•Terrestrial plants experience multiple stresses when they are submerged, caused both by oxygen deficiency due to reduced gas diffusion in water, and by shade due to high turbidity of the floodwater. It has been suggested that responses to submergence are de facto responses to low light intensity.
•We investigated the extent to which submergence and shade induce similar acclimation responses by comparing two terrestrial Rumex species that differ in their responses to flooding.
•Our study confirms that there are strong similarities between acclimation responses to shade and submergence. Petiole length, specific leaf area (SLA), chlorophyll parameters and underwater light-compensation points changed at least qualitatively in the same direction. Maximum underwater photosynthesis rate, however, did discriminate between the functionality of the responses, as the acclimation to submergence appeared to be more effective than acclimation to shade at saturating light.
•We conclude that acclimation to submergence involves more than an increase in SLA to achieve the significant reduction of diffusion resistance for gas exchange between leaves and the water column.
Different environmental stimuli may induce similar morphological modifications of leaves (Arntz & Delph, 2001; Meinzer, 2003) and roots (Schmidt & Schikora, 2001). For example, the morphology of leaves that are shaded or submerged may appear the same. However, these leaves may have different functional characteristics in either situation. To distinguish if an apparent trait similarity is also functionally the same, comparative in-depth studies are needed to gain understanding of the complex relationship between environmental signals and the resulting phenotypes.
In this study we investigate explicitly to what extent the acclimation responses to shade and submergence are convergent in terrestrial floodplain species. Two terrestrial Rumex species that differ in their ability to respond to submergence (Nabben et al., 1999) allowed us to investigate whether these differential responses to submergence are consistent with their responses to shade. First, we investigated responses to shade, including typical shade acclimation of photosynthesis (lower maximum photosynthesis rates, lower dark respiration and lower light-compensation points, Sims & Pearcy, 1994; Terashima & Hikosaka, 1995; Givnish et al., 2004) as well as well known morphological (specific leaf area, SLA and petiole length) and biochemical (chlorophyll content and a : b ratio) parameters (Björkman, 1981; Givnish, 1988; Terashima et al., 2001). Next we compared these morphological and biochemical responses to shade with those under water. If these responses to submergence are in fact shade responses, we expect changes under water to mimic the changes in shaded drained conditions, irrespective of the light regime under water. Finally, we investigated underwater photosynthesis of plants grown under four different conditions: every combination of drained or submerged with high or low light. If responses to submergence are in fact shade responses, we expect drained shade-acclimated plants to have underwater photosynthesis characteristics similar to those of submergence-acclimated plants. If gas diffusion is more limiting for plant growth under water than light (Bowes, 1987; Maberly & Madsen, 2002), submergence-acclimated plants will behave similarly under both light conditions under water, whereas drained plants – shade-acclimated or not – will do worse.
Materials and Methods
Two floodplain species, Rumex palustris Sm. and Rumex thyrsiflorus Fingerh., were studied in this experiment. Rumex palustris is a characteristic species for the low elevated and thus frequently flooded sites of the floodplain, whereas R. thyrsiflorus occurs at drier and higher elevations (Blom et al., 1994; Nabben et al., 1999). The species differ in resistance to submergence, as R. palustris has a much higher survival time under water than R. thyrsiflorus (Nabben et al., 1999). Furthermore, when submerged, R. palustris is able to develop leaves with a more elongated morphology than the leaves that are formed in air (Voesenek et al., 2004), whereas the leaves of R. thyrsiflorus do not elongate as much when submerged (cf. Rumex acetosa in Voesenek et al., 1993a). Seeds of these species were collected from the floodplains near Nijmegen (the Netherlands) and germinated for 10 d on moistened filter paper in Petri dishes at temperatures of 22°C during daytime [photosynthetic photon fluence rate (PPFD) 20 µmol m−2 s−1] and 10°C at night. The seedlings were transplanted to pots of 6 × 6 × 8 cm, containing a mixture of one volume potting soil (Hortimea Group, Lent, the Netherlands) and one volume sand. Plants were grown for a further 20 d (on average, see below) in a growth chamber (PPFD at leaf level ≈220 µmol m−2 s−1 (sodium lamps SON-T plus 600 W and fluorescent light TLD Reflex 36 W/840R, Philips, Eindhoven, the Netherlands); day : night cycle 16 h light : 8 h dark; temperature 20 ± 2°C). The positions of the pots were changed regularly to homogenize the growth conditions among replicate plants.
