Notice: Wiley Online Library will be unavailable on Saturday 27th February from 09:00-14:00 GMT / 04:00-09:00 EST / 17:00-22:00 SGT for essential maintenance. Apologies for the inconvenience.
• The capability to withstand and to recover from severe summer droughts is becoming an important issue for tree species in central Europe, as dry periods are predicted to occur more frequently over the coming decades.
• Changes in leaf gas exchange, chlorophyll a fluorescence and leaf compounds related to photoprotection were analysed in young Quercus pubescens trees under field conditions during two summers (2004 and 2005) of progressive drought and subsequent rewatering.
• Photochemistry was reversibly down-regulated and dissipation of excess energy was enhanced during the stress phase, while contents of leaf pigments and antioxidants were almost unaltered. Plant water status was restored immediately after rewatering. Net photosynthesis (Pn) measured at ambient CO2 recovered from inhibition by drought within 4 wk. Pn measured at elevated CO2– to overcome stomatal limitations – was restored after a few days.
• A network of photoprotective mechanisms acted in preserving the potential functionality of the photosynthetic apparatus during severe drought, leading to a rapid recovery of photosynthetic activity after rewatering. Thus, Q. pubescens seems to be capable of withstanding and surviving extreme drought events.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
In the context of future climate projections, increased probability of drought and heatwaves in central Europe as well as rising CO2 concentrations during the coming decades will affect vegetation structure and plant productivity (Schär et al., 2004; Ciais et al., 2005). Limited water availability in conjunction with high temperature and radiation are known to affect plant productivity and growth greately (Boyer, 1982). Thus, photosynthetic CO2 fixation is suppressed under drought stress by enhanced diffusive resistances within the leaf (closure of stomata and decline of mesophyll conductance) and by drought-induced impairments of metabolic processes (Flexas et al., 2004, 2006; Reddy et al., 2004). Whether stomatal or nonstomatal limitations play the major role in drought-induced inhibition of the photosynthetic activity has been a focus of debate in recent years, leading to a number of studies in this area field (Cornic & Fresneau, 2002; Flexas et al., 2004, 2006; Grassi & Magnani, 2005). Among these, some reports are in favor of stomatal limitations as the main drought-induced process, because maximum CO2 assimilation could be restored by applying high external CO2 or by removing leaf epidermis (Schwab et al., 1989; Epron & Dreyer, 1993b; Cornic & Fresneau, 2002). On the other hand, it has been documented in some cases that maximum photosynthesis of drought-stressed plants could not be restored completely by high CO2, suggesting increased nonstomatal limitations (Tezara et al., 1999; Lawlor, 2002). To date, a widely accepted theory posits that, under mild to moderate drought conditions, it is predominately diffusive resistances that limit photosynthesis, whereas under severe drought, when stomatal conductance (gs) drops below 0.1–0.05 mol H2O m−2 s−1, metabolic limitations become dominant (Flexas et al., 2004, 2006).
After stress, impaired photosynthetic electron flow results from an imbalance between the generation and utilization of photosynthetic electrons. Thus, excess excitation energy has to be safely removed, mainly by thermal dissipation (nonphotochemical quenching, NPQ), to prevent formation of reactive oxygen species (ROS), such as hydrogen peroxide (H2O2) and superoxide radicals (Szabo et al., 2005). Plants also possess several antioxidative defense mechanisms, for example the ascorbate-glutathione cycle, which operates mainly in the chloroplast (Mittler, 2002). Besides scavenging of H2O2 and the regeneration of NADP, ascorbate is also involved in the xanthophyll cycle, where it facilitates the conversion of violaxanthin to zeaxanthin. This cycle is known to act at the photosystem II (PSII) antenna, where it protects the photosynthetic apparatus from being photo-damaged during high light stress (Demmig-Adams & Adams, 1996). Besides that, carotenoids like xanthophylls and carotenes can also act as direct scavengers of ROS and stabilize the light-harvesting complexes within the thylakoid membranes (Kuhlbrandt et al., 1994; Bassi & Caffarri, 2000). α-Tocopherol, a lipophilic molecule abundantly present in the chloroplast, is also involved in the detoxification of ROS, correlating positively with high light and/or drought stress (Munne-Bosch, 2005), as well as with leaf aging (Hansen et al., 2003).
