The p97 segregase cofactor Ubxn7 facilitates replisome disassembly during S-phase
2022; Elsevier BV; Volume: 298; Issue: 8 Linguagem: Inglês
10.1016/j.jbc.2022.102234
ISSN1083-351X
AutoresZeynep Tarcan, Divyasree Poovathumkadavil, Aggeliki Skagia, Agnieszka Gambus,
Tópico(s)Microtubule and mitosis dynamics
ResumoComplex cellular processes are driven by the regulated assembly and disassembly of large multiprotein complexes. While we are beginning to understand the molecular mechanism for assembly of the eukaryotic DNA replication machinery (replisome), we still know relatively little about the regulation of its disassembly at replication termination. Recently, the first elements of this process have emerged, revealing that the replicative helicase, at the heart of the replisome, is polyubiquitylated prior to unloading and that this unloading requires p97 segregase activity. Two different E3 ubiquitin ligases have now been shown to ubiquitylate the helicase under different conditions: Cul2Lrr1 and TRAIP. Here, using Xenopus laevis egg extract cell-free system and biochemical approaches, we have found two p97 cofactors, Ubxn7 and Faf1, which can interact with p97 during replisome disassembly during S-phase. We show only Ubxn7, however, facilitates efficient replisome disassembly. Ubxn7 delivers this role through its interaction via independent domains with both Cul2Lrr1 and p97 to allow coupling between Mcm7 ubiquitylation and its removal from chromatin. Our data therefore characterize Ubxn7 as the first substrate-specific p97 cofactor regulating replisome disassembly in vertebrates and a rationale for the efficacy of the Cul2Lrr1 replisome unloading pathway in unperturbed S-phase. Complex cellular processes are driven by the regulated assembly and disassembly of large multiprotein complexes. While we are beginning to understand the molecular mechanism for assembly of the eukaryotic DNA replication machinery (replisome), we still know relatively little about the regulation of its disassembly at replication termination. Recently, the first elements of this process have emerged, revealing that the replicative helicase, at the heart of the replisome, is polyubiquitylated prior to unloading and that this unloading requires p97 segregase activity. Two different E3 ubiquitin ligases have now been shown to ubiquitylate the helicase under different conditions: Cul2Lrr1 and TRAIP. Here, using Xenopus laevis egg extract cell-free system and biochemical approaches, we have found two p97 cofactors, Ubxn7 and Faf1, which can interact with p97 during replisome disassembly during S-phase. We show only Ubxn7, however, facilitates efficient replisome disassembly. Ubxn7 delivers this role through its interaction via independent domains with both Cul2Lrr1 and p97 to allow coupling between Mcm7 ubiquitylation and its removal from chromatin. Our data therefore characterize Ubxn7 as the first substrate-specific p97 cofactor regulating replisome disassembly in vertebrates and a rationale for the efficacy of the Cul2Lrr1 replisome unloading pathway in unperturbed S-phase. DNA replication is one of the most fundamental processes in life and its faultless execution is essential for normal cell fate. Until recently, the final stage of eukaryotic DNA replication, the termination stage, was mostly unexplored. DNA replication initiates from thousands of replication origins, the positions within the genome where replicative helicases become activated and start unwinding DNA. These then move in opposite directions away from each other, creating two DNA replication forks. The replicative helicase is composed of Cdc45, Mcm2-7 hexamer, and GINS complex (CMG complex) (1Moyer S.E. Lewis P.W. Botchan M.R. Isolation of the Cdc45/Mcm2-7/GINS (CMG) complex, a