System performance in measurements with single MCPs and GCPs
The measuring system allows suitable samples to be inspected first via the ocular of the microscope and then, for the actual fluorescence measurements, the field of view to be narrowed down with the help of an iris diaphragm. In this way, the background signal is minimized and it can be assured that the assessed fluorescence originates almost exclusively from the selected cells. Figure 2a shows typical recordings from single protoplasts, which differ in size and chloroplast number. The corresponding photographs of a Vicia MCP (top), GCP (centre) and Arabidopsis GCP (bottom) are displayed in Fig. 2b. Both GCP and MCP showed the well-known dark–light induction kinetics (Kautsky effect). Additional information on fluorescence quenching and, hence, on the status of the chloroplasts, was obtained by repetitive application of saturating light pulses. The signal/noise ratio (S/N, evaluated during a saturation pulse applied to a dark-adapted sample) was very satisfactory not only with a single MCP (S/N > 100), but also with a GCP of Vicia (S/N = 50). For the latter, however, in order to reach the same signal amplitude as with the MCP, amplification had to be increased by a factor of 27·4. Similar fluorescence information was obtained with an extraordinary small single GCP (diameter = 6·7 μm) of Arabidopsis. In this case the fluorescence intensity was ≈ 180 times lower than with a MCP of Vicia and the S/N was only ≈ 5. While this S/N is too low for quantitative assessment of fluorescence parameters by a single recording, it can be readily improved by averaging a number of such recordings, a feature provided by the WINCONTROL software which is distributed with the microscopy–PAM fluorometer (data not shown).
Figure 2. . Typical recordings of dark–light induction transients of single protoplasts with fluorescence quenching analysis by the saturation pulse method and corresponding micrographs of the investigated samples. (a) Fluorescence responses of Vicia mesophyll cell protoplasts (MCP; top), Vicia guard cell protoplasts (GCP; centre) and Arabidopsis GCP (bottom). The measuring light (ML) intensity was 0·8 μmol m−2 s−1 of photosynthetically active radiation. Saturating light pulses (SP) inducing maximal fluorescence yields, Fm and Fm′, were applied at 40 s following the onset of ML and every 20 s after onset of actinic light (AL). The intensity of saturation pulses was 4280 and 1550 μmol m−2 s−1 for Vicia MCP/GCP and Arabidopsis GCP, respectively (see text for further details). (b) The attached micrographs show typical sizes and chloroplast contents of Vicia MCP (top), Vicia GCP (centre) and Arabidopsis GCP (bottom) as seen under a light microscope. The bar represents 10 μm.
Download figure to PowerPoint
In the following, the fluorescence parameters measured by the new approach will be briefly explained and a comparison with earlier single guard cell fluorescence recordings will be made. When the weak measuring light is switched on (ML on), the minimal fluorescence yield, Fo, is assessed which is characteristic for the dark-adapted state with all PS II reaction centres being open (see Fig. 2a). It is important to note that the measuring beam intensity is sufficiently low to allow continuous recording of Fo, i.e. there is no accumulation of reduced acceptors at PS II, which would lead to an induction phenomenon, as in the original work of Zeiger and coworkers ( Zeiger et al. 1980 ; Melis & Zeiger 1982). Despite such low measuring light, the Fo information is obtained within fractions of a second, which compares favourably with the 15 min time required for obtaining an Fo image by the technique of Oxborough & Baker (1997). Upon application of a brief (0·8 s) saturating light pulse, the fluorescence yield of the dark-adapted sample is transiently increased to its maximal level, Fm. The increase of fluorescence yield from Fo to Fm is called variable fluorescence, Fv, and the ratio Fv/Fm has been shown to correspond to the potential maximum quantum yield of light energy conversion in PS II ( Butler 1978; Björkman 1987). After fluorescence yield has returned close to the Fo level, actinic light is turned on (AL on) and repetitive saturation pulses are applied (at 20 s intervals) to assess the change of maximal fluorescence yield during illumination (Fm′). The Kautsky effect observed in the GCP followed the pattern of a typical dark–light induction curve well known from numerous studies on whole leaves ( Briantais et al. 1986 ; Krause & Weis 1991; Schreiber et al. 1994 ). There were no major differences between the curves of GCP (centre panel) and MCP (top panel), contrary to earlier single-cell measurements of Melis & Zeiger (1982) and Mawson & Zeiger (1991). While the definite cause of this discrepancy is not clear, it may be relevant that in the study of Melis & Zeiger (1982) leaf segments were enclosed in a 1 mm thick cuvette, and in the study of Mawson & Zeiger (1991) epidermal peels were enclosed between coverslips. In view of the high dark respiratory O2 uptake rate and the pronounced O2 requirement for photosynthetic electron transport in guard cells ( Mawson 1993), it appears possible that in these earlier studies the induction curves of guard cells were influenced by local, partial anaerobiosis (see also the section on guard cells under anaerobiosis below).