Apart from the great number of reports on plant responses to drought stress and on the mechanisms of photoprotection during stressful conditions, little is known about the capacity of recovery and the underlying processes during recovery from drought (Flexas et al., 2006). Up until now there have been only a few studies on recovery from drought stress, providing evidence of a dependency of the rate of recovery on the previously experienced stress (Kirschbaum, 1988; Souza et al., 2004; Cai et al., 2005; Miyashita et al., 2005). Furthermore, enhanced protection against oxidative stress, which may prevent irreversible damage, seems also to be relevant for withstanding severe stresses and for providing a rapid and complete recovery (Reddy et al., 2004).
Consequently, the understanding of how plants respond to increased drought in the field and the effect on productivity and growth may not be the only points of interest. Equally, the capacity and rate of recovery from drought stress are important aspects in terms of growth and survival in a rapidly changing climate (‘global warming, extreme events’). As an increased frequency of drought events is predicted for the coming decades, further investigations are needed to elucidate the mechanisms of recovery from drought in the field. This could also improve predictions of ecosystem productivity or better irrigation systems.
The following experiments were carried out on deciduous, young pubescent oaks (Quercus pubescens), which typically grow on rocky, shallow soils with a poor water-retaining capacity. These trees play an essential role in protecting slopes from erosion and are widespread in Switzerland on sunny south-facing slopes of the Wallis, Jura and Tessin regions. Changes in photosynthetic traits were studied during two consecutive summers under ongoing drought conditions and upon rewatering under field conditions. Previous reports on deciduous oaks during summer drought in the field have already described several aspects of drought-induced responses of the photosynthetic machinery, but were lacking information about the recovery phase (Epron & Dreyer, 1993a,b; Damesin & Rambal, 1995). The main focus of this study was on processes involved in the photosynthetic adaptation to stress, especially the reversibility of drought-induced down-regulation of the photosynthetic machinery after rewatering. Contributions of stomatal and nonstomatal limitations are discussed in this context. Thus, the following questions were addressed:
• How is drought-induced suppression of CO2 fixation reversed after rewatering?
• What are the processes involved in the capacity of drought-tolerance and of recovery in Q. pubescens?
Materials and Methods
Plant material and experimental setup
Six-year-old pubescent oaks (Quercus pubescens Willd.), approx. c. 1.6 m high, from the cantonal forestry nursery of Bern were planted in 50 l pots (diameter, 0.5 m) on 21 October 2003. The soil contained c. 50% coarse clay, 40% brown soil and 10% humus. Twenty-four plants were kept under natural conditions in a flat field in the Botanical Garden of Bern (46°57′08″N, 7°26′42″E, 499 m asl) for the following 2 yr. During the summers of 2004 and 2005 they were subjected to two different treatments, well watered (control saplings) and nonwatered (stressed saplings), arranged in a completely randomized design. Stressed and control saplings were identical in both years, imposing two extreme summer droughts on the same plants. To avoid shading by twigs and leaves of neighboring plants, a distance of 30 cm was maintained among pots, which represents a stem-to-stem-distance of 80 cm (diameter of crown, 70 cm). The treatments were started when leaves had been completely developed and matured. All oaks were well irrigated until the onset of the experiments and control plants were then continuously watered to field capacity every 2–3 d during the experimental period. Drought stress was induced by withholding water, and shielding from rainfall was achieved by covering the pots with plastic sheeting c. 10 cm above the soil. However, on days of increased humidity (heavy rainfall), the leaf status of stressed plants might have been slightly affected. In both years, stressed plants were kept without water until net photosynthesis was almost completely inhibited during the late morning and the early afternoon. In total, water was withheld for 49 d in 2004 (2 July to 20 August) and rewatered afterwards until leaf senescence in October. In 2005 the stressed saplings were kept without water for 36 d (13 June to 20 July) and then irrigated again until October. When rewatering was started, the pots were watered to field capacity. Throughout the experiment, predawn leaf water potential (ΨPD), leaf temperature, gas exchange, chlorophyll a (Chla) fluorescence and various leaf compounds were determined on southwest-orientated leaves from the upper crown of at least three randomly selected trees per treatment. Early senescing leaves, which were observed during severe drought on a few plants at the uppermost part of the tree crown, were not taken for data acquisition.
