The contribution of photosynthesis to the red light response of stomatal conductance

To determine the contribution of photosynthesis on stomatal conductance, we contrasted the stomatal red light response of wild type tobacco ( Nicotiana tabacum L. cv W38) with that of plants impaired in photosynthesis by antisense reductions in the content of either cytochrome b 6 f complex (anti- b / f plants ) or Rubisco (anti-Ssu plants). Both transgenic genotypes showed a lowered content of the antisense target proteins in guard cells as well as in the mesophyll. In the anti- b / f plants CO 2 assimilation rates were proportional to leaf cytochrome b 6 f content, but there was little effect on stomatal conductance and rate of stomatal opening. To compare the relationship between photosynthesis and stomatal conductance wild type plants and anti-Ssu plants were grown at 30 and 300 µ mol photon m -2 s -1 irradiance (LL and ML, respectively). Growth in ML increased CO 2 assimilation rates and stomatal conductance in both genotypes. Despite the significantly lower CO 2 assimilation rate in the anti-Ssu plants, the differences in stomatal conductance between the genotypes were non-significant at either growth irradiance. Irrespective of plant genotype, stomatal density in the two leaf surfaces was two-fold higher in ML than in LL-grown plants and conductance normalized to stomatal density was unaffected by growth irradiance. We conclude that the red light response of stomatal conductance is independent of the concurrent photosynthetic rate of the guard cells or of that of the underlying mesophyll. Furthermore we suggest that the correlation of photosynthetic capacity and stomatal conductance observed under different light environments is caused by signals largely independent of photosynthesis.


The contribution of photosynthesis to the red light response of stomatal conductance
Irene Baroli, Dean Price, Murray Badger, and Susanne von Caemmerer* Molecular Plant Physiology Group, Research School of Biological Sciences, Australian National University, Canberra, Australian Capital Territory 2601, Australia

ABSTRACT
To determine the contribution of photosynthesis on stomatal conductance, we contrasted the stomatal red light response of wild type tobacco (Nicotiana tabacum L. cv W38) with that of plants impaired in photosynthesis by antisense reductions in the content of either cytochrome b 6 f complex (anti-b/f plants ) or Rubisco (anti-Ssu plants). Both transgenic genotypes showed a lowered content of the antisense target proteins in guard cells as well as in the mesophyll. In the anti-b/f plants CO 2 assimilation rates were proportional to leaf cytochrome b 6 f content, but there was little effect on stomatal conductance and rate of stomatal opening. To compare the relationship between photosynthesis and stomatal conductance wild type plants and anti-Ssu plants were grown at 30 and 300 µmol photon m -2 s -1 irradiance (LL and ML, respectively). Growth in ML increased CO 2 assimilation rates and stomatal conductance in both genotypes. Despite the significantly lower CO 2 assimilation rate in the anti-Ssu plants, the differences in stomatal conductance between the genotypes were non-significant at either growth irradiance. Irrespective of plant genotype, stomatal density in the two leaf surfaces was two-fold higher in ML than in LL-grown plants and conductance normalized to stomatal density was unaffected by growth irradiance. We conclude that the red light response of stomatal conductance is independent of the concurrent photosynthetic rate of the guard cells or of that of the underlying mesophyll. Furthermore we suggest that the correlation of photosynthetic capacity and stomatal conductance observed under different light environments is caused by signals largely independent of photosynthesis.

