PIP-Box-Mediated Degradation Prohibits Re-Accumulation of Cdc6 During S Phase
Total Page:16
File Type:pdf, Size:1020Kb
ß 2014. Published by The Company of Biologists Ltd | Journal of Cell Science (2014) 127, 1336–1345 doi:10.1242/jcs.145862 RESEARCH ARTICLE PIP-box-mediated degradation prohibits re-accumulation of Cdc6 during S phase Linda Clijsters1 and Rob Wolthuis1,2,* ABSTRACT levels of Cdt1 greatly increase owing to formation of a stable complex with its inhibitor geminin (Ballabeni et al., 2004; Cdc6 and Cdt1 initiate DNA replication licensing when cells exit Clijsters et al., 2013). When cells satisfy the spindle checkpoint mitosis. In cycling cells, Cdc6 is efficiently degraded from anaphase and exit mitosis, cyclin B1 is degraded through the E3 ligase onwards as a result of APC/C–Cdh1 activity. When APC/C–Cdh1 is anaphase-promoting complex/cyclosome containing Cdc20 switched off again, at the end of G1 phase, Cdc6 could thus re- (APC/C–Cdc20), which inactivates Cdk1, a key licensing accumulate, risking the re-licensing of DNA as long as Cdt1 is inhibitor (Pines, 2011). Geminin, a second key licensing present. Here, we carefully investigated the dynamics of Cdt1 and inhibitor in mammals, is also degraded by APC/C–Cdc20 when Cdc6 in cycling cells. We reveal a novel APC/C–Cdh1-independent the spindle checkpoint is satisfied (Clijsters et al., 2013). This degradation pathway that prevents nuclear Cdc6 re-accumulation at means that, in proliferating cells, the competence to license the G1-S transition and during S phase. Similar to Cdt1, nuclear DNA replication greatly increases at the metaphase-to-anaphase clearance of Cdc6 depends on an N-terminal PIP-box and the Cdt2- transition. Immediately after anaphase, however, Cdc6 is containing CRL4 complex. When cells reach G2 phase, Cdc6 degraded through APC/C containing Cdh1 (also known as rapidly re-accumulates but, at this time, Cdt1 is mostly absent and FZR1 in human). This schedule of events suggests that, in expression of Cdc6 is limited to the cytoplasm. We propose that rapidly cycling cells, there is a short window of time at the end of Cdk1 contributes to the nuclear export of Cdc6 at the S-to-G2 mitosis when licensing competence peaks (Clijsters et al., 2013). transition. In summary, our results show that different control Cdt1, after its release from geminin, remains present during mechanisms of Cdc6 restrain erroneous licensing of DNA mitotic exit and G1 phase but is lost when PCNA is loaded onto replication during G1, S and G2 phase. DNA (Clijsters et al., 2013). Cdt1 degradation requires a so-called PIP-box motif that is typically recognized by the E3 ligase cullin KEY WORDS: CRL4–Cdt2, Cdc6, Cdt1, PCNA, PIP-box, Cell cycle ring (CRL)4 complex containing Cdt2 (CRL4–Cdt2). This Cdt1 degradation motif is only functional when CRL4–Cdt2 interacts with the DNA-loaded form of PCNA (Havens and Walter, 2009; INTRODUCTION Havens et al., 2012). Thereby, Cdt1 clearance coincides robustly Cell division control protein 6 (Cdc6) is an essential factor of the with the start of S phase, reducing the chance that any newly pre-replication complex (preRC). Together with Cdt1, Cdc6 loads formed DNA is re-licensed and subsequently re-replicated. Another the mini-chromosome maintenance (MCM) complex, which Cdt1 destruction mechanism might rely on phosphorylation and consists of the six MCM proteins (numbered 2 to 7, hereafter subsequent destruction via the E3 ligase SCF–Skp2 (Liu et al., referred to as MCM2–7) on origins of DNA replication. This 2004; Nishitani et al., 2006). The risk of DNA re-replication is also process is called DNA replication licensing. Once licensed, reduced by the degradation of Cdc6 during mitotic exit. However, replication origins can be activated by cyclin–Cdk2 complexes by the end of G1 phase, the APC/C switches off and Cdc6 is newly and Cdc7 kinase. DNA replication initiation factors, such as synthesized (Bassermann et al., 2014; Duursma and Agami, 2005; Cdc45 and GINS complex factors, subsequently facilitate the Hsu et al., 2002; Piatti et al., 1996; Rape and Kirschner, 2004). In process of DNA unwinding that is carried out by MCM2– fact, when quiescent cells re-enter the cell cycle, the first 7. Replication begins when the homotrimeric DNA-clamp opportunity to license DNA replication would arise at this point, proliferating cell nuclear antigen (PCNA) is loaded onto the when newly synthesized Cdc6 is protected from degradation by chromatin, enabling recruitment of DNA polymerases (Masai phosphorylation, even though APC/C–Cdh1 is highly active in et al., 2010). Altogether, this means that the onset of S phase is quiescent cells. (Mailand and Diffley, 2005). By contrast, we found largely controlled by regulatory events that take place during the that rapidly proliferating cells have, probably, already licensed preceding mitosis and G1 phase. their origins during mitotic exit, before APC/C–Cdh1 is activated Early in mitosis, cells start preparing for DNA replication (Clijsters et al., 2013). In these cells, expression of Cdc6 at the end licensing: Cdc6 translocates to condensed chromosomes, and the of G1, when Cdt1 levels are still high, could pose a threat to genomic integrity. Here, we investigated how cycling cells control Cdc6 and prohibit re-licensing as they enter, and progress through, 1Division of Cell Biology (B5) and Division of Molecular Carcinogenesis (B7), The Sphase. Netherlands Cancer Institute (NKI-AVL), Plesmanlaan 121, 1066 CX, Amsterdam, The Netherlands. 