In the PAM measurements of Fig. 2a, upon the onset of continuous actinic light, fluorescence yield first rapidly rises to a peak, P, and then more slowly declines to a quasi-stationary level, S, before it briefly rises to a secondary maximum, M, and then finally reaches the terminal steady-state level, T. It may be mentioned that before the advent of PAM fluorimetry the M-peak phenomenon and the slow fluorescence decline from M to T were interpreted in terms of Calvin cycle activity and proton gradient formation, respectively. Such a phenomenological approach, however, has been replaced by a more quantitative assessment using the saturation pulse quenching analysis ( Walker 1992). The repetitive fluorescence increases (Fm′–F) induced by each saturation pulse contain the essential information for quenching analysis, resulting in assessment of the PS II quantum yield and energy status of the chloroplast ( Genty et al. 1989 ). Any decrease of Fm′ with respect to the original Fm reflects non-photochemical quenching, whereas the difference Fm′–F is caused by photochemical quenching, which is a relative measure of open PS II reaction centres.
The so-called quenching coefficients, qN and qP, were defined ( Schreiber et al. 1986 ; Bilger & Schreiber 1986), the calculation of which, however, requires knowledge of minimal fluorescence yield, Fo′, in a given state of preillumination. In order to assess Fo′, the actinic light must be turned off and the electron pool at the acceptor side of PS II must be quickly reoxidized with the help of far-red light, before relaxation of non-photochemical quenching sets in. In the present study, no use of far-red light could be made because of excessive disturbance of the photomultiplier. Therefore, instead of qN and qP, the fluorescence parameters NPQ ( Bilger & Björkman 1990) and ΔF/Fm′ ( Genty et al. 1989 ) are presented, the calculation of which does not require knowledge of Fo′.
In intact leaves and isolated chloroplasts, a major component of non-photochemical quenching is energy-dependent quenching, qE, which is caused by acidification of the thylakoid internal space. It can be distinguished from other forms of non-photochemical quenching by its rapid reversal upon light-off. In the measurements of Fig. 2a, in order to gain information on the reversal of non-photochemical quenching, saturation pulses were also applied during the dark period following light-off. The dark intervals between the pulses were stepwise increased in order to minimize the illumination effect. It is apparent that Vicia GCP displayed substantial light-induced non-photochemical quenching, most of which was reversed within 1 min following light-off ( Fig. 2a, centre). Unless guard cell fluorescence is governed by different quenching mechanisms than mesophyll fluorescence, which appears highly unlikely in view of the very similar phenomenology of fluorescence induction, these data argue for substantial energy-dependent quenching in guard cells. Notably, also in guard cells during the first minute of illumination there is a pronounced decrease of Fm′, with partial recovery during prolonged illumination. In leaves, the recovery of high Fm′ values depends on the presence of CO2 and Calvin cycle activity ( Schreiber et al. 1986 ). Hence, this phenomenon is believed to reflect the onset of the Calvin cycle following light activation, when ATP consumption is initiated and consequently there is a drop in transthylakoidal ΔpH. If this rationale is also applied to guard cell chloroplasts, they would appear to display very similar membrane energization by photosynthetic electron flow coupled to vectorial proton translocation, activation of Calvin cycle enzymes and steady-state CO2-dependent electron flow. However, more extensive work will be required to validate this conclusion.
