Chromosome Segregation in Closed Mitosis Under an Excess of Nuclear Envelope

Abstract

Background: Two major types of cell division occur in eukaryotic cells regarding the dismantlement or not of the nuclear envelope (NE) in mitosis, open and closed mitosis, respectively. In the budding yeast Saccharomyces cerevisiae, the prototypical model for closed mitosis, the Nem1-Spo7 phosphatase complex, which regulates lipid metabolism, plays a key role in coordinating NE expansion throughout the cell cycle. Indeed, Nem1 depletion leads to abnormal NE evaginations in interphase, which protrude the ribosomal DNA (rDNA) and the nucleolus. However, the specific impact of these NE and chromosome organization abnormalities during chromosome segregation in anaphase remains poorly understood.

Results: Our study investigated chromosome segregation and NE dynamics during closed mitosis in relation to the presence or absence of Nem1. Nem1 was depleted by means of the auxin degron system. Nem1 depletion led to the formation of chromatin protrusions in interphase, particularly at the rDNA locus, as it has been reported before for nem1 mutants. These protrusions persisted into anaphase and were associated with delayed recoiling of the rDNA-bearing chromosome XII right arm, resulting in lagging chromatin during late anaphase. Additionally, cells can maintain nucleus-vacuole junctions (NVJs) during anaphase, suggesting that vacuoles may play a role in shaping NE morphology during chromosome segregation.

Conclusion: Our findings suggest that the Nem1-Spo7/lipin regulation of the NE size is crucial for the timely segregation of the rDNA-bearing chromosome during closed mitosis. Thus, the NE homeostasis actively contributes to chromosome segregation and the spatial organization of chromosomes in subsequent cell cycles. In addition, the persistent association between the NE and vacuoles in anaphase further underscores how cumbersome organelle interactions can become during closed mitosis, opening inspiring research avenues.

Keywords: Lipin; Nem1; Nvj1; Saccharomyces cerevisiae; chromosome segregation; closed mitosis; nuclear envelope; rDNA; vacuole.

Figure 1

Chromosome XII segregation in wild type cells growing in nutrient rich and complete minimum media. The wild type strain (FM2658) carries the following labels to follow anaphase and chromosome segregation under the fluorescence microscope: Sec61‐eCFP (nuclear envelope), Net1‐mCherry (rDNA), and the TetR‐YFP/tetOs system for spotting the centromere‐proximal rDNA flank (tetO:450) on the right arm of chromosome XII and its corresponding telomere (tetO:1061). The strain was grown to log‐phase at 25°C in the corresponding nutrient‐rich (YPD) or complete minimum (SCcshl) media (mean ± SEM, n = 3). Micrographs were taken either directly (A and B) or after fixation in formaldehyde followed by nuclear DNA staining with DAPI (C–G). (A) Representative cells at different cell cycle stages. Early and late anaphase were differentiated based on the morphology and length of Sec61 hourglass NE. (B) The degree of resolution and segregation of the right arm of chromosome XII (cXIIr) was assessed relative to the NE morphology. In early anaphase (short hourglass NE), all cells presented three tetOs spots, denoting resolution of tetO:450 only. In late anaphase (long hourglass NE), nearly 90% of cells had four tetOs spots, implying that the whole cXIIr has been resolved. (C) Single cell analysis of nuclear length relative to the cXIIr resolution/segregation status (3 vs. 4 tetOs) (mean ± SD). Nuclear length was determined by measuring distal DAPI signals in the anaphase nucleus. Note that full cXIIr resolution associates with a more elongated nucleus, although overlap exists. Plots of a single experiment are shown. (D) Average nuclear length relative to the cXIIr resolution/segregation status (mean ± SEM, n = 3). Note that partial and full cXIIr resolution center around nuclear lengths of ∼4.2 and 6.2 µm, respectively. However, there is no difference between the growth media. (E) The degree of rDNA segregation in all anaphases (early plus late). Note that there is a ∼15% increase in the SCcshl medium. (F) The degree of resolution of the cXIIr before rDNA segregation. (G) The degree of DAPI bridges after rDNA segregation. Note that these chromatin bridges are rare in both growth media.

Figure 2

Interphase chromatin protrusions after Nem1 depletion. The strain (FM2748) is isogenic to the one shown in Figure 1, but carries the following labels: Hta2‐mCherry (chromatin), Net1‐eCFP (rDNA), and TetR‐YFP/tetO:1061 (cXIIr telomere). It also carries the nem1:aid* allele and the OsTIR1 system for depleting Nem1‐aid* after auxin (IAA) addition. (A) Representative micrographs of log‐phase FM2748 cells without and with rDNA chromatin protrusions. Cells in G1, S/G2 and G2/M cells are shown. In all cases, protrusions (white arrows) correspond to the rDNA locus. (B) Quantification of chromatin protrusions relative to cell cycle stages shown in (A), growth media (YPD vs. SCcshl), and − vs. + IAA for 24 h (mean ± SEM, n = 3). FM2748 (nem1:aid*) and its wild type (WT) parental FM2707 (as FM2748 but with a wild type NEM1) were assessed. Note that depletion of Nem1 causes a single rDNA protrusion in most cases. They are observed in G1 and S/G2/M. (C) Characterization of rDNA shape in the +IAA (24 h) protrusions (mean ± SEM, n = 3). (D) Relative location of the cXIIr telomere in the +IAA (24 h) protrusions (mean ± SEM, n = 3). (E) As in (B) but before (0 h) and just 3 h after IAA addition in the FM2748 strain (mean ± SEM, n = 3). (F) Characterization of the rDNA shape in the +IAA (3 h) protrusions (mean ± SEM, n = 3). (G) Relative position of the cXIIr telomere in the +IAA (3 h) protrusions (mean ± SEM, n = 3). (H) Protrusion length after 3 and 24 h in IAA (mean ± SD). One way ANOVA followed by the Tukey test was applied for statistical assessment (***p < 0.001; **p < 0.01; *p < 0.05; ns, p > 0.05).

