pmc logo image
Logo of mbcJournal URL: http://www.molbiolcell.org
Journal List > Mol Biol Cell > v.20(4); Feb 15, 2009

Formats:
Mol Biol Cell. 2009 February 15; 20(4): 1252–1267.
doi: 10.1091/mbc.E08-10-1047.
PMCID: PMC2642744
Copyright © 2009 by The American Society for Cell Biology
Systematic Analysis in Caenorhabditis elegans Reveals that the Spindle Checkpoint Is Composed of Two Largely Independent Branches
Anthony Essex,*† Alexander Dammermann,* Lindsay Lewellyn,*† Karen Oegema,*† and Arshad Desaicorresponding author*†
*Department of Cellular and Molecular Medicine, Ludwig Institute for Cancer Research, University of California, San Diego, La Jolla, CA 92093; and
†Biomedical Sciences Graduate Program, School of Medicine, University of California, San Diego, La Jolla, CA 92093
Kerry S. Bloom, Monitoring Editor
corresponding authorCorresponding author.
Address correspondence to: Arshad Desai (Email: abdesai/at/ucsd.edu).
Received October 21, 2008; Revised December 3, 2008; Accepted December 5, 2008.
Small right arrow pointing to: This article has been cited by other articles in PMC.
Abstract
Kinetochores use the spindle checkpoint to delay anaphase onset until all chromosomes have formed bipolar attachments to spindle microtubules. Here, we use controlled monopolar spindle formation to systematically define the requirements for spindle checkpoint signaling in the Caenorhabditis elegans embryo. The results, when interpreted in light of kinetochore assembly epistasis analysis, indicate that checkpoint activation is coordinately directed by the NDC-80 complex, the Rod/Zwilch/Zw10 complex, and BUB-1—three components independently targeted to the outer kinetochore by the scaffold protein KNL-1. These components orchestrate the integration of a core Mad1MDF-1/Mad2MDF-2-based signal, with a largely independent Mad3SAN-1/BUB-3 pathway. Evidence for independence comes from the fact that subtly elevating Mad2MDF-2 levels bypasses the requirement for BUB-3 and Mad3SAN-1 in kinetochore-dependent checkpoint activation. Mad3SAN-1 does not accumulate at unattached kinetochores and BUB-3 kinetochore localization is independent of Mad2MDF-2. We discuss the rationale for a bipartite checkpoint mechanism in which a core Mad1MDF-1/Mad2MDF-2 signal generated at kinetochores is integrated with a separate cytoplasmic Mad3SAN-1/BUB-3–based pathway.
INTRODUCTION
Kinetochores assemble on centromeric DNA to connect spindle microtubules to sister chromatids and enable their segregation (Cheeseman and Desai, 2008 blue right-pointing triangle). Improper segregation can generate aneuploid daughter cells, which in turn may promote apoptosis or tumorigenesis (Rajagopalan and Lengauer, 2004 blue right-pointing triangle). To prevent aneuploidy, a kinetochore-based signaling pathway called the spindle checkpoint monitors chromosome–microtubule attachments and inhibits anaphase onset until all chromosomes have successfully bioriented, i.e., the two sister chromatids have attached to spindle microtubules emanating from opposing spindle poles (Musacchio and Salmon, 2007 blue right-pointing triangle). The presence of even a single unattached kinetochore is sufficient to inhibit progression into anaphase in somatic cells (Rieder et al., 1995 blue right-pointing triangle).
Screens for budding yeast mutants unable to arrest in the presence of microtubule-depolymerizing drugs identified as mitotic arrest deficient (Mad)1, Mad2, and Mad3 and budding uninhibited by benzimidazole (Bub)1 and Bub3 as molecular components of the checkpoint (Hoyt et al., 1991 blue right-pointing triangle; Li and Murray, 1991 blue right-pointing triangle). Mps1, a kinase essential for spindle pole body duplication, was subsequently also shown to be required for the checkpoint (Weiss and Winey, 1996 blue right-pointing triangle). Vertebrates and flies have additional proteins essential for checkpoint signaling, including Rod, Zwilch, and Zw10 (RZZ), which copurify as a complex and are interdependent for their kinetochore localization (Williams et al., 2003 blue right-pointing triangle; Buffin et al., 2005 blue right-pointing triangle; Karess, 2005 blue right-pointing triangle; Kops et al., 2005 blue right-pointing triangle), and the kinesin-like motor protein CENP-E (Abrieu et al., 2001 blue right-pointing triangle). Another difference between vertebrates and yeast is that the Mad3-like vertebrate protein BubR1 contains a C-terminal Bub1-like kinase domain (Murray and Marks, 2001 blue right-pointing triangle). Localization interdependencies, turnover dynamics, and biochemical interactions among the checkpoint proteins have been primarily studied in vertebrates and yeast and indicate that Bub1 is at the top of the checkpoint protein kinetochore localization hierarchy (Sharp-Baker and Chen, 2001 blue right-pointing triangle; Gillett et al., 2004 blue right-pointing triangle; Johnson et al., 2004 blue right-pointing triangle; Meraldi et al., 2004 blue right-pointing triangle; Rischitor et al., 2007 blue right-pointing triangle) and that downstream components such as Mad2 are rapidly exchanging at unattached kinetochores to communicate the checkpoint signal to the cytoplasm (Musacchio and Salmon, 2007 blue right-pointing triangle).
Checkpoint activation delays sister chromatid separation and mitotic exit by preventing the anaphase-promoting complex/cyclosome (APC/C), an E3-ubiquitin ligase, from inducing the destruction of securin and cyclin B (Peters, 2002 blue right-pointing triangle; Yu, 2002 blue right-pointing triangle). The checkpoint sequesters or inhibits Cdc20 (Hwang et al., 1998 blue right-pointing triangle; Kim et al., 1998 blue right-pointing triangle), which is essential for APC/C activation and substrate recognition (Yu, 2007 blue right-pointing triangle). The precise mechanism of Cdc20 inhibition by the checkpoint is a current topic of investigation. Recent structural and in vitro studies have shown that a kinetochore-bound Mad1–Mad2 complex interacts with free Mad2 and modifies its conformation to make it a more potent inhibitor of APC-Cdc20 (Sironi et al., 2002 blue right-pointing triangle; Luo et al., 2004 blue right-pointing triangle; De Antoni et al., 2005 blue right-pointing triangle; Vink et al., 2006 blue right-pointing triangle; Mapelli et al., 2007 blue right-pointing triangle; Yang et al., 2008 blue right-pointing triangle). However, Mad2 is unlikely to be the sole Cdc20 inhibitor. BubR1 has been shown to directly bind Cdc20 and subunits of the APC/C (Tang et al., 2001 blue right-pointing triangle; Sironi et al., 2002 blue right-pointing triangle). Bub1 has also been shown to bind and phosphorylate Cdc20 (Tang et al., 2004a blue right-pointing triangle). Finally, a complex named mitotic checkpoint complex containing BubR1 (Mad3 in yeast and worms), Bub3, Mad2 and Cdc20 that displays much higher APC/C inhibitory activity than purified Mad2 in vitro has been purified from HeLa cells as well as budding yeast (Hardwick et al., 2000 blue right-pointing triangle; Fraschini et al., 2001 blue right-pointing triangle; Sudakin et al., 2001 blue right-pointing triangle).
The early C. elegans embryo has emerged as an important model for studying kinetochore assembly and function. In vivo assembly epistasis analysis has comprehensively defined the relationships between kinetochore constituents, including proteins that direct assembly of centromeric chromatin (Maddox et al., 2007 blue right-pointing triangle) and proteins that provide the core microtubule binding activity of the kinetochore (Desai et al., 2003 blue right-pointing triangle; Cheeseman et al., 2004 blue right-pointing triangle, 2006 blue right-pointing triangle). These studies revealed a central role for the scaffold-like protein KNL-1 in outer kinetochore assembly, including the targeting of Bub1, the upstream kinase involved in spindle checkpoint activation (Desai et al., 2003 blue right-pointing triangle). The role of KNL-1 family proteins in checkpoint signaling is conserved in vertebrates (Kittler et al., 2007 blue right-pointing triangle; Kiyomitsu et al., 2007 blue right-pointing triangle). A delay in mitosis after treatment with microtubule-depolymerizing drugs has been documented in the gonad and in embryos (Kitagawa and Rose, 1999 blue right-pointing triangle; Nystul et al., 2003 blue right-pointing triangle; Encalada et al., 2005 blue right-pointing triangle; Stein et al., 2007 blue right-pointing triangle; Tarailo et al., 2007 blue right-pointing triangle; Hajeri et al., 2008 blue right-pointing triangle), and spindle checkpoint proteins have been implicated in cessation of activity under anoxia (Nystul et al., 2003 blue right-pointing triangle) and starvation-induced arrest of germ cell precursors (Watanabe et al., 2008 blue right-pointing triangle).
Here, we develop a controlled monopolar spindle formation-based assay in the early C. elegans embryo to systematically analyze the relationship between kinetochore structure and checkpoint activation. Our results indicate that checkpoint activation is coordinately directed by three components—the NDC-80 complex, the Rod/Zwilch/Zw10 complex, and BUB-1—that are targeted independently of one another by the outer kinetochore scaffold protein KNL-1. Mad3SAN-1, unlike the other checkpoint proteins, does not enrich at unattached kinetochores. Surprisingly, a subtle (2.5-fold) increase in Mad2MDF-2 levels can bypass the requirement of Mad3SAN-1 as well as BUB-3 for checkpoint activation. We propose that a core Mad1MDF-1/Mad2MDF-2 signal generated at kinetochores is integrated with a largely independent cytoplasmic Mad3SAN-1/BUB-3–based signal to achieve APC/C inhibition.