Quenching analysis, effective quantum yield and relative electron transport rate (ETR)
The microscopy–PAM is equipped with an extensive data acquisition system (PAM control EPROM in conjunction with WINCONTROL software) which allows automated assessment of the relevant fluorescence levels and on-line calculation of the essential fluorescence-derived parameters. These include the effective quantum yield of PS II, the so-called Genty parameter ΔF/Fm′ = (Fm′–F)/Fm′ ( Genty et al. 1989 ), the apparent relative ETR = ΔF/Fm′× PAR ×c (where PAR is the photon flux density of incident photosynthetically active radiation, and the constant c corresponds to the absorption factor) ( Schreiber et al. 1994 ) and the NPQ parameter of non-photochemical quenching, as defined by (Fm–Fm′)/Fm′ ( Bilger & Björkman 1990). As to the fluorescence-derived ETR parameter, it should be emphasized that it provides a relative measure of ETR only. The calculation of absolute rates would not only require knowledge of the fraction of incident quanta absorbed by the guard cell chloroplasts, but also of the energy distributed to PS II. The various fluorescence-derived parameters were determined repetitively during the course of a dark–light induction curve or after each of a series of consecutive illumination steps at increasing light intensities (so-called light curves). Such dark–light induction curves and light curves were recorded several times using the same computer-controlled illumination program and then the data were averaged. In this way, the relatively large variability observed between different protoplasts of the same preparation was removed and systematic differences could be assessed. In Fig. 3, averaged values of ΔF/Fm′, ETR, and NPQ measured during dark–light induction are presented for Vicia GCP and MCP. It should be pointed out that the data variability, which is expressed in the mean ± standard deviation (SD), was mainly due to actual differences in the photosynthetic activity of the individual protoplasts, rather than to recording noise or errors in the quenching analysis. This may be concluded from the fact that ± SD amounted to 10–50%, particularly large in the case of NPQ, whereas the S/N ratio was in the order of 50 (corresponding to ≈ 2% noise; see original data in Fig. 2a and Figs 4–6). Actually, when, for example NPQ was measured repetitively with the same sample with 1 min dark intervals between consecutive measurements, ± SD was only 1·5% for n = 9 (data not shown).
Figure 3. . Characteristic fluorescence parameters of Vicia guard cell protoplasts (GCP) and mesophyll cell protoplasts (MCP) as assessed during dark–light induction by quenching analysis. (a) Effective photosystem II quantum yield, ΔF/Fm′ (Genty parameter). (b) Relative electron transport rate, ETR. (c) Parameter of non-photochemical quenching, NPQ. The constant c in the equation for the ETR calculation (see text) was assumed to be 0·42. Actinic intensity was 67 μmol m−2 s−1. The displayed data are the means ± standard deviation (SD) for MCP (n = 13) and GCP (n = 16). For other conditions, see legend to Fig. 2 and Materials and methods.
Download figure to PowerPoint
Figure 4. . Effect of anaerobiosis on the fluorescence characteristics of Vicia guard cell protoplasts (GCP) during dark–light induction with repetitive application of saturation pulses. (a) Original recording, control sample. (b) Original recording, anaerobic sample after treatment with glucose/glucose oxidase (see MATERIALS AND METHODS). (c) On-line calculated values of effective quantum yield, ΔF/Fm′, for control (open squares) and anaerobic sample (closed squares). Other conditions, see legend to Fig. 2.
