Shugoshin is required for efficient metaphase arrest in response to microtubule depolymerizing drugs

SGO1 (FungiDB ORF ID: CNAG_05516) encodes the only form of shugoshin that is expressed in C. neoformans. The putative 645-aa-long protein coded by SGO1 has an N-terminal coiled-coil domain and a C-terminal basic SGO motif (Figs 1A, S1A and S1B). Multiple sequence alignment of amino acid sequences revealed the presence of conserved residues required for PP2A binding (S1B Fig) [37]. The SGO motif in C. neoformans shugoshin also possesses a conserved lysine residue (K382) that is required for the binding of shugoshin to phosphorylated histone H2A (S1B Fig) [26,27,38]. Previous studies reported that shugoshin plays a role in ensuring proper kinetochore-MT attachments during mitosis (see reviews [39,40]). Mutants affecting kinetochore-MT interactions and proteins participating in the error correction machinery and the SAC, are hypersensitive to MT depolymerizing drugs. To test if Sgo1 contributes to the fidelity of mitotic chromosome segregation, we created sgo1 null mutants in C. neoformans. While Sgo1 was found to be dispensable for growth, sgo1 null mutant cells displayed sensitivity to a MT-depolymerizing drug, thiabendazole (TBZ), similar to that of SAC mutant mad2 (Fig 1B). These results suggest that sgo1Δ mutant cells in the basidiomycete yeast C. neoformans have phenotypes similar to those of ascomycete yeasts such as Saccharomyces cerevisiae, Schizosaccharomyces pombe and Candida albicans [23,25,41].

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Shugoshin is required for a robust spindle assembly checkpoint arrest.

(A) Schematic of the predicted motifs present in Sgo1 of C. neoformans. (B) A ten-fold serial dilution spotting assay to score for the sensitivity of strains CNVY108 (SGO1 GFP-H4), CNSD110 (GFP-H4 sgo1Δ), SHR741 (GFP-H4 mad2Δ), CNSD148 (GFP-H4 sgo1Δ mad2Δ) on plates containing the indicated concentrations of thiabendazole (TBZ). No drug represents DMF (dimethyl formamide) only. The plates were incubated at 30°C for 24 h. (C) Left, schematic of the assay to score for cell cycle arrest phenotypes when treated with nocodazole. Cells treated with nocodazole for 120 and 240 min were scored for cell cycle arrest phenotypes. Right, bar graphs representing the proportion of large-budded cells in the indicated strains, N = 3, n>100, cells counted in each experiment. Two-way ANOVA with Tukey’s multiple comparison test was used to estimate the significance. ****p<0.0001, ***p<0.001, **p = 0.0019/0.0072. Non-significant p-values were omitted from the graph. (D) Left, an outline of the assay to study the effect of nocodazole on the ploidy content of indicated strains using flow cytometry. Right, histograms depicting ploidy changes associated with nocodazole treatment.

https://doi.org/10.1371/journal.pgen.1011552.g001

Next, we sought to test if enhanced sensitivity to the MT poison observed in sgo1Δ mutant cells accentuate when combined with mad2Δ mutant. To test this, we created sgo1Δ mad2Δ double mutants. Such double mutant cells exhibited synthetic lethality in the presence of TBZ (Fig 1B), indicating that both Sgo1 and Mad2 are required for an efficient SAC response in MT-depolymerizing conditions. This striking synthetic lethality observed suggests that shugoshin is involved in functions such as biorientation apart from promoting the SAC response, which is lost in mad2Δ. If shugoshin is required for an efficient SAC response in C. neoformans, it is expected that the sgo1Δ mutant, when combined with a kinetochore mutant, would enhance sensitivities to TBZ. To explore this possibility, we used null mutant cells of a recently reported basidiomycete-specific kinetochore linker protein bridgin (Bgi1). It has been shown that bgi1Δ mutants are viable but exhibit gross chromosomal missegregation [33]. bgi1Δ mutants, combined with mad2Δ, displayed enhanced sensitivity to TBZ compared to each of the single mutants alone [33]. The spotting assay indicated that sgo1Δ bgi1Δ double mutant cells are hypersensitive to TBZ, similar to that of mad2Δ bgi1Δ (S2 Fig). Together, these results reveal that shugoshin plays a role in proper kinetochore-MT interactions and/or monitors defective attachments in response to spindle poison.

MT-depolymerizing drugs arrest wild-type cells at metaphase in the large-budded stage with an undivided nucleus. Only when the defects are rectified will such cells proceed to complete the cell division. To verify whether sgo1Δ mutants can induce metaphase arrest when MTs are depolymerized, we performed a microscopy-based assay to score for the cell cycle defects associated with sgo1Δ mutants by treating cells with nocodazole (Fig 1C). Nocodazole-treated wild-type cells of C. neoformans arrest at the large-budded stage with a visibly condensed chromatin mass (Fig 1C, right). After 240 min of nocodazole treatment, >90% of wild-type cells were at the large-budded stage with a condensed undivided nucleus (Figs 1C and S1C–S1E). As expected, mad2Δ, a bonafide checkpoint-defective mutant, failed to arrest cells at the large-budded stage (Fig 1C, right). In the sgo1Δ mutants, though we initially observed an increase in the proportion of large-budded cells compared to mad2Δ mutants, after 240 min post-nocodazole treatment, the large-budded cell population significantly reduced (Fig 1C, right). Moreover, unlike wild-type, most large-budded cells in sgo1Δ mutants had an uncondensed chromatin mass similar to mad2Δ mutant cells (S1C–S1E Fig). In the case of sgo1Δ mad2Δ double mutants, no significant differences in the proportion of large-budded cells were observed when compared to sgo1Δ and mad2Δ single mutants (Fig 1C, right). Estimating the ploidy content in nocodazole-treated cells using flow cytometry (Fig 1D) revealed that wild-type cells arrested at 2n ploidy level. On the other hand, mad2Δ cells shifted towards 4n, indicating a failure in arresting cells at metaphase and initiating the next cell cycle events. Unlike mad2Δ, sgo1Δ cells displayed a ploidy content of 2n and 4n, revealing the presence of cells that are delayed at metaphase and cells that have initiated the next round of DNA replication. Similar results were also obtained in sgo1Δ mad2Δ double mutants. Taken together, we show that shugoshin is required for an efficient metaphase arrest in response to unattached kinetochores induced by MT-depolymerizing drugs.

sgo1Δ mutant cells fail to maintain the SAC component Bub1 at kinetochores in response to prolonged microtubule depolymerization

