Genetic evidence linking TORC1 to cohesin.

(A) Model structure of Mis4. The G1487D substitution confers a thermosensitive growth (Ts) phenotype. The cell growth assay shows the suppression of the Ts phenotype by the indicated mutations. (B) Composition of fission yeast TORC complexes (adapted from Hayashi et al, 2007). (C-D) Down-regulation of TORC1 suppresses the Ts growth defect of mis4-G1487D. (C) Model structure of Mip1 highlighting amino-acid substitutions in the suppressor mutants (RNC :Raptor N-terminal CASPase-like domain, ARM: Armadillo-type fold, WD40: WD-40 repeats). All mip1 mutant alleles from the genetic screen showed reduced phosphorylation of the S6-kinase Psk1, a known TORC1 substrate. Conversely, mip1-Y533A which is deficient for Psk1 binding and phosphorylation, suppressed the Ts phenotype of mis4-G1487D. (D) The TORC1 inhibitor rapamycin rescued mis4-G1487D but not eso1 and rad21 mutants. (E) Genetic upregulation of TORC1 exacerbates the Ts phenotype of mis4-G1487D. The deletion of gcn2 or tsc2 was essentially neutral, although the suppression by mip1-R401G was reduced in a tsc2 deleted background. Deletion of gtr1, iml1 or gtr1-Q61L mimicking the GTP-bound, inactive state of GATOR exacerbated the Ts phenotype of mis4-G1487D while mimicking the active GDP-bound form (gtr1-S16N) was neutral. (F) Psm3K105N-K106N confers a Ts phenotype when combined with the deletion of pph3, encoding the catalytic subunit of PP4, a phenotype efficiently rescued by mip1-R401G.

The incidence of chromosome segregation defects in mis4-G1487D is modulated by TORC1 and the growth medium.

(A) Exponential growing cells at 25°C were shifted at 36.5°C for 24 hours. The growth curve shows that mip1-R401G suppressed the mis4-G1487D temperature growth defect. Cell survival after the 24 hours temperature shift was addressed by plating serial 5-fold cell dilutions at permissive temperature for mis4-G1487D (4×104 cells in the first row). Cell survival was reduced ∼25 fold in mis4-G1487D and restored to wild-type levels when combined with mip1-R401G. (B) Exponentially growing cells at 25°C were shifted at 36°C for one doubling. DNA was stained with DAPI (red, pseudo-colour) and tubulin detected by indirect immunofluorescence (green). Lagging DNA (white arrow) was defined as DAPI-stained material on the anaphase spindle (length>5µm). (C) Rapamycin reduces the incidence of anaphases with lagging DNA. Rapamycin (RAPA, 200ng/ml) or solvent alone (DMSO) was added to cycling cells at 25°C and the cultures shifted to 36°C for one doubling of the cell population. Samples were treated as in B. (D) The mip1 mutant reduced the frequency of anaphases with lagging DNA in the psm3K105-K106N pph3Δ background. Cells were treated as in B. (E) Spindle shrinkage events (shrinkage of more than 0.5µm) were detected by live analysis of cells undergoing anaphase at 36.5°C (Fig.2 supplement 2). (F) Anaphase cells with displaced spindle. Cells were cultured at 36.5°C, fixed and stained with Calcofluor to visualize septa (blue). Shown are mean and SD from three independent experiments (n>65 cells per strain and per experiment). (G) Measurement of inter-kinetochore distance. Cells were shifted to 36.5°C for 5 hours and subsequently chilled at 4°C for 30 minutes. Cells were fixed, DNA stained with DAPI and the distance between Mis6 signals was measured. The images on the top show scattered kinetochore pairs in the mis4-G1487D mutant. Bottom, graph showing the distribution of inter-kinetochore distances (wt, n=46 ; mis4-G1487D, n=48; mis4-G1487D mip1-R401G, n=48). (H) Top, cell growth assay showing that the Ts growth defect of mis4-G1487D is more severe in YES+A medium that in EMM2. Bottom, the frequency of anaphases with lagging DNA is higher in YES+A than in EMM2. The inactive form of the TORC1 inhibitor Gtr1 (Gtr1-Q61L) increased the frequency of abnormal anaphases while the active form was neutral (Gtr1-S16N). Cells were treated as in B. (*) (**) (***), P-values from two-sided Fisher’s exact test.