Figure 5. . Effect of dark incubation of a Vicia epidermal peel underneath a microscope coverslip. (a) Control in the open system for 30 min. (b) After 30 min incubation in the closed system. The temperature was ≈ 25 °C. For other conditions, see legend to Fig. 2 and Materials and methods.
Download figure to PowerPoint
Figure 6. . Effect of fusicoccin (FC) on the fluorescence characteristics of single Vicia guard cell pairs in an epidermal peel. (a) Control; (b) presence of FC. The peels were incubated in a buffered KCl solution for 3·5 h in the dark, with and without 6·7 mmol m−3 FC being present (see MATERIALS AND METHODS). In both cases, the medium contained 0·27% dimethylsulphoxide (DMSO). Actinic intensity was 67 μmol m−2 s−1 and saturation pulses had an intensity of 2904 μmol m−2 s−1.
Download figure to PowerPoint
A major finding, which in principle was already apparent from the single recordings in Fig. 2a, was that the observed fluorescence parameters and their general induction patterns were quite similar for MCP and GCP. While ΔF/Fm (maximal variable fluorescence of dark-adapted sample, also referred to as Fv/Fm) was close to 0·8 (the maximal potential quantum yield of charge separation in PS II) in MCP as well as in GCP, the effective quantum yield of PS II during illumination, ΔF/Fm′, was systematically lower in GCP than in MCP ( Fig. 3a). Corresponding differences were also apparent in the ETR data ( Fig. 3b). During the course of an induction curve, the difference was most pronounced ≈ 40 s following the onset of illumination. The light-induced rise of NPQ was almost identical in MCP and GCP ( Fig. 3c). However, relaxation of NPQ during illumination (reflecting Calvin cycle activation) was retarded and less pronounced in GCP. Furthermore, differences were observed in the kinetics of dark relaxation of NPQ, with the rapid phase (presumably reflecting the decay of energy-dependent quenching) showing a larger amplitude in GCP.
In Fig. 7, so-called rapid light curves (RLC) of Vicia GCP and MCP are compared. Such RLC, which involve relatively short illumination periods at increasing light intensities, give insight into the light saturation properties of a sample ( Schreiber et al. 1997 ; White & Critchley 1999). When recorded with a light-adapted sample, RLC data reflect the dependence of the relative photosynthetic ETR on quantum flux density in a given state of light adaptation. These data showed even more clearly than the ETR data of Fig. 3b that the electron transport capacity of GCP is distinctly lower than that of MCP, with half saturation at 58 μmol quanta m−2 s−1 PAR in GCP and at 132 μmol quanta m−2 s−1 PAR in MCP. Such measurements involving the determination of effective PS II quantum yield and the calculation of relative ETR in single guard cells were not possible using the non-modulated fluorescence technique of earlier work (e.g. Mawson & Zeiger 1991). While similar information under quasi-steady-state conditions could well be obtained with the help of the special fluorescence imaging system of Oxborough & Baker (1997), such measurements so far have not been reported. However, it has already been noted by Shimazaki & Zeiger (1985) that non-cyclic photophosphorylation saturates at lower light intensities in guard cells than in mesophyll cells.
Figure 7. . Rapid light response curves of the relative electron transport rate, ETR, of Vicia guard cell protoplasts (GCP) and mesophyll cell protoplasts (MCP). Actinic light was applied during consecutive 30 s periods with stepwise increasing intensity. At the end of each illumination period a saturation pulse was applied to assess the effective quantum yield on which the calculation of ETR is based. The data are displayed as means ± standard deviation (SD) for GCP (n = 11) and MCP (n = 11). Other conditions as for Fig. 2.
Download figure to PowerPoint
It is known that on a chlorophyll basis the O2 evolution rate in GCP exceeds that of MCP by a factor of 3–4 ( Shimazaki, Gotow & Kondo 1982; Shimazaki 1989). This difference, however, cannot be reflected by the present ETR measurements, as without knowledge of the absolute fraction of incident light reaching PS II, ETR is just a relative measure of electron transport (see above).
