Anatomical and Trait Analyses Reveal a Silicon-Carbon Trade-Off in the Epidermis of Sedges

Abstract

In recent years, the detection of numerous negative correlations between silicon (Si) and carbon (C)-based compounds in plants has suggested trade-offs between different stress resistance and/or mechanical support strategies. However, nearly all studies have involved whole-leaf analysis, and it is unclear how the trade-off operates mechanistically, at the cellular level. Here we combined leaf trait measurements and microscopic analyses (electron microscopy with elemental X-ray mapping and X-ray microtomography) of 17 species from a high-Si family: Cyperaceae. Accumulation of Si was strongly negatively correlated with C-based compounds, particularly tannins. Our microscopical investigations showed that the accumulation of phenolics and deposition of silica were mutually exclusive in the outer epidermal cell walls. This trade-off was independent of that between the construction of tough, sclerenchyma-rich leaves and growth potential (the leaf economics spectrum). We also identified a strong negative correlation between Si and accumulation of epicuticular waxes. Previous whole leaf analyses were, in effect, hiding the locations of the trade-off between Si and C-based compounds in plants. The epidermal location of this trade-off and the specific involvement of tannins and waxes suggest the existence of different strategies to resist environmental stresses. Our study provides key insights into plant Si utilization and highlights the multidimensionality of plant stress resistance strategies.

Keywords: epicuticular waxes; leaf anatomy; leaf economics spectrum; plant defense strategies; secondary metabolites; sedges; silicification; trade‐offs.

Figure 1

Leaf silicon (Si) concentrations in 17 sedges. The central horizontal bar in each box shows the median, the box represents the interquartile range (IQR) and the whiskers show the location of the most extreme data points that are still within a factor of 1.5 of the upper or lower quartiles. Black dots show outliers. Data are ranked by increasing species‐mean leaf Si concentration (red dots). Results of ANOVA conducted among all species are given.

Figure 2

Relationships between leaf concentrations of silicon (Si) and carbon‐based compounds. (a) Fourier‐transform infrared (FTIR) spectra from 3000 to 2750 cm⁻¹, colored by leaf Si concentrations. Leaf Si concentrations were log‐transformed to improve visualization. Species‐mean spectra are shown (individual‐level spectra are shown in Supporting Information: Figure S5). Vertical dotted lines show peaks at 2850 and 2920 cm⁻¹, which are typical of epicuticular waxes (Khambatta et al. 2021). (b–f) Relationships between leaf Si concentrations and the area below the 2920 cm⁻¹ peak from panel a (indicating wax compounds), tannins, leaf C concentrations, phenolics and lignin obtained by FTIR. Dots and error bars indicate species means ± standard errors. Statistics are given, as well as slopes (solid lines) when models were significant (panels b–d). The equation was given for the Si‐C relationship, to highlight that the slope was close to −1. Yellow dots indicate Mesomelaena tetragona and Chorizandra enodis, for which detailed microscopic analyses were conducted. All the statistics and slopes that are presented are using all the species. The Si‐tannins relationship (panel c) was, however, also tested without Schoenus rigens because it does not fit well with the relationship (labeled in panel c) and results are presented in the text and Supporting Information: Figure S6 (R² = 0.54 and p < 0.01). The Si‐phenolics relationships (panel e) were tested without Schoenus rigens and Cyathochaeta equitans for similar reasons (labeled in panel e), and the result is presented in the text (R² = 0.30 and p = 0.03).

