Human Wapl is a cohesin-binding protein that promotes sister-chromatid resolution in mitotic prophase

Abstract

Background: The linkage between duplicated chromosomes (sister chromatids) is established during S phase by the action of cohesin, a multisubunit complex conserved from yeast to humans. Most cohesin dissociates from chromosome arms when the cell enters mitotic prophase, leading to the formation of metaphase chromosomes with two cytologically discernible chromatids. This process is known as sister-chromatid resolution. Although two mitotic kinases have been implicated in this process, it remains unknown exactly how the cohesin-mediated linkage is destabilized at a mechanistic level.

Results: The wings apart-like (Wapl) protein was originally identified as a gene product that potentially regulates heterochromatin organization in Drosophila melanogaster. We show that the human ortholog of Wapl is a cohesin-binding protein that facilitates cohesin's timely release from chromosome arms during prophase. Depletion of Wapl from HeLa cells causes transient accumulation of prometaphase-like cells with chromosomes that display poorly resolved sister chromatids with a high level of cohesin. Reduction of cohesin relieves the Wapl-depletion phenotype, and depletion of Wapl rescues premature sister separation observed in Sgo1-depleted or Esco2-depleted cells. Conversely, overexpression of Wapl causes premature separation of sister chromatids. Wapl physically associates with cohesin in HeLa-cell nuclear extracts. Remarkably, in vitro reconstitution experiments demonstrate that Wapl forms a stoichiometric, ternary complex with two regulatory subunits of cohesin, implicating its noncatalytic function in inactivating cohesin's ability to interact with chromatin.

Conclusions: Wapl is a new regulator of sister chromatid resolution and promotes release of cohesin from chromosomes by directly interacting with its regulatory subunits.

Figure 1

Cell Cycle Dynamics of Wapl in HeLa Cells. (A) Total cell lysates were prepared from HeLa cells (lane 1) and MCF7 cells (lane 2), and analyzed by immunoblotting with an antibody against an N-terminal recombinant fragment of Wapl. (B) HeLa cells were synchronized by means of double-thymidine block and release. Total cell lysates were prepared at time intervals, and analyzed by immunoblotting with the antibodies indicated. Mitosis-specific mobility shift of Incenp and the level of cyclin B were used as markers to monitor cell cycle progression. α-tubulin was used as a loading control. The final lane contains a lysate prepared from an asynchronous culture (Asyn). (C) HeLa cells were arrested in mitosis with nocodazole for 16 h and then released. Total cell lysates were prepared at time intervals, and analyzed as above. (D) HeLa cells were pre-extracted with a buffer containing 0.5% Triton X-100, fixed with paraformaldehyde, and stained with anti-Wapl and DAPI. Bar, 10 μm. (E) Cell lysates were prepared from asynchronously growing HeLa cells (-Noco) and from HeLa cells treated with nocodazole for 16 hr (+Noco). The total cell lysates (lanes 1 and 5) were separated into soluble cytoplasmic fraction (S2, lanes 2 and 6), soluble nucleoplasmic fraction (S3, lanes 3 and 7) and chromatin-enriched fractions (P3, lanes 4 and 8), and they were analyzed by immunoblotting with the indicated antibodies (top four rows). The lowest row represents the level of histones recovered in each lane, which were visualized by Coomassie blue stain.

Figure 2

Depletion of Wapl Produces an Increased Population of Prometaphase Cells. (A) HeLa cells were transfected with two independent Wapl siRNAs and analyzed for protein levels by immunoblotting. Mock-transfected cells without siRNAs served as controls. (B) Chromosome spreads were prepared (without colcemid treatment) from the control and Wapl-depleted cells after hypotonic treatment, and stained with anti-aurora B (AuB) and anti-Smc2 (left panels). Bar, 5 μm. (C) The mitotic figures on the spreads were classified into “prometaphase-like” (B, lower panels) and “metaphase-like” (B, upper panels), and their numbers are plotted (Colcemid -). Chromosome spreads were also prepared from cells treated with colcemid for 3 hr, and analyzed similarly (Colcemid +). (D) The control and Wapl-depleted cells were fixed and stained with anti-Mad2 or anti-Bub1. The frequency of mitotic cells positive for Mad2 and Bub1 was plotted. (E) HeLa cells were treated with thymidine 8 h after mock-transfection (control) or Wapl siRNA transfection, incubated for another 15 h, and then released from the thymidine block. The frequency of mitotic cells was assessed by staining with anti-phospho-H3 (at serine 10) and DAPI. (F) In the same experiments as described in (E), cell lysates were prepared at the indicated time points, and analyzed by immunoblotting. The upper faint band in the third row (α-tubulin) is cyclin B signal as the blot was first probed with cyclin B and then with α-tubulin without stripping. (G) and (H) HeLa cells were mock-transfected (control) or transfected with Wapl siRNA, incubated for 24 h, and treated with nocodazole for 16 h. After release from the nocodazole arrest, the cells were analyzed as in (E) and (F), respectively.

