph).
Comment: 3. It is worth mentioning that CUL4-RBBP7 was indicated in CENP-A loading after mitoic exit (e.g. Mouysset J et al., JCS, 2015). Whether this is related to BUB3 ubiquitination described here might be interesting to compare. Other late mitotic events regulated by non-APC/C mediated ubiquitylation: CUL3 complexes are required for aurora B localization (Sumara, I. et al, Dev Cell, 2007).

Minor points: Comment: 1. BUB3 is often not regarded as a key player for physical inhibition of the APC/C. What do the authors think about that?
Response: Thank you for this important comment, and indeed, we wish to highlight that our study emphasizes that BUB3 plays a more critical role in mitosis than originally thought. Our understanding is that the activity of the APC/C is regulated throughout the cell cycle by several mechanisms, one involving BUB3, which would interact with CDC20 and inhibit the formation of APC/C-CDC20 . Our data are in line with observations in mice carrying a disruption of the Bub3 gene, which leads to embryonic lethality of lagging chromosomes, micronuclei and chromatin bridging among others (PMID: 10995385, #63 in the revision), and in observations suggesting that patients carrying the BUB3 mutations show mosaic aneuploidy (PMID: 23747338; #64 in the revision). These observations are discussed in the revised submitted paper (page 15, first paragraph).

Comment: 2. BUB3 level is constant throughout the cell cycle in HeLa cells. One major difference is that the HCT116 cells used in this study have functional p53. Does p53 status affect BUB3 ubiquitination?
Response: We thank the reviewer for this very interesting suggestion. We tested the hypothesis that p53 status affects BUB3 degradation directly by generating RepID KO in two additional cell lines with dysfunctional p53: H1299 (a Non-Small Cell Lung Cancer, p53-/-) and DMS114 (Small Cell Lung Cancer, mutant p53). The new data show (new Supplementary Fig. 2b-f, also shown below) that all RepID KO cell lines exhibit a mitotic delay after nocodazole release and suggest that the delayed mitotic and BUB3 ubiquitination are independent of the status of p53.

Response:
Yes, the IFs in Fig. 6b, d, e were performed using a same protocol and with identical thresholds during image captures to prevent signal saturation. Fig. 4d-f. left).

Reviewer #2 (Remarks to the Author):
Comment: In this manuscript, the authors identified a novel role of RepID-CRL4, known for modulating DNA replication, in the regulation of the degradation of BUB3, which triggers the termination of SAC and enables chromosome segregation. They investigated the mechanism and found that in mitosis, RepID is disassociated from CRL4 and replaced by RBBP7. RBBP7 ubiquitinates BUB3 and triggers BUB3 degradation, leading to mitotic exit. This study identified a previously unrecognized role of RepID in mitotic checkpoint and demonstrated an interesting switch of DCAFs to regulate the progression of cell 7 cycle. The manuscript is well written, and the data are clearly presented. However, several points still need to be addressed.
Response: We thank the reviewer for this evaluation of the manuscript, as well as for the important comments and suggestions. Below we address the specific points raised by the reviewer. Fig. 2d-f, shown to the left). These experiments demonstrate that reconstituted RepID (FL) recovered mitotic exit and BUB3 ubiquitination in three cell lines, suggesting that the mitotic problems observed in RepID-deficient cells are caused by RepID deficiency. For these analyses, we have generated RepID KO in two additional cell lines: H1299 (Non-Small Cell Lung Cancer) and DMS114 (Small Cell Lung Cancer), both showing a mitotic delay that can be prevented by reconstitution with RepID. Fig.2B, is the WD40 domain of RepID sufficient to mediate the interaction with BUB3? Does BUB3 use the same domain for the interactions with RepID and RBBP7? This could be a possible mechanism for DCAF switching. The mechanism of how switching of RepID to RBBP7 is induced in mitosis is not addressed in this report.

Comment: 2. For the interaction of RepID with BUB3 shown in
Response: We thank the reviewer for this comment, which raises an important and relevant point. The WD domain by itself is not sufficient to mediate interaction, but we performed additional experiments using a RepID construct lacking the RepID WD40 domain (F2-5 construct). Our results suggest that depletion of the WD40 domain of RepID was sufficient to prevent the interaction with BUB3. We added the new data in the revised Fig. 2b (right).
Cell cycle analysis after release from mitotic block (d-e), and in vivo ubiquitination analysis (f) in HCT116, H1299 and DMS114 cells with WT, KO and reconstituted RepID.
Soluble nuclear and chromatin-bound fractions from U2OS cells expressing the indicated FLAG-RepID mutants were immunoprecipitated with FLAG antibodies and analyzed by immunoblotting. RepID and RBBP7 in vivo. In vitro ubiquitination assay would be needed to show that BUB3 is truly the substrate of RBBP7-containing CRL4.