Why are sgo1Δ mutant cells not arrested when treated with an MT poison at metaphase? It is possible that sgo1Δ mutant cells arrest at metaphase immediately after being treated with the MT poison but were unable to maintain the arrest upon prolonged exposure to the drug. To test this, we localized the SAC component Bub1 (Fig 2A) tagged with GFP in wild-type and sgo1Δ cells. A recent report [34] suggested that Bub1 localizes to kinetochores in C. neoformans and plays a central role in the SAC-mediated arrest. We synchronized C. neoformans cells by arresting them at G2 using hypoxia [42] and released them in the presence of nocodazole (refer to Materials and Methods for further details) (Fig 2B). Inducing physical hypoxia in C. neoformans results in ’unbudded’ G2 arrest. These G2-arrested cells enter mitosis synchronously when subjected to extensive aeration. Upon release into nocodazole-containing media, wild-type cells accumulated at the large-budded stage (S3 Fig) with GFP-Bub1 signals, which were maintained until the end of the experiment (Fig 2C and 2D). In the case of sgo1Δ cells, the proportion of Bub1-positive large-budded cells increased similarly to that of wild-type until 120 min. However, beyond 200 min post-nocodazole treatment, a sudden decline in the proportion of Bub1-positive large-budded cells was observed in the mutant (Fig 2C and 2D). Studies in S. pombe have shown that the kinase-dead version of Bub1 is checkpoint-proficient but fails to maintain prolonged SAC arrest [27]. A similar observation was reported in C. neoformans, where bub1-kd mutant cells failed to sustain the arrest induced by nocodazole [34]. To understand how checkpoint responses differ in sgo1Δ from bub1-kd, we compared GFP-Bub1 localization dynamics in hypoxia-synchronized cells treated with nocodazole (Fig 2C and 2D). Though neither sgo1Δ nor bub1-kd mutant cells could maintain the arrest like the wild-type, bub1-kd cells were able to maintain the arrest longer than sgo1Δ cells.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Shugoshin ensures Bub1 is retained at kinetochores to maintain the spindle assembly checkpoint arrest.

(A) Schematic of the SAC pathway and the evolutionarily conserved role of Sgo1 and Aurora B^Ipl1 in detecting erroneous kinetochore-MT attachments. (B) An experimental workflow followed to determine the GFP-Bub1 localization dynamics under nocodazole-treated conditions. (C) Sub-cellular localization of GFP-Bub1 localization in IL102 (SGO1), IL143 (bub1-kd) and CNSD176 (sgo1Δ). For representation, cells treated with nocodazole for 120 and 280 min were shown. Scale bar, 5 μm. (D) Quantification of the proportion of large-budded cells with GFP-Bub1 signal in indicated strains treated with nocodazole. N = 3, n>100 cells were analyzed, and the shaded regions above and below the line graphs represent error bands corresponding to SD. (E) Quantification of GFP-Bub1 fluorescence intensity values at 120 min post-nocodazole treatment. N = 3, n>100 cells analyzed for each experiment. White circles represent the average intensity values of replicates. A mean of the average intensity values of replicates is represented as a black bar, and error bars represent SD. An unpaired t-test was used to test the significance, ** p = 0.0092, ns = 0.1131 (F) Microfluidics assays to determine the retention time of GFP-Bub1 at kinetochores in response to nocodazole treatment. Temporal heat maps of 30 randomly selected cells are shown. The heat map represents the changes in kinetochore localization signal of GFP-Bub1 over time. Each bright track on the y-axis of the heat map represents GFP-Bub1 signals from an individual cell (median of the brightest 5 pixels in each cell divided by the overall cell median brightness). The length of each bright track along the x-axis represents the time (min). The time of addition of nocodazole (2.5 μg/mL) is considered as 0 h. Images are taken every 2 min for 15 h. Assays were performed using strains IL102 (GFP-BUB1), IL143 (GFP-bub1-kd), CNSD176 (GFP-BUB1 sgo1Δ), and IL089 (GFP-BUB1 mad1Δ) strains. (G) Bar graphs representing the quantitative analysis of the GFP-Bub1 retention time at kinetochores obtained from microfluidics assays in the above-indicated strains. n = 30, error bars represent SEM.

https://doi.org/10.1371/journal.pgen.1011552.g002

Moreover, Bub1 signal intensities in sgo1Δ and bub1-kd mutants were significantly reduced compared to the wild-type (Fig 2E). To further support these observations, we resorted to a single-cell microfluidics assay to quantitatively measure the checkpoint response in individual cells treated with the MT poison (Fig 2F and 2G). Using this system, we probed the GFP-Bub1 dynamics in wild-type, sgo1Δ, bub1-kd, sgo1Δ bub1-kd, and mad1Δ mutant cells treated with nocodazole. Similar to the results observed in the experiment performed with the synchronized cells, neither sgo1Δ nor bub1-kd mutant cells could maintain checkpoint arrest like wild-type as indicated by the lower mean kinetochore residence times of Bub1 (Fig 2G). Checkpoint defective mad1Δ mutant cells were used as a control. The sgo1Δ bub1-kd double mutant behaved like the sgo1Δ single mutant (S4A and S4B Fig). These results reveal that sgo1Δ cells, similar to bub1-kd cells, are proficient in checkpoint activation but cannot maintain checkpoint signals during prolonged treatment with nocodazole.

Aurora B^Ipl1 maintenance at kinetochores is affected in sgo1Δ mutants during prolonged checkpoint arrest

Shugoshin interacts with CPC proteins and helps in either the recruitment or maintenance of CPC components at centromeres/kinetochores in S. cerevisiae, S. pombe and humans [40,43]. Previously, we established the role of Aurora B^Ipl1 (a CPC component) in regulating nuclear division during mitosis in C. neoformans [44]. Although it has been shown that the entry of Aurora B^Ipl1 into the nucleus coincides with the nuclear envelope breakdown (NEBD), its localization to kinetochores remained unexplored. In this study, we colocalized Aurora B^Ipl1 with the inner kinetochore protein CENP-A at different cell cycle stages (S5A Fig). During G2, when kinetochores are clustered, we could only detect weak signals of Aurora B^Ipl1 that intensified as cells entered metaphase. Apart from the kinetochore localization, we could also observe spindle-like localization of Aurora B^Ipl1 during anaphase (S5A Fig). Upon treatment with nocodazole, we observed colocalization of Aurora B^Ipl1 with kinetochores (S5B Fig). To test whether kinetochore localization of Aurora B^Ipl1 is affected in sgo1Δ cells in C. neoformans, we synchronized both wild-type and sgo1Δ mutant cells by inducing physical hypoxia followed by a release in nocodazole-containing media (Fig 2B). We scored for Aurora B^Ipl1-3xGFP-positive large-budded cells. In sgo1Δ, we observed a reduced proportion of Aurora B^Ipl1-positive large-budded cells over time compared to wild-type (Figs 3A, 3B and S6). Similar to the Bub1 levels in sgo1Δ mutant cells, a reduction in the intensity of the Aurora B^Ipl1 signals in the sgo1Δ cells was also observed compared to the wild-type (Fig 3C). Taken together, we conclude that Sgo1 is required to maintain Bub1 and Aurora B^Ipl1 at kinetochores when MTs are depolymerized.

Aurora B^Ipl1 is required for the maintenance of Bub1 at kinetochores in response to microtubule depolymerization

Reduced Aurora B^Ipl1 levels at kinetochores in sgo1Δ cells prompted us to test whether the absence of Aurora B^Ipl1 leads to failure of checkpoint arrest when treated with MT depolymerizing agents. For conditional depletion of Aurora B^Ipl1, we placed the gene under the GAL7 promoter. We treated Aurora B^Ipl1 –(over)expressed (permissive, galactose media) and Aurora B^Ipl1-depleted (non-permissive, glucose media) cells with nocodazole (Fig 3D). We assayed for large-budded cells with a condensed chromatin mass as observed at metaphase. Cells (over)expressing Aurora B^Ipl1 and treated with nocodazole displayed a higher proportion of large bud arrested cells. In contrast, cells depleted of Aurora B^Ipl1 failed to arrest at the large-budded stage and exhibited an increase in the proportion of tri-budded cells (S7A and S7B Fig). Moreover, approximately 50% of the large-budded cells displayed an uncondensed nuclear mass, similar to what was observed in mad2Δ mutants (S7A and S7B Fig). These results suggest that Aurora B^Ipl1 is required for an efficient metaphase arrest when cells are treated with a spindle poison in C. neoformans.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Shugoshin-dependent Aurora B^Ipl1 pool at kinetochores ensures maintenance of Bub1 to prolong the checkpoint arrest.