Plant water status and climatic data
Predawn leaf water potential, ΨPD, was determined between 06:00 and 07:00 h with a ‘Scholander pressure chamber’ (SKPM, Skye Instruments Ltd, Powys, UK). Changes in pot weight of stressed and control plants were monitored weekly. Lower initial pot weight in 2005 is presumably the result of changes in soil structure (compacting) during winter. During the measuring period solar irradiance was recorded frequently with a portable light meter LI-250 A (Li-Cor, Lincoln, NE, USA) and leaf temperature was measured with a handheld infrared thermometer (Oakton TempTestr IR, Cole-Palmer Int., Vernon Hills, IL, USA). Solar irradiance, air temperature and daily precipitation in 2004 and 2005 were recorded continuously at a meteorological station nearby (Bern-Liebefeld, MeteoSwiss, CH).
Leaf gas exchange and chlorophyll a fluorescence measurements
Photosynthetic traits were measured in situ on mature, southwest-orientated and fully sun-exposed leaves of the crown of three to five different trees per treatment (one leaf per plant) during the late morning (09:00–11:00 h) and early afternoon (13:00–15:00 h). Photon flux density (PFD) ranged between 1000 and 1800 µmol m−2 s−1 during the periods of measurements. Measurements of gas exchange and Chla fluorescence were carried out on the same leaf. Leaf gas exchange was recorded with an open infrared gas exchange analyzer (IRGA) system (CIRAS-1, PP-Systems, Hitchin, UK). Net CO2 assimilation rate (Pn), calculated stomatal conductance for water vapor (gs) and calculated intercellular CO2 concentration (Ci) were determined. External CO2 was provided by using a CO2 cartridge plugged into the CIRAS-1 in order to adjust and maintain the CO2 concentration in the leaf cuvette during the measurements (at 392 ± 5 ppm CO2 for ambient CO2 and at 1388 ± 12 ppm CO2 for high CO2). High CO2 was used as a tool to overcome stomatal limitations and was applied only during the gas exchange measurements. Temperature and relative humidity inside the leaf cuvette were always close to ambient air values.
Chlorophyll a fluorescence was recorded with two portable pulse amplitude modulated fluorometers, using the FMS-1 (Hansatech, King's Lynn, Norfolk, UK) in 2004 and the PAM-2000 (Heinz Walz GmbH, Effeltrich, Germany) in 2005. The following parameters were determined and calculated according to Maxwell & Johnson (2000): Fv/Fm, the maximum quantum efficiency of photosystem II; ΦPSII, quantum yield of PS II electron transport; qP, the proportion of open PSII reaction centers; and NPQ, the efficiency of heat dissipation. In 2004 ΦPSII, qP and NPQ were recorded under actinic light of 280 µmol photons m−2 s−1, provided by the light source of the FMS-1 fluorometer and applied for 10 min before recording. In 2005, ΦPSII and NPQ were recorded with the PAM-2000 under ambient light, using a leaf-clip holder 2030B (Heinz Walz). After recording the fluorescence parameters in the light-adapted state, Fv/Fm was determined on the same leaf after 20 min in the dark. In 2005 qP was not determined.
Leaf discs (diameter, 1.4 cm) were taken from sun-orientated leaves of five plants per treatment (i.e. stressed, control) before sunrise and around noon, immediately frozen in liquid nitrogen and stored at −80°C before analysis.
Ascorbate was analyzed via reversed-phase HPLC (Davey et al., 2003), using a HPLC System Gold (Beckmann Coulter, Fullerton, CA, USA). In brief, frozen leaf discs were ground to powder and ice-cold extraction buffer (6% metaphosphoric acid, 2 mm EDTA and 1% insoluble polyvinylpolypyrrolidone) was added. The obtained extract was centrifuged at 24 000 g (Sigma 3-18K, Osterode, Germany) for 12 min at 4°C. Supernatant was stored on ice and the remaining pellet was mixed with extraction buffer and centrifuged again. The unified supernatants were thereafter used for reversed-phase HPLC analysis. A 250 mm × 4.6 mm HYPERSIL (ODS, 5 µm) column was used for sample separation and a multiarray UV/VIS detector (195–300 nm) for peak determination (A243nm). Data acquisition and quantification were carried out with the 32 Karat Software 5.0 (Beckmann Coulter).