Increased cohesin binding to Cohesin Associated Regions in the raptor mutant mip1-R401G.

(A) Chromosomal sites examined were Cohesin Associated Regions along the arms (438, 1806, 2898 and 3323) and centromere of chromosome 2, the non-transcribed spacer region (NTS) of the rDNA gene cluster on chromosome 3 and the chromosome 1 right telomere. Within the centromere, the central core (cc2) which is the site of kinetochore assembly, the imr and dg repeats that flank the central core on either side and at tRNA rich domains that delineate the centromere. (B) The mip1-R401G mutation restores cohesin binding in mis4-G1487D cells. Cohesin binding to chromatin was monitored by Rad21-FLAG ChIP in G1 cells (cdc10-129 arrest) at 36.5°C. Error bars = SD from 4 ChIPs, except for mis4-G1487D (3 ChIPs). ***P ≤ 0.001, **P ≤ 0.01, *P ≤ 0.05 by two-tailed, unpaired t-test with 95% confidence interval. (C-D) The mip1-R401G mutation increased Mis4 and Rad21 binding in mis4+ cells. ChIP assays were made with cdc10-129 arrested cells. (C) Mean ratios +/- SD were calculated from 2 independent experiments and 4 technical replicates per experiment. (D) Mean ratios +/- SD were calculated from 4 independent experiments and 4 technical replicates per experiment. ***P ≤ 0.001, **P ≤ 0.01, *P ≤ 0.05, by two-tailed, one sample t-test with 95% confidence interval.

TORC1 components co-purify with Rad21 and Mis4.

(A) Affinity purifications (triplicate samples) were made from G1-arrested cells (cdc10-129 arrest) and proteins analysed by label-free mass spectrometry. (B-C) Reciprocal co-immunoprecipitation experiments. Western blots were probed with the indicated antibodies. (B) Rad21-PK IP (cdc10-129 arrested cells). (C) Tor2-FLAG IP (cycling cells).

TORC1 signalling affects the phosphorylation levels of Psm1-S1022 and Mis4-S183.

(A) Mis4-GFP and Cohesin (Rad21-PK) were affinity-purified from G1-arrested cells (cdc10-129 arrest). Triplicate samples were analyzed by label-free mass spectrometry. Mis4-S183 and Psm1-S1022 phosphorylation levels were reduced in the mip1-R401G background. (B) Cohesin was purified by Rad21-FLAG immunoprecipitation from cdc10-129 arrested cells and probed with anti-Psm1-S1022p and anti-Psm1 antibodies. Mean ratios +/- SD from three experiments. (C) Mis4-GFP was affinity-purified from cdc10-129 arrested cells and probed with anti-Mis4-S183p and anti-GFP antibodies. Mean ratios +/- SD from three experiments. (D-E) Effect of rapamycin on the phosphorylation status of Psm1-S1022 and Mis4-S183. Rapamycin (RAPA, 200ng/ml) or solvent alone (DMSO) was added to cycling cells at 25°C and the cultures shifted to 36°C for one doubling of the cell population. (C) Mis4-GFP was affinity purified and probed with anti-Mis4-S183p and anti-GFP antibodies. Mean ratios +/- SD from two experiments. (D) Cohesin was purified by Rad21-FLAG immunoprecipitation and probed with anti-Psm1-S1022p and anti-Psm1 antibodies. Mean ratios +/- SD from two experiments

Mimicking the non phosphorylated state of Psm1-S1022 and Mis4-S183 alleviates mis4-G1487D phenotypes.

(A) Cell growth assays. (B) Anaphases with lagging DNA. Exponentially growing cells at 25°C were shifted at 36°C for 3.5hrs (one doubling). DNA was stained with DAPI and tubulin detected by indirect immunofluorescence. Lagging DNA was defined as DAPI-stained material on the anaphase spindle (length>5µm). P-values from two-sided Fisher’s exact test. (C) Anaphase cells with displaced spindle. Cells were cultured at 36°C, fixed and stained with Calcofluor to visualize septa. Mean and SD from three independent experiments (n>70 cells per strain and per experiment). P-value from two-sided Fisher’s exact test. (D) Inter-kinetochore distance. Cells were treated as in Figure 2G (n=51 for each mutant). Data from Figure 2G are shown for comparison (dotted lines). (E) Rad21 ChIP. Exponentially growing cells at 25°C were shifted at 36°C for 3.5hrs (one doubling). Mean and SD from 6 ratios and 2 independent experiments. **P ≤ 0.01, *P ≤ 0.05, by two-tailed, one sample t-test with 95% confidence interval.

