Plant tissues in 3D via X-ray tomography: simple contrasting methods allow high resolution imaging

Abstract

Computed tomography remains strongly underused in plant sciences despite its high potential in delivering detailed 3D phenotypical information because of the low X-ray absorption of most plant tissues. Existing protocols to study soft tissues display poor performance, especially when compared to those used on animals. More efficient protocols to study plant material are therefore needed. Flowers of Arabidopsis thaliana and Marcgravia caudata were immersed in a selection of contrasting agents used to treat samples for transmission electron microscopy. Grayscale values for floral tissues and background were measured as a function of time. Contrast was quantified via a contrast index. The thick buds of Marcgravia were scanned to determine which contrasting agents best penetrate thick tissues. The highest contrast increase with cytoplasm-rich tissues was obtained with phosphotungstate, whereas osmium tetroxide and bismuth tatrate displayed the highest contrast increase with vacuolated tissues. Phosphotungstate also displayed the best sample penetration. Furthermore, infiltration with phosphotungstate allowed imaging of all plants parts at a high resolution of 3 µm, which approaches the maximum resolution of our equipment: 1.5 µm. The high affinity of phosphotungstate for vasculature, cytoplasm-rich tissue, and pollen causes these tissues to absorb more X-rays than the surrounding tissues, which, in turn, makes these tissues appear brighter on the scan data. Tissues with different brightness can then be virtually dissected from each other by selecting the bracket of grayscale to be visualized. Promising directions for the future include in silico phenotyping and developmental studies of plant inner parts (e.g., ovules, vasculature, pollen, and cell nuclei) via virtual dissection as well as correlations of quantitative phenotypes with omics datasets. Therefore, this work represents a crucial improvement of previous methods, allowing new directions of research to be undertaken in areas ranging from morphology to systems biology.

Figure 1. Sample mounting techniques for samples of different sizes.

(A) Mounting of large samples (>10 mm). Samples are mounted in acryl foam and scanned in a solvent atmosphere, which provides optimal signal to background ratio. (B–D) Mounting of medium-sized samples (1–10 mm). For medium-sized samples, movements of solvent surface cause significant sample movement; therefore samples are best scanned immersed in the solvent. (B) Single sample are mounted in pipette tips, which are inserted in aluminium tubes and glued to the latter. (C) In the pipette tips, paraffin wax is used as a seal and to stabilize the sample on the lower end of the tip. PARAFILM is used to seal the upper end of the tip. (D) Sample batch with batch holder. Batch scanning is required to scan large numbers of samples. Samples mounted in pipette tips are mounted in a 1 ml syringe tube and stabilized with resin. (E, F) Mounting of small samples (>1 mm). (E) Small sample with sample holder. (F) Small samples require maximal stabilization and contrast enhancement. Such samples are best scanned after having been critical point dried and embedded in a drop of epoxy glue directly on aluminium stubs.

Figure 2. Contrast improvement over time for different floral tissues.

(A) Contrast index vs. time for all contrasting agents on stamen filament. Filament cells are highly vacuolated, and the highest contrast index values are obtained with bismuth tartrate and osmium tetroxide and lead citrate, all of which bind strongly to the cell wall components. (B) Contrast index vs. time for all contrasting agents on ovules. Ovule cells are highly cytoplasmic, the highest contrast index values are obtained with phosphotungstate, that binds proteins and cell membranes. Abbreviations: Bi = bismuth tartrate; I.lugol = Lugol’s solution; Mn = potassium permanganate; OsO4 = osmium tetroxide; OsFeCN = osmium tetroxide with ferrocyanate; Pb = lead citrate; U = uranyl acetate; W/EtOH = phosphotungstate in 70% EtOH; W/FAA = phosphotungstate in FAA. Samples infiltrated with alcoholic and aqueous iodine are not detectable under the used scanning conditions.

Figure 3. Reproducibility and speed of contrast improvement: distribution of half-saturation of contrast agents per tissue.

