Mitosis: a matter of getting rid of the right protein at the right time

Review

TRENDS in Cell Biology

Vol.16 No.1 January 2006

Chromosome Segregation and Aneuploidy series

Mitosis: a matter of getting rid of the right protein at the right time Jonathon Pines Wellcome/Cancer Research UK Gurdon Institute & Dept of Zoology, Tennis Court Road, Cambridge, UK CB2 1QN

There are two major problems for the cell to solve in mitosis: how to ensure that each daughter cell receives an equal and identical complement of the genome, and how to prevent cell separation before chromosome segregation. Both these problems are solved by controlling when two specific proteins are destroyed: securin, an inhibitor of chromosome segregation, and cyclin B, which inhibits cell separation (cytokinesis). It has recently become clear that several other proteins are degraded at specific points in mitosis. This review (which is part of the Chromosome Segregation and Aneuploidy series) focuses on how specific proteins are selected for proteolysis at defined points in mitosis and how this contributes to the proper coordination of chromosome segregation and cytokinesis.

Introduction Mitosis is conventionally divided up into discrete stages according to the morphology of the cell (Figure 1, but see reference [1] for an alternative view [1]). In cells that undergo an open mitosis, prophase ends with nuclear envelope breakdown, and the subsequent stage, prometaphase, is defined by the search-and-capture behaviour of microtubules as the kinetochores are attached to the spindle. (See the accompanying articles in this series by Tim Yen and by Benjamin Pinsky and Sue Biggins [91,92].) Once all the kinetochores have correctly attached to the mitotic spindle, the cell is defined as being in metaphase, and chromosomes proceed to align on a ‘metaphase plate’. By this definition, metaphase can be a remarkably defined length of time that is likely to be set by how long it takes to degrade particular proteins (see below). Metaphase ends with the rapid and almost synchronous separation of all the sister chromatids, which begin to segregate to opposite poles of the spindle (anaphase A), followed by elongation of the spindle itself (anaphase B). Once each set of sister chromatids has reached opposite spindle poles, the chromatids begin to decondense, the nuclear envelope re-forms and the mitotic spindle disassembles (telophase). During anaphase and telophase, the cell itself begins to divide (cytokinesis) to generate two genetically identical daughter cells, although in animal cells these do not complete separation until abscission that, in mammalian cell culture, can take Corresponding author: Pines, J. ([email protected]). Available online 5 December 2005

place hours after cells re-enter interphase. As we shall see, several of these events are coordinated by proteolysis. Cells are driven into and through mitosis by the mitotic cyclin-dependent kinases (CDKs) working in concert with several other protein kinases such as members of the Polo, Aurora and NIMA families (reviewed in [2]). The kinases are often coordinated by recruitment to a substrate only after it has been phosphorylated by an upstream kinase. For example, polo kinases use their polo-box to bind to sites previously phosphorylated by mitotic CDKs [3]. Moreover, the cyclins themselves have recently been shown to contain a phospho-peptide binding site in their conserved ‘cyclin fold’ [4]. The mitotic kinases are antagonized by phosphatases, and it is the balance between these that controls several steps in mitosis. Although phosphorylation is a rapidly reversible event, inactivating a kinase or phosphatase by proteolysis can make phosphorylation or dephosphorylation effectively irreversible and confer directionality. The ability to select a specific protein for rapid proteolysis is conferred by the ubiquitin–proteasome system (UPS) (reviewed in [5]). Proteins degraded by the UPS are tagged with a multi-ubiquitin chain that is recognized by the proteasome cap. Proteasomes appear to be constitutively active throughout the cell cycle; therefore, substrate selection is primarily controlled by when and where proteins are ubiquitylated. Ubiquitin is transferred onto the 3-amino group of a lysine residue of a substrate by a ubiquitin-conjugating enzyme (UBC) working in concert with a ubiquitin ligase. The ubiquitin can subsequently be removed by deubiquitinating enzymes (DUBs), a large number of which are encoded in the genome. Some DUBs are components of the proteasome cap, where they have a general role in ‘proof reading’ substrate selection or recycling ubiquitin. It is highly likely that others will be found that are required for the proper regulation of mitosis. Although ubiquitylation can target a protein to the proteasome, it can also perform other important roles – for example, in endocytosis and signal transduction. Thus, some proteins might be ubiquitylated in mitosis for purposes other than destruction, but this function is outside the scope of this review. The entry to mitosis can be regulated by proteolysis (Box 1), but it is in mitosis itself that the UPS has its most defined cell cycle roles. In mitosis, most of the specificity in substrate selection is conferred by the ubiquitin ligase, of which the most prominent is a multisubunit complex

www.sciencedirect.com 0962-8924/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.tcb.2005.11.006

56

Review

TRENDS in Cell Biology

Vol.16 No.1 January 2006

Antephase

G2 phase Prophase Pro-metaphase Metaphase

NEBD

Anaphase

Telophase

Chromosome Chromatid APCCdh1 Cytokinesis Spindle disassembly congression separation

Spindle checkpoint

Time

Cyclin A Nek2A HOXC10

Securin Cyclin B

Cdc20 Plk1

Aurora A

After anaphase: Kip1, Cin8, Ase1, slk1, ndc10 Anillin, Geminin, Aurora B TRENDS in Cell Biology

Figure 1. Schematic of when specific substrates are degraded as cells progress into, and through, mitosis. After the APC/C is activated at nuclear envelope breakdown (NEBD), it recognizes and degrades different proteins at different times according to specific cues, such as whether the spindle checkpoint is active and when Cdh1 replaces Cdc20. See text and Table 1 for details.

called the anaphase-promoting complex or cyclosome (APC/C). The APC/C has the primary role in ensuring correct chromosome segregation and in coordinating mitosis with cytokinesis. Thus, the question of how the APC/C is activated and how it recognizes its substrates is key to understanding how mitosis is regulated. The multifaceted anaphase-promoting complex/cyclosome The APC/C comprises up to 13 different subunits in yeast, and 11 subunits in animal cells (reviewed in [6,7]). The catalytic subunits are APC11, a RING-finger protein, APC2, a protein with homology to the cullin family, and Doc1, a subunit that is important for substrate recognition and/or extending the poly-ubiquitin chain on a substrate [8–10]. The function of the other subunits is unclear, but there is evidence that they might also be important in substrate recognition. Several of these subunits contain protein– protein interaction domains of the tetratricopeptide repeat (TPR) family and are multiply phosphorylated in mitosis, which is required to activate the APC/C [11,12] but could also alter substrate binding affinities. Several of the phosphorylation sites have been mapped and are mitotic cyclin–CDK and Polo-like kinase sites, of which the cyclin–CDK sites are the most important for activating the APC/C [11,12]. In addition to its core components, the APC/C requires a member of the WD40 family for activity. Three different WD40 proteins can act with the APC/C: Cdc20, Cdh1 and Ama1 (these are the names of the proteins in budding yeast). These proteins have a conserved isoleucine– arginine (IR) dipeptide motif at their C-terminus that is required for them to bind to the APC/C, apparently to subunits with TPR motifs [13]. The WD40 proteins act at different times in the cell cycle and alter the range of substrates recognized by the APC/C (reviewed in [14]; see Table 1 for a list of APC/C substrates). Cdc20 (fizzy in www.sciencedirect.com