In the first experiment, one group of plants was subjected to high or low light intensities [PPFD at leaf level 140 vs 30 µmol m−2 s−1; referred to as high light (HL) vs low light (LL)] under drained conditions in the same growth chamber as described above. Neutral filters (type ULS 10, Ludvig Svensson, the Netherlands) were used to attain the two light intensities. Aerial photosynthesis characteristics were determined on these plants.
A fully factorial second experiment with the two factors, water level (drained vs complete submergence) and light intensity (HL vs LL) was performed with another group of plants. The plants treated under these conditions enabled morphological, biochemical and underwater photosynthesis comparisons of the acclimation to shade and submergence. Each treatment lasted 12 d, during which leaves present at the start of the experiment acclimated and one to three new leaves were formed. Plants in submergence and drained treatments had different growth rates, therefore duration of the growth period previous to the treatment was chosen in such a way that plants reached a comparable developmental stage at the end of each treatment, based on the total number of leaves. The period from sowing until the start of drained and submerged treatments varied therefore from 26 to 34 d.
All treatments were carried out in opaque polyethylene basins (80 × 60 × 70 cm). The basins used in the drained treatment were kept drained, whereas the basins used for submergence were filled with tap water. The water was circulated with a flow rate of 1.5 l min−1 and filtered to prevent algal growth. Temperature of the water was 20°C. Total dissolved inorganic carbon concentration (measured with an infrared gas analyser, Horiba PIR 2000; Northampton, UK) was 1.3 mm, which resulted at pH 8.4 in a free CO2 concentration of 15 µm. The same neutral filters as used in experiment 1 were used to attain the two light intensities in the basins.
Aerial photosynthesis measurements (experiment 1)
To verify if both species are able to acclimate photosynthetically to shade, we measured aerial photosynthesis characteristics on drained plants that had acclimated to high and low light intensity. These measurements were not performed on submergence-acclimated plants, as these data would be obscured by unpredictable stomatal behaviour caused by desubmergence effects. The aerial gas-exchange measurements were performed on leaves (leaf number 7) of drained plants (n = 4) by measuring CO2 uptake using a portable infrared gas analyser (LI-COR 6400; Lincoln, NE, USA) (370 µmol mol−1 CO2, 80% RH, temperature 20°C in leaf chamber).
Maximum photosynthesis rates were observed at a PPFD of 500 µmol m−2 s−1 for both treatments. Respiration rates were measured in the dark. Apparent quantum yield, a parameter that expresses the initial slope of the photosynthesis–irradiance curve, was calculated by linear regression of the net values obtained at PPFDs of 13, 27 and 40 µmol m−2 s−1. Light-compensation points were calculated from extrapolation of data from these same PPFDs to the light intensity where respiration equals assimilation and thus net photosynthesis is zero.
Morphological and biochemical parameters (experiment 2)
Morphological responses to shade and submergence are reflected by petiole length and changes in SLA. The petiole length of the largest leaf was recorded (n = 10), and the SLA calculated from the total dry weight of three leaf punches (1.07 cm2 each) per plant (n = 6–8). This resulted in relatively high SLA values, as vein material was excluded from the samples.
Biochemical acclimation responses are reflected by changes in chlorophyll content and relative investment in chlorophyll a : b. Chlorophyll contents of the youngest fully grown leaves were determined spectrophotometrically (Shimadzu Benelux, Den Bosch, the Netherlands) after extraction with 96% ethanol (n = 6–10). Equations of Wintermans & de Mots (1965) were used to calculate the chlorophyll a and b pigment concentrations.
