The stomatal response to CO 2 is linked to changes in guard cell zeaxanthin*

Transcription

1 Plant, Cell and Environment (1998) 21, ORIGINAL ARTICLE OA 220 EN The stomatal response to CO 2 is linked to changes in guard cell zeaxanthin* J. ZHU, L. D. TALBOTT, X. JIN & E. ZEIGER Department of Biology, University of California, Los Angeles, CA , USA ABSTRACT The mechanisms mediating CO 2 sensing and light CO 2 interactions in guard cells are unknown. In growth chamber-grown Vicia faba leaves kept under constant light (500 µmol m 2 s 1 ) and temperature, guard cell zeaxanthin content tracked ambient [CO 2 ] and stomatal apertures. Increases in [CO 2 ] from 400 to 1200 cm 3 m 3 decreased zeaxanthin content from 180 to 80 mmol mol 1 Chl and decreased stomatal apertures by 7 0 µm. Changes in zeaxanthin and aperture were reversed when [CO 2 ] was lowered. Guard cell zeaxanthin content was linearly correlated with stomatal apertures. In the dark, the CO 2 - induced changes in stomatal aperture were much smaller, and guard cell zeaxanthin content did not change with chamber [CO 2 ]. Guard cell zeaxanthin also tracked [CO 2 ] and stomatal aperture in illuminated stomata from epidermal peels. Dithiothreitol (DTT), an inhibitor of zeaxanthin formation, eliminated CO 2 -induced zeaxanthin changes in guard cells from illuminated epidermal peels and reduced the stomatal CO 2 response to the level observed in the dark. These data suggest that CO 2 -dependent changes in the zeaxanthin content of guard cells could modulate CO 2 - dependent changes of stomatal apertures in the light while a zeaxanthin-independent CO 2 sensing mechanism would modulate the CO 2 response in the dark. Key-words: Vicia faba; stomatal response to CO 2 ; xanthophyll cycle; zeaxanthin. INTRODUCTION Stomata have been shown to respond to changes in [CO 2 ] in over 50 species (Morison 1987). In general, increases in intercellular [CO 2 ] (C i ) reduce stomatal apertures and decreases in C i lead to stomatal opening. These responses are optimally suited to couple photosynthetic activity in the mesophyll and stomatal conductance in the epidermis (Mott 1990). Interactions between the stomatal response to CO 2 and other signals such as light, relative humidity, water stress, Correspondence: Dr Eduardo Zeiger. Fax (310) ; zeiger@biology.ucla.edu *This work was supported by NSF #DCB and DOE #90ER Blackwell Science Ltd temperature, leaf age and nutrients have been extensively studied (Morison 1987). In the intact leaf, the stomatal responses to CO 2 and light are functionally coupled and hard to separate experimentally. It is well established, however, that guard cells have independent responses to light and CO 2 (Sharkey & Raschke 1981; Zeiger 1983), that guard cells respond to CO 2 in both the light and in darkness, and that stomatal sensitivity to CO 2 increases with incident radiation (Heath & Russell 1954; Morison & Jarvis 1983; Wong et al. 1987). The mechanisms mediating CO 2 sensing in guard cells are not well understood (Assmann 1993). Guard cells have two well-characterized carboxylation reactions catalysed by Rubisco and phosphoenol pyruvate (PEP) carboxylase, which yield sugars and malate, respectively. Both reactions have been invoked as potential CO 2 sensing mechanisms (Zeiger et al. 1987; Raschke et al. 1988). The problem with these hypotheses, however, is that higher [CO 2 ] should increase the rate of synthesis of sugars and malate and thus cause stomatal opening, rather than closing. Results from a recent study argue against CO 2 sensing by osmoregulatory mechanisms (Talbott et al. 1996). Pulses of CO 2 applied to Vicia leaves caused very similar stomatal responses irrespective of whether guard cell osmoregulation was mediated by sucrose or by potassium and its counterions. Findings showing that the xanthophyll zeaxanthin is involved in signal transduction in guard cells (Srivastava & Zeiger 1995a,b) suggested another possible CO 2 sensing mechanism. Studies on the xanthophyll cycle of mesophyll chloroplasts have shown that the de-epoxidation reaction that catalyses the conversion of violaxanthin into zeaxanthin is regulated by lumen ph (Hager 1980). Lumen ph depends on the balance between light-driven electron transport and carbon fixation rates, and recent studies have shown that, under constant irradiation, decreases in ambient [CO 2 ] increased the zeaxanthin content of chloroplasts from cotton leaves (Gilmore & Björkman 1994). The steady-state yield of chlorophyll a fluorescence from guard cell chloroplasts is sensitive to changes in [CO 2 ] (Melis & Zeiger 1982; Cardon & Berry 1992), indicating that changes in [CO 2 ] alter the balance of ATP and NADPH in guard cell chloroplasts and have the potential to affect lumen ph. Therefore, changes in CO 2 could cause changes in the zeaxanthin content of guard cell chloroplasts. In the present study, we used growth chamber-grown Vicia faba leaves to test the hypothesis that the xanthophyll 813