MATERIALS AND METHODS
Strains and Culture Conditions
All C. elegans strains were maintained at 20°C. Strain genotypes are listed in Table 1. The strains OD108 (expressing a GFP fusion with MDF-2), OD109 (expressing a GFP fusion with SAN-1), and OD133 (expressing a GFP fusion with BUB-3) were all generated by cloning the coding (BUB-3 and MDF-2) or genomic (SAN-1) sequences into the Spe1 site of pIC26 (Cheeseman et al., 2004 blue right-pointing triangle) and by integrating the constructs into DP38 [unc-119 (ed3)] by ballistic bombardment (Praitis et al., 2001 blue right-pointing triangle) with a PDS-1000/He Biolistic Particle Delivery System (Bio-Rad, Hercules, CA). Fluorescence intensity measurements in the nuclear region during early prometaphase (immediately after nuclear envelope breakdown [NEBD]) in the AB cell indicate that the GFP::Mad3SAN-1 and GFP::BUB-3 proteins are expressed at similar levels (mean ± SD in arbitrary units: 100 ± 16 [n = 8] for GFP::Mad3SAN-1 and 81 ± 22 [n = 9] for GFP-BUB-3) and that the GFP::Mad2MDF-2 protein is expressed at an approximately threefold higher level relative to the other two (300 ± 44 [n = 16]). The strain RB1391 [san-1(ok1580) I; referred to as Mad3san-1Δ] was obtained from the CGC. The strain AG170 was a generous gift from the laboratory of Dr. A. Golden. Two-color strains were constructed by mating as described previously (Green et al., 2008 blue right-pointing triangle).
Table 1.
Table 1.
Worm strains used in this study
RNA Interference (RNAi)
Double-stranded RNA (dsRNA) was prepared as described previously (Oegema et al., 2001 blue right-pointing triangle). Oligonucleotides used for dsRNA production are listed in Table 2. L4 worms were injected with dsRNA and incubated for 45–48 h at 20°C. For double depletions, dsRNAs were mixed to obtain equal concentrations of >0.75 mg/ml for each RNA. Western blots were performed as described previously (Desai et al., 2003 blue right-pointing triangle).
Table 2.
Table 2.
Double-stranded RNAs used in this study
Microscopy
All images for the timing assays and immunofluorescence were acquired on a DeltaVision deconvolution microscope (Applied Precision, Issaquah, WA) equipped with a CoolSNAP charge-coupled device camera (Roper Scientific, Trenton, NJ) at 20°C. Z-sections were acquired at 2-μm steps by using a 100×, 1.3 numerical aperture (NA) Olympus U-Planapo objective with 2 × 2 binning and a 480 × 480 pixel area at 20-s intervals, and each exposure was 100 ms. Z-stacks were projected and imported into MetaMorph (Molecular Devices, Sunnyvale, CA) to rotate and scale images. Immunofluorescence was performed as described previously (Oegema et al., 2001 blue right-pointing triangle; Desai et al., 2003 blue right-pointing triangle). Polyclonal antibodies against BUB-1, BUB-3 (amino acids 189-329), ZwilchZWL-1 (amino acids 1-200), and Mad2MDF-2 (splice variant Y39A2AR.30A amino acids 2-203) were generated as described previously (Oegema et al., 2001 blue right-pointing triangle; Desai et al., 2003 blue right-pointing triangle). All images acquired using a specific strain or specific antibody were scaled identically.
For GFP::BUB-3, GFP::Mad3SAN-1, and GFP::Mad2MDF-2 localization, embryos were filmed using a spinning disk confocal mounted on an inverted microscope (TE2000-E; Nikon, Tokyo, Japan) equipped with a 60 × 1.4 NA Plan Apochromat lens (Nikon), a krypton-argon 2.5-W water-cooled laser (Spectra Physics, San Jose, CA), and an electron multiplication back-thinned charge-coupled device camera (iXon; Andor Technology, Belfast, Ireland). Acquisition parameters, shutters, and focus were controlled by MetaMorph software (MDS Analytical Technologies, Winnersh, United Kingdom). Then, 5 × 1 μm RFP/GFP z-series with no binning and a single central reference differential interference contrast (DIC) image with no binning were collected every 20 s. Exposures were 300 ms for both green fluorescent protein (GFP) and red fluorescent protein (RFP), and 200 ms for DIC (laser power, 50%).
To specifically measure kinetochore-localized GFP::Mad2MDF-2, a subtraction approach (Dammermann et al., 2008 blue right-pointing triangle) was used. See Supplemental Figure 3 legend for details.
RESULTS
Controlled Monopolar Spindle Formation in the C. elegans Embryo Elicits a Cell Cycle Delay that Requires Conserved Spindle Checkpoint Components
To quantitatively monitor spindle checkpoint signaling in C. elegans embryos, we triggered checkpoint activation by generating monopolar spindles. In C. elegans, RNAi-mediated depletion of proteins required for centriole duplication results in a bipolar first division, which serves as a useful internal control, followed by subsequent monopolar divisions (Figure 1Figure 1.A; O'Connell et al., 2001 blue right-pointing triangle). Monopolar spindles have both unattached kinetochores and kinetochores not under tension and have been shown to activate the checkpoint in other organisms (Kapoor et al., 2000 blue right-pointing triangle). This approach avoids drug treatments, which are difficult due to the impermeable eggshell surrounding the embryos.
Figure 1.
Figure 1.
Figure 1.
Controlled monopolar spindle formation in C. elegans embryos results in a spindle checkpoint-mediated cell cycle delay. (A) Selected frames from time-lapse sequences of the first (P0), second (AB), and third (P1) divisions of embryos expressing GFP::histone (more ...)
We quantified the time from NEBD to chromosome decondensation (DCON) in embryos expressing GFP-histone H2b (to mark the chromosomes) and GFP-γ-tubulin (to mark the spindle poles). NEBD was defined by diffusion of free GFP-histone H2b out of the nucleoplasm and DCON as the disappearance of fluorescent punctae throughout the decondensing chromatin (Figure 1Figure 1.B). Monopolar spindles were generated by depleting the kinase ZYG-1 or the centriole structural protein SAS-6 (Bettencourt-Dias and Glover, 2007 blue right-pointing triangle). In both control and centriole duplication-inhibited embryos, the timing of NEBD–DCON was unaltered in the first mitotic division. By contrast, the same interval in the subsequent monopolar mitotic divisions was significantly elongated in both the anterior AB cell (Figure 1Figure 1.C) and the posterior P1 cell (data not shown). In all subsequent experiments, we only present analysis of mitotic timing in the first embryonic division and in the AB cell.
To determine whether the delay in cells with monopolar spindles was due to spindle checkpoint activation, we codepleted the conserved checkpoint protein Mad2MDF-2. Mad2MDF-2 codepletion did not affect the timing of the first bipolar division (Figure 1Figure 1.C and Supplemental Movie S1), but it abolished the cell cycle delay triggered by monopolar spindle formation (Figure 1Figure 1., B and C, and Supplemental Movie S2). Mad2MDF-2 depletion on its own did not affect the NEBD–DCON interval in either division (Supplemental Figure S1A). Similar results were obtained for both ZYG-1– and SAS-6–depleted embryos, establishing that the delay in mitotic exit is due to the presence of monopolar spindles and not due to a specific role for the targeted proteins in cell cycle progression. We conclude that controlled generation of monopolar spindles elicits a Mad2MDF-2-dependent cell cycle progression delay in the C. elegans embryo.
The C. elegans homologues of proteins implicated in checkpoint signaling are indicated in Figure 1Figure 1.D together with the consequences of their RNAi-mediated depletion. C. elegans has a Mad3-like protein (Mad3SAN-1) instead of a BubR1-like kinase and lacks an Mps1-like kinase, which is also absent in other related nematodes with sequenced genomes. Unlike depletion of other checkpoint proteins, depletion of BUB-1, ROD-1, or ZwilchZWL-1 resulted in penetrant embryonic lethality, reflecting functions for these proteins in chromosome segregation in addition to their role in checkpoint signaling. Depletion of Zw10CZW-1 resulted in penetrant sterility consistent with a previously described nonmitotic function for Zw10 (independently of Rod and Zwilch) in membrane trafficking (Hirose et al., 2004 blue right-pointing triangle), which is required for oocyte production.
We next examined the consequences of depleting components of the spindle checkpoint pathway in the monopolar spindle assay. Individual depletions of each protein abolished the monopolar spindle-induced mitotic delay (Figure 1Figure 1.E). By contrast, none of the depletions affected the timing of the first bipolar division (Figure 1Figure 1.E). Abolishing checkpoint signaling by depletion of Mad1MDF-1 also did not alter kinetochore-spindle microtubule interactions, as assessed by quantitative analysis of spindle pole separation (Supplemental Figure S1B; Oegema et al., 2001 blue right-pointing triangle). We conclude that controlled monopolar spindle formation generates a reproducible spindle checkpoint-mediated cell cycle delay in the early C. elegans embryo.
