Flooding imposes stress upon terrestrial plants since it severely hampers gas exchange rates between the shoot and the environment. The resulting oxygen deficiency is considered to be the major problem for submerged plants. Oxygen microelectrode studies have, however, shown that aquatic plants maintain relatively high internal oxygen pressures under water, and even may release oxygen via the roots into the sediment, also in dark. Based on these results, we challenge the dogma that oxygen pressures in submerged terrestrial plants immediately drop to levels at which aerobic respiration is impaired. The present study demonstrates that the internal oxygen pressure in the petioles of Rumex palustris plants under water is indeed well above the critical oxygen pressure for aerobic respiration, provided that the air-saturated water is not completely stagnant. The beneficial effect of shoot acclimation of this terrestrial plant species to submergence for gas exchange capacity is also shown. Shoot acclimation to submergence involved a reduction of the diffusion resistance to gases, which was not only functional by increasing diffusion of oxygen into the plant, but also by increasing influx of CO2, which enhances underwater photosynthesis.
Plants growing in flood-prone environments experience submerged conditions during parts of their life cycle. For terrestrial species, such complete submergence imposes stress because gas exchange rates between the shoot and the environment are severely reduced, since the diffusion rates of gases are 104 times lower in water than in air (Armstrong 1979). As a result, oxygen deficiency is considered to be the major factor negatively affecting survival and growth of submerged plants (Vartapetian & Jackson 1997; Voesenek et al. 2004), because it leads to energy deficits due to hampered aerobic metabolism (Crawford & Brändle 1996).
Many studies have investigated the responses of floodplain species that reduce the negative effects of oxygen deficiency (Armstrong, Brändle, & Jackson 1994; Blom 1999). Several species are able to change their shoot morphology in such a way that contact with the air can be restored, for example by enhanced elongation and reorientation of the leaves (Voesenek et al. 2003). Another common response of floodplain species is the increase of the amount of air spaces (aerenchyma) within the shoot and the roots to optimize within-plant gas transport (Colmer 2003). Furthermore, plants are able to switch to anaerobic metabolic pathways to prevent energy deficits and thus cell death (Perata & Alpi 1993).
Despite this broad palette of responses plants may employ to overcome the negative effects of submergence, oxygen pressures within the plant are considered to drop rapidly to very low levels when plants are submerged (Raskin & Kende 1984; Stünzi & Kende 1989; Rijnders et al. 2000). We have, however, reasons to challenge this dogma. This is inspired by recent gas exchange studies on aquatic plants showing that internal oxygen pressures in the shoots of these plants may in fact be sufficiently high to maintain aerobic respiration under natural conditions, even in the absence of photosynthesis (Greve, Borum & Pedersen 2003). Pedersen et al. (1998) and Revsbech et al. (1999) were the first to show that submerged seagrasses and rice plants release oxygen into the sediment, during daytime conditions and, unexpectedly, also at night. The oxygen release during daytime was fed by leaf photosynthesis (Smith, Dennison, & Alberte 1984), and in the dark by passive diffusion of oxygen from the water column via the aerenchymatous shoot into the roots (Pedersen et al. 1998; Pedersen, Binzer, & Borum 2004).
Building on these results, we studied gas exchange rates between the shoot of a terrestrial plant and the surrounding water. We used Rumex palustris, a submergence-resistant species (Nabben, Blom, & Voesenek 1999) as a model. This species allows us to investigate not only the default underwater gas exchange capacity of a non-acclimated terrestrial plant, but also whether shoot acclimation to submergence affects the underwater gas exchange capacity of the plant. Rumex palustris is a temperate floodplain species (Blom et al. 1994), which upon submergence shows typical wetland species characteristics, such as enhanced petiole elongation (Voesenek et al. 2004) and adventitious root and aerenchyma formation (Visser, Blom, & Voesenek 1996). More importantly for the current study, this species also shows a high plasticity in leaf morphology upon submergence (Mommer, unpublished data).
We expected that the submerged leaf morphology facilitates gas exchange under water. We therefore measured not only the effect of the oxygen pressure of the water column on the internal oxygen pressure of the shoot, but also the contribution of underwater photosynthesis to the internal oxygen pressure with varying CO2 concentrations in the surrounding water.