Drosophila) acts in all cells; it is most important in embryonic cell cycles and in early mitosis in somatic cells. When there are unattached kinetochores in the cell, Cdc20 is inactivated by the spindle checkpoint to prevent anaphase (see below and the accompanying article by

Box 1. Proteolysis at the entry to mitosis and a brief word about Antephase and Chfr After mammalian cells have begun to condense their chromosomes but before the cell commits to mitosis, certain types of stress – such as microtubule or topoisomerase poisons – can cause the cell to delay nuclear envelope breakdown and entry to mitosis [75,76]. This response is called the antephase checkpoint and activates the p38 stress kinase [75,76]. The Chfr ubiquitin ligase has been implicated as a regulator of this checkpoint [75], and mice lacking Chfr are tumour prone and their cells develop aneuploidy in culture [77]. Chfr has been proposed to target the Xenopus Polo-like kinase [78] and the mammalian Aurora A kinase [77] for proteolysis to impose the delay on mitosis. However, Chfr primarily conjugates ubiquitin through lysine 63 [79], which tends to correlate with activating a damage signaling pathway rather than targeting a substrate to the proteasome, and the antephase checkpoint still functions in the presence of a variety of proteasome inhibitors [75]. The paradigm for how most cells trigger mitosis – budding yeast are an exception here – is that the principal mitotic cyclin–CDK complex accumulates in interphase but is held in check by the Wee1 family of kinases that phosphorylate and inactivate the complex. To trigger mitosis, cells alter the balance between Wee1 and the antagonistic phosphatase – Cdc25 – that activates the cyclin–CDK, and the levels of both Wee1 and Cdc25 can be regulated by proteolysis. In Xenopus egg extracts, and in fission and budding yeast, Wee1 is stabilized by checkpoints that delay mitosis [80–82]. Cells also alter the stability of Cdc25 to regulate mitotic entry. In mammalian cells, there are three forms of Cdc25: Cdc25A, Cdc25B and CDC25C, of which only Cdc25A is essential for mitosis [83]. Cdc25A is targeted for degradation by SCFb-TrCP in response to unreplicated or damaged DNA after phosphorylation by the Chkfamily of kinases (reviewed in [84]). Cdc25A and cyclin-B–Cdk1 form a positive-feedback loop at mitosis because Cdc25A is stabilized by phosphorylation by cyclin-B–Cdk1.

Review

TRENDS in Cell Biology

Vol.16 No.1 January 2006

57

Table 1. Cell-cycle-related APC/C substrates

a

Protein Cyclin A Nek2A HoxC10

Timing of destruction Pro-metaphase Pro-metaphase Pro-metaphase

APC/C activator Cdc20 Cdc20 Cdc20

Checkpoint-dependent? No No No

Function Mitotic regulator Centrosome regulator Transcription factor

Cyclin B Securin

Metaphase Metaphase

Cdc20 Cdc20

Yes Yes

Mitotic regulator Separase inhibitor

Cdc20 UbcH10/vihar Cdh1

Anaphase Anaphase/telophase G1 and G0

Cdh1 ? Cdh1

No No No

APC/C activator APC/C cofactor APC/C activator

Cyclin B3 Plk1 Aurora A Aurora B Survivin

Anaphase Anaphase Anaphase Anaphase/telophase After anaphasea

Cdh1 Cdh1 Cdh1 Cdh1 Cdh1

No No No No No

Mitotic/meiotic regulator Mitotic regulator Mitotic regulator Mitotic regulator Localizes Aurora B

Ndc10 Slk1 Ase1/Prc1 Kip1 Cin8 Anillin ECT2

After After After After After After After

anaphasea anaphasea anaphasea anaphasea anaphasea anaphasea anaphasea

Cdh1 Cdh1 Cdh1 Cdh1 Cdh1 Cdh1 Cdh1

No No No No No No No

Kinetochore component Spindle component Spindle component Spindle motor Spindle motor Cytokinesis regulator Cytokinesis regulator

Geminin

After anaphasea

Cdh1

No

DNA replication inhibitor

Hsl1 Skp2 Cks1 SnoN

G1 G1 G1 G1

Cdh1 Cdh1 Cdh1 Cdh1

No No No No

Swe1 inhibitor SCF component CDK cofactor Inhibitor of TGFb genes

Determined by immunoblot or using in vitro extracts. Exact timing not yet determined using an in vivo assay.

Roger Karess). APC/C bound to Cdc20 (APC/CCdc20) seems primarily to recognize substrates with ‘destruction box’ (D box) motifs (see Box 2). By contrast, Cdh1 (fizzy-related in Drosophila) does not seem to be present in most embryonic cell cycles and is most important for ubiquitylation in anaphase and on through the following G1 phase. APC/C bound to Cdh1 (APC/CCdh1) can recognize substrates with either a D box or a KEN box (see Box 2) and thus has a wider range of substrates than APC/CCdc20. Ama1 only acts in meiosis; remarkably, one of the APC/C subunits in mitotic cells inhibits Ama1 to prevent it acting prematurely in meiosis and in mitotic cells [15,16]. Ubiquitin ligases other than the APC/C also have important roles in mitosis, in particular members of

the SCF (Skp1–Cullin–F-box) family. These are modular complexes with core components – a cullin, the Rbx1 RING-finger protein and the Skp1 protein – plus a variable protein of the F-box family that is responsible for recruiting the substrate [17]. The ligase recognizes a phosphorylated domain (phospho-degron) on the substrate, meaning that, unlike the APC/C, the timing of ubiquitylation is controlled by modifying the substrate and not by altering the activity of the ligase. Activating the APC/C and recognizing its prometaphase substrates When cyclin-B–Cdk1 is fully activated, cells are committed to mitosis. At the same time, cells become

