A Bub1-Mad1 interaction targets the Mad1-Mad2 complex to unattached kinetochores to initiate the spindle checkpoint

Abstract

Recruitment of Mad1-Mad2 complexes to unattached kinetochores is a central event in spindle checkpoint signaling. Despite its importance, the mechanism that recruits Mad1-Mad2 to kinetochores is unclear. In this paper, we show that MAD-1 interacts with BUB-1 in Caenorhabditis elegans. Mutagenesis identified specific residues in a segment of the MAD-1 coiled coil that mediate the BUB-1 interaction. In addition to unattached kinetochores, MAD-1 localized between separating meiotic chromosomes and to the nuclear periphery. Mutations in the MAD-1 coiled coil that selectively disrupt interaction with BUB-1 eliminated MAD-1 localization to unattached kinetochores and between meiotic chromosomes, both of which require BUB-1, and abrogated checkpoint signaling. The identified MAD-1 coiled-coil segment interacted with a C-terminal region of BUB-1 that contains its kinase domain, and mutations in this region prevented MAD-1 kinetochore targeting independently of kinase activity. These results delineate an interaction between BUB-1 and MAD-1 that targets MAD-1-MAD-2 complexes to kinetochores and is essential for spindle checkpoint signaling.

Figure 1.

A kinetochore protein two-hybrid library screen identifies a MAD-1–BUB-1 interaction. (A, left) Schematic of checkpoint signaling mediated by kinetochore-anchored Mad1–Mad2 complexes. The inactive (O, open) and active (C, closed) conformers of Mad2 are depicted. (right) Summary of requirements for MAD-1–MAD-2 targeting to unattached kinetochores in C. elegans. (B) Summary of kinetochore protein two-hybrid screen. Strong indicates growth in low and high stringency conditions; weak indicates growth only under low stringency conditions. (C) Mapping of the region of MAD-1 that interacts with BUB-1 (FL). (D) Biochemical analysis of the MAD-1–BUB-1 interaction. Experimental schematic is on the left. Bead eluates were analyzed by anti–BUB-1 immunoblotting (top right) and Coomassie staining (bottom right). The data shown are representative of three experiments.

Figure 2.

MAD-1 accumulation requires kinetochore-localized BUB-1. (A) The fluorescence intensity of GFP::MAD-1 at unattached kinetochores in monopolar spindles was quantified and expressed as a fraction of that in controls. n is the number of embryos; error bars are the 95% confidence interval of the mean. Ctrl, control. (B) Schematic showing the deleted/mutated regions in the indicated KNL-1 mutants. Graph plots the ratio of BUB-1::GFP to KNL-1::mCh for each KNL-1 variant normalized by dividing by the same ratio for WT KNL-1::mCh. Endogenous KNL-1 was depleted in all conditions. n is number of embryos analyzed. Error bars are the 95% confidence interval of the mean. See also Fig. S1 B. (C and D) MAD-1 and BUB-1 targeting to unattached kinetochores in strains expressing the indicated KNL-1 variants after depletion of ZYG-1 and endogenous KNL-1. n is the number of embryos analyzed. Graph plots the ratio of GFP::MAD-1 (in a mad-1Δ background) or BUB-1::GFP to KNL-1::mCh at kinetochores of monopolar spindles in the indicated strains. Error bars are the 95% confidence interval of the mean. Bars: (A and C) 5 µm; (B) 2 µm.

Figure 3.

Unbiased and targeted mutagenesis of the MAD-1 coiled coil. (A) Schematic summarizing identification of MAD-1 mutants generated by unbiased mutagenesis that fail to bind BUB-1. See also Fig. S2 A. (B) Schematic of targeted heptad mutagenesis and summary of results. Mutations that resulted in no or weak BUB-1 binding are indicated. The P504A mutation prevents interaction with MAD-2. (C) Biochemical analysis of MAD-1^3A conducted as in Fig. 1 D. The blot and gel image is representative of three experiments.

Figure 4.

Analysis of BUB-1 interaction-defective MAD-1 mutants in vivo. (A) GFP::MAD-1 variants expressed from single-copy transgene insertions (Fig. S2 F), propagated in a mad-1Δ background, and analyzed by anti–MAD-1 immunoblotting. α-Tubulin (α-tub) serves as a loading control. (B and C) Images and quantification of GFP::MAD-1 variant localization to unattached kinetochores. Unpaired t tests show that MAD-1^3A is significantly reduced relative to MAD-1^WT (P < 0.0001) and MAD-1^P504 (P = 0.0004); MAD-1^D423A is also significantly reduced (P < 0.0001 relative to MAD-1^WT, and P = 0.0066 relative to MAD-1^P504). The WT value is reproduced from Fig. 2 A. n is number of embryos analyzed. Error bars are the 95% confidence interval of the mean. (D) Immunoblotting of untagged MAD-1 variants. α-Tubulin serves as a loading control. (E) Quantification of time from NEBD to onset of cortical contractility in the presence of monopolar spindles. n is number of embryos analyzed. The red dotted line indicates mitotic duration in mad-1(RNAi), where the checkpoint is inactive. Error bars are the 95% confidence interval of the mean. (F) MAD-1 accumulation between separating homologous chromosomes in early anaphase of oocyte meiosis I. Time (seconds) on the bottom left of merge images are relative to anaphase onset. Images on the bottom left show that this localization depends on BUB-1. Images on the bottom right show loss of this localization for MAD-1^3A but not MAD-1^P504A. 3–11 embryos were imaged per condition. Ctrl, control. (G) Nuclear periphery enrichment of MAD-1 for the indicated conditions. Dotted line indicates the embryo outline. More than eight embryos were imaged per condition. Bars, 5 µm.

Figure 5.

Identification and analysis of BUB-1 mutations affecting MAD-1 kinetochore7 localization. (A) Identification of the C-terminal region of BUB-1 as the interaction domain for MAD-1 and results of a compensatory two-hybrid screen using MAD-1^D423A. TPR, tetratricopeptide repeat. (B) Location of N781 and other activity-related residues in the BUB-1 kinase domain (Protein Data Bank accession no. 3E7E; Kang et al., 2008). Residue numbering: Ce N781 (Hs N879), Ce L777 (Hs L875), Ce K718 (Hs K821), Ce K847 (Hs K946), and Ce D814 (Hs D917). Panel on the right shows two-hybrid analysis of the MAD-1–BUB-1 interaction with N781K and L777K mutations. (C) Schematic of bub-1 transgene and analysis of embryo viability after endogenous BUB-1 depletion for the indicated variants. At least 7 worms and >445 total embryos were scored per condition. Error bars are the 95% confidence interval of the mean. Chr I, chromosome I. (D) Images and quantification of GFP::MAD-1 at unattached kinetochores for the indicated BUB-1 variants, normalized relative to the WT transgene control. n is number of embryos analyzed; error bars are the 95% confidence interval of the mean. Bar, 5 µm. (E) Kinase activity assay of WT and D814N mutant BUB-1 with GST-fused C. elegans histone H2a as substrate. The blot and gel image are representative of three experiments.