MATERIALS AND METHODS
Plant material and treatments
Rumex palustris Sm. seeds were collected from the floodplains near Nijmegen (The Netherlands) and germinated for 10 d in a Petri dish on moistened filter paper at temperatures of 22 °C during daytime [photosynthetic photon flux density (PPFD) 20 µmol m−2 s−1] and 10 °C at night. Then the seedlings were transplanted to pots of 6 cm × 6 cm × 8 cm, containing a mixture of one volume potting soil and one volume sand and placed in a climate room [PPFD approximately 250 µmol m−2 s−1 (sodium lamps Philips SON-T plus 600 W and fluorescent light TDL Reflex 36 W/840R; Eindhoven, The Netherlands); day/night cycle: 16 h/8 h; temperature 20 °C] for 28 d.
To test the effect of shoot acclimation to submergence on internal oxygen pressure, one group of plants was then fully submerged, whereas the other group was kept drained. The submerged plants were flooded in a glass aquarium of 30 cm × 60 cm × 40 cm, which was filled with tap water. Atmospheric air was bubbled through the water to maintain air-equilibrium concentrations of O2 (21 kPa) and CO2 (15 µm or 0.04 kPa). A pump (Type 40C; Rena, Annecy, France) provided circulation of the water. In both treatments, PPFD was 140 µmol m−2 s−1; the day/night cycle was 16 h/8 h, and the temperature 20 °C. The treatments lasted 12–18 d, and the plants had developed at least two new leaves during the treatments.
Oxygen microelectrode measurements
Three aquaria containing 4.5 L water were placed in a thermostat-controlled bath to maintain the temperature at 20 °C. The water in the aquaria consisted of partly de-ionized water with an alkalinity of 1.9 meq L−1 and a conductivity of 220 µS cm−1. These two parameters together with pH and temperature, allowed the free CO2 concentration in the water to be calculated according to Mackereth, Heron, & Talling (1978). A Clark-type oxygen microelectrode (OX10; Unisense, Århus, Denmark) placed in one of the aquaria monitored the oxygen pressure of the water surrounding the shoots. Desired O2 and CO2 concentrations of the water were achieved by bubbling the water with mixtures of O2, CO2 and N2 obtained by mixing pure O2, CO2 and N2 (Air Liquide, Ballerup, Denmark) using mass flow controllers (Brooks Instruments, Veenendaal, The Netherlands). Increasing CO2 concentrations of the water concomitantly lowered the pH of the water from 8.5 at very low CO2 concentrations to 6.5 at the highest concentrations, but this pH range has been shown not to affect underwater photosynthesis in other amphibious plants (Madsen & Sand-Jensen 1991; Sand-Jensen & Frost-Christensen 1998; Andersen & Pedersen 2002). Because of the bubbling, water was circulated in the aquaria with a flow velocity of 2–5 cm s−1, which minimized the formation of boundary layers. Light was provided with glass-fibre tubes attached to lamps (Osram Halogen Xenophot, Osram, Munich, Germany) with a PPFD of 450 µmol m−2 s−1, which was sufficient to light-saturate underwater photosynthesis for this species (Mommer, unpublished data)
Oxygen microelectrode measurements were performed in the petiole of the fifth developing leaf, which in all cases had developed during the treatment. Whole plants were mounted horizontally in the aquaria in such a way that the petiole was fixed 3 mm below the lamina and water moved freely around all lamina. Before transfer, roots with soil were gently transferred from the pot into a plastic bag (10 cm × 12 cm) with a piece of raw cotton around the root–shoot junction preventing the soil from being flushed out during the measurements. The set-up was chosen in such a way that the lamina could be considered as the oxygen source in light because of photosynthesis, but also in dark, because of passive oxygen diffusion from the water layer into the leaf blades. This passive diffusion from the water into the plant mainly occurred through the lamina rather than through the petiole surface, since the lamina represents a much larger gas exchange area with a considerably higher surface area : volume ratio. The roots acted as a sink for internal oxygen, because they were placed in oxygen-deficient soil and were not in direct contact with the surrounding water.
