The APC/C (anaphase-promoting complex/cyclosome) is an E3 ubiquitin ligase that targets specific substrates for degradation by the 26S proteasome. APC/C activity depends on two cofactors, namely Cdc20 (cell division cycle 20) and Cdh1, which select the appropriate targets for ubiquitination. It is well established that APC/C is a target of the SAC (spindle assembly checkpoint) during mitosis and has critical roles in controlling the protein levels of major regulators of mitosis and DNA replication. In addition, recent studies have suggested new cell-cycle-independent functions of APC/C in non-mitotic cells and specifically in neuronal structure and function. Given the relevant functions of APC/C in cell proliferation and neuronal physiology, modulating APC/C activity may have beneficial effects in the clinic.
- anaphase-promoting complex/cyclosome (APC/C)
- cell cycle regulation
- cell division cycle 20 (Cdc20)
- spindle assembly checkpoint (SAC)
Cell cycle progression is tightly controlled through protein degradation of critical regulators at specific stages. Pioneer biochemical and genetic studies identified in 1995 a catalytic activity required for degradation of cyclins, the activator subunits of major kinases [known as Cdks (cyclin-dependent kinases)] involved in progression through the different phases of the cell cycle. The corresponding protein complex was named APC/C (anaphase-promoting complex/cyclosome) since its activity allowed sister chromatid separation and the transition from metaphase to anaphase and the subsequent exit from mitosis (reviewed in ).
The APC/C is an E3 ubiquitin ligase that assembles polyubiquitin chains on substrates (e.g. cyclins), targeting these proteins for destruction by the 26S proteasome [1,2]. This complex is composed of at least 12 core subunits and additional cofactors, Cdc20 (cell division cycle 20) and Cdh1, which interact only transiently with APC/C and are required for regulating its activity (Table 1). Whereas Cdc20 activates APC/C during early mitosis, Cdh1 is responsible for APC/C activity from late mitosis to the G1–S transition . Both Cdc20 and Cdh1 contain a C-terminal WD40 domain that mediates the ability to recognize APC/C substrates by interacting with specific elements such as the D-box , KEN-box , A-box  or O-box .
Cdks and the APC/C regulate each other during cell cycle progression. Cdc20 can only associate with APC/C when several subunits of APC/C have been phosphorylated by mitotic kinases such as Cdk1 or Plk1 (Polo-like kinase 1). In contrast, APC/C–Cdh1 interaction does not depend on phosphorylation of APC/C proteins. Instead, phosphorylation of Cdh1 by Cdk1 or Cdk2 during S-phase, G2-phase and early mitosis impairs its interaction with APC/C until late stages of mitosis. Thus Cdk activity determines whether APC/C is inactive (S-phase), activated by Cdc20 in early-mid-mitosis (due to high Cdk activity) or activated by Cdh1 during the exit of mitosis and G1-phase (due to low Cdk activity). On the other hand, APC/C controls Cdk activity by targeting for degradation A-type cyclins (during G2/M-phase) or B-type cyclins (metaphase–anaphase transition) in addition to other substrates (Table 1). The reciprocal control between Cdks and APC/C modulates the protein levels and activity of major regulators of the cell division cycle, allowing the proper transitions between DNA replication and mitosis [1,2].
In addition to the major function of APC/C in controlling the cell cycle, additional roles for APC/C components have been identified recently in non-mitotic cells such as neurons, muscle cells and lens fibre cells (see below). Thus the APC/C not only regulates cell cycle progression but may also be involved in cellular differentiation and neuronal physiology.
APC/C–Cdc20 and the SAC (spindle assembly checkpoint)
Cdc20 binds APC/C in early mitosis once several core subunits of the complex have been phosphorylated. Yet, its activity is restrained by different mechanisms. The most important is the SAC or mitotic checkpoint, a surveillance mechanism that detects unattached chromosomes and inhibits APC/C–Cdc20 activity, delaying anaphase onset until all the chromosomes are properly attached and bi-oriented at the metaphase plate . How the SAC inhibits APC/C–Cdc20 is not fully understood. The MCC (mitotic checkpoint complex) is a possible SAC effector. It is made up of three SAC genes, namely Mad2, BubR1/Mad3 and Bub3, and these along with Cdc20 bind to the APC/C as a pseudosubstrate, impairing its ubiquitin-ligase activity and delaying the metaphase–anaphase transition. Diverse genetic, biochemical and structural studies have led to a model where the formation of a soluble MCC is facilitated by specific interaction between Mad1 and Mad2 at the unattached kinetochores, generation of soluble Mad2 with the proper protein conformation and ultimately the interaction of Mad2 with Cdc20 and BubR1 . Recent evidence suggests that Mad1–Mad2 complexes bound at unattached kinetochores facilitate binding of soluble Mad2 to Cdc20, and this latter complex functions as a diffusible precursor to accelerate the formation of BubR1–Cdc20 inhibitory complexes [8,9]. Thus the APC/C is kept inactive either by the sequestering of Cdc20 in the MCC or by full interaction between the APC/C and MCC (APC/CMCC) in which Cdc20 seems to be displaced and the APC/C locked in a closed state, preventing binding and ubiquitination of substrates  (Figure 1). Recent studies have also postulated that the SAC inhibits APC/C activity by targeting Cdc20 for destruction. Most Cdc20 is not in complex with Mad2. Instead, Cdc20 forms a stable complex with BubR1 and Bub3. Mad2 catalyses the binding between Cdc20 and BubR1, which presents Cdc20 to the APC/C, targeting this cofactor for destruction [11–13].
