High light-induced switch from C 3 -photosynthesis to Crassulacean acid metabolism is mediated by UV-Aublue light

Journal of Experimental Botany, Vol. 53, No. 373, pp. 1475 1483, June 2002 High light-induced switch from C 3 -photosynthesis to Crassulacean acid metabolism is mediated by UV-Aublue light Thorsten E. E. Grams 1,2,4 and Stephan Thiel 1,3,5 1 GSF-National Research Center for Environment and Health, Institute of Soil Ecology, Department of Environmental Engineering, Ingolstädter Landstraße, D-85764 Neuherberg, Germany 2 Forest Botany, Department of Ecology, Technische Universität München, Am Hochanger 13, D-85354 Freising, Germany 3 Forschungszentrum Karlsruhe, Institut für Meteorologie und Klimaforschung, Bereich Atmosphärische Umweltforschung, Kreuzeckbahnstraße 19, D-82467 Garmisch-Partenkirchen, Germany Received 10 September 2001; Accepted 27 January 2002 Abstract The high light-induced switch in Clusia minor from C 3 -photosynthesis to Crassulacean acid metabolism (CAM) is fast (within a few days) and reversible. Although this C 3 ucam transition has been studied intensively, the nature of the photoreceptor at the beginning of the CAM-induction signal chain is still unknown. Using optical filters that only transmit selected wavelengths, the CAM light induction of single leaves was tested. As controls the opposite leaf of the same leaf pair was studied in which CAM was induced by high unfiltered radiation (c. 2100 mmol m 2 s 1 ). To evaluate the C 3 -photosynthesisucam transition, nocturnal CO 2 uptake, daytime stomatal closure and organic acid levels were monitored. Light at wavelengths longer than 530 nm was not effective for the induction of the C 3 ucam switch in C. minor. In this case CAM was present in the control leaf while the opposite leaf continued performing C 3 -photosynthesis, indicating that CAM induction triggered by high light conditions is wavelength-dependent and a leaf internal process. Leaves subjected to wavelengths in the range of 345 530 nm performed nocturnal CO 2 uptake, (partial) stomatal closure during the day (CAM-phase III), and decarboxylation of citric acid within the first 2 d after the switch to high light conditions. Based on these experiments and evidence from the literature, it is suggested that a UV-Aublue light receptor mediates the light-induced C 3 -photosynthesisucam switch in C. minor. Key words: Citrate decarboxylation, Clusia minor, C 3 -photosynthesisucam switch, spectral irradiance, UV-Aublue light receptor. Introduction Crassulacean acid metabolism (CAM) is a typical ecophysiological adaptation of plants to arid conditions. Its characteristics are nocturnal CO 2 -uptake via the enzyme phosphoenolpyruvate carboxylase (PEPCase, CAMphase I sensu Osmond, 1978) and subsequent accumulation of the fixed CO 2 in the form of organic acids (in the case of Clusia these are malate and citrate) in the vacuole. Apart from periods of high stomatal conductance in the morning and evening hours (CAM-phases II and IV, respectively), CAM plants show a typical daytime stomatal closure (CAM-phase III) paralleled by decarboxylation of the stored organic acids in the cytoplasm. Subsequently, the CO 2 released is assimilated into carbohydrates by 1,5-ribulose bisphosphate carboxylaseu oxygenase (Rubisco) and used in the C 3 -photosynthesis carbon reduction cycle. In Clusia the CAM-phases II and IV are often very pronounced and, therefore, 4 Present address: Department of Integrative Biology, University of California, 3060 Valley Life Science Building, Berkeley, CA 94720, USA. 5 To whom correspondence should be sent at: Forschungszentrum Karlsruhe, Institut für Meteorologie und Klimaforschung, Bereich Atmosphärische Umweltforschung, Kreuzeckbahnstraße 19, D-82467 Garmisch-Partenkirchen, Germany. Fax: q49 8821 183 295. E-mail: Stephan.Thiel@imk.fzk.de ß Society for Experimental Biology 2002

1476 Grams and Thiel CAM-phase III is less obvious (for a review of CAM in the genus Clusia see Lüttge, 1999). Several plant species are able to switch from the more common mode of photosynthesis (C 3 ) to CAM, either during their ontogenetic development (e.g. Mesembryanthemum crystallinum; Adams et al., 1998) or in response to particular environmental conditions (e.g. drought). Within the tropical tree genus Clusia the phenomenon of facultative CAM is widespread and manifold (Lüttge, 1999). The highly flexible species C. minor has proved to be a useful model plant for the study of CAM induction because the C 3 ucam-switch is fast (within 1 d), reversible, and can be studied reliably by gas exchange measurements and analysis of accumulated organic acids. Several environmental stresses such as water shortage (Franco et al., 1991, 1992; Borland et al., 1996, 1998; Grams et al., 