Effects of photosynthetic acclimation to shade and submergence were investigated by measuring underwater photosynthetic characteristics, defined as oxygen production or consumption in water. Therefore plants from all treatments were submerged in transparent Perspex cuvettes (820 or 1200 ml), depending on the size of the plants. The cuvettes were filled with 5.0 mm NaHCO3, which was adjusted to pH 6.5 with 7.5% HCl, yielding a free CO2 concentration of 2.2 mm. Before pH adjustment the O2 concentration was lowered to 50% of air saturation by flushing with N2 to ensure a steep diffusion gradient from the leaves to the surrounding water. Measurements were performed on whole shoots (n = 6–8) from which the roots had been excised. After 1 h acclimation, oxygen release and uptake were recorded using a YSI 5331 oxygen probe electrode (YSI, Yellow Springs, OH, USA) connected to a Diamond Micro Sensor (Ann Arbor, MI, USA). A rotary pump (Rena type C40, Annecy, France) provided a constant circulation of 10 l min−1 in order to minimize the development of boundary layers. Water temperature was kept constant at 20 ± 1°C using a Haake DC 50 cooling system (Karlsruhe, Germany). Real-time logging of the oxygen concentration was performed with a Grant 1600 SQ data logger (Cambridge, UK). Average oxygen production was calculated from the logged data, after the system had stabilized for 15 min.
Underwater photosynthesis parameters were measured at the same light intensities as described above for the aerial photosynthesis measurements. Different light intensities in the underwater photosynthesis setup (PPFD 500, 40, 27, 13 and 0 µmol m−2 s−1) were achieved by attenuating the light (SON-T plus 600 W, Philips, Eindhoven, the Netherlands) with metal gauze screens. The cuvette with plant was placed horizontally, and mirrors around the cuvette reflected the light coming from above so that self-shading was minimized. Dark respiration measurements were performed at 80% of air saturation, which does not limit respiration in Rumex species (Laan et al., 1990). After the underwater photosynthesis measurements, shoot dry weight and leaf area were determined. Shoot dry weight was determined after drying for 48 h at 80°C. Leaf area was determined with a portable leaf area meter (LI-COR 3000).
Within a species, the data were analysed using two-way anova and Scheffé post hoc tests across different levels of light intensities and water levels with significance levels set to 0.05 (sas version 8.2; SAS Institute, Cary, NC, USA). Data were ln-transformed in several cases in order to have equal variances. Differences between aerial photosynthesis parameters of drained HL and LL plants were analysed using Student's t-tests (P = 0.05).
Responses to shade (experiments 1 and 2)
Both R. palustris and R. thyrsiflorus showed classic shade responses under drained conditions for all parameters investigated. The aerial photosynthesis measurements (experiment 1) indicated that both species show entirely normal shade acclimation. As expected, maximum photosynthesis rates were reduced and accompanied by low dark respiration rates in response to low light (Table 1). Furthermore, light-compensation points tended to decrease, and apparent quantum yields were unaffected (Table 1).
Table 1. Aerial photosynthesis parameters of drained Rumex thyrsiflorus and Rumex palustris, grown at high and low light intensities
Pmax, maximum net photosynthesis rate at PPFD of 500 µmol m−2 s−1 (µmol CO2 m−2 s−1); Rd, dark respiration rate (µmol CO2 m−2 s−1); α, apparent quantum yield calculated by linear regression of the net values obtained at PPFD 13, 27 and 40 µmol m−2 s−1 (µmol CO2 µmol−1 photons); light-compensation points (LCP) were calculated from extrapolation of data from these same PPFDs (µmol photons m−2 s−1).
Data are means ± SE, n = 4. Student's t-test was performed to distinguish shade effects.
Significance levels: ***, P < 0.001; **, P < 0.01; *, P < 0.05; ns, not significant.