Box 2. How does the APC/C recognize its substrates? The motif recognized by the APC/C on its substrates has become increasingly unclear. Initially, the ‘destruction box’ (D box) was identified as a nine-amino-acid sequence conserved in B-type cyclins that was required for proteolysis and could confer mitotic instability when transferred to another protein. However, as more APC/C substrates have been identified, the D box sequence has become increasingly redundant (it is now R/KxxL/I/M/V) and is only recognized by APC/CCdc20 in some, usually unstructured, regions in a protein, whereas APC/CCdh1 is able to recognize it in a broader range of contexts [85]. Even more mysterious is the KEN box (the consensus is KEN), which is required for the degradation of some APC/CCdh1 substrates but is highly context dependent [85], not sufficient in itself to confer instability on another protein and cannot be recognized by APC/CCdc20. The difference in the abilities of APC/CCd20 and APC/CCdh1 to recognize D boxes and KEN boxes could be simply explained if Cdc20 and Cdh1 themselves recognize and bind APC/C substrates, and www.sciencedirect.com

some evidence supports this, particularly for Cdh1. Cdh1 can bind directly to some substrates in a D-box- or KEN-box-dependent manner [86,87], and recent elegant studies by Kraft et al. using photocrosslinking have mapped the binding sites to one face of the bpropeller of the WD40 domain [88]. Crosslinking also demonstrated that Cdh1 interacts primarily with the APC3 subunit [88]. However, it is less clear how APC/CCdc20 recognizes its substrates. Some studies indicate that Cdc20 binds to substrates [89], but there is also strong evidence that the mitotic APC/C itself, without Cdc20, can bind directly to D boxes [90]. Indeed, the Doc1 subunit of the APC/C has been implicated as a protein required to recognize some substrates in budding yeast [8,9] and resembles Cdc20 and Cdh1 in having an IR motif at its C-terminus. Of course, it is possible that some substrates are recruited by Cdc20 or Cdh1, and others by the APC/C itself, which would help to explain how the APC/C can act on a wide variety of different substrates at distinct times in mitosis.

Review

58

(a)

TRENDS in Cell Biology

Vol.16 No.1 January 2006

(b)

Activating the APC/C

Separating sister chromatids

Spindle checkpoint Cleaved Scc1 Emi1?

Mad2 BubR1

Cdc20

APC/C

Inactive

Phospho-APC/C + Cdc20

Active

Cyclin B1–Cdk1 Plk1 Plk1 Aurora

Separase

Securin

Phospho-APC/C + Cdc20 Shugoshin Cohesin complex TRENDS in Cell Biology

Figure 2. Schematics of how the APC/C is activated (a) and how sister chromatid separation is regulated (b). (a) The APC/C is activated by a combination of phosphorylation by cyclin–CDK and Plk1 kinases, and by binding to Cdc20. Cdc20 itself might be held in check by Emi before mitosis and is certainly regulated by the spindle checkpoint in prometaphase. (b) Sister chromatids are held together by cohesin complexes on both chromosome arms and at the centromeres. The cohesin complexes on the arms are released by phosphorylation in prophase and prometaphase by the Plk1 and Aurora kinases, but cohesin at the centromeres is protected by the Shugoshin protein. Cohesin at the centromeres can be removed when the separase enzyme is activated after its inhibitor, securin, is degraded. Separase cleaves the Scc1 component of the cohesin complex.

committed to exit from mitosis through cyclin-B–Cdk1 phosphorylating the APC/C, to generate the form of the APC/C bound by Cdc20. By contrast, Cdh1 is able to bind and activate unphosphorylated APC/C (reviewed in [7]). Therefore, to keep Cdh1 from prematurely activating the APC/C in G2 phase, it is kept inactive by phosphorylation by G2 cyclin–CDK activity [18,19]. In animal cells, Cdh1 is also sequestered and inactivated by Rca1 in Drosophila or its vertebrate homolog Emi1 (Figure 2) [20–22]. Rca1 and Emi1 are cysteine-rich F-box proteins that bind to the substrate-binding site of Cdh1 and apparently regulate Cdh1 in an SCF-independent manner. A second member of the family in vertebrates is called Emi2 or Xerp1 and has an important role in maintaining meiotic arrest in unfertilized eggs [23–26]. In vertebrate somatic cells, Emi1 is also reported to bind and inactivate Cdc20, and so a second condition required to activate the APC/C in early vertebrate mitosis might be the degradation of Emi1 in late prophase. Emi1 is first phosphorylated by the mitotic polo-like kinase [27,28] and subsequently ubiquitylated by SCFb-TrCP [29] (Figure 2). So, when is the APC/C activated? To know this, we must identify the earliest APC/C substrate. The mitotic cyclins were the first APC/C substrates to be characterized, and they get their name from their dramatic instability once www.sciencedirect.com

cells enter mitosis. On immunoblots or by following radiolabelled proteins, cyclin A always disappears before cyclin B, and, in highly synchronous invertebrate eggs, the disappearance of cyclin B correlates with anaphase. However, to understand how the degradation of a specific protein is controlled, the key event is when its degradation starts, not when the protein has disappeared. To this end, live-cell imaging of fusion proteins made between a protein of interest and green fluorescent protein (GFP) have proved very useful. The fusion proteins act as markers for the UPS-dependent destruction of the endogenous protein because the GFP tag is unfolded and degraded by the proteasome along with the protein to which it is attached. Thus, assaying the amount of fluorescence gives a measure of the amount of protein in the cell. Using this live-cell assay, the earliest APC/C substrate known to date is cyclin A [30,31]. Cyclin A begins to be degraded at, or just after, nuclear envelope breakdown. Degradation of the Nek2A kinase that regulates centrosomes separation and the HOXC10 transcription factor parallel that of cyclin A on immunoblots of synchronized cells [32,33]. The evidence that they are APC/C substrates is that all three proteins can be ubiquitylated by the APC/C in vitro, and all three