1997b, 1998), high irradiance, dayunight temperatures (Haag-Kerwer et al., 1992; Borland et al., 1993; Grams et al. 1997a; Roberts et al., 1998; Herzog et al., 1999) and nitrogen supply (Franco et al., 1991) have been shown to be effective at inducing the C 3 ucam transition. Although in C. minor the exposure to high irradiance is one of the most intensively studied CAM-inducing factors, to date the nature of the initial photoreceptor sensing the high radiation has remained unclear. Plants employ a series of photoreceptors, absorbing in different regions of the solar spectrum. The energy absorbed by a photoreceptor starts a cascade of physiological processes that enables plants to monitor magnitude ( quantity ) and wavelength distribution ( quality ) of the radiation environment (photosensing). For example, in the facultative CAM plant Kalanchoë blossfeldiana, CAM is induced by short days via red-light controlled synthesis of PEPCase mediated through the phytochrome system (Brulfert et al., 1982, 1988). Other possible photoreceptors or signalling pigments are chlorophyll (light harvesting for photosynthesis) and the blue light receptor (cryptochrome). The latter is known as a photoreceptor, for example, for phototropism, stomatal aperture and entrainment of endogenous rhythms in plants and animals (Baskin and Iino, 1987; Batschauer, 1998; Christie et al., 1998; Cashmore et al., 1999). The aim of the present study was to determine which of these photosensitive pigments elicits the high light (here photosynthetically active radiation (PAR) of c. 2000 mmol m 2 s 1 ) induced C 3 ucam transition in C. minor. To answer this question, the spectral range of the radiation applied to the leaves was manipulated using optical filters. The gas exchange patterns and organic acid accumulation was then studied in response to the light manipulations as a reliable way to monitor the CAM induction. Schmitt et al. showed in a rewatering experiment that C. minor is able to perform CAM in one leaf and C 3 -photosynthesis in the opposite leaf of the same leaf pair (Schmitt et al., 1988). This encouraged the use of one leaf as a control where the leaf was switched from C 3 -photosynthesis to CAM using high light conditions, while the opposite leaf was tested by selecting wavelengths and monitoring leaf responsiveness with respect to CAM induction. Materials and methods Spectroradiometric measurements Spectral data were recorded with a spectroradiometer system which consists of a double-monochromator, a photomultiplier, electronic devices (high voltage supply, amplifier and ADC), and an entrance optic (quartz fibre with a quartz diffuser plate; for details see Thiel et al., 1996). The system was calibrated against a 100 W quartz halogen lamp of a known spectral emission, determined via comparison with a 1000 W lamp (calibrated by the Physikalisch Technische Bundesanstalt, Braunschweig, Germany). The spectral irradiance of the light source used was monitored at the distance of the halogen lamp to the upper leaf surface during the experiments. Two sets of experiments were performed (see below: experimental design) using first UG11, OG550 (Schott, Mainz, Germany) and LCLS-550 filters (Laser Components, Olching, Germany, see Fig. 1a for spectral transmittance) and, second, a series of long pass filters (WG345, GG395, GG400, GG455, OG530, OG550, all from Schott, Mainz, Germany, see Fig. 1b for spectral transmittance). Plants Clusia minor L. plants were derived from cuttings from four individuals, all belonging to one clone, which originated from a plant that was collected in the Sierra de San Luis, Venezuela (Popp et al., 1987). This clone is well characterized in at least 15 publications and was therefore chosen for this study. Cuttings were grown in standard soil (Fruhsdorfer Einheitserde LD80) in a glasshouse, kept well watered and were fertilized regularly (using a 1 : 1 mixture of Hakaphos green and Hakaphos blue, Campo, Münster, Germany). At the age of 3 months, plants were acclimated for 3 6 months in a programmed climate chamber to low irradiance (12u12 h dayunight at c. 120 mmol m 2 s 1 ), constant temperature of 25 8C and relative humidity of 50%. The experimental plants were at least 6-months-old and had three to four new leaf pairs which developed under the low light conditions in the climate chamber. The first or second fully developed leaf pair was chosen for the experiments in which plants were kept well watered. Experimental set-up and assessment of leaf gas exchange Gas exchange of individual leaves was assessed using a minicuvette system (Fa. Walz, Effeltrich, Germany) with two climate- and light-controlled Plexiglas chambers (volume: 0.7 l) each enclosing one leaf of the same leaf pair. Measurements were performed under standardized conditions (cuvette temperature of 25.0"0.1 8C and dew point of 12.0"0.1 8C). Due to the effective climate control of the minicuvette system the increase of leaf temperature after the switch to high irradiance was small (2.4"1.5 8C) and similar for control and treatment leaves. Hence, the increase of vapour pressure deficit (VPD) with increasing irradiance was small. However, in C. minor, VPD is

