Action at a distance during cytokinesis

Abstract

Animal cells decide where to build the cytokinetic apparatus by sensing the position of the mitotic spindle. Reflecting a long-standing presumption that a furrow-inducing stimulus travels from spindle to cortex via microtubules, debate continues about which microtubules, and in what geometry, are essential for accurate cytokinesis. We used live imaging in urchin and frog embryos to evaluate the relationship between microtubule organization and cytokinetic furrow position. In normal cells, the cytokinetic apparatus forms in a region of lower cortical microtubule density. Remarkably, cells depleted of astral microtubules conduct accurate, complete cytokinesis. Conversely, in anucleate cells, asters alone can support furrow induction without a spindle, but only when sufficiently separated. Ablation of a single centrosome displaces furrows away from the remaining centrosome; ablation of both centrosomes causes broad, inefficient furrowing. We conclude that the asters confer accuracy and precision to a primary furrow-inducing signal that can reach the cell surface from the spindle without transport on microtubules.

Figure 1.

Microtubules in live urchin embryos. All panels show single confocal sections. (A and A′) 16-cell purple urchin embryo; A′ shows a 2× enlarged view of the lower cell (indicated by white mark in A). Microtubules approach the cortex everywhere before anaphase onset (1 min, 30 s); during anaphase, just before furrowing, many astral microtubules penetrate both polar and equatorial cortex (arrowheads in A′). (B) Vegetal view, 28-cell sand dollar embryo; i–vi are 2× enlarged views as indicated. Astral microtubules frequently cross spindle midplane before and during anaphase (i–iii), approach within 1 µm of the equatorial surface before furrowing (ii–iv), and curve inward as the furrow ingresses (v and vi). Arrowheads point to exemplars. (C) Eight-cell sand dollar embryo; single microtubules grow as far as the cell surface in all directions (equatorially in the 01:56 frame, tropically in the 01:08 frame, and toward the pole in the 03:32 frame). Astral microtubules reach the polar cortex most densely in anaphase (08:48) but also reach the equator before furrowing (frame 10:48). (C′) Enlargement of successive frames for microtubules indicated by arrowheads in C (intensities squared to enhance contrast). Video 1 corresponds to A–C. Time is indicated in minutes:seconds.

Figure 2.

The furrow forms in a microtubule-poor region. (A) Single superficial confocal sections of a 16-cell sand dollar embryo, slightly compressed (EMTB-3G). (02:00) All three cells in metaphase; few microtubule ends are visible. (07:00) Left cell begins cleavage, right cell is still in metaphase, and the top cell has likely entered anaphase. (13:00) Right cell begins to cleave; fewer microtubules approach the cell surface (bright dots) in the incipient furrow than outside it. (A′) “Kymocubes” made by 3D rendering sequence in A; white lines denote frames in A. The whole of each of three cells is shown in the kymocube. Bright dots (open arrowheads in A–C) indicate brief cortical visits by microtubule ends; these predominate in metaphase. Vertical streaks (closed arrowheads in A–C) indicate kinematically stable microtubule ends, which appear before furrowing, and are scarce in the equatorial zone. (B) Kymograph of 5-µm medial strip (indicated by box) ∼1 µm beneath the surface of a slightly flattened eight-cell sand dollar embryo. The arrow indicates furrow initiation time; vertical streaks above this point indicate stable microtubules at the cortex before furrowing. These are notably fewer in the equator. (C) Surface rendering of single medial sections of uncompressed sand dollar blastomeres, each covering anaphase through cytokinesis. Long streaks (stable microtubules at the cell surface) are evident well before furrowing. Broad smears appear during furrowing, notably in the area around the crotch; these reflect microtubules growing along the cortex (see also A′ and B). (D) Single superficial sections from a 6.5-h X. laevis embryo mosaically expressing GFP-rGBD (red) and 3C-EMTB (cyan). Surface microtubules disappear in metaphase, then reappear (16:52) ∼90 s before active Rho appears in the equator (18:28). Microtubules are largely absent from a >20-µm-wide band inhabited by the Rho zone; arrows indicate equatorial microtubules that disappear during furrowing. Time is indicated in minutes:seconds.

