Opinion: Stomatal responses to light and CO2 depend on the mesophyll

Authors


K. A. Mott. Fax: +1 435 797 1575; e-mail: kmott@biology.usu.edu

ABSTRACT

The mechanisms by which stomata respond to red light and CO2 are unknown, but much of the current literature assumes that these mechanisms reside wholly within the guard cells. However, responses of guard cells in isolated epidermes are typically much smaller than those in leaves, and there are several lines of evidence in the literature suggesting that the mesophyll is necessary for these responses in leaves. This paper advances the opinion that although guard cells may have small direct responses to red light and CO2, most of the stomatal response to these factors in leaves is caused by an unknown signal that originates in the mesophyll.

INTRODUCTION

The effects of light and CO2 on stomata have been studied for decades. Despite this, there is still debate over the mechanisms responsible for these responses. Much of the recent literature tacitly assumes that the mechanisms reside in the guard cells, because there is a considerable body of literature suggesting that guard cells can respond directly to light and CO2 (for reviews, see Vavasseur & Raghavendra 2005; Shimazaki et al. 2007). However, it remains an open question as to whether all of the stomatal response to light and CO2 is because of perception by the guard cells, or whether other mechanisms located in the mesophyll may be operating in addition to those in the guard cells. In this Opinion article, I first briefly review the literature on light and CO2 responses to show that the mechanisms by which stomata respond to these parameters are not fully known. I then examine the differences in the responses to light and CO2 between isolated epidermes and leaves to show that most of the data in the literature suggest that responses to light and CO2 are much smaller in isolated epidermes than in leaves. Finally, I discuss several lines of evidence that support a role for the mesophyll in stomatal responses to light and CO2. Based on these considerations, I suggest that at least part, and possibly most, of the stomatal response to light and CO2 is the result of an unknown signal or signals produced by the mesophyll.

This paper is not intended as an exhaustive review of stomatal responses to light and CO2, and because of the sheer number of manuscripts in this area, it is impossible to cite all relevant work. I have, however, attempted to provide a balanced view of the available literature and cover most of the work that pertains directly to the question at hand.

SEPARATION OF THE LIGHT AND CO2 RESPONSES

Stomata in leaves respond rapidly and reversibly to both light and intercellular CO2 concentration (ci). These responses vary in magnitude among species and even within a species under different growth conditions (e.g. Talbott, Srivastava & Zeiger 1996; Doi, Wada & Shimazaki 2006). In a normally functioning leaf, the responses to light and ci will interact with one another, and unless ci is artificially kept constant or photosynthesis is saturated with light, some of the observed stomatal response to light will inevitably be caused by changes in ci. The degree to which these two responses interact has been, and continues to be, the subject of debate in the literature. Early studies suggested that much or all of the light response was actually a response to CO2 (Heath 1950), but it was established fairly quickly that stomata showed a response to light that was not caused by changes in CO2 concentration (Heath & Russell 1954; Kuiper 1964; Heath 1965; Fischer 1968). Measurements of the relative magnitudes of the CO2 and light responses show that most of the stomatal response to light is a direct response to light and that the contribution of the ci response is small in comparison (Sharkey & Raschke 1981). This conclusion is supported by many studies that show a weak response of stomata to ci at constant light (e.g. Morison & Jarvis 1983; Mott 1988) and by data showing that stomatal conductance (gs) responds strongly to light even when ci is maintained constant by manipulating ambient CO2 concentration (Messinger, Buckley & Mott 2006). The only challenge to this idea in the recent literature is a study in which a single guard cell was illuminated with a beam of red light (Roelfsema et al. 2002). This treatment did not cause hyperpolarization of the guard cell plasma membrane (which is associated with ion transport and stomatal opening, see below), whereas illuminating the guard cell and the surrounding mesophyll with the same intensity of red light did cause hyperpolarization. From these data, the authors concluded that stomatal responses to red light are mostly a response to changes in ci rather than a direct response to light.

THE LIGHT RESPONSE

Stomata show at least two separate light responses – one that shows an action spectrum with a peak near 450 nm, and one that shows an action spectrum that coincides with the absorption spectrum of chlorophyll. These have been called the ‘blue light’ and ‘red light’ responses, respectively (Shimazaki et al. 2007). The blue light response saturates at low light fluxes and is usually studied by adding low fluxes of blue light in the presence of a high-intensity red light (Shimazaki et al. 2007). Thus, although blue light is more efficient per photon than red light in opening stomata, most of the stomatal response to white light is caused by the red light effect (Zeiger & Field 1982).