UV-Aublue light induced CAM 1477 Fig. 2. Spectral irradiance of the used light sources (control and treatment) under high light conditions (c. 2100 mmol m 2 s 1, open and closed circles) and low light conditions (c. 120 mmol m 2 s 1, thin line and squares). The thick line shows the ratio of spectral irradiance of both lamps under high light conditions with a deviation between the lamps in the range of "10%. Fig. 1. Spectral transmittance of the optical filters used in the experiments. (a) Filters used in the first experimental set. Circles: UG11 filter, 2 mm thickness. Dashed line: short pass filter -550 nm, 3 mm thickness. Solid line: long pass filter )550 nm, 2 mm thickness. (b) Cut-off filters used in the second experimental set. From left to right: WG345, GG395, GG400, GG455, OG530, OG550, all 2 mm thick. Filter names give, theoretically, wavelength of 50% transmittance. a microwave oven (Popp et al., 1996), malate and citrate were determined enzymatically in hot water extracts of the dried leaf material according to Möllering (Möllering, 1974, 1985). The titratable Clusia leaf acidity at dusk is known to be negatively correlated with the CAM capacity (Zotz and Winter, 1993). Therefore, differences between control and treatment leaves at dusk were used to evaluate the CAM induction: CAM is not (or to a lesser extent) expressed in the treatment leaf, if its organic acid concentration is higher than in the control leaf. Statistics Significant levels were tested by means of Student s t-test. not effective for the C 3 ucam transition (Herzog, 1994). The rates of leaf gas exchange were calculated according to von Caemmerer and Farquhar (von Caemmerer and Farquhar, 1981). As a light source, halogen lamps were used (Master Line, 20W, type no. 17717, Philips, The Netherlands), which were selected for similar spectral distributions (Fig. 2). Each lamp was mounted in a light unit at a distance of 15 cm to the leaf. Slots allowed the installation of optical filters between lamp and leaf. To avoid external light the Plexiglas chambers were covered with aluminium foil. One leaf was switched from C 3 -photosynthesis to CAM by high light (control leaf ), while the opposite leaf of the same leaf pair was used for the induction manipulations using selected wavelengths. This treatment leaf was measured for its responsiveness to the light manipulation and light-induced C 3 ucam switch. The use of optical filters reduce the spectral irradiance within the experimental wavelength range (transmittance below 1.0, see Fig. 1). Therefore, the irradiance for the control leaf was reduced to the same order of magnitude by quartz glass, NG11 or NG4 filters (Schott, Mainz, Germany). Organic acid analyses Leaves were sampled at the end of each experiment (end of light phase) and stored at 75 8C. After drying of the leaf samples in Experimental design The aim of a first set of experiments was to contain the spectral range of the effective radiation. Figure 3 outlines the experimental scheme that was followed. Each experiment is represented by a box that gives the filter used and a number indicating whether the C 3 ucam switch occurs ( 1 ) or not ( 0 ). In the first experiment the UG11 filter tested the effectiveness of radiation from the ultraviolet part of the spectrum and radiation above 670 nm, while most of the photosynthetically active radiation was blocked. If the radiation that passed through the UG11 filter was effective for the C 3 ucam switch (denoted as 1 in Fig. 3), an additional experiment with a filter that transmits only radiation above 550 nm would be necessary to decide which of both possible receptors ( phytochrome or UV-B receptor) was responsible for the CAM induction. If, on the other hand, the UG11 filter blocked the effective radiation (i.e. no CAM switch, denoted as 0 ) further experiments would be needed as shown in the right branch of Fig. 3: first, a filter which mainly transmits radiation between 400 nm and 550 nm ( -550 nm ) and, secondly, a filter which only transmits radiation above 550 nm can be used to detect the involved photoreceptor. In a second set of experiments several cut-off filters (see Fig. 1b for spectral transmittance) were employed to study, in more detail, the spectral response of the CAM induction within the wavelength range determined in the preceding experiments.