2 814 J. Zhu et al. cycle of the guard cell chloroplast might be involved in CO 2 sensing. Stomata from growth chamber-grown Vicia leaves have an enhanced CO 2 response and thus provide a useful experimental system for the study of CO 2 sensing (Talbott et al. 1996). Stomata in intact leaves and in detached epidermis were used for parallel measurements of stomatal apertures and guard cell zeaxanthin content, as a function of varying ambient [CO 2 ]. Obtained results implicate the xanthophyll cycle of guard cells in the regulation of the stomatal response to CO 2. MATERIALS AND METHODS Plant material Plants were grown as described previously (Talbott et al. 1996). Briefly, seeds of Vicia faba L. cv. Windsor Long Pod (Bountiful Gardens Seeds, Willits, CA, USA) were planted in pots using commercial potting soil (Sunshine Mix #1, American Horticulture Supply, Camarillo, CA, USA). Plants were grown in a walk-in growth chamber (PGV-36, Conviron, Asheville, NC, USA) at 85 ± 3% RH, 12 h day, 500 µmol m 2 s 1, 25 ± 0 5 C/12 h night, 15 ± 0 5 C. Plants were watered 4 times a day with an automatic watering system and fertilized ( mix, Grow More Research and Manufacturing Co., Gardena, CA, USA) once a week. Fully expanded, recently matured and unshaded leaves from the third and fourth nodes of 4 5-week-old plants were used for experiments. In vivo CO 2 pulse experiments Carbon dioxide levels in the growth chamber were increased by addition of 100% CO 2 into the fan compartment of the growth chamber using a mass flow controller, as described previously (Talbott et al. 1996). Chamber [CO 2 ] was continuously monitored with an infrared gas analyser (Model EGM-1, PP Systems, Hitchin Herts, UK). At the onset of the light period, chamber [CO 2 ] was around cm 3 m 3 depending on the ambient [CO 2 ] of the room in which the chamber was located. Chamber CO 2 levels were raised from the initial level to 1200 cm 3 m 3 in 200 cm 3 m 3 increments and maintained for 1 h at each concentration to allow for the attainment of steady-state apertures at each [CO 2 ]. To lower chamber [CO 2 ], the flow of CO 2 was decreased, allowing photosynthetic carbon fixation in the canopy and leakage to remove CO 2 until the desired level was reached. Epidermal peel experiments Detached Vicia leaves were wrapped in wet paper towels and kept for 2 5 h in the dark, except as indicated. Abaxial epidermal peels were removed with forceps under dim room light and incubated in Petri dishes containing 1 0 mol m 3 KCl, 0 1 mol m 3 CaCl 2, and 1 0 mol m 3 Mes/NaOH (ph 6 8), either under 150 µmol m 2 s 1 white light (Sylvania DAH 300 W projector bulb, GTE Products Co., Winchester, KY, USA) or in darkness. The Petri dishes were placed in a 25 C constant temperature bath to prevent heating. The solution was aerated with 800 cm 3 m 3 CO 2 -containing air during the first hour of incubation, and then switched to CO 2 -free, 400 cm 3 m 3 or 800 cm 3 m 3 CO 2 -containing air for an additional hour. At the end of each treatment, a few peels were used for measurement of stomatal apertures (see below) and the rest were sonicated on ice for 30 s in a Branson Sonifier (model 250, Branson Ultrasonics Corp., Danbury, CT, USA) at a continuous duty cycle and a power setting of 7, to eliminate contamination from mesophyll chloroplasts. After sonication, the peels were rinsed thoroughly in tap water and stored at 80 C for pigment analysis by HPLC (Zhu et al. 1995). Measurement of stomatal apertures Apertures were measured with an Olympus BH-2 microscope (Olympus Corp., Lake Success, NY, USA) connected to a digital camera (JE2362 A, Javelin Electronics, Torrance, CA, USA). Abaxial epidermal strips from either intact leaves or the epidermal peel experiments were mounted on a slide for image analysis as described previously (Talbott et al. 1996). Apertures of 30 stomata from three or more epidermal strips were determined at each data point. Pigment extraction and analysis In the intact leaf experiments, 5 6 leaves were immersed in ice water immediately after the CO 2 treatments to minimize xanthophyll cycle activity, and the abaxial epidermes were detached and sonicated within 6 min. Mesophyll tissue was obtained from the same leaves used for epidermis preparation. Pigments were extracted under dim room light on ice. Clean epidermal peels were ground thoroughly (20 min) to ensure complete pigment extraction in a mortar in a mixture of 1 cm 3 acetone and about 500 mg anhydrous Na 2 SO 4, and 0 5 cm 3 NaHCO 3 -saturated hexane. KCl (1 0 cm 3, 0 2 kmol m 3 ) was then added and the mixture centrifuged at 1500 g for 3 min. The upper solvent phase of the extracts was collected and evaporated in vacuo. The dried samples were either analysed immediately after extraction or flushed with N 2 and stored at 80 C. Samples were chromatographed on a Beckman HPLC system (a 421 controller, a 165 detector and two 110 A pumps) provided with a Spectra-Physics SP4270 integrator and an Alltech Spherisorb ODS-1 5µ column, using the method described by Gilmore & Yamamoto (1991). Pigments were separated by elution with solvent A (acetonitrite:methanol: 0 1 kmol m 3 Tris-HCl buffer; 72:8:3, v/v; ph 8 0 adjusted with 10 kmol m 3 KOH) for 4 min at a flow rate of 2cm 3 min 1, followed by a 2 5 min linear gradient of 0 100% solvent B (methanol:hexane; 4:1, v/v), and then a 7 5 min elution with 100% solvent B. Pigments were quantified by integrating the area under the 440 nm absorption

3 Guard cell zeaxanthin and stomatal response to CO peak using empirically determined response factors. Response factors used in the calculation were for violaxanthin, for antheraxanthin, for zeaxanthin, for chlorophyll b, and for chlorophyll a. They were obtained by separating leaf pigment extracts by HPLC and collecting fractions containing pigments of interest, based upon published HPLC profiles (Gilmore & Yamamoto 1991). The collected fractions were dried and resuspended in hexane, acetone or ethanol. The identity of the pigment was confirmed by its absorption spectra and its concentration determined spectrophotometrically. Response factors were determined from peak areas obtained from different concentrations of the pigment (M. Quiñones, unpublished results). RESULTS Changes in stomatal apertures and guard cell zeaxanthin content in intact leaves as a function of ambient [CO 2 ] Typical pigment profiles from growth chamber-grown mesophyll and guard cells are shown in Fig. 1. Under constant light and temperature, abaxial guard cells from intact, growth chamber-grown Vicia leaves showed large changes in zeaxanthin content as a function of changes in ambient [CO 2 ] (Figs 1 & 2a). Guard cell zeaxanthin decreased from 181 mmol mol 1 Chl to 79 mmol mol 1 Chl when chamber [CO 2 ] increased from 400 to 1200 cm 3 m 3. As reported previously (Talbott et al. 1996), stomatal apertures also showed large changes as a function of ambient [CO 2 ] (Fig. 2a). A plot of stomatal apertures as a function of guard cell zeaxanthin content showed that the two parameters were linearly related (Fig. 2b). Figure 3 shows the changes in violaxanthin, antheraxanthin and zeaxanthin the three components of the xanthophyll cycle of mesophyll and guard cell chloroplasts as a function of changes in [CO 2 ]. The xanthophyll cycle of mesophyll chloroplasts was insensitive to the changes in [CO 2 ] under the experimental conditions used, and showed no detectable zeaxanthin under all tested [CO 2 ] (Fig. 3, inset). On the other hand, the xanthophyll cycle from guard cell chloroplasts showed a very high CO 2 sensitivity (Fig. 3). The reversibility of these responses to ambient [CO 2 ] was tested in experiments in which chamber [CO 2 ] was Figure 1. Typical HPLC chromatography profiles of photosynthetic pigments from guard cells under 400 and 900 cm 3 m 3 CO 2, and mesophyll cells under 400 cm 3 m 3 CO 2 from growth chamber-grown Vicia faba. Light fluence rate at the adaxial leaf surface was 500 µmol m 2 s 1 in the chamber. Individual pigment is indicated above the corresponding peak. Neo, neoxanthin; Vio, violaxanthin; Ant, antheraxanthin; Lut, lutein; Zea, zeaxanthin; Chl b, chlorophyll b; Chl a, chlorophyll a; β-car, β-carotene.