Fluorescence characteristics of single guard cells under anaerobiosis
The fluorescence-derived ETR values which reflect the relative rate of energy conversion in PS II ( Genty et al. 1989 ; Schreiber et al. 1994 ), do not necessarily correspond to the rate of CO2 fixation. There are other forms of electron transport passing PS II, such as the Mehler ascorbate peroxidase (MAP) cycle which depends on O2 instead of CO2 as the electron acceptor [for reviews see Schreiber et al. (1995) and Asada et al. (1998) ]. Actually, Gotow et al. (1988) reported that the rate of CO2 fixation was only 8% of photosynthetic O2 evolution in GCP. And Mawson (1993) showed that the net rate of photosynthetic O2 evolution in V. faba GCP was strongly O2 dependent. As shown in Fig. 4, this earlier finding is impressively confirmed by chlorophyll fluorescence quenching analysis in single-cell measurements (compare traces in Fig. 4a & b). After removal of O2 using the glucose/glucose oxidase trap ( Fig. 4b), part of the PS II reaction centres are already closed in the dark, as revealed by a doubling of the apparent Fo and a corresponding decrease of the apparent Fv/Fm. Photochemical quenching is rapidly eliminated upon onset of actinic illumination. Saturation pulses induce fluorescence quenching instead of increases of fluorescence. The latter feature suggests that the primary acceptor of PS II, QA, is fully reduced by the actinic light and that there is transient accumulation of reduced pheophytin during saturating light pulses ( Heber et al. 1985 ). A total block of PS II activity in the absence of O2 is also reflected by the calculated values of effective PS II quantum yield (ΔF/Fm′) as shown in Fig. 4c.
A central role of O2-dependent electron flow was also apparent when epidermal peels were submersed underneath a microscope coverslip. Due to strong O2 uptake by respiration in guard cells ( Mawson 1993) and limited O2 diffusion from the surrounding air, within 10–30 min (depending on the temperature and the size of the epidermal fragments) the fluorescence responses of the submersed guard cells indicated strong inhibition of photosynthetic electron flow ( Fig. 5). It is conceivable that intermediate stages of anaerobiosis will yield fluorescence responses intermediate between those shown in Fig. 5a and b. While the severe inhibition of photosynthetic electron flow is clearly revealed by the saturation pulse technique, the continuous fluorescence response, which essentially was measured in earlier work on single guard cells ( Melis & Zeiger 1982; Mawson & Zeiger 1991), just shows a slow down of the fluorescence decline from P to T.
These data suggest that O2 is a competent electron acceptor in guard cell chloroplasts, as has been previously shown for intact leaves and spinach chloroplasts. In the latter case, an important role of the MAP cycle in ΔpH generation and light protection has been demonstrated ( Schreiber et al. 1995 ). It can be readily shown that there is ascorbate peroxidase activity in Vicia GCP, as addition of H2O2 (0·5 mol m−3) to an O2-limited sample in the presence of ascorbate (5 mol m−3) restored photochemical quenching (data not shown), similar to previous observations with spinach chloroplasts ( Neubauer & Schreiber 1989). However, an alternative explanation for the pronounced O2 requirement of guard cell photosynthetic electron flow must be considered on the basis of the work of Mawson (1993). The latter showed that in GCP the decrease of photosynthetic O2 evolution with decreasing O2 concentration is closely correlated with a decrease of respiratory O2 uptake, suggesting close metabolic coupling between the two activities. Hence, it appears feasible that photosynthetic electron flow in guard cells depends on the export of reducing equivalents from the chloroplasts to the cytosol via a PGA/DHAP shuttle system ( Shimazaki et al. 1989 ), which are oxidized by oxidative phosphorylation in the mitochondria. Oxidation of photosynthetically derived reducing equivalents by mitochondria has also been previously suggested for C3 mesophyll cells ( Krömer, Stitt & Heldt 1988). On the basis of the present data, we are unable to favour one of the two alternatives. A third alternative was suggested by Cardon & Berry (1992) who studied the steady-state fluorescence yield of single guard cells at 21% and 2% O2 in dependence of CO2 concentration. These authors interpreted a quenching of the steady-state fluorescence yield at a low CO2 concentration upon transition from 21 to 2% O2 as evidence for photorespiration in guard cells, analogous to related changes in mesophyll cells. Whether or not this suggestion is justified should be verified by corresponding experiments with the microscopy–PAM making use of fluorescence quenching analysis. In this context, it should be pointed out that the present study reports on the responses of isolated protoplasts and guard cells in epidermal peels only, which do not necessarily in all respects reflect the biochemical and physiological responses of guard cells in intact leaves. Therefore, it is intended to extend the present work to intact leaves, for example by studying guard cells in the albino parts of variegated leaves [see earlier work of Melis & Zeiger (1982) and Cardon & Berry (1992)].
