Cellular basis of growth suppression by submergence in azuki bean epicotyls

Abstract

Background and Aims

Complete submergence severely reduces growth rate and productivity of terrestrial plants, but much remains to be elucidated regarding the mechanisms involved. The aim of this study was to clarify the cellular basis of growth suppression by submergence in stems.

Methods

The effects of submergence on the viscoelastic extensibility of the cell wall and the cellular osmotic concentration were studied in azuki bean epicotyls. Modifications by submergence to chemical properties of the cell wall; levels of osmotic solutes and their translocation from the seed to epicotyls; and apoplastic pH and levels of ATP and ethanol were also examined. These cellular events underwater were compared in etiolated and in light-grown seedlings.

Key Results

Under submergence, the osmotic concentration of the cell sap was substantially decreased via decreased concentrations of organic compounds including sugars and amino acids. In contrast, the viscoelastic extensibility of the cell wall was kept high. Submergence also decreased ATP and increased the pH of the apoplastic solution. Alcoholic fermentation was stimulated underwater, but the resulting accumulated ethanol was not directly involved in growth suppression. Light partially relieved the inhibitory effects of submergence on growth, osmoregulation and sugar translocation.

Conclusions

A decrease in the levels of osmotic solutes is a main cause of underwater growth suppression in azuki bean epicotyls. This may be brought about by suppression of solute uptake via breakdown of the H⁺ gradient across the plasma membrane due to a decrease in ATP. The involvement of cell wall properties in underwater growth suppression remains to be fully elucidated.

INTRODUCTION

Flooding is one of the major abiotic stresses that influence growth and productivity of terrestrial plants. Particularly severe injury is brought about by complete and sustained submergence. The physiological causes of the injury of plants by submergence have been extensively studied (Jackson and Ram, 2003), and many recent studies focus on the physiological, biochemical and molecular basis for submergence tolerance in particular species or lines (Greenway and Gibbs, 2003; Jackson and Colmer, 2005; Voesenek et al., 2006; Bailey- Serres and Voesenek, 2008). However, much still remains to be clarified at the cellular level regarding the mechanism of growth suppression by submergence in terrestrial plants.

The cell wall provides the protoplasts with the structural rigidity which directly determines the size and shape of the plant cell. At the same time, the apoplast, consisting mainly of the cell wall, is the site where the plant cell first makes contact with various environmental stimuli or stresses and thus plays important roles in plant responses to the outer environment (Hoson, 1998, 2002). Many aquatic and semi-aquatic plant species show stimulated elongation growth of shoot organs under conditions of submergence, which contributes to maintaining the apical parts above water. Submergence- induced stem elongation is primarily caused by an increase in cell wall extensibility due to changes in hormonal interactions (Voesenek et al., 2006). The involvement of some wall proteins, such as expansins and xyloglucan endotransglucosylase/hydrolases, in underwater-accelerated elongation has also been reported (Cho and Kende, 1997; Ookawara et al., 2005; Vreeburg et al., 2005). Nevertheless, little is known about the changes in the mechanical or chemical properties of the cell wall in ordinary terrestrial species, whose growth is suppressed rather than stimulated by submergence.

A reduction in cellular water potential to a more negative value creates the driving force for water uptake that drives an irreversible expansion of the cell wall. It is generally accepted that energy deficit caused by a shortage of O₂ for respiration and CO₂ for photosynthesis is one of the major problems for submerged plants (Greenway and Gibbs, 2003; Jackson and Ram, 2003; Voesenek et al., 2006; Bailey-Serres and Voesenek, 2008). The amount of carbohydrates is often positively correlated with the level of submergence tolerance, and high reserves of carbohydrates and slower rates of depletion during submergence are adaptive traits for plant survival underwater (Jackson and Ram, 2003). The shortage of carbohydrates may increase osmotic potentials, leading to increased water potentials that would slow underwater growth. However, in deepwater rice, accelerated elongation of internodal cells is related to an increase in cell wall extensibility and not to a decrease in osmotic potential (Kutschera and Kende, 1988). In other plant materials grown underwater, the relationship between the growth rate and the cellular osmotic potential has not been reported.