2Section of Oncogenetics, Department of Clinical Genetics and CCA/V-ICI Research Program Oncogenesis, VUmc Medical Faculty, van de RESULTS Boechorststraat 7, 1081 BT Amsterdam, The Netherlands. APC/C–Cdh1- and KEN-box-independent degradation of Cdc6 *Author for correspondence ([email protected]) Cdc6 degradation starts in anaphase in a manner strictly dependent on APC/C–Cdh1 (Fig. 1A) (Clijsters et al., 2013). Received 5 November 2013; Accepted 17 December 2013 Cdc6 is absent from early G1 phase onwards and, in most cells, Journal of Cell Science 1336 RESEARCH ARTICLE Journal of Cell Science (2014) 127, 1336–1345 doi:10.1242/jcs.145862 Fig. 1. APC/C–Cdh1- and KEN-box-independent degradation of Cdc6. (A) Mitotic U2OS cells [transfected with siRNA against CDH1 (si-CDH1) or a control siRNA (mock)] were either treated with 10 mM Cdk1 inhibitor RO-3306 or left untreated. Extracts from these cells were then subjected to western blotting for the indicated proteins. (B) U2OS cells expressing Cdc6–Venus were imaged by differential interference contrast (DIC) and fluorescence microscopy. Upper panel, no siRNA. Lower panel, si-CDH1. Images of the indicated phases of mitosis are shown. NEB, nuclear envelope breakdown. (C) U2OS cells expressing Cdc6–Venus and depleted of Cdh1 were filmed as in B. The intensity of fluorescence was plotted against time after NEB. Left, curves of individual cells. Right, average fluorescence of all cells compared with controls. (control, n511, mean6s.e.m.; si-CDH1, n59, mean6s.e.m.). (D) U2OS cells expressing KEN-Cdc6–Venus were imaged by DIC and fluorescence microscopy. Captured images of the indicated phases of mitosis are shown. (E) The intensity of fluorescence was plotted against time after NEB. Left, curves of individual cells. Right, average fluorescence of all cells compared to controls (control, n511, mean6s.e.m.; KEN-Cdc6–Venus, n511, mean6s.e.m.). does not re-accumulate until cells are at least several hours into S destabilize Cdc6 from the G1-S transition onwards, when APC/ phase. Remarkably, this means that in cycling cells, Cdc6 appears C–Cdh1 is not active. later in the cell cycle than other typical APC/C–Cdh1 substrates, Firstly, we followed fluorescent Cdc6–Venus by time-lapse such as Aurora A (Clijsters et al., 2013). Here, we investigated fluorescence microscopy in cells treated with siRNA. Cdc6– whether additional destruction mechanisms might exist that can Venus remained largely stable during and after mitosis, as well as Journal of Cell Science 1337 RESEARCH ARTICLE Journal of Cell Science (2014) 127, 1336–1345 doi:10.1242/jcs.145862 in G1 phase. However, Cdc6 was rapidly degraded several hours together with a fluorescent marker of S phase, Cherry-tagged later (Fig. 1B,C). In Cdh1 siRNA-treated cells, fluorescent Cdc6 PCNA. In most control cells, Cdc6–Venus was degraded levels declined specifically from within the nucleus (Fig. 1B, 90 substantially before Cherry–PCNA re-distributed towards and 120 minutes). To further investigate the apparent existence of discrete nuclear foci (dots), which represent DNA replication an APC/C–Cdh1-independent Cdc6 degradation pathway, we factories (Fig. 2A) (Burgess et al., 2012). However, in cells used a destruction box mutant of Cdc6, named KEN-Cdc6– depleted of Cdh1, Cdc6–Venus remained stable until the end of Venus, in which Lys81, Glu82 and Asn83 were mutated to Ala. G1 phase. Following the end of G1 phase, Cdc6 levels began to Indeed, KEN-Cdc6–Venus was stable at anaphase but suddenly diminish, a process which started in the nucleus, while at the disappeared later in G1 phase, again, with fluorescence declining same time, Cherry–PCNA dots formed (Fig. 2A; an example is specifically in the nucleus (Fig. 1D, 120 minutes). KEN-Cdc6– shown in Fig. 2B, compare 60 minutes to 90 minutes). In the Venus degradation started in a switch-like fashion (Fig. 1E). A presence of Cdh1, KEN-Cdc6–Venus also started to be degraded suggested D-box recognition motif comprising Arg56 and Leu59 when Cherry–PCNA dots appeared (Fig. 2A). This suggested did not contribute to the degradation pattern of Cdc6 in our live- that the APC/C–Cdh1-independent pathway that destabilizes cell degradation assay (Petersen et al., 2000) (supplementary Cdc6 is linked to the timing of PCNA loading. Decrease of Cdc6 material Fig. S1). We hypothesize from these experiments that a levels was not related to a reduction in protein synthesis as previously unknown APC/C-independent Cdc6 destruction treatment of cells with proteasome inhibitor MG-132 led to a mechanism is initiated by an event scheduled at a time around rapid accumulation of Cdc6, regardless of whether cells were in the end of G1 phase.