While O2 can function as an electron acceptor in the MAP cycle only during illumination (reduction at PS I acceptor side), the data presented in Figs 4 and 5 suggest that it also plays a role in the reoxidation of the intersystem electron transport chain in the dark. The observed increase of Fo argues for reduction of the plastoquinone pool by reduced stroma components which in air is compensated by O2-dependent reoxidation. Such a ‘chlororespiration’ ( Bennoun 1982) has been reported for higher plant chloroplasts ( Garab et al. 1989 ; Gruszecki, Bader & Schmid 1994) and also for algae ( Schreiber & Vidaver 1974; Ting & Owens 1993; Bennoun 1994). The present results suggest a rather strong flux of electrons from reduced substrates to O2 via the plastoquinone pool, which displays the characteristics of chlororespiration.
Effect of FC on single guard cell fluorescence characteristics
The fungal toxin FC is well known to stimulate stomatal opening by hyperactivating the H+-pumping ATPase in the plasma membrane of guard cells ( Lohse & Hedrich 1992). The effect of FC on chlorophyll fluorescence characteristics of single guard cell pairs was studied in epidermal peels of Vicia. As shown in Fig. 6, there was a strong effect of 6·7 mmol m−3 FC on the dark–light fluorescence induction characteristics when applied for 3·5 h. The same amount of FC, when applied over several hours, induced pronounced stomatal opening in the dark, with saturation occurring at 3·5 h. The kinetics of the increase of stomatal aperture during dark incubation with 6·7 mmol m−3 FC are depicted in Fig. 8. Analogously, Fig. 9 shows the gradually developing suppression of effective PS II quantum yield, ΔF/Fm′, measured during continuous illumination (as in the experiment of Fig. 6), after different incubation times with FC (compare corresponding data of stomatal aperture in Fig. 8). Control experiments with intact spinach chloroplasts as well as with Vicia MCP did not show any effect of FC on CO2-dependent electron flow (data not shown). Phenomenologically, the FC effect on GCP fluorescence parameters resembles that of O2 depletion, although not to the same extent as by the glucose/glucose oxidase trap (compare Fig. 4). This was not only true for the loss of photochemical and non-photochemical quenching, but also for the increase of Fo and of the steady-state fluorescence yield.
Figure 8. . Kinetics of the fusicoccin (FC)-induced increase of stomatal aperture observed microscopically in Vicia epidermal peels. Closed symbols, epidermal peels submerged for the indicated time in buffer solution containing 6·7 mmol m−3 FC in the dark. Open symbols, control with epidermal peels submerged in buffer solution without FC in the dark (see MATERIALS AND METHODS). The displayed data are the means ± standard deviation (SD) of n = 3 (60 stomata).