Figure 3.

Nocodazole-insensitive microtubules exhibit no orientation bias in urchin blastomeres. (A–C) Single sections, all expressing EMTB-3G; (A′–C′) 2× enlarged views of cells indicated by the white marks in A–C. (A) Eight-cell purple urchin embryo; 20 µM nocodazole was added at time 00:00. Within 2 min, most astral microtubules disassembled; remaining microtubules are randomly oriented and most end well short of the cortex. Even so, furrowing is accurate and timely. Arrowheads indicate surviving nonequatorial astral microtubules. (B) 16-cell purple urchin embryo; 10 µM nocodazole was added at time 00:00. The top left cell entered anaphase around time 0; none of the other cells in view left metaphase nor furrowed. This one cell cleaved despite nearly complete absence of astral microtubules, and the furrow crossed the spindle midzone. (C) 16-cell sand dollar embryo; 10 µM nocodazole was added at time 00:00, at which time three of four macromeres have entered anaphase; the east cell enters anaphase shortly thereafter (03:00). In all cells, numerous astral microtubules, pointing in all directions, persist >5 min after nocodazole addition (arrowheads). Stable microtubules may approach the cortex in west and south cells (03:00), but none can be seen to do so in north and east cells; all commence furrowing at a normal time and complete cytokinesis. Note that in all cases, stable microtubules become brighter because EMTB-3G liberated by disassembly becomes available for binding. Video 3 includes A–C. Time is indicated in minutes:seconds.

Figure 4.

Cleavage occurs despite diminution of the aster and cortical microtubules by TSA. All panels are single confocal sections. (A) 16-cell purple urchin embryo (EMTB-3G), treated with 20 µM TSA at time 00:00. A′ shows a 2× enlarged view of the indicated cell (white mark in A). Within minutes of TSA addition in metaphase, asters were reduced to nearly nothing (04:00). Despite reduction in spindle length, cells enter anaphase (07:20) and complete cytokinesis (11:20–16:10). No microtubules connecting the mitotic apparatus to the cortex are visible (see Video 4). (B) 16-cell purple urchin embryo (cyan, EMTB-3G; yellow, mC-H2B); 15 µM TSA was added at time 00:00. After a delay in metaphase, five of six cells in focus initiate cytokinesis. B′ shows a 2× enlarged view of the indicated cell (white mark in B). (C) One cell within a 16-cell sand dollar embryo (EMTB-3G), treated with 25 µM TSA ∼10 min before time 00:00. Although the aster regrows slightly, the cortex is ∼20 µm away from the nearest spindle microtubule ends; nevertheless, cytokinesis completes accurately (see Video 4). Arrowheads in A and C indicate metaphase plate before (04:00 and 04:48), and the gap after (07:20 and 10:48), anaphase onset. Time is indicated in minutes:seconds.

Figure 5.

Myosin occupies a wider-than-normal zone in asterless cells. Cyan, anti-tubulin; yellow, Hoechst; magenta, anti-phospho-myosin. (A and B) Untreated 16-cell sand dollar embryos. Projection of 24 0.5-µm sections, vegetal view (A); and 19 0.6-µm sections, side view (B). (C and D) 16-cell sand dollar embryos fixed 10 min after adding 25 µM TSA. (C) A projection of 18 0.5-µm sections through the middle portion of the macromeres. (D) A projection of 28 0.5-µm sections through the middle portion of the mesomeres. A–D are from the same batch fixed at the same time. Normally, phospho-myosin is virtually absent outside of ∼10 µm furrow zone. In TSA-treated cells, phosphomyosin is not as thoroughly excluded from the poles but is still enriched equatorially, with definite zones around ingressing furrows (brackets in D), even though no spindle microtubules approach the cortex. A′–D′ are 2× enlarged views of the cells indicated by the white marks in A–D.