INTRODUCTION
Stomata function as hydraulic valves on the surface of aerial parts of plants, with the guard cells that surround each pore rapidly adjusting their turgor to optimise photosynthetic CO 2 uptake and minimize transpirational water loss from leaves.
Stomata respond to a variety of signals, either received from the environment or produced within the plant, which lead to changes in the activities of ion or solute channels regulating guard cell turgor. Stomatal opening is induced by low CO 2 concentrations, high light intensity and high humidity, and closing is promoted by high CO 2 concentrations, darkness, drought and the plant hormone abscisic acid (Outlaw, 2003).
In C 3 species stomatal opening in response to light is thought to be induced by distinct mechanisms depending on the wavelength of incident light. Blue light is perceived directly by phototropins (Kinoshita et al., 2001;Doi et al., 2004) and activates a signalling cascade that results in fast stomatal opening under background red light (Shimazaki et al., 2007). The opening response of stomata to red light requires higher irradiance than blue light and shares characteristics of photosynthesis in its action spectra in the red region (Sharkey and Raschke, 1981). Furthermore the red light response can be abolished by 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) , a photosystem II inhibitor in whole leaf, epidermal strips and guard cell protoplasts (Sharkey and Raschke, 1981;Tominaga et al., 2001;Olsen et al., 2002;Messinger et al., 2006). Using isolated guard cell protoplasts, Tominaga et al. (2001) showed that DCMU inhibited proton pumping in red light suggesting that guard-cell chloroplasts provide ATP required for H+ pumping in the guard cell plasma membrane.
It has also been suggested that the guard cell response to red light is in part an indirect response to red-light-driven intercellular CO 2 uptake in the mesophyll (Roelfsema et al., 2002). For example, Roelfsema et al. (2006) have shown that chloroplast-containing guard cells in albino sections of variegated leaves do not respond to photosynthetically active radiation, but are sensitive to blue light and CO 2 , bringing into question a direct role of guard cell photosynthesis on red-light mediated stomatal opening in intact leaves.
With the exception of the orchid Paphiodedilum, guard cells from all species studied to date contain chloroplasts. Chlorophyll fluorescence measurements (Cardon and Berry, 1992;Goh et al., 1999;Lawson et al., 2002Lawson et al., , 2003 and biochemical and immunolocalization experiments (Ueno, 2001;Zeiger et al., 2002) suggest that guard cell chloroplasts have the capacity for electron transport, Rubisco-mediated CO 2 assimilation and photorespiration. Guard cell photophosphorylation has been postulated as a significant energy source driving stomatal opening in red light (Tominaga et al., 2001). These results have suggested a role of guard cell photosynthesis in the red light response of stomata; however, the mechanism underpinning this link has remained elusive.
Across species and under a variety of growth conditions plants regulate their transpiration and photosynthetic rates in parallel, maintaining a balance between the stomata-mediated supply of CO 2 to the mesophyll chloroplasts and their photosynthetic demand for CO 2 . This results in the conservation of the ratio of intercellular (C i ) to ambient (C a ) CO 2 partial pressures (pCO 2 ) within the leaf (Wong et al., 1979(Wong et al., , 1985Hetherington and Woodward, 2003). This empirical direct correlation between photosynthesis and stomatal conductance was central to initial models of stomatal control of photosynthesis (Farquhar and Wong, 1984;Ball et al., 1987) and has been carried over to more recent models (Jarvis and Davies, 1998;Dewar, 2002;Buckley et al., 2003). However, the underlying regulatory mechanism is still unclear. It has been proposed that guard cells sense the metabolic status of the mesophyll via a diffusible factor that is a product of photosynthetic activity in the mesophyll (Wong et al., 1979;Lee and Bowling, 1992) and that stomatal aperture would be inversely proportional to the pool size of such metabolites (Farquhar and Wong, 1984). Possible metabolites include ATP, NADPH or RuBP, the concentration of which depends strongly on the balance between chloroplast electron transport and the carboxylation reaction catalysed by Rubisco. However, exogenous target gene was effective in guard cells as well as in the mesophyll. However, those plants also maintained wild type values of conductance under ambient CO 2 concentrations and a light source with a mix of red and blue light. (von Caemmerer et al., 2004). Most of the gas exchange measurements made on transgenic plants with impaired photosynthesis have so far been made under white light and this raises the question of whether stomatal conductance would be affected when red light is the only source of illumination.
This report examines the contribution of photosynthetic activity to the stomatal response to red light in intact plants. We contrasted the red light response of stomata of wild type tobacco with that of antisense plants impaired in photosynthetic CO 2 assimilation either by a decrease in chloroplast electron transport rate and ATP synthesis or by a decrease in Rubisco activity and ATP consumption and find that these impairments do not affect stomatal conductance. To further explore the relationship between photosynthesis and stomatal conductance we also examine the stomatal response of wild type and antisense Rubisco plants to growth irradiance.
Remarkably, despite the large difference in photosynthetic rates, the transpiration machinery of wild type and anti-Ssu plants responded in the same manner to the different light growth conditions.