In the present study, the effects of submergence on the viscoelastic extensibility of the cell wall and the cellular osmotic concentration in azuki bean epicotyls were examined. The modifications by submergence of the processes that determine these growth-regulating parameters were also studied. To avoid the modification of submergence effects by light, most analyses were carried out using etiolated seedlings. Because underwater photosynthesis strongly affects the degree of submergence tolerance, these cellular events were also compared in both etiolated and light-grown seedlings.

MATERIALS AND METHODS

Plant materials and growth experiments

Seeds of azuki bean, Vigna angularis (Willd.) Ohwi & H.Ohashi ‘Erimowase’, were soaked in running tap water for 1 d at 30 °C and germinated on gauze lining a plastic dish filled with water at 25 °C in the dark. After 5 d , seedlings with 30–35 mm long epicotyls were selected, and 10 mm of the sub-hook region (3–13 mm below the hook) was marked with Indian ink. Marked seedlings were transplanted into rock wool blocks (M25S30, Nittobouseki, Tokyo, Japan) set in a plastic dish containing water and incubated either in air or under complete submergence at 25 °C in the dark. After incubation, the length of the marked regions was measured with a scale and then excised. The fresh weight of the excised segments was measured with an electronic balance. The seed (cotyledons) was also excised from the seedlings and dried at 50 °C before weighing. All manipulations were done under dim green light. In some experiments, germinated seedlings were grown for 5·5 d under continuous white light (approx. 30 µmol m⁻² s⁻¹) at 25 °C. The sub-hook region of uniform epicotyls was marked as above and the marked seedlings were grown in air or underwater at 25 °C under unchanged light conditions.

Determination of the mechanical properties of the cell wall

The marked regions excised from the epicotyls were immediately boiled in 80 % (v/v) ethanol for 10 min and stored in fresh 80 % ethanol until use. Before measurement of mechanical properties of the cell wall, ethanol-killed segments were rehydrated for 3 h at 4 °C with several changes of water. The viscoelastic extensibility of the cell wall was measured with a tensile tester (Tensilon RTM-25, Toyo Baldwin, Tokyo, Japan). The segments were fixed between two clamps 5 mm apart and stretched by lowering the bottom clamp by 20 mm min⁻¹ until a load of 10 g was produced. The viscoelastic extensibility of the cell wall (strain load⁻¹) was determined by measuring the rate of the increase in load just before it reached 10 g.

Determination of the level of cell wall polysaccharides

Methanol-fixed samples were rehydrated and homogenized in water with a mortar and pestle, washed with water, acetone, a methanol:chloroform mixture (1 : 1, v/v) and finally ethanol. The cell wall material was treated with 2 U mL⁻¹ porcine pancreatic α-amylase (EC 3·2·1·1, type I-A, Sigma, St Louis, MO, USA) in 50 mm sodium acetate buffer (pH 6·5) at 37 °C for 3 h and then washed with water. After the amylase treatment, pectic substances were extracted from the cell wall materials three times (15 min each) with 50 mm EDTA at 95 °C. The hemicellulose was then extracted three times (8 h each time) with 4 % (w/v) KOH and three times (8 h each time) with 24 % (w/v) KOH containing 0·02 % (w/v) NaBH₄ at 25 °C. The fractions extracted with 4 and 24 % KOH were designated as the hemicellulose-I (HC-I) and hemicellulose-II (HC-II) fractions, respectively. The HC-I and HC-II fractions were neutralized with acetic acid and then dialysed against water. The alkali-insoluble fraction (cellulose fraction) was washed successively with 0·03 m acetic acid and ethanol, and then dried at 40 °C. The cellulose fraction was dissolved in 72 % (v/v) sulfuric acid for 1 h at 25 °C and then diluted with a 29-fold volume of water. The total sugar content in each fraction was determined by the phenol–sulfuric acid method (Dubois et al., 1956) using glucose as the standard. The amount of xyloglucans in the HC-II fraction was determined by the iodine staining method (Kooiman, 1960).

