INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS AND DISCUSSION MATERIALS AND METHODS LITERATURE CITED

The thinning of the stratospheric ozone layer and the resulting increase in solar UV-B radiation (280-320 nm) at the earth's surface have led to major research efforts in studying the effects of UV-B on photosynthetic organisms. Based on a survey of a large number of long-term field studies of vascular plants published between 1976 and 1999, Searles et al. (2001) concluded that artificially elevated UV-B induced an increase in content of UV-B-absorbing compounds, but had little or no effect on morphological parameters and leaf photosynthesis as measured by means of gas exchange and chlorophyll (Chl) fluorescence. More recent studies also did not show effects of substantially elevated UV-B on photosynthetic performance of several plant species (Nogués and Baker, 2000; Lud et al., 2001; Bassman et al., 2002). It appears that in many cases, plants are capable of acclimating and, thereby, achieve efficient protection against increased UV-B levels (for review, see Allen et al., 1998). In leaves of tropical tree seedlings grown in simulated, differently sized tree-fall gaps, the level of UV screening substances was positively related to the radiation dose determined by the duration of daily sun exposure (Krause et al., 2001). However, when the protective mechanisms are overtaxed, detrimental effects of elevated UV-B radiation on photosynthesis do occur (for review, see Teramura and Ziska, 1996). In particular, photosystem (PS) II has been found to be UV-B-sensitive, whereas PSI appeared to be unaffected by UV-B. Inhibition of photosynthetic CO₂ assimilation and, specifically, effects on the activity, synthesis, and degradation of Rubisco were seen in a number of studies (Strid et al., 1990; Huang et al., 1993; Nogués and Baker, 1995; Rao et al., 1995; Allen et al., 1997; Bassman et al., 2001; Keiller and Holmes, 2001; Takeuchi et al., 2002).

In most investigations, artificial UV-B supplemental to the ambient solar UV-B level was applied. Fewer studies have been designed to evaluate the effects of present levels of solar UV-B radiation on photosynthesis. With suitable optical filters, one can test whether the ambient UV-B (and UV-A) contributes to photoinhibition of photosynthesis caused by exposure of plant leaves to full sunlight. In particular, the high UV-B flux within tropical latitudes may adversely affect photosynthesis (Madronich et al., 1995; Ziska, 1996). When shade-acclimated tropical tree seedlings were suddenly exposed for short periods (15-75 min) to direct sunlight, as may occur when tree-fall gaps open in the forest, PSII was found to be sensitive to ambient UV-B (and UV-A) radiation (Krause et al., 1998, 1999). In sun leaves from the outer canopy of mature tropical trees, UV-B effects on PSII were also observed in certain cases, depending on acclimation and developmental stage of the leaves (Krause et al., 1999). A number of studies have shown that in aquatic organisms, solar UV-B may inhibit photosynthetic reactions, predominantly those of PSII (e.g. Herrmann et al., 1997; Figueroa and Gómez, 2001; Häder et al., 2001). In a study of shade-grown grape (Vitis vinifera) plants, performed at mid-latitude (49°N), leaves responded to the UV-B component of full sunlight by enhanced inhibition of potential PSII efficiency and, perhaps independently, of CO₂ assimilation (Kolb et al., 2001). In Antarctic vascular plants studied during periods of ozone depletion, ambient UV-B did not reduce the rates of photosynthetic O₂ evolution per unit leaf area, but rates were diminished on a Chl or dry mass basis (Xiong and Day, 2001). The authors concluded from fluorescence analyses that, independent of PSII limitation, inhibition of photosynthetic enzyme reactions had occurred in the upper mesophyll of the leaves. It should be noted that solar UV-B has also been shown to cause DNA damage in photosynthetic organisms (Ballaré et al., 2001; Buma et al., 2001).

