Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 23 (2), 371-83

Changes in Ect2 Localization Couple Actomyosin-Dependent Cell Shape Changes to Mitotic Progression

Affiliations

Changes in Ect2 Localization Couple Actomyosin-Dependent Cell Shape Changes to Mitotic Progression

Helen K Matthews et al. Dev Cell.

Abstract

As they enter mitosis, animal cells undergo profound actin-dependent changes in shape to become round. Here we identify the Cdk1 substrate, Ect2, as a central regulator of mitotic rounding, thus uncovering a link between the cell-cycle machinery that drives mitotic entry and its accompanying actin remodeling. Ect2 is a RhoGEF that plays a well-established role in formation of the actomyosin contractile ring at mitotic exit, through the local activation of RhoA. We find that Ect2 first becomes active in prophase, when it is exported from the nucleus into the cytoplasm, activating RhoA to induce the formation of a mechanically stiff and rounded metaphase cortex. Then, at anaphase, binding to RacGAP1 at the spindle midzone repositions Ect2 to induce local actomyosin ring formation. Ect2 localization therefore defines the stage-specific changes in actin cortex organization critical for accurate cell division.

Figures

Figure 1
Figure 1
Ect2 Alters the Dynamics of Mitotic Cell Rounding (A and B) Time-lapse phase contrast images of HeLa cells rounding up before mitosis with cell length (Feret's diameter) indicated by red line showing cells treated with (A) control siRNA and (B) Ect2 siRNA. Images taken every 2 min. Scale bar, 20 μm. (C) Mean length of 22 cells during progression through mitosis, aligned so that time point 0 represents nuclear envelope breakdown (NEB). Error bars denote SD. Colored vertical lines show mean timing of mitotic events with shaded areas showing SD. Mitotic events were visualized using the expression of histone H2B-mRFP (for chromatin condensation and anaphase) and tubulin-GFP (centrosome separation and microtubule nucleation at spindle). NEB was recorded as the time point at which free tubulin-GFP dimers are able to enter the nucleus. (D) Comparison of rounding in cells treated with control siRNA (n = 20 cells) and Ect2 siRNA (n = 23). Error bars denote SD. (E) Box plot showing time taken to round up at mitosis for control siRNA (n = 31) compared to three nonoverlapping siRNAs targeting Ect2 (n = 27, 33, and 25) in control HeLa cells, and in HeLa cells expressing mouse Ect2-GFP at endogenous levels (n = 28 cells in each condition). Central line shows median, boxes are quartiles and whiskers show complete range. Cells were imaged 24 hr post-RNAi and only the first division after RNAi treatment was analyzed. The percentage of cells that then go on to fail cytokinesis in each condition is indicated. (F) Western blot showing knockdown of human Ect2 but not mouse Ect2-GFP (upper band) by three siRNAs targeting Ect2. See also Figure S1.
Figure 2
Figure 2
Ect2 Is Required for the Organization of a Rigid, Cortical Actin Cytoskeleton in Mitosis (A and B) Time-lapse confocal images of HeLa cells entering mitosis labeled with LifeAct-GFP and histone H2B-mRFP treated with control siRNA (A) and Ect2 siRNA (B). Time is in minutes. Three different z planes, 4 μm apart are shown. See also Movie S1. (C) XZ projections of metaphase cells labeled with LifeAct-GFP and histone H2B-mRFP. Confocal Z sections were taken every 200 nm through living cells covering the full height of the cell. (D) Graph showing the mean height of cells in interphase and metaphase treated with control siRNA or Ect2 siRNA (n = 10–15 cells per condition). Error bars show SD. (E) Confocal micrographs of fixed metaphase HeLa cells stained to show the actin cytoskeleton in control siRNA and Ect2 siRNA cells. Actin is labeled with phalloidin-TRITC in red, tubulin in green and 4′,6-diamidino-2-phenylindole (DAPI) in blue, scale bars 10 μm. (F) Quantification of the ratio of cortical/cytoplasmic actin in control and Ect2 RNAi cells in interphase and mitosis. Mean signal intensity in a 3 × 3 pixel box was measured in the actin channel at two locations: 0.5 μm from the cell edge (cortex) and 5 μm from the cell edge (cytoplasm). Four sites per cell were measured and the graph shows the mean values for 15 cells per condition with error bars denoting SD. (G) Diagram of the optical stretcher set-up used to measure cell compliance. (H) Graph showing mean compliance J(t) (see Experimental Procedures for detail) over time as cells are subjected to optical stretching for 4 s comparing control siRNA cells in interphase (n = 60 cells) and mitosis (n = 63) and Ect2 siRNA mitotic cells (n = 45). Error bars denote SEM. See also Figure S2.
Figure 3
Figure 3
Ect2 Controls Mitotic Rounding via RhoA, Rho Kinase, and Myosin II (A) Box plot comparing rounding times of control cells (n = 22 cells) compared to cells treated with 100 ng/ml nocodozole (Nz) to depolymerize microtubules (n = 16), RacGAP1 siRNA (n = 20), 2 μg/ml C3 transferase to inhibit Rho (n = 22), 50 μM ROK inhibitor Y-27632 (n = 18) and 50 μM blebbistatin to inhibit myosin II (n = 25). Central line shows median, boxes are quartiles, and whiskers show complete range. (B) Graph showing the cell length through time for conditions in Figure 3A. Error bars denote SD. (C and D) Pseudo-colored FRET ratio images showing RhoA activity in cells arrested in prometaphase by treatment with 5 μM STLC, comparing a control siRNA cell (C) to an Ect2 siRNA treated cell (D). (E) Graph showing mean total RhoA FRET efficiency in control siRNA cells (n = 20) and Ect2 siRNA (n = 16) cells. FRET efficiency was calculated using acceptor photo-bleaching (see Experimental Procedures). Error bars denote SD. (F and G) Representative confocal images of control (F) and Ect2 siRNA (G) prometaphase cells during mitotic rounding stained for phospho-myosin light chain. Insets show tubulin staining and DNA (DAPI stain, blue). (H) Quantification of the Ect2 siRNA p-myosin II phenotype. The ratio of cortical/cytoplasmic phospho-myosin was calculated by measuring mean signal intensity in a 3 × 3 pixel box at four locations at the cortex of the cell, and four locations 5 μm into the cytoplasm. The graph shows the mean values for 11 cells per condition with error bars denoting SD. Scale bars, 5 μm. See also Figure S3.