Figure 3

Principal component analysis (PCA) showing the trait space of the studied sedges. PCA is based on a matrix with 15 traits for the 17 studied species in a, and cross‐sections of four selected species viewed under fluorescent light in b. In a, the color gradient indicates regions of highest (red) to lowest (beige) occurrence of probability of species in the traits space, and contour lines show 0.25, 0.50, and 0.99 quantiles. Black dots indicate species position in the functional space. The two yellow dots show the two species with high and low scores on PC2 (Mesomelaena tetragona and Chorizandra enodis, respectively), for which detailed microscopic analyses were conducted. PCA was run on log‐transformed data and results are presented in Supporting Information: Tables S2 and S3. LDMC is leaf dry matter content; LMA is leaf mass per area; FTIR is Fourier‐transformed infrared spectroscopy. The four gray dots in the PCA represent two species with high scores on PC1 and two species with low scores on PC1 for which cross‐sections were analyzed in panel b (panels a and b are linked with Roman numbers). In b, Lepidosperma sp. C. Tauss 2598 top right; Schoenus subfascicularis bottom right; Lepidosperma rostratum top left; Schoenus rigens bottom left. In these images, structural tissues appear blue while photosynthetically‐active mesophyll cells appear pink. All scale bars: 200 µm.

Figure 4

Cross‐sections of Chorizandra enodis (a) and Mesomelaena tetragona (b) viewed under bright‐field light with or without FeCl₃ staining, at two magnifications. Images on the right were acquired with an FeCl₃ histochemical test, while images on the left were without. The dark‐brown color after FeCl₃ staining indicates the presence of phenolics and tannins (Cavalcanti et al. 2014) that were mostly detected in epidermal regions (red arrows). The much more pronounced change in color for C. enodis than for M. tetragona, after staining, indicates higher tannin/phenolic concentrations, in line with organ‐level concentrations (Figure 2c). Scale bars for low‐magnification images are 200 µm (first row for each species), and those for high‐magnification images are 50 µm (second row for each species).

Figure 5

X‐ray µCT images highlighting tissue density in Chorizandra enodis and Mesomelaena tetragona. A color lookup table was applied to differentiate low‐density/background (black) through to higher‐density tissues and mineral particles (blue to white to red). In a and b, 50 transverse µCT slices (totaling 190 µm thickness) are shown as an average projection image for one representative replicate of C. enodis in a, and M. tetragona in b. Similar slices for the two other replicates are shown in Supporting Information: Figure S15. Air spaces appear in black (e.g., in the mesophyll region of both species, as indicated with green arrows, and in the middle of the culm for C. enodis). Water‐filled parenchyma cells (orange stars) appear in dark blue, while sclerenchymatic tissues appear in lighter blue (orange arrows), indicating higher density. Red zones indicate areas of much higher density, such as mineral deposits. The 3‐D volume in c shows a series of thresholds to highlight the silica deposits in a 150‐slice region (570 µm) of one individual M. tetragona culm (the six images show only the red areas from panel b). The top left image shows only the densest zones of the culm, to which mineral layers of lower density were gradually added (from I to VI; see M&M). All scale bars = 250 µm.

Figure 6

Cell‐specific silicon (Si) and carbon (C) concentrations in Chorizandra enodis and Mesomelaena tetragona acquired using CryoSEM‐EDX. In a (C. enodis) and b (M. tetragona), from left to right, culm anatomical schematics, combined Si‐C maps showing the location of Si (green) and C (purple), and quantitative Si heatmaps for two selected maps by species. For heatmaps, note the difference in [Si] scales between the two species. Five regions of interest were analyzed for Si and C on 10 maps from three replicates for C. enodis and on nine maps from two replicates for M. tetragona (see Materials and Methods for details): (1) the lumen of epidermal cells (epidermal CL), (2) the cell wall between epidermal cells and the cuticle (epidermal CW), (3) mesophyll cells, (4) parenchyma cells, and (5) sclerenchyma cells. Results are shown as boxplots in panels c and d. The central horizontal bar in each box shows the median, the box represents the interquartile range (IQR) and the whiskers show the location of the most extreme data points that are still within a range of 1.5 of the upper or lower quartiles. The y‐scale for Si has been root square‐transformed to improve visualization. ANOVAs were conducted to test the difference in Si and C concentrations in the epidermal CW between species and the results were highly significant (***). All scale bars: 100 µm.