Cytochrome f and Rubisco content in the epidermis of wild type and transgenic tobacco
We used three different phenotypes of tobacco, the wild type and two lines with low photosynthetic CO 2 assimilation rates, generated by antisense technology: anti-b/f plants, which carry an antisense construct directed against the Rieske ironsulphur subunit of the chloroplast cytochrome b 6 f complex (Price et al., 1998), and homozygous anti-Ssu plants with 10-15 % of the wild type content of Rubisco (Ruuska et al., 1998). To minimize developmental effects of the transgenes, our plants were grown under conditions that provided similar rates of growth for the wild type and transgenic genotypes (see Materials and Methods). Figure 1 shows that the known leaf phenotype of the transgenic plants is also expressed in the epidermal tissue. As reported previously (Price et al., 1998) whole leaf cytochrome f content of anti-Ssu plants was also decreased ( Fig. 1). This is consistent with published data (Jiang and Rodermel, 1995).

Photosynthetic rates and stomatal conductances under red measuring light in wild type and anti-b/f plants
We used red light in our gas exchange experiments to induce photosynthesis independently of any stimulation of the blue light response of stomata, which is mediated by phototropins (Shimazaki et al., 2007). Attached leaves were equilibrated to ambient pCO 2 (362 µbar) in the gas exchange chamber in the dark for at least 20 min before red light of 1000 µmol photons m -2 s -1 was turned on. Wild type plants gradually attained an average steady state net rate of CO 2 assimilation of 7.5 µmol photons m -2 s -1 , whereas anti-b/f plants exhibited net rates of CO 2 assimilation ranging from wild type values to only 0.8 % of the wild type rate ( Fig. 2A). There was a direct correlation between the steady state photosynthetic rates under red light and cytochrome b 6 f content in all plants (Fig. 3A). Both the wild type and anti-b/f plants showed variability in the rate of stomatal opening and the maximal stomatal conductance in red light (Fig. 2B). We found that stomatal conductance in both wild type and transgenics did not always reach a complete steady state under red light and therefore reported the maximal conductance. There was little effect of low photosynthetic rate on stomatal parameters, and even the anti-b/f plant with the lowest photosynthetic rate in the range (0.8% of wild type) showed a maximal conductance that was 50% of the average wild type value (Fig 2A and 2B).
Except for plants with wild-type rates of CO 2 assimilation, the anti-b/f plants maintained a ratio of intercellular to ambient CO 2 concentration (C i /C a ) higher than in the wild type (Fig. 2C). There was a marked proportionality of C i /C a and cytochrome b 6 f content, with the anti-b/f plant with the lowest photosynthetic rate showing a C i /C a ratio close to 1 (Fig. 3C). Figure

Photosynthetic rates and stomatal conductances under red measuring light in wild type and anti-Ssu plants and the effect of growth irradiance
To gain further insight on the relationship between stomatal conductance and photosynthesis, we contrasted the red light response of stomata in wild type and anti-SSu plants which contained between 10-15% of wild type Rubisco content. These 288±0.03 mol m -2 s -1 for wild type and anti-Ssu, respectively, and was higher than that of LL-grown plants, which had conductances of 0.16±0.06 and 0.13±0.02 mol m -2 s -1 for wild type and anti-Ssu respectively. However, the differences in stomatal conductance between wild type and anti-Ssu plants at either light intensity were nonsignificant (P=0.05). The low CO 2 assimilation rates and relatively unchanged stomatal conductances in anti-SSu plants resulted in higher C i /C a ratios for the transgenic plants than for wild type (Fig. 5C). The initial transient lowering of C i /C a results from the fact that CO 2 assimilation rate increases more rapidly with irradiance than stomatal conductance.