Figure 3

Chromosome XII and rDNA segregation patterns in cells depleted of Nem1 while growing in nutrient rich and complete minimum media. (A) Cell cycle distribution of log‐phase cultures in YPD and SCcshl with (−IAA) or without (+IAA) Nem1 (mean ± SEM, n = 3). (B) Schematic illustration of cXII segregation. In early anaphase (unresolved rDNA): (a) the nucleus commences its elongation through the division axis while the nucleolus/rDNA remains on one edge; (b) cXIIr resolves from centromere to telomere, which unzips the rDNA to form a bridge between mostly segregated nuclear masses. In late anaphase (segregated rDNA): (c), distal cXIIr regions are the last to be resolved and lag behind the rest of the genome; (d and e) rDNA compaction pulls distal regions to complete segregation. This latter process can occur asynchronously so that one sister cXIIr can appear in an advanced segregation state (d). (C) Representative micrographs of the drawings of (B). (D) Quantification of the different rDNA and cXIIr segregation figures (mean ± SEM, n = 3). Note that Nem1 depletion increases the percentage of late anaphase cells with lagging cXIIr. The growth media had little influence on this pattern (unpair t test for “type e” in −IAA vs. +IAA render p < 0.001 for both YPD and SC). (E) Transitions between the five anaphase types in the presence and absence of Nem1 (N, number of cells followed by time‐lapse super‐resolution microscopy; each colored line corresponds to a cell). Time zero corresponds to when the cell entered anaphase (type a).

Figure 4

The NE‐vacuole axis in anaphase with and without Nem1. An asynchronous log‐phase culture of FM2748 was stained with MDY‐64 dye under conditions that label both the NE and the vacuolar membrane. MDY‐64 is read in the CFP channel, and its brightness overshadowed Net1, which could not be assessed here. (A) Seven representative anaphase cells. Note the close interaction between the NE and vacuoles. Lagging cXIIr often bends around or is attached to vacuole(s). (B) Single cell analysis of the NE perimeter at different cell cycle stages in the presence and absence of Nem1 (mean ± SD). One way ANOVA was applied for statistical assessment. The effect of Nem1 for each cell cycle stage was the only post hoc comparison included (*p < 0.05; ns, p > 0.05). (C) As in (B), but only counting late anaphase cells (i.e., elongated or binucleated Hta2 with two tetO:1061). The two‐tailed Student's t test was applied. (D) Vacuole number per cell (all cell cycle stages) with and without Nem1 (mean ± SEM, n = 3; unpair t test for “>3 vacuoles” renders a p = 0.0134).

Figure 5

Nvj1‐Vac8 nucleus‐vacuole junctions with and without Nem1. (A) Representative log‐phase cells from a NVJ1‐YFP VAC8‐mCherry NEM1 strain (FM2793) growing in SCcshl. The top images were taken by wide‐field fluorescence microscopy (WF), while the bottom images were taken by Airyscan 2 super‐resolution microscopy (SR). Four mitotic cells are numbered in the WF example. Note the variability of NE labelling by Nvj1. Cells #3 and #4 have Nvj1/Vac8 spots indicating NVJs. In the SR example, the cell on the left is in early anaphase and shows an Nvj1‐Vac8 patch. (B) As in (A) but with the addition of MDY‐64 dye. Both examples are from SR images and include cells in late anaphase. The upper cell contains an Nvj1‐Vac8 spot, whereas the lower cell does not show NVJ, but the NE is still tightly associated with the vacuole (white arrowhead). Scale bar corresponds to 5 µm. (C) Percentage of cells displaying Nvj1‐Vac8 spots in a NVJ1‐YFP VAC8‐mCherry nem1:aid* strain (FM3311) grown overnight in SCcshl with (−IAA) and without (+IAA) Nem1 (mean ± SEM, n = 3). Cells were grouped according to the cell cycle stage they were in (anaph is anaphase). (D) Length of Nvj1 patches (mean ± SD; all cells included regardless of cell cycle stage). The two‐tailed Student's t‐test was used to compare between the presence and absence of Nem1.

Figure 6

Summary and models of how an excess of NE leads to lagging chromosome arms. In the presence of Nem1 (upper diagram), the NE is maintained at a size appropriate for NE elongation in anaphase. We propose that the NE surface is short enough to exert some tension on the NE bridge of the long hourglass late anaphase nucleus, which might favor karyokinesis and stabilization of type e late anaphases. In the absence of Nem1 (lower diagram), the excess of NE would be absorbed by the bridge, resulting in less tension and affecting karyokinesis and NE retraction, and thus the recoiling of cXIIr (more type c late anaphases). This in turn would lead to the perpetuation of the rDNA/cXIIr protrusions across generations.