(A) Sub-cellular localization of Aurora B^Ipl1-3xGFP localization in CNSD190 (AURORA B^Ipl1-3xGFP mCherry-CENP-A SGO1) and CNSD196 (AURORA B^Ipl1-3xGFP mCherry-CENP-A sgo1Δ) strains treated with nocodazole. For representation, cells treated with nocodazole for 120 and 280 min were shown. Scale bar, 5μm. (B) Quantification of the proportion of large-budded cells with Aurora B^Ipl1-3xGFP signals in the indicated strains. N = 3, n>100 cells were analyzed, and the shaded regions above and below the line graph represent error bands corresponding to SD. (C) Quantification of Aurora B^Ipl1-3xGFP fluorescence intensity values at 120 min post-nocodazole treatment. N = 3, n>58 cells for each experiment. White circles represent the average intensity values of replicates. The mean of the average intensity values of replicates is represented as a black bar. Error bars represent SD. An unpaired t-test was used to test the significance, ** p = 0.0013. (D) Schematic to assess the response of Aurora B^Ipl1 (over)expression or depletion to nocodazole treatment. (E) Sub-cellular localization of GFP-Bub1 localization in CNSD205 (GAL7-3xFLAG-AURORA B^IPL1 GFP-BUB1) Aurora B^Ipl1 (over)expressed and repressed conditions treated with nocodazole. Cells treated with nocodazole for 120 and 240 min are shown for representation. Scale bar, 5 μm. (F) Bar graphs representing the proportion of large-budded cells with or without of GFP-Bub1 signals at 120 and 240 min of nocodazole treatment. N = 3, n>100 cells analyzed for each experiment. Error bars represent SD. (G) Quantification of GFP-Bub1 fluorescence intensity values in Aurora B^Ipl1 (over)expressed and repressed conditions at 120 min of nocodazole treatment. White circles represent the average intensity values of replicates. The mean of the average intensity values of the replicates is represented as a black bar. Error bars represent SD. N = 3, n> 79 cells analyzed for each experiment. An unpaired t-test was used to test the significance, *p = 0.0439.

https://doi.org/10.1371/journal.pgen.1011552.g003

To confirm these observations further, we examined GFP-Bub1 localization dynamics in cells depleted of Aurora B^Ipl1 and treated with nocodazole (Fig 3E). Aurora B^Ipl1-depleted cells subjected to nocodazole treatment for 120 and 240 min were able to load Bub1 initially but failed to maintain Bub1 at kinetochores upon prolonged treatment with nocodazole (Fig 3E and 3F), similar to what was observed in sgo1Δ mutant cells. We also noticed a significant reduction in the levels of Bub1 at kinetochores in Aurora B^Ipl1-depleted cells similar to that of sgo1Δ mutants (Fig 3G). These results reveal that Aurora B^Ipl1 in C. neoformans helps maintain Bub1 at kinetochores in the presence of nocodazole.

Shugoshin antagonizes the recruitment of PP1 at kinetochores in response to nocodazole treatment

The kinase activity of Aurora B^Ipl1 and the phosphatase activity of PP1 are antagonistic at kinetochores [45,46]. We hypothesized that a reduction in Aurora B^Ipl1 levels leads to increased PP1 levels (Fig 4A), and higher PP1 levels may lead to the loss of Bub1 from kinetochores. We expressed PP1-GFP (FungiDB ORF ID CNAG_03706; S8 Fig) in a CENP-A-mCherry expressing strain (CNKB003). First, we studied the effect of sgo1Δ on PP1 localization during nocodazole treatment (Fig 4B). Both wild-type and sgo1Δ cells exhibited GFP-PP1 signals at kinetochores during the metaphase-arrested stage. Quantifying the fluorescence signal intensity of PP1 revealed a marginal but significant increase in the GFP-PP1 signal intensities at kinetochores in sgo1Δ compared to wild-type cells (Fig 4C). A similar result was observed when Aurora B^Ipl1-depleted cells were treated with nocodazole (Fig 4D and 4E). The SAC can be silenced by dephosphorylation of MELT repeats on Knl1^Spc105 with the help of PP1. As a result, checkpoint components are removed from kinetochores [20,47–51]. Premature recruitment of PP1 to kinetochores is detrimental and leads to chromosome mis-segregation [22]. Based on the results described above, we propose that the increase in PP1 levels at kinetochores leads to the loss of Bub1 signals in the absence of shugoshin or Aurora B^Ipl1, which interferes with the SAC maintenance in MT depolymerizing conditions.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Sgo1-mediated Aurora B^Ipl1 recruitment prevents excess PP1 phosphatase at the kinetochore.

(A) The effect of sgo1Δ on maintaining a balance between Aurora B^Ipl1 and PP1 levels at kinetochore in C. neoformans. (B) Sub-cellular localization of CNKB003 (GFP-PP1 mCherry-CENP-A SGO1) and CNSD180 (GFP-PP1 mCherry-CENP-A sgo1Δ) cells treated with nocodzole (1 μg/mL) for 120 min. Yellow ovals highlight PP1-GFP signals. (C) Quantification of GFP-PP1 fluorescence signal intensity values in the indicated strains treated with nocodazole. N = 3, n>100 cells were analyzed for each experiment. White circles represent the average GFP-PP1 signal intensities calculated for each experimental repeat. An unpaired t-test was used to determine statistical significance. **p = 0.0044. The mean of the average intensity values of replicates is represented as a black bar. Error bars represent SD. (D) Microscopic images of CNKB023 (GFP-PP1 mCherry-CENP-A GAL7-3xFLAG AURORA B^IPL1) Aurora B^Ipl1 (over)expressed and repressed conditions treated with nocodazole (1 μg/mL) for 120 min. (E) Quantification of fluorescence intensity values of GFP-PP1 signals in conditions mentioned above. N = 4, n = 100 cells for each experiment. White circles represent the average intensity values of replicates. Error bars represent SD. The mean of the average intensity values plotted is represented as a black bar. Paired t-test was used to determine statistical significance. **p = 0.0069.

https://doi.org/10.1371/journal.pgen.1011552.g004

The tension-sensing function of shugoshin is conserved in C. neoformans

Shugoshin senses tensionless attachments by recruiting several factors that play a role in correcting tensionless attachments [40]. Cohesin is a molecular glue that holds two sister chromatids together during mitosis until metaphase. Cohesin plays an essential role in generating tension that resists the pulling force of MTs, aiding high-fidelity chromosome segregation [52,53]. Loss of cohesion at centromeres results in tensionless kinetochore-MT attachments. To test if Sgo1 in C. neoformans detects tensionless attachments, we depleted one of the cohesin subunits, Scc1/Mcd1/Rad21, by placing it under the GAL7 promoter both in the wild-type and sgo1Δ mutant cells. C. neoformans Scc1 (FungiDB ORF ID CNAG_01023) harbors N- and C-terminal Scc1 domains (S9A and S9B Fig). We show that like S. cerevisiae and S. pombe, SCC1 in C. neoformans is essential for viability (S9C Fig). Upon depletion of Scc1, a high proportion of wild-type cells are arrested at the large-budded stage (S9D–S9F Fig). In contrast, the sgo1Δ mutants depleted of Scc1 did not show such an increase in the proportion of large-budded cells, indicating that shugoshin plays a role in detecting tensionless kinetochore-MT attachments and delays the cell cycle progression.