The present study aims to clarify whether the high solar UV-B of tropical latitudes causes a primary inactivating effect on photosynthetic CO₂ assimilation independent of (or in parallel with) inactivation of PSII. In addition, we investigated the response of PSI to ambient UV-B. Under controlled conditions, exposure of shade leaves of seedlings and sun leaves of mature tropical trees to excessive visible light caused a decrease in the efficiency of the photooxidation of the PSI reaction center pigment, P700, by far-red light, whereas the photochemical capacity of PSI remained largely unaffected (Barth et al., 2001). The decline in the efficiency of the P700 photoreaction is probably a universal response of PSI to high-light stress and may be caused by enhanced charge recombination in the PSI reaction center that confers photoprotection (Barth et al., 2001). Thus far, it is unknown whether such alteration of PSI photochemistry is also induced by the UV-B component of solar radiation. Therefore, in the present investigation, the effects of solar UV-B on PSI, as well as on the rate of net CO₂ assimilation in shade-acclimated tree seedlings, were assessed and compared with the effects on PSII.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS AND DISCUSSION MATERIALS AND METHODS LITERATURE CITED

CO₂ Assimilation in Comparison with PSII

When shade-acclimated leaves of tropical tree seedlings were exposed around midday for short periods to direct sunlight, subsequently measured net CO₂ assimilation rates were substantially reduced. Figure 1 shows data from typical experiments obtained prior to high light treatments (controls) and from the same leaves after exposure to direct sunlight in the presence or absence of UV-B. In leaves of A. excelsum (Bertero and Balb.) Skeels (Fig. 1A), a significant UV-B effect on net CO₂ uptake can be seen at intermediate light intensities. Similar patterns were observed in leaves of V. surinamensis (Rol.) Warb. (Fig. 1B) and C. longifolium Willd. (Fig. 1C), although the UV-B effect was not significant. There was no marked difference in the overall response of CO₂ assimilation between A. excelsum, a pioneer tree, and the late successional species, V. surinamensis and C. longifolium. Seedlings of the three species are known to be shade tolerant and had been acclimated to the same shade conditions. In agreement with the literature (e.g. Bassman et al., 2001), no effect of sun exposure on rates of dark respiration was detectable for any of these species (Fig. 1, A-C).

View larger version (21K): [in this window] [in a new window]   Figure 1.   Effects of sun exposure of shade-acclimated seedlings of tropical trees on net CO2 assimilation. Rates are depicted as a function of PAR. Leaves of one pioneer species (Anacardium excelsum) and two late successional species (Virola surinamensis and Calophyllum longifolium) were examined before () and after the plants were exposed for short periods to direct sunlight in the presence () and absence () of ambient UV-B radiation. Experimental conditions for A, B, and C, respectively, are exposure time, 34, 24, and 45 min; PAR dose, 3.0, 2.5, and 2.8 mol m2; and UV-B dose, 3.9, 3.1, and 4.0 kJ m2. Means and SD are presented; controls: n = 4 (A), n = 6 (B and C); sun-exposed plants, n = 3.

A strong tendency to enhanced photoinhibition of CO₂ assimilation in the presence of solar UV-B was observed in all experiments conducted as part of the present investigation, although leaf responses varied widely. To evaluate all CO₂ assimilation and PSII measurements concurrently, data from individual leaves were ranked according to the degree of remaining activity determined after sun exposure in the presence and absence of UV-B. The resulting graphs (Figs. 2 and 3) clearly demonstrate the enhancing effect of ambient UV-B on photoinhibition of net CO₂ assimilation and of PSII. Figure 2 depicts CO₂ uptake rates from eight independent experiments (four with V. surinamensis and two each with A. excelsum and C. longifolium). The top panels show the control rates measured prior to sun exposure at 20 µmol m² s¹ PAR (Fig. 2A) and near-light saturation at 350 µmol m² s¹ (Fig. 2B). The bottom panels show the normalized rates after sun exposure (Fig. 2, C and D). In general, CO₂ uptake rates of leaves that had been shielded from UV-B were substantially greater than those from leaves exposed in the presence of ambient UV-B radiation. Only in the extreme case where solar visible plus UV-A light had lowered CO₂ uptake close to the compensation point, little additional UV-B effect was apparent (Fig. 2, C and D, right-hand parts).

View larger version (33K): [in this window] [in a new window]   Figure 2.   Net CO2 uptake of shade-acclimated tree seedlings measured at 20 µmol m2 s1 (A and C) and 350 µmol m2 s1 photosynthetically active radiation (PAR; B and D) before sun exposure (controls, A and B) and after exposure in the presence and absence of ambient UV-B radiation (C and D). Rates recorded subsequent to sun exposure have been normalized based on the respective control values. Negative values indicate rates below the compensation point. Data are from eight independent sun exposure experiments using seedlings from three tree species. The mean value obtained at 20 µmol m2 s1 PAR was significantly lower after sun exposure in the presence than in the absence of UV-B (P < 0.01). Data points representing individual leaves from different plants have been ordered according to increasing degree of inhibition of CO2 assimilation caused by the preceding exposure to direct sunlight. For experimental details of sun exposure see "Materials and Methods." Circles, A. excelsum; triangles, V. surinamensis; squares, C. longifolium; open symbols, exposure in the absence of UV-B; closed symbols, exposure in the presence of UV-B.