Once all chromosomes have become bi-orientated on the metaphase plate, the checkpoint activity is extinguished. How the SAC is turned off is poorly understood. APC/C activation seems to be a rapid process involving APC/C-dependent multi-ubiquitination regulated by the E2 enzyme UBE2C (ubiquitin-conjugating enzyme E2 C)/UbcH10 and p31Comet, which promote APC/C-dependent Cdc20 ubiquitination, leading to the release of Cdc20 from the Mad2–BubR1 complex . Released Cdc20 rapidly binds to APC/C, promoting anaphase onset. By contrast, when the SAC is active, the deubiquitinating enzyme USP44 (ubiquitin-specific peptidase 44) deubiquitinates the APC/C co-activator Cdc20, stabilizing the Cdc20–Mad2–BubR1 complex and preventing the premature activation of the APC/C [14,15].
APC/C–Cdc20 initiates anaphase by targeting the separase inhibitors, securin [PTTG1 (pituitary-tumour transforming gene 1 protein)] and cyclin B1, for destruction [16,17]. Securin is a small protein that functions as stoichiometric inhibitor of separase, whereas cyclin B1–Cdk1 complexes inhibit separase by direct binding and phosphorylation . Degradation of securin and cyclin B thus results in inactivation of Cdk1 and activation of separase, which in turn cleaves cohesins to liberate sister chromatids at anaphase onset (Figure 1). In late anaphase, when Cdk1 activity is extinguished and mitotic phosphatases have been activated, removal of the inhibitory phosphates of Cdh1 results in the formation of active APC/C–Cdh1 complexes (Figure 1). In Saccharomyces cerevisiae, separase activation results in release of the Cdc14 phosphatase from the nucleolus, promoting Cdc14 activity and mitotic exit . However, how mitotic cells activate the phosphatases required for dephosphorylation of Cdk substrates remains unclear in most other organisms [19,20].
Securin and cyclin B1 seem to be the only essential targets of APC/C–Cdc20, but this complex also ubiquitinates other substrates such as cyclin A, Nek2a (never in mitosis in Aspergillus nidulans-related kinase 2a) and p21Cip1 (Figure 1). These substrates are targeted for degradation before the metaphase–anaphase transition and their proteolysis is not delayed by the SAC [21–23]. p21Cip1 proteolysis by APC/C–Cdc20 in G2/M may lead to full activation of Cdk1 necessary for mitotic entry . Recent evidence suggests that degradation of these SAC-independent substrates is achieved by recruiting these targets to the APC/C in prometaphase before checkpoint activation [10,24].
In summary, Cdc20 is an essential activator of the APC/C during early mitosis, when Cdk activity is high and Cdh1 is inhibited by Cdk-dependent phosphorylation. After nuclear envelope breakdown, APC/C–Cdc20 targets cyclin A, Nek2 and p21Cip1 for degradation during prometaphase. The presence of unattached kinetochores inactivates APC/C either by sequestering Cdc20 or by generating locked APC/CMCC complexes. Once all chromosomes are bipolarly attached, Mad2 signal disappears from the kinetochores and active APC/C–Cdc20 complexes target securin and cyclin B for degradation, resulting in inactive Cdk1, active separase, cohesin cleavage and release and separation of sister chromatids at anaphase onset (Figure 1).
Role of APC/C–Cdh1 in the exit from mitosis and interphase
Once Cdk1 is inactivated at anaphase onset, Cdh1 phosphates are removed by phosphatases during anaphase, resulting in active APC/C–Cdh1 complexes (Figure 1). APC/C–Cdh1 allows mitotic exit by targeting for destruction Plk1 and Aurora kinases among other substrates (Table 1) and completing the elimination of mitotic cyclins. In addition, Cdh1 also degrades Cdc20, leading to the complete switch from APC/C–Cdc20 to APC/C–Cdh1 complexes during the exit from mitosis .