Effect of growth irradiance on stomatal density and index in wild type and anti-Ssu plants
The drastic increase in stomatal conductance in plants grown at ML compared

Relationship between photosynthetic rate and stomatal conductance in wild type and transgenic tobacco
As it has been shown that there can be a strong correlation between CO 2 assimilation rates and stomatal conductance over a range of growth conditions and leaf ages (Wong et al., 1979), reviewed in (Hetherington and Woodward, 2003)), we were interested to see how our data on transgenic tobacco would fit with the expected linear trend. Figure 7 (filled circles) shows the co-variation of stomatal conductance and CO 2 assimilation rates for young wild type tobacco plants grown in environmental cabinet conditions, at CO 2 concentration of 1000 µmol mol -1 .
Differences in growth light intensity and plant-to-plant variation produced a range of wild type net rates of CO 2 assimilation between approximately 7 and 20 µmol m -2 s -1 .
In these plants stomatal conductance was directly proportional to net CO 2 assimilation rates. Transgenic plants with reduced photosynthetic rates, caused either by decreased Rubisco content (Fig. 7, open circle, square and diamond) or by low cytochrome b 6 f complex (Fig. 7, triangles), maintain stomatal conductances higher than expected from their low CO 2 assimilation rates and thus break the linear relationship of conductance and photosynthetic rate observed for the wild type.

Light response of CO 2 assimilation rate and stomatal conductance
We also examined the fluency response of CO 2 assimilation rate and stomatal conductance to varying intensities of red light in wild type and anti-SSu plants (Fig.   8). Leaves from ML-grown plants were acclimated in the dark for a minimum of 20 minutes before the red light was turned on for 30 min at each irradiance. Stomatal opening continued even after 50 min in the light and we decided to make measurements at defined time interval of 30 min. CO 2 assimilation rate was similar for wild type and anti-SSu plants at low light but saturated for the anti-SSu plants at a low rate around 300 µmol quanta m -2 s -1 whereas it continued to increase for wild type leaves (Fig. 8A). Stomatal conductance on the other hand was similar for wild type and anti-SSu plants (Fig. 8B). The largest increase in conductance occurred in the first step form dark to 50 µmol quanta m -2 s -1 ; however conductance continued to increase up to 1500 µmol quanta m -2 s -1 in both genotypes and the response was distinctly biphasic. The different response of CO 2 assimilation rate and stomatal conductance to irradiance in the anti-SSu plants resulted in greater ratios of C i /C a compared to wild type (Fig. 8C). The humidity of the chamber was not controlled after the initial adjustment and led to a decrease in leaf to air vapour pressure difference which was however similar in wild-type and anti-SSu plants (Fig.8D).

The red light response of stomata in transgenic tobacco with impaired photosynthesis
We have used transgenic tobacco with low capacity for either chloroplast electron transport (anti -b/f plants) or CO 2 fixation capacity (anti-Ssu plants) to probe the contribution of photosynthetic capacity to stomatal opening in red light. In both types of transgenic plants there was an effective decrease in the the amount of the proteins targeted by antisense technology in the guard cells, as shown by immunoblotting (Fig. 1). Thus it is expected that guard cells from these transgenic plants will share at least some of the deficiencies in photosynthetic performance that have been described before for whole leaves. In fact, (von Caemmerer et al., 2004) have observed a strong correlation of photosynthetic performance in guard cells and mesophyll cells of anti-Ssu plants comparing measurements of guard cell chlorophyll fluorescence with that of the underlying mesophyll.
The two transgenic genotypes have contrasting phenotypes. The antisense RNA decrease in Rubisco content has been shown to cause an imbalance between the capacity of the photosynthetic carbon reduction cycle to fix CO 2 and the chloroplast's capacity for electron transport, resulting in an increase in the pool size of RuBP and ATP (Quick et al., 1991a;Hudson et al., 1992), and zeaxanthin (Ruuska et al., 2000a).
Conversely, in the anti-b/f plants, low cytochrome b/f content and hence low electron transport rates cause a decrease in RuBP content, altered redox state (Price et al., 1998;Ruuska et al., 2000a) together with a lowered capacity for zeaxanthin formation via the xanthophyll cycle (Hurry et al., 1996). Despite the differences in photosynthetic properties and rates, we observed no difference in steady state stomatal conductance and stomatal opening in red light between wild type and the transgenics plants. Our results clearly indicate that in intact, attached leaves, the response of stomata to a dark to light transition utilizing red light as irradiance under ambient pCO 2 is independent of the concurrent photosynthetic rate of the guard cells or of the underlying mesophyll .
The opening response of stomata to red light has frequently been linked to photosynthesis because the stomatal response saturates at similar irradiance to photosynthesis and can be abolished by photosystem II inhibitors. Our results confirm that stomatal conductance continues to increase with increasing red irradiance in both wild type and anti-SSu plants (Fig. 8) continued to respond to increasing irradiance in the anti-SSu plants although CO 2 assimilation rate was saturated a low irradiance also suggests that the stomatal response to red light is not linked to the response of CO 2 assimilation rate to red light.
A reduction in the b/f content in our transgenic line leads to a near linear decrease in CO 2 assimilation rate (Fig.3) as has been previously observed (Price et al., 1998). It is thus surprising that this reduction of chloroplast electron transport mediated by the reduction of the cytochrome b/f content does not have a proportional effect on the stomatal red light response and strongly suggests that the red light response of stomata is not quantitatively linked with chloroplast electron transport of guard cells or the mesophyll. All our transgenic plants by necessity have some chloroplast electron transport as they can be grown autotrophically. Thus, we can't exclude the possibility that a complete inhibition of guard cell chloroplast electron transport is required to decrease the extent of the stomatal red light response.
The lack of a stomatal phenotype in the anti b/f plants suggests that routes other than photophosphorylation can provide the energy required for stomatal opening.
The importance of guard cell respiration as an energy source to drive opening has been pointed out (Parvathi and Raghavendra, 1995). Recent experiments on plants with reduced tricarboxylic acid cycle activity but normal chloroplast electron transport rates support the suggestion that mitochondrial function is necessary to maintain optimal stomatal opening and transpiration rates (Nunes-Nesi et al., 2007).