1478 Grams and Thiel Results Representative results of the first set of experiments are shown in Fig. 4: during the first 24 h all leaves were exposed to a radiation regime similar to the growth conditions in the climate chamber (i.e. low photosynthetic Fig. 3. Processing chart for the first set of experiments. The first experiment was conducted using the UG11 filter, which blocks mainly photosynthetically active radiation (Fig. 1a). If CAM is induced ( 1 ), the following experiment will be performed using the )550 long pass filter. If CAM is not induced ( 0 ), the next experiment will be conducted with the -550 nm short pass filter (and so on). This sequence of experiments determines the photoreceptor which is responsible for high light CAM induction. Please refer to text for details. photon flux density (PPFD) of about 120 mmol m 2 s 1, see Fig. 2 for spectral irradiance). Under these conditions all leaves showed the typical gas exchange pattern of C 3 -photosynthesis: CO 2 uptake and stomatal opening was restricted to daytime and plants displayed respiratory CO 2 loss during the night. At time zero the PPFD of the control leaf (Fig. 4a, c, e) was increased to about 2100 mmol m 2 s 1 and leaves subsequently switched from C 3 -photosynthesis to CAM with nocturnal CO 2 -uptake anduor a clearly expressed CAM-phase III. The corresponding opposite leaf received only radiation of selected wavelengths. Radiation which passed through the UG11 filter (mainly -400 nm, see Fig. 1a) was not effective at inducing the C 3 ucam switch (Fig. 4b). Moreover, since photosynthetically active radiation was blocked by this filter, plants did not take up CO 2 either during the day, and stomata stayed closed day and night. By contrast, wavelengths below 550 nm (filter -550 in Fig. 1a) induced a switch to CAM with a clearly expressed CAM-phase III and a small CO 2 -uptake during the second night after the irradiance increase (Fig. 4d). The following experiment (Fig. 4f) shows that CAM is not induced by wavelengths above 550 nm (filter )550 in Fig. 1a): plants remained in the C 3 -mode of photosynthesis showing only daytime stomatal opening and Fig. 4. Net CO 2 exchange (J CO2, m ) and stomatal conductance (g H2 O, ) of individual C. minor leaves under increasing irradiance. At time zero the light conditions were switched from approximately 120 to 2100 mmol m 2 s 1 for control leaves (a, c, e), while the corresponding opposite leaves of the same leaf pairs received only selected wavelengths (b, d, f, respectively). Following the processing chart (Fig. 3) the UG11, -550 and )550 filters were used. Experiments shown are representative for a total of three experiments. Dark bars indicate night-time.

CO 2 uptake, while stomata closure and some respiratory CO 2 loss was observed during the night. The second set of experiments is shown in Fig. 5: similar to the earlier figures, the gas exchange patterns of control leaves are presented on the left hand side (Fig. 5a, c, e, g, i, k), while the spectral responsivenesses of high light CAM induction (corresponding opposite leaves of the same leaf pairs) are shown to the right (Fig. 5b, d, f, h, UV-Aublue light induced CAM 1479 j, l). Under low light conditions ( 24 to 0 h) all leaves showed C 3 -photosynthesis with daytime stomatal opening and CO 2 -uptake, while stomata stayed closed during the night and respiratory CO 2 loss was recorded. After the irradiance was increased all control leaves switched to CAM with nocturnal CO 2 -uptake, (partial) stomatal closure during the day (CAM-phase III; see g H2 O)anduor daytime citrate breakdown (Fig. 6). In some control Fig. 5. Net CO 2 exchange (J CO2, m ) and stomatal conductance (g H2 O, ) of individual C. minor leaves under increasing irradiance. At time zero the light conditions were switched from approximately 120 to 2100 mmol m 2 s 1 for control leaves (a, c, e, g, i, k), while the corresponding opposite leaves of the same leaf pairs received only selected wavelengths (b, d, f, h, j, l, respectively). Dark bars indicate night-time.