4 816 J. Zhu et al. The effect of the de-epoxidase inhibitor dithiothreitol (DTT) on CO 2 -dependent zeaxanthin formation and stomatal movements in the light and in the dark The results of experiments with intact leaves showing that guard cell zeaxanthin content tracks CO 2 -dependent stomatal movements in the light but not in the dark (Fig. 2) suggest that the metabolic processes associated with the stomatal responses to CO 2 in the dark might differ from those in the light. Experiments with epidermal peels made it possible to use the inhibitor of violaxanthin de-epoxidase, DTT, to test whether inhibition of zeaxanthin formation altered the stomatal response to CO 2. Stomata from abaxial epidermis detached from dark-adapted (2 5 h) leaves were incubated for 1 h in a solution aerated with 800 cm 3 m 3 CO 2 either in 150 µmol m 2 s 1 white light or in darkness, and then treated with 400 cm 3 m 3 CO 2 or CO 2 -free air for an additional hour in the presence or absence of 3 mol m 3 DTT (Srivastava & Zeiger 1995a). Table 1 shows the net changes in apertures after the second hour of incubation. In the dark, stomata transferred from 800 to 400 cm 3 m 3 CO 2 and to CO 2 -free air opened about 0 6 and 1 4 µm, respectively, and these responses were insensitive to 3 mol m 3 DTT. In the light, the stomatal responses to CO 2 were larger than in the dark, and 3 mol m 3 DTT reduced the response by 50% (Table 1). The DTT-insensitive component of the stomatal response to Figure 2. (a) CO 2 -dependent changes in stomatal apertures and guard cell zeaxanthin (Z) in leaves of growth chamber-grown Vicia faba under constant light (open symbols) and in the dark (closed symbols); (b) Relationship between guard cell zeaxanthin and stomatal apertures in leaves of growth chambergrown Vicia faba as a function of ambient [CO 2 ] under constant light. Light fluence rate at the adaxial leaf surface was 500 µmol m 2 s 1 in the chamber. Chamber CO 2 levels were increased from about 400 to 1200 cm 3 m 3 in 200 cm 3 m 3 increments. Each [CO 2 ] was maintained for 1 h prior to measurements. Each value represents an average of four experiments ± SE. raised from 400 to 1000 cm 3 m 3 and then lowered to 400 cm 3 m 3 in a stepwise fashion. Both guard cell zeaxanthin content and stomatal apertures tracked the changes in [CO 2 ] in both directions (Figs 4a & b). In the dark, average stomatal apertures at ambient [CO 2 ] were about 7 µm (vs. nearly 14 µm in the light) and apertures decreased by about 3 3 µm over the cm 3 m 3 range, compared to about 7 µm in the light. Guard cell zeaxanthin remained constant in the cm 3 m 3 range and decreased only slightly at 1000 cm 3 m 3 (Fig. 2a). There was no significant correlation between stomatal apertures in the dark and the zeaxanthin content in guard cells (r = 0 21). Figure 3. Antheraxanthin (A, ), violaxanthin (V, ), and zeaxanthin ( ) content in abaxial guard cell chloroplasts from leaves of growth chamber-grown Vicia, expressed as the percentage of the total xanthophyll cycle pool, as a function of ambient [CO 2 ] under constant light. Inset: A, V, and Z content in Vicia mesophyll chloroplasts under the same conditions. Light fluence rate and ambient [CO 2 ] were as described in Fig. 2. Each value represents an average of four experiments ± SE.

5 Guard cell zeaxanthin and stomatal response to CO CO 2 in the light was of the same magnitude as the total response to CO 2 in the dark. Figure 5(a) shows the zeaxanthin content in guard cells from epidermal strips incubated in 150 µmol m 2 s 1 white light under three different [CO 2 ], in the presence or absence of 3 mol m 3 DTT. Zeaxanthin content in guard cells incubated in 800 cm 3 m 3 CO 2 increased in parallel with stomatal apertures when transferred to 400 cm 3 m 3 or CO 2 -free air, and the increase in zeaxanthin was inhibited by DTT. In the absence of DTT, stomatal apertures were closely correlated with the zeaxanthin content of guard cells (Fig. 5b). Figure 4. Reversibility of the CO 2 -dependent changes in stomatal apertures (a) and guard cell zeaxanthin (b) in growth chambergrown Vicia leaves kept under constant light. Light fluence rate at the adaxial leaf surface was 500 µmol m 2 s 1. Chamber CO 2 levels were increased from about 400 to 1000 cm 3 m 3 in 200 cm 3 m 3 increments, and then decreased to 400 cm 3 m 3 in a stepwise fashion. Each [CO 2 ] was maintained for 1 h prior to measurements. Arrows indicate the sequence of the [CO 2 ] treatments. Each value represents an