Quantitation of cellular dynamics in growing Arabidopsis roots with light sheet microscopy

Abstract

To understand dynamic developmental processes, living tissues have to be imaged frequently and for extended periods of time. Root development is extensively studied at cellular resolution to understand basic mechanisms underlying pattern formation and maintenance in plants. Unfortunately, ensuring continuous specimen access, while preserving physiological conditions and preventing photo-damage, poses major barriers to measurements of cellular dynamics in growing organs such as plant roots. We present a system that integrates optical sectioning through light sheet fluorescence microscopy with hydroponic culture that enables us to image, at cellular resolution, a vertically growing Arabidopsis root every few minutes and for several consecutive days. We describe novel automated routines to track the root tip as it grows, to track cellular nuclei and to identify cell divisions. We demonstrate the system's capabilities by collecting data on divisions and nuclear dynamics.

Figure 1. Imaging system.

(A) Plant micro-chamber details. Left panel, view from the top; Right panel, view from the side showing inflow and outflow of perfusion medium. B, beads; CT, capillary tube; Ch, channel; Co, collar. (B) Optics. L, laser; F1, excitation filter; PC, Pockels cell; BE, beam expander; I, iris; CL, cylindrical lens; IGC, plant imaging and growing micro-chamber on a stage moving in xyz; O, objective; F2, emission filter; C, camera on a stage moving in xz. (C) 3D rendering of the laser sheet scanning the corner of the cuvette where the root is growing; P, plant (D) Long-term sustained root growth. Root tip speed (blue line) as measured during tracking in a typical experiment; average fluorescent intensity normalized to the beginning of experiment, detected from the same root expressing 35S::H2B::YFP (green line).

Figure 2. Deconvolution and segmentation.

(A) Arabidopsis root expressing 35S::H2B::YFP. Maximum intensity projections before (left panel) and after (right panel) deconvolution; the horizontal axis corresponds to the objective axial dimension; bar, 50 µm. (B) Cell divisions from same root in (A), time-lapse. Three cells divide giving rise to daughters labeled with “a” and “b”; bar, 5 µm. (C) Density plot of segmented nuclei. Projection along the longitudinal axis of all the nuclei recognized by the segmentation, accumulated in a 29 hour period. Five regions used in Table S1: I, towards objective; II, away from laser source; III, away from objective; IV towards laser source; V center of the root (ρ<30 µm).

Figure 3. Trajectories.

(A) Cylindrical coordinates. Schematic of the root tip and the cylindrical coordinate system used in text: m, distance from the tip on the main longitudinal axis of the root; ρ, distance from the center of the root; θ, angle with respect to the axis normal to m and with maximal component z (see Fig. 1B), where θ = 0 towards the objective and θ = π/2 towards the laser source. (B) Nuclear displacements along a typical trajectory, displayed along the main axis of the root (m, top panel), the m-normal closest to z (ρcosθ, middle panel) and the m-normal closest to x (ρsinθ, bottom panel); bar, 10 µm. (C) High frequency (left panel, freq.>0.23 hr⁻¹; bar, 1 µm) and low frequency (right panel, freq.<0.23 hr⁻¹; bar, 20 µm) components of the trajectory shown in (B).

Figure 4. Analysis of collective nuclear velocities.

Heatmaps and plots for each cylindrical component of the velocity, from data collected in a 29 hour temporal interval. In the heatmaps, only bins containing 36 or more nuclei are shown; in the plots, the solid black line is the median per 5 micron bin, while the dashed black lines are the 25% and 75% quartiles. (A) m component of the velocity; “early”, temporal window 0–21 hours; “late”, temporal window 21–29 hours; the red dotted line is the linear fit of v_m for the region 80–300 µm. (B) ρ component of the velocity. (C) θ component of the velocity multiplied by the position ρ in the radial dimension.

Figure 5. Cell divisions.

(A) Nuclear displacements along a typical trajectory showing branching, displayed along the main axis of the root (m, top panel), m-normal closest to z (ρcosθ, middle panel) and the m-normal closest to x (ρsinθ, bottom panel); arrow, branching point indicating cell division; bars, 20 µm. (B) Temporal distribution of the cell divisions detected in the 29 hour interval analyzed (true positives, N = 340). (C) Position and orientation of the same cell divisions, superimposed to a smoothed density estimate of all the nuclei segmented at all time points throughout the same 29 hour interval; main direction of divisions: red, Δm; blue, ρΔθ; green, Δρ. (D) Polar histograms for the angles of the cell division orientation to Δm, in the plane defined by Δm and Δρ (top panel) and by Δm and ρΔθ (bottom panel); color code as in (C).