Fig 1.
Premature loss of centriole association in STLC-arrested cells.
(A) After double thymidine block and release, HeLa cells were treated with DMSO or STLC (10 μM) and immunostained with the centrin-2 (green) and CEP135 (red) antibodies. DNA was stained with DAPI (blue). Scale bar, 10 μm. As illustrated, centriole configuration was determined by the ratio of centrin-2 and CEP135 foci. (B) HeLa cells were synchronously released from G1/S phase, treated with STLC or nocodazole, and cultured for the indicated time periods. Centriole association was determined with the centrin-2 and CEP135 antibodies. The experiments were repeated 3 times with over 150 cells at each time point. (C) Asynchronous HeLa, U2OS and RPE-1 cells were treated with STLC (5 μM) for the indicated time periods and immunostained with the centrin-2 and CEP135 antibodies. The mitotic cells with separated centrioles were counted. The experiments were repeated twice with over 100 cells at each time point. Values are means and standard deviations.
Fig 2.
Premature centriole separation in separase-depleted cells.
(A) Immunoblot analysis was performed to determine cellular levels of the full-length and C-terminal fragment of separase in the siSeparase-transfected HeLa cells. (B) The siCTL- or siSeparase-transfected HeLa cells were enriched to S phase with a single thymidine block and release, and treated with RO3306 or STLC for 20 h. The cells were fixed and coimmunostained with the centrin-2 (green) and CEP135 (red) antibodies. Based on the ratio of centrin-2 and CEP135 foci, the number of cells with separated centrioles was counted. Scale bar, 10 μm. The experiments were repeated 3 times with over 100 cells at each group. Values are means and standard deviations. (C) HeLa cells were synchronously released from G1/S phase in the presence of RO3306 (10 μM) or STLC (10 μM). The cell lysates at the indicated time points were subjected to immunoblot analyses with the indicated antibodies.
Fig 3.
Premature centriole separation in CDC20-depleted cells.
(A) Immunoblot analysis was performed to determine the CDC20 levels in siCDC20-transfected RPE-1 cells. (B) The CDC20-depleted mitotic cells were immunostained with antibodies specific to centrin-2 (green) and CEP135 (red). Scale bar, 10 μm. (C) The CDC20-depleted RPE-1 cells were synchronized with a thymidine block and release, and fixed for immunostaining at the indicated time points. The control cells were treated with STLC for M phase arrest. The number of mitotic cells with separated centrioles was counted at each time points. Experiments were repeated three times with over 150 cells per each group. Values are means and standard deviations. (D) HeLa cells were treated with the STLC (10 μM) for 20 h and their chromosomes were spread and stained with DAPI.
Fig 4.
Effects of BI2536 on the premature centriole separation.
(A) The RPE-1 cells were treated with STLC (10 μM) as soon as they were released from a single thymidine block. Eight hours later, the cells were treated with BI2536 (200 nM) for the indicated time periods, and immunostained with the centrin-2 (green) and CEP135 antibodies (red). Scale bar, 10 μm. (B) The number of cells with separated centrioles were counted. The experiments were repeated three times, with over 150 cells at each group. Values are means and standard deviations.
Fig 5.
Effects of nocodazole on the premature centriole separation.
(A) HeLa cells were synchronously released from G1/S phase in the presence of nocodazole or STLC. Nine hours later, the second drug was added and cultured for the indicated time periods. The cells were immunostained with the centrin-2 and CEP135 antibodies and the number of cells with separated centrioles was counted. The experiments were repeated twice with over 100 cells at each group. Values are means and standard deviations. (B) The separase-depleted HeLa cells were treated with RO3306 or STLC for 20 h. Nocodazole (0, 50 or 200 ng/ml) was added along with RO3306 or at the 9th hour after the STLC treatment. The cells with separated centrioles were counted and statistically analyzed as above.
Fig 6.
Induction of the premature centriole separation after the removal of nocodazole.