average of three experiments ± SE. DISCUSSION Under the constant, relatively low light levels of the growth chamber environment, mesophyll from Vicia leaves had no detectable zeaxanthin, as shown previously for growth chamber-grown cotton (Brugnoli & Björkman 1992). In contrast, guard cells had high zeaxanthin content at ambient [CO 2 ], comprising about 45% of the xanthophyll cycle pool (Fig. 3). These high levels of guard cell zeaxanthin are comparable to those found at midday in greenhouse-grown guard cells (Srivastava & Zeiger 1995b). The xanthophyll cycle of guard cell chloroplasts was very sensitive to changes in [CO 2 ], and guard cell zeaxanthin decreased to about 20% of the xanthophyll cycle pool when [CO 2 ] increased from 400 to 1000 cm 3 m 3. Concomitant increases in violaxanthin content indicated that the xanthophyll cycle was modulated by the changes in ambient [CO 2 ]. There were no detectable changes in the xanthophyll cycle pigments of mesophyll chloroplasts in response to these CO 2 manipulations. Studies of the xanthophyll cycle of mesophyll chloroplasts have shown that, at high irradiances, a decrease in C i stimulates violaxanthin de-epoxidation and causes an increase in zeaxanthin content (Gilmore & Björkman 1994), and in non-photochemical fluorescence quenching (Demmig-Adams & Adams 1992). A similar mechanism Increase in stomatal aperture (µm) light dark CO 2 (cm 3 m 3 ) DTT +DTT DTT +DTT ± 0 15a 0 63 ± 0 05a* 0 61 ± 0 09* 0 56 ± 0 07* ± 0 25b 1 55 ± 0 09b** 1 37 ± 0 20** 1 45 ± 0 32** a Significantly different (one-way ANOVA; F = , P = 0 020; at the 0 05 level). b Significantly different (one-way ANOVA; F = , P = 0 004; at the 0 05 level). *Not significantly different (one-way ANOVA; F = 0 231, P = 0 799; at the 0 05 level). **Not significantly different (one-way ANOVA; F = 0 134, P = 0 876; at the 0 05 level). Table 1. Stomatal response to CO 2 in abaxial epidermal peels incubated in the light or in darkness in the presence or absence of 3 mol m 3 DTT. Epidermal peels from dark-adapted (2 5 h) leaves were incubated in a medium (1 0 mol m 3 KCl, 0 1 mol m 3 CaCl 2, and 1 0 mol m 3 Mes/NaOH ph 6 8) aerated with 800 cm 3 m 3 CO 2 -containing air in either 150 µmol m 2 s 1 white light or in darkness for 1 h and then transferred to a medium in equilibrium with 800, 400 and 0 cm 3 m 3 CO 2 -containing air for an additional 1 hr. DTT was added to the medium at the end of the first hour of incubation. The values shown are the net aperture changes measured under 0 or 400 cm 3 m 3 CO 2 in comparison with the aperture under 800 cm 3 m 3 CO 2 at the end of the 2 h incubation. Each value represents an average of three experiments ± SE

6 818 J. Zhu et al. Figure 5. (a) Effect of DTT on CO 2 -dependent changes in guard cell zeaxanthin in the light; (b) relationship between guard cell zeaxanthin and stomatal apertures in abaxial epidermal peels as a function of [CO 2 ] in the light in the absence of DTT. Conditions were as described in Table 1. The DTT concentration was 3 mol m 3. Each value represents an average of four experiments ± SE. appears to operate at low irradiance in chloroplasts of abaxial guard cells (Figs 2 & 3). The large changes in guard cell zeaxanthin content in response to changes in ambient [CO 2 ] suggest correspondingly large, CO 2 -mediated effects on lumen ph (Hager 1980). Changes in [CO 2 ] could alter lumen ph via changes in the rate of consumption of ATP and NADPH by the carboxylation reaction catalysed by Rubisco (Gotow et al. 1988; Cardon & Berry 1992). A second CO 2 -sensitive sink for reducing equivalents from the guard cell chloroplast could be provided by PEP carboxylase activity in the cytosol. A 3-phosphoglycerate/dihydroxyacetone phosphate shuttle has been suggested to export ATP and reducing equivalents from the chloroplast to the cytosol (Shimazaki et al. 1989) and would alter the energy status of the thylakoid membrane in a [CO 2 ]-dependent manner. Under constant light, temperature and relative humidity, changes in stomatal apertures were linearly related to changes in guard cell zeaxanthin content over the cm 3 m 3 range of [CO 2 ] in intact leaves (Fig. 2b) and under incubation in 0, 400, and 800 cm 3 m 3 CO 2 in detached epidermis (Fig. 5b). In intact leaves, the changes were reversible (Fig. 4) and both aperture and zeaxanthin changes saturated in the same range of [CO 2 ] (Figs 2 and 4). This response pattern argues for a causal relationship between the two parameters. Stomatal