candidate for the eukaryotic DNA replication fork helicase.Proc. Natl. Acad. Sci. U. S. A. 2006; 103: 10236-10241Crossref PubMed Scopus (535) Google Scholar) and is positioned at the tip of replication forks forming a platform for replisome assembly (2Gambus A. Jones R.C. Sanchez-Diaz A. Kanemaki M. van Deursen F. Edmondson R.D. et al.GINS maintains association of Cdc45 with MCM in replisome progression complexes at eukaryotic DNA replication forks.Nat. Cell Biol. 2006; 8: 358-366Crossref PubMed Scopus (582) Google Scholar). Once established, the replication forks replicate chromatin until they encounter forks coming in opposite directions from neighboring origins. At this point the termination of replication forks takes place, with removal of the replisome from fully duplicated DNA being its final stage (3Dewar J.M. Budzowska M. Walter J.C. The mechanism of DNA replication termination in vertebrates.Nature. 2015; 525: 345-350Crossref PubMed Google Scholar). In higher eukaryotes (Xenopus laevis egg extract, Caenorhabditis elegans embryos, mouse embryonic fibroblasts, and human cells), replisome removal in S-phase is driven by the Cul2Lrr1 ubiquitin ligase, which ubiquitylates Mcm7 within the terminated CMG helicase complex with lysine 48 (K48)-linked ubiquitin chains (4Sonneville R. Moreno S.P. Knebel A. Johnson C. Hastie C.J. Gartner A. et al.CUL-2LRR-1 and UBXN-3 drive replisome disassembly during DNA replication termination and mitosis.Nat. Cell Biol. 2017; 19: 468-479Crossref PubMed Scopus (54) Google Scholar). The modified CMG is then recognized by the p97 segregase and removed from chromatin allowing for disassembly of the whole replisome built around the helicase (5Moreno S.P. Bailey R. Campion N. Herron S. Gambus A. Polyubiquitylation drives replisome disassembly at the termination of DNA replication.Science. 2014; 346: 477-481Crossref PubMed Scopus (123) Google Scholar). Any helicase complexes that fail to be unloaded in S-phase are alternatively unloaded in mitosis. Disassembly of these complexes in mitosis also depends on p97 segregase function, but this time requires the E3 ubiquitin ligase TRAIP (6Priego Moreno S. Jones R.M. Poovathumkadavil D. Scaramuzza S. Gambus A. Mitotic replisome disassembly depends on TRAIP ubiquitin ligase activity.Life Sci. Alli. 2019; 2e201900390PubMed Google Scholar). Consequently, disassembly in mitosis is driven by alternative species of ubiquitin chains that decorate Mcm7, namely K6- and K63-linked ubiquitin chains. TRAIP ubiquitin ligase can act also during S-phase: it interacts with the replisome and either ubiquitylates CMGs that converge at interstrand crosslinks or ubiquitylates a protein crosslinked to DNA (DPC) that blocks progression of the replication fork (7Wu R.A. Semlow D.R. Kamimae-Lanning A.N. Kochenova O.V. Chistol G. Hodskinson M.R. et al.TRAIP is a master regulator of DNA interstrand crosslink repair.Nature. 2019; 567: 267-272Crossref PubMed Scopus (77) Google Scholar, 8Larsen N.B. Gao A.O. Sparks J.L. Gallina I. Wu R.A. Mann M. et al.Replication-coupled DNA-protein crosslink repair by SPRTN and the proteasome in Xenopus egg extracts.Mol. Cell. 2019; 73: 574-588.e577Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar). p97 segregase (also referred to as VCP, Cdc48, CDC-48, and Ter94) is a hexameric AAA+ ATPase family member that uses the free energy of ATP binding and hydrolysis to extract ubiquitylated protein targets from stable protein complexes, chromatin, or membranes. As a result, p97 is essential for protein homeostasis in the cell and the dynamic behavior of a multitude of multiprotein assemblies (9Meyer H. Bug M. Bremer S. Emerging functions of the VCP/p97 AAA-ATPase in the ubiquitin system.Nat. Cell Biol. 2012; 14: 117-123Crossref PubMed Scopus (567) Google Scholar). The substrate specificity of p97 recognition is believed to be achieved by a number of regulatory cofactors (reviewed in (10Hanzelmann P. Schindelin H. The interplay of cofactor interactions and post-translational modifications in the regulation of the AAA+ ATPase p97.Front. Mol. Biosci. 