Shugoshin is enriched at the spindle pole bodies (SPBs) and spindle during mitosis

Thus far, we reveal an unknown role of Sgo1 in maintaining Bub1 through the balance of Aurora B^Ipl1 and PP1 at kinetochores in C. neoformans. Next, we examined the cell cycle localization dynamics of Sgo1 in C. neoformans. Shugoshin localizes to the kinetochore during cell division in most species studied [40] (S11 Fig). To find the sub-cellular localization of Sgo1, we expressed GFP-Sgo1 from its native locus in the CENP-A-mCherry tagged strain. We synchronized the cells using hypoxia to examine the cell cycle stage-specific localization of Sgo1. No signals of Sgo1 could be observed during most of the interphase. In G2, where all centromeres cluster as a single punctum [30,33], we observed Sgo1 signals near clustered centromeres (Fig 5A). Upon entry into mitosis, Sgo1 appeared as two distinct puncta that flanked the bar-like signals of CENP-A. Apart from the punctate signals of Sgo1, we observed weak signals along the pole-to-pole axis during metaphase and anaphase that resembled spindle-like localization. To examine whether Sgo1 localizes to spindle pole bodies (SPBs), we expressed Tub4-mCherry (γ-tubulin) in the GFP-Sgo1 expressing strain. GFP-Sgo1 signals colocalized with SPBs during G2 and mitotic stages (Fig 5B), indicating that Sgo1 localizes predominantly to SPBs during the mitosis in C. neoformans.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 5. Kinetochore proximal association of Sgo1 in MT-depolymerizing conditions.

(A-D) Left, microscopic images of GFP-Sgo1 localized with a kinetochore marker, CENP-A, and an SPB marker, Tub4, at different stages of the mitotic cell cycle, in nocodazole treated and Cdc20 depleted conditions. Scale bar, 5 μm. (A) CNSD125 (GFP-SGO1 mCherry-CENP-A). (B) CNSD130 (GFP-SGO1 TUB4-mCherry). (C) Nocodazole (1 μg/mL) treated condition. Top panel, CNSD125. Bottom panel, CNSD130. (D) Cdc20 depletion condition, CNSD167 (GFP-SGO1 mCherry-CENP-A GAL7-3xFLAG-CDC20). Right, line scan profiles of both GFP (Sgo1) and mCherry (CENP-A, Tub4) fluorescence signals. Distance plotted on the x-axis represents the length of the line drawn on the image to measure fluorescence intensity. Gray values on the y-axis represent the fluorescence signal intensities.

https://doi.org/10.1371/journal.pgen.1011552.g005

To test if the spindle-like localization during mitosis depends on the spindle MT integrity, we arrested cells at metaphase using two methods. First, we used nocodazole to depolymerize MTs that activate the SAC and arrest cells at metaphase. Second, we depleted Cdc20, an anaphase-promoting complex/cyclosome (APC/C) component. Depletion or inactivation of Cdc20 leads to metaphase arrest in S. cerevisiae [54]. Upon treating C. neoformans cells with nocodazole, we observed that Sgo1 colocalized with the SPB (Fig 5C). On the other hand, depletion of Cdc20 by placing it under the galactose regulatable promoter (GAL7) resulted in metaphase arrest, and the spindle-like localization of shugoshin along the pole-to-pole axis was maintained (Fig 5D). These observations led us to conclude that the spindle-like localization of Sgo1 requires intact MTs during mitosis.

The kinase activity of Bub1 is dispensable for the function and localization of shugoshin during mitosis in C. neoformans

Studies from several yeast species and humans reveal that Bub1 is one of the major factors that recruit shugoshin to kinetochores. Although Bub1 plays a critical role, several other proteins such as cohesin, heterochromatin protein (HP1), CPC, and CENP-A also contribute to the targeting of shugoshin to kinetochores, see review [40] (Fig 6A). In the present study, to test if the Bub1 kinase activity required for shugoshin’s function is evolutionarily conserved in C. neoformans, we utilized the kinase-dead (bub1-kd) mutant of Bub1. In vitro assays reported earlier [34] revealed that Bub1 could phosphorylate the reconstituted human H2A containing nucleosomes, and kinase-dead mutants of Bub1 fail to phosphorylate. We performed a genetic interaction assay by scoring the TBZ sensitivity of sgo1Δ, bub1-kd, and sgo1Δ bub1-kd mutants (Fig 6B). If both Sgo1 and Bub1 kinase activity function in the same pathway in C. neoformans, one would expect little to no change in the sensitivity of double mutants compared to either of the single mutants, sgo1Δ and bub1-kd. By dilution spotting assay, we observed that the bub1-kd sgo1Δ double mutant cells behaved like that of the sgo1Δ mutant (Fig 6B). Moreover, the sensitivity of sgo1Δ mutant cells to TBZ is higher than that of bub1-kd mutant. There are two possibilities for the above observations. First, the enhanced sensitivity of the sgo1Δ mutant could be due to the additional functions of the shugoshin that do not rely on the kinase activity of Bub1. Second, the kinase activity of Bub1 does not play a significant role in dictating the kinetochore/kinetochore proximal localization of Sgo1. To test these possibilities, we have localized GFP-Sgo1 and performed a genetic interaction assay using mutants known to disrupt Bub1 kinase-mediated targeting of shugoshin in other species.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 6. The kinase activity of Bub1 is dispensable for shugoshin-mediated maintenance of the spindle assembly checkpoint.

(A) Schematic of the conserved pathway of shugoshin recruitment to kinetochores. (B) A ten-fold serial dilution spotting assay to score for the sensitivity of the CNSD110 (sgo1Δ), IL143 (GFP-bub1-kd), CNSD173 (sgo1Δ GFP-bub1-kd) double mutants and SHR866 (mad2Δ) to thiabendazole. (C) A ten-fold serial dilution spotting assay to score for the sensitivity of the CNSD125 (GFP-SGO1 mCherry-CENP-A), CNSD215 (GFP-SGO1 mCherry-CENP-A bub1-kd-3XFLAG), CNSD219 (GFP-SGO1 mCherry-CENP-A h2a-T121A), CNSD218 (sgo1Δ GFP-sgo1-K382A-SH1 mCherry-CENP-A), CNSD207 (sgo1Δ GFP-SGO1-SH1) and CNSD110 (sgo1Δ). (D) Microscopic images of GFP-Sgo1 localized with CENP-A at metaphase stage in wild-type and mutant backgrounds. Scale bar, 5 μm. (E) Top, Schematic showing the line scan analysis method adopted to score for the fluorescence intensity of GFP-Sgo1 along the spindle and SPBs. Middle, line scan profile of GFP-Sgo1 in the indicated strains synchronized using thiabendazole (10 μg/mL) treatment. Bottom, line scan profile of GFP-Sgo1 in the indicated strains synchronized using hypoxia. Distance on the x-axis represents the length of the line drawn on the image to measure fluorescence intensity. The y-axis represents the fluorescence signal intensity values. N = 3, n>20 cells were analysed for each experiment. Each line graph represents mean fluorescence intensity values calculated from three experimental repeats and error bars represent SD. (F) Top, Schematic of the assay used to arrest cells at metaphase by depleting Scc1. Below, microscopic images of GFP-Sgo1 localized with CENP-A metaphase in Scc1 depleted conditions. (G) Fold enrichment of Sgo1 at kinetochores in metaphase arrested cell obtained by cross-linking chromatin immunoprecipitation (ChIP) followed by quantitative polymerase chain reaction (qPCR). CEN2 and CEN5 represent centromeres 2 and 5, non-CEN region represents LEU2 locus. N = 3, and error bars represent SD.

