Increased spatial and axial resolution achieved by 3D STED in HL-60 cells. (A) Confocal (left), 3D STED (middle), and merged (right) image of an xz slice of the ventral membrane of an HL-60 cell labeled with CellMask Deep Red. Scale bar is 1 μm. (B) Representative fluorescence intensity profile along the axial direction in confocal (green) and 3D STED (magenta). (C) Histogram of axial resolution in confocal (green) and 3D STED (purple) imaging mode. 3D STED imaging provides a 4.4× increased axial resolution (160 ± 16 nm for 3D STED versus 704 ± 31 nm for confocal). The axial resolution is determined by the full-width half-maximum of a Gaussian fit to axial intensity profiles of the confocal and 3D STED images (n = 1,600 profiles). (D) Estimation of the lateral (xy) spatial resolution offered by 3D STED through imaging fluorescent beads (40 nm) in confocal (left) and 3D STED (right) mode with the same settings as in live-cell experiments. (E) Histogram of spatial resolution in confocal (green) and 3D STED (purple) imaging mode. 3D STED imaging provides a 1.6× increased spatial resolution (160 ± 6 nm for 3D STED versus 255 ± 17 nm for confocal). Spatial resolution is calculated by the full-width half-maximum of 2D Gaussian fits to the beads (n = 102 beads). Scale bar is 1 μm. (F) Integrated intensity of thresholded WASP signal across xyz stacks reveals significant enrichment of WASP to 200-nm beads compared with 500-nm beads (nonneck invaginations) when normalized to bead surface area. Mean normalized WASP intensity per unit area is 4.08 ± 0.59 a.u./nm² for 200-nm beads and 0.97 ± 0.15 a.u./nm² for 500-nm beads. P = 8.81E-05 by an unpaired two-tailed t test on the normalized WASP intensities at all beads. n_{200 nm} = 18 beads collected from two experiments and n_{500 nm} = 14 beads collected from one experiment. AU, arbitrary units.