Typically, petiole length (Fig. 1; experiment 2) and SLA (Fig. 2; experiment 2) increased significantly under shaded conditions compared with HL conditions. As expected, shade acclimation under drained conditions did not reveal significant differences in chlorophyll content per unit surface area, but expressed per unit dry weight, shade-acclimated plants contained significantly more chlorophyll than HL-grown plants (Table 2; experiment 2). Chlorophyll a : b ratios decreased significantly under shaded conditions (Table 2), indicating a higher relative investment in light-harvesting capacity than in carbon-assimilating capacity.
Table 2. Chlorophyll parameters of Rumex thyrsiflorus and Rumex palustris grown at high and low light intensities in drained and submerged conditions
Data are means ± SE, n = 6–10. Data were ln transformed before statistical analysis. Different letters in the columns within species indicate significant differences (P < 0.05).
Chlorophyll (mg m−2)
141 ± 7 ab
126 ± 6 a
158 ± 7 b
142 ± 4 ab
Chlorophyll (mg g−1 d. wt)
7.2 ± 0.86 a
12.27 ± 0.47 b
10.37 ± 0.59 b
10.43 ± 0.47 b
Chlorophyll a : b ratio
4.0 ± 0.28 a
2.69 ± 0.06 b
2.75 ± 0.10 b
2.85 ± 0.05 b
Chlorophyll (mg m−2)
127 ± 4 A
148 ± 3 A
144 ± 13 A
137 ± 4 A
Chlorophyll (mg g−1 d. wt)
5.3 ± 0.40 A
18.6 ± 1.65 B
15.3 ± 1.49 B
14.07 ± 0.47 B
Chlorophyll a : b ratio
3.4 ± 0.05 A
2.7 ± 0.03 B
3.13 ± 0.05 C
2.8 ± 0.05 B
Responses to submergence (experiment 2)
Both species showed significant petiole elongation on submergence, when HL plants from the submerged treatment were compared with drained HL plants (Fig. 1; significant main water level effect, Table 3). Low light under water induced additional elongation (Fig. 1). However, the petioles of R. palustris elongated under water several-fold more than those of R. thyrsiflorus.
Table 3. Results of anova across levels of light intensity (L) and water level (W) and their interaction for both Rumex species
SLA, specific leaf area (m2 g−1 d. wt); Pmax, maximum net photosynthesis rate under water (µmol m−2 s−1); Rd, dark respiration rate under water (µmol m−2 s−1); LCP, light-compensation point under water (µmol m−2 s−1); α, apparent quantum yield under water (µmol µmol−1).
Significance levels: ***, P < 0.001; **, P < 0.01; *, P < 0.05; ns, not significant.
Specific leaf area (SLA) increased in response to submergence, although this increase was less pronounced than in shade (Fig. 2). Light conditions under water did not affect the change in SLA (Fig. 2). The chlorophyll parameters changed similarly in response to submergence to the way they changed in response to low light (significant interaction L × W for chlorophyll per leaf area and chlorophyll a : b ratio; Tables 2, 3).
The typical differences in aerial light-response curves that were observed between HL- and LL-grown drained plants were not apparent in underwater photosynthesis measurements. Maximum net photosynthesis rates under water were similar for drained HL and LL plants of both species (Fig. 3). Apparently shade acclimation has no detectable effect on photosynthesis under water.
A strong change in underwater photosynthesis rate, however, was observed in the submergence-acclimated plants: these plants showed significantly higher maximum photosynthesis rates than drained ones, except for submerged LL R. palustris, in which the maximum photosynthesis rate was lower (Fig. 3).
Light-compensation points (Fig. 4) were highest in the HL drained plants in both species, and were significantly lower in LL drained and submerged plants. Dark respiration (Fig. 3) and apparent quantum yield (Fig. 5) were not affected by the treatments in R. thyrsiflorus. In contrast with R. thyrsiflorus, these parameters changed significantly in R. palustris (Figs 3, 5): submerged HL plants had lower respiration rates and higher apparent quantum yields than the drained plants, which may contribute to survival under water. Surprisingly, submerged LL R. palustris plants showed opposite responses.