Response:
We thank to reviewer for this suggestion. As suggested, we performed in vitro ubiquitination assays with immunopurified flagged-CUL4A, CUL4B, DDB1, RepID, BUB3 and RBBP7. The revised manuscript reports these assays and shows that RBBP7 (but not RepID) is a catalytic DCAF required to ubiquitinate BUB3 (new Supplementary  Fig. 3i, left).

Comment: 4. In Fig4b, the interaction of RBBP7 with CUL4 but not with BUB3 is shown. Is the interaction between RBBP7 and BUB3 increased in mitosis? In the absence of RepID, is the interaction between RBBP7 and BUB3 affected when cells enter the mitosis?
Response: We thank the reviewer for this comment, which raises an interesting point. In the revision, we include experiments examining the interactions between RBBP7 and CRL4 and between RBBP7 and BUB3. Interactions were quantified in nocodazole-released HCT116 WT and RepID KO cells. The data are presented in the new Supplementary Fig. 3b  (right). Indeed, interactions between RBBP7 and CUL4A or between RBBP7 and BUB3 increased in metaphase (30 min post release; at later time points, BUB3 is degraded in RepID WT cells so interactions cannot be detected).

Reviewer #3 (Remarks to the Author):
Comment: In this manuscript Aladjem and colleagues investigate BUB3 degradation during the cell cycle. BUB3 is a critical component of the spindle assembly checkpoint (SAC). Degradation of BUB3 would silence the SAC and promote anaphase onset. The authors investigate the role of the CRL4 E3 ubiquitin ligase in BUB3 degradation. Their major conclusions are that CRL4 is recruited to chromatin by RepID, that during mitosis CRL4 dissociates from RepID and binds to RBBP7 which is the complex of CRL4 that ubiquitinates BUB3. BUB3 is protected from CRL4-mediated ubiquitination during interphase by association with PML. The findings that RepID deficient cells were delayed exiting mitosis and entering G1, and showed compromised geminin, cyclin B and securin degradation, is the starting point for this study with the proposal that RepID 'passes over' CRL4 to RBBP7 in mitosis. This is linked to increased levels of BubR1 associated with the APC/C in RepID deficient cells. Bub3 levels decline in RepID proficient but not RepID deficient cells after release from exposure to nocodazole. This is UPSdependent. CUL4A-B and RBBP7 depletion also leads to reduced BUB3 degradation.
In vitro ubiquitination assay using purified proteins, followed by immunoblotting with BUB3 antibody.
RepID WT and KO cells transfected with FLAG-tagged RBBP7 were synchronized by nocodazole and released into fresh medium. Chromatin fractions were immunoprecipitated by FLAG antibody (detecting RBBP7) and analyzed by immunoblotting with the indicated antibodies.
Altogether the authors perform a wide range of cell-based experiments. However overall, their model is not completely convincing without further experiments to validate the proposal that CRL4-RBBP7 is the E3 ligase directly responsible for ubiquitinating BUB3.
Comment: 6. The major concern is the absence of evidence that  For this model to be convincing the authors should show that reconstituted . In addition since the authors claim that CRL4-RepID does not ubiquitinate BUB3, even though RepID interacts with BUB3 (whether this is direct or indirect was not shown), the activity of CRL4-RepID towards BUB3 should also be tested.
Response: We thank the reviewer for raising these important points. To assess if RBBP7 directly ubiquitinates BUB3, we performed additional in vitro experiments (new Supplementary Fig. 3i, left). Our data show that the reconstituted CRL4-RBBP7, but not CRL4-RepID, can ubiquitinate BUB3. These data suggest that the presence of RBBP7 in the reconstituted complex being critical for BUB3 ubiquitination. RepID, which is required to recruit CUL4 to chromatin, was required for BUB3 ubiquitination in cells but was dispensable in vitro.
Comment: 8. The authors appear to assume that is kinetochore-associated BUB3 that is degraded by CRL4-RBBP7. However have they considered BUB3 in the complex with the mitotic checkpoint complex?
Response: Yes, thank you, we agree with the reviewer that RepID-induced BUB3 degradation could occur either in the context of the kinetochore or within the mitotic checkpoint complex. We have modified the discussion (page 14, second paragraph) and our model (Fig. 7) to emphasize this point.
Comment: 9. Lines 67-69. The mechanism of SAC silencing is incorrect. The APC/C is activated because the MCC dissociates from the APC/C-Cdc20 complex due to Cdc20 (of the MCC) of BubR1 ubiquitination.
Response: We thank the reviewer for pointing out this issue. We changed the text according to the reviewer's suggestion.
In conclusion, we would like to thank all the reviewers again for their insightful comments and suggestions. We believe that these suggestions helped improve the manuscript and helped strengthen the conclusions, and we are grateful for the opportunity to submit this revision for your consideration.
Supplementary Fig. 3i: In vitro ubiquitination assay using purified proteins, followed by immunoblotting with BUB3 antibody.