Rapamycin reduces the frequency of chromosome loss.

(A) The rate of chromosome loss was estimated by measuring the breakdown of diploids after growth in medium containing rapamycin (RAPA, 200 ng/ml) or solvent alone (DMSO). Data are mean ± SD, n=2 per group, **P< 0.01, unpaired t-test. (B) Loss rate of the CM3112 minichromosome. The loss rate was determined in the presence of rapamycin (RAPA, 200 ng/ml) or solvent alone (DMSO). Data are mean ± SEM, n=4 per group (n=10 for the wild-type), ***P< 0.001, repeated measures two-way ANOVA followed by Bonferroni post-hoc test.

mis4-G1487D affects gene response to environmental changes.

(A) RNA-sequencing (biological triplicates) was performed on cells growing in rich medium at 25°C (V25), after one cell doubling at 36.5°C (V36.5) and on G1-arrested cells (cdc10-129) at 36.5°C (G1_36.5). For nitrogen starvation, cells grown in EMM2 were deprived from nitrogen for 24H at 25°C (T0-N). One half of the culture was shifted to 36.5°C for 4 days (T4D36.5) while the other half was left at 25°C for 4 days (T4D25). The Venn diagrams show the overlap number of differentially expressed genes (FDR<0.05) for each experimental condition. P-value for under- or over-enrichment were calculated based on the cumulative distribution function of the hypergeometric distribution. (B) Differentially expressed genes in mis4-G1487D across all experiments. There were 408 hits involving 338 genes. The lower graphs show that most of the affected genes are specific to a single experimental condition. (C) Gene features and GO-term analysis for the 338 mis4-regulated genes. (D) Overlap between genes differentially expressed in mis4-G1487D and mip1-R401G versus wild-type in all experimental conditions. (E) Among the 338 genes differentially expressed in mis4, 100 show a similar (+/- 10%) fold change in the mis4, mip1 and mis4 mip1 mutant strains. See also Figure 8—figure supplement 6.

(A) Cell growth assay showing the suppression of the Ts phenotype of mis4-G1487D by the indicated mutations. Plates were incubated for 6 days at the indicated temperatures. The caa1 mutant is slow growing and 6 days were required to see a weak suppressor effect. (B) The deletion of tor1, encoding the catalytic subunit of TORC2, shows a negative genetic interaction with mis4-G1487D.

Live analysis of mis4-G1487D undergoing anaphase at 36.5°C.

A merotelic kinetochore is evidenced by the stretching of the outer kinetochore protein Ndc80 (line scan time 28 to 36 min).

Live analysis of anaphase cells revealed spindle shrinkage events.

(A) Example of a mis4-G1487D cell showing spindle shrinkage during anaphase. (B) As shown in Figure 2E the proportion of mitotic cells with the occurrence of spindle shrinkage (>0.5µm; red curves) is reduced in the mip1-R401G background (wt, n=30; mis4-G1487D, n=34; mis4-G1487D mip1-R401G, n= 34).

(A) Sister-chromatid cohesion assay at the ade6 locus.

Exponentially growing cells at 25°C were shifted at 36°C for one doubling. Fixed cells were treated for detection of GFP dots and tubulin by indirect immunofluorescence. The sample image on the left shows mis4-G1487D cells. DNA appears in blue, tubulin in red and the GFP-marked ade6 locus in green. Separated GFP dots (>0.5µm) were scored in G2 cells as judged by the presence of the interphase array of microtubules and a single nucleus. P-value from two-sided Fisher’s exact test. (B-C) Psm3-K106 acetylation levels. (B) Exponentially growing cells at 25°C were shifted at 36.5°C. HU (12mM) was added at the time of the shift and cells incubated for 3.5 hours. The level of Psm3-K106 acetylation was determined by western blotting. Cell cycle arrest in early S-phase was checked by DNA-content analysis (Right). (C) Rapamycin (200ng/ml) or solvent alone (DMSO) was added to exponentially growing cells at 25°C and the cultures shifted to 36.5°C for one doubling of the cell population. Protein extracts were analyzed by western blotting using antibodies detecting phosphorylated Psk1 and the cohesin core subunits Rad21, Psm1 and Psm3. The acetylated form of Psm3 was detected with anti Psm3-K106ac antibodies.

