Quantitative analysis of chromosome condensation in fission yeast

Abstract

Chromosomes undergo extensive conformational rearrangements in preparation for their segregation during cell divisions. Insights into the molecular mechanisms behind this still poorly understood condensation process require the development of new approaches to quantitatively assess chromosome formation in vivo. In this study, we present a live-cell microscopy-based chromosome condensation assay in the fission yeast Schizosaccharomyces pombe. By automatically tracking the three-dimensional distance changes between fluorescently marked chromosome loci at high temporal and spatial resolution, we analyze chromosome condensation during mitosis and meiosis and deduct defined parameters to describe condensation dynamics. We demonstrate that this method can determine the contributions of condensin, topoisomerase II, and Aurora kinase to mitotic chromosome condensation. We furthermore show that the assay can identify proteins required for mitotic chromosome formation de novo by isolating mutants in condensin, DNA polymerase ε, and F-box DNA helicase I that are specifically defective in pro-/metaphase condensation. Thus, the chromosome condensation assay provides a direct and sensitive system for the discovery and characterization of components of the chromosome condensation machinery in a genetically tractable eukaryote.

Fig 1

Live imaging of chromosome condensation in fission yeast. (A) The three-dimensional distance between two fluorescent marker arrays located on the same chromosome arm is expected to decrease during mitotic chromosome condensation; (B) different chromosome loci were fluorescently labeled by integration of TetO or LacO arrays and expression of TetR-tdTomato (TetR-tdTom) or LacI-GFP; (C) images of mitosis in a cell in which chromosome I was labeled with the 1.2-Mb cen-arm array (strain C2926) (see Movie S1 in the supplemental material).

Fig 2

Quantitative analysis of condensation dynamics. (A) Three-dimensional distances (d) between GFP- and tdTomato-labeled arrays were measured in the image stacks. (B) Distances between a GFP-labeled LacO array at the centromere (cen) and tdTomato-labeled TetO arrays positioned on the left arm (arm) of chromosome I (chrI) with increasing spacing (s) from cen were recorded by time-lapse microscopy at 36°C (strains C2572, C2574, C2570, C2568, and C2566). The measurements from the indicated number (n) of individual cells were aligned to anaphase onset (time zero), and mean distances ± standard deviations (blue bars) were plotted. (C) Maximum cen-arm distances during G₂ phase (d_max) were plotted against the spacing between the marker arrays (Table 1). The gray line indicates a logarithmic fit (R² = 0.998). (D) As in panel B, plotting of distances between a tdTomato-labeled TetO array 1.5 Mb from the left telomere and GFP-labeled LacO arrays positioned with various spacings on the left arm of chromosome I (strains C2774, C2779, and C2724). (E) Distances between a GFP-labeled LacO array 1.95 Mb from the left telomere of chromosome I (armI) and a tdTomato-labeled TetO array on the right arm of chromosome II (armII) were plotted as in panel B (closed circles; strain C3245). Distances from the 1.2-Mb arm-cen combination with a similar G₂ phase distance are plotted for comparison (open circles; strain C2570).

Fig 3

Chromosome condensation in metaphase-arrested cells. (A) Thiamine (VB₁) was added to exponentially growing cultures of yeast cells expressing slp1⁺ from either its endogenous promoter or the P_nmt41 promoter (strains C1277 and C2489). The fraction of cells in interphase, metaphase, or anaphase judged by immunofluorescence staining of tubulin and Hoechst staining of DNA was scored for 100 cells at the indicated time points after VB₁ addition. (B) Cells containing the 1.2-Mb cen-arm arrays and expressing slp1⁺ from P_nmt41 and mCherry-labeled α-tubulin (strain C3365) were enriched in G₂ phase and incubated for 1 h in the presence of VB₁ before imaging at 32°C. Individual time traces were aligned to the first time that a stable metaphase spindle had formed (see Movie S3 in the supplemental material), and mean distances ± standard deviations (blue bars) were plotted. (C) Asynchronous cultures of cells expressing slp1⁺ from P_nmt41 and eCFP-labeled α-tubulin (strain C2852) were grown at 30°C for 3 h in the presence of VB₁ to arrest cells in metaphase. Then, BCM was added to half of the culture for an additional 1 h, and microtubule depolymerization was confirmed by fluorescence microscopy (see Fig. S2C in the supplemental material). Distances between 1.2-Mb cen and arm arrays in 100 cells from asynchronous or arrested populations were measured, and mean values ± standard deviations were plotted; P values were calculated from Student's t tests. (D) As in panel C, but the HDAC inhibitor TSA was added instead of BCM.

Fig 4

Chromosome condensation in condensin, Topo II, and Aurora kinase mutants. (A) Temperature-sensitive single or double mutant strains of condensin's Cut14 subunit (cut14-208) or Topo II (top2-191) containing the 1.2-Mb cen-arm array (strains C3005, C3061, and C3136) were imaged at 36°C. Time traces were aligned to anaphase onset, and mean distances ± standard deviations (blue bars) of the indicated number of cells (n) were plotted. (Insets) Sigmoid fits for each plot (red) compared to the average fit of 4 independent plots of wild-type cells (green) (see Fig. S3D in the supplemental material). (B) Wild-type or analog-sensitive ark1-as3 Aurora kinase mutant cells containing the 1.2-Mb cen-arm array (strains C2926 and C3130) were imaged after addition of 1NM-PP1 or dimethyl sulfoxide (DMSO) solvent at 32°C. Distance-time traces of the indicated number of cells were aligned to anaphase onset, and average distances ± standard deviations (blue bars) were plotted for each time point. (Insets) Sigmoid-fit curves (red) compared to the average-fit curve of 4 independent graphs obtained for wild-type cells in the absence of the ATP analogue at 32°C (green) (see Fig. S3E in the supplemental material). No sigmoid curve fit could be obtained for ark1-as3 in the presence of 1NM-PP1.

Fig 5

Chromosome condensation during S. pombe meiosis. (A) Meiosis in a strain with heterozygous dots. Anaphase I was defined by the directed movement of the centromere-proximal marker arrays toward one cell pole (left); homologous recombination between the cen and arm markers could result in arm marker splitting during anaphase I (right). Anaphase II was defined by sister marker splitting. (B) Sporulation of diploid cells in which one of the two homologs of chromosome I was labeled with the 1.2-Mb cen-arm array (strain C3064) was induced at 25°C. Distance-time traces were aligned to the first time point at which cen arrays moved toward one cell pole (anaphase I onset) or to splitting of the cen arrays (anaphase II onset), and separate mean distance ± standard deviation (blue bars) plots were generated. (C) Images of a cell passing through both meiotic divisions (see Movie S7 in the supplemental material).

Fig 6

Novel condensin mutants identified by the condensation assay. Screening of a library of conditional S. pombe mutants identified eight novel temperature-sensitive alleles in the five condensin subunits (white stars). Previously characterized condensin mutants are indicated as a reference (gray stars).