Review

TRENDS in Cell Biology

proteins contain D boxes necessary for their destruction, although the D boxes in cyclin A and Nek2A are more extended than the normal nine amino acids [31,32]. Furthermore, cyclin A is stabilized in Drosophila cells carrying a mutant version of Cdc20 (fizzy) [34,35]. At present, it is unclear whether each of these proteins has to be degraded so early in mitosis, although expressing an ectopic nondegradable version of cyclin A does prevent Drosophila embryo and mammalian tissueculture cells from initiating anaphase [30,31,35]. While APC/CCdc20 is busily ubiquitylating cyclin A, Nek2A and HOXC10, the chromosomes are attaching to the spindle, and, in somatic cells, the spindle-assembly checkpoint machinery is activated to prevent the cells from prematurely entering anaphase. This sets up an as-yetunresolved puzzle because most models of the checkpoint propose that it works by inactivating or sequestering Cdc20 to prevent it from interacting with the APC/C (reviewed in [36]). As yet, we have no clear idea how the APC/C can still recognize proteins in the presence of the checkpoint. It might be that substrates such as cyclin A are recognized very efficiently by the APC/C so that even a small amount of Cdc20 is sufficient to promote their degradation, or that they bind to Cdc20 in a manner that prevents the spindlecheckpoint proteins from inactivating Cdc20. Spindle checkpoint-dependent APC/C substrates The key event in mitosis is the removal of the anaphase and cytokinesis inhibitors only after all the sister chromatids have correctly attached to the spindle. These inhibitors are securin and cyclin B, respectively, both substrates of APC/CCdc20, and live-cell imaging has revealed that they begin to be degraded at the same time in human cells: when the last unattached kinetochore is captured by a spindle microtubule – that is, when the spindle checkpoint is inactivated [37,38]. Eliminating the spindle checkpoint in somatic cells advances cyclin B and securin destruction to begin at the same time as the destruction of cyclin A [38]. This means that the spindle checkpoint is an integral part of every mitosis in somatic cells; its elimination leads to aneuploidy and is inviable in animal cells [93]. Although the spindle checkpoint sets the timing for cyclin B and securin destruction in somatic cells, there is a lag between when the APC/C is activated and when cyclin B begins to be degraded, even in systems where there is no checkpoint, such as cleaving invertebrate embryos and frog egg extracts. This is obviously crucial to keeping the cell in mitosis long enough for the spindle properly to segregate sister chromatids. At present, we do not really know how this works. Emi1 is one of the proteins proposed to instigate the lag by binding and inhibiting Cdc20 from G2 phase until Emi1 is degraded by SCFb-TrCP. However, Emi1 begins to be degraded before the lag period, and the Drosophila ortholog of Emi1, Rca1, does not bind Cdc20 and is not required to regulate mitosis until Cdh1 (fizzy-related) is made after zygotic transcription is turned on [21]. A second timer has been proposed to be the Mad2 and BubR1 proteins independent of their role in the mitotic checkpoint [39]. This is based on siRNA studies where www.sciencedirect.com

Vol.16 No.1 January 2006

59

reducing Mad2 levels in mammalian cultured cells accelerates the average time from nuclear envelope breakdown to anaphase, regardless of whether there are unattached kinetochores. Reducing BubR1 and Mad2 together further accelerates progress to anaphase, indicating that Mad2 and BubR1 might together constitute the timer [39]. However, this model has yet to be tested in an embryonic system. Separating sister chromatids In the PtK1 rat kangaroo cell line, which has only 11 chromosomes, the capture of the last kinetochores can be accurately assayed and is consistently w23 min before anaphase [40]. This raises the questions of what sets the time from chromosome attachment to anaphase – perhaps the time taken to degrade securin and cyclin B – and how do sister chromatids separate synchronously at anaphase? Sister chromatids are held together by cohesin complexes that assemble during DNA replication (reviewed in [41]) (Figure 2). Cohesin complexes comprise a heterodimer of two structural maintenance of chromosomes (SMC) proteins and two sister chromatid cohesion (Scc) proteins that have been proposed to form a ring that holds sister chromatids together. In vertebrate cells, most of the cohesin complexes on the chromosome arms are removed in prophase by phosphorylation by the Plk and Aurora protein kinases [42,43], but the complexes at the centromeres are protected by the Shugoshin protein [44–47] (which was originally identified as the protein that protects centromeric complexes from cleavage in meiosis I in yeast, reviewed in [48]). All the yeast cohesin complexes, and centromeric animal cohesin complexes, are subsequently released by the separase protease that cleaves the Scc1 subunit of the complex (Figure 2). Securin both activates and inhibits separase, and a nondegradable version of securin prevents sister chromatid separation in the yeasts and animal cells [41]. Therefore, in principle, anaphase could simply be triggered by separase as it is released from inhibition by securin. However, it is difficult to reconcile the almost simultaneous separation of all sister chromatids with a gradual increase in free separase over, for example, the 20 min period in which securin is released in PtK1 cells. One explanation might be that there is a second signal that triggers the final separation of the chromosomes. In budding yeast, the Polo-like kinase ortholog, Cdc5, must phosphorylate Scc1 to make it a substrate for separase [49], and this might be coordinated on all chromosomes at the same time. However, this mechanism is not apparently conserved in animal cells, where separase does not require a phosphorylated form of cohesin as its substrate; instead, Plk1 helps cohesin subunits to disassemble in prophase [42]. An obvious candidate to impose synchrony in animal cells is the Shugoshin protein that protects centromeric cohesion, but, as yet, it is unclear how Shugoshin is inactivated and when. Shugoshin does appear to be a substrate of the APC/C, but the exact time at which it is degraded is not known [46]. Another mechanism that could provide a second signal has been observed in Xenopus egg extracts. Here, the separase protein is phosphorylated and inhibited by mitotic cyclin–CDK activity [50]; therefore, separase can

60

Review

TRENDS in Cell Biology

only cleave the centromeric cohesin complexes when mitotic cyclin–CDK activity falls below a certain threshold. This might also explain why securin is not essential in mammalian cells [51]. However, this threshold appears to be set fairly high because it requires 1.5–2-fold more cyclin–CDK activity than is present in normal mitotic cells to inhibit separase in Xenopus extracts [50] or in living mammalian cells [38]. Early in mitosis, cyclin-A–CDK activity could contribute to inhibiting separase, and DNA damage does delay anaphase in Drosophila embryos through stabilizing cyclin A [52]. However, as cyclin A is mostly degraded by metaphase, it cannot explain how sister chromatid cohesion can persist in cells lacking securin when they are arrested in metaphase. Recently, mouse cells with a non-phosphorylatable separase have been generated that also lack securin. These cells are viable and able properly to regulate sister chromatid separation except when mitosis is arrested in the presence of spindle poisons. Thus, mitotic cyclin–CDK activity is important to inhibit separase in a prolonged metaphase arrest, but the question of what regulates the timing of sister chromatid separation in a normal, unperturbed, mitosis remains open [95]. Cut phenotypes Although cyclin B and securin are degraded at the same time in animal cells and are under the same control by the spindle checkpoint, they are not dependent on one another. A nondegradable version of securin will prevent sister chromatid separation, but cyclin B proteolysis continues on schedule such that the cell attempts cytokinesis in the presence of unseparated chromosomes. This generates a ‘cut’ (cell untimely torn) phenotype in fission yeast cells, where the septum divides the nucleus [53]. In animal cells, the outcome is variable, depending on the position of the chromosomes in the cell; sometimes all the chromosomes are partitioned into only one daughter cell, in other cases, the cleavage furrow attempts to divide the chromosome mass and, in most cases, eventually regresses to generate one tetraploid cell. A large number of fission yeast ‘cut’ mutants have been isolated, and it is interesting to note that some of these are mutations in different APC/C subunits. This might be evidence that particular interaction domains on specific APC/C subunits have a role in recognizing different substrates. Alternatively, some substrates might be recognized at higher affinity than others. Spatial control of proteolysis Remarkably, the spindle-assembly checkpoint can rapidly inactivate cyclin B1 and securin proteolysis even after it has begun. Adding taxol or nocodazole to metaphase cells arrests them because the drugs re-impose the spindle checkpoint and turn off cyclin B1 and securin destruction [37,38]. When this experiment is performed with taxol in mammalian cells, there is a striking re-localization of cyclin B1 to the spindle poles and chromosomes, indicating that cyclin B1 might need to flux onto the spindle to be degraded [37]. In agreement with this, in the cell-division cycles of Drosophila embryos, although the bulk of the www.sciencedirect.com