View larger version (17K): [in this window] [in a new window]   Figure 3.   Potential PSII efficiency (FV/FM) in leaves of shade-acclimated tree seedlings after sun exposure in the presence and absence of ambient UV-B radiation. FV/FM ratios were recorded 15 to 20 min (A) and about 5 h (B) subsequent to sun exposure and are given in the percentage of the control values of the respective leaves determined prior to exposure. Open symbols, absence of UV-B; closed symbols, presence of UV-B. Symbols for plant species as for Figure 2. The data points were ranked from lowest to highest degree of PSII photoinhibition and represent 96 individual leaves from a total of 48 plants (two leaves per plant). Control values of FV/FM (means ± SD) were: A. excelsum, 0.805 ± 0.001 (n = 24); V. surinamensis, 0.779 ± 0.008 (n = 48); and C. longifolium, 0.799 ± 0.010 (n = 24). Mean values of FV/FM were significantly more reduced by sun exposure in the presence than in the absence of UV-B (15 min after exposure, P < 0.05; 5 h after exposure, P < 0.01).

The response of PSII to sun exposure in the shade-acclimated tree seedlings is shown in Figure 3. In the control leaves prior to treatments, maximum photochemical efficiency of PSII in the dark-adapted state (F_V/F_M) ratios were very uniform (see legend to Fig. 3), but varied strongly after sun exposure. Data were obtained from the same leaves (and one additional leaf of each plant) used for CO₂ uptake measurements (compare with Fig. 2). All data points from UV-B-shielded leaves indicated a higher potential PSII efficiency than in leaves exposed in the presence of UV-B (Fig. 3, A and B). The UV-B effect was most distinct when the degree of photoinhibition was intermediate. The comparison of F_V/F_M recorded 15 to 20 min (Fig. 3A) and about 5 h (Fig. 3B) subsequent to sun exposure indicated a greater UV-B effect after the 5-h period in low light. This resulted from lack of recovery of potential PSII efficiency of the UV-B-exposed leaves, whereas after 5 h, onset of recovery was seen in leaves pre-exposed in the absence of UV-B. The fluorescence data confirm results of an earlier study (Krause et al., 1999). A comparison of Figures 2 and 3 shows that inhibition of net CO₂ assimilation did not correlate with the decline of F_V/F_M ratios. Measurement in limiting light (20 µmol photons m² s¹) revealed the strongest UV-B effect on net CO₂ uptake in the data region where F_V/F_M was relatively mildly influenced by UV-B (compare left-hand parts of Fig. 2C and 3, A and B). This may be interpreted as a reduction in quantum yield of CO₂ assimilation (_P), whereas electron transport activity is presumably only slightly inhibited. Moreover, the CO₂ data obtained under saturating light (Fig. 2D) exhibited a steeper decline than the F_V/F_M data (Fig. 3). Thus, in the presence and (to a lesser extent) in the absence of UV-B, a photoinhibition of CO₂ assimilation obviously occurred independently of the partial inactivation of the electron transport system. These results were obtained when shade-acclimated leaves were exposed to full sunlight for only short periods (less than 1 h), and they demonstrate a direct impact of solar UV-B radiation on photosynthetic parameters. The only other investigation that shows an effect of solar UV-B on rates of net CO₂ assimilation per unit leaf area was that of Kolb et al. (2001). In that study with shade-acclimated grape leaves, the UV-B effect was seen only after several days of exposure to full sunlight, whereas in our case, the UV-B-induced reduction in CO₂ uptake was instantaneous.

