MAD3 encodes a novel component of the spindle checkpoint which interacts with Bub3p, Cdc20p, and Mad2p

Abstract

We show that MAD3 encodes a novel 58-kD nuclear protein which is not essential for viability, but is an integral component of the spindle checkpoint in budding yeast. Sequence analysis reveals two regions of Mad3p that are 46 and 47% identical to sequences in the NH(2)-terminal region of the budding yeast Bub1 protein kinase. Bub1p is known to bind Bub3p (Roberts et al. 1994) and we use two-hybrid assays and coimmunoprecipitation experiments to show that Mad3p can also bind to Bub3p. In addition, we find that Mad3p interacts with Mad2p and the cell cycle regulator Cdc20p. We show that the two regions of homology between Mad3p and Bub1p are crucial for these interactions and identify loss of function mutations within each domain of Mad3p. We discuss roles for Mad3p and its interactions with other spindle checkpoint proteins and with Cdc20p, the target of the checkpoint.

Figure 3

Antibodies to Mad3p reveal that it is a nuclear protein, and that its abundance is not cell cycle regulated. (a) Immunoblots of yeast extracts with anti-Mad3p antibodies. Whole cell extracts made from wild-type (KH 34), mad3Δ.2 (KH 173), mad3-1 (KH 45), and mad3-2 (KH 160) strains were immunoblotted with affinity-purified anti-Mad3p antibodies. A strong signal at ∼58 kD is seen in the wild-type and mad3-1 extracts, which is missing in the mad3Δ and mad3-2 strains indicating that it is specific for Mad3p. (b) Immunofluorescence staining of yeast with anti-Mad3p antibodies. An asynchronous culture of wild-type cells containing myc-tagged Mad3p expressed from a multi-copy plasmid were stained with DAPI, affinity-purified anti-Mad3p antibodies, and anti-Kar2p as a marker for the nuclear envelope and endoplasmic reticulum. Mad3p levels are very variable due to its expression from a multi-copy vector, but in all cases is confined to the nucleus, as delineated by the Kar2p signal in the nuclear envelope. Similar results were obtained with an anti-myc antibody. (c) Mad3p levels are not cell cycle regulated. Wild-type cells (KH 34) were arrested with the mating pheromone alpha-factor and then released into YPD with and without the addition of 15 μg/ml nocodazole. Samples were taken every 20 min for a period of 2 h, whole cell extracts were prepared, and then immunoblotted with affinity-purified antibodies specific for Mad3p, Mad1p, and the mitotic cyclin Clb2p. Unlike Mad1p, which becomes hyperphosphorylated upon nocodazole treatment, and Clb2p, which is regulated at the level of abundance, Mad3p levels are constant and the protein displays no obvious signs of modification.

Figure 4

Mad3p and Bub1p both have two-hybrid interactions with Bub3p and the target of the spindle checkpoint, Cdc20p. (a) A haploid strain containing a fusion between the Gal4 DNA binding domain and Mad3p or Bub1p was crossed with strains containing fusions between the transcriptional activation domain and the indicated checkpoint proteins or Snf4p. β-Galactoside activity of the resulting diploids is shown in Miller units and is the average of at least three independent crosses. (b) Mapping the Bub3p and Cdc20p interaction domains of Mad3p. A series of Mad3p truncations was tested for Bub3p and Cdc20p interaction in the two-hybrid assay, identifying two distinct interaction domains.