Vol.16 No.1 January 2006

population of cyclin B1 is not degraded, a subpopulation around the spindle is destroyed, and this is required for cells to enter anaphase [54]. In these embryos, a wave of cyclin B1 proteolysis appears to begin at the centrosomes and spreads to the middle of the spindle. Furthermore, in mutant embryos where the centrosomes detach from the spindle, cyclin B1 is degraded on the detached centrosome but not on the rest of the spindle [55]. These experiments indicate that cyclin B1 ubiquitylation might be spatially regulated in cells, and the phenotype of embryos lacking the Drosophila UBC10-family member vihar combined with the localization of the vihar protein to the spindle and spindle poles indicate that some of the spatial control of cyclin B destruction might be orchestrated by vihar. Immunofluorescence studies in Drosophila and mammalian cells have revealed that the APC/C is localized to the spindle, in particular to the spindle poles [12,56], and, in prophase and pro-metaphase, to unattached kinetochores [56]. Possibly, the ubiquitylation of APC/CCdc20 substrates is spatially regulated to facilitate the close coupling between the spindle checkpoint and the APC/C (reviewed in [57]). Leaving mitosis In somatic cells, the decline in cyclin-B–CDK activity allows the APC/C to bind Cdh1. In budding yeast, the Cdc14 phosphatase is responsible for dephosphorylating Cdh1, but it is unclear whether this holds true in animal cells. The result of binding Cdh1 is that the APC/C now recognizes a wider set of substrates: those with D boxes and those with KEN boxes. One of these substrates is Cdc20 itself [58], meaning that there is a complete switch from APC/CCdc20 to APC/CCdh1, and one consequence of this is that the spindle-assembly checkpoint machinery can no longer turn off the APC/C. The proteins targeted by APC/CCdh1 include regulatory proteins such as the mitotic kinases, and geminin, an inhibitor of DNA replication, whose ubiquitylation – but not necessarily destruction – allows cells to re-license origins of replication as they reenter interphase [59]. They also include proteins that are functional components of mitosis- or cytokinesis-specific structures such as the mitotic spindle, cytokinetic furrow [60] and kinetochores, which must be disassembled to return the cell to its interphase state. The mitotic regulators targeted by APC/CCdh1 include Cdc5/Plk1 and the Aurora A kinase [61–64]. Live-cell imaging reveals that these proteins are degraded at different times in anaphase, indicating that there are further controls on the timing of when Cdh1 can recognize its substrates [63]. For Aurora A, this might be through modification of a second motif, the A box or D-boxactivating domain, that is required for Aurora A to be ubiquitylated and whose phosphorylation inhibits destruction in vitro [62,64]. Both these protein kinases can also be inactivated by alternative pathways such as dephosphorylation or, for Aurora A, dissociation from its activating partner TPX2 mediated by the p97 AAAATPase, which is required for spindle disassembly in Xenopus extracts [65]. Thus, proteolysis is not essential to inactivate them, but it does appear to promote efficient mitotic exit. For example, a non-degradable version of Plk1

Review

TRENDS in Cell Biology

perturbs cytokinesis and interferes with coordination between the position of the cleavage furrow and the mitotic spindle [63]. Indeed, it appears that none of the APC/CCdh1 substrates must be degraded for cells to exit from mitosis [66]; the most profound effects in cells lacking Cdh1 are on the regulation of events and decisions in G1 phase, in maintaining quiescence [67] and in post-mitotic cells [7]. Degrade cyclin B? No FEAR! When cells are unable to degrade cyclin B, they are unable to exit from mitosis because they cannot inactivate Cdk1. In animal cells, a non-degradable form of cyclin B usually blocks cells in anaphase [35,68], but low levels block in telophase and high levels in metaphase – probably through inhibiting separase [38]. In the Drosophila embryo, one effect of high levels of a non-degradable cyclin B is to prevent passenger proteins such as the Aurora B kinase and its INCENP partner from leaving the kinetochore and binding to the central spindle [69]. Thus, there does not appear to be any one crucial substrate that is phosphorylated by cyclin-B–Cdk1 that prevents exit from mitosis and cytokinesis. Rather, there are likely to be many different substrates with different affinities for the kinase, whose dephosphorylation is required for different aspects of chromosome movement, cytokinesis, spindle disassembly, chromosome decondensation and nuclear reformation. These observations also indicate that there is unlikely to be a positive-feedback loop driving animal cells out of mitosis once cyclin–CDK activity has dropped below a threshold. This is in contrast to budding yeast, where the Cdc14 phosphatase drives cells out of mitosis by dephosphorylating Cdh1 and stimulating the transcription of the Sic1 CDK inhibitor (reviewed in [70]). The difference between the systems is likely to arise because budding yeast cells have a problem to solve that is not shared by animal cells. Budding yeast set up the plane of division – the bud neck – before they set up their mitotic apparatus. Therefore, the cells have to position one of the poles of the spindle correctly in the bud, and, to achieve this, they must maintain the cell in mitosis after sister chromatid separation when the spindle elongates in anaphase. To this end, there are two waves of cyclin B (Clb2) destruction in budding yeast [71]. The first wave is driven by APC/ CCdc20 in metaphase/anaphase, and a second, more profound, destruction is triggered by Cdc14 and driven by APC/CCdh1 at the end of anaphase B, accompanied by an increase in the mitotic cyclin–CDK inhibitor Sic1. By contrast, animal cells set up the plane of division only after chromosomes have defined the metaphase plate, meaning that securin and cyclin B can be degraded together because the chromatids will segregate away from the incoming cytokinetic furrow. To ensure that budding yeast cells do not exit from mitosis until the spindle is properly positioned between the mother cell and the bud, the activation of APC/CCdh1 and consequent destruction of the remaining mitotic cyclin, Clb2, is regulated by the FEAR and MEN pathways. However, as this problem is unique to cells that divide by budding, the fission yeast homologs of the MEN pathway control the septation initiation network (SIN pathway) that controls septation but not cyclin www.sciencedirect.com