The above conclusion is supported by the plots of _P (obtained under strictly limiting light) versus F_V/F_M (Fig. 4). The data, recorded after the plants had been exposed to full solar radiation (Fig. 4A) or in the absence of UV-B (Fig. 4B), were expressed relative to control values measured prior to exposure. Slopes of calculated regression lines were steep and could not be extrapolated to the origin or to the pair of control values (100%). Correlation coefficients were low, particularly in the UV-B-exposed leaves (see legend to Fig. 4). In total, the _P values depicted in Figure 4A (presence of UV-B) were significantly lower (P < 0.05) than in Figure 4B (minus UV-B). In the presence of UV-B, _P was more severely inhibited than F_V/F_M (P < 0.01). This difference was also seen, but was less distinct (P < 0.05), in the absence of UV-B (see legend to Fig. 4). When rates of CO₂ uptake were plotted versus F_V/F_M, pictures similar to Figure 4 were obtained (data not shown). These results contrast with those of studies using artificial visible light, where a close empirical correlation between optimal quantum yield of photosynthesis and F_V/F_M (at room temperature) was observed, with regression lines extending to the origin (Leverenz and Öquist, 1987; Krause and Somersalo, 1989; Giersch and Krause, 1991; Krause and Weis, 1991). Figure 4 indicates a large increase in the degree of inhibition of CO₂ assimilation when only small changes in potential PSII efficiency occur.

View larger version (18K): [in this window] [in a new window]   Figure 4.   Quantum yield of CO2 uptake (P) depicted as function of FV/FM. Data were obtained from shade-acclimated leaves of tree seedlings after short-term sun exposure in the presence (A) and absence (B) of ambient UV-B. Relative values of P were calculated from the slope of the linear part of the light-saturation curves of CO2 exchange (compare with Fig. 1) and are presented in the percentage of the respective values determined prior to sun exposure. Symbols for plant species as for Figure 2. Control values of P (means ± SD) were: A. excelsum, 0.059 ± 0.005 (n = 4); V. surinamensis, 0.056 ± 0.006 (n = 6); and C. longifolium, 0.061 ± 0.009 (n = 8). Correlation coefficients (r2) of the regression lines were: A, 0.55; B, 0.72. Means of data determined subsequent to sun exposure were: P = 35.1% ± 15.6%, FV/FM = 50.6% ± 10.7% (A); P = 49.4% ± 17.6%, FV/FM = 63.4% ± 7.6% (B). According to t tests, the difference between A and B was significant for P (P < 0.01) and FV/FM (P < 0.01). Means of P were significantly lower than means of FV/FM in A (P < 0.01) and B (P < 0.05).

The gas exchange analysis (measured transpiration rates and calculated intercellular CO₂ concentration) did not indicate increased stomatal limitation of CO₂ uptake in response to light stress (data not shown). Thus, the inhibition of CO₂ assimilation presumably was caused by restriction of enzymatic reactions in the chloroplasts. Damage to enzyme proteins such as Rubisco (see "Introduction") could result from active oxygen species formed under UV light (Hideg et al., 2000). The absence of stomatal limitation and loss of Rubisco was observed in greenhouse-grown oilseed rape (Brassica napus) exposed to high supplemental UV-B (Allen et al., 1997). Moreover, active oxygen species might affect the complex thiol-mediated light activation system of the Calvin cycle. Reductions in light activation could, to a lesser extent, also be caused by solar UV-A and even by visible light. Strong white light was reported to inhibit CO₂ assimilation in isolated intact chloroplasts, which was related to an accumulation of Fru-1.6-bisphosphate and a decrease in the ribulose-1.5-bisphosphate pool size (Giersch and Robinson, 1987). An inhibition of light activation of the Calvin cycle that is partly overcome by saturating light could explain why the UV-B effect on CO₂ assimilation was strongest when CO₂ uptake was measured in limiting light (Fig. 2C).

Response of PSI to Solar UV-B Radiation

Exposure of shade-grown leaves to natural sunlight led to a decrease in the efficiency of P700 photooxidation (Figs. 5 and 6), similar to results from previous experiments with artificial white light (Barth et al., 2001). The alteration in the efficiency of P700 oxidation was strongly enhanced by ambient UV-B radiation. Figure 5 demonstrates the dependence of P700 oxidation on the intensity of far-red light for an experiment with seedlings of A. excelsum. Exposure of leaves to direct sunlight altered the slopes of the light-response curves, indicating a reduction in the efficiency of P700 oxidation, which was substantially more prominent after sun exposure in the presence than in the absence of ambient UV-B.