Download figure to PowerPoint
Figure 9. . Effective quantum yield during illumination, ΔF/Fm′, in single Vicia guard cell pairs as a function of the preceding dark incubation time in the absence and presence of 6·7 mmol m−3 fusicoccin (FC). The quantum yield was determined at the end of a 4·7 min illumination period. The displayed data are the means ± standard deviation (SD) of n = 6 (1 h incubation), n = 8 (2 h) and n = 10 (3·5 h). Other experimental conditions as for Fig. 6.
Download figure to PowerPoint
As expected, the effect of FC on the apparent ETR in guard cell chloroplasts (determined by ΔF/Fm′) was prevented by 100 mmol m−3 vanadate, a specific inhibitor of plasma membrane H+-ATPase ( Amodeo, Srivastava & Zeiger 1992) ( Fig. 10). In parallel experiments, vanadate also prevented FC-induced stomatal opening in the dark (data not shown). These findings are in agreement with the suggestion that uncontrolled H+-ATPase activity in response to FC and the resulting depletion of the ATP pool in the cytosol led to the suppression of photosynthetic electron flow in the guard cell chloroplasts. Indeed, when ATP depletion was counteracted by the addition of glucose (2 mol m−3), which stimulates ATP generation by respiration, the FC-induced inhibition of guard cell photosynthetic electron flow was much less pronounced ( Fig. 11). Furthermore, inhibition of oxidative phosphorylation by KCN restored the full effect of FC in the presence of glucose with a half-maximal effect at ≈ 25 mmol m−3 KCN ( Fig. 12).
Figure 10. . Reversal of the fusicoccin (FC) effect on effective quantum yield, ΔF/Fm′, in single Vicia guard cell pairs by vanadate. Incubation was for 3·5 h. Vanadate dissolved in water was added to a final concentration of 100 mmol m−3. Other experimental conditions as for Fig. 6. The displayed data are the means ± standard deviation (SD) of n = 5.
Download figure to PowerPoint
Figure 11. . Effect of glucose on the suppression of ΔF/Fm′ by fusicoccin (FC) in single Vicia guard cell pairs. Incubation was for 3·5 h. Glucose was added to a final concentration of 2 mol m−3. Other experimental conditions as for Fig. 6. The displayed data are the means ± standard deviation (SD) of n = 8.
Download figure to PowerPoint
Figure 12. . Effect of increasing KCN concentration on ΔF/Fm′ in single Vicia guard cell pairs in the presence of fusicoccin (FC) and glucose (a). The control value of ΔF/Fm′ in the absence of glucose is shown in (b). Incubation was for 3·5 h in the dark. Other experimental conditions as for Fig. 10. The displayed data are the means ± standard deviation (SD) of n = 5.
Download figure to PowerPoint
While all the data presented in Figs 6, 8–11 in principle seem to agree with the suggestion that the observed FC effect is related to ATP depletion of the cytosol, it is not clear why such ATP depletion should cause the inhibition of electron transport within the chloroplasts. As an alternative explanation, we have considered the possibility that, as a consequence of cytosolic ATP depletion, there is further stimulation of O2 uptake due to the quasi-uncoupled state of oxidative phosphorylation, with the consequence of local O2 depletion within the guard cells. However, this explanation, which would be in line with the observed phenomenology (compare Figs 5 and 6), appears to be ruled out by the effect of glucose, which restores photosynthetic electron transport in the presence of FC, while at the same time it is expected to stimulate O2 uptake by respiration. Alternatively, the FC effect may be also seen in the context of the complex metabolic coupling between the photosynthetic activity of guard cell chloroplasts and oxidative phosphorylation in guard cell mitochondria, both of which are linked via the cytoplasma with sucrose–glucose metabolism on the one hand and stomatal opening via activation of the proton pumping ATPase, followed by ion uptake and malate biosynthesis on the other. Future studies on mutants defective in the FC response of guard cells, as well as on metabolic and photosynthesis mutants, will help to elucidate the molecular mechanisms underlying the FC effect on photosynthesis.