apertures in intact leaves also tracked ambient [CO 2 ] in the dark, in the absence of changes in guard cell zeaxanthin content (Fig. 2a). As reported in previous studies, the stomatal response to [CO 2 ] in the dark was smaller than in the light (Heath & Russell 1954; Morison & Jarvis 1983; Wong et al. 1987). In epidermal peels kept under constant irradiation, the net increase in stomatal aperture in response to a decrease in [CO 2 ] from 800 to 0 cm 3 m 3 CO 2 was about twice as large as that measured in the dark (Table 1). Decreases in [CO 2 ] caused an increase in guard cell zeaxanthin content in the light (Fig. 5a) but not in the dark (Fig. 2a). The inhibitor of violaxanthin de-epoxidase, DTT, prevented the CO 2 -dependent zeaxanthin increases in the light (Fig. 5a) and inhibited net aperture increases to the levels observed in the dark controls (Table 1). Thus, both in the intact leaf and in epidermal peels, net changes in apertures in response to CO 2 were larger in the light than in the dark, and were accompanied by changes in guard cell zeaxanthin content only in the light. The tight coupling between CO 2 -dependent aperture and zeaxanthin changes in the light, the co-directionality of the aperture and zeaxanthin changes, and the inhibition by DTT of the CO 2 -dependent zeaxanthin increase and of the aperture increases in the light to the level observed in the dark, suggest that zeaxanthin could play a role in the sensory transduction of CO 2 signals in guard cells. Zeaxanthin would transduce prevailing [CO 2 ] at the guard cell chloroplast into modulated stomatal apertures in the light, while the stomatal response to CO 2 in darkness would be dependent on a second, zeaxanthin-independent sensory transduction pathway (Hedrich & Marten 1993). Pharmacological, action spectroscopy, and genetic studies have recently implicated zeaxanthin as a blue light sensor in guard cells (Srivastava & Zeiger 1995a; Quiñones et al. 1996; Zeiger & Zhu 1998). Interactions between the stomatal responses to blue light and CO 2 have been reported (Assmann 1988; Lascève et al. 1993), and measurements of guard cell zeaxanthin content over a daily course of stomatal movements in a greenhouse have shown that zeaxanthin content in guard cells tracked incident radiation to the leaf and was closely correlated with stomatal apertures (Srivastava & Zeiger 1995b). Available data indicate that the xanthophyll cycle of guard cells has the same biochemical properties as its mesophyll counterpart (Karlsson et al. 1992; Masamoto et al. 1993; Srivastava & Zeiger 1995b) but it is regulated in a quite different way. On a chlorophyll basis, guard cell chloroplasts have a nearly 4-fold higher capacity for electron transport and oxygen evolution, substantially lower rates of carbon fixation, and a larger LHCII unit size (Zeiger et al. 1981; Shimazaki & Zeiger 1987; Mawson 1993; Zhu et al. 1997) than mesophyll chloroplasts. High rates of electron transport and low rates of carbon fixation can be expected to favour lumen acidification and zeaxanthin formation at low photon

7 Guard cell zeaxanthin and stomatal response to CO flux densities, as observed (Srivastava & Zeiger 1995b). Stomata from greenhouse-grown leaves have an enhanced light sensitivity (Talbott et al. 1996) and the xanthophyll cycle of these guard cells is more sensitive to light than its mesophyll counterpart (Srivastava & Zeiger 1995b). Stomata from growth chamber-grown leaves have an enhanced CO 2 sensitivity, and the xanthophyll cycle of these guard cells is dramatically more sensitive to CO 2 than that of mesophyll (Fig. 3). These contrasting properties underscore the regulatory differences of the xanthophyll cycle from mesophyll and guard cell chloroplasts, which is presumably associated with the distinct function of the two chloroplast types: carbon fixation in the mesophyll, and sensory transduction in the guard cell. Modulation of the xanthophyll cycle of guard cells by both light and ambient [CO 2 ] would integrate light and CO 2 sensing at the guard cell chloroplast into a single mechanism mediating light CO 2 interactions on stomatal apertures (Morison 1987). In such a mechanism, the modulation of guard cell zeaxanthin content by lumen ph could sense environmental and metabolic signals regulating stomatal apertures via their effect on lumen ph. An empirically verifiable implication of this hypothesis is that recently characterized acclimations of the stomatal responses to blue light and to CO 2 (Assmann 1992; Frechilla et al. 1997) could underlie changes in the sensitivity of the xanthophyll cycle of guard cells to light and CO 2. A major question emerging from the zeaxanthin hypothesis is the nature of the mechanism transducing zeaxanthin-mediated signal processing within the chloroplast to extrachloroplastic targets elsewhere in the guard cell. Light sensing at the guard cell chloroplast has been shown to mediate red light-stimulated outward electrical currents from guard cell protoplasts (Serrano et al. 1988), and the chloroplast appears to be the site of perception of blue light in blue light-stimulated stomatal opening (Srivastava & Zeiger 1995a). A metabolite of guard cell photosynthesis (Serrano et al. 1988; Cardon & Berry 1992) and calcium fluxes across the chloroplast envelope (Kinoshita et al. 1995) have been proposed as possible second messengers. Recent studies have shown that cytosolic free calcium appears to be a component of the CO 2 signal transduction pathway in stomatal guard cells (Webb et al. 1996). Further studies on the role of zeaxanthin on signal transduction in guard cells could provide insight on this important question. REFERENCES Assmann S.M. (1988) Enhancement of the stomatal response to blue light by red light, reduced intercellular concentrations of CO 2, and low vapor pressure differences. Plant Physiology 87, Assmann S.M. (1992) Effects of light quantity and quality during development on the morphology and stomatal physiology of Commelina communis. Oecologia 92, Assmann S.M. (1993) Signal transduction in guard cells. Annual Review of Cell Biology 9, Bilger W. & Björkman O. (1991) Temperature dependence of violaxanthin de-epoxidation and non-photochemical fluorescence quenching in intact leaves of Gossypium hirsutum L. & Malva parviflora L. Planta 184, Brugnoli E. & Björkman O. (1992) Growth of cotton under continuous salinity stress: Influence on allocation pattern, stomatal and non-stomatal components of photosynthesis and dissipation of excess light energy. Planta 187, Cardon Z.G. & Berry J. (1992) Effects of O 2 and CO 2 concentration on the steady-state fluorescence yield of single guard cell pairs in intact leaf discs of Tradescantia albiflora. Plant Physiology 99, Demmig-Adams B. (1990) Carotenoids and photoprotection in plants. A role of the xanthophyll zeaxanthin. Biochimica et Biophysica Acta 1020, Demmig-Adams B. & Adams W.W. III (1992) Photoprotection and other responses of plants to high light stress. Annual Review of Plant Physiology and Plant Molecular Biology 43, Frechilla S., Talbott L.D. & Zeiger E. (1997) The blue light response of guard cells acclimates to growth environment. Plant Physiology 114 (Suppl.), 98. Gilmore A.M. & Björkman O. (1994) Adenine nucleotides and the xanthophyll cycle in leaves I. Effects of CO 2 - and temperaturelimited photosynthesis on adenylate energy charge and violaxanthin de-epoxidation. Planta 192, Gilmore A.M. & Yamamoto H.Y. (1991) Resolution of lutein and zeaxanthin using a non-endcapped lightly carbon loaded C-18 high performance liquid chromatographic column. Journal of Chromatography 543, Hager A. (1980) The reversible, light-induced conversions of xanthophylls in the chloroplasts. In Pigments in Plants (ed. F-C. Czygan), pp Gustav Fisher, Stuttgart. Heath O.V.S. & Russell J. (1954) Studies in stomata behaviour. VI. An investigation of the light response of wheat stomata with the attempted elimination of control by the mesophyll. Part 2: Interactions with external carbon dioxide and general discussion. Journal of Experimental Botany 5, Hedrich R. & Marten I. (1993) Malate-induced feedback regulation of plasma membrane anion channels could provide a CO 2 sensor to guard cells. The EMBO Journal 12, Karlsson P.E., Bogomolni R.A. & Zeiger E. (1992) High performance liquid chromatography of pigments from guard cell protoplasts and mesophyll tissue of Vicia faba L. Photochemistry and Photobiology 55, Kinoshita T., Nishimura M. & Shimazaki K.I. (1995) Cytosolic concentration of Ca 2+ regulates the plasma membrane H + - ATPase in guard cells of fava bean. The Plant Cell 7, Lascève G., Gautler H., Jappé J. & Vavasseur A. (1993) Modulation of blue light response of stomata of Commelina communis by CO 2. Physiologia Plantarum 88, Lorimer G.H., Badger M.R. & Andrews T.J. (1976) The activation of ribulose 1,5-bisphosphate carboxylase by carbon dioxide and magnesium ions. Equilibria, kinetics, a suggested mechanism, and physiological implications. Biochemistry 