https://doi.org/10.1371/journal.pgen.1011552.g006

The lysine residue K382 in Sgo1 in C. neoformans corresponds to the lysine K492 in Sgo1 in humans (S1B Fig). This residue is known to play a role in binding to phosphorylated H2A [26,38]. A mutation in the lysine residue of shugoshin in fission yeast (K298) or humans (K492) results in the loss of kinetochore localization of shugoshin in mitosis [26,38]. Similarly, mutating the S121 (budding/fission yeast) or T121 (humans) residue of histone H2A also abrogates the kinetochore localization of the shugoshin [26,38]. To test the role of these conserved residues in C. neoformans, we mutated K382 in Sgo1 (S1B Fig) and T121 in H2A to alanine (S10A Fig). The plasmid constructs harboring either wild-type SGO1 or a sgo1-K382A allele were integrated into the SAFE HAVEN locus (chromosome 1) [55] of the sgo1Δ strain. An overlap PCR construct harboring the h2a-T121A mutation was transformed into a strain expressing GFP-SGO1 and mCherry-CENP-A. If the kinase activity of Bub1 is indeed dispensable for shugoshin’s checkpoint function during mitosis, the strain carrying the GFP-sgo1-K382A allele would rescue the sensitivity to spindle poisons. Similarly, strain harboring the h2a-T121A allele would not show any sensitivity. Spotting assays on TBZ plates revealed that both sgo1-K382A and h2a-T121A mutants behaved like wild-type cells (Fig 6C). To further confirm these results, we localized Sgo1 in the above-mentioned mutant strains. Since Sgo1 exhibits both SPB and spindle-like localization during metaphase, we have obtained metaphase cells using hypoxia or thiabendazole arrest followed by release to score for the localization (refer to materials and methods for details). In both budding and fission yeasts, bub1-kd, h2a-SA, and sgo-KA mutants abolish kinetochore localization of shugoshin [26]. In sharp contrast, we observed that sgo1-K382A, h2a-T121A, and bub1-kd mutants in C. neoformans did not abolish the SPB or spindle-like localization of Sgo1 during metaphase (Fig 6D and 6E). Sgo1 levels were unaffected in the above mutants (S10B Fig). These results confirm that the kinase activity of Bub1 is dispensable for the recruitment of shugoshin and its SAC maintenance function in C. neoformans.

In C. neoformans, Sgo1 predominantly localizes to SPBs and the spindle during metaphase. The metaphase spindle localization of Sgo1 overlaps with CENP-A (Fig 5A). To probe if the kinetochores at the metaphase stage bind to Sgo1, we performed chromatin immunoprecipitation (ChIP) assay in Scc1-depleted cells (Fig 6F, top panel). Microscopy analysis of Scc1-depleted cells showed both SPB and spindle-like localization of Sgo1 (Fig 6F, bottom panel) as observed previously (Fig 5). ChIP assay indicates that there is only a minor enrichment of Sgo1 at centromeres in C. neoformans (Fig 6G), suggesting that the association of Sgo1 with centromeres could be transient in nature and/or only a minor pool of Sgo1 is associated with kinetochores and the majority could be along the spindle.

Strains and media growth conditions

C. neoformans cultures were grown and maintained on YPD (1% yeast extract, 2% peptone, and 2% dextrose) and YPG (1% yeast extract, 2% peptone, and 2% galactose) media at 30°C unless specified. YPD+1M sorbitol or YPG+1M sorbitol plates were used for transformation. For the selection of transformants, 100 μg/mL of nourseothricin (NAT, clonNAT, Werner BioAgents), 180 μg/mL neomycin (G-418, Sigma-Aldrich), and 220 μg/mL of hygromycin (HiMedia) were used.

Biolistic transformation

The biolistic transformation was done based on a previously described method [71]. Approximately 5 mL of C. neoformans culture was grown overnight, centrifuged at 4,000 rpm for 5 min, and resuspended in 300 μL sterile dH₂O. Around 300 μL of cell suspension was spread on the YPD+1M sorbitol plate and dried. Gold micro-carrier beads (0.6 μm) coated with 2–5 μg of DNA were prepared, and 10 μL of 2.5 M CaCl₂, 2 μL of 1M spermidine free base (Sigma-Aldrich) were added. The mixture was vortexed for 30 s, incubated for 10 min at room temperature, centrifuged, washed with 500 μL of 100% ethanol, and resuspended in 10 μL of 100% ethanol. The solution was placed on a micro-carrier disk and allowed to dry. Dried disks were placed in the biolistic transformation apparatus (Biolistic® PDS-1000/He Particle Delivery System). Helium gas pressure of approximately 1300 psi under a vacuum of a minimum of 25 inches of Hg was generated to bombard the cells with micro-carrier beads. The cells were incubated for 5–6 h in YPD+1M sorbitol medium at 30°C. Cells were then plated on YPD+NAT/NEO/HYG selection plate and incubated at 30°C for 3–4 days.

Strains and plasmids

Lists of strains, plasmids, and primers used in this study are provided in S1–S3 Tables. The construction of C. neoformans strains and plasmids is detailed below.

GFP-SGO1 strains

The fluorescent fusion protein of shugoshin was generated by cloning the 1200 bp homology region and the SGO1 promoter into the pVY7 plasmid containing the GFP and NAT sequences. GFP was tagged to the N-terminal region of the SGO1 gene. The primers used for generating the clone are listed in S3 Table. The 1200 bp homology region and a 500 bp SGO1 promoter region were amplified from the SGO1 gene using C. neoformans (H99) genomic DNA. The individual PCR fragments were gel purified. The amplified regions were cloned using BamHI, SacI, and NcoI sites. The final clone was linearized using HindIII and transformed into H99 and CNVY101. The transformants obtained were confirmed by PCR.

sgo1Δ strains

An overlap PCR strategy was used to delete SGO1 gene in the desired strains. Around 1000 bp region upstream to the start codon (ATG) and 1000 bp downstream region of the stop codon of the SGO1 ORF was amplified using C. neoformans (H99) genomic DNA. The NEO resistance marker fragment was amplified from the pLK25 plasmid (S2 Table). The individual PCR fragments were gel purified. Using these individual PCR amplified fragments, a final overlap cassette of ~3800 bp was constructed using primers (S3 Table). The final cassette was used to transform C. neoformans. The transformants obtained were confirmed by PCR. To delete the SGO1 gene in different strain backgrounds, the NEO marker was swapped with either NAT or HygB genes amplified from the pVY7 and pSH7G plasmids (S2 Table) using the same set of primers. The sgo1Δ cassette was used to transform H99, CNVY108, SHR741, SHR830, IL102, IL143, CNKB003 and CNSD190 strains. The transformants obtained were screened by PCR.