View larger version (25K): [in this window] [in a new window]   Figure 5.   P700 oxidation as a function of far-red light intensity measured as absorbance change at 810 nm in shade-acclimated leaves of A. excelsum before and after exposure to direct sunlight. Exposure time, 18 min; PAR dose, 2.2 mol m2; UV-B dose, 2.6 kJ m2. Means and SD are given. , control values prior to exposure (n = 6); , absence of UV-B (n = 3); , presence of ambient UV-B (n = 3).

View larger version (18K): [in this window] [in a new window]   Figure 6.   Increase of KS of P700 oxidation under far-red light caused by exposure of shade-grown tree seedlings to direct sunlight in the presence (closed symbols) and absence (open symbols) of ambient UV-B radiation. The relative increase in KS is expressed as the ratio of KS values determined after sun exposure to the corresponding control values prior to exposure. Ratios obtained from individual leaves were ranked from low to high values. Four pairs of extremely high KS values ranging from 14- to 51-fold (presence of UV-B) and 6.5- to 11-fold increase (absence of UV-B) have been omitted. Symbols for plant species: circles, A. excelsum; triangles, V. surinamensis; and squares, C. longifolium. Control KS values (determined prior to sun exposure) were (W m2): A. excelsum, 0.3 ± 0.1 (n = 12); V. surinamensis, 0.4 ± 0.2 (n = 24); and C. longifolium, 0.4 ± 0.1 (n = 6). The mean value of KS increase was significantly higher after sun exposure in the presence than in the absence of UV-B (P < 0.01).

In Figure 6, the sunlight-induced decrease in the efficiency of P700 oxidation under far-red light is expressed as an increase in the saturation constant, K_S, i.e. the far-red intensity required for half-maximal P700 absorbance change in the steady state (compare with Barth et al., 2001). The data were obtained from eight experiments in which the response of CO₂ assimilation and PSII were also investigated (compare with Figs. 2-4). The relative increases in the values of K_S (with reference to control measurements prior to sun exposure) have been arranged from lowest to highest. The resulting graph (Fig. 6) indicates a much stronger increase in K_S induced by exposure to full sunlight than in the absence of UV-B. The UV-B effect is strongest in the region where a substantial increase in K_S is seen also in leaves exposed to visible plus UV-A light (right-hand part of Fig. 6). Highest K_S values (not shown) reached a nearly 50-fold increase in the presence, and an 11-fold increase in the absence, of ambient UV-B. There was a negative correlation between K_S values and F_V/F_M, i.e. K_S tended to increase when the potential PSII efficiency decreased due to light stress (data not shown), but recovery studies (Barth et al., 2001) indicated that changes in K_S and F_V/F_M occur independently.

As in the laboratory study by Barth et al. (2001), exposure to sunlight did not significantly affect the photochemical capacity of PSI represented by the maximum P700 absorbance change, A_{810 max} (Fig. 5). Thus, the increase in K_S does not result from a population of fully inactive PSI, but might be related to changes in the function of the FeS centers that delay electron transfer to ferredoxin. Such effect may facilitate recombination of radical pairs and, thereby, dissipation of excessive photon energy (for discussion, see Barth et al., 2001). The FeS centers of PSI are known to be the primary targets of PSI photoinhibition (Sonoike et al., 1995; Tjus et al., 1999). Charge recombination in the PSI reaction center has been reviewed by Brettel (1997) and studied by Teicher et al. (2000) in thylakoid membranes from barley (Hordeum vulgare) leaves subjected to illumination at chilling temperatures. Recombination in PSI was also suggested to occur in leaves of cucumber (Cucumis sativus) in response to chilling under low visible light (Kim et al., 2001). Enhanced recombination manifested by an increase in K_S, as observed here, may indicate an incipient inactivation of PSI that prevents severe damage to the reaction center.

The present investigation shows that exposure of shade-acclimated leaves to direct sunlight, simulating the opening of a tree-fall gap in the tropical forest, strongly affects quantum yield and light-saturated rate of net CO₂ assimilation. It is apparent that this occurs independently of a simultaneous photoinhibition of PSII. The solar UV-B component of sunlight contributes substantially to these effects. Moreover, solar UV-B promotes a decrease in the efficiency of P700 oxidation by far-red light that might be caused by accelerated charge recombination in the PSI reaction center and probably represents a photoprotective response.