During G1, APC/C–Cdh1 maintains low levels of Cdk activity by degrading cyclins, the Cdk activator Cdc25A as well as two components of the SCF [Skp1 (S-phase kinase-associated protein 1)/cullin/F-box] complex, namely Skp2 and Cks1, which degrade p27Kip1 and p21Cip1 Cdk inhibitors (Table 1). Thus G0/G1 is maintained by a combination of Cdh1 and Cdk inhibitors that prevent Cdk activity. Under these conditions, the members of the retinoblastoma protein family act as transcriptional repressors to avoid the expression of cell cycle genes during G0/G1. In addition, APC/C–Cdh1 controls DNA replication through regulation of prereplication complex formation by targeting geminin (the Cdt1 inhibitor) and other replication regulators, such as Cdc6, Tk1 and Tmpk (Table 1). At the G1–S transition, the mitogenic-dependent increase in Cdk activity leads to phosphorylation and inactivation of Cdh1. In addition, APC/C is negatively regulated during S-phase by either Emi1  or degradation of Cdh1 , leading to geminin stabilization and preventing DNA reduplication after replication origins have fired.
Despite its multiple functions during the cell cycle, Cdh1 is not essential in yeasts, Caenorhabditis elegans or Drosophila ( and references herein). Genetic ablation of Cdh1 in mammals results in embryonic lethality during mid-gestation owing to placental defects [27,28]. These defects are due to the requirements of Cdh1 for endoreplication, a process in which cells, such as placental giant trophoblasts or Drosophila salivary gland cells, undergo repeated rounds of DNA replication without entering mitosis in order to acquire polyploidy and increase cell size [27,29]. Indeed, Cdh1-null mice are viable when these placental defects are restored, indicating that Cdh1 is also dispensable for the cell cycle in mammals ( and M. Eguren and M. Malumbres, unpublished work). Yet, Cdh1 deficiency leads to cellular defects in the timing of mitotic exit, cytokinesis and DNA replication that lead to chromosomal aberrations and genomic instability [27,28].
Cell-cycle-independent functions of APC/C
The APC/C was suggested to play a role in the nervous system when high expression of several APC/C components was reported in postmitotic neurons . The APC/C has been found to act locally in pre- and post-synaptic compartments in C. elegans and Drosophila to regulate synaptic strength in mature neurons. Thus the APC/C regulates the number of glutamate receptors on the postsynaptic side in the ventral nerve cord of C. elegans  and controls presynaptic organization and synaptic size in Drosophila by targeting liprin-α for degradation, an activity that probably depends on the Cdh1 cofactor  (Figure 2). In mammals, Cdh1 acts as a regulator of axonal growth and patterning during development in normal brain [33–35]. Cdh1 knockdown by RNAi (RNA interference) resulted in increased axons and disrupted axonal patterning without affecting dendritic morphology . Subsequent studies have reported that Cdh1 mediates the degradation of two nuclear proteins, Id2  and the transcriptional co-repressor SnoN , whose stabilization increases axonal growth (Figure 2). In cultured postmitotic neurons and neuroblastoma cells, Cdh1 depletion leads to apoptosis . This cell death is accompanied by increased cyclin B levels and BrdU (bromodeoxyuridine) incorporation, suggesting that postmitotic neurons could be more sensitive to apoptotic cell death induced by an aberrant re-entry into the cell cycle [36,37]. APC/C–Cdh1 activity down-regulates cyclin B protein levels as an essential survival mechanism, thus preventing inappropriate cell cycle entry and apoptosis in postmitotic neurons. The bioenergetic and antioxidant status of neurons is also controlled by continuous degradation of a key glycolytic enzyme, Pfkfb3 (6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase-3), by APC/C–Cdh1 .
A recent study has proposed a new centrosomal-dependent activity of APC/C–Cdc20 to regulate dendrite morphogenesis . Cdc20 depletion impairs dendrite growth and branching in cerebellar granule neurons, but it does not affect axonal growth. In these neurons, the centrosome-associated protein HDAC6 (histone deacetylase 6) promotes the polyubiquitination of Cdc20 and stimulates the activity of centrosomal APC/C–Cdc20. This interaction is required for APC/C–Cdc20-dependent degradation of Id1, a transcription factor that regulates dendrite size . Thus APC/C–Cdc20-mediated degradation of Id1 modulates the proper growth of dendrites in these specialized cells (Figure 2).
Control of neuronal function by the APC/C seems to have relevant consequences in vivo. Cdh1 heterozygous mice display synaptic and behaviour defects including alterations in neuromuscular co-ordination and learning [27,28]. Since Cdh1 may have relevant roles in maintaining the quiescence state of neuronal progenitors , it will be interesting to dissect to what extent these phenotypes depend on cell-cycle-dependent versus -independent functions of the APC/C. Similarly, the physiological requirements for Cdc20 in postmitotic neurons deserve further investigation in vivo.
