Micro-CT for multiscale analysis of radial O 2 profiles in roots of Melilotus siculus

Transcription

1 Micro-CT for multiscale analysis of radial O 2 profiles in roots of Melilotus siculus Pieter Verboven 1*, Ole Pedersen 2,3, Els Herremans 1, Quang Tri Ho 1, Timothy David Colmer 2, Bart M. Nicolaï 1 1 BIOSYST-MeBioS, Faculty of Bioscience Engineering, KU Leuven, W. de Croylaan 42, 3001 Leuven, Belgium 2 School of Plant Biology, Faculty of Natural and Agricultural Sciences, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia 3 Freshwater Biological Laboratory, Department of Biology, University of Copenhagen, Helsingørsgade 51, DK-3400 Hillerød, Denmark 4 Centre for Ecohydrology, The University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. Aims Plants in waterlogged soils need to cope with the lack of O 2 available to roots from the soil. Waterlogging tolerant species possess mechanisms that enhance internal root aeration and thus growth in anoxic soil, and tissues can also tolerate short-term O 2 deficiency via anaerobic catabolism of sugars in fermentation as well as avoid or repair oxidative damage. Aerenchyma development and formation of a barrier to radial O 2 loss enhance internal O 2 supply to roots. Aerenchymatous phellem forms externally from a phellogen and contains large volumes of intercellular gas-filled spaces, and can be observed easily as white spongy tissue on outer parts of stems, hypocotyl, roots and nodules. Despite the apparent importance of aerenchymatous phellem for waterlogging tolerance in some dicotyledonous species, the structure of gas spaces and O 2 profiles within phellem and adjacent stelar tissue is still not well documented. To this end, a 3-D imaging method is required that can distinguish clearly the gas-filled spaces from cells within tissues. Computed tomography (CT) uses x-ray radiation, which can penetrate plant tissues (Mendoza et al., 2007; Verboven et al., 2008; ; Verboven et al., 2011; Herremans et al., 2013). The present study aimed to enhance understanding of O 2 supply to phellem-containing roots in anoxic media. To this end, micro-ct was used to image in 3-D the phellem gas-space structure. These images were used to calculate O 2 fluxes and model O 2 profiles, which were compared to O 2 microelectrode traces. Tissue porosity and respiration were also measured. Experiments used an annual legume, Melilotus siculus, as this species produces extensive aerenchymatous phellem on hypocotyl and roots and is of interest as a pasture species for waterlogged, saline land. Method Plant growth and analysis methods tissue characterization are explained in Verboven et al., (2011). An example of the phellem covered root of Melilotus siculus and a cross section is given in Figure 1. Root samples were taken for micro-ct imaging (Skyscan 1172, Kontich, Belgium). Diskshaped cross-sectional samples of minimum 5 mm thickness were cut from the hypocotyl and main root axis (tap root) with a scalpel and further manipulated with tweezers. Samples were taken from three random plants at different distances from the hypocotyl-root junction. If the disk shaped samples were too large to scan, pie-shaped sub-samples from the disks were taken to reduce the sample size and increase imaging resolution. The samples were wrapped in parafilm to prevent dehydration. To mount samples on the imaging stage, they were put inside a hollow Styrofoam cylinder (diameter 1 cm) that was sealed onto the stage with double-sided tape. The Styrofoam is almost invisible on the micro-ct images. Micro-CT