1480 Grams and Thiel Fig. 6. Differences (values of control leaves minus values of treatment leaves) in organic acid concentrations between control and treatment leaves of the same leaf pair at dusk. Negative values indicate a lower acid concentration in the control leaf relative to the treatment leaf and, thus, no or only small CAM capacity of the treatment leaf. Results of the first set of experiments are shown in (a) and (b) (with n ¼ 3 6"1 SE) while (c) and (d) give the data for the second set of experiments). leaves (Fig. 5e, g) the high light-induced C 3 ucam switch was relatively slow and the expression of phase III only started during the second day under high light conditions. However, CAM induction of these leaves was confirmed by low organic acid levels at dusk (see below: Fig. 6). Wavelengths above 345 and 395 nm (Fig. 5b, d) proved to be effective for the C 3 ucam switch in C. minor: stomatal conductance (g H2 0) and CO 2 gas exchange showed all four phases of CAM (Osmond, 1978). Leaves at )400 and )455 nm (Fig. 5f, h) showed a slow CAM induction similar to the corresponding control leaves (Fig. 5e, g). Thus, wavelengths above 400 and 455 nm are still effective, although the C 3 ucam switch was slowed down: CO 2 -uptake was small during the second night and CAM-phase III was hardly expressed. The experiments using wavelengths above 530 and 550 nm (Fig. 5j, l) confirmed the results presented in Fig. 4: wavelengths above 550 nm are not effective for the CAM induction. The results presented above (Figs 4, 5) are summarized in Table 1 showing integrated nocturnal CO 2 -exchange under the different light conditions (low irradiance and high irradiance, including and excluding blue light). In the case of low irradiance, both control and treatment leaves showed CO 2 -loss during the night. After the irradiance was increased in the range of 345 530 nm ( UV-A blue light ) to high light conditions, treatment leaves had similar nocturnal CO 2 -uptake as control leaves under unfiltered radiation (8.21 and 7.19 mmol m 2, respectively; P)0.862). However, if blue light was excluded after the irradiance increase, integrated nocturnal CO 2 -exchange was negative (CO 2 -loss) for the treatment leaves, while the opposite control leaves switched to CAM ( 3.24 and 4.35 mmol m 2 for treatment and control leaves, respectively; P-0.006). At the end of the experiments (end of light phase) leaves of the gas-exchange measurements were harvested and analysed for organic acid levels (Fig. 6). The differences in malate or citrate concentrations within one leaf pair (concentration of control leaf minus concentration of treatment leaf) indicates the degree of CAM induction in the treatment leaves. If the difference is close to zero, both leaves used about the same amount of organic acid for daytime decarboxylation as an internal CO 2 source during CAM phase III. A negative value (higher organic acid level in the treatment leaf at dusk) indicates that the treatment leaf did not (or to a lesser extent) switch to CAM since less organic acid was decarboxylated. Differences in malate concentrations between control and treatment leaves (Fig. 6a, c) did not give a clear trend with respect to the high lightinduced C 3 ucam switch. However, differences in citrate concentrations at dusk confirmed that CAM was induced by wavelengths below 550 nm (Fig. 6b), since the treatment and control leaves decarboxylated the same amount of citrate during the day (i.e. -550 -value is not significantly different from zero). Wavelengths which passed through the UG11 and )550 filters were not effective for the citrate decarboxylation: in both cases treatment leaves decarboxylated no citrate or decarboxylated less citrate than the control leaves (see negative values in Fig. 6b; both values are significantly different from -550 citrate-value, P-0.05). Similar conclusions can be drawn from the citrate data of the second experimental set (Fig. 6d): wavelengths above 455 nm appear not, or only partially, to be effective for the high light-induced C 3 ucam switch (negative values), whereas experiments with wavelengths above 345, 395 and 400 led to no or only minor differences in citrate concentrations between treatment and control leaves, indicating that both leaves switched to CAM. Discussion Most of the Clusia species studied so far are facultative CAM plants, in which CAM is induced by various environmental factors like drought, increased daytime temperature or increased irradiance (Grams et al., 1998; Lüttge, 1999). CAM induction caused by high irradiance is well known and has been studied intensively, especially in C. minor (Schmitt et al., 1988; Franco et al., 1991; Haag-Kerwer et al., 1992; Borland et al., 1993, 1996; Roberts et al., 1998; Herzog et al., 1999). It is shown here that the photoreceptor responsible for the CAM