Systematic Analysis Subdivides the Protein Constituents of the Kinetochore into Three Classes Based on Their Roles in Spindle Checkpoint Activation
The protein components of the C. elegans kinetochore can be partitioned into different functional groups. A set of three proteins (CENP-AHCP-3, CENP-CHCP-4, and KNL-2) form the centromeric chromatin foundation for kinetochore assembly (Buchwitz et al., 1999 blue right-pointing triangle; Moore and Roth, 2001 blue right-pointing triangle; Oegema et al., 2001 blue right-pointing triangle; Maddox et al., 2007 blue right-pointing triangle). The conserved KNL-1/Mis12 complex/Ndc80 complex (KMN) network assembles on this foundation to form the core microtubule binding site of the kinetochore (Desai et al., 2003 blue right-pointing triangle; Cheeseman et al., 2004 blue right-pointing triangle, 2006 blue right-pointing triangle). KNL-1 serves as a scaffold that recruits not only the microtubule-binding NDC-80 complex but also other outer kinetochore proteins such as the RZZ complex, the kinase BUB-1, the CENP-F–like proteins HCP-1/2, and the microtubule-binding protein CLASPCLS-2 (Desai et al., 2003 blue right-pointing triangle).
To investigate their role in spindle checkpoint activation, we systematically analyzed the consequences of depleting kinetochore components on the monopolar spindle-induced cell cycle delay. Because the chromosome missegregation associated with several of these depletions made chromosome decondensation difficult to score, we used an alternative method to time cell cycle progression by measuring the interval from NEBD to onset of cortical contractility (OCC) in a strain coexpressing mCherry-Histone H2b and a GFP-tagged plasma membrane marker (Figure 2Figure 2.B and Supplemental Movies S3 and S4). Cortical contractility is tightly linked to mitotic exit and is a frequently used visual marker in live imaging studies (Canman et al., 2000 blue right-pointing triangle; Kurz et al., 2002 blue right-pointing triangle). We defined OCC as the transition of the membrane from a roughly circular conformation to a rectangular conformation (in embryos with bipolar spindles) or to the appearance of membrane “blebs” (in embryos with monopolar spindles; Figure 2Figure 2.B, arrowheads). Using this assay, we confirmed that monopolar spindles trigger a Mad2MDF-2-dependent increase in the NEBD–OCC interval relative to controls (Figure 2Figure 2.C and Supplemental Movie S4).
Figure 2.
Figure 2.
Figure 2.
Systematic analysis comparing the position of components in the kinetochore assembly hierarchy to their role in checkpoint activation. (A) Summary of the kinetochore assembly pathway in C. elegans embryos. (B) Selected frames from time-lapse sequences (more ...)
Next, we depleted each of the kinetochore components on their own and in conjunction with ZYG-1 and measured the NEBD–OCC intervals for the first two mitotic divisions. None of the tested proteins affected the NEBD to OCC interval during the first bipolar mitotic division (Figure 2Figure 2.C and Supplemental Movie S3). By contrast, analysis of cell cycle timing in the AB cell partitioned the targeted kinetochore components into three classes. The first class (I), which includes CENP-AHCP-3, CENP-CHCP-4, KNL-3 (data not shown), KNL-1, and NDC-80, is composed of proteins required for the monopolar spindle-induced delay; depletions of proteins in this class did not result in a significant cell cycle delay relative to controls (Figure 2Figure 2.C). The second class (II), which includes MCAKKLP-7 and the nonessential kinetochore protein KBP-5 (Figure 2Figure 2.C; data not shown), was dispensable for the monopolar spindle-induced delay. The third class (III) includes proteins whose depletion induces a cell cycle delay on their own, regardless of whether spindles were bipolar or monopolar; HCP-1/2, which are functionally analogous to CENP-F in vertebrates (Moore and Roth, 2001 blue right-pointing triangle; Cheeseman et al., 2005 blue right-pointing triangle; Encalada et al., 2005 blue right-pointing triangle; Tarailo et al., 2007 blue right-pointing triangle; Hajeri et al., 2008 blue right-pointing triangle), fell into this class (Figure 2Figure 2.C). The delay in HCP-1/2–depleted embryos was abolished by Mad2MDF-2 or Mad3SAN-1 codepletion, but it was of lower magnitude compared with the delay induced by monopolar spindles (Figure 2Figure 2.C). Codepletion of ZYG-1 did not increase the delay resulting from HCP-1/2 depletion, indicating that in addition to performing a function that prevents checkpoint activation, HCP-1/2 also make a positive contribution that increases the magnitude of the checkpoint signal.
In addition to the systematic analysis of kinetochore proteins described above, we also analyzed whether the inner centromere-localized Aurora BAIR-2 kinase subunit of the chromosomal passenger complex or the putative single Shugoshin family protein SGO-1 in C. elegans (C33H5.15; Kitajima et al., 2005 blue right-pointing triangle) are required for checkpoint signaling. We did not observe abrogation of the monopolar-spindle induced cell cycle delay after inactivation of Aurora BAIR-2 by using a temperature-sensitive mutant allele (or707ts; Severson et al., 2000 blue right-pointing triangle; Supplemental Figure S2A) or after sgo-1(RNAi) (Supplemental Figure S2B).
When considered in light of the assembly hierarchy of the kinetochore (Figure 2Figure 2.A), the above-mentioned data confirm that checkpoint signaling requires a core kinetochore scaffold. In addition, the results suggest that recruitment of three different components (the NDC-80 complex, the RZZ complex, and BUB-1) by KNL-1 is critical for checkpoint activation.
Checkpoint Signaling Status after Inhibition of the Three Classes of Kinetochore Constituents Correlates with GFP::Mad2MDF-2 Enrichment at Unattached Kinetochores
Checkpoint activation correlates with the enrichment of specific components of the pathway, most prominently Mad2, on unattached kinetochores (Musacchio and Salmon, 2007 blue right-pointing triangle). This enrichment is thought to reflect the local kinetochore-catalyzed reaction that generates the inhibitor of the APC/C. To correlate Mad2 recruitment with the functional analysis of checkpoint signaling, we generated a strain stably coexpressing GFP::Mad2MDF-2 and mCherry-Histone H2b. In the early mitotic divisions of control embryos, GFP::Mad2MDF-2 fluorescence is detected at the nuclear envelope/nucleoplasm beginning in prophase. After NEBD, GFP::Mad2MDF-2 remains present as a “cloud” of diffuse fluorescence surrounding the chromatin until anaphase onset, at which point it rapidly dissipates (Figure 3Figure 3.A and Supplemental Movie S5). Thus, no significant kinetochore localization of GFP::Mad2MDF-2 is observed in control embryos. In embryos depleted of ZYG-1 or SAS-6, GFP::Mad2MDF-2 localization was indistinguishable from controls during the first bipolar mitotic division (data not shown). However, during the second monopolar division, GFP::Mad2MDF-2 accumulated on the away-from-pole side of the chromatin after NEBD, reaching its peak intensity within 2 min (Figure 3Figure 3., A and C, and Supplemental Movie S5) followed by decay of the signal. Thus, the accumulation of GFP::Mad2MDF-2 at kinetochores correlates with functional checkpoint signaling.
Figure 3.
Figure 3.
Figure 3.
The accumulation of GFP::Mad2MDF-2 on chromosomes associated with monopolar spindles correlates with checkpoint activation. (A) Selected frames from time-lapse sequences of embryos expressing GFP::Mad2MDF-2 and mCherry::H2B that have normal bipolar spindles (more ...)
Immunoblotting indicated that the GFP::Mad2MDF-2 transgene was expressed at ~1.5 times the level of endogenous Mad2MDF-2 (Figure 3Figure 3.B) and that it caused a monopolar spindle-induced delay in the Mad2MDF-2 deletion strain mdf-2(tm2190) (Supplemental Figure S3A). GFP::Mad2MDF-2 localization was qualitatively similar on monopolar spindles generated in the deletion mutant strain. We also observed partial rescue of the variable and low brood size phenotype of the mdf-2(tm2190) strain (data not shown). Because the transgene is expressed under the pie-1 promoter (Green et al., 2008 blue right-pointing triangle), a lack of full rescue may reflect restricted expression.
We next analyzed the recruitment of GFP::Mad2MDF-2 to unattached kinetochores after depletion of the three classes of kinetochore components (Figure 3Figure 3.D). GFP::Mad2MDF-2 failed to accumulate on monopolar spindle-associated chromosomes after depletion of class I components, which are essential for checkpoint signaling. By contrast, depletion of class II components, which are not required for the monopolar spindle induced delay, did not affect the kinetochore accumulation of GFP::Mad2MDF-2. AuroraBAIR-2 inhibition, which does not abrogate the checkpoint-induced delay, also did not affect kinetochore accumulation of GFP::Mad2MDF-2 (Supplemental Figure S2C). Consistent with the fact that their depletion triggers the checkpoint even in the absence of monopolar spindles, depletion of the class III components HCP-1/2 induced GFP::Mad2MDF-2 accumulation both in the presence and absence of monopolar spindles (Figure 3Figure 3.D). These results support a strict correlation between the ability of unattached kinetochores to induce a cell cycle delay and their ability to recruit GFP::Mad2MDF-2, providing strong support for the model that the kinetochore scaffold-based local recruitment of Mad2MDF-2 is required to generate the signal that inhibits APC/C activity.