Is there a link between photosynthetic processes and the red light response of stomata?
The fact that the transgenic plants used in this study maintain normal conductances but low photosynthetic rates results in higher than wild type C i values for a given ambient CO 2 partial pressure (Fig. 3). The lack of sensitivity of guard cells to C i has been observed in transgenic plants with low Rubisco or cytochrome b 6 f content before (Quick et al., 1991a;Price et al., 1998;von Caemmerer et al., 2004).
However, because those experiments were performed under white light or a red/blue light source, they did not rule out the possibility of an equal, direct blue light stimulation of opening in wild type and transgenic plants, which could be independent of photosynthesis. To our knowledge, this is the first report to use red light to address this question in intact plants. The nature of the red light response mechanism remains unresolved. The experiments in this paper would argue that the effect of red light absorption and utilisation by photosynthesis on conductance is not as direct as previously thought. Perhaps another, so far unidentified, photoreceptor is involved.
However, the opening response to red light intensity (Fig. 8) clearly shows that no matter how light perception is achieved, stomatal conductance does respond to high light flux levels in a manner not dissimilar to photosynthesis, although there is a distinctly biphasic nature to it with both low and high light response regions. Clearly, more research needs to be done to find alternative mechanisms to explain the red light response of stomata.
Recent mathematical models that attempt to link guard cell photosynthesis with stomatal function hypothesize that the response of stomatal conductance is controlled by the balance between electron transport capacity and Rubisco capacity, and Zeaxanthin and ATP have been proposed as possible metabolic links (Zhu et al., 1998;Buckley et al., 2003). In our transgenic lines the balance between electron transport and Rubisco capacity has been perturbed in opposite directions and our results suggest that the pool size of either metabolite is not the main determinant of stomatal opening under red light. Due to their low electron transport rates relative Rubisco capacity, anti-b/f plants have a substantially decreased zeaxanthin pool (Hurry et al., 1996), however, their maximal stomatal conductance and stomatal opening rate are similar to those of the wild type (Fig. 4). Conversely, anti-Ssu plants have a decreased Rubisco relative to electron transport capacity and increased levels of zeaxanthin (Ruuska et al., 2000b); nevertheless, stomatal conductance is similar to wild type. A logical interpretation of our results is that in general they do not support the hypothesis of a direct link between the stomatal response to CO 2 and the photosynthetic process.
Sucrose, either synthesized inside the guard cell or imported from the apoplast, has been proposed to play a key role as an osmoregulatory solute in stomatal movements Outlaw, 2003). Low leaf sugar content has been reported in anti-Ssu plants (Quick et al., 1991b), even under elevated CO 2 growth conditions (Masle et al., 1993). We don't know whether these transgenic plants can maintain their apoplastic sucrose concentrations despite the lowered bulk leaf concentrations, however the lack of a stomatal phenotype in anti-Ssu plants would suggest that metabolites other than sucrose can act as osmoregulators during stomatal opening.