For the analysis of total soluble carbohydrates, frozen leaf discs were ground to powder in liquid nitrogen and mixed with chilled extraction buffer (20 mm sodium phosphate (pH 7.5), 1% (w/v) polyvinylpolypyrrolidone and 0.1% (v/v) β-mercaptoethanol). The extract was filtered through miracloth and centrifuged at 16 000 g for 2 min. The insoluble fraction was washed again two times with extraction buffer (without polyvinylpolypyrrolidone) and soluble fractions were then unified. Aliquots of the soluble leaf extract were used to determine the content of total soluble sugars, using the anthron reagent for quantification (Stieger & Feller, 1994).
Student's t-test analysis on independent samples of leaf gas exchange parameters (n = 3–5) and leaf compounds (n = 3–6) was performed, testing for significant differences between stressed and control oaks on each day measurements were taken.
Climatic conditions and water status
The climatic conditions during the experimental periods of 2004 and 2005 were typical of summer in the Swiss lowlands. The maximum PFD ranged between 1520 and 1940 µmol m−2 s−1 and the total precipitation from May to September was 457 mm in 2004 and 499 mm in 2005. Daily maximum air temperatures did not exceed 34°C (Fig. 1), while leaf temperatures of stressed and control leaves exceeded air temperatures by several degrees during most of the day, reaching values up to 43°C in stressed plants (Fig. 2). Differences in leaf temperature between control and stressed plants increased significantly with progression of drought, where differences up to 5°C were determined on clear and sunny days during the afternoon (Fig. 2c,d). Leaf temperature of stressed and control plants did not differ significantly after stressed plants had been rewatered for c. 1 wk.
During the imposed drought in 2004 and 2005, the loss of soil water was reflected in the progressive decline of pot weight (Fig. 3a,b). The minimal pot weight was reached after 7 wk in 2004 and after 3 wk in 2005. It then remained unchanged until the onset of rewatering. Well-watered plants (controls) showed almost no changes in pot weight throughout the whole experimental period (data not shown). According to the depletion of soil water, ΨPD decreased gradually with ongoing drought stress and dropped to values below −3 MPa after 3–6 wk in 2004 and after 2 wk in 2005 (Fig. 3c,d). High humidity (after rainfall during the night) led to a slight increase of ΨPD in stressed and control leaves (day 40, 2004; day 28, 2005), indicating a little absorption of water into the leaf. Control plants showed only slight changes in the ΨPD throughout the experimental period, ranging between −1.1 MPa and −0.2 MPa in both years. Besides an age-related increase of biomass in 2005, less rainfall and warmer climatic conditions after the start of the experiment in 2005 and/or changes in water-retaining capacity of the soil (lower initial pot weight) compared with 2004 might have contributed to an accelerated loss of water and decline of ΨPD (Figs 1, 3). After rewatering, soil and leaf water status were restored immediately in both years (Fig. 3).
Gas exchange parameters
The loss of soil and leaf water under progressive drought in both years was followed by a gradual decline in net CO2 assimilation (Pn) and stomatal conductance (gs), as shown in Figs 4–6. In parallel the internal CO2 (Ci) increased. In both years, the values of Pn, gs and Ci of well-watered plants ranged predominantly between 10 and 20 µmol CO2 m−2 s−1, 100–300 mmol H2O m−2 s−1 and 200–280 ppm, respectively. Stressed and control oaks often displayed lower Pn and gs values in the afternoon than in the morning (compare Figs 4 and 5 with Fig. 6), which was confirmed by gas exchange data from the course of the day, taken during the experiment (data not shown). Pn was almost completely inhibited after withholding water for 33 d and 14 d in 2004 and 2005, respectively. Even negative Pn values were measured in 2004 (day 33, Pn(am) =−0.13 ± 0.04 µmol m−2 s−1, Pn(pm) = −0.17 ± 0.03 µmol m−2 s−1) and 2005 (day 14, Pn(pm) = −0.14 ± 0.1 µmol m−2 s−1) during days of high solar irradiance and elevated temperatures, indicating a net CO2 production in the light. This was also supported by a higher leaf-internal (i.e. Ci) than ambient CO2 concentration. Increased external CO2 concentration during gas exchange measurements resulted in an almost twofold increase of Pn in watered plants (Figs 4, 5). In contrast, the exposition to high CO2 had little effect on the Pn values of stressed oaks under severe drought.