2 scans were taken at 55 kev and 180 µa; the exposure time was 560 ms and the rotation step 0.3. Total scan time was 35 min, resulting in 2000 cross-sectional image slices of 2000 x 2000 pixels each. The image pixel size depended on sample size: from 2.5 µm for a sample of 5 mm³ to 7.5 µm for a sample of 15 mm³. Volume renderings and quantitative calculation of porosity on the sample were performed by 3-D image segmentation and isosurface representations with Avizo Fire software (Visualization Group Sciences, Merignac, France). A microscale gas exchange model was solved over the 3-D microstructure geometry using the finite volume method to obtain characteristic diffusion properties of the tissues. 3-D rendered tomographic images of root tissues with edge dimensions of 660 µm were discretised into cubical control volumes with an edge of 6.6 µm. Diffusion model equations (see Ho et al., 2011) were discretised over the finite volume grid to yield a linear system of algebraic equations on the unknown concentrations at the nodes. The linear equation system was solved by the conjugate gradient method available in Matlab (The Mathworks, Natick, MA). The program was run on a 16 GB RAM node (Opteron 250; Xenon 5420 and Xenon 5560) of the high performance computer at the KU Leuven (Leuven, Belgium). A macroscale diffusion-reaction model that was developed earlier (Ho et al., 2011) was then used to predict O 2 profiles in the tap root due to diffusion and respiration in the roots. The aim of the model was to demonstrate the 3-D nature of observed radial profiles of oxygen partial pressure along the tap root using a multiscale modeling approach (Ho et al., 2013). Figure 1. Example of a 6-week old Melilotus siculus grown in stagnant agar nutrient solution for the final 3 weeks, and a transverse section of phellem and stele from a lateral root. (a) Photograph of the shoot base and hypocotyl-root junction, showing the first 10 cm of the root system (total root length average 20 ± 0.81 cm). Arrow indicates the approximate location of the hypocotyl-root junction. (b) Unstained partial transverse section of a phellem-covered lateral root showing the aerenchymatous phellem (ap), phellogen (pg) and secondary xylem (sx). (taken from Verboven et al., 2011). Results Fig. 2 plots a CT cross section and the rendered 3-D volume obtained from one of the root samples. The denser stele region can be clearly distinguished from the porous phellem tissue on both tap and lateral roots. The stele contains only a very small amount of gas spaces. The gas spaces are mostly concentrated in a ring zone beneath the phellem-stele interface (under the phellogen), additional gas spaces appear at the central axis of the root (Fig. 3a). The phellem has a wellconnected air space (Fig. 3b). Fig. 3c displays the size and 3-D connections of the pores in the stele under the phellogen. The pores are generally smaller than in the porous phellem with fewer visible connections. The pores in the stele have a few tiny connections to the pores in the phellem; these connections are marked in Fig. 3d.

3 Figure 2. CT cross section and reconstructed 3-D view of a quarter section of the tap root with phellem at a position of 1.5 cm below the hypocotyl of Melilotus siculus: (a) CT scan, (b) stele, (c) complete phellem, (d) subsample of the phellem layer (7.3 µm pixel size). The figures are showing the tap root (tr) and lateral roots (lr) with the central stele (s) covered by aerenchymatous phellem (ap); on the lateral roots some remainders of agar (a) are present and the whole sample is wrapped in parafilm (pa). Colours are green for stele and yellow for phellem tissue. (taken from Verboven et al., 2011). Applying the microscale gas diffusion model to the micro-ct images, the radial O 2 diffusivity of the tissues was calculated for the different samples. The apparent values (~ m 2 s - 1 ) are very low in the stele. Due to the abundant presence of gas spaces in the phellem, the sample across the phellem-stele interface has a much higher O 2 diffusivity (~ m 2 s -1 ) and the highest value (~ m 2 s -1 ) is found in the phellem, closer to that of O 2 in air ( m 2 s -1 ). The diffusivity values in the phellem decrease with distance down the root: at 1.5 cm it is less than half the value at the hypocotyl-root junction.

