Spatial positive feedback at the onset of mitosis

Abstract

Mitosis is triggered by the activation of Cdk1-cyclin B1 and its translocation from the cytoplasm to the nucleus. Positive feedback loops regulate the activation of Cdk1-cyclin B1 and help make the process irreversible and all-or-none in character. Here we examine whether an analogous process, spatial positive feedback, regulates Cdk1-cyclin B1 redistribution. We used chemical biology approaches and live-cell microscopy to show that nuclear Cdk1-cyclin B1 promotes the translocation of Cdk1-cyclin B1 to the nucleus. Mechanistic studies suggest that cyclin B1 phosphorylation promotes nuclear translocation and, conversely, nuclear translocation promotes cyclin B1 phosphorylation, accounting for the feedback. Interfering with the abruptness of Cdk1-cyclin B1 translocation affects the timing and synchronicity of subsequent mitotic events, underscoring the functional importance of this feedback. We propose that spatial positive feedback ensures a rapid, complete, robust, and irreversible transition from interphase to mitosis and suggest that bistable spatiotemporal switches may be widespread in biological regulation.

Figure 1. Abrupt Cyclin B1 Nuclear Translocation Prior to Nuclear Envelope Breakdown

(A) A representative HeLa cell expressing cyclin B1-YFP (top) and the mitotic biosensor (MBS-RFP) (bottom) after release from G1/S block. Centrosome separation (top), cyclin B1 translocation (top), and NEB (bottom) occur in quick succession. Scale bars: 10 μm. (B) Quantitation of nuclear accumulation of cyclin B1-YFP after release from thymidine block of the cell shown in (A) (green) and three other representative cells (blue). For each cell, the last data point shown represents the time of NEB. (C) Translocation kinetics are well-approximated by the logistic equation. Cyclin B1-YFP translocation was measured for more than 100 cells. Individual time courses were b fitted to the logistic equation $y=a+\frac{b}{1+e^{-\left(\frac{t-t_0}{\tau }\right)}}$ and were scaled to their fitted maximum and minimum values (b and a respectively) and half-maximal times (t₀). Ten scaled time courses are shown (light blue). Rise times (τ), were calculated from the curve fits for all >100 cells and are expressed as means ± S.D. The fitted logistic equation curve with τ = 3.2 min is shown in dark blue. Note that with the logistic equation, y goes from 27% to 73% of its maximal value over a time interval of 2τ. See also Figure S1 and Movies S1 and S2.

Figure 2. In Cdk1AF-Expressing Cells, NLS-Cyclin B1-CFP Expression Promotes Cyclin B1-YFP Translocation

(A) Cumulative percentage of cells that showed cyclin B1 nuclear translocation (closed circles) and NEB (open circles) after release from G1/S block. Cells were expressing Cdk1AF (AF) or Cdk1-WT (WT), plus cyclin B1-YFP (B1) and either NLS-cyclin B1-CFP (NLS-B1) or histone H2B-CFP (H2B). At least 100 cells were analyzed for each experimental condition. One of three independent experiments yielding similar results. (B) Median times of nuclear import for the experimental conditions described in (A). In each condition at least 100 cells were analyzed. Error bars represent the median absolute deviations calculated for the first 50% of the cells to accumulate cyclin B1 in the nucleus. p values were calculated by Mann-Whitney and Kolmogorov-Smirnov tests. (C) Cumulative percentage of cells that showed cyclin B1 nuclear translocation for cells transfected with Cdk1AF (AF) and cyclin B1-YFP (B1-YFP) plus either NLS-cyclin B1-CFP (NLS-B1), cyclin B1-CFP (B1-CFP), or histone H2B-CFP (H2B). (D) Median times of nuclear import for the experimental conditions described in (C). Error bars and p values were calculated as described in (B). See also Figure S2.

Figure 3. Induced Translocation of Active Cdk1-Cyclin B1 Complexes to the Nucleus Triggers Mitosis and Induces Spatial Positive Feedback

(A) Schematic of the experimental approach by which the rapamycin-analog iRap causes cyclin B1-FRB to accumulate in the nucleus (see also Figure S3). (B) Two representative cells expressing mCherry-NLS3-FKBP2 (top), cyclin B1-FRB-YFP (middle), cyclin B1-CFP (bottom) and treated with 5 μM iRap in the presence of 1 μg/ml doxycycline to induce Cdk1AF expression. Scale bars: 10 μm. See also Movies S3, S4. (C) Quantitation of the accumulation of cyclin B1-FRB-YFP (green) and cyclin B1-CFP (blue) for the cells shown in (B). For the binucleate cells both the import into the smaller nucleus (open points) and the larger nucleus (solid points) are shown. (D) Cumulative percentage of cells that underwent NEB. Cells were treated with 5 μM iRap and/or 1 μg/ml doxycycline as indicated. More then 100 cells were analyzed for each experimental condition. (E) Percentage of Cdt1-cerulean positive cells (G1/early S ) and Geminin-cerulean (Late S/G2) positive cells that underwent mitosis within 300 minutes of iRap treatment. (F) Representative images of one cell ectopically expressing cyclin B1-FRB-YFP (top), cyclin B1-CFP (bottom) and Lyn-FKBP2-mCherry (middle) treated with 5 μM iRap. Scale bars: 10 μm. See also Figure S3 and Movie S5.