GFP::Mad2MDF-2 Accumulation at Kinetochores Is Unaffected By Depletion of Mad3SAN-1 and Is Reduced, but Not Eliminated, by Depletion of BUB-3
We next investigated GFP-Mad2MDF-2 localization at unattached kinetochores after depletion of conserved checkpoint pathway proteins (Figure 1Figure 1.D). We expected that because all of these proteins are required for the monopolar spindle-induced delay (Figure 1Figure 1.E), their depletion would eliminate GFP::Mad2MDF-2 localization, as observed for class I kinetochore components. This was indeed the case after depletion of Mad1MDF-1, BUB-1, or ROD-1 (Figure 3Figure 3.E). However, depletion of Mad3SAN-1 had no significant effect on GFP::Mad2MDF-2 localization at unattached kinetochores (Figure 3Figure 3.E, Supplemental Figure S3B, and Supplemental Movie S8). To confirm this result, we repeated the analysis using a viable null mutant of san-1 (san-1(ok1580); referred to subsequently as Mad3san-1Δ) that, similar to Mad3SAN-1 depletion by RNAi, is unable to generate a monopolar spindle-induced cell cycle delay (Supplemental Figure S4A). Even in the Mad3san-1Δ strain, we did not see a significant reduction in the accumulation of GFP::Mad2MDF-2 at unattached kinetochores compared with controls (Figure 3Figure 3.F and Supplemental Movie S9). Depletion of BUB-3 reduced the accumulation of GFP::Mad2MDF-2 but did not eliminate its kinetochore localization (Figure 3Figure 3.E and Supplemental Movie S7). Quantitative analysis of the peak GFP::Mad2MDF-2 fluorescence on chromosomes of monopolar spindles confirmed these observations (Supplemental Figure S3B). We conclude that Mad3SAN-1 and BUB-3 are not essential for the accumulation of GFP::Mad2MDF-2 at unattached kinetochores.
Mad3SAN-1 Does Not Enrich at Unattached Kinetochores When the Spindle Checkpoint Is Active
Mad3SAN-1 is not required for Mad2MDF-2 to accumulate at unattached kinetochores. To determine whether the converse is also true, we generated a strain coexpressing mCherry-Histone H2b and GFP::Mad3SAN-1. Expression of the Mad3SAN-1 transgene restored a monopolar spindle-induced cell cycle delay in the Mad3san-1Δ strain (Supplemental Figure S4A). In control embryos, GFP::Mad3SAN-1 showed diffuse localization in the vicinity of chromatin at prometaphase, which seemed significantly reduced by metaphase; there was no signal above background in other stages of mitosis (Figure 4Figure 4.A and Supplemental Movie S10). Surprisingly, we did not detect enrichment of GFP::Mad3SAN-1 at kinetochores of monopolar spindle-associated chromosomes; instead, we observed a diffuse localization pattern similar to that in control embryos with bipolar spindles (Figure 4Figure 4.B and Supplemental Movie S10). This localization pattern was unchanged in the absence of a wild-type Mad3SAN-1 allele (Figure 4Figure 4.C and Supplemental Movie S11) and was eliminated by RNAi-mediated depletion of Mad3SAN-1 (Supplemental Figure S4B). Codepletion of Mad2MDF-2, Mad1MDF-1, or BUB-1 had no significant effect on this diffuse localization; by contrast, in BUB-3–depleted embryos, the GFP signal was significantly diminished (Figure 4Figure 4.B and Supplemental Movies S12 and S13). The latter observation suggests that Mad3SAN-1 protein may be destabilized after depletion of BUB-3; we were unable to confirm this due to lack of a suitable anti-Mad3SAN-1 antibody. We conclude that Mad3SAN-1 is not coenriched on unattached kinetochores and that its stability may be dependent on BUB-3.
Figure 4.
Figure 4.
Figure 4.
Analysis of Mad3SAN-1 and BUB-3 localization in control and checkpoint-activated embryos. (A and B) Selected frames from time-lapse sequences of embryos coexpressing mCherry::Histone H2b and GFP::Mad3SAN-1 are shown for the indicated conditions. GFP::Mad3 (more ...)
BUB-3 Exhibits Basal Kinetochore Localization That Is Enriched at Unattached Kinetochores in a BUB-1–dependent but Mad1MDF-1/Mad2MDF-2-independent Manner
We next generated a strain coexpressing GFP::BUB-3 and mCherry-Histone H2b and performed experiments similar to those performed for Mad3SAN-1. Both endogenous BUB::3 (Supplemental Figure S5A) and GFP::BUB-3 (Figure 4Figure 4.D) were detected at kinetochores of control embryos. BUB::3 is first detectable on condensing chromosomes in late prophase and reaches maximal fluorescence intensity as paired lines on kinetochores at metaphase (Figure 4Figure 4.D and Supplemental Movie S14). It rapidly dissipates from kinetochores in early anaphase and is no longer detectable by late anaphase/early telophase. In cells with monopolar spindles, GFP::BUB-3 becomes enriched on the unattached kinetochores on the chromosomal face away from the pole (Figure 4Figure 4.E and Supplemental Movie S14). We conclude that BUB-3 has a basal kinetochore localization that is amplified when the checkpoint is active.
We next wanted to investigate the relationship between BUB-3 enrichment and Mad2MDF-2 enrichment at unattached kinetochores. We did not observe an effect of depleting either Mad1MDF-1 or Mad2MDF-2 on the enrichment of BUB-3 at unattached kinetochores (Figure 4Figure 4.E). We also did not observe an effect of depleting Mad3SAN-1 (Figure 4Figure 4.E), indicating that BUB-3 levels and localization are independent of Mad3SAN-1. By contrast, depletion of BUB-1 eliminated BUB-3 localization on both control bipolar (data not shown) and monopolar spindles (Figure 4Figure 4.E and Supplemental Movie S15); depletion of the RZZ complex subunit ROD-1, reduced the level of BUB-3 at unattached kinetochores, although localization was still evident (Figure 4Figure 4.E and Supplemental Movie S16), but depletion of the Ndc80 complex did not have a significant effect (Supplemental Figure S5C).
In converse experiments, BUB-3 depletion had no effect on BUB-1 (Figure 4Figure 4.F) or RZZ complex kinetochore localization (data not shown). Because BUB-3 depletion does not lead to embryonic lethality, whereas depletion of BUB-1 or ROD-1 leads to penetrant lethality, these results suggest that BUB-3 is not essential for the other chromosome segregation functions of BUB-1 and the RZZ complex. We conclude that BUB-3 exhibits basal kinetochore localization and accumulates at checkpoint signaling kinetochores in a BUB-1–dependent manner.
The NDC-80 Complex, the RZZ Complex, and BUB-1 Converge Downstream of KNL-1 to Direct the Accumulation of Mad2MDF-2 and BUB-3 and Checkpoint Activation
The NDC-80 complex, BUB-1, and the RZZ complex are all dependent on KNL-1 for their kinetochore localization (Desai et al., 2003 blue right-pointing triangle; Cheeseman et al., 2004 blue right-pointing triangle) and are all essential for checkpoint activation. Previous work has shown that NDC-80 complex is recruited to kinetochores independently of BUB-1 and the RZZ complex (Desai et al., 2003 blue right-pointing triangle; Gassmann et al., 2008 blue right-pointing triangle). Consistent with this, localization of BUB-3, which depends on BUB-1, is independent of the NDC-80 complex (Supplemental Figure S5C). We extended this analysis to show that BUB-1 and the RZZ complex also target to kinetochores independently of each other (Figure 4Figure 4.G). Thus, three components with distinct functions that are independently targeted to kinetochores by KNL-1 are integrated to direct Mad2MDF-2 and BUB-3 recruitment and checkpoint activation (Figure 4Figure 4.H). Interestingly, the kinetochore targeting of Mad2MDF-2 and BUB-3 reflect different, largely independent, pathways downstream of NDC-80, BUB-1, and the RZZ complex (Figure 4Figure 4.H). The kinetochore accumulation of Mad2MDF-2 (and presumably also Mad1MDF-1) requires NDC-80, BUB-1, and the RZZ complex and is enhanced by (but does not require) BUB-3. The kinetochore localization of BUB-3 requires BUB-1 and is enhanced by the presence of the RZZ complex, but it does not require Mad1MDF-1 or Mad2MDF-2. The existence of distinct pathways for the recruitment of Mad2MDF-2 and BUB-3 may facilitate the integration of different inputs during spindle checkpoint activation.
A Subtle Increase in Mad2MDF-2 Levels Bypasses the Requirement for Mad3SAN-1 and BUB-3 to Elicit a Kinetochore-dependent Monopolar Spindle-induced Cell Cycle Delay
In the strain expressing both endogenous and GFP::Mad2MDF-2, basal cell cycle timing was unaffected and monopolar spindles increased the NEBD–DCON interval (Figure 5Figure 5.A). Because this increase was dependent on Mad1MDF-1 (Figure 5Figure 5.A) and KNL-1 (data not shown), it reflects kinetochore-dependent signaling and excludes the trivial possibility that overexpression of Mad2MDF-2 is causing a cell cycle delay by general cytoplasmic inhibition of the APC/C. Strikingly, depletion of Mad3SAN-1 or BUB-3 did not eliminate the monopolar spindle-induced delay in this strain (Figure 5Figure 5.A). The same result was obtained after crossing the GFP::Mad2MDF-2 transgene into the Mad3san-1Δ strain background (Figure 5Figure 5.B). Importantly, the monopolar spindle-induced delay in the Mad3san-1Δ strain expressing the GFP::Mad2MDF-2 transgene required Mad1MDF-1 (Figure 5Figure 5.B), indicating that the Mad3SAN-1-independent delay was kinetochore dependent. We did not observe a bypass of the requirement for Mad1MDF-1 in the strain expressing GFP::BUB-3 (Supplemental Figure S5D), indicating that kinetochore-localized Mad1MDF-1/Mad2MDF-2 is indispensable for checkpoint signaling and that the bypass only works one-way.
Figure 5.
Figure 5.
Figure 5.
A subtle increase in Mad2MDF-2 levels bypasses the requirement for Mad3SAN-1 and BUB-3 in generating a monopolar spindle-induced cell cycle delay. (A and B) The mean NEBD to DCON interval in the AB cell is plotted for the indicated conditions/strains. (more ...)