TUB4-mCherry strain

An overlap PCR strategy was used to tag Tub4 with mCherry at C-terminus. Primers used for overlap cassette construction are listed in S3 Table. Around 1000 bp region upstream to the stop codon (without including the stop codon) and 1000 bp 3’ UTR region after the stop codon was amplified from the TUB4 ORF using C. neoformans (H99) genomic DNA. The mCherry sequence with antibiotic resistance marker NEO was amplified from the pLK25 plasmid (S2 Table). The individual PCR fragments were gel purified. A final overlap cassette of ~4500 bp was constructed using individual PCR fragments. The final cassette was transformed into CNSD127 strain. The transformants obtained were confirmed by PCR.

GAL7-3xFLAG tagged SCC1, AURORA B^IPL1, CDC20 strains

To place the desired gene under regulatable GAL7p along with an N-terminal 3xFLAG sequence, an overlap PCR strategy was used. Around 1000 bp regions upstream (US) and downstream (DS) of the start codon (ATG) of the ORFs of interest were amplified using C. neoformans (H99) genomic DNA. GAL7 promoter with HygB resistance gene was amplified from pSH7G plasmid, and the 3xFLAG sequence was introduced using a primer (S3 Table). The individual PCR fragments were gel purified. A final overlap cassette was assembled using the individual purified fragments. The cassettes generated were transformed into C. neoformans and the positive transformants were confirmed by PCR. The 3xFLAG tagged conditional mutants of SCC1, AURORA B^IPL1, and CDC20 generated using the strategy mentioned above are listed in S1 Table.

AURORA B^IPL1-3xGFP mCherry-CENP-A strain

The H99 strain expressing Aurk B^Ipl1-3xGFP and mCherry-CENP-A was generated by the random integration of the plasmid pLKB71 (S2 Table). The mCherry-CENP-A construct is expressed using the H3 promoter. To create the strain, the plasmid backbone of pLKB71 was digested with HindIII and transformed into CNNV113, and the transformants were selected on the YPD-agar plate containing hygromycin selection. The colonies obtained were screened using a Zeiss Axio Observer 7 widefield microscope for the transformants positive for the mCherry-CENP-A signal.

GFP-PP1 mCherry-CENP-A strain

The N-terminal GFP fusion protein of PP1 was generated by cloning around 1000 bp homology region and PP1 promoter into the pVY7 plasmid containing GFP and NAT sequences. Around 955 bp PP1 promoter region and 1000 bp PP1 homology region were amplified from PP1 ORF using C. neoformans (H99) genomic DNA. The PCR amplicons were gel purified. The digested amplicons of promoter and homology region were cloned into pVY7 using SacI, NcoI, and SpeI sites, respectively. The final clone was linearized using BglII and transformed into the CNVY101 strain, and the positive transformants were confirmed by PCR.

GFP-sgo1-K382A-SH1, GFP-SGO1-SH1 and GFP-sgo1-K382A-SH1 mCherry-CENP-A strains

The N-terminal GFP fusion protein of wild-type and mutant (K382A) allele of SGO1 was generated by cloning SGO1 promoter, SGO1 ORF, and 3’UTR into the pSD29 plasmid (S2 Table) containing the GFP, HygB and SAFE HAVEN sequences. Around 2340 bp of SGO1 ORF 315 bp of SGO1 promoter, and 436 bp of 3’UTR were amplified from the SGO1 gene using C. neoformans (H99) genomic DNA. The K382A mutation in the SGO1 ORF was introduced using overlap PCR. The primers used are listed in S3 Table. The individual PCR fragments were gel purified. The purified fragments of the SGO1 promoter, SGO1 ORF, and 3’UTR were cloned sequentially using SpeI, BamHI, HpaI, NheI, and ApaI sites. The final clone was linearized using XcmI and transformed into CNSD113. The transformants obtained were confirmed by PCR, and Sanger sequencing was performed to confirm the presence of the K382A mutation. The strain harbouring GFP-sgo1-K382A allele at SAFE HAVEN locus was transformed with mCherry-CENP-A-NAT plasmid linearized using BglII and the positive transformants were confirmed by microscopy.

h2a-T121A GFP-SGO1 mCherry-CENP-A strain

A 1503 bp fragment of H2A ORF and 3’ UTR was amplified by PCR using C. neoformans (H99) genomic DNA and the T121A mutation was introduced into the 1503 bp fragment using overlap PCR strategy. A second fragment containing a Hygromycin marker was amplified from pSH7G. The homology region downstream of H2A 3’ UTR was amplified by PCR using C. neoformans (H99) genomic DNA. The primers used for generating the construct are listed in S3 Table. The three PCR fragments were subsequently fused by overlap PCR. The final overlap product was used to transform the CNSD125 strain using the biolistic transformation method, and the transformants obtained were confirmed by PCR, and Sanger sequencing was performed to confirm the presence of mutation within the gene.

bub1-kd 3xFLAG GFP-SGO1 mCherry-CENP-A strain

The 3xFLAG-tagged bub1-kd cassette was generated by cloning the 1200 bp homology region of BUB1 ORF harboring point mutations into the pSH7G plasmid containing the HYG sequences. The primers used for generating the clone are listed in S3 Table. The 1200 bp homology region was amplified from the bub1-kd allele using genomic DNA obtained from the IL143 (GFP-bub1-kd) strain. The individual PCR fragments were gel purified. The amplified regions were cloned using SpeI and KpnI sites. The final clone was linearized using ApaI and transformed into CNSD125 background using the biolistic transformation method. The transformants obtained were confirmed by PCR.

Thiabendazole sensitivity assay

For assaying sensitivity to thiabendazole, the O.D₆₀₀ values of overnight cultures of wild-type and mutants used in this study were estimated, and serial dilutions starting from 10⁵ cells to 10¹ cells were spotted on YPD-agar, DMF (Dimethyl Formamide); control, and YPD-agar plates with different concentrations of thiabendazole (2.0, 2.5, 3.0 and 4.0 μg/mL). The plates were incubated at 30°C for 24 to 48 h and imaged using a Bio-Rad chemidoc imaging system.

Synchronization of C. neoformans by inducing physical hypoxia

For the synchronization of C. neoformans cells, we have used a protocol developed by Ohkusu et al., [42] which utilizes a two-step protocol for the synchronization. C. neoformans cells, when grown in hypoxia conditions, result in unbudded G2 arrest, and cells release synchronously once shifted to extensive aeration conditions. Hypoxia synchrony is a two-step process. First, cells were grown at moderate aeration to 4.0 O.D₆₀₀/mL in a 250 mL flask (Borosil Erlenmeyer flask with a screw cap was used for this experiment) with 50 mL of 1% YPD at 100 rpm, 30°C. In the second step, the culture was diluted 1.5 times with fresh 1% YPD (final volume of 125 mL) and then incubated further for 5 h. The second step induces physical hypoxia, and at the end of the incubation, most cells get arrested at the unbudded G2 stage.

For the synchronization experiments, we used an overnight culture grown in 1% YPD to inoculate the secondary culture at 0.1 O.D₆₀₀/mL in 1% YPD and grown to 1 O.D₆₀₀/mL at 180 rpm, 30°C. This log phase culture was used for hypoxia synchrony, as mentioned above. The percentage of cells in the unbudded G2 stage was confirmed using a microscope. The culture was diluted to 1 O.D₆₀₀/mL using fresh media, transferred to a sterile falcon or conical flask with appropriate culture volumes, and released under extensive aeration conditions (180 rpm, 30°C).