Measurement of the molecular mass of xyloglucans

The HC-II fractions were lyophilized and the samples dissolved in 50 mm potassium-phosphate buffer (pH 7·2). A portion of the solution was injected into the gel permeation column (TSK-GEL 5000 PW, Tosoh, Tokyo, Japan) of an HPLC system (LC-6A, Shimadzu, Kyoto, Japan) equipped with a refractive index detector (RID-6A, Shimadzu). The sample was eluted with 50 mm potassium-phosphate buffer (pH 7·2) at a flow rate of 1 mL min⁻¹. The elution pattern was monitored by refractive index. Fractions were collected with a fraction collector (model-203, Gilson, Middleton, WI, USA) at 0·5 min intervals. The xyloglucan content in each fraction was determined by the iodine staining method. The weight-average molecular mass of xyloglucans was calculated with the elution profiles of molecular mass markers, dextrans of 10, 40, 70, 120 and 500 kDa (Sigma).

Determination of the osmotic properties of cell sap

The marked regions excised from epicotyls were rinsed with water and placed in parallel on a stainless mesh in the barrel of a plastic syringe. After removal of excess water on the surface of the segments by flash centrifugation, the segments were frozen with liquid nitrogen and kept at –80 °C. For use, frozen segments were placed on a stainless steel mesh in the barrel of a 12 mL plastic syringe (Top Surgical MFG, Tokyo, Japan) cut off at the 4 mL mark and thawed at room temperature. The cell sap was collected from frozen–thawed segments by centrifugation at 1500 g for 10 min at 4 °C. The osmotic concentration of the collected cell sap was measured with a vapour pressure osmometer (Model 5500 C, Wescor, Logan, UT, USA).

The amount of total osmotic solute in the epicotyl regions was estimated as the product of the osmotic concentration and the fresh weight. The content of sugars in the cell sap was determined by the phenol–sulfuric acid method and expressed as glucose equivalents, and that of amino acids by the ninhydrin method (Moore and Stein, 1954) as leucine equivalents. The amount of potassium ions was determined by flameless atomic absorption spectrophotometry using a polarizing Zeeman atomic absorption spectrometer (Z-5000, Hitachi, Tokyo, Japan).

Determination of ATP content

ATP was determined according to the method of Wulff and Doppen (1985) and Jaworek and Welsch (1985). Briefly, ATP was extracted from frozen tissue powder with ice-cold 10 % perchloric acid (v/v) with 5 mm EDTA. The extract was centrifuged at 18 500 g for 10 min at 4 °C and the supernatant was collected. KOH at 3 m was then added to the supernatant to adjust the pH to 7–8. After the pH adjustment, the solution was centrifuged at 18 500 g for 10 min at 4 °C, and the supernatant was stored at –80 °C until assay. The supernatant was then mixed with the test solution containing 500 mm HEPES buffer, 900 mm MgCl₂, 180 mm glucose, 9·9 mm NADP and 1·4 U mL⁻¹ glucose-6-phosphate dehydrogenase (EC 1·1·1·49, Oriental Yeast, Tokyo, Japan). After measuring the absorbance at 340 nm, 2 U mL⁻¹ hexokinase (EC 2·7·1·1, Calbiochem, Darmstadt, Germany) was added to the mixture solution, and then the mixture was incubated for 20 min at 25 °C. ATP content was determined by measuring the difference in absorbance at 340 nm before and after the addition of hexokinase.

Determination of ethanol content

Frozen segments were placed on a small column (Micro Bio-Spin Column, Bio-Rad, Hercules, CA, USA) and thawed at room temperature. Cell sap was obtained by centrifugation at 1500 g for 10 min at 4 °C. The concentration of ethanol in the sap was determined using an ethanol assay kit (Roche, Darmstadt, Germany) according to the manufacturer's instructions.