Because GFP::Mad2MDF-2 was expressed from the transgene at 1.5 times the level of endogenous Mad2MDF-2 (Figure 3Figure 3.B), the total level of Mad2MDF-2 in the GFP::Mad2MDF-2 strain was ~2.5 times that in controls. These results suggest that a subtle increase in Mad2MDF-2 levels is sufficient to bypass the requirement for Mad3SAN-1 or BUB-3 to elicit a monopolar spindle-induced kinetochore-dependent cell cycle delay. If this were true, then restoring Mad2MDF-2 expression to endogenous levels should reverse this effect. To test this prediction, we used dsRNAs targeting GFP and ZYG-1 to simultaneously eliminate expression of the GFP-Mad2MDF-2 transgene and generate monopolar spindles. In this condition, a delay in the NEBD–DCON interval was observed that was not significantly different from ZYG-1 depletions alone (Figure 5Figure 5.A); the lack of any GFP signal on the chromosomes confirmed the efficacy of the GFP dsRNA (Figure 5Figure 5.C). When we then additionally codepleted Mad3SAN-1 or BUB-3, the monopolar spindle-induced delay in the NEBD-DCON interval was eliminated, indicating that the bypass of the requirement for Mad3SAN-1 and BUB-3 is dependent on the expression of the GFP-Mad2MDF-2 transgene (Figure 5Figure 5.A).
It is possible that the GFP::Mad2MDF-2 fusion is functionally altered in terms of APC/C inhibitory activity; however, neither basal cell cycle timing nor the extent of the kinetochore-dependent delay, both of which are sensitive to APC/C inhibition, were significantly affected by its presence. We conclude that a subtle increase in Mad2MDF-2 levels bypasses the requirement for Mad3SAN-1 and BUB-3 in kinetochore-dependent spindle checkpoint signaling.
DISCUSSION
Systematic Analysis of the Requirements for Spindle Checkpoint Activation Indicates a Central Role for the KMN Network
Here, we use controlled monopolar spindle formation to perform a systematic analysis of the requirements for checkpoint activation and Mad2MDF-2 recruitment in the C. elegans embryo. Our analysis comparing the classification of kinetochore proteins into functional groups based on phenotypic analysis and their position in the kinetochore assembly hierarchy to their role in checkpoint activation (Figure 6Figure 6.A), strongly supports the model that a kinetochore-triggered reaction is central to checkpoint activation. Specifically, all tested inhibitions that abrogate outer kinetochore assembly, including depletion of the centromeric histone CENP-A, prevented checkpoint activation. This result seems contradictory to checkpoint-dependent mitotic delays reported after inhibition of the centromeric histone CENP-A in Drosophila embryos and vertebrate cells (Regnier et al., 2005 blue right-pointing triangle; Blower et al., 2006 blue right-pointing triangle). It is possible that this reflects a difference in kinetochore assembly pathways between these systems. Alternatively, the high stability of CENP-A, which does not affect intrinsic turnover-independent RNAi-mediated depletion in C. elegans (Oegema and Hyman, 2006 blue right-pointing triangle) but does affect depletion efficacy in mutant Drosophila embryos that contain maternal product (Blower et al., 2006 blue right-pointing triangle) or in chicken cells where expression of a rescuing transgene is turned off (Regnier et al., 2005 blue right-pointing triangle), may account for the difference.
Figure 6.
Figure 6.
Figure 6.
Summary of the systematic analysis and a model for the bypass of Mad3SAN-1 and BUB-3 by a subtle increase in Mad2MDF-2 levels. (A) Summary of the requirements for checkpoint activation & GFP::Mad2MDF-2 accumulation at unattached kinetochores. (more ...)
Our systematic analysis identifies the core microtubule binding site of the kinetochore, the KMN network, as the most downstream stably kinetochore-localized protein group required for checkpoint activation. Specifically, three components with distinct functions that are independently targeted to kinetochores by the scaffold protein KNL-1—the NDC-80 complex, the RZZ complex, and BUB-1—are all critical for checkpoint activation. The NDC-80 complex directly associates with KNL-1 in the KMN network (Cheeseman et al., 2004 blue right-pointing triangle) and has been implicated in checkpoint signaling in other systems (McCleland et al., 2003 blue right-pointing triangle; Gillett et al., 2004 blue right-pointing triangle; Meraldi et al., 2004 blue right-pointing triangle). Based on work in human cells, the association of KNL-1 with BUB-1 is also likely to be direct (Kiyomitsu et al., 2007 blue right-pointing triangle). At least in C. elegans, where a Zwint-like intermediate protein bridging KNL-1 and the RZZ complex is not present, the RZZ complex may also directly associate with KNL-1. Together, these findings suggest an analogy to signaling networks in which different inputs integrated by scaffold proteins control signaling reactions. In the case of the spindle checkpoint, mechanical inputs from two independently targeted microtubule-binding activities of distinct functions, one resident in the Ndc80 complex and the second in the dynein/dynactin motor complex targeted by the RZZ complex, are likely integrated with BUB-1 in the context of the KNL-1 scaffold. Investigating the mechanism of integration will require developing a means to model the checkpoint reaction in vitro with a faithful facsimile of the activation base provided by the kinetochore—such an effort should be facilitated by the reconstitution of the C. elegans KMN network (Cheeseman et al., 2006 blue right-pointing triangle).
Core Checkpoint Pathway in C. elegans
In C. elegans, the core checkpoint pathway is simplified relative to other metazoan systems—no Mps1-like kinase exists and a Mad3, instead of a BubR1-like protein is present. It is possible that this simplification is linked to weakening of the checkpoint to accommodate the large diffuse kinetochores on the holocentric chromosomes of this organism. At least in the second embryonic division, which is the focus of our work, monopolar spindles are only able to extend the mitotic phase of the cell cycle twofold. Alternatively, the relatively small magnitude of the delay at the two-cell stage may reflect the large cytoplasm-to-nuclear ratio in the blastomeres at two-cell stage, consistent with the previously established relationship between the checkpoint signal efficacy and the nuclear-cytoplasmic ratio observed in Xenopus embryos (Minshull et al., 1994 blue right-pointing triangle). A fast-acting temperature-sensitive mutant that permits generation of monopolar spindles in later embryonic cell divisions, in which the cells are smaller, should help distinguish between these possibilities in future work.
Depletion of the core checkpoint proteins had no effect on basal cell cycle timing, but all were essential for the monopolar spindle-induced cell cycle delay. In addition, recruitment of Mad2MDF-2 to kinetochores was observed only when the checkpoint was activated—no significant accumulation at kinetochores was evident in control embryos. By contrast, both BUB-1 and BUB-3 localized to kinetochores even without checkpoint activation. This is consistent with the idea that BUB-1 provides an essential function in chromosome segregation that is required for embryonic viability. These results are generally analogous to what has been reported in budding yeast, in whichMad1 and Mad2 localization is only observed after drug-induced microtubule depolymerization and where Bub1 and Bub3 mutants are significantly more sick than Mad1 and Mad2 mutants (Warren et al., 2002 blue right-pointing triangle; Gillett et al., 2004 blue right-pointing triangle). Several noncheckpoint functions for Bub1 family kinases have been reported in yeast and vertebrates (Johnson et al., 2004 blue right-pointing triangle; Tang et al., 2004b blue right-pointing triangle; Kitajima et al., 2005 blue right-pointing triangle; Vaur et al., 2005 blue right-pointing triangle; Boyarchuk et al., 2007 blue right-pointing triangle), and at least one of these functions (targeting of CENP-F–like proteins HCP-1/2 to kinetochores) is conserved in C. elegans embryos (Encalada et al., 2005 blue right-pointing triangle; data not shown).
In addition to the core checkpoint proteins and the KMN network, we also observed a positive contribution to checkpoint signaling from HCP-1/2. Depletion of these proteins in cells with either bipolar or monopolar spindles triggers a Mad2MDF-2/Mad3SAN-1-dependent cell cycle delay, but the magnitude of this delay is less than that when HCP-1/2 are present. Synthetic genetic screens have identified HCP-1, but not HCP-2, as a contributor to checkpoint signaling in C. elegans (Tarailo et al., 2007 blue right-pointing triangle; Hajeri et al., 2008 blue right-pointing triangle)—our results extend these studies by showing that HCP-1/2 are not required for Mad2MDF-2 enrichment at kinetochores; HCP-1/2 may control the extent of Mad2MDF-2 accumulation or they may act at a different step that affects the potency of the inhibitory signal. Analogous conclusions have been made from studies on vertebrate CENP-F (for discussion, see Tarailo et al., 2007 blue right-pointing triangle; Hajeri et al., 2008 blue right-pointing triangle). Finally, MCAKKLP-7 was dispensable for both checkpoint activation and Mad2MDF-2 kinetochore localization. This result is in contrast to a previous report that MCAKKLP-7 is required for the checkpoint based on differential interference-contrast imaging of nocodazole-treated embryos (Encalada et al., 2005 blue right-pointing triangle). The reason for this discrepancy is currently unclear; we note that inhibition of kinesin-13s in vertebrates has not suggested an involvement in checkpoint activation (e.g., see (Manning et al., 2007 blue right-pointing triangle).
Mad3 Versus BubR1 in the Core Checkpoint Pathway
C. elegans is the only metazoan analyzed to date that lacks a BubR1-like kinase and instead has a truncated Mad3-like protein. An interesting emerging pattern is that the presence of a BubR1-like kinase correlates with the presence of a CENP-E–like kinetochore-localized kinesin motor (Chan et al., 1999 blue right-pointing triangle; Abrieu et al., 2000 blue right-pointing triangle). Worms and fungi, which have Mad3 instead of BubR1, lack CENP-E. The described functional links between CENP-E and the BubR1 kinase during checkpoint signaling in vertebrates are consistent with this pattern (Mao et al., 2005 blue right-pointing triangle).