Our data on short-term responses to high solar UV-B radiation do not contradict previous conclusions by Searles et al. (2001) that, on average, photosynthesis is not significantly affected in terrestrial plants acclimated to elevated levels of UV-B, which simulate a realistic stratospheric ozone depletion. However, when drastic changes in light conditions occur, e.g. upon formation of a gap in the forest as simulated in the present study, a UV-B dosage not previously experienced by the leaves evidently does affect the photosynthetic apparatus.

The changes in photosynthetic activities reported here were slowly reversible. Daily sun exposure with photon dosages similar (or higher) to those applied here led to an acclimative adjustment of photosynthetic pigment composition and to an increase in the level of UV-absorbing compounds in seedlings of the three species studied (Krause et al., 1999, and G.H. Krause, E. Grube, O.Y. Koroleva, and K. Winter, unpublished data).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS AND DISCUSSION MATERIALS AND METHODS LITERATURE CITED

Experiments were performed at the Tupper Center, Smithsonian Tropical Research Institute, in Panama City (9°N, 49°W).

Plant Material

Seedlings of Anacardium excelsum (Bertero and Balb.) Skeels (Anacardiaceae), Virola surinamensis (Rol.) Warb. (Myristicaceae), and Calophyllum longifolium Willd. (Clusiaceae; nomenclature according to Croat, 1978) were cultivated in a shaded greenhouse (neutral shade, PAR 10-60 µmol m² s¹) in pots (40 cm high, 15 × 15-cm wide) filled with soil without fertilizer. At the time of the experiments, the seedlings were 6 to 9 months old, 20 to 30 cm tall, and possessed three to nine leaves. Measurements were taken on the youngest mature leaves. A. excelsum is a pioneer tree, whereas V. surinamensis and C. longifolium are late-successional trees of tropical forests in Central and South America. Seedlings of all three species are shade tolerant. Contents of photosynthetic pigments (determined with HPLC) and soluble UV-B-absorbing compounds (extracted with ethanol/water mixtures) of A. excelsum and C. longifolium did not show significant differences between leaves from the shaded greenhouse and from the shaded understory of natural forests in Central Panama (G.H. Krause, E. Grube, O.Y. Koroleva, and K. Winter, unpublished data). Understory seedlings of V. surinamensis were not investigated.

Sun Exposure

Plants were exposed to direct sunlight for 20 to 50 min on days of low cloudiness between 11:00 AM and 1:30 PM, local time, simulating the conditions in a typical forest gap caused by the fall of a single tree. Exposure times were varied depending on the degree of sunlight fluctuations caused by clouds. Three plants were placed under a frame covered with 0.13-mm polyester plastic film (Mylar D; DuPont, Wilmington, DE) that excluded most UV-B but had a high transmission for UV-A and visible light. Three additional plants were placed under 0.08-mm Aclar film (22A; Allied Signal, Pottsville, PA) with high transmittance of UV-B, UV-A (320-400 nm), and PAR (400-700 nm). For details of the optical properties of these plastic films, see Krause et al. (1999). The plastic covers had slits for air circulation, which was enhanced by fans. The pots were shielded from the sun with aluminum foil.

PAR was measured with a quantum sensor (LI 189 B; LI-COR, Lincoln, NE) and UV-B was measured with a radiometer (IL 1400 A; International Light, Newburyport, MA). Measured integrated PAR doses of exposure varied from 2.2 to 3.6 mol m² and UV-B doses under Aclar film from 2.6 to 4.8 kJ m². The UV-B sensor has a fixed wavelength sensitivity (with a tail reaching in the UV-A region); it provided an approximate measure of UV-B energy. Weighted biologically effective UV-B radiation was not determined. Leaf temperatures were measured on the lower leaf surface with a leaf clip holder (2030-B) of a fluorometer system (PAM 2000; Walz, Effeltrich, Germany). In eight experiments, leaf temperatures (mean ± SD) in full sun were 40°C ± 2°C and 39°C ± 3°C in the presence and the absence of UV-B, respectively. Corresponding air temperature under the plastic films was 33°C ± 2°C, and ambient air temperature was 32°C ± 2°C. Subsequent to sun exposure, the plants were returned to the shaded greenhouse until the measurements of photosynthetic activities (below) were performed.

