Figure 4.
Dynein counteracts left-handed twist in the anaphase spindle. (A) Schematic diagram of LGN, NuMA, and dynein localization in anaphase cells. Dynein and NuMA cluster microtubule minus ends and localize to spindle poles, while LGN–NuMA–dynein complexes localize to cortical crescents where they exert pulling forces on astral microtubules. The purple arrow indicates the direction of dynein stepping, and the gray arrow indicates the direction of force on astral microtubules. (B) Immunofluorescence images (single z-planes) of anaphase MCF10A cells stained for NuMA (magenta) and DNA (blue). Scale bars = 5 µm. (C) Quantification of NuMA enrichment at the cell cortex relative to NuMA intensity in the cytoplasm (see Materials and methods), in immunofluorescence images of MCF10A cells transfected with siRNA targeting luciferase (control) or LGN. The dashed line indicates a cortex/cytoplasm ratio of 1, i.e., no cortical enrichment. Black lines represent mean ± SD. n = 22 cells from 2 independent days in each condition. ****P = 1.55 × 10−10, two-sample t test. (D) Confocal images of live MCF10A cells transfected with siRNA targeting luciferase (siCtrl), dynein heavy chain, or LGN labeled with SiR-tubulin (see also Video 4). Maximum intensity projections of a 2-µm thick low region (magenta) and a 2-µm thick high region (green) relative to the spindle midplane are overlaid. Dashed lines highlight individual microtubule bundles in each region. The helicity of each spindle is indicated in the top right. Positions of spindle poles are indicated by white circles. Scale bars = 3 µm. (E) Helicity of anaphase spindles calculated from SiR-tubulin intensity. Black lines represent mean ± SD. n = 65, 45, and 78 spindles pooled from N = 6, 5, and 7 independent experiments for siCtrl, siDHC, and siLGN, respectively. ***P = 9.43 × 10−5, **P = 9.03 × 10−4, one-way ANOVA with Tukey’s post-hoc test. (F) Helicity of metaphase spindles calculated from SiR-tubulin intensity. Black lines represent mean ± SD. n = 39, 38, and 35 spindles pooled from N = 4, 4, and 2 independent experiments for siCtrl, siDHC, and siLGN, respectively. n.s., not significant, one-way ANOVA with Tukey’s post-hoc test. (G) Proposed models for twist regulation in the anaphase spindle. Motors within the spindle midzone generate left-handed torques (blue arrows), which are partially counteracted by LGN-NuMA-dynein complexes at the cell cortex. These cortical complexes could actively generate right-handed torques on astral microtubules (magenta arrows, Model 1), exert outward pulling on spindle poles (Model 2), or serve to anchor astral microtubules in the cortex (Model 3). Together, these active and passive torques establish weak global left-handed twist in the anaphase spindle.

Dynein counteracts left-handed twist in the anaphase spindle. (A) Schematic diagram of LGN, NuMA, and dynein localization in anaphase cells. Dynein and NuMA cluster microtubule minus ends and localize to spindle poles, while LGN–NuMA–dynein complexes localize to cortical crescents where they exert pulling forces on astral microtubules. The purple arrow indicates the direction of dynein stepping, and the gray arrow indicates the direction of force on astral microtubules. (B) Immunofluorescence images (single z-planes) of anaphase MCF10A cells stained for NuMA (magenta) and DNA (blue). Scale bars = 5 µm. (C) Quantification of NuMA enrichment at the cell cortex relative to NuMA intensity in the cytoplasm (see Materials and methods), in immunofluorescence images of MCF10A cells transfected with siRNA targeting luciferase (control) or LGN. The dashed line indicates a cortex/cytoplasm ratio of 1, i.e., no cortical enrichment. Black lines represent mean ± SD. n = 22 cells from 2 independent days in each condition. ****P = 1.55 × 10−10, two-sample t test. (D) Confocal images of live MCF10A cells transfected with siRNA targeting luciferase (siCtrl), dynein heavy chain, or LGN labeled with SiR-tubulin (see also Video 4). Maximum intensity projections of a 2-µm thick low region (magenta) and a 2-µm thick high region (green) relative to the spindle midplane are overlaid. Dashed lines highlight individual microtubule bundles in each region. The helicity of each spindle is indicated in the top right. Positions of spindle poles are indicated by white circles. Scale bars = 3 µm. (E) Helicity of anaphase spindles calculated from SiR-tubulin intensity. Black lines represent mean ± SD. n = 65, 45, and 78 spindles pooled from N = 6, 5, and 7 independent experiments for siCtrl, siDHC, and siLGN, respectively. ***P = 9.43 × 10−5, **P = 9.03 × 10−4, one-way ANOVA with Tukey’s post-hoc test. (F) Helicity of metaphase spindles calculated from SiR-tubulin intensity. Black lines represent mean ± SD. n = 39, 38, and 35 spindles pooled from N = 4, 4, and 2 independent experiments for siCtrl, siDHC, and siLGN, respectively. n.s., not significant, one-way ANOVA with Tukey’s post-hoc test. (G) Proposed models for twist regulation in the anaphase spindle. Motors within the spindle midzone generate left-handed torques (blue arrows), which are partially counteracted by LGN-NuMA-dynein complexes at the cell cortex. These cortical complexes could actively generate right-handed torques on astral microtubules (magenta arrows, Model 1), exert outward pulling on spindle poles (Model 2), or serve to anchor astral microtubules in the cortex (Model 3). Together, these active and passive torques establish weak global left-handed twist in the anaphase spindle.

This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal