SEARCH

SEARCH BY CITATION

Keywords:

  • aerenchyma;
  • ethylene;
  • hydrogen peroxide (H2O2);
  • MT2b;
  • Oryza sativa (rice);
  • oxygen radical;
  • reactive oxygen species

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Gas spaces (aerenchyma) form as an adaptation to submergence to facilitate gas exchange. In rice (Oryza sativa), aerenchyma develop by cell death and lysis, which are poorly understood at the cellular level.
  • Aerenchyma formation was studied in rice stems by light microscopy. It was analyzed in response to submergence, ethylene and hydrogen peroxide (H2O2) treatment, and in the MT2b::Tos17 mutant. O2· was detected with nitroblue tetrazolium and an epinephrine assay. H2O2 was detected with 3,3′-diaminobenzidine.
  • Aerenchyma develop constitutively in all internodes of the deep-water rice variety Pin Gaew 56, but are absent from the nodes. Constitutive aerenchyma formation was also observed in two lowland rice varieties, albeit to a lesser degree. A larger number of aerenchyma are present in older internodes, and at the top of each internode, revealing developmental gradients. Submergence or treatment with the ethylene-releasing compound ethephon promoted aerenchyma formation in all genotypes analyzed. Pre-aerenchymal cells contain less starch, no chloroplasts, thinner cell walls and produce elevated levels of O2· and H2O2 compared with other parenchymal cells. Ethephon promotes O2· formation and H2O2 promotes aerenchyma formation in a dose-dependent manner. Further-more, genetic downregulation of the H2O2 scavenger MT2b enhances aerenchyma formation.
  • Aerenchyma formation is mediated by reactive oxygen species.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Plants are aerobic organisms and a lack of oxygen presents a severe stress. Low oxygen conditions in plants can be brought about by high metabolic activity, soil waterlogging or flooding. Soil waterlogging and flooding are environmental conditions that are frequently encountered by plants growing in aquatic or semi-aquatic habitats. Plant roots are most often exposed to a lack of oxygen as gas exchange is strongly decreased in waterlogged soils. However, when plant shoots become flooded, they also experience oxygen shortage. The problem of oxygen supply becomes more severe in turbid waters or with deep floods, which cause low light conditions that prevent photosynthetic oxygen production. Plants, such as rice (Oryza sativa), endure frequent flooding by adaptations found not only in roots, but also in stems. One basic strategy to cope with flooding is to improve gas exchange. Gas-filled spaces (aerenchyma) are constitutively found in aquatic and semi-aquatic plants, and are considered to be an efficient mechanism to ameliorate low oxygen stress. Aerenchyma facilitate gas exchange between aerial and submerged plant parts by reducing the diffusion resistance to gas exchange imposed by cells.

Aerenchyma can form by lysigeny or schizogeny. Lysigenous aerenchyma formation is observed in rice or maize roots, where cortex cells undergo programmed cell death. Subsequently, the cell corpses, including the cell walls, are degraded and resorbed, resulting in a gas space (Kawai et al., 1998; Evans, 2003). Aerenchyma form along the length of a plant organ, such as the root. In rice, they are also present in stems and leaves (Arashi & Nitta, 1955; Matsukura et al., 2000; Colmer & Pedersen, 2008). In wetland species, such as Sagittaria lancifolia and rice, which are well adapted to frequent flooding, aerenchyma develop constitutively (Raskin & Kende, 1983; Webb & Armstrong, 1983; Schussler & Longstreth, 1996). However, there is large variation in the development of adaptive traits, even among different rice cultivars (Fukao et al., 2006; Hattori et al., 2009).

Ethylene has been identified as the hormonal signal that mediates aerenchyma formation in wheat, maize, rice and Arabidopsis (Drew et al., 1981; Konings, 1982; Jackson et al., 1985; Justin & Armstrong, 1991; He et al., 1996; Watkin et al., 1998; Gunawardena et al., 2001; Mühlenbock et al., 2007; Shiono et al., 2008), whereas, in Juncus effuses, root aerenchyma form independent of ethylene (Visser & Bögemann, 2006). The diffusion rate of the gaseous hormone ethylene is greatly reduced on submergence, resulting in rapid accumulation in submerged tissues. Although the ethylene signaling pathway has been elucidated in detail, molecular events that result in the induction and execution of programmed cell death during aerenchyma formation are less well understood. In the maize root, Ca2+, protein phosphorylation and G-protein signaling appear to be integral signal components, as reagents that increase cytosolic Ca2+, protein phosphatase inhibitors and constitutive activation of G-protein promote aerenchyma formation at normoxic conditions (He et al., 1996). In rice roots, cell death has been shown to begin with the largest cell in the mid-cortex and to expand radially to neighboring cells (Kawai et al., 1998).

Programmed cell death is a developmental program that is indispensable for plant development, and represents an important strategy of plants for adaptation to biotic and abiotic stresses. The hypersensitive response (HR) describes local cell death in response to pathogen attack and helps to limit the spread of the pathogen (Zurbriggen et al., 2010). Epidermal cell death precedes the emergence of adventitious roots at the nodes of submerged rice plants. As is the case with aerenchymal cell death, epidermal cell death is induced by ethylene (Mergemann & Sauter, 2000). Hydrogen peroxide (H2O2) has been shown recently to act as a signal that is required for ethylene-induced epidermal cell death and is sufficient to promote it (Steffens & Sauter, 2005, 2009). Reactive oxygen species (ROS) have also been recognized as important signaling molecules in the hypersensitive cell death response (Lamb & Dixon, 1997). HR is characterized by an oxidative burst during which high levels of H2O2 are produced.

In this study, we analyzed aerenchyma formation in the rice stem. Based on the hypothesis that parenchymal cell death may be accompanied by ROS production, we analyzed the presence of H2O2 and the superoxide anion radical (O2·) in the rice stem. Furthermore, we pursued the question of whether an increase in H2O2 would be sufficient to induce parenchymal cell death and thus promote aerenchyma formation. In addition to H2O2 treatment, we employed a rice mutant with an insertional mutation in the MT2b gene, which encodes a metallothionein. In rice suspension cells, constitutive genetic downregulation of MT2b transcript levels resulted in the accumulation of H2O2, showing that MT2b acts as a ROS scavenger (Wong et al., 2004). In addition, both hypersensitive cell death and epidermal cell death were enhanced in the MT2b::Tos17 mutant (Wong et al., 2004; Yuan et al., 2008; Steffens & Sauter, 2009). The data provided in this study show that aerenchyma formation is also controlled by H2O2, indicating that ROS play a key role in the regulation of diverse cell death processes in rice.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Plant material and growth conditions

Seeds of Oryza sativa L., deep-water rice cv Pin Gaew 56 (PG56; indica rice) were cultivated. Stem sections were prepared from 11–14-wk-old plants from 20 mm below the basal node of the first, second or third internode with a total length of 200 mm, as described previously (Sauter, 1997). Seeds of the lowland rice cultivars Kinmaze (japonica rice), Nipponbare (japonica rice) and the MT2b::Tos17 line in the Nipponbare background were imbibed in 50 μM S-nitroso-N-acetylpenicillamine (SNAP; Molecular Probes, Eugene, Oregon, USA), a nitric oxide (NO)-releasing compound, to increase germination rates. Stem sections were excised from 18–20-wk-old plants. For the treatment with H2O2, stem sections were excised 5 mm below the lower node and had a total length of 185 mm. Up to 15 stem sections were placed in a 150-ml beaker containing 20 ml of an aqueous solution of 150 μM ethephon (2-chloroethanephosphoric acid; Sigma-Aldrich, Steinheim, Germany) or 50 ml of H2O2 (Roth, Karlsruhe, Germany) at the concentrations indicated. Plastic cylinders covered the beakers to ensure high humidity. For submergence treatment, plants were partially submerged in a 600-l plastic tank filled with tap water with c. 300 mm of the leaf tips remaining above the surface. Plants were grown and stem sections were incubated in a growth chamber at 27°C at 70% relative humidity and 16 h light at 150 μmol m2 s−1.

Analysis of aerenchyma in the stem

For microscopic analyses, internodal sections, 5 mm in length, were excised from the second youngest internode of cv PG56, cv Nipponbare and MT2b::Tos17. Sections of cv PG56 were embedded in Technovit 7100, according to the manufacturer’s instructions (Heraeus Kulzer GmbH, Wehrheim, Germany), and cross-sections, 25 μm thick, were cut. The cv Nipponbare and MT2b::Tos17 were embedded in LRwhite; cross-sections, 1 μm thick, were cut with a Leica RM 2255 microtome (Leica, Bensheim, Germany) and observed under bright-field illumination using an Olympus BX41 microscope (Olympus, Hamburg, Germany). For transmission electron microscopy (TEM), 10-mm-long internodal sections from internodes of cv Nipponbare were fixed, dehydrated and embedded in LRwhite. Cross-sections, 60–80 nm thick, were cut with a Leica Ultracut UCT, prepared for TEM analysis and analyzed with a Philips 208 S transmission electron microscope.

Aerenchyma

Early- and late-stage aerenchyma were counted in cross-sections obtained at 10, 20, 30 or 40 mm above the second, third or fourth node as indicated. Aerenchyma formation was expressed as a percentage of the maximal number of aerenchyma that can be formed in the stem of a given genotype, as indicated in Table 1.

Table 1.   Maximal number of aerenchyma that can develop in the stem of deep-water rice cv PG56 and lowland rice cultivars Kinmaze and Nipponbare [wild-type (wt) and MT2b::Tos17 line]
Number of aerenchyma
cv PG56cv Kinmazecv Nipponbare
  1. Results are averages (± SE) from 14–37 stem sections analyzed. Numbers are not significantly different (< 0.001; ANOVA, Tukey test).

Wild-typeWild-typeWild-typeMT2b::Tos17
28.5 (± 0.3)27.6 (± 0.4)28.6 (± 0.3)27.8 (± 0.3)

Staining methods and superoxide anion radical assay

For the staining of dead cells, 1-mm-long internodal sections were excised at the positions indicated, stained with 2% (w/v) Evans blue for 10 s and washed three times with water (Mergemann & Sauter, 2000). To visualize starch, internodal sections were stained with Lugol’s solution (1 : 2 I/KI) for 60 s and washed three times with water. To detect superoxide anion radicals, 1 mM nitroblue tetrazolium (NBT; Sigma-Aldrich) was applied in 10 mM potassium phosphate buffer, pH 7.8 (Hückelhoven et al., 2000). To detect H2O2, the 3,3′-diaminobenzidine (DAB) Enhanced Liquid Substrate System was used according to the manufacturer’s instructions (Sigma-Aldrich). To detect O2· and H2O2in situ, 6-mm-long internodal cross-sections were vacuum infiltrated with either NBT or DAB solution for 15 min. NBT staining was visualized 90 min, and DAB staining was visualized 180 min, after vacuum infiltration. In all cases, staining was observed with a binocular (Olympus).

For the quantitative analysis of O2·, 2-mm-thick tissue sections were excised 20 mm above the third node after the treatment of stem sections with 150 μM ethephon for the times indicated. Three slices each were weighed, and ground with a pestle in 500 μl of 50 mM Tris-HCl at pH 7.8. The extract was centrifuged for 10 min at 8000 g; 300 μl of the supernatant were transferred to 1 ml of 1 mM epinephrine and 1 mM KCN in 50 mM Tris-HCl. O2· evolution was assayed spectrophotometrically (Sagi & Fluhr, 2001) by measuring the change in absorbance at 480 nm after 1 h, which occurs as a result of the oxidation of epinephrine to adrenochrome by O2·.

Statistical analysis

Statistical analyses were performed with Minitab (Minitab Inc., State College, Pennsylvania, PA, USA). Percentages were transformed with arcsine√(x/100) to obtain normally distributed data. Comparison of means was analyzed for statistical significance with an analysis of variance (ANOVA) and Tukey test, or with a two-sample t-test. Constant variance and normal distribution of the data were verified before statistical analysis, and the P value was set to P < 0.001 if one of both conditions was not achieved. The P value for the Pearson product moment correlation is indicated in the figure and table legends.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Parenchyma cells that form aerenchyma in the stem are morphologically distinct

In order to characterize the parenchymal cells that form aerenchyma in the rice stem, light micrographs were taken from cross-sections of the second youngest internode of deep-water rice cv PG56 (Fig. 1a–f). Morphological studies were carried out on cross-sections from untreated plants. Cells that formed aerenchyma appeared to be distinct from other parenchymal cells in that they were lighter than the green parenchymal cells that surrounded them (Fig. 1a). These lighter colored cells that have the potential to develop into aerenchyma were hence termed pre-aerenchymal cells. A cross-section displayed, on average, 28.5 pre-aerenchymal areas that were located in between the outer ring of vascular bundles (Fig. 1a–f, Table 1). Three to five parenchymal cells were present between the (pre-)aerenchyma and epidermis. The average number of (pre-)aerenchymal areas was 27.6 in cv Kinmaze and 28.6 in cv Nipponbare, and was thus a conserved feature in deep-water and lowland rice (Fig. 2a, Table 1). Staining with Lugol’s solution revealed that pre-aerenchymal cells contained less starch than the surrounding parenchymal cells, which were stained dark (Fig. 1b). Cell death, as visualized with Evans blue staining, was observed not only in pre-aerenchymal cells, but also in the cell layers proximal to the lacunae and in vascular bundles (Fig. 1c). Analysis of cross-sections of cv Nipponbare by TEM revealed that pre-aerenchymal cells had a thinner cell wall and were larger than surrounding parenchymal cells (Fig. 2b,c). Furthermore, pre-aerenchymal cells had a large vacuole and little cytoplasm that appeared to be devoid of chloroplasts (Fig. 2b,c).

image

Figure 1. Submergence and ethylene promote aerenchyma formation in internodes of rice. Cross-sections were cut 20 mm above the third node of deep-water rice cv PG56 plants. White or black dashed lines in (a–f) surround areas (pre-aerenchyma) that will develop into aerenchyma, and the cell layer that will not be lysed but has similar features to pre-aerenchymal cells. (a) Pre-aerenchymal cells are lighter in colour than the surrounding green parenchymal cells. (b) A cross-section was stained for starch with Lugol’s solution. The dark staining indicates that starch accumulates mainly in parenchymal cells surrounding (pre-)aerenchyma. Cells that later on are lysed to form aerenchyma contain little starch. (c) A cross-section was stained with Evans blue to visualize dead cells. Staining is obvious in the vasculature in pre-aerenchymal cells and in the parenchymal cell layers proximal to the lacunae. (d–f) Three stages of aerenchyma formation were distinguished. (d) In pre-aerenchyma, all cells are still present. (e) In early-stage aerenchyma, one or a few cells have disappeared; the aerenchyma is indicated by an asterisk. (f) In late-stage aerenchyma, one layer of light-colored, chloroplast-free parenchymal cells occasionally persists. (g) Plants were partially submerged (gray bars) with the third node c. 60 cm beneath the water surface, or kept in air as a control (white bars). Aerenchyma are expressed as a percentage of the maximal number of aerenchyma that can be formed in the stem (see Table 1). The number of aerenchyma formed at 20 mm above the third node was enhanced significantly after 3 d of partial submergence (< 0.001; ANOVA, Tukey test). Results are the means (± SE) from 27–54 stems analyzed per treatment in three biological experiments. (h) Stem sections containing the second internode were treated with 150 μM ethephon (gray bars), an ethylene-releasing compound, or without effector as a control (white bars). Results are the means (± SE) from 40–48 stems analyzed per treatment in three biological experiments. Aerenchyma formation at 20 mm above the third node was enhanced significantly by ethephon after 2 d. Different letters indicate statistically significant differences (< 0.001; ANOVA, Tukey test). (i) Dose-dependent aerenchyma formation in response to ethephon. Stem sections were treated for 2 d and aerenchyma were analyzed at 10 mm above the third node. The asterisk indicates a statistically significant difference between the control and 150 μM ethephon treatment (< 0.05, two-sample t-test).

Download figure to PowerPoint

image

Figure 2. Pre-aerenchymal cells are distinct from their surrounding parenchymal cells. (a) Light micrograph of a cross-section from the second youngest internode, 20 mm above the third node, of lowland rice cv Nipponbare plants. Pre-aerenchyma (surrounded by a dashed line) are located between vascular bundles (vb). (b, c) Transmission electron micrographs from the same region as described in (a). Pre-aerenchymal cells have thinner cell walls (thin arrow) than surrounding parenchymal cells (thick arrow), are highly vacuolated and have no chloroplasts.

Download figure to PowerPoint

Aerenchyma formation in the rice stem is enhanced by submergence and by ethylene

Aerenchyma form from seemingly intact cells (Fig. 1d, pre-aerenchyma) that may, however, already have lost their membrane integrity, resulting in Evans blue staining as observed in Fig. 1(c). Cell lysis begins at the center of the pre-aerenchymal area (early-stage aerenchyma) and expands radially, such that, later on, one layer of light-shaded, highly vacuolated cells remains that eventually also disappears (late-stage aerenchyma; Fig. 1e,f).

Deep-water rice plants of cv PG56 were partially submerged for up to 4 d and the percentage of aerenchyma was determined at a distance of 20 mm above the third node after 1, 2, 3 and 4 d (Fig. 1g). The percentage of aerenchyma formed was determined from the number of early- and late-stage aerenchyma in relation to the total number of (pre-)aerenchymal areas. As a control, plants were kept in air. Approximately 70% of constitutive aerenchyma formation was observed in control plants. On submergence, the percentage increased significantly within 3 d, reaching close to 100%. Subsequently, stem sections, including the second youngest internode, were excised and treated with or without 150 μM ethephon for up to 4 d and aerenchmya formation was again analyzed 20 mm above the third node. The percentage of aerenchyma increased from 64.6% in controls to 89.7% after 2 d, and further increased to 94.8% after 3 d and to nearly 100% after 4 d of ethephon treatment (Fig. 1h). Although, in untreated stems, the majority of aerenchyma were in the early stage (Table 2), ethylene treatment not only increased significantly the total number of aerenchyma, but also the percentage of late-stage aerenchyma (< 0.001, ANOVA, Tukey test). The lowland rice cultivars Nipponbare (Fig. 5) and Kinmaze (Supporting Information Fig. S1) showed fewer constitutively formed aerenchyma when compared with the deep-water rice variety. However, the induction of aerenchyma formation was observed in all three genotypes.

Table 2.   Percentages of early- and late-stage aerenchyma formed after submergence or treatment with 150 μM ethephon (+ 150 μM E) in the second internode of deep-water rice cv PG56, as summarized in Fig. 1g,h
DaysControlSubmergedControl+ 150 μM E
Aerenchyma (%)
EarlyLateEarlyLateEarlyLateEarlyLate
  1. After 2, 3 and 4 d of ethephon treatment, the percentages of late aerenchyma are significantly different (< 0.001, ANOVA, Tukey test).

044.822.7  38.521.8  
142.832.347.125.834.322.135.839.9
234.541.147.227.835.329.324.765.1
337.530.026.769.635.135.225.169.8
436.825.721.673.032.242.85.491.8

A dose-dependent response to ethylene was studied in deep-water rice cv PG56 at 10 mm above the third node (Fig. 1i). Ethephon concentrations up to 15 μM were ineffective in promoting aerenchyma formation within 2 d. Constitutively, c. 30% of aerenchyma were formed at this distance. In the presence of 50 or 150 μM ethephon, aerenchyma formation was c. 60%, reminiscent of a threshold response. These results not only confirmed that aerenchyma were induced by ethylene in the rice stem, but revealed that fewer aerenchyma were constitutively present at 10 mm above the third node when compared with 20 mm above the third node.

To follow up on this observation, we next analyzed how many aerenchyma were present constitutively in the first, second and third internodes of cv PG56 plants. In addition, aerenchyma formation was determined after plants had been submerged for 3 d. Aerenchyma were counted 10, 20, 30 and 40 mm above each node (Fig. 3a). Unlike older inter-nodes, the youngest internode possesses an intercalary meristem that extends 5 mm above the second node, from which cells are released into the elongation zone, which extends up to 15 mm above the node (Sauter & Kende, 1992). After 3 d of submergence, the elongation zone extends up to 35 mm above the node (Kende et al., 1998). No aerenchyma were observed in the elongation zone of the youngest internode, 10 mm above the second node. Aerenchyma were observed within the differentiation zone of the youngest internode, reaching 75.3% at 40 mm above the node. The second internode showed slightly more aerenchyma, especially 20 mm above the third node. The third internode showed, overall, more constitutively formed aerenchyma, but a general pattern was observed in all internodes: few aerenchyma existed at the base of each internode; an abrupt increase in aerenchyma numbers occurred between 10 and 20 mm above each node; and a gradual increase was observed further up, reaching 90% in the oldest internode. No aerenchyma were present at the nodes (Fig. 3b). Partial submergence resulted in increased aerenchyma formation throughout each internode, indicating that all tissues were responsive to the cell death and cell lysis signal(s).

image

Figure 3. Developmental regulation of aerenchyma formation. (a) Deep-water rice cv PG56 plants were partially submerged for 3 d (3 d subm.) with c. 30 cm of the leaf tips remaining above the water surface. The percentage of aerenchyma formed was determined at 10, 20, 30 and 40 mm above the second, third and fourth nodes. Each internode was characterized by few aerenchyma at the base and increasing numbers towards the top. On submergence, aerenchyma numbers increased along the length of each internode; however, a gradient from base to top persisted. Per treatment, 12–15 plants were analyzed in two independent experiments. (b) Cross-sections of the second, third and fourth nodes show a spongy tissue in the center and no gas spaces in the parenchyma, indicating that neither the central pith cavity nor the aerenchyma traverse the nodes.