The most significant difference between Mad3SAN-1 in C. elegans and BubR1 in other metazoans is with respect to kinetochore localization. The BubR1-like proteins in Drosophila and vertebrates localize to kinetochores, whereas we find that a functional C. elegans GFP:Mad3SAN-1 does not. Interestingly, chromatin immunoprecipitation and microscopy failed to detect budding yeast Mad3 at kinetochores under spindle depolymerization conditions that significantly enriched Mad1 and Mad2 at kinetochores (Gillett et al., 2004 blue right-pointing triangle). This similarity suggests that Mad3-like proteins, compared with BubR1-like protein kinases, are not enriched at kinetochores and, by inference, act primarily in the cytoplasm/nucleoplasm. However, contrary to this suggestion, fission yeast Mad3 localizes to kinetochores (Millband and Hardwick, 2002 blue right-pointing triangle). Experiments in which the Mad3s are switched between the two yeasts and C. elegans may help define the signals that control Mad3 localization and elucidate its site of action with respect to checkpoint signaling. Whether the kinetochore localization of Mad3 in fission yeast or BubR1 in vertebrate cells is essential for checkpoint signaling has not been established. Recent studies in vertebrates are leading to the conclusion that, similar to our findings in C. elegans for Mad3SAN-1, the checkpoint signaling function of BubR1 is independent of kinetochores (Kulukian and Cleveland, personal communication); the kinetochore localization of BubR1 may contribute to a distinct noncheckpoint role in chromosome segregation (Lampson and Kapoor, 2005).
In C. elegans, Mad3SAN-1 and Mad2MDF-2 are both required for the monopolar spindle-induced cell cycle delay in the early embryo. However, that subtle overexpression of Mad2MDF-2 bypasses the requirement for Mad3SAN-1 as well as BUB-3 indicates that Mad2MDF-2 is functionally more important. Consistent with this idea, the developmental phenotypes associated with deletion of Mad3SAN-1 are significantly weaker than those resulting from mutations in Mad1MDF-1 and Mad2MDF-2, which lead to pronounced defects in germline development and embryo production (Kitagawa and Rose, 1999 blue right-pointing triangle; Stein et al., 2007 blue right-pointing triangle). We speculate that in the germline, the core Mad1–Mad2 mechanism may be up-regulated independently of Mad3 to protect against aneuploidy. It is also possible that, similar to meiosis in budding yeast (Shonn et al., 2003 blue right-pointing triangle), the Mad1–Mad2 mechanism may provide an additional function important for chromosome segregation. Further work on these two interacting branches of the checkpoint pathway in the context of developmental regulation may provide insight into both the basal checkpoint signaling mechanism and its adaptation in different contexts.
Mad1MDF-1/Mad2MDF-2 Versus Mad3SAN-1/BUB-3: Two Branches of the Checkpoint Signaling Pathway
The most interesting theme emerging from our systematic analysis was the partitioning of the kinetochore-dependent checkpoint signaling pathway into two largely independent branches. Mad2MDF-2 (and presumably also Mad1MDF-1) accumulate at kinetochores and, in the situation where Mad2MDF-2 levels are elevated, support kinetochore-dependent checkpoint activation independently of BUB-3 and Mad3SAN-1. Conversely, BUB-3 targets to and become enriched at kinetochores in the absence of Mad2MDF-2, although in this case no checkpoint signal is generated. The independence of Mad1–Mad2 kinetochore localization from Mad3SAN-1 is supported by work in yeast (Gillett et al., 2004 blue right-pointing triangle; Vanoosthuyse et al., 2004 blue right-pointing triangle) and by BubR1 depletion in human cells (Johnson et al., 2004 blue right-pointing triangle; Meraldi et al., 2004 blue right-pointing triangle). Although Mad3SAN-1 does not localize to kinetochores upon checkpoint activation, two lines of evidence support a functional link to BUB-3. First, subtle overexpression of Mad2MDF-2 bypassed depletion of either BUB-3 or Mad3SAN-1. Second, BUB-3 depletion resulted in a significant decrease in GFP-Mad3SAN-1 signal, which suggests that the protein may be destabilized—such an effect is typically observed for proteins that are associated with each other. Together, these results suggest that a core Mad1MDF-1-Mad2MDF-2 signaling mechanism, which involves conversion of the free “open” form of Mad2 (Mad2-O) to the Cdc20-inhibiting “closed” form (Mad2-C) by a kinetochore-bound Mad1–Mad2 complex (Figure 6Figure 6.B; Musacchio and Salmon, 2007 blue right-pointing triangle), cooperates with a Mad3SAN-1/BUB-3 dependent cytoplasmic mechanism to inhibit the APC/C (Figure 6Figure 6.B); under normal conditions neither mechanism is sufficient to induce a cell cycle delay. Consistent with this idea, a Bub3–BubR1 complex has been purified from human cells and suggested to inhibit APC/C activity on its own in a manner similar to Mad2 (Tang et al., 2001 blue right-pointing triangle). The bypass we document here suggests that elevating Mad2 levels enhances Mad2-C formation to a point where the Mad1–Mad2 mechanism is sufficient to induce a kinetochore-dependent cell cycle delay in the absence of the BUB-3/Mad3SAN-1 branch (Figure 6Figure 6.B). This result suggests that Mad2 levels are limiting for Mad2-C formation in vivo and may be tightly controlled to allow integration of the Mad2-C mechanism with Mad3/BUB-3.
In addition to functioning with Mad3SAN-1 in the cytoplasm, BUB-3 may also act at the kinetochore, because it does enrich there and its depletion reduces the ability of Mad2MDF-2 to enrich at kinetochores. Because both BUB-1 and Mad3 use a similar and mutually exclusive interaction mechanism to associate with BUB-3 (Wang et al., 2001 blue right-pointing triangle; Larsen et al., 2007 blue right-pointing triangle), it is tempting to speculate that there are two pools of BUB-3: a population that enriches at kinetochores complexed with BUB-1 and a population that associates with Mad3SAN-1 that acts cytoplasmically. It is unclear what effect there is, if any, of kinetochore cycling of BUB-3, presumably via its direct association with BUB-1. It is possible that kinetochore-cycled BUB-3 is modified to potentiate its association with Mad3SAN-1 in the cytoplasm, proving another kinetochore-dependent input.
It is interesting to speculate on why the checkpoint signaling pathway is organized into two interacting branches. One possibility is that synergy between the branches may confer a property to the checkpoint signaling circuit that satisfies its difficult-to-reconcile requirements for potency and lability (Nasmyth, 2005 blue right-pointing triangle). An attractive alternative possibility is that the Mad1–Mad2 and Mad3/BubR1 mechanisms provide independent inhibitory signals that are responsive to different states—lack of attachment for the Mad1–Mad2 mechanism and lack of tension for the Mad3/BubR1 mechanism. In support of the latter possibility, a Mad3 phosphorylation site targeted by the error correction kinase Aurora B was recently identified and shown to be specifically required for detecting a defect in tension but not in attachment in budding yeast (King et al., 2007 blue right-pointing triangle). The two branches may integrate these different inputs to control the stability of Cdc20, which is modulated by checkpoint activation (Pan and Chen, 2004 blue right-pointing triangle). Further work on the relationship between the Mad1–Mad2 and Mad3/BUB-3 branches may help provide insight into the reasons for this bipartite architecture of the spindle checkpoint pathway.
Supplementary Material
[Supplemental Materials]
Click here to view.
ACKNOWLEDGMENTS
We thank members of the Oegema and Desai laboratories for discussions and advice; Paul Maddox for help with imaging; Julien Espeut and Reto Gassmann for help with biochemistry; and Andy Golden, Ann Rose, and Pamela Padilla for strains and other reagents. This work was supported by a University of California, San Diego Genetics training grant (to A. E.), National Institutes of Health grant GM-074215 (to A. D.), and funding from the Ludwig Institute for Cancer Research (to A. D. and K. O.).
Abbreviations used:
DCONdecondensation
NEBDnuclear envelope breakdown
OCConset of cortical contractility.

Footnotes
This article was published online ahead of print in MBC in Press (http://www.molbiolcell.org/cgi/doi/10.1091/mbc.E08-10-1047) on December 24, 2008.