Table 1. Integrated nocturnal CO 2 -exchange of control and treatment leaves under different light conditions UV-Aublue light induced CAM 1481 Experiments of Figs 4 and 5 are summarized with respect to blue light (345 530 nm) application. Low irradiance : summary of all experiments under low light conditions. High irradiance including UV-Aublue light High spectral irradiance in the UV-A- and blue range: summary of experiments with filters -550, )345, )395, )400, and )455. High irradiance excluding UV-Aublue light No or negligible radiation from the UV-A- and blue range: summary of experiments with filters UG11, )530 and )550. P-values show if treatment leaves are significant different from control leaves. Means"1 SE(n) ¼ 10). Integrated nocturnal CO 2 exchange (mmol m 2 ) Low irradiance High irradiance Including UV-Aublue light Excluding UV-Aublue light Treatment leaves 8.83"0.96 8.21"4.22 3.24"0.63 Control leaves 8.27"0.89 7.19"4.01 4.35"2.30 Significance level P)0.669 P)0.862 P-0.006 induction by high irradiance is most effective in the range of 345 to 530 nm (UV-Aublue light, Figs 4, 5, 6; Table 1). Since light above 530 nm was not effective for this CAM induction, a phytochrome induction system can be excluded as the primary chromophore involved in the high light CAM induction process. Obviously, CAM induction caused by high irradiance requires the attainment of a certain threshold value, which can be estimated from a biologically weighted radiation quantity as follows: ð E biol ¼ E L ðlþ3a F ðlþ3 S biol ðlþdl where E biol represents the biological effective irradiance (here for the C 3 ucam transition), E L the spectral irradiance of the radiation source, A F the filter transmittance and S biol the action spectrum of the CAM induction. If the action spectrum of alfalfa hypocotyl phototropism (mediated by a blue light receptor; Baskin and Iino, 1987) is taken to estimate E biol for the high light CAM induction, the applied low light level (C 3 -photosynthesis conditions) equals 167 mw m 2, while for unfiltered radiation under high light conditions (CAM conditions) a value of 2750 mw m 2 is calculated. These values also show that the amount of UV-Aublue light passing through the UG11 filter and resulting in an biologically effective radiation of 59 mw m 2 is below the threshold value for the CAM induction. On the other hand the values for )400 and )455 (2479 and 1392 mw m 2, respectively) are above the threshold, since CAM was still induced by these wavelengths. Despite the fact that a threshold value for the CAM induction has to be reached, the C 3 ucam transition might be a gradual process rather than an all-or-none response, since the amount of nocturnally fixed CO 2 is known to saturate with the integrated light sum from the previous day (Schmitt et al., 1988). The fact that CAM induction does not occur in experiments with wavelengths higher than 530 nm, shows that the maximum of the action spectrum for CAM induction is in the range of blue light. However, the investigation of a dose response relationship above the threshold and the determination of an action spectrum remains a task for further research. High irradiance in the range of 345 530 nm induces nocturnal CO 2 uptake anduor (partial) daytime stomatal closure (phase III of CAM, sensu Osmond, 1978; see Table 1; Figs 4, 5) as well as decarboxylation of citric acids (Fig. 6). Concentrations of malate at the end of the light phase stayed low (2.1"1.8 mmol m 2, data not shown), since nocturnally accumulated malate will be broken down completely during the day. However, these experiments monitored only the beginning of the C 3 ucam transition, which will result, if environmental conditions remain unchanged, in a higher nocturnal CO 2 uptake and organic acid (malate and citrate) turnover. Similar findings have been reported previously (Herzog et al., 1999) which showed that in C. minor diurnal malate oscillations are not present during the first day under high irradiance (i.e. an increase from 100 to 1000 mmol m 2 s 1 ), while citrate breakdown had already started (cf. Fig. 6). Likewise, a much closer correlation was found between PPFD and citrate breakdown than malate breakdown during the first days of CAM induction in C. minor and it was concluded that light-induced