Figure 4. Timely Completion of Mitosis Requires Abrupt Activation and Redistribution of Cdk1-Cyclin B1

(A, B) Cdk1AF expression makes cyclin B1 translocation more graded and makes the duration of mitosis longer and more variable. As a proxy for the duration mitosis we measured the time of completion of cyclin degradation minus the time of cyclin B1 translocation. Rise times were calculated by fitting traces of ~100 cells to the logistic equation as described in Figure 1. Ten individual traces are shown here for each condition. The individual shown for the Cdk1-WT-transfected cells are the same as those shown in Figure 1C. (C, D) Wee1 knockdown makes cyclin B1 translocation more graded and makes the duration of mitosis longer and more variable. Ten individual traces are shown here for each condition. See also Figure S4. (E, F) Cdc25-CD expression makes cyclin B1 translocation more graded and makes the duration of mitosis longer and more variable. Ten individual traces are shown here for each condition. See also Figure S4. (G, H) Leptomycin B treatment makes cyclin B1 translocation more graded and makes the duration of mitosis longer and more variable. Ten individual traces are shown here for each condition. Timing of events was monitored for approximately 100 cells for each condition. See also Figure S4 and Movie S6. (I, J) iRap-mediated induction of cyclin B1 translocation in the absence of Cdk1AF makes cyclin B1 translocation more graded and makes the duration of mitosis longer and more variable. Ten individual traces are shown here for each condition. In each panel the timing of events was monitored for approximately 100 cells for each condition.

Figure 5. Cyclin B1 Phosphorylation is Required for the Pre-NEB Translocation of Cyclin B1 to the Nucleus and for Spatial Positive Feedback

(A, B) Cyclin B1-WT-YFP generally translocates to the nucleus ~5 min prior to NEB. In contrast, cyclin B1-5SA translocates to the nucleus concomitantly with NEB. Panel A shows representative cells. The arrows show when cyclin B1 translocation was maximal. Scale bar represents 10 μm. Panel B shows binned translocation timings for more than 100 cells. (C–E) Cyclin phosphorylation is required for spatial positive feedback. Cyclin B1-FRB-YFP was induced to translocate to the nucleus by iRap treatment. Panel C shows one representative cell, where cyclin B1-5SA-CFP translocation lags substantially behind that of cyclin B1-FRB-YFP. The arrows show when cyclin B1 translocation was maximal. Scale bar represents 10 μm. Panel D quantifies the translocation shown in panel C. Panel E shows four other cells. See also Figure S5 and Movie S7.

Figure 6. Cyclin B1 Phosphorylation Is Required for the Stable Association of Cyclin B1 with Mitotic Chromosomes

(A) Cell fractionation. Cells were released from a double thymidine block and lysed at various times. Mitotic cells were obtained by releasing double thymidine blocked cells into nocodazole for 12 hours. Lysates were separated into a low speed pellet, N, and a low speed supernatant, C, and were blotted for total cyclin B1, phosphorylated cyclin B1, and histone H3 (a nuclear marker). (B) Fluorescence confocal microscopy. Cells were synchronized by double thymidine block and release. Non-transfected cells were stained with cyclin B1 or pS126-cyclin B1 antibodies (green) plus DAPI (red). Cyclin B1-5SA-YFP (green) transfected cells were stained with DAPI (red). Scale bar represents 10 μm. See also Figure S6.

Figure 7. Translocation Promotes Cyclin B1 Phosphorylation, Completing the Positive Feedback Loop

(A) Control (DMSO-treated) HeLa cells (top) and leptomycin B-treated HeLa cells (bottom) 8 h post release from a double thymidine block. Cells were stained for total cyclin B1 (red), pS126-cyclin B1 (green), and DNA (blue). (B) Quantitation of the cyclin B1 and pS126-cyclin B1 staining in interphase cells treated with DMSO (blue) or leptomycin B (red). (C) Control (DMSO-treated) HeLa cells (top) iRap-treated HeLa cells (bottom) after 5 hours. Cells were expressing stained for total cyclin B1 (red), pS126-cyclin B1 (green), and DNA (blue). (D) Quantitation of the cyclin B1 and pS126-cyclin B1 staining in interphase cells treated with DMSO (blue) or iRap (red). (E) Schematic model of Cdk1-cyclin B1 translocation and autophosphorylation. (F) A simplified, three-species model of Cdk1-cyclin B1 translocation and autophosphorylation. (G) Steady-state responses of the model. Parameters were: k_imp = 0.25; k_exp = 1; k_phos = 100; k_dephos = 0.2; and, for the Hill function, K = 1 and n = 4. The steady-state responses were solved numerically using Mathematica 7.0 (Wolfram). (H) Dynamical response of the model. The ODEs were solved numerically using Mathematica 7.0. The dashed line is $x_{\mathit{tot}}(t)=\frac{1.5}{1+e^{-t/\tau }}$, with τ = 7.2 min. See also Figure S7.