Figure 6.

Active Rho occupies a wider-than-normal zone in asterless cells. (A) GFP-rGBD at similar stages of furrowing in control (left) and TSA-treated (right) purple urchin embryos; single confocal sections. Rho zones are broader and fainter in TSA-treated cells compared with untreated siblings, but zones are centered equatorially and match furrows. Plots show representative fits between intensity data (red) and a fit curve (blue) that measures zone width along the cell outline (see Materials and methods). (B) Measured zone widths normalized by pole-to-pole cell length, plotted against integrated intensity, minus baseline, within the curve fit as shown in A, expressed as a fraction of baseline. (C) Normal Rho zones, eight-cell purple urchin embryo (red, GFP-rGBD; cyan, 3C-EMTB). (D) Eight-cell purple urchin embryo (same probes as C) treated with 20 µM TSA at time 00:00. Uniform cortical Rho activity during metaphase (00:00) disappears as cells enter anaphase, as in normal cells (Bement et al., 2005). Rho zones (brackets) are barely detectable above background, yet furrows develop and complete with minor delay (compare times in C), which implies that cells normally express more equatorial Rho activity than they require. Time is indicated in minutes:seconds.

Figure 7.

Asters alone induce furrows only if they are far enough apart. (A–C) Single-plane sequences of anucleate, centrosome-containing cytoplasts (sand dollars). (A) Anucleate cytoplast with four centrosomes, two in focus. In normal cells this size, furrowing would begin between 03:00 and 07:00, but no furrow occurs even 20 min later. Centrosomes are closer than the distance to the equator. (B) Anucleate cytoplast with two centrosomes, farther apart than the distance to the equator. A furrow initiates at the expected time and place, and proceeds to completion. (C) An anucleate cytoplast with four centrosomes variously spaced. A deeply ingressing furrow bisects the well-spaced centrosomes; shallow furrows form over closely spaced centrosomes. Time is indicated in minutes:seconds. (D) Long-term differential interference contrast sequence of two halves of a bisected sand dollar zygote. The film began shortly after the unbisected zygote would have divided; because each half has only one centrosome, neither divided at first mitosis. All subsequent divisions proceeded normally in the nucleated half (right), creating a perfect blastula. The anucleate half cleaved abortively in its first attempt. After that, four centrosomes (dots in the 1:51:40 frame) were variously spaced. The second attempt made deep furrows between the widely separated asters, but none between the closely spaced pair. Subsequent divisions were variously successful. Time is indicated in hours:minutes:seconds. (E) Aster separation strongly correlates to the furrow extent. See Materials and methods for measurement. Video 6 corresponds to A–C.

Figure 8.

Centrosome ablation displaces furrows. (A–D) Single-plane recordings of sand dollar embryos expressing EMTB-3G alone (A) or GFP-rGBD (red) and 3C-EMTB (cyan; B–D); time is shown in minutes:seconds after last irradiation. Arrows, ablation sites; dotted lines, furrow plane. (A) Single-pole ablation in a moderately large cell. Few astral microtubules remain; furrowing occurs over the spindle end rather than the midzone (see Video 7). (B) Two single-pole ablations. Furrows form over and close upon the ablated end. In the top cell, chromosomes are partitioned by the furrow; in the bottom cell, they are not. Rho activity zones in ablated cells are similar to normal cells but shifted. (C) Double-pole ablation in a large cell. A broad furrow with barely detectable Rho activity forms above the spindle midplane and closes between spindle halves. (D) A normal cell (left), singly ablated cell (top), and doubly ablated cell (right). The furrow in the doubly ablated cell is broad, with dilute Rho activity, but closes accurately. The furrow in the singly ablated cell is shifted away from the midplane. (E) Distances (percentage of cell length) between spindle midplane and furrow plane in control, double-, and single-ablated cells (dots) superimposed upon normal curves computed from mean and standard deviation. Video 9 corresponds to B–D.