Download figure to PowerPoint

ROS accumulate in pre-aerenchymal cells and H2O2 promotes aerenchyma formation

To analyze the involvement of ROS in aerenchyma formation, we first determined whether the ROS species H2O2 and O2· were present in pre-aerenchymal cells. H2O2 was visualized with DAB and was detected as a brown precipitate in cross-sections 20 mm above the third node (Fig. 4b). H2O2 appeared to be particularly abundant in the vasculature and was also detected in pre-aerenchymal cells. O2· was detected with NBT, resulting in a blue precipitate (Fig. 4c). O2· was present in pre-aerenchymal cells and in the cell layer surrounding aerenchyma. Not all pre-aerenchyma cell areas showed elevated H2O2 and O2· levels. We subsequently studied the effect of exogenously applied H2O2 on aerenchyma formation. Stem sections were treated with H2O2 at concentrations between 0.01% (v/v) and 3% (v/v) for 3 d, and aerenchyma were counted 10 and 20 mm above the third node (Fig. 4d,e). Aerenchyma formation was promoted by H2O2 in a dose-dependent manner (Fig. 4d,e). At 10 mm, aerenchyma formation was enhanced significantly with 1% (v/v) H2O2. At 20 mm, a significant increase was observed with 0.3% (v/v) H2O2. At this distance, aerenchyma numbers are constitutively high at 75.9%, but were increased by H2O2 up to 96.8%. A time course experiment was carried out with the application of 3% (v/v) H2O2 for up to 3 d. The percentage of aerenchyma increased within 2 d at both 10 mm and 20 mm above the third node (Fig. 4d,e), revealing similar kinetics to those observed with ethephon.

image

Figure 4. Hydrogen peroxide (H2O2) promotes aerenchyma formation. Cross-sections were obtained from 20 mm above the third node of cv PG56 stem segments that were unstained (a), stained brown with 3,3′-diaminobenzidine (DAB) to visualize H2O2 (b) or stained blue with nitroblue tetrazolium (NBT) to visualize O2· (c). (d, e) Left-hand panels: stem sections were treated with H2O2 at concentrations between 0.01% and 3% (v/v) for 3 d. Aerenchyma formation was determined at 10 mm (d) and 20 mm (e) above the third node. Results are averages (± SE) from 13–16 stem sections analyzed per treatment in two experiments. Different letters indicate statistically significant differences (< 0.001; ANOVA, Tukey test). (d, e) Right-hand panels: stem sections were treated with 3% (v/v) H2O2 for up to 3 d (gray bars) and the percentages of aerenchyma were determined at 10 mm (d) and 20 mm (e) above the third node (control, white bars.) Results are averages (± SE) from 31–36 stems analyzed per treatment in three experiments. Different letters indicate statistically significant differences (< 0.001; ANOVA, Tukey test). (f) Stem sections were treated with or without 150 μM ethephon (E) for 2, 4, 8, 16 or 24 h, and the generation of O2· was determined at each time point in controls (circles) and treated (bars) samples. Results are means (± SE) from 12–16 measurements. The asterisk at 16 h indicates a statistically significant difference between control and ethephon treatment (< 0.05, two-sample t-test).

Download figure to PowerPoint

To analyze whether ethylene promotes the formation of ROS, stem sections were incubated with or without 150 μM ethephon for up to 24 h, and O2· generation was measured using an assay with epinephrine as a substrate, which is oxidized to adrenochrome in the presence of O2· (Fig. 4f). O2· generation in untreated stems remained nearly constant. Treatment with ethephon resulted in a significant increase in O2· generation after 16 h. A similar result was obtained with the lowland rice cv Nipponbare, with a significant increase in O2· generation after 16 h of ethephon treatment (data not shown). The peak in O2· generation preceded aerenchyma formation induced by ethephon (Fig. 1h).

Downregulation of MT2b enhances aerenchyma formation by submergence

Downregulation of the ROS scavenger gene MT2b has been shown previously to promote the death of epidermal cells above adventitious roots (Steffens & Sauter, 2009). To analyze whether MT2b was involved in aerenchymal cell death, aerenchmya formation was determined in the homozygous MT2b::Tos17 insertion line. The MT2b::Tos17 mutant was generated in the Nipponbare background; hence, wild-type (wt) cv Nipponbare was analyzed as a control. wt and MT2b::Tos17 plants were partially submerged for up to 3 d and the percentages of aerenchyma formed were determined 20 mm above the third node (Fig. 5). The average numbers of aerenchyma per cross-section did not differ between wt cv Nipponbare and MT2b::Tos17 (Table 1). Approximately 20% of aerenchyma developed constitutively in both lines, with wt having seemingly fewer constitutive aerenchyma than MT2b::Tos17 (Fig. 5). In wt, the percentage increased to 39.5% after 2 d of submergence and to 48.5% after 3 d of submergence, indicating that aerenchyma formation in cv Nipponbare was induced by submergence with a lag phase longer than 1 d. In MT2b::Tos17, 55.7% aerenchyma developed after 2 d and 66.9% after 3 d of submergence. After 3 d, the differences between wt and mutant were statistically significant (two-sample t-test, < 0.001).

