Loss of SUN1/2 results in changes of spindle tilt and morphology. (A) Schematic representation of the experimental approach to determine spindle pole-to-pole distance and tilt. Stacks of 0.3-µm sections were analyzed. Spindle length (d) was measured as the distance of the centrosomes in 3D. Spindle tilt is represented by the angle α, which is calculated as the arcsin of the pole-to-pole distance in z divided by d. (B) Immunofluorescence of spindles in control RNAi or SUN1/2-depleted cells using α-tubulin and pericentrin antibodies. Representative images are shown as three scans of different z sections and their maximum intensity projections. (C) Quantitative analysis of tilted spindles displayed by the distribution of cells into classes of spindle angles (α in degrees) from n ≥ 121 (three experiments; means ± SD). (D) HeLa cells stably expressing H2B-mCherry and centrin2-GFP were depleted of SUN1/2 by RNAi for 72 h. Images show mitotic cells stained for α-tubulin and pericentrin. (E) Bar plot representing the mean percentage of defective spindles (aberrant MTs and segregated/supernumerary spindle poles) upon SUN1/2 RNAi compared with control cells (three experiments; means ± SD; n ≥ 40 per condition and experiment). (F) Cells as in D were stained for pericentrin. (G) Bar plot representing the mean percentage of segregated/supernumerary interphase centrosomes upon SUN1/2 RNAi compared with control cells (three experiments; n ≥ 86 per condition; means ± SD). Bars: (B and D) 5 µm; (F) 10 µm.