2017; 4: 21Crossref PubMed Google Scholar)). In C. elegans embryos, the CDC-48 cofactors UFD-1/NPL-4 and UBXN-3 (Faf1 in higher eukaryotes) were shown to be required for replisome removal from chromatin in both S-phase and in mitosis (4Sonneville R. Moreno S.P. Knebel A. Johnson C. Hastie C.J. Gartner A. et al.CUL-2LRR-1 and UBXN-3 drive replisome disassembly during DNA replication termination and mitosis.Nat. Cell Biol. 2017; 19: 468-479Crossref PubMed Scopus (54) Google Scholar, 11Xia Y. Fujisawa R. Deegan T.D. Sonneville R. Labib K.P.M. TIMELESS-TIPIN and UBXN-3 promote replisome disassembly during DNA replication termination in Caenorhabditis elegans.EMBO J. 2021; 40e108053Crossref Scopus (6) Google Scholar). UFD-1/NPL-4 form a heterodimer, essential for most chromatin-related roles of p97, while UBXN-3 provides higher substrate specificity. Interestingly, Ufd1/Npl4 were also shown to interact with p97 and the replisome during replication termination in Xenopus egg extract (4Sonneville R. Moreno S.P. Knebel A. Johnson C. Hastie C.J. Gartner A. et al.CUL-2LRR-1 and UBXN-3 drive replisome disassembly during DNA replication termination and mitosis.Nat. Cell Biol. 2017; 19: 468-479Crossref PubMed Scopus (54) Google Scholar). However, while Ufd1/Npl4 are evolutionarily conserved and well characterized, the number and variability of the minor, substrate specific, cofactors of p97 grows through evolution, reflecting the increasing complexity of p97 regulation. So far, roughly three times more cofactors have been identified in mammals than in C. elegans (4Sonneville R. Moreno S.P. Knebel A. Johnson C. Hastie C.J. Gartner A. et al.CUL-2LRR-1 and UBXN-3 drive replisome disassembly during DNA replication termination and mitosis.Nat. Cell Biol. 2017; 19: 468-479Crossref PubMed Scopus (54) Google Scholar, 12Buchberger A. Schindelin H. Hanzelmann P. Control of p97 function by cofactor binding.FEBS Lett. 2015; 589: 2578-2589Crossref PubMed Scopus (123) Google Scholar). Importantly, additional cofactors of p97, which provide substrate specificity towards the terminated replisomes, are as yet to be determined in vertebrates. Here, we sought to identify p97 cofactors that are facilitating replisome disassembly during S-phase. While we were able to identify two new cofactors for this process, Ubxn7 and Faf1, our findings revealed that the Ubxn7 cofactor specifically is crucial for efficient and fast disassembly of replisomes, as it creates bridges between the essential factors of this process: Cul2Lrr1, ubiquitylated Mcm7, and the p97 complexes. Using the X. laevis egg extract model system, we have previously shown that the unloading of terminated replicative helicases during S-phase depends on formation of K48-linked ubiquitin chains on the Mcm7 subunit of the CMG helicase by Cul2Lrr1 ubiquitin ligase (5Moreno S.P. Bailey R. Campion N. Herron S. Gambus A. Polyubiquitylation drives replisome disassembly at the termination of DNA replication.Science. 2014; 346: 477-481Crossref PubMed Scopus (123) Google Scholar). Such modified Mcm7 is in turn recognized and unfolded by p97. We therefore first confirmed that p97 interacts with replicating chromatin in X. laevis egg extract with kinetics similar to replication fork components (Fig. 1A). We found that p97 is a highly abundant protein in Xenopus egg extract (13Heubes S. Stemmann O. The AAA-ATPase p97-Ufd1-Npl4 is required for ERAD but not for spindle disassembly in Xenopus egg extracts.J. Cell Sci. 2007; 120: 1325-1329Crossref PubMed Scopus (21) Google Scholar), and only a small proportion of it interacts with chromatin during DNA replication, with the interaction peaking during the exponential stage of replication, when large numbers of replication forks are moving through the chromatin and converging during termination (Fig. 1A). To determine the portfolio of p97 cofactors that interact with p97 during DNA replication termination in egg extract and which may direct p97 to the terminated replisome, we aimed to immunoprecipitate (IP) p97 from a chromatin fraction and analyze its interactors. Firstly, we blocked replisome disassembly by inhibiting p97 activity with NMS-873 (p97i), which is a highly specific allosteric inhibitor of p97 ATPase activity binding to p97 D2 domain (14Her N.G. Toth J.I. Ma C.T. Wei Y. Motamedchaboki K. Sergienko E. et al.p97 composition changes caused by allosteric inhibition are suppressed by an on-target mechanism that increases the enzyme's ATPase activity.Cell Chem. Biol. 2016; 23: 517-528Abstract Full Text Full Text PDF PubMed Google Scholar). We have shown previously that inhibition of p97 ATPase activity stops replisome unloading from chromatin (4Sonneville R. Moreno S.P. Knebel A. Johnson C. Hastie C.J. Gartner A. et al.CUL-2LRR-1 and UBXN-3 drive replisome disassembly during DNA replication termination and mitosis.Nat. Cell Biol. 2017; 19: 468-479Crossref PubMed Scopus (54) Google Scholar). Critically, this treatment does not stop p97 from interacting with substrates or chromatin (Fig. S1A) and should stabilize p97–substrate complexes on chromatin as the substrates cannot be processed. Subsequently, we isolated chromatin with accumulated terminated replisomes, immunoprecipitated p97 and analyzed interacting factors by mass spectrometry (MS). Such analysis revealed numerous components of replication machinery interacting with p97, including those which reside in the replisome built around the CMG helicase but also other DNA replication and DNA damage repair factors (Table S1). In order to focus our analysis on potential p97 cofactors, which direct p97 specifically to the terminated replisomes, results from the p97 interactome were compared with those of an Mcm3 IP, which was performed in conditions blocking replisome disassembly. Briefly, the extract was replicated in the presence of a dominant negative ATPase-dead mutant of p97, described previously, and Mcm3 was immunoprecipitated to isolate terminating replisomes (4Sonneville R. Moreno S.P. Knebel A. Johnson C. Hastie C.J. Gartner A. et al.CUL-2LRR-1 and UBXN-3 drive replisome disassembly during DNA replication termination and mitosis.Nat. Cell Biol. 2017; 19: 468-479Crossref PubMed Scopus (54) Google Scholar, 13Heubes S. Stemmann O. The AAA-ATPase p97-Ufd1-Npl4 is required for ERAD but not for spindle disassembly in Xenopus egg extracts.J. Cell Sci. 2007; 120: 1325-1329Crossref PubMed Scopus (21) Google Scholar). This comparison allowed us to determine which of the p97 cofactors identified in the p97 immunoprecipitation are also interacting with the terminated replisomes, as we appreciate that p97 does have other substrates on replicating chromatin (Fig. 1B). In doing this, we identified both major cofactors Ufd1 and Npl4, which are known to facilitate chromatin functions of p97 segregase and were shown previously by us and others to act in replisome disassembly during S-phase (4Sonneville R. Moreno S.P. Knebel A. Johnson C. Hastie C.J. Gartner A. et al.CUL-2LRR-1 and UBXN-3 drive replisome disassembly during DNA replication termination and mitosis.Nat. Cell Biol. 2017; 19: 468-479Crossref PubMed Scopus (54) Google Scholar, 15Dewar J.M. Low E. Mann M. Raschle M. Walter J.C. CRL2Lrr1 promotes unloading of the vertebrate replisome from chromatin during replication termination.Genes Dev. 2017; 31: 275-290Crossref PubMed Scopus (62) Google Scholar). Interestingly, only two minor cofactors were identified to interact with both the replisome and p97: Faf1 and Ubxn7 (human UBXD7). Both of these factors have been shown previously to interact preferentially with p97 when in complex with Ufd1/Npl4 (16Hanzelmann P. Buchberger A. Schindelin H. Hierarchical binding of cofactors to the AAA ATPase p97.Structure. 