image

Figure 5. Downregulated MT2b expression promotes aerenchyma formation during submergence. MT2b::Tos17 and wild-type (wt) plants of the lowland rice cv Nipponbare were partially submerged (gray bars) or kept in air as a control (white bars). Aerenchyma formation was enhanced significantly in both lines after 2 d of submergence at a distance of 20 mm above the third node (< 0.001, ANOVA, Tukey test). Different letters indicate statistically significant differences. After 3 d of submergence, significantly more aerenchyma had formed in MT2b::Tos17 relative to wt, as indicated by an asterisk (two-sample t-test, < 0.001). Results are means (± SE) from 17–58 stems analyzed per genotype and treatment in two experiments.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Improved gas exchange is considered to be a hallmark of flooding tolerance, and is observed in essentially all wetland species. Improved gas exchange is achieved externally through leaf gas films that transport oxygen from leaves which remain above the water surface to submerged plant parts (Raskin & Kende, 1985), or that enlarge the gas–water interface in completely submerged plants (Pedersen et al., 2009). Improved gas exchange is further achieved through internal air spaces. As soil waterlogging or flooding affects primarily the roots, much research has focused on the study of aerenchyma formation in roots (e.g. Raskin & Kende, 1983; Webb & Armstrong, 1983; Schussler & Longstreth, 1996; Colmer, 2003). However, aerenchyma formation also occurs in stems and other shoot organs, where it may be especially important during deep floods when shoots become partially submerged, such as in rice (Arashi & Nitta, 1955). In this study, developmental and submergence-induced aerenchyma formation was analyzed in the rice stem. The deep-water rice variety PG56 showed a higher percentage of aerenchyma formed constitutively than the lowland rice varieties Kinmaze and Nipponbare, with the total number of aerenchymal tracts being the same in the two varieties. In the rice stem, aerenchyma were formed in between the vascular bundles which were located in an outer ring below the epidermis. The conserved number of aerenchyma may be a result of the conserved anatomy of rice with a defined number of vascular bundles in the stem.

The number of aerenchyma formed increased from bottom to top in a recurrent pattern that was observed in each internode. Nodes did not display gas spaces. Aerenchyma are constitutively formed in many aquatic or semi-aquatic plant species. On flooding, enhanced aerenchyma formation is observed in (semi-)aquatic, but also in nonaquatic species, as shown for maize (Drew et al., 1989; He et al., 1994), Spartina patens (Burdick, 1989), Luffa cylindrica (Shimamura et al., 2007) and Rumex palustris (Laan et al., 1991; Pierik et al., 2009), to name only a few. However, not all plants can form aerenchyma, as reported for Brassica napus (Voesenek et al., 1999). Consequently, the porosity value in aerated rapeseed roots is 3%, whereas rice roots show a constitutive porosity of 25–30% (Armstrong, 1971). During submergence, the porosity value remains constant in rapeseed and increases to 40–50% in rice roots. In the rice stem, partial submergence for 3 d resulted in an increase in aerenchyma formation throughout all internodes, indicating that all pre-aerenchymal cells were responsive to the cell death and cell lysis promoting signal. Despite the overall increase in aerenchyma numbers following submergence, a gradient remained in each internode, with fewer aerenchyma at the base than at the top. For instance, twice as many aerenchyma were counted at distances of 30 and 40 mm than at 10 mm above the third node. This gradient within the stem is exacerbated by the fact that nodes do not form any aerenchyma, indicating that neither aerenchyma nor the central pith cavity form a continuous gas space even when submerged. It is, however, conceivable that gas flow across the node occurs through intercellular spaces. Gas flow measurements, such as were performed in Phragmites australis (Afreen et al., 2007), will be required to determine the actual resistance to gas diffusion in the internodes and at the stem nodes of rice. It has been proposed that aerenchyma not only improve gas exchange, but also reduce the number of cells requiring oxygen for respiration (Kawai et al., 1998). In maize roots, aerenchyma form even under drought conditions (Zhu et al., 2010). Recombinant inbred maize lines with high root cortical aerenchyma produced higher shoot biomass and more seed when grown under water-stressed (drought) conditions than did lines which formed very few aerenchyma. The improved drought tolerance in lines with high aerenchyma formation has been proposed to be a result of reduced metabolic cost in the roots caused by reduced cell numbers (Zhu et al., 2010).