Oxygen microelectrodes are nowadays widely used in plant sciences to measure oxygen pressures in tissues (Revsbech 1989; Armstrong 1994; Pedersen et al. 2004). The conical tip of the microelectrodes used in this study (OX10; Denmark) was so small (φ = 10 µm) that it could easily be inserted into the petiole without damaging the plant tissue. In situ partial oxygen pressures within the plant's petiole can thus be monitored very accurately, since the response time of these electrodes is fast (90% stability within < 3 s) and their sensitivity to stirring so small (< 2%) that their oxygen consumption is negligible. Data were logged from a picoamp meter (PA8000; Unisense) every 15 s with an A/D converter (AD 16; PicoTech, St Neots, UK). Electrode signals were converted to oxygen concentrations using individual calibrations between 0 and 100% of air-saturation before and after each series of measurements. Measurements were replicated three times.
Plant parameters measured included the area (LI3000; Li-Cor Inc., Lincoln, NE, USA) of the leaf at which oxygen measurements were conducted, total leaf area of the plant, and dry weights of the lamina, petioles and roots, which were determined after drying for 48 h at 80 °C (n = 6). Aerenchyma content of the petiole, a measure for within-plant gas transport capacity, was measured using the microbalance method as described in Visser & Bögemann (2003) (n = 10). Differences between submergence-acclimated and non-acclimated plants were statistically analysed with Student's t-test (P < 0.05). If needed, data were ln-transformed prior to analysis in order to have equal variances.
The effect of submergence on plant morphology was clearly visible (Table 1). As expected, petioles of the submergence-acclimated plants were significantly larger in comparison with the non-acclimated controls. The aerenchyma content of the petiole tissue was also significantly higher in the submergence-acclimated plants (Table 1), indicating an increased within-plant transport capacity. The differences in porosity and length of diffusion pathway between the leaf types did not influence our microelectrode data, because the microelectrodes were inserted very close to the lamina, and thereby as close to the oxygen source as possible. Total number of leaves and total biomass remained unaffected upon submergence and the area of leaf 5 was not significantly different between the plant types either.
Table 1. Plant parameters of submergence-acclimated and non-acclimated plants (means ± SE), statistically tested with Student's t-tests
P < 0.001; NS, not significantly different. Number of replicates is six, except for aerenchyma content (n = 10).
We examined the relationship between the oxygen pressure of the water and the oxygen pressure within the plant when submerged (Fig. 1). The plants were submerged in the dark in this experiment, to be certain that only passive diffusion of oxygen from the water column into the plant occurred. Internal oxygen pressures reached a steady state within half an hour after manipulation of the oxygen pressure of the water in which the plants were measured, which was mainly due to the time needed to reach the desired gas pressure in the water. Internal oxygen pressures were proportional to the oxygen pressure in the water in both submergence-acclimated and non-acclimated plants, but still lower than the oxygen pressure of the water column (Fig. 1). In air-saturated water (oxygen pressure 21 kPa) submergence-acclimated plants reached stable internal oxygen pressures of 17 kPa, whereas the non-acclimated plants only reached 9 kPa. The effect of shoot acclimation to submergence on gas exchange between the water column and the lamina was thus substantial: the internal oxygen pressure was almost a factor of two higher in the submergence-acclimated plants compared with the non-acclimated plants.
Diffusion of CO2 into the plant in the light
A significant increase of the internal oxygen pressure was observed in both non-acclimated and acclimated plants when the light was switched on after a dark period (Fig. 2), even though during this experiment the availability of free CO2 in the water was rather low [free CO2 concentration < 8 µm (< 0.02 kPa)]. The absolute increase in oxygen pressure in light compared with the dark was 3.5 and 5 kPa for the acclimated and non-acclimated plants, respectively, or, expressed as relative increase proportional to the dark oxygen concentration, 22 and 43%.