Pn and gs of previously stressed oaks recovered gradually after rewatering had started, while the concentration of Ci in previously stressed saplings was immediately restored to the control value. Pn increased by four times to c. 30% of the control values after 1 d of rewatering in 2005, to 60% after 4 d and to 100% after c. 3 wk. Comparable results were obtained for Pn in 2004, just lacking data of the first day of recovery. However, Pn was already completely restored to control values after 1 wk in both years, when high external CO2 was applied during gas exchange measurements. The recovery of gs was slightly retarded during the first 2 wk of rewatering compared with Pn, which resulted in an increased intrinsic water use efficiency for CO2 (compare Pn and gs in Figs 4–6). Finally, Pn and gs were restored to control values after 26 and 28 d of rewatering in 2004 (day 75) and 2005 (day 65), respectively.
In 2004 leaves were adapted to a PFD of 280 µmol m−2 s−1 before recording (Fig. 7), while the measurements in 2005 were taken under ambient irradiance (Fig. 8). By reducing the incident light intensity to 280 µmol m−2 s−1, it was intended to provide information about the reversibility of photo-inhibitory processes during the day under nonexcessive light in drought-stressed and control plants. In spite of the different light treatments, the maximum efficiency of PSII (Fv/Fm) and the quantum yield of PSII (ΦPSII) in the late morning showed a decline with progressive drought stress in both years (Figs 7, 8). Fv/Fm and ΦPSII in stressed leaves deviated from control values after 21 and 14 d of withholding water in 2004 and 2005, respectively. In general, ΦPSII was markedly lower under ambient light (2005) than under moderate light (2004), while the differences between control and stressed plants were highest during increased drought stress in both years. On the other hand, stressed and nonstressed oaks showed Fv/Fm ratios above 0.8 before sunrise during severe drought stress (Fig. 8a). Under moderate light in 2004, the ratio of open PSII reaction centers (qP) also declined with ongoing drought stress (Fig. 7). The NPQ values of drought-stressed saplings in 2004 ranged between 1.1 and 1.6, whereas the values in control plants ranged between 0.6 and 1.0 (Fig. 7d). Although slightly higher NPQ values were detected in drought-stressed than in control oaks under ambient irradiance in 2005, this difference was rather small in relation to the overall high NPQ values in both treatments (Fig. 8e,f). After 1 d of rewatering, Fv/Fm and ΦPSII were only partially restored to control values (Fig. 8a–d), while after 4 d a complete restoration of all fluorescence parameters was observed in both years (Figs 7, 8).
Pigments and antioxidants
The content of xanthophyll cycle pigments (violaxanthin, antheraxanthin and zeaxanthin, VAZ) slightly increased in drought-stressed compared with control leaves during the drought period in 2004 (Fig. 9e), while in 2005 almost no alteration in the VAZ-pool size was detected in stressed and control leaves (Fig. 9f). After rewatering, the differences in the VAZ-pool size of stressed and control plants leveled off to similar values after 1 d in 2005 and after c. 4 d in 2004. In 2004 and 2005, determination of the de-epoxidation state of the VAZ-pool (DEPS) at noon resulted in a continuously high DEPS ratio (90%) in stressed oaks during the drought periods, while control plants never exceeded 90% (Fig. 9c,d). DEPS ratios of control plants were linked to the PFDs around sampling time, while stressed plants did not show this linkage. When PFD had dropped before sampling, a marked reduction of DEPS on days 32 (2004) and 34 (2005) was determined, while only little effect on the DEPS in stressed saplings was observed (Fig. 9a–d). This phenomenon was also observed during the initial phase of rewatering (i.e. 2005), but leveled off after 4 d.
The content of chlorophyll a and b (Chla + b) ranged between 150 and 250 µmol m−2 in both years, with only minor differences between stressed and nonstressed oaks (Fig. 9g,h). The content of other photosynthesis-related pigments (i.e. β-carotene, neoxanthin, lutein) also remained stable in stressed and control plants during both years (data not shown).
Furthermore, the content of α-tocopherol increased slightly in all oak leaves over the experimental period, while differences between stressed and control leaves were almost negligible (Tables 1, 2). The leaf content of total ascorbate ranged between 0.43 and 0.92 mol m−2 in 2004 and 2005, showing little difference between comparable stressed and control saplings (Tables 1, 2). A similar pattern was observed for the ratio of reduced to total ascorbate, where almost constant ratios were detected in stressed and control oaks (70–90% during 2004 and 2005). The content of soluble carbohydrates was slightly higher in stressed than in control leaves during the drought period (Tables 1, 2), but a clear trend was missing because of the variations in stressed and control plants throughout the experimental periods.
