(a) Side view of the central components of a digital scanned laser light-sheet fluorescence microscope. The illumination lens illuminates a thin volume by rapidly scanning a micrometer-sized laser beam through the specimen. Fluorescence is detected at a right angle to the illuminated plane by the detection lens. The intensity of the laser beam is modulated in synchrony with the scanning process. (b) Cross-sections of DSLM and DSLM-SI illumination profiles, recorded at 10× magnification. A uniform intensity distribution is used in the standard DSLM light-sheet illumination mode (LS). In the DSLM-SI mode, the spatial laser light intensity distribution is modulated in a sinusoidal fashion. The patterns shown here as reflections of a mirror oriented at 45° to the incident laser beam range from four sine periods (SI frequency s = 4; SI-4) to 180 sine periods (SI frequency s = 180; SI-180) across the field of view. The width of the image corresponds to a field size of 1.5 mm. Center-to-center distances of the maxima in the patterns range between 412.5 μm (SI-4) and 9.2 μm (SI-180).

(a) Maximum-intensity projection of a DSLM image stack of a 3.5-day old Medaka fish embryo with Sytox Green nuclear staining, recorded with standard light-sheet illumination (10× magnification). Stack dimensions are 435 images, recorded at a z-dimension spacing of 3 μm, covering a total volume of 1,516 μm × 1,516 μm × 1,305 μm. (b) Magnification of the boxed region in a, recorded in standard light-sheet mode (top) and in structured-illumination mode (SI frequency s = 32; SI-32; bottom). (c) Intensity plot along the lines indicated in b. In both plots, the raw intensity values were normalized by the same factor (global maximum of both images). Shading highlights structures in the structured-illumination image that are not visible in the light-sheet mode image. (d) Maximum-intensity projections of a DSLM time-lapse recording of a membrane- and nuclei-labeled zebrafish embryo injected with ras-eGFP mRNA and H2A-mCherry mRNA at the one-cell stage. Membranes were imaged using structured illumination (SI frequency s = 25; SI-25), and nuclei were imaged using standard light sheet illumination (LS). (m.p.f., minutes post fertilization.) Images were deconvolved with ten iterations of the Lucy-Richardson algorithm²⁰. Scale bars, 200 μm (a), 50 μm (b) and 100 μm (d). Fluorescence was detected with a Carl Zeiss C-Apochromat 10×/0.45 W objective. Recording speeds were six DSLM images per second and two structured illumination–reconstructed images per second.

(a) Maximum-intensity projections of DSLM image stacks of a nuclei-labeled Drosophila embryo, using standard light-sheet illumination (LS, left) and structured illumination (SI frequency s = 20; SI-20; right) at the indicated times. The image pairs show the same volume at the same time point. (b) Maximum-intensity projections of a DSLM-SI multiple-view time-lapse recording of a nuclei-labeled Drosophila embryo at the indicated times. Images were deconvolved with the Lucy-Richardson algorithm (five iterations). (c) Lateral snapshots of the fly digital embryo. Nuclei were automatically detected in the four views of the developing Drosophila embryo. The resulting point clouds were fused. Color bar indicates directed regional nuclei movement speeds over 10-min periods. Scale bars, 100 μm. Fluorescence was detected with a Carl Zeiss Plan-Apochromat 20×/1.0 W objective. Recording speeds were six DSLM images per second and two structured illumination reconstructed images per second.

(a,b) DSLM images of a nuclei-labeled stage 5 Drosophila embryo (before internal structures have developed; a) and at stage 8 (b) at two different depths (z) inside the specimen, using standard light-sheet illumination (LS) and structured-illumination patterns with different frequencies (SI-1 to SI-101). The dashed line separates the region containing the background and out-of-focus signal that has been removed by DSLM-SI (left) from the in-focus structures where the contrast has been enhanced (right). Scale bar, 100 μm. Fluorescence was detected with a Carl Zeiss Plan-Apochromat 20×/1.0 W objective. Recording speeds were five DSLM images per second and 1.7 structured illumination–reconstructed images per second. (c,d) Quantification of image contrast as a function of structured-illumination frequency at the indicated imaging depths inside the specimen, for the time points shown in a and b, respectively. The horizontal line indicates the image contrast obtained with standard light-sheet illumination (LS). Colored spheres indicate structured-illumination frequencies for which the maximum image contrast was obtained.