ESCO1/2's roles in chromosome structure and interphase chromatin organization

Abstract

ESCO1/2 acetyltransferases mediating SMC3 acetylation and sister chromatid cohesion (SCC) are differentially required for genome integrity and development. Here we established chicken DT40 cell lines with mutations in ESCO1/2, SMC3 acetylation, and the cohesin remover WAPL. Both ESCO1 and ESCO2 promoted SCC, while ESCO2 was additionally and specifically required for proliferation and centromere integrity. ESCO1 overexpression fully suppressed the slow proliferation and centromeric separation phenotypes of esco2 cells but only partly suppressed its chromosome arm SCC defects. Concomitant inactivation of ESCO1 and ESCO2 caused lethality owing to compromised mitotic chromosome segregation. Neither wapl nor acetyl-mimicking smc3-QQ mutations rescued esco1 esco2 lethality. Notably, esco1 esco2 wapl conditional mutants showed very severe proliferation defects associated with catastrophic mitoses and also abnormal interphase chromatin organization patterns. The results indicate that cohesion establishment by vertebrate ESCO1/2 is linked to interphase chromatin architecture formation, a newly identified function of cohesin acetyltransferases that is both fundamentally and medically relevant.

Keywords: ESCO1/2; centromere; chromatin territories; chromosome segregation; cohesin acetyltransferases; sister chromatid cohesion.

Figure 1.

ESCO2, but not ESCO1, is critical for proliferation and centromere integrity. (A) Schematic representation of the ESCO1 gene locus and gene targeting knockout construct. (Closed boxes) Exons; (Marker) drug resistance genes; (gray box) the sequence encoding the acetyltransferase domain of ESCO1. (B) RT–PCR. Disruption of the ESCO1 gene was ultimately verified by RT–PCR using an ESCO1-specific primer set. The ESCO2 gene was used as a control. (C) Growth curves of the indicated cell lines. Cells (1 × 10⁵) of the indicated genotypes were inoculated in 1 mL of medium, counted, and passaged every 24 h. (D) The SMC3-K105, K106 acetylation level was measured. Whole-cell lysates were prepared from cells of the indicated genotypes. Acetylated SMC3, SMC3, and PCNA (loading control) were detected by Western blotting. (E) Chromosomes from metaphase spreads were classified into three groups, and >100 metaphase cells were analyzed for each genotype. The results of two independent experiments are plotted. (F) Metaphase spread samples were prepared by the cytospin method after incubation with 100 ng/mL colcemid for 1 h. The distances between CENP-T signals were measured for >400 chromosomes. The same trend was confirmed from an independent biological experiment. P-values were calculated by Student's t-test.

Figure 2.

Combined deficiency in ESCO1 and ESCO2 causes lethality. (A) Scheme of the conditional double mutant's establishment. (B) Depletion of the ESCO2-3AID-6Flag protein and measurement of Ac-SMC3, SMC3, and α-tubulin (loading control) by immunoblotting. The results were confirmed with lysates from an independent biological experiment. (C) Growth curves. Cells (1 × 10⁵) of the indicated genotypes were inoculated in 1 mL of medium, counted, and passaged every 12 h. Auxin (500 µM final) was added at time 0 as required. (D) Chromosomes from metaphase spreads were analyzed as in Figure 1E. Auxin (500 µM final) was added 6 h before cell collection and analysis. (E) Lagging chromosomes in anaphase cells. At least 50 anaphase cells were analyzed for each experiment. The results of two independent experiments are plotted. Auxin (500 µM final) was added 6 h before cell collection.

Figure 3.

Proliferation and centromeric separation defects of esco2 cells are compensated for by ESCO1 overexpression. (A) ESCO1 mRNA levels were measured by quantitative PCR. (B) ESCO1-Flag protein level and the acetylation level of SMC3 were measured by Western blotting. (C) Growth curves are as in Figure 1C. (D) The distances between CENP-T signals were measured for >400 chromosomes, as in Figure 1F. (E) Growth curves are as in Figure 1C. Doxycycline (Dox; 1 µg/mL final) was added at time 0, when indicated. (F) The measurement of lagging chromosomes in anaphase cells was as in Figure 2E. Dox was added 24 h before sample collection, as indicated. (G) Chromosomes from metaphase spreads were analyzed for cohesion defects following the classification procedure outlined in Figure 1E.

Figure 4.