CO₂ Gas Exchange

About 1 to 2 h before and 1 to 3 h after sun exposure of plants, CO₂ gas exchange was measured on one leaf per plant using a portable photosynthesis system (LI-6400; LI-COR) equipped with a light source (6400-02B LED; LI-COR). In addition, transpiration was recorded. Leaf temperature in the chamber was maintained at 28°C. Prior to measurements, plants were preilluminated in a controlled-environment chamber (EGC, Chagrin Falls, OH) for 30 min at about 150 µmol m² s¹ PAR and at an air temperature of 24°C. Air flow through the photosynthesis cuvette was 250 µmol s¹; ambient air containing 360 to 380 µL L¹ CO₂ was used.

Prior to sun exposure, net CO₂ uptake at 350 µmol m² s¹ PAR and dark respiration were recorded for 6 min each on four plants. Light saturation curves of CO₂ assimilation were determined on two plants using the following protocol: 6 min at 350 µmol m² s¹; 2 min at 150, 100, 50, 20, and 10 µmol m² s¹; and 6 min in the dark. The rates obtained served as control values. Subsequent to sun exposure, light-saturation curves were determined for all six plants.

_P was calculated from the initial slopes of the light saturation curves; only curves were considered that exhibited at low actinic light an approximately linear phase with high regression coefficient (r² > 0.9). In particular, in strongly inhibited leaves (e.g. see Fig. 1B), such a linear phase was absent, and an initial slope could not be determined.

Chl a Fluorescence

F_V/F_M served as a measure of potential efficiency of PSII (see Giersch and Krause, 1991; Krause and Weis, 1991). The F_V/F_M ratio was determined with a chlorophyll fluorometer (PAM 2000; Walz, Effeltrich, Germany) equipped with a portable personal computer (Poquet Computer, Santa Clara, CA) and data acquisition software (DA-2000; Walz). Measurements were performed on the same leaves used for gas exchange measurements and on an additional leaf of each plant of similar orientation and age. Control measurements were done about 1 h prior to sun exposure, after 10 min of dark adaptation. After the end of sun exposure, the leaves were dark adapted for 15 min, and F_V/F_M was measured again. An additional measurement was performed about 5 h after exposure (after a 10-min dark adaptation). Immediately before F_V/F_M recording, weak far-red light from a diode of the PAM system was applied for 3 s to fully oxidize intersystem electron carriers for exact determination of initial fluorescence, F_O. The decrease in the "dark-adapted" F_V/F_M ratio caused by sun exposure indicates the degree of PSII photoinhibition (Giersch and Krause, 1991). For further details of the measuring protocol, see Barth and Krause (1999).

Photooxidation of P700

Photochemical activity of the PSI reaction center was assessed as described by Barth et al. (2001), by means of P700 absorbance changes at 810 nm (Klughammer and Schreiber, 1998) using a dual-wavelength emitter-detector unit (ED-P700DW; Walz) connected to a fluorometer system (PAM 101/102/103; Walz). Far-red light to photooxidize P700 was supplied by a diode with maximum emission at 735 nm (model 102-FR; Walz) controlled by the PAM 102, supplemented by a lamp (KL-150; Schott, Mainz, Germany) with optical filters RG9 (Schott) and Calflex C (Balzers, Liechtenstein; maximum intensity at 738 nm). Leaves were preilluminated for 5 min with white light (about 170 µmol m² s¹) from a lamp (KL-1500; Schott). Relative absorbance changes were subsequently recorded in the steady state as a function of far-red intensity. From the data, a "saturation constant" (K_S), defined as the far-red flux at which one-half of the maximum absorbance change (A810_max/2) in the steady state is reached (Barth et al., 2001), was calculated. The maximum absorbance change (A810_max) served as a relative measure of the photochemical capacity of PSI.

Measurements were performed 1 to 3 h before and 2 to 4 h after sun exposure of plants (subsequent to recording of gas exchange) with leaf discs (diameter of 1.45 cm) placed into a temperature-controlled cuvette (24°C) and ventilated with a moistened air stream. The changes in PSI activity induced by sun exposure did not show any tendency of reversal during the measuring period.

Statistics

To assess the significance of UV-B effects on photosynthetic parameters, the data from eight experiments with three species (sun-exposed in the presence and absence of UV-B, respectively) were averaged, and t tests were made. Probabilities of error (P) are given in the legends of Figures 2 through 4 and 6. In Figures 1 and 5, typical experiments with three plants each are depicted; for reasons of clarity, means and SD are presented instead of raw data.