In the literature, acclimation to submergence has been hypothesized to be an acclimation to shade (Bowes & Salvucci, 1989; Frost-Christensen & Sand-Jensen, 1995; Wells & Pigliucci, 2000; Boeger & Poulson, 2003). Our results underline this contention, as the morphological and biochemical responses to shade and submergence were in the same direction, although they could differ by an order of magnitude. Moreover, underwater light-compensation points were lower in both shade-acclimated and submergence-acclimated plants as compared with drained HL-grown plants. The difference between the acclimation responses, however, became clear when measuring underwater photosynthesis at saturating light intensities. Maximum underwater photosynthesis rates were much higher in submergence-acclimated than in shade-acclimated plants, suggesting that acclimation to submergence is directed towards optimization of gas exchange, rather than to optimization of light capture.
Morphological and biochemical responses to shade and submergence are similar
Many of the morphological responses to shade and submergence were, at least qualitatively, the same. Still, significant interactions between light and water level were observed for petiole length and SLA in both Rumex species (Figs 1, 2; Table 3). Submergence affected petiole length significantly more than shade, which may be explained by mechanical support from the water column, allowing longer petioles to emerge under water. In contrast to petiole elongation, SLA was affected more strongly by the shade treatment.
Acclimation to shade and submergence results in similar underwater light-compensation points
Acclimation to shade and submergence resulted in the same efficient use of light intensities, as low underwater light-compensation points were lower in shade and submergence-acclimated plants compared with HL drained plants (Fig. 4). The light-compensation point is determined by the amount of respiratory CO2 release in photorespiration and dark respiration as compared with CO2 fixation. For lower respiratory carbon losses, less CO2 fixation, and thus less light absorption, is required to reach net positive photosynthesis. Plants with lower CO2 compensation points are thus expected to show lower respiration rates, and that is what we observed for submergence-acclimated R. palustris HL plants. Respiratory costs were also lower in shade-acclimated plants of both species when measured in air (Table 1), but under water this could be detected only in shade-acclimated R. palustris plants. In addition, the decreased light-compensation points will also have resulted from increased light-harvesting capacity in shade-acclimated and submergence-acclimated plants compared with drained HL plants as acclimation reduced the chlorophyll a : b ratio.
Maximum underwater photosynthesis distinguishes shade and submergence
Underwater photosynthesis measurements at saturating light discriminated between acclimation responses to shade and submergence. Acclimation to submergence was far more beneficial than acclimation to shade for underwater photosynthesis, as submergence-acclimated plants showed higher maximum rates of underwater photosynthesis compared with drained plants of either light intensity (Fig. 3). Surprisingly, the maximum underwater photosynthesis rate of both HL- and LL-acclimated drained plants was similar (Fig. 3), whereas the aerial photosynthesis rates were much lower in the shade-treated plants (Table 1). Typically, shade-acclimated plants have reduced maximum photosynthesis rates in air (Table 1; Björkman, 1981; Givnish, 1988), and the fact that this was not apparent when photosynthesis was measured under water (Fig. 3) suggests that other components of shade acclimation can be beneficial for maximum photosynthesis rate under water.
The beneficial effect of shoot acclimation to submergence – and, to a lesser extent, acclimation to low light intensity – is likely to originate from a reduction in diffusion resistance to gas exchange between leaves of plants and their aquatic environment, analogous to aquatic plants (Nielsen, 1993; Frost-Christensen & Sand-Jensen, 1995). An increased SLA indicates that the relative surface area that is in contact with the water layer is larger. Under water, gases enter the plant via cuticle and epidermis cells, and thus increased gas-exchange area will increase underwater gas-exchange capacity. The drained shaded plants showed the highest SLA values, but this resulted in lower rather than higher maximum photosynthesis rates under water compared with submergence-acclimated plants. This suggests that acclimation to submergence involves more than simply the increased SLA observed in LL-treated drained plants. As the major limitation for gas diffusion between the plant and the water layer is likely to be the cuticle (Frost-Christensen et al., 2003), effective acclimation to submergence is probably caused by reduced cuticle resistance (Mommer et al., 2004).