Figure 5

Functional analysis of the two homology regions of Mad3p confirms that they form two distinct interaction domains and shows that both are required for checkpoint function. (a) Wild-type, but not homology region II mutant, Mad3p was coimmunoprecipitated with Bub3p. Extracts were made from four yeast strains: a control strain lacking a myc tag (KH 34); a wild-type strain containing myc-tagged Bub3p (KH 228); a mad3Δ strain containing the plasmid pRJ001 (encoding the mad3 homology region I mutant GIGS₁₅₉ > AAAA) and myc-tagged Bub3p (RJ 10); and a mad3 homology region II mutant (mad3-1) containing myc-tagged Bub3p (RJ 11). Bub3p was immunoprecipitated from these extracts and the lysates and immunoprecipitates were then immunoblotted with anti-Mad3p antibodies. (b) Wild-type, but not homology region I mutant, Mad3p was coimmunoprecipitated with Cdc20p. Extracts were made from four yeast strains: a mad3Δ strain (a Ura⁻ derivative of KH173); a wild-type strain (KH 34); a mad3Δ strain containing the plasmid pRJ001 (encoding the mad3 homology region I mutant); and a mad3 homology region II mutant (KH 45). All four strains contained pLH68 which encodes HA-tagged Cdc20p. Mad3p and Cdc20p were immunoprecipitated from these extracts and the immunoprecipitates were then immunoblotted with anti-Mad3p antibodies. *Immunoglobulin heavy chains from the immunoprecipitation. (c) Benomyl sensitivity of homology region I and region II mad3 mutants. Cells were spotted out on YPD plates and plates with 7.5 and 12.5 μg/ml benomyl and then photographed after 3 d growth at 24°C. (d) The central portion of Mad3p is sufficient for Bub3p binding. Mad3-GST fusions (Nterm contains amino acids 1–237 and middle contains 176–409) were mixed with reticulocyte lysates containing radiolabeled Bub3p, and the GST fusions were pulled down with glutathione agarose beads to determine whether they had bound Bub3p.

Figure 6

Cell cycle regulation of Mad3p complexes and the lack of dependency of Mad3p–Bub3p complex formation on other checkpoint proteins. (a) Levels of Mad3p–Cdc20p and Mad3p–Mad2p interaction vary through the cell cycle, but the level of Mad3p–Bub3p complex remains constant. Wild-type yeast (KH 34) was grown to log phase and then arrested in G1 (with the mating pheromone alpha factor), in S phase (with hydroxyurea) or in mitosis (with nocodazole). Lysates were prepared from these cells, and from a control strain lacking Mad3p (KH 173), and Mad3p was immunoprecipitated using affinity-purified antibodies. The immunoprecipitates were then separated by SDS-PAGE and immunoblotted with antibodies specific for Mad1p, Cdc20p (anti-HA), Mad3p, Bub3p, and Mad2p. (b) Mad3p–Bub3p complex formation does not require the other known checkpoint proteins. Strains KH 232–242, all of which contain myc-tagged Bub3p, were grown to log phase, whole cell extracts were prepared, and Mad3p was immunoprecipitated. The immunoprecipitates were then separated by SDS-PAGE and immunoblotted with antibodies specific for Bub3p (anti-Myc) and Mad3p. *Immunoglobulin heavy chains from the immunoprecipitation.

Figure 7

Dependence of Mad3p–Cdc20p, Mad3p–Mad2p, and Cdc20p–Mad2p complex formation on the other checkpoint proteins. Strains KH 243–256, all of which contain the temperature-sensitive cdc26Δ and pLH68L2 (CDC20-HA), were grown to log phase and then nocodazole was added to their growth media for 3 h at 37°C. Whole cell extracts were prepared and split into two aliquots from which Mad3p and Cdc20p were immunoprecipitated. The immunoprecipitates were then separated by SDS-PAGE and immunoblotted with antibodies specific for Cdc20p (anti-HA), Mad3p and Mad2p. (a) Mad3p immunoprecipitations reveal that the Mad3p–Cdc20p complex was low in all checkpoint mutants, except for bub2, and was entirely absent in mad2. A Mad3p–Mad2p association could only be detected in wild-type and bub2 strains. (b) Cdc20p immunoprecipitations reveal that wild-type levels of Mad2p–Cdc20p were present in mad3 and bub2 extracts. The Mad2p–Cdc20p levels were reduced in all other mutants.

Figure 8

A model describing Mad3p interactions. The two regions of homology between Mad3p and Bub1p are labeled I and II. Mad3p associates with Bub3p throughout the cell cycle. Once Cdc20p accumulates, Mad3p can also be found in association with Cdc20p and Mad2p. The formation or stability of this complex is affected by mad1, bub1 and mps1 mutations, but the roles of those proteins remain unclear (see text for details).