Figure 3

Depletion of Wapl Produces Prometaphase-like Chromosomes with Poorly Resolved Sister Chromatids and a High Level of Cohesin. (A) HeLa cells were mock-treated (the first row) or treated with siRNA specific to Plk1 (the second row), aurora B (the third row) or Wapl (the fourth row). Chromosome spreads were prepared without colcemid treatment, and were stained with anti-topoisomerase II and anti-Smc1. The DNA was counterstained with DAPI. Bar, 10 μm. (B) Close-ups of chromosomes from the control and siRNA-treated cells. Areas indicated by the rectangles in (A) panels, are magnified here. Bar, 5 μm.

Figure 4

Functional Interactions among Wapl, Cohesin and Cohesin Regulators. (A) HeLa cells were mock transfected (Control) or transfected with Scc1, Wapl, or Scc1 and Wapl siRNAs, and cell lysates were prepared and analyzed by immunoblotting analysis using the antibodies indicated. (B) Chromosome spreads were prepared (without colcemid treatment) from the control and siRNA-transfected cells, and stained with an anti-Smc2 antibody and CREST serum. Merged images are also shown on the right. Bar, 10 μm. (C) Quantification of mitotic parameters observed in the control and siRNA-treated cells. Abnormal mitotic cells include cells displaying chromosome misalignment and abnormal anaphase-like cells indicative of premature chromatid separation. (D) Quantification of mitotic defects and chromosome phenotypes observed in control cells and cells transfected with siRNAs specific to Wapl, Sgo1, or Wapl+Sgo1 (left). The efficiency of depletion in each sample was assessed by immunoblotting (right). (E) A similar set of experiment was performed using siRNAs specific to Wapl, Esco2, and Wapl+Esco2, and analyzed as in (D). The asterisk above the Esco2 band indicates a non-specific band.

Figure 5

Overexpression of Wapl Causes Segregation Defects Reminiscent of Those Found in Cohesin-deficient Cells. (A) 293T cells were transfected with a plasmid harboring Wapl cDNA under the control of the CMV promoter in pcDNA3.1 (lane 3) or with the corresponding empty vector (lane 2). Cell lysates were prepared 42 hr after transfection and analyzed by immunoblotting along with a lysate from untransfected cells (lane 1). (B) The control and Wapl cDNA-transfected cells were fixed and stained with antibodies against α-tubulin and Sgo1. The DNA was counterstained with DAPI. Shown are representative mitotic figures for each population of cells. Bar, 10 μm. (C) Quantitation of abnormalities observed in total cells (B) and in chromosome spreads (D). (D) Chromosome spreads were prepared from the control and Wapl cDNA-transfected cells, and stained with an antibody against Smc2 (a condensin subunit) and a CREST serum. The DNA was counterstained with DAPI. Bar, 10 μm.

Figure 6

Wapl Interacts with Cohesin in HeLa Cell Nuclear Extracts. (A&B) HeLa cell nuclear extracts were subjected to immunoprecipitations with control IgG (lanes 1 and 2), anti-Smc1 (lanes 3 and 4) or anti-Wapl (lanes 5 and 6). The precipitates were washed with a buffer containing 0.1 M KCl (lanes 1, 3 and 5) or 0.5 M KCl (lanes 2, 4 and 6), and analyzed by silver staining (A) or immunoblotting using the antibodies indicated (B). 5% of the input sample was loaded in the first lane (input) of panel (B). (C) The same extracts were used for immunoprecipitations using control IgG (lanes 1 and 2), anti-Pds5A (lanes 3 and 4) or anti-Pds5B (lanes 5 and 6). The precipitates and the input sample were analyzed as above. (D) A HeLa nuclear extract was loaded onto a 5–20% sucrose gradient and centrifuged at 36,000 rpm for 15 h in an SW50.1 rotor. Fractions were subjected to SDS-PAGE and analyzed by immunoblotting. The positions of three protein standards (BSA [4.5S], catalase [11.3S] and thyroglobulin [19.4S]) are indicated.

Figure 7

In Vitro Reconstitution Reveals that Wapl Directly Associates with the Non-SMC Dimer of Cohesin. (A) Wapl was co-expressed in Sf9 cells with Smc1 and Smc3 (lanes 1, 4–6), Smc1, Smc3 and Scc1 (lanes 2, 7–9), or Smc1, Smc3, Scc1 and SA1 (lanes 3, 10–13). Cell lysates were prepared (lanes 1–3; input) and subjected to immunoprecipitations with the antibodies indicated (lanes 4–13). The precipitates were divided into two aliquots, resolved by SDS-PAGE, and analyzed by silver stain (Silver) and immunoblotting using the antibodies indicated (Blot). (B) Wapl was co-expressed in Sf9 cells with Scc1 alone (lanes 1, 4 and 5), SA1 alone (lanes 2, 6 and 7) or Scc1 and SA1 together (lanes 3, 8–10). Cell lysates were prepared (lanes 1–3; input) and reciprocal immunoprecipitations were performed using the antibodies indicated (lanes 4–10). The precipitates were analyzed as described in (A). (C) A schematic diagram of the order of subunit interactions. Wapl can associate only with the holocomplex of cohesin or the Scc1-SA dimer. A hypothetical model for Wapl’s action is shown at the right end of the upper panel.