4 Figure 3. 3-D reconstructions of the roots of Melilotus siculus grown in stagnant solution: (a) (a) 1.5 cm below the hypocotyl-root junction (2.4 µm pixel size).the figure shows the central stele (s) covered by aerenchymatous phellem (ap). Colours are green for stele, yellow for phellem tissue and blue for air. (b) Pore network in the phellem of the sample in (a). The skeleton shows connections of the pores, the colours indicate the radii of the pores (1.65 µm, blue, to 18 µm, red). (c) Pore network in the stele of the sample in (a). The skeleton shows connections of the pores, the colours indicate the radii of the pores (1.22 µm, blue, to 10 µm, red). (d) Pore network across the phellem-stele interface of the sample in (a). The arrow indicates small connections between the pores in the phellem and in the stele (1.22 µm, blue, to µm, red). (taken from Verboven et al., 2011). The diffusion-respiration model was solved for a 3-D root model to obtain radial O 2 concentration profiles through stele and phellem at different positions along the root. The computed radial and axial profiles are compared to measured profiles in Fig. 4. The uniform radial O 2 concentration in the phellem is predicted well, the phellem O 2 concentration decreases with distance from the hypocotyl junction and the gradients in the stele are comparable. The predicted and measured profiles show a maximum of the axial stelar O 2 concentration at the 4 cm position. The peaks in stelar O 2 concentration at 0.5 cm cannot be reproduced by the model because the stele is treated as a homogeneous material with a constant diffusivity, while the peaks are due to discrete pores in the microstructure. Conclusion Micro-computed tomography (micro-ct) was employed to visualise in 3-D the microstructure of the aerenchymatous phellem in roots of Melilotus siculus. Tissue porosity and respiration were also measured for phellem and stelar tissues. A multiscale 3-D diffusion-respiration model compared predicted O 2 profiles in roots, with those measured using O 2 microelectrodes. Micro-CT confirmed the measured high porosity of aerenchymatous phellem (44-54%) and low porosity of stele (2-5%) A network of connected gas spaces existed in the phellem but not within the stele. O 2 partial pressures were high in the phellem but fell below the detection limit in the thicker upper part of the stele, consistent with the

5 poorly-connected low porosity and high respiratory demand. The presented model integrates and validates micro-ct with measured radial O 2 profiles for roots with aerenchymatous phellem, confirming the existence of near anoxic conditions at the centre of a stele in basal parts of the root, coupled with only hypoxic conditions towards the apex. (a) (b) Figure 4. Predicted versus measured radial profiles of po 2 (a) at 0.5, 3.0, 4.0 and 9.0 cm below the root-hypocotyl junction. The longitudinal profiles (b) are plotted on the axis of the stele and in the phellem. Measurements in (b) are averages with standard error bars from all data points in the phellem (n = +/- 50) and at few data points in the center of the stele (n = 5). The solid lines are the result of the model. In (a) the symbols are the measured values at 0.5 cm ( ), 3 cm ( ), 4 cm ( ) and 9 cm( ). In (b) the symbols are the average measured values in the stele center ( ) and in the phellem ( ).(taken from Verboven et al., 2011). References: 1. Herremans, E., Verboven, P., Bongaers, E., Estrade, P., Verlinden, B.E., Wevers, M., Hertog, M.L.A.T.M., Nicolai, B.M. Characterisation of Braeburn browning disorder by means of X-ray micro-ct. Postharvest Biology and Technology 75, , Ho, Q.T., Verboven, P., Verlinden, B.E., Herremans, E., Wevers, M., Carmeliet, J., Nicolaï, B.M. A three-dimensional model multiscale model for gas exchange in fruit. Plant Physiology 155, , Ho, Q.T., Carmeliet, J., Datta, A., Defraeye, T., Delele, M., Herremans, E., Opara, L., Ramon, H., Tijskens, E., van der Sman, R., Van Liedekerke, P., Verboven, P., Nicolaï, B.M. Multiscale modeling in food engineering. Journal of Food Engineering, 114: , Mendoza, F., Verboven, P., Mebatsion, H.K., Kerckhofs, G., Wevers, M., Nicolaï, B. Three-dimensional pore space quantification of apple tissue using X-ray computed microtomography. Planta 226, , Verboven, P., Kerckhofs, G., Mebatsion, H.K., Ho, Q.T., Temst, K., Wevers, M., Cloetens, P., Nicolaï, B.M. 3-D gas exchange pathways in pome fruit characterised by synchrotron X-ray computed tomography. Plant Physiology 147, , Verboven P., Pedersen O., Herremans E., Ho Q.T., Nicolaï B.M. Colmer T.D., Teakle N. Root aeration via aerenchymatous phellem 3-D micro-imaging and radial O 2 profiles in Melilotus siculus. New Phytologist 193, , 2011.