ClearSee: a rapid optical clearing reagent for whole-plant fluorescence imaging

Abstract

Imaging techniques for visualizing and analyzing precise morphology and gene expression patterns are essential for understanding biological processes during development in all organisms. With the aid of chemical screening, we developed a clearing method using chemical solutions, termed ClearSee, for deep imaging of morphology and gene expression in plant tissues. ClearSee rapidly diminishes chlorophyll autofluorescence while maintaining fluorescent protein stability. By adjusting the refractive index mismatch, whole-organ and whole-plant imaging can be performed by both confocal and two-photon excitation microscopy in ClearSee-treated samples. Moreover, ClearSee is applicable to multicolor imaging of fluorescent proteins to allow structural analysis of multiple gene expression. Given that ClearSee is compatible with staining by chemical dyes, the technique is useful for deep imaging in conjunction with genetic markers and for plant species not amenable to transgenic approaches. This method is useful for whole imaging for intact morphology and will help to accelerate the discovery of new phenomena in plant biological research.

Keywords: Arabidopsis thaliana; Clearing; Confocal microscopy; Deep imaging; Physcomitrella patens; Two-photon microscopy; Whole plant.

Fig. 1.

Screening of chemical clearing solutions for Arabidopsis plant tissues. (A) Fixed leaves were incubated with chemical solutions (#1-24). Autofluorescence of extracted chlorophyll was measured after incubation for 7 days. (B-D) Recombinant Venus proteins were incubated with chemical solutions. The fluorescent signal intensities were measured after 1 day of incubation for the first (B), second (C) and third screening (D). Mean±s.e. shown (n=3). (E) Fixed seedlings were incubated in ClearSee or PBS (control) for 2 weeks. In the righthand panel, the samples are shown on the illuminator. Scale bars: 5 mm.

Fig. 2.

Optical clearing of Arabidopsis leaf using ClearSee. (A) Fixed UBQ10pro::H2B-mClover leaves were incubated in clearing solutions for 4 days and placed on a grid sheet. Note that grid lines are clearly observed in the grid sheet with ClearSee-treated and chloral hydrate-based solution-treated leaf, whereas retention of green coloration and only limited transparency are shown in PBS-treated and Scale-like solution-treated leaves. (B) Treated UBQ10pro::H2B-mClover leaves were observed by fluorescence microscopy. Images of H2B-mClover were acquired using a U-FBNA (excitation 470-495 nm, emission 510-550 nm) filter. (C) Treated UBQ10pro::H2B-mClover mesophyll cells were observed by 2PEM with 950 nm excitation. Images were acquired in sequential 6 nm bandwidths spanning the wavelength range 463.9-649.2 nm to generate a lambda stack containing 32 images. (D) Autofluorescence spectrum in leaves treated with various clearing solutions. The measurement regions are indicated by white circles in C. Mean values±s.e. shown (n=3 regions). Scale bars: 1 mm in A; 30 µm in B,C.

Fig. 3.

Comparison of imaging penetration for CLSM and 2PEM in ClearSee-treated Arabidopsis root tips. (A) DR5rev::3xVenus-N7 (green); RPS5Apro::H2B-tdTomato (magenta) root treated with ClearSee for 4 days (ClearSee), or after (fixed) and before (live) fixation without ClearSee treatment. Optical xy and xz sections were generated from 150 z-stack images with 1.0 µm intervals by CLSM with 488 nm and 561 nm excitation (confocal) and 2PEM with 950 nm excitation (two-photon). Beneath are cross-sections at the positions indicated by the colored lines (1, transition zone; 2, meristematic zone). The top of the xz section images is facing the objective lens. (B,C) Fluorescence intensities of DR5rev::3xVenus-N7 recorded at positions 1 and 2. Scale bars: 30 µm.

Fig. 4.

Application of ClearSee for whole-leaf imaging and staining with chemical dyes. (A-D) UBQ10pro::H2B-mClover leaves fixed with PFA (A,B) and treated with ClearSee for 4 days (C,D). Optical xy (left) and xz (right) maximum-intensity projections were generated from 100 z-stack images with 1.0 µm intervals by CLSM with 488 nm excitation (A,C) and by 2PEM with 950 nm excitation (B,D). The color bar indicates depth from the leaf surface. (E) Cell wall stained with Calcofluor White (cyan) in ClearSee-treated UBQ10pro::H2B-mClover (yellow) leaves observed by 2PEM with 950 nm excitation. Left image shows optical xy section. The xz image on the right is a cross-section at the position indicated by the magenta line. Arrowheads indicate stomata. Scale bars: 100 µm.

Fig. 5.

Application of ClearSee for multicolor imaging of the whole pistil. (A) UBQ10pro::H2B-mClover pistil treated with ClearSee for 6 days. Optical xy and xz sections were generated from 410 z-stack images with 1.0 µm intervals by 2PEM with 950 nm excitation. (B) Pistil pollinated with LAT52pro::mTFP1 and LAT52pro::Venus pollen and treated with ClearSee for 5 weeks. Maximum intensity projections for xy view and xz sections were generated from 60 z-stack images with 6.0 µm intervals by 2PEM with 950 nm excitation. Each image on the left in A and the bottom images in B represent xz cross-sections at the positions indicated by the colored lines (1, stigma; 2, style; 3, ovary). Asterisks indicate discharged pollen tubes. The top of the xz section images is facing the objective lens. Scale bars: 100 µm.

Fig. 6.

ClearSee is applicable for long-term storage. Pistil pollinated with LAT52pro::mTFP1, LAT52pro::sGFP, LAT52pro::Venus, and LAT52pro::mApple pollen and treated with ClearSee for 5 months. Maximum intensity projection for xy sections was generated from 96 z-stack images with 3.0 µm intervals by 2PEM with 990 nm excitation. Images were acquired in sequential bandwidths of 8 nm spanning the wavelength range 460-648 nm to generate a lambda stack containing 19 images. Scale bar: 50 µm.

Fig. 7.

Phloem patterning in the whole seedling. (A-E) SUC2pro::RCI2A-mCitrine seedling treated with ClearSee for 7 days. Maximum intensity projection for xy view was generated from 67 z-stack images with 10 µm intervals by 2PEM with 950 nm excitation. Boxed regions in A are magnified in B-E. (F,G) Reconstituted 3D image of seedling with rosette leaves expressing SUC2pro::RCI2A-mCitrine after ClearSee treatment for 7 days. Arrowheads indicate spiral xylem vessels. Arrow indicates extension of phloem into rosette leaf from root. Scale bars: 1 mm in A; 100 µm in B-E.

Fig. 8.

Clearing of a leafy gametophore of Physcomitrella patens with ClearSee. A leafy gametophore of the H2B-mRFP line of P. patens treated with ClearSee for 4 days. Images were collected in the ranges of 570-668 nm for H2B-mRFP and 672-701 nm for autofluorescence with 561 nm excitation by CLSM. (A) Maximum-intensity projections were generated from 325 z-stack images with 1.0 µm intervals for living and ClearSee-treated gametophores. (B) Optical slice of the apical region of gametophore covered with juvenile gametophore leaves. Scale bars: 100 µm.