Vol.16 No.1 January 2006

61

destruction and mitotic exit (reviewed in [70]). Moreover, animal cells lack several of the MEN components, and those that are present appear to be required for abscission – the final cut that separates daughter cells that is the very last event in cytokinesis – and not for exit from mitosis. Concluding remarks Ubiquitin-mediated proteolysis mediated by the APC/C is a rapid and decisive mechanism to control progress through mitosis, to aid in cytokinesis and to return cells to their interphase state. In budding yeast, cytokinesis requires another ubiquitin ligase – SCFGrr1 – that is recruited to the region of the mother–bud neck where it binds and degrades the Hof1/Cyk2 protein to allow the efficient contraction of the actomyosin ring [72]. Genetic screens in the yeasts and C. elegans have indicated that other ubiquitin ligases might be also be involved in regulating mitosis [73,74], although in each case it is important to determine whether these are direct effects or the consequences of entering mitosis with damaged or unreplicated DNA. In animal cells, conditional knockouts of the core APC subunits, APC2 and APC11, have revealed an important role for these proteins in maintaining cells in their quiescent state [94], and recent evidence from invertebrate systems has indicated a role in synaptic plasticity. In mitosis, the APC/C has the crucial role in selecting the right substrate at the right time, in part through associating with different WD40 proteins, but elucidating exactly how it selects its substrates and how it responds to the spindle checkpoint is essential to a proper understanding of how mitosis is regulated. Acknowledgements I sincerely apologize to all those whose work I have had to refer to through reviews. I am deeply indebted to all the members of my lab, past and present, for their dedication, partnership in science and many lively discussions. I am particularly thankful to Claire Acquaviva, Lori Clay, Paul Clute, Fay Cooke, Barbara Di Fiore, Nicole den Elzen, Anja Hagting, Mark Jackman, Catherine Lindon, Takahiro Matsusaka and Adam Walker for all their dedication and perseverance in analysing proteolysis in living cells. I am also very grateful to my colleagues at the Gurdon Institute and around the world, especially Mary Dasso, Bill Earnshaw, Iain Hagan, Alexy Khodjakov, Sally Kornbluth, Danny Lew, Erich Nigg, Jan Michael Peters, Conly Rieder and Ted Salmon for expert advice, stimulating discussions and so often for putting me right. Cancer Research UK, the MRC and the EU make research in my laboratory possible.

References 1 Pines, J. and Rieder, C.L. (2001) Re-staging mitosis: a contemporary view of mitotic progression. Nat. Cell Biol. 3, E3–E6 2 Nigg, E.A. (2001) Mitotic kinases as regulators of cell division and its checkpoints. Nat. Rev. Mol. Cell Biol. 2, 21–32 3 Elia, A.E. et al. (2003) Proteomic screen finds pSer/pThr-binding domain localizing Plk1 to mitotic substrates. Science 299, 1228–1231 4 Mimura, S. et al. (2004) Phosphorylation-dependent binding of mitotic cyclins to Cdc6 contributes to DNA replication control. Nature 431, 1118–1123 5 Hershko, A. and Ciechanover, A. (1998) The ubiquitin system. Annu. Rev. Biochem. 67, 425–479 6 Passmore, L.A. (2004) The anaphase-promoting complex (APC): the sum of its parts? Biochem. Soc. Trans. 32, 724–727 7 Peters, J.M. (2002) The anaphase-promoting complex: proteolysis in mitosis and beyond. Mol. Cell 9, 931–943

62

Review

TRENDS in Cell Biology

8 Carroll, C.W. et al. (2005) The APC subunit Doc1 promotes recognition of the substrate destruction box. Curr. Biol. 15, 11–18 9 Passmore, L.A. et al. (2003) Doc1 mediates the activity of the anaphase-promoting complex by contributing to substrate recognition. EMBO J. 22, 786–796 10 Carroll, C.W. and Morgan, D.O. (2002) The Doc1 subunit is a processivity factor for the anaphase-promoting complex. Nat. Cell Biol. 4, 880–887 11 Rudner, A.D. and Murray, A.W. (2000) Phosphorylation by Cdc28 activates the Cdc20-dependent activity of the anaphase-promoting complex. J. Cell Biol. 149, 1377–1390 12 Kraft, C. et al. (2003) Mitotic regulation of the human anaphasepromoting complex by phosphorylation. EMBO J. 22, 6598–6609 13 Vodermaier, H.C. et al. (2003) TPR subunits of the anaphasepromoting complex mediate binding to the activator protein CDH1. Curr. Biol. 13, 1459–1468 14 Vodermaier, H.C. (2001) Cell cycle: waiters serving the destruction machinery. Curr. Biol. 11, R834–R837 15 Oelschlaegel, T. et al. (2005) The yeast APC/C subunit Mnd2 prevents premature sister chromatid separation triggered by the meiosisspecific APC/C-Ama1. Cell 120, 773–788 16 Penkner, A.M. et al. (2005) Mnd2, an essential antagonist of the anaphase-promoting complex during meiotic prophase. Cell 120, 789–801 17 Vodermaier, H.C. (2004) APC/C and SCF: controlling each other and the cell cycle. Curr. Biol. 14, R787–R796 18 Zachariae, W. et al. (1998) Control of cyclin ubiquitination by CDKregulated binding of Hct1 to the anaphase promoting complex. Science 282, 1721–1724 19 Lukas, C. et al. (1999) Accumulation of cyclin B1 requires E2F and cyclin-A-dependent rearrangement of the anaphase-promoting complex. Nature 401, 815–818 20 Dong, X. et al. (1997) Control of G1 in the developing Drosophila eye: rca1 regulates Cyclin A. Genes Dev. 11, 94–105 21 Grosskortenhaus, R. and Sprenger, F. (2002) Rca1 inhibits APCCdh1(Fzr) and is required to prevent cyclin degradation in G2. Dev. Cell 2, 29–40 22 Reimann, J.D. et al. (2001) Emi1 is a mitotic regulator that interacts with Cdc20 and inhibits the anaphase promoting complex. Cell 105, 645–655 23 Liu, J. and Maller, J.L. (2005) Calcium elevation at fertilization coordinates phosphorylation of XErp1/Emi2 by Plx1 and CaMK II to release metaphase arrest by cytostatic factor. Curr. Biol. 15, 1458–1468 24 Schmidt, A. et al. (2005) Xenopus polo-like kinase Plx1 regulates XErp1, a novel inhibitor of APC/C activity. Genes Dev. 19, 502–513 25 Tung, J.J. et al. (2005) A role for the anaphase-promoting complex inhibitor Emi2/XErp1, a homolog of early mitotic inhibitor 1, in cytostatic factor arrest of Xenopus eggs. Proc. Natl. Acad. Sci. U. S. A. 102, 4318–4323 26 Rauh, N.R. et al. (2005) Calcium triggers exit from meiosis II by targeting the APC/C inhibitor XErp1 for degradation. Nature, doi: 10. 1038/nature04093 27 Hansen, D.V. et al. (2004) Plk1 regulates activation of the anaphase promoting complex by phosphorylating and triggering SCFbetaTrCPdependent destruction of the APC Inhibitor Emi1. Mol. Biol. Cell 15, 5623–5634 28 Moshe, Y. et al. (2004) Role of Polo-like kinase in the degradation of early mitotic inhibitor 1, a regulator of the anaphase promoting complex/cyclosome. Proc. Natl. Acad. Sci. U. S. A. 101, 7937–7942 29 Margottin-Goguet, F. et al. (2003) Prophase destruction of Emi1 by the SCF(betaTrCP/Slimb) ubiquitin ligase activates the anaphase promoting complex to allow progression beyond prometaphase. Dev. Cell 4, 813–826 30 den Elzen, N. and Pines, J. (2001) Cyclin A is destroyed in prometaphase and can delay chromosome alignment and anaphase. J. Cell Biol. 153, 121–136 31 Geley, S. et al. (2001) APC/C-dependent proteolysis of human cyclin A starts at the beginning of mitosis and is not subject to the spindle assembly checkpoint. J. Cell Biol. 153, 137–148 32 Hames, R.S. et al. (2001) APC/C-mediated destruction of the centrosomal kinase Nek2A occurs in early mitosis and depends upon a cyclin A-type D-box. EMBO J. 20, 7117–7127 www.sciencedirect.com