Table 1. Leaf compounds of oak (Quercus pubescens) saplings determined during drought stress and rewatering (after 49 d) in 2004
Day of experiment (2004)
AsA + DHA, total ascorbate content (the sum of the reduced (AsA) and the oxidized (DHA) form of ascorbate); refer to text for other terms. Two treatments were applied: well watered (control) and nonwatered (drought), followed by rewatering after 49 d. Means and SE of at least three trees are shown. Significant differences between drought and control values are indicated (*, P = 0.05).
Table 2. Leaf compounds of oak (Quercus pubescens) saplings determined during drought stress and rewatering (after 37 d) in 2005
Day of experiment (2005)
AsA + DHA, total ascorbate content (the sum of the reduced (AsA) and the oxidized (DHA) form of ascorbate); refer to text for other terms. Two treatments were applied: well watered (control) and nonwatered (drought), subsequently followed by rewatering after 37 d. Means and SE of at least three trees are shown. No significant differences (P = 0.05) between stressed and control values were detected at each date.
nd, not determined.
α-tocopherol (µmol m−2)
51.8 ± 5.5
76.6 ± 2.0
118.2 ± 22.5
85.3 ± 2.0
117.0 ± 11.8
109.2 ± 11.1
127.3 ± 13.0
65.6 ± 5.0
90.0 ± 9.2
78.2 ± 17.3
98.3 ± 37.0
95.5 ± 16.1
100.3 ± 10.5
106.0 ± 3.8
AsA + DHA (mol m−2)
0.53 ± 0.13
0.78 ± 0.12
0.71 ± 0.05
0.70 ± 0.02
0.69 ± 0.06
0.62 ± 0.04
0.43 ± 0.02
0.59 ± 0.10
0.68 ± 0.06
0.65 ± 0.04
0.64 ± 0.10
0.66 ± 0.02
AsA/(AsA + DHA) (%)
80.1 ± 7.7
77.6 ± 3.6
75.2 ± 4.6
75.9 ± 1.0
80.0 ± 8.1
82.6 ± 7.2
69.5 ± 0.5
77.6 ± 3.2
67.7 ± 0.8
81.6 ± 5.7
90.7 ± 2.6
66.2 ± 2.9
Soluble carbohydrates (mmol m−2)
0.32 ± 0.04
0.39 ± 0.07
0.30 ± 0.04
0.24 ± 0.03
0.25 ± 0.02
0.21 ± 0.01
0.33 ± 0.05
0.43 ± 0.16
0.33 ± 0.04
0.40 ± 0.07
0.31 ± 0.07
0.23 ± 0.01
0.26 ± 0.05
0.37 ± 0.02
During the summer of 2004 and 2005, net CO2 fixation (Pn) of watered Q. pubescens saplings was similar to that previously reported for oak trees in the field, ranging between 10 and 20 µmol CO2 m−2 s−1 (Epron & Dreyer, 1993b; Damesin & Rambal, 1995). Pn of pubescent oak declined in parallel with gs after drought stress, following the course of soil and leaf water depletion (pot weight, ΨPD) during the drought periods in both years. As Pn decreased in parallel with gs, stomatal limitations seemed to account mainly for this reduction of photosynthesis. Stomatal closure protects against further water loss and irreversible cell dehydration under progressing drought.
On the other hand, exposure of stressed leaves to CO2-enriched air in order to overcome stomatal limitations for CO2 entry had only little effect on Pn (Fig. 4). This observation suggests increased limitations to photochemistry by nonstomatal processes during progression of drought. Other reports also support this view, as maximum photosynthetic capacity in several C3 plants could not be restored when gs dropped below 100–50 mmol H2O m−2 s−1 (Flexas et al., 2004, 2006) and/or when Pn was reduced by > 80% (Cornic & Fresneau, 2002).