15, Masamoto K., Kinoshita T. & Shimazaki K. (1993) Light-induced de-epoxidation of violaxanthin in guard cell protoplasts of Vicia faba. Plant Cell Physiology 34, Mawson B.T. (1993) Modulation of photosynthesis and respiration in guard and mesophyll cell protoplasts by oxygen concentration. Plant, Cell and Environment 16, Morison J.I.L. (1987) Intercellular CO 2 concentration and stomatal response to CO 2. In Stomatal Function (eds E. Zeiger, G.D. Farquhar & I.R. Cowan), pp Stanford University Press, Stanford. Morison J.I.L. & Jarvis P.G. (1983) Direct and indirect effects of light on stomata, II. In Commelina communis L. Plant, Cell and Environment 6, Mott K.A. (1990) Sensing of atmospheric CO 2 by plants. Plant, Cell and Environment 13,

8 820 J. Zhu et al. Quiñones M.A., Lu Z. & Zeiger E. (1996) Close correspondence between the action spectra for the blue light responses of the guard cell and coleoptile chloroplasts, and the spectra for blue light-dependent stomatal opening and coleoptile phototropism. Proceedings of the National Academy Sciences, USA 93, Raschke K. (1975) Simultaneous requirement of carbon dioxide and abscisic acid for stomatal closing in Xanthium strumarium L. Planta 125, Raschke K., Hedrich R., Reckmann U. & Schroeder J.I. (1988) Exploring biophysical and biochemical components of the osmotic motor that drives stomatal movement. Botanica Acta 101, Serrano E., Zeiger E. & Hagiwara S. (1988) Red light stimulates an electrogenic proton pump in Vicia guard cell protoplasts. Proceedings of the National Academy Sciences, USA 85, Sharkey T.D. & Raschke K. (1981) Separation and measurement of direct and indirect effects of light on stomata. Plant Physiology 68, Shimazaki K., Gotow K. & Kondo N. (1982) Photosynthetic properties of guard cell protoplasts from Vicia faba L. Plant Cell Physiology 23, Shimazaki K.I., Terada J., Tanaka K. & Kondo N. (1989) Calvin- Benson cycle enzymes in guard-cell protoplasts from Vicia faba L. Plant Physiology 90, Shimazaki K. & Zeiger E. (1987) Red light-dependent CO 2 uptake and oxygen evolution in guard cell protoplasts of Vicia faba L. Evidence for photosynthetic CO 2 fixation. Plant Physiology 84, 7 9. Srivastava A. & Zeiger E. (1992) Fast fluorescence quenching from isolated guard cell chloroplasts of Vicia faba is induced by blue light and not by red light. Plant Physiology 100, Srivastava A. & Zeiger E. (1995a) The inhibitor of zeaxanthin formation, dithiothreitol, inhibits blue-light-stimulated stomatal opening in Vicia faba. Planta 196, Srivastava A. & Zeiger E. (1995b) Guard cell zeaxanthin tracks photosynthetically active radiation and stomatal apertures in Vicia faba leaves. Plant, Cell and Environment 18, Talbott L.D., Srivastava A. & Zeiger E. (1996) Stomata from growth-chamber-grown Vicia faba have an enhanced sensitivity to CO 2. Plant, Cell and Environment 19, Talbott L.D. & Zeiger E. (1993) Sugar and organic acid accumulation in guard cells of Vicia faba in response to red and blue light. Plant Physiology 102, Webb A.A.R., McAinsh M.R., Mansfield T.A. & Hetherington A.M. (1996) Carbon dioxide induces increases in guard cell cytosolic free calcium. The Plant Journal 9, Wong S.C., Cowan I.R. & Farquhar G.D. (1987) Leaf conductance in relation to assimilation in Eucalyptus pauciflora Sieb. Ex Spreng. Plant Physiology 62, Zeiger E. (1990) Light perception in guard cells. Plant, Cell and Environment 13, Zeiger E., Armond P. & Melis A. (1981) Fluorescence properties of guard cell chloroplasts. Evidence for linear electron transport and light-harvesting pigments of photosystems I and II. Plant Physiology 67, Zeiger E., Gotow K., Mawson B. & Taylor S. (1987) The guard cell chloroplast: properties and function. In Proceedings of the 7th International Photosynthesis Congress, Vol. 4 (ed. J. Biggins), pp Martinus Nijhoff, Dordrecht. Zeiger E. & Zhu J. (1998) Role of zeaxanthin in blue light photoreception and the modulation of light CO 2 interactions in guard cells. Journal of Experimental Botany 49, Zhu J., Gomez S.M., Mawson B.T., Jin X. & Zeiger E. (1997) The coleoptile chloroplast: Distinct distribution of xanthophyll cycle pigments, and enrichment in Photosystem II. Photosynthesis Research 51, Zhu J., Zeiger R. & Zeiger E. (1995) Structural and functional properties of the coleoptile chloroplast: photosynthesis and photosensory transduction. Photosynthesis Research 44, Received 6 October 1997; received in revised form 27 February 1998; accepted for publication 2 April 1998