2011; 19: 833-843Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar). To support this finding, we first confirmed by p97 IP and Western blotting that p97 can indeed interact with Ubxn7 and Faf1 on S-phase chromatin when replisome disassembly is blocked (Fig. 1C). The finding of Faf1 was somewhat unsurprising as it is already known to play a role in maintaining replication fork stability in C. elegans and human cell lines (17Franz A. Pirson P.A. Pilger D. Halder S. Achuthankutty D. Kashkar H. et al.Chromatin-associated degradation is defined by UBXN-3/FAF1 to safeguard DNA replication fork progression.Nat. Commun. 2016; 710612Google Scholar), and indeed, the C. elegans homolog of Faf1 (UBXN-3) is essential for replisome unloading in S-phase and in mitotic prophase (4Sonneville R. Moreno S.P. Knebel A. Johnson C. Hastie C.J. Gartner A. et al.CUL-2LRR-1 and UBXN-3 drive replisome disassembly during DNA replication termination and mitosis.Nat. Cell Biol. 2017; 19: 468-479Crossref PubMed Scopus (54) Google Scholar, 11Xia Y. Fujisawa R. Deegan T.D. Sonneville R. Labib K.P.M. TIMELESS-TIPIN and UBXN-3 promote replisome disassembly during DNA replication termination in Caenorhabditis elegans.EMBO J. 2021; 40e108053Crossref Scopus (6) Google Scholar). In contrast, this is the first time that Ubxn7 has been implicated in the process of replisome unloading. What is already known about Ubxn7, is that inhibition of the human homolog UBXD7 in human cells leads to hyperaccumulation of DNA damage sensors after UV damage (18Puumalainen M.R. Lessel D. Ruthemann P. Kaczmarek N. Bachmann K. Ramadan K. et al.Chromatin retention of DNA damage sensors DDB2 and XPC through loss of p97 segregase causes genotoxicity.Nat. Commun. 2014; 5: 3695Crossref PubMed Scopus (80) Google Scholar), although it is best known as a regulator of degradation of the hypoxia inducible factor Hif1α (19Alexandru G. Graumann J. Smith G.T. Kolawa N.J. Fang R. Deshaies R.J. UBXD7 binds multiple ubiquitin ligases and implicates p97 in HIF1alpha turnover.Cell. 2008; 134: 804-816Abstract Full Text Full Text PDF PubMed Scopus (243) Google Scholar). Like Faf1, Ubxn7 belongs to the ubiquitin-associated (UBA)–ubiquitin regulatory X (UBX) family of p97 cofactors. This means that it interacts with ubiquitylated proteins via its UBA domain and with p97 via its UBX domain. It also contains a UAS domain of unknown function and an ubiquitin-interacting motif (UIM) (Fig. S2A) (19Alexandru G. Graumann J. Smith G.T. Kolawa N.J. Fang R. Deshaies R.J. UBXD7 binds multiple ubiquitin ligases and implicates p97 in HIF1alpha turnover.Cell. 2008; 134: 804-816Abstract Full Text Full Text PDF PubMed Scopus (243) Google Scholar). Interestingly, while targeting Hif1α for degradation, UBXD7 simultaneously interacts with active (neddylated) cullin ligase Cul2VHL through its UIM domain and with p97 through its UBX domain (Fig. S2, A and B) (20Bandau S. Knebel A. Gage Z.O. Wood N.T. Alexandru G. UBXN7 docks on neddylated cullin complexes using its UIM motif and causes HIF1alpha accumulation.BMC Biol. 2012; 10: 36Crossref PubMed Scopus (46) Google Scholar). Given that the same factors appear to be involved in replisome disassembly, we hypothesized that the association of p97 with its cofactor Ubxn7 could provide