Vol.16 No.1 January 2006

33 Gabellini, D. et al. (2003) Early mitotic degradation of the homeoprotein HOXC10 is potentially linked to cell cycle progression. EMBO J. 22, 3715–3724 34 Dawson, I.A. et al. (1995) The Drosophila cell cycle gene fizzy is required for normal degradation of cyclins A and B during mitosis and has homology to the CDC20 gene of Saccharomyces cerevisiae. J. Cell Biol. 129, 725–737 35 Sigrist, S. et al. (1995) Exit from mitosis is regulated by Drosophila fizzy and the sequential destruction of cyclins A, B and B3. EMBO J. 14, 4827–4838 36 Musacchio, A. and Hardwick, K.G. (2002) The spindle checkpoint: structural insights into dynamic signalling. Nat. Rev. Mol. Cell Biol. 3, 731–741 37 Clute, P. and Pines, J. (1999) Temporal and spatial control of cyclin B1 destruction in metaphase. Nat. Cell Biol. 1, 82–87 38 Hagting, A. et al. (2002) Human securin proteolysis is controlled by the spindle checkpoint and reveals when the APC/C switches from activation by Cdc20 to Cdh1. J. Cell Biol. 157, 1125–1137 39 Meraldi, P. et al. (2004) Timing and checkpoints in the regulation of mitotic progression. Dev. Cell 7, 45–60 40 Rieder, C.L. et al. (1994) Anaphase onset in vertebrate somatic cells is controlled by a checkpoint that monitors sister kinetochore attachment to the spindle. J. Cell Biol. 127, 1301–1310 41 Nasmyth, K. (2001) Disseminating the genome: joining, resolving, and separating sister chromatids during mitosis and meiosis. Annu. Rev. Genet. 35, 673–745 42 Waizenegger, I.C. et al. (2000) Two distinct pathways remove mammalian cohesin from chromosome arms in prophase and from centromeres in anaphase. Cell 103, 399–410 43 Losada, A. et al. (2002) Cohesin release is required for sister chromatid resolution, but not for condensin-mediated compaction, at the onset of mitosis. Genes Dev. 16, 3004–3016 44 Kitajima, T.S. et al. (2005) Human Bub1 defines the persistent cohesion site along the mitotic chromosome by affecting Shugoshin localization. Curr. Biol. 15, 353–359 45 McGuinness, B.E. et al. (2005) Shugoshin prevents dissociation of cohesin from centromeres during mitosis in vertebrate cells. PLoS Biol. 3, e86 46 Salic, A. et al. (2004) Vertebrate shugoshin links sister centromere cohesion and kinetochore microtubule stability in mitosis. Cell 118, 567–578 47 Tang, Z. et al. (2004) Human Bub1 protects centromeric sisterchromatid cohesion through Shugoshin during mitosis. Proc. Natl. Acad. Sci. U. S. A. 101, 18012–18017 48 Watanabe, Y. (2005) Sister chromatid cohesion along arms and at centromeres. Trends Genet. 21, 405–412 49 Alexandru, G. et al. (2001) Phosphorylation of the cohesin subunit Scc1 by Polo/Cdc5 kinase regulates sister chromatid separation in yeast. Cell 105, 459–472 50 Stemmann, O. et al. (2001) Dual inhibition of sister chromatid separation at metaphase. Cell 107, 715–726 51 Mei, J. et al. (2001) Securin is not required for cellular viability, but is required for normal growth of mouse embryonic fibroblasts. Curr. Biol. 11, 1197–1201 52 Su, T.T. and Jaklevic, B. (2001) DNA damage leads to a Cyclin A-dependent delay in metaphase-anaphase transition in the Drosophila gastrula. Curr. Biol. 11, 8–17 53 Yanagida, M. (1998) Fission yeast cut mutations revisited: control of anaphase. Trends Cell Biol. 8, 144–149 54 Huang, J. and Raff, J.W. (1999) The disappearance of cyclin B at the end of mitosis is regulated spatially in Drosophila cells. EMBO J. 18, 2184–2195 55 Wakefield, J.G. et al. (2000) Centrosomes have a role in regulating the destruction of cyclin B in early Drosophila embryos. Curr. Biol. 10, 1367–1370 56 Acquaviva, C. et al. (2004) The anaphase promoting complex/cyclosome is recruited to centromeres by the spindle assembly checkpoint. Nat. Cell Biol. 6, 892–898 57 Pines, J. and Lindon, C. (2005) Proteolysis: anytime, any place, anywhere? Nat. Cell Biol. 7, 731–735 58 Pfleger, C.M. and Kirschner, M.W. (2000) The KEN box: an APC recognition signal distinct from the D box targeted by Cdh1. Genes Dev. 14, 655–665