A possible contribution of elevated leaf temperatures (> 40°C) to the inhibition of photosynthesis in the stressed oaks might be assumed as well, because the Pn values were lowest (Pn < 0) when leaf temperatures were highest (Figs 2, 4, 6). In a recent study by Haldimann & Feller (2004), short-term exposure of Q. pubescens leaves to elevated temperatures up to 45°C did not cause irreversible photo-damage (PSII inactivation was reversible). Although the duration of elevated leaf temperatures might be different in the present study, heat-induced damages seemed to be little, as leaf temperature did not exceed 43°C throughout the experimental periods (Fig. 2) and PSII seemed to be well protected (see below). Moreover, there is also evidence that the combination of drought and heat stress might even improve thermal stability of the thylakoid membranes (Havaux, 1992; Ghouil et al., 2003).
Removal of excess excitation energy via thermal dissipation increased during drought-induced suppression of Pn, as indicated by higher NPQ values in stressed than in control plants and a continuously high de-epoxidation state (DEPS > 90%) of abundantly present xanthophyll cycle pigments (VAZ). In parallel, the utilization of excitation energy for photochemistry was down-regulated during the day, as shown in a strong but not complete suppression of PSII activity (ΦPSII, Fv/Fm) in stressed oaks (Figs 7, 8). Down-regulation of PSII activity and increased NPQ in severely stressed oak leaves during the day was only partially reversed after removal of excessive light (Fig. 7). Furthermore, ratios of DEPS remained constantly high in stressed oak leaves in spite of changes in solar irradiance, while control plants adjusted their DEPS quickly to these changes (Fig. 9). These results suggest a persisting impairment of photochemistry during severe drought, at least partially because of enhanced oxidative stress and inactive PSII units. Nevertheless, the potential functionality of the photosynthetic apparatus seemed to be preserved during almost complete inhibition of Pn, as indicated by the complete restoration of Fv/Fm overnight (Fig. 8). In previous studies on Quercus species, a reversible down-regulation of maximum PSII activity during the day has been reported under mild to moderate summer drought, when Pn values remained above zero (Epron & Dreyer, 1993a; Damesin & Rambal, 1995). A possible destabilization of PSII might have been counteracted by an accelerated turnover of the D1 protein and reorganization of the PSII, as proposed for pea plants under drought (Giardi et al., 1996).
Damage to the photosynthetic apparatus was also little or rapidly repaired, as reflected in maintained contents and composition of the pigments of the photosynthetic apparatus (Chla+b and associated carotenoids, e.g. β-carotene, lutein and neoxanthin) in stressed and control oaks. Changes in the size of the VAZ-pool were also small (Fig. 9), while a maintained high reduction potential and high amount of antioxidants (ascorbate, α-tocopherol) suggest minimized oxidative stress and irreversible damage in oak leaves under drought conditions (Tables 1, 2). These observations are in contrast to other studies, where loss of Chl was observed during drought and a possible photoprotective function was assumed by reducing PS II antenna size and an increased VAZ/Chl ratio (Faria et al., 1998; Munne-Bosch & Alegre, 2000).
Changes in soluble carbohydrate content were only small in pubescent oak leaves under drought conditions, suggesting no or only a minor effect on osmotic adjustment. However, there are also other ways in which plants adjust osmotic potential during drought (Reddy et al., 2004).
Despite these considerations, a water regulative function by maintaining leaf water potentials above a certain threshold might be assumed for the oaks of this study (Fig. 3), as this has been described as a characteristic of Quercus species under drought conditions (Rambal, 1992; Damesin & Rambal, 1995). Leaf water potentials of stressed oaks remained above the critical value of −4 MPa for loss of turgor (Kyriakopoulos & Larcher, 1976; Hinckley et al., 1983). Although some leaves at the tree crown of a few plants showed early senescence during severe drought, a loss of hydraulic conductivity by cavitation or embolisms within the xylem seemed to be of minor relevance, because leaf water status was restored immediately after rewatering.
During the rewatering phase, photosynthetic performance recovered gradually. Initially, lasting nonstomatal limitations seemed to prevail, because Pn, gs and PSII activity (Fv/Fm, ΦPSII) of previously stressed oaks recovered only partially, while leaf water status and internal CO2 were restored within a day of rewatering. Applying high CO2 concentrations to leaves of stressed oaks during the initial rewatering phase resulted in a substantial increase in Pn, but restoration to control values was still not complete (Fig. 5). Besides metabolic impairments, it is equally likely that the concentration of CO2 was not high enough to overcome diffusive resistances within the leaf, suggesting a possibly lower mesophyll conductance (gm) or chloroplast conductance. In fact, gm was shown to be finite and to differ considerably between species, whereas it is still not clear if changes in gm and gs occur at the same rate (Loreto et al., 1992; Centritto et al., 2003; Warren & Adams, 2006). A clear distinction between metabolic and diffusive impairments at the mesophyll/chloroplast level during the delayed recovery of photosynthesis cannot be made here, but the term ‘nonstomatal limitations’ refers to both aspects.
A delayed response of the xanthophyll cycle (i.e. DEPS) to changing light conditions, as already observed during severe drought, was still present during the first days of rewatering (Fig. 9). On the other hand, overnight restoration of down-regulated PSII activity (Fv/Fm, Fig. 8) indicated reversibility of midday photo-inhibition, implying a preserved functionality of the photosynthetic apparatus or at least rapid repairing mechanisms. Moreover, as also observed during drought, maintained amounts of photosynthetic pigments and antioxidants (i.e. ascorbate, α-tocopherol) among comparable leaves of stressed and control plants emphasize the high capacity of Q. pubescens to protect and preserve its photosynthetic apparatus during the rewatering phase.
Whether an end-product inhibition of photosynthesis resulting from an elevated pool of soluble carbohydrates, as suggested by Souza et al. (2004), contributed to the delayed recovery of photochemistry of stressed oaks during rewatering remains unanswered, because data during the first days of rewatering are insufficient (Tables 1, 2).
Photosynthesis reacted faster than stomata upon rewatering after drought-induced suppression, which resulted in increased intrinsic water use efficiency for net CO2 assimilation (Pn : gs ratio). Such observations have also been documented for rehydrated bean plants after exposure to different degrees of drought stress (Miyashita et al., 2005), whereas in other studies such an increase in Pn : gs ratio during rewatering was not detected (Cai et al., 2005; Ennahli & Earl, 2005). The reason for this difference in the recovery of Pn and gs might be species-specific or stress-specific and requires further investigations. Here it might be speculated that physiological adjustments at the leaf level to minimize water loss under stressful conditions remained during rewatering in order to prevent future stress. Also drought-induced changes in leaf morphology or in biomass allocation (e.g. root : shoot ratio), especially in the context of (long-term) adaptation to drought, have been quoted by others (Niinemets & Kull, 1998; Breda et al., 2006).
Furthermore there are indications that intensity and duration of stress affect the velocity of recovery after the relief of stress (Miyashita et al., 2005; Flexas et al., 2006). Although differences in the velocity and duration of drought stress have been observed in 2004 and 2005, the recovery pattern of previously stressed oaks was almost the same in both years. This might be the result of the maximal degree of drought stress that was reached in both years (Pn∼ 0), which then led to a similar response after rewatering.
In summary, a similar pattern of photosynthetic adjustments during drought and rewatering cycles was observed in Q. pubescens saplings, although the velocity of drought progression and the climatic conditions differed between the experimental periods of summer 2004 and 2005. Thermal dissipation of excess energy and electron transfer to acceptors other than CO2 seemed to be important factors in preserving the potential functionality of the photosynthetic apparatus in Q. pubescens saplings during severe stress. Additionally, a comprehensive ROS scavenging system, including xanthophylls, α-tocopherol and ascorbate, was presumably acting sufficiently in cellular detoxification of ROS and preservation of integer photosynthesizing membranes. Therefore, the maximum efficiency of the photosynthetic machinery (i.e. PSII activity; Pn under high CO2) was quickly restored within a few days after the relief of drought stress, while recovery of field Pn was delayed by stomatal limitations.
Besides long-term adaptations on the whole plant level, which were not addressed in this study, it can be concluded that a tight network of photoprotective mechanisms minimize stress-induced damages to the photosynthetic apparatus in two consecutive extreme summer droughts. Thus, a rapid recovery of photosynthetic performance was achieved, even after an almost complete suppression of Pn and very low leaf water potentials during several weeks. This characterizes and underlines the capacity of Q. pubescens to withstand and survive extreme summer droughts.
The authors thank Prof. Dr P. Jahns and his team at the Institute of Plant Biochemistry, University of Duesseldorf (Germany) for making their HPLC system available for analysis and their helpful support. This work was supported by the NCCR climate research program of the Swiss National Science Foundation.