When the free CO2 concentration of the water increased (Fig. 3), internal oxygen pressures rose, indicating that photosynthetic rates increased. Submergence-acclimated plants responded already at 40 µm (0.1 kPa) free CO2 in the water column, whereas non-acclimated plants were unresponsive until between 250 and 460 µm (0.65–1.2 kPa) free CO2 in the water column. At the latter concentration, internal oxygen pressures of submergence-acclimated plants were already saturated.
Micro-electrode measurements across the submerged petiole showed that the exact depth of insertion of the microelectrode did not affect the readings of oxygen pressures (Fig. 4). Apparently, all cells within the petiole are located close to the intercellular air spaces, and these air spaces are sufficiently large (aerenchyma content of > 20%) to provide a homogeneous oxygen concentration in the radial direction and to prevent the occurrence of anaerobic cores in the petiole. Conversely, the internal oxygen pressure did decrease along the petiole with increasing distance from the lamina (Fig. 5), indicating that the lamina was the major source and the roots the major sink for oxygen within the plant. Unexpectedly, the relationship between internal oxygen pressure and distance from lamina was a linearly decreasing function (Fig. 5), which is in contrast with exponential predictions of gas diffusion models, deduced from experimental work with roots (Armstrong 1979, 1994; Christensen, Revsbech, & Sand-Jensen 1994). This may be explained by the fact that the green petioles are also photosynthetically active and thus partly compensate for the decrease with distance.
Diffusion of oxygen into the plant in the dark
The internal oxygen pressure of the petioles of R. palustris plants correlated strongly with the oxygen pressure of the water column in dark (Fig. 1), which indicates that the water column functioned as an important source of oxygen for the plant. This passive diffusion of oxygen from the water column into the plant has been observed for aquatic macrophytes (Sand-Jensen, Prahl, & Stokholm 1982; Pedersen et al. 1998; Greve et al. 2003; Pedersen et al. 2004) but to our knowledge, this is the first time to show this process in submerged terrestrial species. At ambient oxygen pressure, the internal oxygen pressures of acclimated, but surprisingly also of non-acclimated R. palustris petioles were well above the critical internal oxygen pressure (0.8 kPa) (Armstrong & Gaynard 1976; Armstrong & Webb 1985) (Fig. 1) that is considered to be sufficient for maximum aerobic respiration.
Strikingly, the internal oxygen pressures of submergence-acclimated plants were considerably higher than those of non-acclimated plants (Fig. 1). Moreover, the internal oxygen pressures of petioles of submergence-acclimated plants were almost similar to the oxygen pressures of the water column. This shows clearly that shoot acclimation to submergence is particularly functional with respect to gas exchange capacity between the water column and the plant. The effect cannot be caused by a difference in gas exchange surface, as the leaves at which the measurements were conducted were not significantly different in leaf area between acclimated and non-acclimated plants (Table 1).
Our results contrast at first sight with the data presented by Raskin & Kende (1984), Stünzi & Kende (1989) and Rijnders et al. (2000), in which internal oxygen pressures dropped to levels where aerobic respiration is impaired. The experiments of Rijnders et al. (2000) on the same species showed that the internal oxygen pressure of a non-acclimated plant dropped within 30 min from 21 to 4 kPa, and after 20 h it was even less than 0.5 kPa. In the present experiment, within half an hour a stable oxygen pressure of 17 and 9 kPa was observed in the petioles of acclimated and non-acclimated plants, respectively, when the oxygen concentration of the water was air-saturated. The difference between the experiments can be explained by a difference in the thickness of boundary layers around the leaves. In the present study we explicitly aimed at minimizing boundary layer effects, whereas in the other studies the solution surrounding the plant was not stirred, thereby leading to significant boundary layers around the leaves. True stagnant conditions are rare in the natural aquatic habitats occupied by amphibious plant species. Not only river systems, but even littorals of lakes are characterized by flow velocities that are sufficiently high to minimize boundary layer effects (Westlake 1967; Sand-Jensen & Pedersen 1999). This indicates that dark conditions are not necessarily as detrimental for the shoot under water as proposed in the former studies (Raskin & Kende 1984; Stünzi & Kende 1989; Rijnders et al. 2000).