The mathematical model for the saturation of infiltration agents (C(t) = C_∞*e^(−1/kt) ) is regressed on 1000 random permutations of data points (one per time point). The coefficients obtained via these regressions allow the calculation of 1000 half-saturation times by applying the equation: (A) Half-saturation times distribution for stamen filaments. (B) Half-saturation times distribution for ovules. In both vacuolated and cytoplasmic tissues, the fastest and most consistent (narrow spread) reagents are the most reactive ones: permanganate and osmium tetroxide (both strong oxidants), and bismuth tartrate (which is in 2N sodium hydroxide). Although making for the highest contrast increase in cytoplasmic tissues, phosphotungstate appears relatively slow (due to high saturation values) and comparatively little reliable (large spread of half-saturation values). Abbreviations: Bi = bismuth tartrate; I.lugol = Lugol’s solution; Mn = potassium permanganate; OsO4 = osmium tetroxide; OsFeCN = osmium tetroxide with ferrocyanate; Pb = lead citrate; U = uranyl acetate; W/EtOH = phosphotungstate in 70% EtOH; W/FAA = phosphotungstate in FAA. Samples infiltrated with alcoholic and aqueous iodine are not detectable under the used scanning conditions.

Figure 4. Tissue penetration: different contrasting agents penetrate the thick buds of Marcgravia caudata with different efficiencies.

(A) Reconstructed transverse sections through buds of M. caudata after 8 days infiltration with selected contrast agents. The sections were chosen at the level of the developing stigma (center) and thecae. Note the thick protective calyptra (fused petals). Scale bar = 1 mm. (B) Air-referenced contrast index profiles through the pictures from part (A). Bismuth tartrate and phosphotungstate clearly outperform other stains for contrast increase and sample penetration. The lead carbonate crystals deposited on the surface of the buds induce two sharp peaks. Osmium tetroxide fails to penetrate thick samples as evidenced by higher contrast index values on the outer portion of the buds, but lower levels on the inside of the bud. Abbreviations: Bi = bismuth tartrate; I.lugol = Lugol’s solution; Mn = potassium permanganate; OsO4 = osmium tetroxide; OsFeCN = osmium tetroxide with ferrocyanate; Pb = lead citrate; U = uranyl acetate; W/EtOH = phosphotungstate in 70% EtOH; W/FAA = phosphotungstate in FAA.

Figure 5. Reproducibility and uniformity, sample damage, and specificity of contrasting agents.

(A) Permanganate typically contrasts one part of the sample and not the rest: A. thaliana flower after 2d infiltration with well contrasted thecae, and barely visible gynoecium. (B) Osmium tetroxide does not penetrate well inside samples and tends to accumulate on the periphery of organs (A. thaliana, 8d infiltration). (C) Lead citrate precipitates in presence of CO2 and forms lead carbonate crystals that accumulate at the sample periphery. The crystals absorb large amounts of X-rays, and cause reconstruction artifacts. (D) A. thaliana flower contrasted with phosphotungstate (8d), showing near ideal properties: good sample penetration and differential contrast increase. (E) and (F) A. thaliana flower, sample degradation after 2d in bismuth tartrate, and 2d in permanganate, respectively. (G) Specific staining of the nucellus of Calycanthus floridus by bismuth tartrate (8d) infiltration (possibly due to the presence of starch). (H) Specific staining of the PTTT of Haplophyllum lissonotum by phosphotungstate (possibly due to the presence of glycoproteins).

Figure 6. Virtual dissection of A. thaliana flower: differential contrast increase allows to handle different floral parts independently.

(A–D) Identical 3D model of the same A. thaliana flower visualized with narrower and narrower selection of pixel grayscale (in D, only the brightest pixels are displayed). A, Full model with broadest selection of pixel grayscale. (B) From the petals, only the vasculature remains. (C),Petals and filaments, and parts of the sepals are removed; the ovary and the pollen grains remain. (D) Most of the flower is removed, and only the ovules, their vasculature and the pollen grains remain. Diameter of peduncke highlighted in (A) = 372 µm.

Figure 7. Special protocol for very small and soft objects.

A. thaliana flower meristems of double line ap1 cal pAP1::AP1-GR mutant (courtesy Toshiro Ito research group, Temasek Life Sciences Laboratory, National University of Singapore, unpublished data) 6d after induction of flower development. Flowers were infiltrated 1 week in phosphotungstate in FAA, critical point dried, sputtered with gold and mounted in glue. (A) 3D model. (B) Reconstructed longitudinal section of the inflorescence. The cells of the meristem are resolved. Asterisk denotes a developing gynoecium. Scale bar = 100 µm.