CAM starts with daytime citrate breakdown (Borland et al., 1996). Thus, following the concept of a limited vacuole storage capacity for H q, only daytime citrate breakdown and, thus, a reduced titratable leaf acidity at dusk allows nocturnal CO 2 uptake (Zotz and Winter, 1993). These findings were confirmed and the citrate concentrations were used to confirm CAMinduction and to evaluate the effective wavelengths for the high light-induced C 3 ucam switch (Fig. 6). The direct impact of blue light on stomatal aperture, as shown for many C 3 plants (Briggs and Huala, 1999), seems to be unlikely if CAM is present. Several studies (Lee and Assmann, 1992; Mawson and Zaugg, 1994; Tallman et al., 1997) report that this stomatal blue-light response is either lost or inhibited in C 3 ucam intermediate plants after the transition to CAM. Rather, the signal transduction pathway of blue light-mediated CAM

1482 Grams and Thiel induction in C. minor might be similar to the short-day CAM induction in K. blossfeldiana, where red light is sensed by the phytochrome system and CAM is induced via de novo synthesis of PEPCase (or phosphoenolpyruvate carboxykinase; Brulfert et al., 1985, 1988). In the habitats of C. minor, increases of irradiance (and simultaneous water shortage) are relatively frequent (e.g. due to leaf senescence of the surrounding vegetation during drought periods). Switching from C 3 -photosynthesis to CAM is an important and useful mechanism to cope with excess radiation energy (Herzog et al., 1999; Lüttge, 1999, 2000). Therefore, the sensing of high irradiance via UV-A blue light and the subsequent switch to CAM gives C. minor a larger ecological amplitude and, in general, a better performance under highly variable radiation conditions. Acknowledgements The authors are grateful to Dr H-D Payer, Dr HK Seidlitz and the staff of the EPOKA team at the GSF-National Research Center for Environment and Health for technical assistance and the opportunity to perform this study in the EPOKA phytotrons. Dr R Matyssek and Dr TE Dawson are thanked for valuable comments on an earlier version of the manuscript. References Adams P, Nelson DE, Yamada S, Chmara W, Jensen RG, Bohnert HJ, Griffiths H. 1998. Growth and development of Mesembryanthemum crystallinum (Aizoaceae). New Phytologist 138, 171 190. Baskin TI, Iino M. 1987. An action spectrum in the blue and ultraviolet for phototropism in alfalfa. Photochemistry and Photobiology 46, 127 136. Batschauer A. 1998. Photoreceptors of higher plants. Planta 206, 479 492. Borland AM, Griffiths H, Broadmeadow MSJ, Fordham MC, Maxwell C. 1993. Short-term changes in carbon-isotope discrimination in the C 3 ucam intermediate Clusia minor L. growing in Trinidad. Oecologia 95, 444 453. Borland AM, Griffiths H, Maxwell C, Fordham MC, Broadmeadow MSJ. 1996. CAM induction in Clusia minor L. during the transition from wet to dry season in Trinidad: the role of organic acid speciation and decarboxylation. Plant, Cell and Environment 19, 655 664. Borland AM, Tecsi LI, Leegood, RC, Walker RP. 1998. Inducibility of crassulacean acid metabolism (CAM) in Clusia species: physiologicalubiochemical characterization and intercellular localization of carboxylation and decarboxylation processes in three species which exhibit different degrees of CAM. Planta 205, 342 351. Briggs WR, Huala E. 1999. Blue-light photoreceptors in higher plants. Annual Review of Cell and Developmental Biology 15, 33 62. Brulfert J, Kluge M, Güclü S, Queiroz O. 1988. Interaction of photoperiod and drought as CAM inducing factors in Kalanchoë blossfeldiana Poelln. cv. Tom Thumb. Journal of Plant Physiology 133, 222 227. Brulfert J, Müller D, Kluge M, Queiroz O. 1982. Photoperiodism and Crassulacean acid metabolism. I. Immunological and kinetic evidence for different patterns of phosphoenolpyruvate carboxylase isoforms in photoperiodically inducible and non-inducible Crassulacean acid metabolism plants. Planta 164, 326 331. Brulfert J, Vidal J, Keryer E, Thomas M, Gadal P, Queiroz O. 1985. Phytochrome control of phosphoenolpyruvate carboxylase synthesis and specific RNA levels during photoperiodic induction in a CAM plant and during greening in a C4 plant. Physiologie Ve ge tale 23, 921 928. Cashmore AR, Jarillo JA, Wu Y-J, Liu D. 1999. Cryptochromes: blue light receptors for plants and animals. Science 248, 760 765. Christie JM, Reymond P, Powell GK, Bernasconi P, Raibekas AA, Liscum E, Briggs WR. 1998. Arabidopsis NPH1: a flavoprotein with the properties of a photoreceptor for phototropism. Science 282, 1698 1701. Franco