Measurement of apoplastic pH

The pH of the apoplastic solution was measured by the method of Soga et al. (2000). The excised regions of azuki bean epicotyls were rinsed with water and placed in parallel on a stainless mesh in the barrel of a plastic syringe. After removal of excess water on the surface of the segments by flash centrifugation, the segments were centrifuged in the longitudinal direction at 1500 g for 20 min at 4 °C to collect the apoplastic solution. The pH of the apoplastic solution was measured immediately after the centrifugation with a pH meter (B-112, Horiba, Kyoto, Japan) equipped with a flat-surface electrode. The activity of malate dehydrogenase in the apoplastic solution was assayed as a measure of cytoplasmic contamination. The activity was low, as reported previously (Soga et al., 2000), and similar in both control and submerged epicotyls.

RESULTS

Growth suppression

Elongation growth occurred only in the sub-hook region in 5-d-old etiolated azuki bean epicotyls. Elongation of this region was suppressed by complete submergence (Fig. 1). Growth suppression underwater was prominent after 3 h and, after 6 h, was about half of the control values. When submerged seedlings were transferred back to air (control) after 3 h underwater, elongation was partially restored.

Fig. 1.

Effects of submergence on elongation of azuki bean epicotyls. Etiolated seedlings with 30–35 mm long epicotyls were incubated either in air (circles) or under complete submergence (squares) at 25 °C in the dark. Elongation of the marked sub-hook region of 10 mm was measured. After 3 h, a portion of the submerged seedlings was transferred back to air (diamond, S → C). Values are means ± s.e. (n = 30).

Cell wall properties

Viscoelastic extensibility of the cell wall, measured with a tensile tester, decreased gradually during incubation in control epicotyls (Fig. 2). However, in submerged epicotyls, no such change was observed. Instead, viscoelastic extensibility remained high under submergence. The transplanting of submerged seedlings back to air made little difference to viscoelastic extensibility.

Fig. 2.

Effects of submergence on the viscoelastic extensibility of cell walls. Seedlings were incubated as in Fig. 1. The marked sub-hook region was excised and boiled in 80 % ethanol. The viscoelastic extensibility of the section was then measured with a tensile tester. Values are means ± s.e. (n = 30).

The biochemical basis of this maintenance of high underwater viscoelasticity was analysed. The levels of both matrix polysaccharides (the sum of the pectic, HC-I and HC-II fractions) and cellulosic polysaccharides per unit length of epicotyl were decreased by submergence (Fig. 3A), indicating that cell wall thinning was induced underwater. The levels of xyloglucans in the HC-II fractions were also lower by 25 % in submerged epicotyls than in the control (data not shown). On the other hand, xyloglucans in the HC-II fraction from submerged epicotyls eluted from a gel filtration column at lower molecular mass regions than did those extracted from control epicotyls (data not shown). The calculated weight-average molecular mass of xyloglucans from submerged epicotyls was significantly lower than that of the control (Fig. 3B).

Fig. 3.

The levels of cell wall polysaccharides (A) and the molecular mass of xyloglucans (B) as affected by submergence. Seedlings were incubated as in Fig. 1 for 6 h and the cell wall materials were prepared from the sub-hook region. Open bars, control; solid bars, submerged epicotyls. Values are means ± s.e. (n = 3).

Cellular osmotic properties

The osmotic concentration of the cell sap in control epicotyls was almost constant or decreased only slightly during incubation (Fig. 4). In contrast, osmotic concentrations in submerged epicotyls declined rapidly; the level being 80 mmol kg⁻¹ lower than in controls after 6 h. When submerged seedlings were transferred back to air conditions, the osmotic concentration was markedly restored.

Fig. 4.

Effects of submergence on osmotic concentration of cell sap. Seedlings were incubated as in Fig. 1. The osmotic concentration was measured with a vapour pressure osmometer. Values are means ± s.e. (n = 5).

The level of total osmotic solutes continued to increase during epicotyl growth in non-submerged controls, but this increase was greatly inhibited by submergence (Fig. 5). The return of submerged seedlings to air after 3 h restored the increase in osmotic solutes. In azuki bean epicotyls, osmotic solutes consist of sugars, and lesser amounts of amino acids and potassium ions (Fig. 6). Submergence strongly depressed sugar and amino acid concentrations, but levels of potassium, a major inorganic solute, were almost unchanged.

Fig. 5.

Effects of submergence on the content of total osmotic solutes. Seedlings were incubated as in Fig. 1. The amount of total osmotic solutes was estimated as the product of the osmotic concentration, shown in Fig. 4, and the fresh weight. Values are means ± s.e. (n = 5).

Fig. 6.

Effects of submergence on the content of major osmotic solutes. Seedlings were incubated as in Fig. 1. Sugar concentration was determined by the phenol–sulfuric acid method and expressed as glucose equivalents, and that of amino acids by the ninhydrin method as leucine equivalents. The concentration of potassium ions was determined by flameless atomic absorption spectrophotometry. Values are means ± s.e. (n = 5).

Cellular metabolism

To clarify the mechanism by which submergence decreases the levels of organic solutes, several other cellular compounds were measured. The concentration of H⁺ in the apoplastic solution was clearly modified by submergence. The apoplastic pH in control epicotyls remained almost constant at approx. 5·7 (Fig. 7) but was promptly and significantly increased by 0·5 units in submerged epicotyls. On return to air, apoplastic pH returned to control levels.

Fig. 7.

Effects of submergence on apoplastic pH. Seedlings were incubated as in Fig. 1 and the apoplastic solution was collected by centrifugation. The pH of the apoplastic solution was measured with a flat-surface pH electrode. Values are means ± s.e. (n = 3).

In control epicotyls, the concentration of ATP in the cell sap was almost constant or decreased only slightly during a 3 h incubation (Fig. 8), but was halved by 3 h submergence. On returning submerged epicotyls to air, control ATP levels were restored. On the other hand, ethanol production was greatly stimulated by submergence, which was suppressed on de-submergence (Fig. 9A). However, exogenously applied ethanol at 10 % (v/v), a concentration several times higher than that produced underwater, had no effect on elongation growth of azuki bean epicotyls at least up to 3 h treatment (Fig. 9B), indicating that ethanol toxicity is not the primary cause of growth suppression.

Fig. 8.

Effects of submergence on ATP. Seedlings were incubated as in Fig. 1 and ATP concentrations in the cell sap determined spectrophotometrically. Values are means ± s.e. (n = 3).

Fig. 9.

Effects of submergence on ethanol production and of exogenous ethanol on elongation growth. (A) Seedlings were incubated as in Fig. 1. The ethanol concentration in the cell sap was determined spectrophotometrically. For ethanol toxicity tests (B), seedlings were transplanted into rock wool blocks containing either water (circles) or 10 % (v/v) ethanol (triangles) and grown in air. Elongation of the sub-hook region was measured. Values are means ± s.e. (n = 3 for A, 15 for B).

Effects of light irradiation

The above-mentioned results were obtained with dark-grown etiolated azuki bean epicotyls. The effects of light on submergence-induced suppression of growth and osmoregulation were examined. In the dark, submergence suppressed the increase in fresh weight of epicotyls by 73 % (Fig. 10A). In the light this suppression was reduced to only 36 %. An increase in levels of organic solutes such as sugars and amino acids during incubation was completely inhibited by submergence in the dark, but only partially in the light (Fig. 10B). During incubation in air, the dry weight of seeds was decreased. Such a decrease was completely inhibited by submergence in the dark, whereas it was inhibited only partially in the light (Fig. 10C). On the other hand, the strong decrease in ATP seen during dark submergence was only slight in the light (Fig. 10D). Also, ethanol production was greatly stimulated by submergence in the dark, but not in the light (Fig. 10E).

Fig. 10.

Effects of light on submergence-induced suppression of growth, osmoregulation and cellular metabolism. Etiolated and light-grown seedlings were submerged for 6 h in the dark (left columns) and in the light (right). The following parameters were then measured: (A) the fresh weight of the sub-hook region of epicotyls; (B) the sugar content in the sub-hook region; (C) the dry weight of the seed; (D) ATP in the sub-hook region; and (E) ethanol in the sub-hook region. Sub., submerged epicotyls. Values are means ± s.e. (n = 3).

DISCUSSION

Overall plant growth necessarily requires a combination of cell division and cell expansion. However, in the present experimental system, azuki bean epicotyl extension is based entirely on cell expansion; cell division is not involved. Cell expansion is accomplished through enlargement of cell volume by water uptake into the vacuole and irreversible cell wall extension. It is generally believed that cell enlargement is initiated by relaxation of the cell wall, followed by water uptake in response to reduced cell water potential (Cosgrove, 1993, 2005). Thus, a decrease in the cell wall extensibility and/or a decrease in the osmotic concentration of the cell sap and associated fall in water potential can be expected when growth of azuki bean epicotyls is suppressed by submergence. In contrast to this expectation, the viscoelastic extensibility of the cell wall decreased gradually in control epicotyls but was maintained at a higher level in slower growing submerged epicotyls (Fig. 2), indicating that the modification of the wall viscoelasticity is not the cause of underwater growth suppression. It is notable that in the aquatic species Rumex palustris, submergence increases cell wall extensibility in association with faster petiole extension (Voesenek et al., 2006). The present results suggest that the maintenance of high viscoelastic extensibility of the cell wall occurs commonly under conditions of submergence not only in aquatic but also in ordinary terrestrial species. In azuki bean epicotyls, grown underwater, cell wall thinning and a molecular mass downshift of xyloglucans are induced (Fig. 3). Such changes are typical of cell walls of plants responding to various environmental signals (Hoson, 1998, 2002) and may be responsible for the maintenance of high viscoelastic extensibility underwater. The results agree with the report that the expression of genes encoding xyloglucan endotransglucosylase/hydrolases was induced in underwater-accelerated elongation (Ookawara et al., 2005).

The osmotic concentration of the cell sap was almost constant or decreased only slightly in control epicotyls but was decreased by submergence (Fig. 4). The decrease occurred promptly after submergence and preceded underwater growth suppression (Fig. 1). Also, when submerged seedlings were transferred back to air, the osmotic concentrations were completely restored (Fig. 4) along with the restoration of the growth rate (Fig. 1). These results suggest that a decrease in the osmotic concentration (equating to an increase in osmotic potential) is one of the main causes of underwater growth suppression in azuki bean epicotyls. There are several reports indicating that the amount of carbohydrate determines the level of submergence tolerance (reviewed in Jackson and Ram, 2003). Because sugars are also major osmotic solutes in azuki bean epicotyls (Fig. 6), a shortage of sugars underwater may be related to the decrease in osmotic concentration, leading to growth suppression.

The seed is the sole source of assimilates for etiolated seedlings, and submergence may interfere with the translocation processes of organic compounds from the seed to epicotyls of azuki bean seedlings. Analyses revealed that it is the level of organic solutes, not of inorganic ions, that is suppressed by submergence (Fig. 6). Also, the rate of decrease in seed dry weight during incubation was much slowed by submergence (Fig. 10), in accordance with an increase in dry weight of epicotyls (data not shown). In an as yet unpublished work, the influence of submergence on the loading and the unloading processes was examined separately. When azuki bean seedlings were submerged just above the seed, the levels of sugars increased during incubation, as in the control. This result suggests that submergence does not directly influence the loading process in the seed. On the other hand, organic compounds translocated from the seed are taken up by sink cells with the aid of the H⁺ gradient across the plasma membrane that is produced by the action of proton pumps (Bush, 1993). As shown in Fig. 7, the apoplastic pH was promptly and significantly increased by submergence, and completely restored to the control levels by returning submerged seedlings to air. Thus, proton co-transport of organic solutes into epicotyl cells is probably inhibited by submergence. Because the level of ATP was also decreased promptly by submergence (Fig. 8), the increase in the apoplastic pH may be brought about by inhibition of the action of proton pumps due to a decrease in the ATP. Invertase is also assumed to be involved in sugar uptake into sink cells (Miyamoto et al., 1993). However, apoplastic invertase activity is not much modified by submergence (data not shown).