Review

TRENDS in Cell Biology

59 Li, A. and Blow, J.J. (2004) Non-proteolytic inactivation of geminin requires CDK-dependent ubiquitination. Nat. Cell Biol. 6, 260–267 60 Zhao, W.M. and Fang, G. (2005) Anillin is a substrate of APC/C that controls spatial contractility of myosin during late cytokinesis. J. Biol. Chem. 280, 33516–33524 61 Shirayama, M. et al. (1998) The Polo-like kinase Cdc5p and the WDrepeat protein Cdc20p/fizzy are regulators and substrates of the anaphase promoting complex in Saccharomyces cerevisiae. EMBO J. 17, 1336–1349 62 Littlepage, L.E. and Ruderman, J.V. (2002) Identification of a new APC/C recognition domain, the A box, which is required for the Cdh1dependent destruction of the kinase Aurora-A during mitotic exit. Genes Dev. 16, 2274–2285 63 Lindon, C. and Pines, J. (2004) Ordered proteolysis in anaphase inactivates Plk1 to contribute to proper mitotic exit in human cells. J. Cell Biol. 164, 233–241 64 Castro, A. et al. (2002) The D-Box-activating domain (DAD) is a new proteolysis signal that stimulates the silent D-Box sequence of Aurora-A. EMBO Rep. 3, 1209–1214 65 Cao, K. et al. (2003) The AAA-ATPase Cdc48/p97 regulates spindle disassembly at the end of mitosis. Cell 115, 355–367 66 Jacobs, H. et al. (2002) Completion of mitosis requires neither fzr/rap nor fzr2, a male germline-specific Drosophila Cdh1 homolog. Curr. Biol. 12, 1435–1441 67 Wirth, K.G. et al. (2004) Loss of the anaphase-promoting complex in quiescent cells causes unscheduled hepatocyte proliferation. Genes Dev. 18, 88–98 68 Wheatley, S.P. et al. (1997) CDK1 inactivation regulates anaphase spindle dynamics and cytokinesis in vivo. J. Cell Biol. 138, 385–393 69 Parry, D.H. et al. (2003) Cyclin B destruction triggers changes in kinetochore behavior essential for successful anaphase. Curr. Biol. 13, 647–653 70 Simanis, V. (2003) Events at the end of mitosis in the budding and fission yeasts. J. Cell Sci. 116, 4263–4275 71 Yeong, F.M. et al. (2000) Exit from mitosis in budding yeast: biphasic inactivation of the Cdc28-Clb2 mitotic kinase and the role of Cdc20. Mol. Cell 5, 501–511 72 Blondel, M. et al. (2005) Degradation of Hof1 by SCF(Grr1) is important for actomyosin contraction during cytokinesis in yeast. EMBO J. 24, 1440–1452 73 Michel, J.J. et al. (2003) A role for Saccharomyces cerevisiae Cul8 ubiquitin ligase in proper anaphase progression. J. Biol. Chem. 278, 22828–22837 74 Hermand, D. et al. (2003) Skp1 and the F-box protein Pof6 are essential for cell separation in fission yeast. J. Biol. Chem. 278, 9671–9677 75 Matsusaka, T. and Pines, J. (2004) Chfr acts with the p38 stress kinases to block entry to mitosis in mammalian cells. J. Cell Biol. 166, 507–516 76 Mikhailov, A. et al. (2004) Topoisomerase II and histone deacetylase inhibitors delay the G2/M transition by triggering the p38 MAPK checkpoint pathway. J. Cell Biol. 166, 517–526

Vol.16 No.1 January 2006

77 Yu, X. et al. (2005) Chfr is required for tumor suppression and Aurora A regulation. Nat. Genet. 37, 401–406 78 Kang, D. et al. (2002) The checkpoint protein Chfr is a ligase that ubiquitinates Plk1 and inhibits Cdc2 at the G2 to M transition. J. Cell Biol. 156, 249–259 79 Bothos, J. et al. (2003) The Chfr mitotic checkpoint protein functions with Ubc13-Mms2 to form Lys63-linked polyubiquitin chains. Oncogene 22, 7101–7107 80 Michael, W.M. and Newport, J. (1998) Coupling of mitosis to the completion of S phase through Cdc34-mediated degradation of Wee1. Science 282, 1886–1889 81 O’Connell, M.J. et al. (1997) Chk1 is a wee1 kinase in the G2 DNA damage checkpoint inhibiting cdc2 by Y15 phosphorylation. EMBO J. 16, 545–554 82 Lew, D.J. (2003) The morphogenesis checkpoint: how yeast cells watch their figures. Curr. Opin. Cell Biol. 15, 648–653 83 Ferguson, A.M. et al. (2005) Normal cell cycle and checkpoint responses in mice and cells lacking Cdc25B and Cdc25C protein phosphatases. Mol. Cell. Biol. 25, 2853–2860 84 Busino, L. et al. (2004) Cdc25A phosphatase: combinatorial phosphorylation, ubiquitylation and proteolysis. Oncogene 23, 2050–2056 85 Zur, A. and Brandeis, M. (2002) Timing of APC/C substrate degradation is determined by fzy/fzr specificity of destruction boxes. EMBO J. 21, 4500–4510 86 Burton, J.L. et al. (2005) Assembly of an APC-Cdh1-substrate complex is stimulated by engagement of a destruction box. Mol. Cell 18, 533–542 87 Burton, J.L. and Solomon, M.J. (2001) D box and KEN box motifs in budding yeast Hsl1p are required for APC- mediated degradation and direct binding to Cdc20p and Cdh1p. Genes Dev. 15, 2381–2395 88 Kraft, C. et al. (2005) The WD40 propeller domain of Cdh1 functions as a destruction box receptor for APC/C substrates. Mol. Cell 18, 543–553 89 Hilioti, Z. et al. (2001) The anaphase inhibitor Pds1 binds to the APC/C-associated protein Cdc20 in a destruction box-dependent manner. Curr. Biol. 11, 1347–1352 90 Yamano, H. et al. (2004) Cell cycle-regulated recognition of the destruction box of Cyclin B by the APC/C in Xenopus egg extracts. Mol. Cell 13, 137–147 91 Chan, G.K. et al. Kinetochore structure and function. Trends Cell Biol. 15, 589–598 92 Pinsky, B.A. and Biggins, S. (2005) The spindle checkpoint: tension versus attachment. Trends Cell Biol. 15, 486–493 93 Dobles, M. et al. (2000) Chromosome missegregation and apoptosis in mice lacking the mitotic checkpoint protein Mad2. C-ell 101, 635–645 94 Wirth, K.G. et al. (2004) Loss of the anaphase-promoting complex in quiescent cells causes unscheduled hepatocyte proliferation. Genes Dev. 18, 88–98 95 Huang, X. et al. (2005) Securin and separase phosphorylation act redundantly to maintain sister chromatid cohesion in mammalian cells. Mol. Biol. Cell 16, 4725–4732

ScienceDirect collection reaches six million full-text articles Elsevier recently announced that six million articles are now available on its premier electronic platform, ScienceDirect. This milestone in electronic scientific, technical and medical publishing means that researchers around the globe will be able to access an unsurpassed volume of information from the convenience of their desktop. The rapid growth of the ScienceDirect collection is due to the integration of several prestigious publications as well as ongoing addition to the Backfiles – heritage collections in a number of disciplines. The latest step in this ambitious project to digitize all of Elsevier’s journals back to volume one, issue one, is the addition of the highly cited Cell Press journal collection on ScienceDirect. Also available online for the first time are six Cell titles’ long-awaited Backfiles, containing more than 12,000 articles highlighting important historic developments in the field of life sciences.

www.sciencedirect.com www.sciencedirect.com

63