GFP-Bub1 and Aurora B^Ipl1-3xGFP retention assay

To study the localization dynamics of GFP-Bub1 and Aur B^Ipl1-3xGFP in response to nocodazole treatment, wild-type, and sgo1Δ mutants were subjected to hypoxia synchrony, and the cells were released into 1% YPD media containing 1 μg/mL nocodazole. Time points were collected every 40 minutes until 280 min. The cells collected were imaged using a Zeiss Axio Observer 7 widefield microscope. A circular ROI of 12*12 or 6*6 pixels was used to measure the signal intensities. The presence or absence of GFP-Bub1 signals was scored manually. The cells that show bright punctate localization of Bub1 above the cell background were considered positive for Bub1.

Imaging GFP-Sgo1 localization in mCherry-CENP-A and TUB4-mCherry strains

For imaging GFP-Sgo1 colocalization with CENP-A- and Tub4-mCherry tagged proteins, we synchronized cells using physical hypoxia. Cells were released into 1% YPD and collected every 10 min starting from 30 to 90 min. Cells were imaged using a Zeiss Axio Observer 7 widefield microscope.

Cell cycle synchronization using thiabendazole (TBZ)

In order to obtain metaphase cells, the overnight-grown culture of Cryptococcus was inoculated at 0.2 O.D₆₀₀/mL and grown to log phase. The log phase culture was treated with 10 μg/mL thiabendazole for 2 h. The cells arrested at G2/M were washed twice with 2% YPD and released into fresh pre-warmed 2% YPD by incubating at 30°C, 180 RPM for 2 min. The cells were harvested by spinning at 5000 RPM for 30 seconds, washed twice with sterile water, and imaged using a Zeiss Axio Observer 7 widefield microscope.

Line scan analysis of GFP-Sgo1 signal at metaphase stage

For measuring the fluorescent intensity of Sgo1 at metaphase in wild-type and the bub1-kd, h2a-T121A, and sgo1-K382A strains, a line of 2-pixel thickness and a length of approximately 3.4 μm was drawn using a line tool available in Fiji [72]. A single Z-plane containing both SPB and spindle signals was considered for intensity measurement. The fluorescence intensity values along the line were measured using the plot profile function. The intensity values of the GFP signal obtained from different cells were averaged and divided by the average background of the cells. The normalized intensity values were plotted using GraphPad Prism.

Budding index calculation

The proportion of unbudded, small-budded, and large-budded cells was calculated based on the budding index (diameter of daughter cell/diameter of mother cell). The diameters of both mother and daughter cells were calculated along the mother-daughter axis using the line tool in Fiji [72].

Fluorescence microscopy

For widefield microscopy, Zeiss Axio Observer 7 equipped with 100x Plan Apochromat 1.4 NA objective, Colibri 7 LED light source, motorized XYZ stage and PCO.edge 4.2 sCMOS camera was used. Images were acquired at 2x2 or 4x4 binning. Zen blue edition (2.3) software was used for controlling the microscope components and acquisition. For imaging, cells grown in culture media were washed twice with autoclaved dH₂O and resuspended in water. The cell suspension was placed on microscope cover glass and imaged.

Time-lapse microfluidic assays

The utility of microfluidics for single-cell analysis in Cryptococcus was demonstrated previously in studies of aging [73]. We used the Alcatras cell traps [74] incorporated into devices allowing for use with multiple strains [75,76]. Devices were moulded in polydimethylsiloxane (PDMS) from a SU8-patterned wafer with an increased thickness of 7 μm to accommodate the larger size of C neoformans cells compared to S cerevisiae (manufactured by Micro-resist, Berlin, design available on request). Imaging chambers for individual strains were isolated by arrays of PDMS pillars separated by 2 μm gaps. This prevents the intermixing of strains while cells experience identical media conditions. Before use, the devices were filled with synthetic complete (SC) media, supplemented with 0.2 g/L glucose, and containing 0.05% w/v bovine serum albumin (Sigma) to reduce cell-cell and cell-PDMS adhesion. Cells pre-grown to logarithmic phase in the same media (lacking the BSA) were injected into the device. An EZ flow system (Fluigent) delivered media at 10 μl per minute to the flow chambers and performed the switch to media containing nocodazole after 5 hours. This media also contained Cy5 dye to allow monitoring of the timing of the media switch. Image stacks were captured at 2-minute intervals at 4 stage positions for each strain, using a Nikon TiE epifluorescence microscope with a 60x oil-immersion objective (NA 1.4), a Prime95b sCMOS camera (Teledyne Photometrics) and OptoLED illumination (Cairn Research). Image stacks had 5 Z-sections, separated by 0.6 μm, captured using a piezo lens positioning motor (Pi).

Time-lapse microfluidic assay image processing and analysis

Time-lapse assays were performed as described [34]. Cell outlines were segmented using the baby algorithm [77]. To quantify the kinetochore retention time of GFP-Bub1 fluorescence, we create a projection of the maximum values from all GFP sections, then divide the median fluorescence of the brightest 5 pixels within each cell by the median fluorescence of the cell as a whole. This ratio reliably measures protein aggregation at an organelle and has been extensively used for quantification [78–80]. In the absence of nocodazole, cells with any outlier values for the ratio, defined as values greater than 3 scaled median absolute deviations away from the median, are removed. Cells present for fewer than 80% of recorded time points are also removed. The ratios were normalized for each strain by dividing the mean value for wild-type cells in the absence of nocodazole, measured in the same experiment. To calculate the kinetochore retention times, a threshold of 1.36 (Fig 2F and 2G) and 1.11 (S4A and S4B Fig) was applied to the ratio values at each time point to determine the presence or absence of Bub1 at the kinetochore (Figs 2G and S4B). Small gaps in localization (due to kinetochore signal dropping briefly out of focus) were removed by applying a 1-dimensional morphological closing to the thresholded image, with a 3-pixel (3 timepoint or 6 minutes) structuring element. For the temporal heat maps (Figs 2F and S4A), the data for 30 cells present for the whole experiment are selected for inclusion from each strain pseudo-randomly using the Matlab rand function. There were a small number of outlier values within the heat maps, resulting from a fault in the LED control equipment. These were removed by replacing any values in the 5th percentile of a cell trace with the median value for all time points. The raw data and code used in data processing and plotting are available on request.

Flow cytometry

5 mL of C. neoformans culture was grown overnight, and 2 O.D₆₀₀ cells were taken and washed with sterile dH2O and fixed in 1 mL of 70% ethanol overnight at 4°C. Fixed cells were centrifuged at 4000 rpm and washed with 1 mL NS buffer(10 mM Tris-HCl pH 7.5, 250 mM sucrose, 1 mM EDTA (pH 8.0), 1 mM MgCl₂, 0.1 mM CaCl₂, and 7 mM β-mercaptoethanol) and resuspended in 200 μL of NS buffer added with 2 μg of RNase and incubated at 37°C for 3–4 h. Propidium Iodide (final concentration 12 μg/mL) was added to this mixture and incubated for 30 min at room temperature in the dark. 50 μL of the sample was suspended in 2 mL of 1x PBS (Phosphate buffered saline) solution, vortexed, and sonicated for 10 s at 30% Amp. 30,000 cells were analyzed by flow cytometry using FL2-A channel on (FACSAria III; BD Biosciences) at a rate of 500–2,000 events/s.

