Myosin II dynamics are regulated by tension in intercalating cells

Abstract

Axis elongation in Drosophila occurs through polarized cell rearrangements driven by actomyosin contractility. Myosin II promotes neighbor exchange through the contraction of single cell boundaries, while the contraction of myosin II structures spanning multiple pairs of cells leads to rosette formation. Here we show that multicellular actomyosin cables form at a higher frequency than expected by chance, indicating that cable assembly is an active process. Multicellular cables are sites of increased mechanical tension as measured by laser ablation. Fluorescence recovery after photobleaching experiments show that myosin II is stabilized at the cortex in regions of increased tension. Myosin II is recruited in response to an ectopic force and relieving tension leads to a rapid loss of myosin, indicating that tension is necessary and sufficient for cortical myosin localization. These results demonstrate that myosin II dynamics are regulated by tension in a positive feedback loop that leads to multicellular actomyosin cable formation and efficient tissue elongation.

Figure 1. Multicellular alignment and formation of myosin II cable-like structures in intercalating cells

(A,B) Wild-type embryos before (A) or after (B) germband elongation (yellow line). (C) Contraction of isolated AP edges (top) leads to local neighbor exchange and contraction of linked AP edges (bottom) leads to multicellular rosette formation. (D) Confocal images from a time-lapse movie of germband cells expressing Myo:GFP before (D) and 5 min (D') or 10 min (D") after the onset of elongation. Anterior left, dorsal up. Scale bar=10 μm. (E) Alignment (red) shows the fraction of AP edges (edges oriented at 75–105° relative to the AP axis) connected to at least one other AP edge (n=14 embryos). The increase in alignment preceded tissue elongation (black line, the AP length of a group of cells normalized to the length at t=0 minus 1). The cumulative fraction of isolated (green), or linked (blue) shrinking edges is shown (n=10 embryos). t=0 indicates the onset of elongation. Error bars show standard error of the mean in all figures. (F) Alignment was reduced in embryos mutant for bcd nos tsl (blue, n=5) (P=5.0×10⁻⁴) or eve (green, n=3) (P=8.6×10⁻⁴) compared to wild type (red). (G) Localization of Myo:GFP in a twist snail mutant embryo. Anterior left, dorsal up.

Figure 2. Myosin II distribution is nonrandom in intercalating cells

(A) Wild-type embryo expressing baz:GFP (red) and myo:mCherry (green, single channel in A'). Anterior left, dorsal up. Scale bar=10 μm. (B) Quantitation of myosin intensity. Edges were scored as myosin-positive (green) if they had a mean pixel intensity I ≥ 1.15 relative to DV edges and were oriented at 60–120° relative to the AP axis. (C–E) Myosin distributions in vivo (red, n=5788 edges in 5 embryos) were not recapitulated in Monte Carlo simulations (blue, n=500,000 simulations). Data shown are for (C) all time points (P=2.0×10^–8), (D) stage 7 (t=0–5 min) (P=4.6×10⁻³), and (E) stage 8 (t=10–20 min) (P=4.8×10⁻⁶).

Figure 3. Multicellular actomyosin cables sustain increased mechanical tension

(A–F) Cells expressing E-cadherin:GFP before (A–F) and after (A'–F') ablation. Arrowheads indicate the vertices attached to the cut edge. Anterior left, dorsal up. Scale bar=5 μm. (A"–F") Kymographs show vertex displacement over time. (G–I) Quantitation of laser ablation experiments. (G) The increase in distance between the vertices attached to the cut edge was greater for linked AP edges (red, n=13) than for isolated AP (blue, n=10) (P=7.0×10⁻⁴) or DV edges (green, n=11) (P=2.9×10⁻⁶). (H) Peak retraction velocities in wild-type uninjected embryos (red bars) were greater for linked AP edges than for isolated AP (P=0.002) or DV edges (P=2.5×10⁻⁵). Peak retraction velocities in injected embryos (blue bars) were lower for AP edges in embryos injected with Y-27632 (n=8 ablations) compared to water-injected controls (n=8) (P=0.0017). Peak retraction velocities were higher for DV edges in embryos injected with Calyculin A (n=3) compared to DMSO-injected controls (n=3). Differences in retraction velocities for AP and DV edges in DMSO-injected embryos (1.07±0.10 vs. 0.53±0.13 μm/s, P=0.03) were abolished in Calyculin-injected embryos (0.87±0.19 vs. 0.91±0.15 μm/s, P=0.90). (I) (Left) Retraction distances correlated with the length of the cable (red bars show the 4 shortest and 4 longest cables, P=0.011), but not with the length of the ablated edge (blue bars show the 4 shortest and 4 longest edges, P=0.95) (n=20 ablations). (Right) Retraction distances correlated with the total myosin level in the cable (red bars show the 4 lowest intensity and 4 highest intensity cables, P=0.008), but not with the level of myosin in the ablated edge (blue bars show the 4 lowest intensity and 4 highest intensity edges, P=0.57). (J) Modeling laser ablation data as the recoil of an elastic fiber in a damped environment revealed similar relaxation times in all experiments. The maximum retraction distance was greater for linked AP edges than for isolated AP (P=0.0022) or DV edges (P=9.4×10⁻⁶).

Figure 4. Myosin II dynamics are regulated by mechanical tension

(A,B) Myosin II levels in living embryos. (A) Linked AP edges had higher levels of Myo:mCherry than isolated AP edges throughout intercalation (all time points, P=2.4×10⁻¹¹; t=0–5 min, P=0.008; t=10–20 min, P=3.6×10⁻¹⁰). (B) Linked AP edges had higher levels of Myo:mCherry than isolated AP edges regardless of edge length (P=0.033, 9.0×10⁻⁶, 0.027, 0.0097, for edges with lengths 1–3, 3–5, 5–7 and 7–9 μm). n=1552 DV, 359 isolated AP, and 2216 linked AP edges in 5 embryos. (C–F) Myo:GFP is stabilized in linked AP edges. (C) Pre-bleach (blue), post-bleach (green) and post-recovery (red) fluorescence intensities (n=13 isolated AP, 22 linked AP edges). (D) The percent recovery of pre-bleach fluorescence was reduced for linked AP edges (P=0.03). (E) The mobile fraction is negatively correlated with pre-bleach fluorescence (R=−0.60, P=2.0×10⁻⁴). Isolated AP edges (blue dots), linked AP edges (red dots). Dotted lines represent the best linear fits. (F) The width (σ) of the bleached region decreased during fluorescence recovery at isolated AP edges (blue bars, P=0.0057), but did not change at linked AP edges (red bars, P=0.65). (G) Myo:GFP intensity in cables decreased after line ablations (red line, 27 edges) compared to control edges anterior or posterior to the ablated region (blue line, 20 edges) (P=0.0018). Measurements were normalized to the fluorescence prior to ablation. (H–J) Aspiration pressure recruits Myo:GFP to the cortex. (H) Myo:GFP intensity at the cortex relative to fluorescence prior to aspiration for the experiment shown in J (blue line). Resille:GFP (red line) and Spider:GFP (green line) were unaffected by aspiration. (I) A micropipette (red lines) was inserted into the perivitelline space and negative pressure (blue arrow) was applied to deform the apical surface of contacting cells. (J) Cross-sections from a wild-type embryo expressing Myo:GFP. Bright-field image (left) shows the position of the micropipette. Scale bar=10 μm. t=0 is the time of aspiration.