The maintenance of the appropriate pH for each sub-cellular compartment is an important requirement for normal functioning of plant cells. A rapid cytoplasmic acidification is induced by submergence as a result of energy deficit, which is the major cause of submergence injury. This could be overcome by transport of excess H⁺ into the vacuole and the cell wall apoplast (Greenway and Gibbs, 2003; Felle, 2005). However, most studies on pH regulation under anoxia or hypoxia have focused on the cytoplasm and to some extent on the vacuole, and little attention has been paid to the apoplast (Felle, 2005). Ethylene is involved in enhanced elongation under complete submergence in R. palustris petioles, which has been shown to induce apoplastic acidification (Vreeburg et al., 2005). On the other hand, apoplastic alkalinization is induced by anoxia in barley leaves (Felle, 2006). In the present study, it was also shown that the apoplastic pH was clearly increased by submergence in azuki bean epicotyls (Fig. 7). In addition, it was recently reported that the activity of H⁺-ATPase on the plasma membrane in pea hypocotyls was decreased under anoxia (Hara et al., 2007). Thus, the regulation of the apoplastic pH occurs widely as the response of plant cells to submergence. One of the expected results of an increase in apoplastic pH by submergence would be a decrease in the viscoelastic extensibility of the cell wall, as observed in hypergravity-induced growth suppression (Soga et al., 2000). However, the viscoelastic extensibility was kept high under conditions of submergence (Fig. 2), suggesting that the modification of cell wall metabolism overwhelmed the change in the apoplastic pH.

When light-grown azuki bean seedlings were submerged in the light, an increase in the fresh weight, an increase in sugar content in epicotyls and a decrease in seed dry weight during incubation were only partially inhibited (Fig. 10A–C). Thus, light is capable of relieving growth suppression and the inhibition of osmoregulation by submergence. Underwater photosynthesis may produce both carbohydrates and O₂. When the seed was excised, the increase in sugar content during incubation was completely inhibited, irrespective of conditions of submergence or light (data not shown), suggesting that the seed acts as the sole source organ even in the light under the present experimental conditions. On the other hand, it is accepted that energy deficit, caused by a shortage of O₂ due to the slow diffusion rate in water, is one of the major problems encountered by plants under conditions of submergence (Greenway and Gibbs, 2003; Jackson and Ram, 2003; Voesenek et al., 2006). Also, in the present study, a submergence-induced decrease in ATP level or stimulation of ethanol production was not clearly observed in azuki bean epicotyls in the light (Fig. 10D, E). These results suggest that in the light-grown seedlings the maintenance of aerobic respiration favours the uptake of translocated sugars into epicotyl cells even under conditions of complete submergence. The data also support the hypothesis that the decrease in the osmotic concentration is one of the main causes of underwater growth suppression.

The term ‘cell wall extensibility’ conveys different meanings in different situations, and various methods have been used for its measurement (Cosgrove, 1993). In the present study, it was shown that the viscoelastic extensibility of the cell wall was kept high during at least 3 h of submergence (Fig. 2). However, the effect of submergence on other wall mechanical parameters remains to be elucidated. Expansins, responsible for chemorheological extension of the cell wall (Cosgrove, 2005), are involved in underwater-accelerated elongation in some aquatic species (Cho and Kende, 1997; Ookawara et al., 2005; Vreeburg et al., 2005). Also, an increase in apoplastic pH by submergence in azuki bean epicotyls (Fig. 7) may inhibit the action of expansins in vivo, because its activity is dependent on acidic pH (Cosgrove, 2005). More detailed analyses of the effects of submergence on the chemorheological nature of the cell wall as well as on the expression of expansin genes in terrestrial species will help to clarify further the cellular basis of growth suppression by submergence in stems.