Figure 1.
The spindle checkpoint is sensitive to tension on bi-oriented chromosomes.
(A) Unattached or incorrectly attached chromosomes produce no tension; correct attachments (bi-orientation) occur when sister chromatids (blue) attach to opposite poles (black dot) via kinetochores (yellow dot). Tension is generated as microtubules attempt to pull chromatids apart but are resisted by cohesin (orange rings). Bi-orientation tension creates separation between kinetochores (L1) and separation within kinetochores (L2). (B) The spindle checkpoint monitors tension either between kinetochores (inter-kinetochore models) or within kinetochores (intra-kinetochore models). (C) To distinguish between models, pericentric chromatin stretching (inter-kinetochore distance) was inhibited by tethering chromatids together with the cross-linking properties of the Lac repressor. Lac operator arrays (gray boxes) are placed on either side of the centromere and either a dimeric (purple) or tetrameric repressor (red) is expressed. The dimeric form of the repressor contains a C-terminal truncation that prevents tethering, while the tetrameric form can crosslink two chromatids.
Figure 2.
Tetrameric Lac repressor inhibits sister chromatid stretching.
(A) To measure the stretching of chromatids, asynchronous populations were treated with alpha factor to arrest cells in G1. Cells were released from G1 into a metaphase arrest generated by depletion of Cdc20, an essential co-activator of the Anaphase Promoting Complex; samples were taken every 30 minutes and scored for separated chromatids (n>100). (B) Both versions of the repressor are fused to GFP to visualize sister chromatids. Separated chromatids appear as two GFP dots, and one GFP dot is categorized as no separation. Scale bar is 3 µm. (C) Inter-kinetochore separation is not inhibited if a LacO array is placed on only one side of the centromere. Both control, dimeric repressor (GFP-LacI2) cells and tetrameric repressor (GFP-LacI4) cells reach maximum percent separation 60 minutes after release from G1; there is no statistical difference between dimer and tetramer cells at all time points. (D) Centromere separation is inhibited if the tetrameric form of the Lac repressor (GFP-LacI4) is expressed and LacO arrays are placed on both sides of the centromere. Control cells expressing the dimeric repressor (GFP-LacI2) reach maximum stretching 60 minutes post-release from G1; the tetrameric strain has fewer cells with visibly stretched chromatids at all time points in metaphase arrest (p<0.005, Student's t-test). More than 100 cells were scored for GFP dots for each strain in each experiment; error bars represent standard deviation across 3 independent trials.
Figure 3.
Holding sister centromeres close together does not disrupt chromosome segregation.
(A) To determine if the tetrameric Lac repressor disrupts kinetochore assembly or error correction mechanisms, cells were grown in YP +2% raffinose at the permissive temperature and synchronized in G1 with alpha factor. Cells were washed and released into an anaphase arrest at the restrictive temperature in either YP +2% glucose or +2% galactose; anaphase arrest was induced by the cdc15-2 temperature sensitive allele. The segregation of GFP-labeled chromatids was assessed three hours after release from G1. (B) DAPI staining was used to confirm anaphase arrest in cdc15-2 cells grown at the restrictive temperature. Anaphase cells contain a DNA mass in both the mother and daughter cell; 99±1.5% of scored cells had a DAPI staining pattern corresponding to anaphase. (C) Correctly segregated chromosomes were scored as cells in which both the mother and daughter cell received a copy of the chromosome (one GFP dot in each cell). Chromosome mis-segregation was scored as cells in which both copies of the chromosome were located in one cell. (D) A conditional centromere was used as a positive control for chromosome mis-segregation. When grown in glucose, the GAL1 promoter placed upstream of CEN3 is turned off and the centromere is functional; however, when grown in galactose, the GAL1 promoter is turned on and the centromere is disrupted by the transcriptional machinery. (E) Cells expressing the tetrameric repressor do not have increased rates of chromosome mis-segregation compared to those expressing the dimeric repressor or cells with their conditional centromere turned on. However, there was a large and statistically significant difference between these three strains and cells with their conditional centromeres turned off, which were as likely to mis-segregate sister chromatids as they were to segregate them properly. 200 cells scored for correct chromosome segregation for each strain in each experiment; error bars are the standard deviation of 3 independent trials.
Figure 4.
Inhibition of chromatin stretch does not delay mitotic progression.
(A) To assay the progression of cells through mitosis, asynchronous populations were treated with alpha factor to arrest cells in G1. Cells were released from G1 and allowed to proceed synchronously through mitosis and into the next cell cycle. Samples were collected every 10 minutes and scored for mitotic progression (B) Cartoon of how the spindle checkpoint can arrest cells in metaphase, delaying anaphase. Anaphase (purple box) is scored in mitotic progression assays by the separation of GFP-labeled chromatids (green dots) into mother and daughter cells. (C) Cells expressing the tetrameric Lac repressor do not delay mitosis compared to control cells expressing the dimeric Lac repressor. Both strains peak in anaphase 60–70 minutes after release from G1 (no statistical difference between populations). The essential spindle checkpoint component Mad2 was deleted in cells expressing the dimeric or tetrameric Lac repressor, and all four strains move through mitosis on the same time scale. There was no statistical significance between any of the four strains at any one of the time points, suggesting that neither form of the Lac repressor delayed mitosis due to checkpoint activation. More than 100 cells were scored for anaphase for each strain in each experiment; error bars represent the standard deviation of at least 3 independent trials.
Figure 5.
Cells with tethered centromeres can activate the spindle checkpoint.
(A) To assay spindle checkpoint activation of cells in the presence of microtubule-depolymerizing drugs, cells were synchronized in G1 with alpha factor, and released into media containing microtubule-depolymerizing drugs (benomyl and nocodazole). Samples were collected every 60 minutes and scored for the large-budded phenotype indicative of checkpoint activation. (B) The ability of both control (dimeric) and tethered (tetrameric) cells to activate the spindle checkpoint was assayed by scoring the percent of cells arrested as large-budded cells, and compared to cells that do not possess a functional checkpoint and cannot arrest (mad2Δ). Cells expressing either the dimeric or tetrameric Lac repressor arrested in the drugs, indicating that the spindle checkpoint was functional in both strains. 200 cells scored for large-budded phenotype; error bars represent the standard deviation of 3 independent trials. (C) To assay the ability of cells to recover from spindle checkpoint activation induced by microtubule-depolymerizing drugs, cells were synchronized in G1 with alpha factor, released into media containing benomyl and nocodazole for 90 minutes then washed and transferred to media without drugs. Samples were collected every 10 minutes and scored for anaphase (GFP dots in both mother and daughter cells) (D) The ability of both control and tethered strains to recover from spindle checkpoint activation was assayed by transiently treating cells with microtubule-depolymerizing drugs. Both strains arrested in the presence of the drugs and did not enter anaphase until after the drugs were washed out at T = 90 minutes (red arrow). 100 cells were scored for anaphase for each strain in each experiment; error bars represent the standard deviation of at least 3 independent trials.
Table 1.
Strains used in this study.