AC, Ball E, Lüttge U. 1991. The influence of nitrogen, light and water stress on CO 2 exchange and organic acid accumulation in the tropical C 3 -CAM tree, Clusia minor. Journal of Experimental Botany 42, 597 603. Franco AC, Ball E, Lüttge U. 1992. Differential effects of drought and light levels on accumulation of citric and malic acid during CAM in Clusia. Plant, Cell and Environment 15, 821 829. Grams TEE, Borland AM, Roberts A, Griffiths H, Beck F, Lüttge U. 1997a. On the mechanism of reinitiation of endogenous crassulacean acid metabolism rhythm by temperature changes. Plant Physiology 113, 1309 1317. Grams TEE, Haag-Kerwer A, Olivares E, Ball E, Arndt S, Popp M, Medina E, Lüttge U. 1997b. Comparative measurements of chlorophyll a fluorescence, acid accumulation and gas exchange in exposed and shaded plants of Clusia minor L. and Clusia multiflora H. B. K. in the field. Trees Structure and Function 11, 240 247. Grams TEE, Herzog B, Lüttge U. 1998. Are there species in the genus Clusia with obligate C 3 -photosynthesis? Journal of Plant Physiology 152, 1 9. Haag-Kerwer A, Franco AC, Lüttge U. 1992. The effect of temperature and light on gas exchange and acid accumulation in the C 3 -CAM plant Clusia minor L. Journal of Experimental Botany 43, 345 352. Herzog B. 1994. Der Einfluss der Wasserdampfdruckdifferenz (VPD) zwischen dem Blattinneren und der Atmosphäre auf die C 3 ucam-umstellung bei Clusia minor L. Dipl.-Biol.- thesis, Technische Universität Darmstadt. Herzog B, Hoffmann S, Hartung W, Lüttge U. 1999. Comparison of photosynthetic responses of the sympatric tropical C 3 species Clusia multiflora H. B. K. and the C 3 -CAM intermediate species Clusia minor L. to irradiance and drought stress in a phytotron. Plant Biology 1, 460 470. Lee DM, Assmann SM. 1992. Stomatal responses to light in the facultative Crassulacean acid metabolism species, Portulacaria afra. Physiologia Plantarum 85, 35 42. Lüttge U. 1999. One morphotype, three physiotypes: sympatric species of Clusia with obligate C 3 photosynthesis, obligate CAM and C 3 -CAM intermediate behaviour. Plant Biology 1, 138 148. Lüttge U. 2000. Light-stress and Crassulacean acid metabolism. Phyton 40, 65 82. Mawson BT, Zaugg MW. 1994. Modulation of lightdependent stomatal opening in isolated epidermis following induction of crassulacean acid metabolism in Mesembryanthemum crystallinum L. Journal of Plant Physiology 144, 740 746.

Möllering H. 1974. Malat. Bestimmung mit Malat- Dehydrogenase und Glutamat-Oxalacetat-Transaminase. In: Bergmeyer HU, ed. Methoden der Enzymology. New York: Academic Press, 1636 1639. Möllering H. 1985. Citrate. determination with citrate lyase, MDH and LDH. In: Bergmeyer HU, ed. Methods of enzymatic analysis. New York: Academic Press, 2 12. Osmond CB. 1978. Crassulacean acid metabolism: a curiosity in context. Annual Review of Plant Physiology and Plant Molecular Biology 29, 379 414. Popp M, Kramer D, Lee H, Diaz M, Ziegler H, Lüttge U. 1987. Crassulacean acid metabolism in tropical dicotyledonous trees of the genus Clusia. Trees Structure and Function 1, 238 247. Popp M, Lied W, Meyer AJ, Richter A, Schiller P, Schwitte H. 1996. Sample preservation for determination of organic compounds: microwave versus freeze-drying. Journal of Experimental Botany 47, 1469 1473. Roberts A, Borland AM, Maxwell K, Griffiths H. 1998. Ecophysiology of the C 3 -CAM intermediate Clusia minor L. in Trinidad: seasonal and short-term photosynthetic UV-Aublue light induced CAM 1483 characteristics of sun and shade leaves. Journal of Experimental Botany 49, 1563 1573. Schmitt AK, Lee HSJ, Lüttge U. 1988. The response of the C 3 - CAM tree, Clusia rosea, to light and water stress. I. Gas exchange characteristics. Journal of Experimental Botany 39, 1581 1590. Tallman G, Zhu J, Mawson BT, Amodeo G, Nouhi Z, Levy K, Zeiger E. 1997. Induction of CAM in Mesembryanthemum crystallinum abolishes the stomatal response to blue light and light-dependent zeaxanthin formation in guard cell chloroplasts. Plant and Cell Physiology 38, 236 242. Thiel S, Döhring T, Köfferlein M, Kosak A, Martin P, Seidlitz HK. 1996. A phytotron for plant stress research: how far can artificial lighting compare to natural sunlight? Journal of Plant Physiology 148, 456 463. von Caemmerer S, Farquhar GD. 1981. Some relationships between the biochemistry of photosynthesis and the gas exchange of leaves. Planta 153, 376 387. Zotz G, Winter K. 1993. Short-term regulation of crassulacean acid metabolism activity in a tropical hemi epiphyte, Clusia uvitana. Plant Physiology 102, 835 841.