(A) The CDC20-depleted RPE-1 cells were synchronously released from a single thymidine block. Nocodazole (200 ng/ml) was treated for 12 h and then washed out. At indicated time points, the cells were fixed for immunostaining with the centrin-2 and CEP135 antibodies to determine centriole separation. The experiments were repeated 3 times with over 100 cells at each group. Values are means and standard deviations. (B) Representative immunostaining results of the CDC20-depleted RPE-1 cells with or without nocodazole wash-out. The cells were immunostained with the centrin-2 (green) and pericentrin (red) antibodies. Scale bar, 10 μm.
Fig 7.
Effects of nocodazole and BI2536 on PCM dispersal in STLC-treated cells.
(A) HeLa cells were cultured in the presence of STLC for 20 h. Nocodazole was added at the 9th hour point. The cells were immunostained with the α-tubulin (green) and γ-tubulin (red) antibodies. The cells were also immunostained with centrin-2 and CEP135 antibodies to determine centriole association. Scale bar, 10 μm. (B, C) HeLa cells were treated with STLC for 20 h. Nocodazole (200 ng/ml) or BI2536 (200 nM) was added at 9th h. The cells were immunostained with antibodies specific to pericentrin, CEP192, CEP215 and γ-tubulin (green) along with centrin-2 (red). Scale bar, 10 μm. (C) The γ-tubulin staining patterns were categorized as discrete and dispersed. The experiments were repeated twice, with over 150 cells at each group. Values are means and standard deviations. (D) HeLa cells were arrested at S phase with thymidine and released in the presence of STLC. At the indicated time points, the cells were coimmunostained with antibodies specific to CEP135 and γ-tubulin to determine centriole association and PCM dispersal, respectively. The experiments were repeated twice, with over 100 cells at each group. Values are means and standard deviations. (E) HeLa cells were treated with STLC for 9 h and then with nocodazole for 11 h. The cells were then transferred to a nocodazole-free medium and cultured for up to 4 h. At the indicated time points, the cells were coimmunostainined with antibodies specific to centrin-2 and pericentrin. DNA was stained with DAPI.
Fig 8.
Depletion of pericentrin enhanced the premature separation of centrioles at M phase-arrested cells.
(A) Pericentrin-depleted HeLa cells were synchronized with a single thymidine block and release, and treated with STLC. Nine hours later, nocodazole or BI2536 was added and cultured for 11 h. The cells were coimmunostained (A) with centrin-2 (green) and CEP135 (red) antibodies to determine centriole association, (B) and with pericentrin (green) and CEP135 (red) antibodies to determine PCM. Scale bar, 10 μm. (C) The experiments were repeated three times with over 100 cells per group and statistically analyzed. Values are means and standard deviations.
Fig 9.
Centriole configurations in STLC-, nocodazole- and BI2536-treated cells by 3D structured illumination microscopy.
(A) G2, metaphase and G1 phase cells were obtained from synchronized populations of HeLa cells. The cells were fixed and coimmunostained with antibodies specific to centrin-2 (green), CEP135 (pink) and CP110 (red) to determine centriole configurations. (B) HeLa cells were synchronously released from a single thymidine block in the presence of STLC. Nine hours after the release, nocodazole or BI2536 was added. Eleven hours later, the M phase arrested cells were coimmunostained with the antibodies specific to centrin-2, CEP135 and CP110. Scale bars, 500 nm and 10 μm.
Fig 10.
A model of centriole separation in M phase-arrested cells.
Associated centrioles in M phase are separated as soon as the cell exit mitosis. Separase plays an essential role in centriole separation. The STLC treatment arrests the cells at M phase, and induces the premature centriole separation along with PCM dispersal. The nocodazole treatment also arrests the cells at M phase, but the centrioles remain associated and PCM looks intact without microtubule emanation. The BI2536 treatment keeps the integrity of PCM around the centriole pair, suggesting that PLK1 is also involved in the PCM dispersal in M phase-arrested cells.