SMC3-K105, K106 acetylation is not singularly responsible for ESCO1's and ESCO2's roles in proliferation. (A) Scheme of the steps involved in establishing conditional ESCO1^−/−/− ESCO2^−/AID SMC3^−/QQ. (B) Chromosomes from metaphase spreads were analyzed for cohesion defects as outlined in Figure 1E. Auxin (500 µM final) was added 6 h before sample collection. (C) Growth curves are as in Figure 2C. Auxin was added at time 0, when indicated. (D) The measurement of lagging chromosomes in anaphase cells was as in Figure 2E. Auxin (500 µM final) was added 6 h before cell collection. (E) Chromosomes from metaphase spreads were analyzed for cohesion defects as outlined in Figure 1E. (F) Growth curves are as in Figure 2C. Auxin was added at time 0, when indicated. (G) Chromosomes from metaphase spreads were analyzed for cohesion defects as outlined in Figure 1E. Type IV indicates chromosomes with fully separated sister chromatids. Auxin was added 6 h before sample collection.

Figure 5.

Cohesin release in esco1 esco2 cells is suppressed by depletion of WAPL but is associated with mitotic arrest and chromosome missegregation. (A) Scheme of the steps involved in establishing the ESCO1^−/−/− ESCO2^−/AID WAPL^−/AID triple conditional mutant. (B) Depletion of ESCO2-3AID-6Flag protein and WAPL-3AID-6Flag protein was confirmed by Western blotting. Similar trends were observed in three independent experiments. (C) Chromosomes from metaphase spreads were classified in four groups (see images for the new type IV [unseparated]) (Fig. 1E), and >100 metaphase cells were analyzed for each genotype. The results of two independent experiments are plotted. Auxin (500 µM final) was added 6 h before cell collection, when indicated. (D) Immunostaining using the anti-Smc3 antibody and DAPI. After incubation with 100 ng/mL colcemid for 1 h, metaphase spread samples were prepared by the cytospin method (Abe et al. 2016). (E) Growth curves are as in Figure 2C. (F) Cell cycle progression for the indicated cell lines was analyzed by propidium iodide staining. Samples were taken at indicated time points. Auxin was added at time point 0. (G) Chromosome bridges and missegregation in anaphase. At least 50 cells for the anaphase plot were analyzed for each experiment. The results of two independent experiments are plotted. The experiments were performed as in Figure 2E. Auxin (500 µM final) was added 6 h before cell collection.

Figure 6.

ESCO1/2-mediated events are implicated in the organization of interphase chromosome territories. (A) Immunostaining using the anti-SMC3 antibody and DAPI. Samples were prepared by the cytospin method. After the centrifugation, cells were fixed with 4% PFA in PBS containing 0.1% Triton X-100. Auxin (500 µM final) was added 6 h before sample collection. (B) DAPI and immunostaining with the anti-SMC3 antibody following nuclear membrane permeabilization. Samples were prepared by the cytospin method. Before the centrifugation, cells were treated with 0.3% Triton X-100 for 5 min. Auxin (500 µM final) was added 6 h before sample collection. The signal intensities versus distance along the indicated line, A to B, were plotted using ImageJ's analytical functions. For merged graphs, each signal intensity was normalized to the mean, and normalized intensities versus micrometer were plotted. Auxin (500 µM final) was added 6 h before sample collection. (C) Immunostaining using an anti-SMC3 antibody and DAPI. Images were taken by stimulated emission depletion (STED) microscopy. Samples were prepared by the cytospin method as in A. Auxin (500 µM final) was added 6 h before cell collection.

Figure 7.

Cohesin binding to interphase chromatin is negatively regulated by ESCO1/2. Immunostaining using the anti-SMC3 antibody and DAPI without (A) or following (D) nuclear membrane permeabilization to allow visualization of cohesin in nuclei (A) or cohesin bound on chromatin (D). Auxin (500 µM final) was added 6 h before sample collection. (B,E) The nucleic area was defined by the DAPI signal. The SMC3 signals that overlapped with DAPI signals were measured and normalized to the SMC3 signal intensity in wild type and then plotted for individual representative experiments in which >100 nuclei were examined for each condition. (Middle line) Median; (box) 25th and 75th percentiles; (bars) 5th and 95th percentiles. (C,F) The median values of three independent experiments were averaged, and plotted. P-values were calculated by Student's t-test. (*) P < 0.05; (**) P < 0.01.