REFERENCES
  • Abrieu A., Kahana J. A., Wood K. W., Cleveland D. W. CENP-E as an essential component of the mitotic checkpoint in vitro. Cell. 2000;102:817–826. [PubMed]
  • Abrieu A., Magnaghi-Jaulin L., Kahana J. A., Peter M., Castro A., Vigneron S., Lorca T., Cleveland D. W., Labbe J. C. Mps1 is a kinetochore-associated kinase essential for the vertebrate mitotic checkpoint. Cell. 2001;106:83–93. [PubMed]
  • Bettencourt-Dias M., Glover D. M. Centrosome biogenesis and function: centrosomics brings new understanding. Nat. Rev. Mol. Cell Biol. 2007;8:451–463. [PubMed]
  • Blower M. D., Daigle T., Kaufman T., Karpen G. H. Drosophila CENP-A mutations cause a BubR1-dependent early mitotic delay without normal localization of kinetochore components. PLoS Genet. 2006;2:e110. [PubMed]
  • Boyarchuk Y., Salic A., Dasso M., Arnaoutov A. Bub1 is essential for assembly of the functional inner centromere. J. Cell Biol. 2007;176:919–928. [PubMed]
  • Buchwitz B. J., Ahmad K., Moore L. L., Roth M. B., Henikoff S. A histone-H3-like protein in C. elegans. Nature. 1999;401:547–548. [PubMed]
  • Buffin E., Lefebvre C., Huang J., Gagou M. E., Karess R. E. Recruitment of Mad2 to the kinetochore requires the Rod/Zw10 complex. Curr. Biol. 2005;15:856–861. [PubMed]
  • Canman J. C., Hoffman D. B., Salmon E. D. The role of pre- and post-anaphase microtubules in the cytokinesis phase of the cell cycle. Curr. Biol. 2000;10:611–614. [PubMed]
  • Chan G. K., Jablonski S. A., Sudakin V., Hittle J. C., Yen T. J. Human BUBR1 is a mitotic checkpoint kinase that monitors CENP-E functions at kinetochores and binds the cyclosome/APC. J. Cell Biol. 1999;146:941–954. [PubMed]
  • Cheeseman I. M., Chappie J. S., Wilson-Kubalek E. M., Desai A. The conserved KMN network constitutes the core microtubule-binding site of the kinetochore. Cell. 2006;127:983–997. [PubMed]
  • Cheeseman I. M., Desai A. Molecular architecture of the kinetochore-microtubule interface. Nat. Rev. Mol. Cell Biol. 2008;9:33–46. [PubMed]
  • Cheeseman I. M., MacLeod I., Yates J. R., 3rd, Oegema K., Desai A. The CENP-F-like proteins HCP-1 and HCP-2 target CLASP to kinetochores to mediate chromosome segregation. Curr. Biol. 2005;15:771–777. [PubMed]
  • Cheeseman I. M., Niessen S., Anderson S., Hyndman F., Yates J. R., 3rd, Oegema K., Desai A. A conserved protein network controls assembly of the outer kinetochore and its ability to sustain tension. Genes Dev. 2004;18:2255–2268. [PubMed]
  • Dammermann A., Maddox P. S., Desai A., Oegema K. SAS-4 is recruited to a dynamic structure in newly forming centrioles that is stabilized by the gamma-tubulin-mediated addition of centriolar microtubules. J. Cell Biol. 2008;180:771–785. [PubMed]
  • De Antoni A., et al. The Mad1/Mad2 complex as a template for Mad2 activation in the spindle assembly checkpoint. Curr. Biol. 2005;15:214–225. [PubMed]
  • Desai A., Rybina S., Muller-Reichert T., Shevchenko A., Hyman A., Oegema K. KNL-1 directs assembly of the microtubule-binding interface of the kinetochore in C. elegans. Genes Dev. 2003;17:2421–2435. [PubMed]
  • Encalada S. E., Willis J., Lyczak R., Bowerman B. A spindle checkpoint functions during mitosis in the early Caenorhabditis elegans embryo. Mol. Biol. Cell. 2005;16:1056–1070. [PubMed]
  • Fraschini R., Beretta A., Sironi L., Musacchio A., Lucchini G., Piatti S. Bub3 interaction with Mad2, Mad3 and Cdc20 is mediated by WD40 repeats and does not require intact kinetochores. EMBO J. 2001;20:6648–6659. [PubMed]
  • Gassmann R., et al. A new mechanism controlling kinetochore-microtubule interactions revealed by comparison of two dynein-targeting components: SPDL-1 and the Rod/Zwilch/Zw10 complex. Genes Dev. 2008;22:2385–2399. [PubMed]
  • Gillett E. S., Espelin C. W., Sorger P. K. Spindle checkpoint proteins and chromosome-microtubule attachment in budding yeast. J. Cell Biol. 2004;164:535–546. [PubMed]
  • Green R. A., Audhya A., Pozniakovsky A., Dammermann A., Pemble H., Monen J., Portier N., Hyman A., Desai A., Oegema K. Expression and imaging of fluorescent proteins in the C. elegans gonad and early embryo. Methods Cell Biol. 2008;85:179–218. [PubMed]
  • Hajeri V. A., Stewart A. M., Moore L. L., Padilla P. A. Genetic analysis of the spindle checkpoint genes san-1, mdf-2, bub-3 and the CENP-F homologues hcp-1 and hcp-2 in Caenorhabditis elegans. Cell Div. 2008;3:6. [PubMed]
  • Hardwick K. G., Johnston R. C., Smith D. L., Murray A. W. MAD3 encodes a novel component of the spindle checkpoint which interacts with Bub3p, Cdc20p, and Mad2p. J. Cell Biol. 2000;148:871–882. [PubMed]
  • Hirose H., Arasaki K., Dohmae N., Takio K., Hatsuzawa K., Nagahama M., Tani K., Yamamoto A., Tohyama M., Tagaya M. Implication of ZW10 in membrane trafficking between the endoplasmic reticulum and Golgi. EMBO J. 2004;23:1267–1278. [PubMed]
  • Hoyt M. A., Totis L., Roberts B. T. S. cerevisiae genes required for cell cycle arrest in response to loss of microtubule function. Cell. 1991;66:507–517. [PubMed]
  • Hwang L. H., Lau L. F., Smith D. L., Mistrot C. A., Hardwick K. G., Hwang E. S., Amon A., Murray A. W. Budding yeast Cdc 20, a target of the spindle checkpoint. Science. 1998;279:1041–1044. [PubMed]
  • Johnson V. L., Scott M. I., Holt S. V., Hussein D., Taylor S. S. Bub1 is required for kinetochore localization of BubR1, Cenp-E, Cenp-F and Mad2, and chromosome congression. J. Cell Sci. 2004;117:1577–1589. [PubMed]
  • Kapoor T. M., Mayer T. U., Coughlin M. L., Mitchison T. J. Probing spindle assembly mechanisms with monastrol, a small molecule inhibitor of the mitotic kinesin, Eg5. J. Cell Biol. 2000;150:975–988. [PubMed]
  • Karess R. Rod-Zw10-Zwilch: a key player in the spindle checkpoint. Trends Cell Biol. 2005;15:386–392. [PubMed]
  • Kim S. H., Lin D. P., Matsumoto S., Kitazono A., Matsumoto T. Fission yeast Slp 1, an effector of the Mad2-dependent spindle checkpoint. Science. 1998;279:1045–1047. [PubMed]
  • King E. M., Rachidi N., Morrice N., Hardwick K. G., Stark M. J. Ipl1p-dependent phosphorylation of Mad3p is required for the spindle checkpoint response to lack of tension at kinetochores. Genes Dev. 2007;21:1163–1168. [PubMed]
  • Kitagawa R., Rose A. M. Components of the spindle-assembly checkpoint are essential in Caenorhabditis elegans. Nat. Cell Biol. 1999;1:514–521. [PubMed]
  • Kitajima T. S., Hauf S., Ohsugi M., Yamamoto T., Watanabe Y. Human Bub1 defines the persistent cohesion site along the mitotic chromosome by affecting Shugoshin localization. Curr. Biol. 2005;15:353–359. [PubMed]
  • Kittler R., et al. Genome-scale RNAi profiling of cell division in human tissue culture cells. Nat. Cell Biol. 2007;9:1401–1412. [PubMed]
  • Kiyomitsu T., Obuse C., Yanagida M. Human Blinkin/AF15q14 is required for chromosome alignment and the mitotic checkpoint through direct interaction with Bub1 and BubR1. Dev. Cell. 2007;13:663–676. [PubMed]
  • Kops G. J., Kim Y., Weaver B. A., Mao Y., McLeod I., Yates J. R., 3rd, Tagaya M., Cleveland D. W. ZW10 links mitotic checkpoint signaling to the structural kinetochore. J. Cell Biol. 2005;169:49–60. [PubMed]
  • Kurz T., Pintard L., Willis J. H., Hamill D. R., Gonczy P., Peter M., Bowerman B. Cytoskeletal regulation by the Nedd8 ubiquitin-like protein modification pathway. Science. 2002;295:1294–1298. [PubMed]
  • Larsen N. A., Al-Bassam J., Wei R. R., Harrison S. C. Structural analysis of Bub3 interactions in the mitotic spindle checkpoint. Proc. Natl. Acad. Sci. USA. 2007;104:1201–1206. [PubMed]
  • Li R., Murray A. W. Feedback control of mitosis in budding yeast. Cell. 1991;66:519–531. [PubMed]
  • Luo X., Tang Z., Xia G., Wassmann K., Matsumoto T., Rizo J., Yu H. The Mad2 spindle checkpoint protein has two distinct natively folded states. Nat. Struct. Mol. Biol. 2004;11:338–345. [PubMed]
  • Maddox P. S., Hyndman F., Monen J., Oegema K., Desai A. Functional genomics identifies a Myb domain-containing protein family required for assembly of CENP-A chromatin. J. Cell Biol. 2007;176:757–763. [PubMed]
  • Manning A. L., Ganem N. J., Bakhoum S. F., Wagenbach M., Wordeman L., Compton D. A. The kinesin-13 proteins Kif2a, Kif2b, and Kif2c/MCAK have distinct roles during mitosis in human cells. Mol. Biol. Cell. 2007;18:2970–2979. [PubMed]
  • Mao Y., Desai A., Cleveland D. W. Microtubule capture by CENP-E silences BubR1-dependent mitotic checkpoint signaling. J. Cell Biol. 2005;170:873–880. [PubMed]
  • Mapelli M., Massimiliano L., Santaguida S., Musacchio A. The Mad2 conformational dimer: structure and implications for the spindle assembly checkpoint. Cell. 2007;131:730–743. [PubMed]
  • McCleland M. L., Gardner R. D., Kallio M. J., Daum J. R., Gorbsky G. J., Burke D. J., Stukenberg P. T. The highly conserved Ndc80 complex is required for kinetochore assembly, chromosome congression, and spindle checkpoint activity. Genes Dev. 2003;17:101–114. [PubMed]
  • Meraldi P., Draviam V. M., Sorger P. K. Timing and checkpoints in the regulation of mitotic progression. Dev. Cell. 2004;7:45–60. [PubMed]
  • Millband D. N., Hardwick K. G. Fission yeast Mad3p is required for Mad2p to inhibit the anaphase-promoting complex and localizes to kinetochores in a Bub1p-, Bub3p-, and Mph1p-dependent manner. Mol. Cell. Biol. 2002;22:2728–2742. [PubMed]
  • Minshull J., Sun H., Tonks N. K., Murray A. W. A MAP kinase-dependent spindle assembly checkpoint in Xenopus egg extracts. Cell. 1994;79:475–486. [PubMed]
  • Moore L. L., Roth M. B. HCP-4, a CENP-C-like protein in Caenorhabditis elegans, is required for resolution of sister centromeres. J. Cell Biol. 2001;153:1199–1208. [PubMed]
  • Murray A. W., Marks D. Can sequencing shed light on cell cycling? Nature. 2001;409:844–846. [PubMed]
  • Musacchio A., Salmon E. D. The spindle-assembly checkpoint in space and time. Nat. Rev. Mol. Cell Biol. 2007;8:379–393. [PubMed]
  • Nasmyth K. How do so few control so many? Cell. 2005;120:739–746. [PubMed]
  • Nystul T. G., Goldmark J. P., Padilla P. A., Roth M. B. Suspended animation in C. elegans requires the spindle checkpoint. Science. 2003;302:1038–1041. [PubMed]
  • O'Connell K. F., Caron C., Kopish K. R., Hurd D. D., Kemphues K.J.Li.Y., White J. G. The C. elegans zyg-1 gene encodes a regulator of centrosome duplication with distinct maternal and paternal roles in the embryo. Cell. 2001;105:547–558. [PubMed]
  • Oegema K., Desai A., Rybina S., Kirkham M., Hyman A. A. Functional analysis of kinetochore assembly in Caenorhabditis elegans. J. Cell Biol. 2001;153:1209–1226. [PubMed]
  • Oegema K., Hyman A. A. Cell division. WormBook. 2006:1–40.