Figure 4
Figure 4
Ect2 Is Phosphorylated throughout Mitosis (A) Gel showing the band shift of Ect2 at mitosis comparing unsynchronized cells (first lane) to cells synchronized at prometaphase by an 18 hr treatment with 5 μM S-trityl-L-cysteine (STLC, second lane). This shift is abolished by addition of 50 μM Roscovitine for 2 hr (third lane), which also reverses cell rounding. (B) Band shift of Ect2 protein over a time course from 9 to 15 hr after release from double thymidine block. Image shown is representative of all experiments (n = 3). (C) Quantification of the gel in (B), showing the percentage of Ect2 protein that is phosphorylated and the percentage of cells in mitosis at each time point. The fraction of phosphorylated Ect2 was calculated by normalizing to background and then dividing the band volume for the phospho-species by the total Ect2 band volume. Mitotic stages were determined by visual inspection of the spindle and DNA following fixation and immunostaining with tubulin and DAPI of a sample of cells at each time point (n = 79–221 cells for each time point) “% cells in mitosis” includes cells in prophase, prometaphase, and metaphase. (D) Phospho-band shift of Ect2 in a synchronized population of cells as they exit mitosis after release from a metaphase arrest. This experiment was repeated twice and the image is representative of both experiments. (E) Quantification of the gel in D, showing the percentage of Ect2 phosphorylated protein compared to the percentage of cells in mitosis (prophase, prometaphase, and metaphase) and at anaphase/cytokinesis.
Figure 5
Figure 5
Ect2 Is Exported from the Nucleus in Early Mitosis (A) Confocal micrograph showing Ect2 localization at each stage of mitosis in fixed cells stained with an antibody against Ect2 (upper panel) and tubulin and DAPI to show mitotic stage (lower panel). Scale bar applies to all images, 10 μm. (B and C) Time-lapse confocal images of a HeLa cell entering mitosis expressing mouse Ect2-GFP (upper panels) and tubulin-RFP (lower panels). Mouse Ect2 is constitutively expressed in a BAC under its endogenous promotor (Hutchins et al., 2010). Two different z planes are shown: the bottom of the cell to show the full extent of the cytoplasm (B), and 8 μm higher (C) at the level of the nucleus. Time is indicated in minutes, with time point 0 being the frame of nuclear envelope breakdown as judged by when tubulin dimers first enter the nucleus. Note increase in Ect2 levels in the cytoplasm before nuclear envelope breakdown in frames −1 and −2. Scale bars, 10 μm. (D) Quantification of time-lapse images in (B) and (C). Six cells were analyzed and measurements aligned, so that time point 0 represents the frame of nuclear envelope breakdown. Mean signal intensity was measured for Ect2 (red line) and tubulin (blue line) in a 6 × 6 pixel box in the nucleus and cytoplasm and the nuclear/cytoplasmic ratio was plotted. The black line shows mean cell length to give an indication of the onset of mitotic rounding. Error bars denote SD. See also Movie S2. See also Figure S4.
Figure 6
Figure 6
Cytoplasmic Ect2 Is Sufficient to Induce Cell Rounding (A) Three different Ect2 constructs were overexpressed in HeLa cells: Ect2-FL-GFP, Ect2-C-GFP, and Ect2-dNLS-GFP. (B) Representative confocal micrographs of cells transfected with Ect2-FL, Ect2-C and Ect2-dNLS showing the actin cytoskeleton stained with phalloidin-TRITC (top panel) and Ect2 construct localization (bottom panel). Note the rounded cell morphology in Ect2-C and Ect2-dNLS cells. (C) Quantification of the percentage of interphase cells displaying the rounded phenotype (n = 80–149 cells). (D and E) Representative confocal micrographs of cells in prophase showing an a nontransfected cell (D) and a cell transfected with Ect2-dNLS (E). The actin cytoskeleton was visualized by phalloidin staining and an anti-GFP antibody was used to indicate transfected cells. Inset shows the cell nucleus, stained with DAPI, to identify mitotic stage. Note rounded cell morphology in E. (F) Phase contrast images of a nontransfected (NT) cell and a cell expressing Ect2-dNLS-GFP at low levels rounding up in early mitosis. Transfected cells are indicated by GFP fluorescence in final panel. See also Movie S3. (G) Box plot comparing the mitotic rounding time of nontransfected cells (n = 21 cells) with those transfected with Ect2-dNLS-GFP (n = 23). To ensure rounding is mitotic rather than apoptotic, only cells that later proceeded to cytokinesis were analyzed. For Ect2-dNLS cells, only cells expressing low levels of the construct that were not already rounded in interphase were analyzed. For box plot, central line shows median, boxes are quartiles, and whiskers show range. For Ect2 dNLS, the median and lower quartile are the same value. Scale bars, 20 μm. See also Figure S5.
Figure 7
Figure 7
A Model for Ect2 Function through Mitosis (A–D) Dynamic changes in Ect2 localization (shown in blue) control actin remodeling throughout mitosis. (A) Ect2 leaves the nucleus in early prophase. (B) Active Ect2 in the cytoplasm is able to activate RhoA and drive mitotic rounding. (C) Ect2 activation of RhoA results in the formation of a rigid actomyosin cortex that assists metaphase spindle assembly (D). At anaphase, Ect2 is relocalized to the central spindle and removed from the poles, resulting in the redistribution of active RhoA and therefore the contractile actomyosin machinery to drive furrowing in the center of the cell. (E) Export of active, phosphorylated Ect2 into the cytoplasm at mitotic onset stimulates a decrease in cell length. At anaphase, Ect2 remains active but its location is modulated by binding to RacGAP1 at the spindle midzone, resulting in elongation of the cell, furrowing, and cytokinesis.