A more detailed analysis of the cells that are lysed to form a gas space in the rice stem revealed that these cells possess characteristic features. They are larger, as also observed for cortex cells in rice roots that form aerenchyma (Kawai et al., 1998). They contain a large vacuole, are lighter in color and appear to be devoid of chloroplasts. Unlike the surrounding parenchymal cells, pre-aerenchymal cells contain little or no starch, and have a thin cell wall. By contrast, the walls in cells that surround aerenchyma are thicker, an observation that was also made in other (semi-)aquatic species, such as Astrocaryum jauari (Schlüter et al., 1993). Thick cell walls around an aerenchyma are thought to prevent its collapse, as the hydrostatic pressure increases with water depth. In rice internodes, pre-aerenchymal cells were occasionally stained with Evans blue, indicating that they had undergone cell death, but were not yet lysed. Overall, cells that eventually form gas spaces in the rice stem do not appear to participate greatly in photosynthetic activity, or contribute to tissue stability, raising the question of why they are maintained to begin with.

Aerenchyma formation in response to submergence has been shown previously to be regulated by ethylene in wheat (Watkin et al., 1998) and maize (Drew et al., 1981; Konings, 1982; Jackson et al., 1985; He et al., 1996; Gunawardena et al., 2001) roots. In rice, ethylene can promote aerenchyma formation in adventitious roots (Justin & Armstrong, 1991), and in internodes, as shown in this study. To date, little else is known about cell death regulation and lysis of cells during gas space formation. ROS have been recognized as central components of cell death signaling (Kotchoni & Gachomo, 2006). Cadmium-induced death of suspension cells of tomato was enhanced by ethylene and was accompanied by elevated levels of H2O2 (Yakimova et al., 2006). In rice, death of epidermal cells above nodal adventitious root primordia was induced by ethylene and H2O2 (Steffens & Sauter, 2009). Pre-aerenchymal cells in the rice internode were characterized by elevated levels of superoxide anion radical and H2O2 when compared with surrounding cells. Induction of aerenchyma formation by ethylene was preceded by an increase in O2·, reminiscent of an oxidative burst, at c. 16 h that preceded ethylene-induced aerenchyma formation. In the mid-cortex of maize roots, aerenchyma formation induced by sulfate starvation was similarly preceded by O2· and H2O2 generation that was observed in some of these cells (Bouranis et al., 2003). Epidermal cell death in rice was also promoted when endogenous levels of H2O2 were increased by the inhibition of catalase activity. Similarly, exogenous application of H2O2 promoted epidermal cell death, even in the presence of 1-methylcyclopropene, which inhibits ethylene perception, indicating that H2O2 acts downstream of ethylene.

Steady-state levels of ROS result from a balance between synthesis and scavenging. ROS detoxification is achieved enzymatically, for example by ascorbate peroxidase, superoxide dismutase or catalase, or nonenzymatically (Jwa et al., 2006). In rice, the metallothionein MT2b acts as a scavenger of H2O2 (Wong et al., 2004). Genetic downregulation of MT2b in rice promoted epidermal cell death and hypersensitive cell death in leaves (Wong et al., 2004; Steffens & Sauter, 2009), and enhanced submergence-induced aerenchyma formation in the stem. These findings support the view that altered ROS scavenging is a highly efficient means to alter ROS levels, such that a physiological cell death response follows. In epidermal cells above adventitious root primordia, MT2b was shown to be downregulated in response to ethylene or H2O2, indicating that reduced ROS scavenging through altered MT2b activity may be a general mechanism of cell death regulation in rice.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Akio Miyao (Genome Research Centre, Ibaraki, Japan) for kindly supplying the MT2b::Tos17 insertion line NE7013. Support of this work by the Deutsche Forschungsgemeinschaft is gratefully acknowledged.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1 Submergence and ethylene promote aerenchyma formation in internodes of lowland rice cv Kinmaze.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

FilenameFormatSizeDescription
NPH_3496_sm_FigS1.ppt115KSupporting info item