Our previous work supports the hypothesis that in prophase the balance between the inward forces generated by kinesin-14 sliding interpolar MTs together (a) and nuclear elasticity (b) and the outward forces generated by dynein pulling on astral MTs (c) governs the spindle length. Here, we test the idea that an additional outward force is exerted by myosin-II (d) indirectly – by contracting the actin cortex and pulling outward on astral MTs cross-linked with the deforming actin network.

A) Injection of the myosin inhibitor Y27632 causes myosin to lose its cortical localization. In Sqh-GFP expressing embryos the spaghetti-squash signal largely disappears from the cortex and becomes diffuse after Y27632 is injected. Bars, 10 µm. B) Graph of pole-pole distance as a function of time (NEB is t = 0 s) for GFP-Tubulin expressing embryos (WT) and myosin inhibited embryos (WT+Y27632). After myosin inhibition pole separation during prophase is greater than in WT.

A) Side views of actin, nucleus and centrosomes (rhodamine-actin, red; GFP-Tubulin, green) and corresponding top views of actin in prophase WT embryos in the presence of the myosin inhibitor Y27632 reveal that the centrosomes segregate around the nucleus, likely being pulled outward by the increasing forces. The resulting vertical component of the force pushes the nucleus upward against the actin cap thinning the cortex and creating the ‘hole’. NEB is t = 0 s. Bars, 10 µm. B) The square root of the actin cap areas as functions of time (NEB is t = 0 s) in WT embryos in the absence and presence of the myosin inhibitor Y27632 and in the Sced embryos.

Hypothesis for each condition (B) is based on the actin distribution as shown in the corresponding line scans (A). The intensity value of 100 has been chosen as the arbitrary cut-off value to discriminate between the ‘soft’ (below 100) and ‘firm’ (above 100) actin patches. In the WT embryo, most of the actin cap is firm, and dynein pulls from the large firm actin cortex (inset b) while kinesin-14 exerts an inward force (inset a). When myosin is inhibited, the area of the firm actin is shifted outward focusing the dynein pulling in the outward direction. In the Sced embryo, many soft actin patches appear, but myosin contracts actin filaments there (inset c) complementing the pulling action of dynein (inset b). When myosin is inhibited in the Sced embryo, dynein pulls from the small infrequent firm patches of actin and does not separate the poles effectively.

A) Pole-pole distances in prophase as functions of time (NEB is t = 0 s) show a different degree of pole separation in Sced embryos in the presence and absence of the myosin inhibitor Y27632. The pole-pole separation in Sced embryos is comparable to that in control, but when myosin is inhibited, centrosomes separate less than in control and Sced embryos. B) Images from a time-lapse movie of a Sced embryo in prophase co-injected with FITC-tubulin and rhodamine-actin in the presence or absence of Y27632. Sced embryos have smaller and less organized actin caps which become even less compact when myosin is inhibited. Images are projections of 8–10 confocal planes taken 0.5 microns apart. Red, actin; green, tubulin. Bars, 10 µm.

A) Images from a time-lapse movie of a WT embryo expressing GFP-tubulin injected with rhodamine-actin and the myosin inhibitor Y27632. When myosin is inhibited, actin distribution in prophase changes compared to control showing a less compact and wider distributed actin structure above the nucleus (note the void area). Images are projections of 8–10 confocal planes taken 0.5 µm apart. B) Kymographs (from −250 to −100 seconds before NEB) of prophase embryos injected with rhodamine actin with or without Y27632 confirm that in the absence of myosin the outer edge of the actin cap expands at a faster rate and to a greater width than in control. Same confocal planes taken at the very top of the actin cap were imaged for both actin and MTs. Red, actin; green, tubulin. C) Images from a time-lapse movie of GFP-EB1 expressing embryos injected with rhodamine-actin with and without Y27632 during prophase. Tracking of MT tips shows that both in WT and myosin-inhibited embryos, MT tips uniformly reach the surfaces of both narrower and wider actin caps (EB1-GFP, green; rhodamine-actin, red). Bars, 10 µm.

A) Side views of actin at the cortex in WT and Sced embryos in the absence and presence of the myosin inhibitor Y27632 reveal different actin distributions and cap sizes. The WT actin cap is compact with an evenly distributed thick layer of actin. When myosin is inhibited, the cap expands remaining thick at the margins and becoming thinner in the center. Note that the nucleus ascends ‘indenting’ the cap. In Sced embryos the cap becomes smaller compared to WT embryos, and when myosin function is inhibited, the actin in the cap becomes disorganized. Corresponding line scans of fluorescence intensity (tubulin, green; actin, red) taken along the actin cap in relation to the centrosomes' position (taken along the pole-pole axis), confirm the qualitative description of the changes in actin distribution. Bars, 5 µm. B) Side view of the centrosomes' position with respect to the nucleus. The nuclear membrane was stained in vivo with fluorescent conjugated WGA in WT embryos expressing GFP-tubulin or in Sced embryos injected with FITC-tubulin. When myosin is inhibited in WT, the centrosomes move further down and around the nucleus than in control. In Sced, the centrosomes are positioned slightly more apically, and when myosin in inhibited, this apical shift and lesser separation becomes more evident. Bars, 10 µm. Next to the side views are corresponding images of a wild-type embryo expressing GFP-tubulin (green) injected with rhodamine-actin (red) and a Sced embryo injected with FITC-tub (green) and rhodamine-actin (red) in the presence and absence of the myosin inhibitor Y27632. The prophase actin cap taken 1.5 µm below the top shows the difference in actin density and distribution between wild-type and Sced embryos in the presence or absence of myosin. Circles mark the position of the centrosomes which in some cases are not clearly visible. Bars, 10 µm.