Several putative photoreceptors for blue light have been identified (Zeiger & Zhu 1998; Kinoshita et al. 2001, 2003), and all of these have been shown to occur in guard cells. Furthermore, blue light has been shown to cause proton extrusion in guard cells in isolated epidermes (Raschke & Humble 1973), and it has been shown to cause swelling (Zeiger & Hepler 1977), proton extrusion (Shimazaki, Iino & Zeiger 1986) and membrane hyperpolarization (Assmann, Simoncini & Schroeder 1985) in isolated guard cell protoplasts. A specific plasma membrane ATPase has been implicated in these effects, and it seems likely that blue light causes proton extrusion by activating a plasma membrane proton pump (Kinoshita & Shimazaki 1999; Kinoshita et al. 2001; Takemiya et al. 2006), which then causes membrane hyperpolarization and K+ influx. These effects are well documented in isolated guard cells and protoplasts, and there is very little contradictory data in the literature. It therefore seems reasonable to conclude that most or all of the blue light response is the result of direct sensing of blue light by the guard cells.

The site of perception and the mode of action for the red light response are much more controversial. Because of the similarities between the action spectrum for the red light response and the absorption spectrum for chlorophyll, it seems likely that chlorophyll is the photoreceptor, and this conclusion is supported by studies showing that the red light response is inhibited when photosynthetic electron transport is blocked by DCMU (Kuiper 1964; Sharkey & Raschke 1981; Tominaga, Kinoshita & Shimazaki 2001; Olsen et al. 2002; Zeiger et al. 2002; Messinger et al. 2006). The controversy arises over how and where the signal is transduced after perception by chlorophyll. Because guard cells of most species contain chloroplasts, the most obvious hypothesis is that red light is perceived by the guard cell chloroplasts. Although there was early disagreement about whether or not guard cell chloroplasts performed the entire set of photosynthetic reactions, the current consensus is that both the light reactions and the dark reactions are present and functional in guard cell chloroplasts. This view is supported by data showing CO2 fixation by guard cells (Shimazaki & Zeiger 1987), and by studies showing fluorescence responses to CO2 that are similar to those seen in the mesophyll (Cardon & Berry 1992; Lawson et al. 2002, 2003).

With the entire suite of photosynthetic reactions occurring in the guard cell chloroplasts, there are many possibilities for a signal based on guard cell photosynthesis. However, as pointed out in several recent reviews (Shimazaki et al. 2007; Lawson 2009), the evidence for a role of guard cell chloroplasts in stomatal movements is contradictory. In contrast to blue light, there are only a few studies that show red light effects in either isolated epidermes or guard cell protoplasts, and many of these show only small responses (Ogawa et al. 1978; Schwartz & Zeiger 1984; Tallman & Zeiger 1988). Although three studies have reported a red light-activated plasma membrane proton pump in guard cells in isolated epidermes (Schwartz & Zeiger 1984; Serrano, Zeiger & Hagiwara 1988; Olsen et al. 2002), other studies have been unable to repeat this result in guard cell protoplasts (Taylor & Assmann 2001) or in intact leaves (Roelfsema et al. 2001). In addition, Zeiger & Hepler (1977) were not able to observe red light-induced swelling of guard cell protoplasts, although they were able to show blue light-induced swelling.

THE CO2 RESPONSE

The mechanism by which CO2 is sensed is also not known. It is generally accepted that the relevant parameter is ci because of experiments using amphistomatous leaves in which ca and ci were manipulated independently of one another (Mott 1988), but sensing of ca or the CO2 concentration in pore have also been suggested (Caemmerer et al. 2004). A response to ci does not distinguish between a sensing mechanism located in the guard cells and one located in the mesophyll, and there are several studies showing that guard cell protoplasts (Gotow, Kondo & Syono 1982) and guard cells in isolated epidermes (Fischer 1968; Young et al. 2006) can respond to CO2. Furthermore, CO2 has been shown to affect plasma membrane anion and K+ channels of guard cells in isolated epidermes (Brearley, Venis & Blatt 1997). It therefore seems probable that at least some of the stomatal response to CO2 is a direct response of the guard cells. This idea is supported by the fact that stomata respond to CO2 in darkness, but the response is qualitatively and quantitatively different from that observed in the light (Messinger et al. 2006). There have been several suggested mechanisms by which guard cells might respond to CO2 (Hedrich et al. 1994; Raschke 2003; Lee et al. 2008), and several genes have been implicated in the CO2 responses (Marten et al. 2008; Vahisalu et al. 2008). There is, however, no accepted mechanism by which guard cells sense CO2 (Vavasseur & Raghavendra 2005), and indeed several authors have suggested that there may be more than one such mechanism (Assmann 1999; Messinger et al. 2006).