a mechanism by which p97 targets Mcm7, ubiquitylated with K48-linked ubiquitin chains, synthesized by Cul2Lrr1. To investigate the importance of Faf1 and Ubxn7 for replisome disassembly during S-phase, we decided to immunodeplete each protein independently from the egg extract and determine the consequences for replisome unloading. To do this, we have raised antibodies against X. laevis Ubxn7 and Faf1 (Fig. S7). Both Faf1 and Ubxn7 could be efficiently immunodepleted from egg extracts to less than 5% of total protein (Figs. 2, A and B and S3B). Neither depletion inhibited the synthesis of nascent DNA in a number of independent immunodepletions (Fig. 2, A and B), suggesting that neither are essential for DNA synthesis completion in the egg extract. Importantly, neither Ubxn7 nor Faf1 immunodepletion affected the level of each other (Fig. S3A). We then followed proteins on chromatin during a replication reaction in IgG-depleted and Ubxn7- or Faf1-depleted extracts. Interestingly, while immunodepletion of Faf1 had a very minor effect on replisome unloading during S-phase (Fig. 2, D and E), immunodepletion of Ubxn7 reproducibly delayed unloading of replisomes (Cdc45, Psf2) in independently immunodepleted extracts (Figs. 2, C and E and S3C), suggesting that although Ubxn7 is not essential for replisome disassembly, it does regulate the efficiency of this process. Next, we wanted to understand how the depletion of Ubxn7 delays the unloading of the replisomes in the egg extract. Induction of replication stress, which affects progression of replication forks, can lead to transient accumulation of replisomes on chromatin. However, Ubxn7 depletion did not lead to the accumulation of DNA damage-associated replication stress, as determined by γH2AX immunoblotting (Fig. 3A), nor the phosphorylation of Chk1 (Fig. S3D). Taken together, this demonstrates that accumulation of replisomes on chromatin in Ubxn7-depleted extract is not due to the effects of replication stress. We next analyzed how Ubxn7 depletion affects the levels of Mcm7 ubiquitylation and chromatin-bound Cul2 and p97 (Fig. 3A). Interestingly, while the levels of p97 and Faf1 on chromatin were slightly increased by Ubxn7 depletion (Fig. 3A), Cul2 markedly accumulated on chromatin in its active neddylated form (Fig. 3A, neddylated Cul2 runs at a higher molecular size on the gel—compare size of unneddylated Cul2 in egg extract and neddylated on chromatin). Similarly, ubiquitylated Mcm7 also accumulated on chromatin, modified with long ubiquitin chains. We then decided to ensure that these phenotypes are actually caused by immunodepletion of Ubxn7 rather than an unidentified protein recognized by Ubxn7 antibody, through supplementing immunodepleted extract with recombinant Ubxn7. As shown in Figure 3B, addition of recombinant Ubxn7 could rescue the delayed unloading of CMG components (Cdc45 and Psf2), the accumulation of Cul2 on chromatin, and the accumulation of long-chain modified Mcm7 (Fig. 3B). These results suggest that Ubxn7 may be fine-tuning the process of replisome unloading, acting as a bridge between Cul2Lrr1, p97, and their shared substrate Mcm7. This bridging allows for fast and efficient processing of Mcm7 ubiquitylated with relatively short ubiquitin chains. Without Ubxn7, Cul2Lrr1 recognizes terminated CMG and ubiquitylates Mcm7, but this ubiquitylated Mcm7 is not recognized and processed quickly by p97 due to lack of Ubxn7. Cul2Lrr1 stays therefore associated with Mcm7 for a longer time (we can observe this as accumulation of active Cul2 on chromatin), resulting in synthesis of longer ubiquitin chains on Mcm7. Although this most likely enables eventual recruitment of p97, the process is less efficient. To determine how Ubxn7 is regulated during replication