Correlation between photosynthetic capacity and stomatal conductance
We used growth light intensity as the environmental variable with which to investigate the effect of low photosynthetic rate in the commonly observed comodulation of stomatal conductance and photosynthesis by environmental conditions (Hetherington and Woodward, 2003). Wild type and anti-Ssu plants responded to an increase in growth irradiance from LL to ML by doubling their CO 2 assimilation rates ( Fig. 5A) and their stomatal conductance (Fig. 5B) and density (Fig. 6). For wild type plants this resulted in a strong linear correlation between CO 2 assimilation rates and conductance, as expected (Fig. 7). Although the response to growth irradiance was similar in wild type and anti-Ssu plants, the latter maintained a high stomatal conductance relative to their decreased CO 2 assimilation rates under the two light conditions, showing that stomatal conductance and photosynthetic rate can be uncoupled by genetic manipulation of Rubisco content (Fig. 7). These results are in agreement with previous findings on transgenic plants with impairments in photosynthesis due to antisense decreases in the levels of Rubisco and other PCR cycle enzymes (Hudson et al., 1992;Lauerer et al., 1993;Haake et al., 1998;Muschak et al., 1999). In contrast with these observations in plants with low photosynthetic rates, the correlation between stomatal conductance and photosynthetic rate was apparently maintained in transgenic plants with decreased mitochondrial respiration (Nunes-Nesi et al., 2007).

Developmental implications
The density of stomata on the leaf epidermis is controlled by the environmental conditions prevailing during leaf expansion, and once determined, it remains unchanged for the lifetime of the leaf. Stomatal densities are higher in plants grown in full sun light or at high light intensities than in plants grown in shade (Willmer and Fricker, 1996). The stomatal index of dicot plants has also been shown to increase with light intensity (Schoch et al., 1980). Advances have been made recently in elucidating the genetic pathway controlling stomatal development (Bergmann, 2006). It is thought that mature leaves sense the environment and produce a systemic signal that determines stomatal density in expanding leaves (Coupe et al., 2006). Our results suggest that this systemic developmental signal is not directly linked to photosynthetic capacity: we show that, when anti-Ssu plants are grown under elevated CO 2 , which prevents a limitation of the carbon fixation reactions, the stomatal developmental program is able to sense light intensities during growths and responds by increasing the stomatal density and index in the same manner as the wild type (Fig. 6). It has been shown previously that a decrease in Rubisco content does not interfere with the acclimation of stomatal conductance to environmental conditions such as growth light intensity under ambient pCO 2 (Lauerer et al., 1993), at elevated pCO 2 (Masle et al., 1993;Sicher et al., 1994), and under different nitrogen nutrition regimes (Quick et al., 1992), but differences in stomatal densities under different growth conditions have not been reported before in these plants,. Taken together, these data suggest that the strong photosynthetic impairment of anti-Ssu plants does not significantly affect their ability to acclimate their transpirational machinery to the prevailing growth conditions. Correlation between CO 2 assimilation rate and stomatal conductance is also observed through out the lifespan of a leaf. Jiang and Rodermel (1995) showed that stomatal conductance followed similar developmental changes with leaf age in anti-Ssu and wild type plants despite their different photosynthetic rates. This is an example where the changes in stomatal conductance are not linked to variation in stomatal numbers.

Conclusion
Our study illustrates the power of the transgenic approach in unravelling correlative links to reveal mechanistic connections. The results show that the red light response of stomata may not be linked to photosynthesis and that further work is required to discover the nature of the red light receptor. Furthermore we have shown that the environmentally induced correlation between stomatal conductance and photosynthetic capacity so frequently observed must be caused by signals not directly related to photosynthesis. The results have major implications for our understanding of stomatal function and demonstrate that photosynthetic metabolism can be manipulated with minimal coupling to stomatal function and aperture. This means that if plants can be genetically engineered for improved photosynthesis this should also lead to improved plant water use efficiency.