The frequency of anaphases with lagging DNA is dependent on the culture medium.

Exponentially growing mis4-G1487D cells at 25°C in the indicated media were shifted at 36.5°C for one doubling of the cell population. DNA was stained with DAPI and tubulin detected by indirect immunofluorescence. Lagging DNA was defined as DAPI-stained material on the anaphase spindle (length>5µm). EMM-GLU is EMM2 in which NH4Cl was replaced by 20 mM glutamate. (*) P-values from two-sided Fisher’s exact test.

(A) The mip1-R401G mutation increases cohesin binding to Cohesin Associated Regions in pph3Δ psm3K105N-K106N cells. Exponentially growing cells at 25°C were shifted to 36°C for one doubling of the cell population. Cohesin binding to chromatin was monitored by ChIP using anti-Rad21 antibodies. Mean and SD from 6 ChIPs. (B) Effect of rapamycin treatment on cohesin binding. Exponentially growing cells at 25°C were shifted to 36°C for one doubling of the cell population. Rapamycin (200ng/ml) or solvent alone (dmso) was added at the time of the temperature shift. Cohesin binding to chromatin was monitored by ChIP using anti-Rad21 antibodies. Mean and SD from 4 ChIPs. ***P ≤ 0.001, **P ≤ 0.01, *P ≤ 0.05, by two-tailed, one sample t-test with 95% confidence interval.

(A) Phosphatase assay showing the specificity of anti-Mis4-S183p antibodies. Mis4-GFP was affinity-purified from cdc10-arrested cells. Phosphatase treatment was performed in the presence of absence of phosphatase inhibitors and proteins analyzed by western blotting with the indicated antibodies. (B) Mis4-GFP immunoprecipitations made from cycling cells. Western-blots were probed with anti-Mis4-S183p and anti-GFP antibodies. (C) Rad21-FLAG immunoprecipitations made from G1 cells (cdc10-129 arrest). Western blots were probed with anti-Psm1-S1022p and anti-Psm1 antibodies.

(A) The phosphorylation of Mis4-S183 is not dependent on Pef1. Mis4-GFP was affinity-purified from pef1+ and pef1Δ cells (cdc10-129 arrest) and triplicate samples analyzed by label-free mass spectrometry. (B-C) The phosphorylation of Psm1-S1022 is dependent on Pef1. (B) Rad21-PK was affinity-purified (triplicate samples) from pef1+ and pef1Δ cells (cdc10-129 arrest) and proteins were analyzed by label-free mass spectrometry (C) Rad21-FLAG was affinity-purified from pef1+ and pef1Δ cells (cdc10-129 arrest) and proteins analyzed by western blotting using the indicated antibodies. Consistent with the mass spectrometry analyses, the anti-Psm1-S1022p antibodies did react with Psm1 of wild-type but not from pef1Δ cells. (D) Psk1 phosphorylation is reduced in pef1Δ. TCA extracts from cycling cells were probed by western blotting with the indicated antibodies. (E) Cell growth assay showing the negative genetic interaction between pef1Δ and mip1-R401G.

(A) Cell growth assay showing the genetic interaction between mis4-G1487D and null mutants of AGC kinase genes. (B) Cohesin binding to chromatin was monitored by Rad21-FLAG ChIP in G1 cells (cdc10-129 arrest). Mean ratios and SD were calculated from 2 independent experiments and 4 technical replicates per experiment. ***P ≤ 0.001, **P ≤ 0.01, *P ≤ 0.05, by two-tailed, one sample t-test with 95% confidence interval. (C) Exponentially growing cells at 25°C were shifted at 36.5°C for one doubling of the cell population and fixed. DNA was stained with DAPI and tubulin detected by indirect immunofluorescence. Lagging DNA was defined as DAPI-stained material lagging on the anaphase spindle (length>5µm). P-values from two-sided Fisher’s exact tests. (D-E) Psm1-S1022 and Mis4-S183 phosphorylation levels in AGC kinase mutants in cdc10-129-arrested cells. Rad21-FLAG (D) and Mis4-GFP (E) immunoprecipitates were analyzed by western blotting using the indicated antibodies and the ratios of signal intensities were calculated.