Enhanced oxygen concentrations (Figs 1, 2 & 3) are not only of great importance for the energy status of the shoot, but also for the roots. The oxygen concentration at the shoot base determines the aeration status of the roots (Connell, Colmer, & Walker 1999; Sorrell et al. 2000) if these organs are interconnected by aerenchymatous tissue (Armstrong et al. 1994), which is the case in R. palustris (Laan et al. 1990). Rumex palustris plants increased the amount of aerenchymatous tissue in the petioles by 50% upon submergence (Table 1), indicating a high gas transport capacity within the acclimated plant (Colmer 2003). It is therefore likely that also the roots of the tolerant R. palustris will benefit from the acclimation of the shoot to the water environment.
Fixation of respired CO2
Light appeared to be beneficial for R. palustris plants even at free CO2 concentrations in the water that are below concentrations that can increase the internal oxygen pressure, as switching on the light rapidly raised the internal oxygen concentration with 22–43% (Fig. 2). Because R. palustris cannot use HCO3– as carbon source for photosynthesis under water (data not shown, cf. R. maritimus and R. crispus; Laan & Blom 1990), we suggest CO2 produced during respiration in the light to be the carbon source that caused the increase of internal oxygen pressure in this experiment, although CO2 uptake from the soil cannot be excluded. Respiration provides a constant delivery of CO2 and fixation of this internally produced CO2 may well contribute to photosynthesis leading to increased internal oxygen pressure in the light. That the putative fixation of respired CO2 might reduce the loss of carbon in the submerged plant considerably is nicely illustrated by experiments of Nabben et al. (1999) and Vervuren, Blom, & de Kroon (2003). They showed that the availability of light under water is beneficial for survival and biomass maintenance of flooded terrestrial plants, including R. palustris, when free CO2 in the water column was almost as low as in our experiment.
Diffusion resistance to CO2
Increased external free CO2 led to elevated internal oxygen pressures through enhanced underwater photosynthesis in both acclimated and non-acclimated plants (Fig. 3). The internal oxygen pressure was already higher in submergence-acclimated plants compared with the non-acclimated plants at the lowest CO2 concentration, because of the earlier mentioned difference in diffusive oxygen flux from the water column into the plant (Fig. 1). Interestingly, internal oxygen pressures in submergence-acclimated leaves already increased at lower external CO2 concentrations than in non-acclimated leaves. This provides additional proof that the resistance to gas exchange with the surrounding water is lower in submergence-acclimated than in non-acclimated plants. Such differences in diffusion resistance between the leaf types are ecologically relevant, because the levels of free CO2 in for example the main stream of the river Rhine and its floodplains range from values of 15–90 µm (0.04–0.24 kPa) (van den Brink et al. 1993), which is well within the range in which acclimated leaves perform better than non-acclimated leaves (Fig. 3).
The cuticle of the lamina will be the major barrier for gas diffusion from the water column into the plant, as stomata are most likely not functioning under water and thus gases need to cross this waxy layer to enter the lamina. Recent work on several amphibious plant species showed that both cuticle thickness and cuticle resistance to oxygen diffusion were lower in leaves that were formed under water compared with aerial leaves of the same species (Frost-Christensen, Bolt Jørgensen, & Floto 2003). It is therefore likely that the differences in gas diffusion resistance between the Rumex leaf types are the result of differences in cuticle thickness and resistance.
Summarizing, this study modifies the dogma that internal oxygen concentrations always decrease to dramatically low levels when plants are submerged. It is still true that the internal oxygen pressure of petioles of R. palustris under water highly depends on the oxygen concentration of the water column, but if the water is not stagnant, oxygen pressures within both acclimated and non-acclimated shoots will not drop to the critical level that prevents aerobic metabolism. We also conclude that acclimation of the shoot to submergence is beneficial to gas exchange between the water column and the plant, since it involves a reduction of the diffusion resistance to gases. This is not only functional for increasing diffusion of oxygen from the water column into the plant, but also for increasing the influx of CO2, which benefits under water photosynthesis.
We like to thank Charlotte Andersen for technical assistance, and Hans de Kroon for valuable comments on an earlier version of the manuscript.