Similar articles

See all similar articles

Cited by 63 articles

See all "Cited by" articles

References

    1. Amano M., Ito M., Kimura K., Fukata Y., Chihara K., Nakano T., Matsuura Y., Kaibuchi K. Phosphorylation and activation of myosin by Rho-associated kinase (Rho-kinase) J. Biol. Chem. 1996;271:20246–20249. - PubMed
    1. Ananthakrishnan R., Guck J., Wottawah F., Schinkinger S., Lincoln B., Romeyke M., Moon T., Käs J. Quantifying the contribution of actin networks to the elastic strength of fibroblasts. J. Theor. Biol. 2006;242:502–516. - PubMed
    1. Barr F.A., Elliott P.R., Gruneberg U. Protein phosphatases and the regulation of mitosis. J. Cell Sci. 2011;124:2323–2334. - PubMed
    1. Beausoleil S.A., Villén J., Gerber S.A., Rush J., Gygi S.P. A probability-based approach for high-throughput protein phosphorylation analysis and site localization. Nat. Biotechnol. 2006;24:1285–1292. - PubMed
    1. Blangy A., Lane H.A., d'Hérin P., Harper M., Kress M., Nigg E.A. Phosphorylation by p34cdc2 regulates spindle association of human Eg5, a kinesin-related motor essential for bipolar spindle formation in vivo. Cell. 1995;83:1159–1169. - PubMed

Publication types

LinkOut - more resources

Feedback