Relevance of the APC/C in human disease: future perspectives
CIN (chromosomal instability) is a frequent characteristic of human solid tumours. In most of the cases, CIN is the consequence of mitotic errors such as chromosome mis-segregation that finally leads to aneuploidy. Cdc20 overexpression is a central feature of the CIN signature [40,41] and Cdc20 is one of the most highly up-regulated genes in many types of malignancies, such as glioblastomas or lung adenocarcinomas . Importantly, p53 negatively regulates Cdc20, and ectopic expression of p53 represses Cdc20 and promotes G2/M arrest, impairing tumour cell growth . Targeting Cdc20 by RNAi enhances the cytotoxicity of chemoradiation, further supporting the therapeutic value of Cdc20 as a molecular target for cancer therapy . However, recent studies on Cdc20 depletion using RNAi have claimed that, despite the essential roles of Cdc20 in the first embryo divisions , the elimination of Cdc20 does not provoke cell cycle arrest or stabilization of the APC/C–Cdc20 substrates cyclin B and securin in somatic cells [46,47]. These results may have important therapeutic implications, invalidating the possible role of Cdc20 as a therapeutic target, and they also suggest that a redundant mechanism controls the metaphase–anaphase transition . One alternative explanation is that only a minimal part of Cdc20 in the cells (not eliminated by RNAi) is required for the APC/C–Cdc20 mitotic function. Future work testing the genetic ablation of Cdc20 in somatic cells will be required to answer this question.
On the other hand, inhibition of Cdh1 may result in many different cellular outcomes. First, lack of Cdh1 should lead to an increase in oncogenic proteins such as Plk1, Aurora A, Cdc6 or Skp2, resulting in enhanced proliferation, a hallmark of cancer. Secondly, it is also possible that Cdh1 depletion may promote senescence due to increases in Ets2 and p16INK4a levels . Moreover, lack of the Cdh1 causes genomic instability, which has been suggested to act both as a tumour suppressor and an oncogene in vivo, resulting in either inefficient proliferation or spontaneous tumours . All these complex phenotypes significantly enhance the importance of investigating the consequences of Cdh1 inhibition in vivo.
Targeting mitotic regulators is currently an attractive approach against proliferation of tumour cells . A recent study has shown that Ras mutant cancer cells undergo an important mitotic stress and are more dependent on key mitotic proteins including several APC/C components . Targeting APC/C proteins could therefore exacerbate this mitotic stress to selectively kill Ras mutant cells. This effect is likely to be common for other oncogenes, suggesting wide uses of inhibiting APC/C function in cancer therapy. Given the current evaluation of proteasome inhibitors in clinical trials and the relevance of the APC/C in cell proliferation and neuronal function, it will be necessary to understand the requirements for Cdc20 and Cdh1 in adult tissues and the consequences of their inhibition in vivo. Hopefully, these studies will suggest putative therapeutic uses of inhibiting APC/C function in human disease.
E.M. and M.E. are supported by fellowships from the Spanish Ministerio de Ciencia e Innovación. The Cell Division and Cancer Group of the CNIO is supported by grants from the Association International for Cancer Research [grant number 08-0188], the OncoCycle Programme [grant number S-BIO-0283-2006] of the Comunidad de Madrid and the Ministerio de Ciencia e Innovación [grant numbers SAF2006-05186 and CSD2007-00017].
We thank members of Cell Division and Cancer Group of the CNIO for comments and helpful discussions.
Ubiquitin–Proteasome System, Dynamics and Targeting: 4th Intracellular Proteolysis Meeting, a Biochemical Society Focused Meeting held at Institut d'Estudis Catalans, Casa de Convalescència, Barcelona, Spain, 27–29 May 2009. Organized and Edited by Bernat Crosas (Institute of Molecular Biology of Barcelona, Spain), Rosa Farràs (Centro de Investigación Príncipe Felipe, Valencia, Spain), Gemma Marfany (University of Barcelona, Spain), Manuel Rodríguez (CIC bioGUNE, Derio, Spain) and Timothy Thomson (Institute of Molecular Biology of Barcelona, Spain)
Abbreviations: APC/C, anaphase-promoting complex/cyclosome; Cdk, cyclin-dependent kinase; CIN, chromosomal instability; HDAC6, histone deacetylase 6; Id2, inhibitor of DNA binding 2; MCC, mitotic checkpoint complex; Nek2a, never in mitosis in Aspergillus nidulans-related kinase 2a; Pfkfb3, 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase-3; Plk1, Polo-like kinase 1; RNAi, RNA interference; SAC, spindle assembly checkpoint; Skp1, S-phase kinase-associated protein 1; UBE2C, ubiquitin-conjugating enzyme E2 C
- © The Authors Journal compilation © 2010 Biochemical Society