Underwater photosynthesis at saturated CO2 does not explain flooding tolerance
Of the two terrestrial Rumex species that we investigated, the flood-tolerant R. palustris showed the highest phenotypic plasticity of the two species to both submergence and shade with respect to shoot morphological parameters (Figs 1, 2) and chlorophyll content (Table 2). It remains to be elucidated whether this higher plasticity of R. palustris leads to higher fitness under submerged conditions, because the difference in flooding tolerance with the less tolerant R. thyrsiflorus was associated with neither higher maximum underwater photosynthesis rates, nor lower light-compensation points or increased apparent quantum yields. Both species photosynthesized equally well in this experiment with unlimited supply of CO2. Earlier attempts to relate underwater photosynthesis to flooding tolerance also failed, as nonacclimated plants were investigated at high CO2 concentrations (Voesenek et al., 1993b; He et al., 1999; Vervuren et al., 1999). We therefore suggest that it is not the capacity for photosynthesis under water per se that determines survival under water, but photosynthetic performance under more natural conditions, where CO2 availability is limited and thus low gas-diffusion resistance becomes more important.
No additional shade acclimation under water
Low light conditions did not induce additional responses under water, with the notable exception of petiole elongation. For this parameter, significant differences were observed between the two light treatments under water. Most effects of submergence overruled the effect of light intensity, suggesting a hierarchy in effects. However, light acclimation can take place under water in real aquatic plant species with respect to SLA (Spence et al., 1973; Goldsborough & Kemp, 1988), and plant height or internode length (Spence & Chrystal, 1970; Chambers & Kalff, 1985). This suggests that the hierarchical effect of submergence over shade may depend on the tolerance of a species to submergence, as aquatic plant species have adapted to an aquatic life in such a way that gas exchange is not a stress factor (Lenssen et al., 2003).
Unexpectedly, the shade-acclimated submerged R. palustris plants showed the lowest maximum photosynthesis rate under water, and relatively high dark respiration rates (Fig. 3). A possible explanation, frequently proposed in aquatic photosynthesis studies, is that the rather high CO2 concentration in the underwater measurements may have had a depressing effect on the maximum photosynthesis rate (Weber et al., 1979; Allen & Spence, 1981; Denny et al., 1983; Pokornýet al., 1985), and may even have damaged cell components. As R. palustris was likely to attain the highest rate of CO2 influx of the two species, because it showed the strongest morphological responses (Figs 1, 2), high CO2 concentrations may specifically have affected maximum photosynthesis in this species. This may have been particularly relevant for the LL submerged plants, because these shaded plants will have experienced a relatively low energy status. The relatively high dark respiration rates may indicate higher costs allocated to repair mechanisms in these plants.
Our study shows that there is indeed a strong similarity between acclimation responses to shade and submergence. Maximum underwater photosynthesis rate, however, clearly discriminated between the functionality of the responses. This suggests that acclimation to submergence involves processes at least partly different from acclimation to shade, with acclimation to submergence being predominantly directed to increased gas-exchange capacity rather than to light capture. This study illustrates that morphological and biochemical responses to environmental cues need to be studied in depth before their physiological functionality can be understood.
It remains intriguing, however, that morphological responses are so convergent, whereas the functional consequences are clearly different. This indicates that initially different signals (shade vs submergence) may converge later in the signal transduction cascade to similar responses. At the same time, other responses are induced that are unique for shade or submergence, as reflected in the underwater maximum photosynthesis rates. We have a long way to go before we understand how low light and low CO2 availability induce at least partly the same morphology of a plant, but the expected low assimilate levels may be a link between both acclimation responses.
We wish to thank Ron Galiart for help with the measurements, and Thijs Pons for valuable comments on an earlier version of the manuscript.