  • Pan J., Chen R. H. Spindle checkpoint regulates Cdc20p stability in Saccharomyces cerevisiae. Genes Dev. 2004;18:1439–1451. [PubMed]
  • Peters J. M. The anaphase-promoting complex: proteolysis in mitosis and beyond. Mol. Cell. 2002;9:931–943. [PubMed]
  • Praitis V., Casey E., Collar D., Austin J. Creation of low-copy integrated transgenic lines in Caenorhabditis elegans. Genetics. 2001;157:1217–1226. [PubMed]
  • Rajagopalan H., Lengauer C. Aneuploidy and cancer. Nature. 2004;432:338–341. [PubMed]
  • Regnier V., Vagnarelli P., Fukagawa T., Zerjal T., Burns E., Trouche D., Earnshaw W., Brown W. CENP-A is required for accurate chromosome segregation and sustained kinetochore association of BubR1. Mol. Cell. Biol. 2005;25:3967–3981. [PubMed]
  • Rieder C. L., Cole R. W., Khodjakov A., Sluder G. The checkpoint delaying anaphase in response to chromosome monoorientation is mediated by an inhibitory signal produced by unattached kinetochores. J. Cell Biol. 1995;130:941–948. [PubMed]
  • Rischitor P. E., May K. M., Hardwick K. G. Bub1 is a fission yeast kinetochore scaffold protein, and is sufficient to recruit other spindle checkpoint proteins to ectopic sites on chromosomes. PLoS ONE. 2007;2:e1342. [PubMed]
  • Severson A. F., Hamill D. R., Carter J. C., Schumacher J., Bowerman B. The aurora-related kinase AIR-2 recruits ZEN-4/CeMKLP1 to the mitotic spindle at metaphase and is required for cytokinesis. Curr. Biol. 2000;10:1162–1171. [PubMed]
  • Sharp-Baker H., Chen R. H. Spindle checkpoint protein Bub1 is required for kinetochore localization of Mad1, Mad2, Bub3, and CENP-E, independently of its kinase activity. J. Cell Biol. 2001;153:1239–1250. [PubMed]
  • Shonn M. A., Murray A. L., Murray A. W. Spindle checkpoint component Mad2 contributes to bioorientation of homologous chromosomes. Curr. Biol. 2003;13:1979–1984. [PubMed]
  • Sironi L., Mapelli M., Knapp S., De Antoni A., Jeang K. T., Musacchio A. Crystal structure of the tetrameric Mad1-Mad2 core complex: implications of a ‘safety belt’ binding mechanism for the spindle checkpoint. EMBO J. 2002;21:2496–2506. [PubMed]
  • Stein K. K., Davis E. S., Hays T., Golden A. Components of the spindle assembly checkpoint regulate the anaphase-promoting complex during meiosis in Caenorhabditis elegans. Genetics. 2007;175:107–123. [PubMed]
  • Sudakin V., Chan G. K., Yen T. J. Checkpoint inhibition of the APC/C in HeLa cells is mediated by a complex of BUBR1, BUB3, CDC20, and MAD2. J. Cell Biol. 2001;154:925–936. [PubMed]
  • Tang Z., Bharadwaj R., Li B., Yu H. Mad2-Independent inhibition of APCCdc20 by the mitotic checkpoint protein BubR1. Dev. Cell. 2001;1:227–237. [PubMed]
  • Tang Z., Shu H., Oncel D., Chen S., Yu H. Phosphorylation of Cdc20 by Bub1 provides a catalytic mechanism for APC/C inhibition by the spindle checkpoint. Mol. Cell. 2004a;16:387–397. [PubMed]
  • Tang Z., Sun Y., Harley S. E., Zou H., Yu H. Human Bub1 protects centromeric sister-chromatid cohesion through Shugoshin during mitosis. Proc. Natl. Acad. Sci. USA. 2004b;101:18012–18017. [PubMed]
  • Tarailo M., Tarailo S., Rose A. M. Synthetic lethal interactions identify phenotypic “interologs” of the spindle assembly checkpoint components. Genetics. 2007;177:2525–2530. [PubMed]
  • Vanoosthuyse V., Valsdottir R., Javerzat J. P., Hardwick K. G. Kinetochore targeting of fission yeast Mad and Bub proteins is essential for spindle checkpoint function but not for all chromosome segregation roles of Bub1p. Mol. Cell. Biol. 2004;24:9786–9801. [PubMed]
  • Vaur S., Cubizolles F., Plane G., Genier S., Rabitsch P. K., Gregan J., Nasmyth K., Vanoosthuyse V., Hardwick K. G., Javerzat J. P. Control of Shugoshin function during fission-yeast meiosis. Curr. Biol. 2005;15:2263–2270. [PubMed]
  • Vink M., Simonetta M., Transidico P., Ferrari K., Mapelli M., De Antoni A., Massimiliano L., Ciliberto A., Faretta M., Salmon E. D., Musacchio A. In vitro FRAP identifies the minimal requirements for Mad2 kinetochore dynamics. Curr. Biol. 2006;16:755–766. [PubMed]
  • Wang X., Babu J. R., Harden J. M., Jablonski S. A., Gazi M. H., Lingle W. L., de Groen P. C., Yen T. J., van Deursen J. M. The mitotic checkpoint protein hBUB3 and the mRNA export factor hRAE1 interact with GLE2p-binding sequence (GLEBS)-containing proteins. J. Biol. Chem. 2001;276:26559–26567. [PubMed]
  • Warren C. D., Brady D. M., Johnston R. C., Hanna J. S., Hardwick K. G., Spencer F. A. Distinct chromosome segregation roles for spindle checkpoint proteins. Mol. Biol. Cell. 2002;13:3029–3041. [PubMed]
  • Watanabe S., Yamamoto T. G., Kitagawa R. Spindle assembly checkpoint gene mdf-1 regulates germ cell proliferation in response to nutrition signals in C. elegans. EMBO J. 2008;27:1085–1096. [PubMed]
  • Weiss E., Winey M. The Saccharomyces cerevisiae spindle pole body duplication gene MPS1 is part of a mitotic checkpoint. J. Cell Biol. 1996;132:111–123. [PubMed]
  • Williams B. C., Li Z., Liu S., Williams E. V., Leung G., Yen T. J., Goldberg M. L. Zwilch, a new component of the ZW10/ROD complex required for kinetochore functions. Mol. Biol. Cell. 2003;14:1379–1391. [PubMed]
  • Yang M., Li B., Liu C. J., Tomchick D. R., Machius M., Rizo J., Yu H., Luo X. Insights into mad2 regulation in the spindle checkpoint revealed by the crystal structure of the symmetric mad2 dimer. PLoS Biol. 2008;6:e50. [PubMed]
  • Yu H. Regulation of APC-Cdc20 by the spindle checkpoint. Curr. Opin. Cell Biol. 2002;14:706–714. [PubMed]
  • Yu H. Cdc 20, a WD40 activator for a cell cycle degradation machine. Mol. Cell. 2007;27:3–16. [PubMed]
Articles from Molecular Biology of the Cell are provided here courtesy of
American Society for Cell Biology

PubMed articles by these authors
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph
See more articles cited in this paragraph