Western blotting

For immunoblotting, whole cell lysates were prepared by harvesting 10 O.D₆₀₀ cells cultured using YPD. The cells were washed twice with sterile water and resuspended in 15% TCA (Trichloroacetic acid), and samples were frozen in a -20°C freezer. The cell suspension was thawed on ice and lysed using a mini-bead beater (Biospec product) with 1 min ON and 5 min OFF for three cycles at 3000 RPM. The cell lysate was centrifuged at 13000 RPM at 4°C for 10 min, and the supernatant was discarded. The pellet was resuspended in 2x sample buffer (0.1 M Tris HCl pH 6.8, 20% glycerol, 4% SDS, 715 mM β-mercaptoethanol and 0.2% bromophenol blue) and incubated for 5 min in a boiling water bath. The samples were cleared by spinning at 13000 RPM for 3 min, and the supernatant was loaded immediately onto an SDS-PAGE gel and separated. The proteins were transferred to a PVDF membrane (Bio-Rad) using the wet transfer method with a transfer time of two and a half hours at 90 V. The transfer efficiency was estimated by staining the blot with Ponceau S solution. The membrane was blocked using 3% skimmed milk and 2% BSA dissolved 1x PBST (0.1% Tween-20) solution at room temperature for 60 minutes. Post blocking, the blot was incubated in 2% skimmed milk and 1% BSA dissolved in 1x PBST solution containing anti-GFP at 1:3000 dilution overnight at 4°C. The membrane was washed thrice for 10 min with 1x PBS solution containing 0.1% tween-20. The blot was incubated with an anti-mouse secondary antibody (1:10000 dilution in 2% skimmed milk and 1% BSA dissolved in 1x PBST solution) for 1 hour at room temperature. The membrane was washed thrice for 10 min with 1x PBS + 0.1% tween-20, and the blot was rinsed in 1x PBS and was developed using Bio-Rad clarity ECL solution and imaged using Bio-Rad Chemidoc imaging system.

Chromatin Immunoprecipitation (ChIP)

The ChIP protocol was adapted from [81,82]. ChIP was performed in metaphase-arrested cells by depleting Scc1. Overnight cultures of CNSD185 strain grown in 2% YPG (permissive media) were washed twice with 2% YPD and inoculated at 0.2 O.D₆₀₀/mL in 150 mL of 2% YPD (Non-permissive) media. The cells were grown for 6 h at 30°C, 180 RPM. Post 6 h incubation, the extent of metaphase arrest was assessed using microscopy, and the culture was fixed by adding freshly prepared methanol-free formaldehyde at 3% final concentration and incubated at 25°C, 100 RPM for 35 minutes. The fixed cells were quenched with 1.5 M Tris HCl pH 8.0 for 5 mins at 25°C, 100 RPM. The cells were pelleted and washed with ice-cold water flash frozen using liquid Nitrogen, and stored at -80°C. The cells were lysed in a buffer containing 50 mM HEPES pH 7.4, 1% Triton-X, 140 mM NaCl, 0.1% Na-deoxycholate, 1 mM EDTA, 1x Protease Inhibitor Cocktail (Sigma) with 10 cycles of bead beating using a mini-beadbeater (BioSpec product) at 3500 RPM for 2 mins ON and 5 mins OFF on ice.

The lysate was transferred to a chilled 15 mL falcon on ice, and the cross-linked chromatin was fragmented using an ultrasonicator (Diagenode Bioruptor Pico) at high frequency, 30 s ON, 30 s OFF mode for 50 cycles to achieve the desired 250–500 size range. The sample was transferred to a chilled 2 mL tube, and centrifuged at 10000 RPM for 10 min at 4°C. The debris-free supernatant was carefully transferred to a fresh 2 mL tube and incubated with 20 μL of control trap beads (Chromo Tek) at 11 RPM for 3 h at 4°C. After incubation, bead-free supernatant was recovered by centrifugation at 500 RCF for 2 min, and GFP-trap and block beads were added to the respective tubes. The tubes were incubated at 11 RPM rotation for 12 h at 4°C. After incubation, the beads were recovered by centrifugation at 1000 RCF for 1 min at RT, and extensive washes were performed at RT with intermittent agitation at 11 RPM for 5 min. First, the beads were washed with 1 mL lysis buffer (50 mM HEPES pH 7.5, 140 mM NaCl, 1% Triton-X 100, 0.1% sodium deoxycholate, 2 mM PMSF), recovered by centrifugation at 1000 RCF for 1 min at RT and then washed 2 times each with the following buffers, with recovery at 1000 RCF for 1 min following each wash: (i) 20 mM Tris HCl pH 8.0, 200 mM NaCl, 2 mM EDTA, 1% Triton-X-100, 0.1% SDS. (ii) 20 mM Tris HCl pH 8.0, 500 mM NaCl, 2 mM EDTA, 1% Triton-X-100, 0.1% SDS. (iii) 20 mM Tris HCl pH 8.0, 250 mM LiCl, 2 mM EDTA, 1% NP-40, 1% sodium-deoxycholate, pH 8.0. (iv) TE: 10 mM Tris HCl pH 8.0, 1 mM EDTA. The IP complex was recovered by incubating the beads with 250 μL of elution buffer (10 mM Tris HCl pH 8.0, 1 mM EDTA, 1% SDS, 200 mM NaCl) at 65°C for 15 min with intermittent flicking. The process was repeated with another 250 μL buffer and pooled to obtain a 500 μL IP sample. The input volume was adjusted to 500 μL with an elution buffer. For de-crosslinking, 20 μL of 5 N NaCl was added to each sample and incubated overnight at 65°C. Next day, 20 μL 1M Tris-HCl pH 6.8, and 10 μL 0.5 M EDTA pH 8.0 were added to each tube and treated with Proteinase K (2 μL/sample) at 42°C for 1 h followed by RNase treatment (5 μL of 10 mg/mL RNase A/sample) at 37°C for 1 h. The DNA was extracted with an equal volume of Tris pH 8.0-saturated phenol:chloroform:isoamyl alcohol (25:24:1) and precipitated (with 1 mL of 100% ethanol with 100 μL 3 M sodium acetate pH 5.3) overnight at -80°C. The DNA was pelleted by centrifugation at 14000 RPM for 30 min at 4°C, washed with chilled 70% ethanol, air-dried, and finally resuspended in 20 μL TE pH 8.0. The ChIP-qPCR reactions were carried out in a total 10 μL volume using CEN and Non-CEN primers listed in S3 Table.

Statistics and microscope image analysis

P-values were assessed by unpaired, two-tailed t-test or two-way ANOVA with Tukey’s multiple comparison test using GraphPad Prism 9.00 (GraphPad software). Error bars represent the standard deviation (SD) or standard error of the mean (SEM) as mentioned for each experiment. N for each experiment is mentioned in the figure legends. Fiji software [72] was used to process and analyze microscope images. Fluorescence intensity values of signals were quantified by drawing a circular region of interest (ROI) of defined pixel size on the sum intensity Z-projected images (Figs 2E, 3C and 3G) or on single Z-plane images (Fig 4C and 4E). The intensity values were calculated by subtracting the product of mean background fluorescence intensity and area of the ROI from integrated density, and the resultant values were plotted using GraphPad Prism.