Mimicking the non phosphorylated state of Psm1-S1022 and Mis4-S183 suppressed pph3Δ psm3-K105K106N phenotypes.

(A) Cell growth assay showing the suppression of the Ts phenotype. (B) Exponentially growing cells at 25°C were shifted at 36°C for one doubling. DNA was stained with DAPI and tubulin detected by indirect immunofluorescence. Lagging DNA was defined as DAPI-stained material on the anaphase spindle (length>5µm). P-values from two-sided Fisher’s exact tests. (C) Exponentially growing cells at 25°C were shifted to 36°C for one doubling of the cell population. Cohesin binding to chromatin was monitored by ChIP using anti-Rad21 antibodies. Mean and SD from 6 ChIPs. ***P ≤ 0.001, **P ≤ 0.01, *P ≤ 0.05, by two-tailed, one sample t-test with 95% confidence interval.

CM3112 loss rate was determined in the presence of rapamycin (RAPA, 200 ng/ml) or solvent alone (DMSO).

Data are mean ± SEM, n=4 per group (n=10 for the wild-type), ***P< 0.001, *P< 0.05 repeated measures two-way ANOVA followed by Bonferroni post-hoc test.

Response to nitrogen starvation and refeeding.

Exponentially growing cells in EMM2 medium at 25°C were washed three times in EMM2-N and incubated in EMM2-N for 16 hours at 25°C (time 0). DNA content analysis showed that all strains arrested with a 1C DNA content. G1-arrested cells were then released into the cell cycle in rich medium at 36.5°C. DNA content analysis showed that all strains progressed through S phase with similar kinetics.

(A) Browser view of mis4-regulated genes showing the distribution bias towards the ends of chromosomes 1 and 2. (B-C) Overlap between the 338 differentially expressed genes in mis4-G1487D versus wild-type with genes upregulated in tor2ts6 (Wei et al., 2021) and genes upregulated during G0 (Zahedi et al., 2023).

Cell survival during quiescence.

Exponentially growing cells in EMM2 medium at 25°C were washed three times in EMM2-N and incubated in EMM2-N for 24 hours at 25°C (time 0). Half of the culture was transferred to 36.5°C while the other half remained at 25°C. Cell viability was monitored over time by methylene blue staining (sample image on the right).

Quiescence exit at 25°C.

Exponentially growing cells in EMM2 medium at 25°C were washed three times in EMM2-N and incubated in EMM2-N for 24 hours at 25°C. Half of the culture was transferred to 36.5°C while the other half remained at 25°C. After 24 hours, cells were plated on YES medium (25°C) to assess cell’s ability to exit G0 and form colonies.

Quiescence exit at 36.5°C.

Exponentially growing cells in EMM2 medium at 25°C were washed three times in EMM2-N and incubated in EMM2-N for 24 hours at 25°C. Half of the culture was transferred to 36.5°C while the other half remained at 25°C. After 3 days, nitrogen (NH4Cl 5g/l) was added (time 0). Exit from G0 was monitored by flow cytometry analysis of DNA content . At 25°C (G0 25°C => exit 25°C) the onset of DNA replication occurred ∼7hrs after nitrogen addition for all strains. At 36.5°C (G0 36.5°C => exit 36.5°C) DNA replication was slow or asynchronous even for the wild-type. The numbers above the peaks indicate the fraction of cells with 1C and 2C DNA content for the 0 and 12h time points. The mis4 mutant behaved mostly as wild-type although a larger fraction of cells did not replicate 12 hours after nitrogen addition. For the two strains harbouring mip1-R401G, a large fraction of cells did not replicate during the duration of the experiment.

Among the 338 genes differentially expressed in mis4-G1487D, 100 (107 occurrences, 100 genes) show a similar fold change (+/−10%) in mis4-G1487D, mip1-R401G and the double mutant.

Graphs show the log2 fold change over wild-type for the 3 mutant strains for each gene_experiment.

Genes differentially expressed in mis4-G1487D showing a change in the opposite direction in the mip1 mutant.

(A) mis4-G1487D log2 fold change over wild-type >0 and mip1-R401G log2 fold change over wild-type <0 ; 43 occurrences, 41 genes. (B) mis4-G1487D log2 fold change over wild-type <0 and mip1-R401G log2 fold change over wild-type >0 ; 21 occurrences, 21 genes. Graphs show the log2 fold change over wild-type for the 3 mutant strains for each gene_experiment.