termination and replisome disassembly in the egg extract, we analyzed its pattern of chromatin binding during a replication reaction in egg extract. Throughout normal, unchallenged replication, Ubxn7 transiently interacts with chromatin with timing concomitant to that of replication fork presence and Cul2Lrr1 (Figs. 4, B and C and S4, A–C (dimethyl sulfoxide [DMSO] control)). Inhibition of replisome unloading with p97 ATPase inhibitor led to an accumulation of not only CMG and Cul2Lrr1 on chromatin as expected (4Sonneville R. Moreno S.P. Knebel A. Johnson C. Hastie C.J. Gartner A. et al.CUL-2LRR-1 and UBXN-3 drive replisome disassembly during DNA replication termination and mitosis.Nat. Cell Biol. 2017; 19: 468-479Crossref PubMed Scopus (54) Google Scholar) but also of Ubxn7 and p97 (Figs. 4A and S4A). We then analyzed the Ubxn7 chromatin-binding pattern upon inhibition of cullin ligases activity. CULi MLN4924 acts through inhibition of Nedd8 activating enzyme NAE and therefore inhibiting neddylation of substrates (21Soucy T.A. Smith P.G. Milhollen M.A. Berger A.J. Gavin J.M. Adhikari S. et al.An inhibitor of NEDD8-activating enzyme as a new approach to treat cancer.Nature. 2009; 458: 732-736Crossref PubMed Scopus (1292) Google Scholar). As members of the cullin family of ubiquitin ligases are the main substrate of neddylation in cells, MLN4924 is primarily inhibiting all cullin activity. We have shown previously that, in X. laevis egg extract, inhibition of cullin activity during DNA replication inhibits Mcm7 ubiquitylation and replisome unloading from chromatin (5Moreno S.P. Bailey R. Campion N. Herron S. Gambus A. Polyubiquitylation drives replisome disassembly at the termination of DNA replication.Science. 2014; 346: 477-481Crossref PubMed Scopus (123) Google Scholar). While inhibition of cullin activity led to accumulation on chromatin of CMG and Cul2Lrr1, as we reported previously (4Sonneville R. Moreno S.P. Knebel A. Johnson C. Hastie C.J. Gartner A. et al.CUL-2LRR-1 and UBXN-3 drive replisome disassembly during DNA replication termination and mitosis.Nat. Cell Biol. 2017; 19: 468-479Crossref PubMed Scopus (54) Google Scholar), levels of chromatin-bound p97 and Ubxn7 were reduced (Figs. 4B and S4B). This result suggests that the key determinants of p97 chromatin binding during DNA replication in the egg extract are either cullin-driven ubiquitylation of substrates and/or neddylation of cullins. Finally, we decided to inhibit the polyubiquitylation of all the potential substrates of p97 during DNA replication by supplementing the extract with a chain terminating mutant of ubiquitin: 6xHIS-UbiNoK. Interestingly, despite inhibition of polyubiquitylation by 6xHIS-UbiNoK, as observed through accumulation of di-monoubiquitylated forms of Mcm7 on chromatin, we can reproducibly observe a higher level of p97 and Ubxn7 binding to chromatin (Figs. 4C and S4C). Importantly, neither inhibition of cullins, nor global polyubiquitylation, affects the extracts' ability to replicate DNA (Fig. S4, B and C). Our findings therefore suggest that Ubxn7 behaves like a p97 cofactor and follows the same patterns of chromatin interaction. Moreover, p97 segregase is not directed to chromatin and terminated replisomes just simply through interaction with polyubiquitylated substrates, as inhibition of polyubiquitylation does not prevent p97 and Ubxn7 from binding to chromatin (Figs. 4C and S4C). It is likely therefore that treatment of replicating extract with the cullin neddylation inhibitor (MLN4924) (Figs. 4B and S4B) downregulates chromatin binding of p97 and Ubxn7 because neddylated cullins provide a binding platform for Ubxn7. To support this hypothesis further, we confirmed that Ubxn7 can interact with p97 not only on chromatin (Fig. 1C) but also in the egg extract (cytoplasm) (Fig. 4D). However, we could not detect an interaction between Ubxn7 and Cul2 in the egg extract (Fig. 4D), as Cul2 is present in the cytoplasmic extract in its inactive, unneddylated form. We went on to perform reciprocal immunoprecipitations of p97, Cul2, and Ubxn7 from chromatin where post-termination replisomes were accumulated due to inhibition of cullin neddylation (CULi) or p97 activity (p97i) (Fig. 4E). The level of inhibition achieved in the samples can be judged by the status of Cul2 in the input; while Cul2 is present in the extract mainly in its inactive, unneddylated form (Fig. 4E, lane 1), it accumulates on chromatin during termination upon inhibition of p97 activity mostly in its active, neddylated form (Fig. 4E, lane 2). CULi treatment, however, leads to Cul2 accumulation on chromatin in its inactive unneddylated form (Fig. 4E, lane 3). The dramatic absence of Ubxn7 on chromatin upon neddylation inhibition, despite the presence of p97 and Faf1 in this input (Fig. 4E, lane 3), clearly suggests that Ubxn7 binding to chromatin strongly depends on cullin neddylation. When present in the chromatin input, Ubxn7 could coimmunoprecipitate neddylated Cul2, Faf1, and a little of p97 (Fig. 4E, lane 10). p97 could interact with Ubxn7, Faf1, and neddylated Cul2, but this can only be detected when replisomes have accumulated in their ubiquitylated form due to inhibition of p97 activity (Fig. 4E, compare lane 6–7). Similarly, despite immunoprecipitation of equal quantities of neddylated and unneddylated Cul2 from each sample, Cul2 could only interact with Ubxn7 when in its neddylated form on p97i treated chromatin (Fig. 4E, lane 8 and 9). Reassuringly, all three, p97, Cul2 and Ubxn7, could coimmunoprecipitate a little of the component of the terminated replisome Cdc45. Altogether, these experiments suggest that Ubxn7, although being a p97 cofactor, is recruited to chromatin during the termination reaction through its interaction with neddylated Cul2. Moreover, Faf1 is likely to form a common complex with p97 and Ubxn7 as we can see it interacting with Ubxn7 when present on chromatin (Fig. 4E, lane 10). Our results suggest that, analogously to Hif1α regulation, Ubxn7 acts as a bridge between Cul2Lrr1, its substrate Mcm7, and the p97 segregase complex. To explore this idea in more detail, we decided to make use of separation-of-function mutants of Ubxn7 that cannot interact with p97 (UBX domain mutated, rUbxn7ΔUBX) or Cul2 (UIM domain mutated, rUbxn7ΔUIM) (Fig. S2, A and C). In human cells, the P459G mutation abolishes interaction with p97, while S297A abolishes interaction with neddylated Cul2, while not affecting p97 interaction (20Bandau S. Knebel A. Gage Z.O. Wood N.T. Alexandru G. UBXN7 docks on neddylated cullin complexes using its UIM motif and causes HIF1alpha accumulation.BMC Biol. 2012; 10: 36Crossref PubMed Scopus (46) Google Scholar). Moreover, L290, A293 and S297 were found to be the most conserved amino acids in UIMs of several human proteins (22den Besten W. Verma R. Kleiger G. Oania R.S. Deshaies R.J. NEDD8 links cullin-RING ubiquitin ligase function to the p97 pathway.Nat. Struct. Mol. Biol. 2012; 19https://doi.org/10.1038/nsmb.2269Crossref PubMed Scopus (62) Google Scholar). We have therefore mutated corresponding P458G in the X. laevis Ubxn7 sequence to create rUbxn7ΔUBX and the corresponding L286E/A289Q/S293A residues to create rUbxn7ΔUIM (Fig.S5A). While we were able to confirm that rUbxn7ΔUBX cannot interact with p97 in the egg extract (Fig. S5B), Ubxn7 and Cul2 do not interact in the egg extract (cytoplasm) and so it is not easy to verify whether the rUbxn7ΔUIM mutation disrupts this. We did observe, however, that adding a high concentra
Referência(s)