THE ROLE OF PHOTOSYNTHESIS

The idea that photosynthesis – either guard cell or mesophyll – is involved in the red light and CO2 responses is attractive for several reasons. Firstly, it could explain the strong correlation between photosynthetic rate and stomatal conductance for plants grown under different light fluxes and different nutrient regimes (Wong, Cowan & Farquhar 1979; Wong et al. 1985a,b,c). Secondly, it could provide a common mechanism for the red light and CO2 responses. It has been noted several times that stomatal conductance is proportional to the amount of excess light energy striking a leaf (Farquhar & Wong 1984; Jarvis & Davies 1998; Buckley, Mott & Farquhar 2003; Messinger et al. 2006). Thus, at a constant value of ci, the amount of excess light energy increases with total light flux, and at a given light flux, the amount of excess light energy increases as ci decreases. If stomatal conductance were somehow regulated in proportion to the amount of excess light energy, it would provide a common mechanism for both light and ci responses. It has been proposed that zeaxanthin could serve as the basis for such a mechanism (Zeiger & Zhu 1998), and there is indirect evidence to support this idea (Tallman et al. 1997; Zhu et al. 1998; Frechilla et al. 1999). It has also been proposed that ATP from photophosphorylation could be the intermediate signal (Farquhar & Wong 1984; Buckley et al. 2003), and there is some evidence in support of this idea as well (Tominaga et al. 2001).

One way to test the involvement of photosynthesis in stomatal response to light or CO2 is to use antisense technology to reduce the activity of various components of the photosynthetic system. These studies have, however, led to conflicting results. Studies using plants with reduced levels of ribulose 1·5-bisphosphate carboxylase/oxygenase (Rubisco) and showing severely reduced rates of photosynthesis have not shown reductions in steady-state stomatal conductance (Quick et al. 1991; Hudson et al. 1992). This result has been confirmed for tobacco plants with antisense reductions in either Rubisco or the cytochrome b6f complex (Hudson et al. 1992; Caemmerer et al. 2004; Baroli et al. 2008). These plants showed similar changes in the target proteins for guard cells and mesophyll cells, which suggests that neither guard cell nor mesophyll photosynthetic rates are important in determining steady-state conductance. However, plants with reduced levels of sedoheptulose-1,7-bisphosphatase (SBPase) activity showed rapid opening in response to red light and higher steady-state conductance values at high light intensities (Lawson et al. 2008). These data were interpreted to mean that stomatal conductance is not controlled by overall photosynthetic rate or capacity, but was instead controlled by photosynthetic electron transport or its end products.

STOMATAL RESPONSES TO LIGHT AND CO2 IN ISOLATED EPIDERMES

Comparisons between isolated epidermes and intact leaves

If the entire sensory and signal transduction systems for light and CO2 are located in the guard cells, then it should be possible to find conditions under which stomatal responses to CO2 and light in isolated epidermes are nearly identical to those in intact leaves. Although many studies report that stomata in isolated epidermes can respond to light and CO2, there are very few that actually compare these responses with those in leaves. In this section, I show that most of these studies find that stomatal responses to light and CO2 are diminished considerably when epidermes are separated from the mesophyll.

One of the first explicit statements of the differences in stomatal behavior between isolated epidermes and leaves was made by Heath (1949). Although that study was conducted before the role of KCl as an osmoticum was known, differences in behaviour between isolated epidermes and leaves have been noted and discussed explicitly several times in the more recent literature (Willmer & Mansfield 1969; Lee & Bowling 1992, 1993, 1995; Roelfsema & Hedrich 2002), and the few studies that directly compare stomatal responses in isolated epidermes and leaves show a wide range of results. On one hand, there are a few studies that show nearly identical responses in isolated epidermes and leaves. For example, Fischer (1968) showed that the response of stomata in isolated epidermes was similar to that in leaf disks when conditions were changed from darkness and ambient CO2 to illuminated and CO2 free air. On the other hand, some studies show essentially no response of stomata to light and CO2 in isolated epidermes (Lee & Bowling 1992; Mott, Sibbernsen & Shope 2008) even though leaves show large responses.

Despite this range of results, most studies show something in between these two extremes, i.e. stomata in isolated epidermes respond to light and CO2, but the responses are smaller than those observed in intact leaves. For example, Hsiao, Allaway & Evans (1973) determined the action spectrum for stomatal opening in isolated epidermes and leaf disks of Vicia faba. They found that these action spectra were similar to one another and closely matched the absorption spectrum of chlorophyll. However, closer examination shows that the stomatal movements in isolated epidermes were much smaller than in leaf disks (Fig. 1), and this trend was also found in their data for stomatal responses to white light in isolated epidermes and leaf disks. Data for the leaf disks were taken at normal CO2 concentrations, while those for isolated epidermes were taken in CO2-free air, so part of the reason that the leaf disks had larger response could easily be because of a change in ci. However, the authors state that when isolated epidermes were measured at normal CO2 concentrations, their response to light was even smaller than in CO2-free air. Similarly, Willmer & Mansfield (1969) studied V. faba and two species of Commelina: sikkimensis and communis. They found ‘. . . major differences in stomatal behaviour between attached and detached epidermis . . .’ with essentially no light effect under any conditions with isolated epidermes of V. faba. Furthermore, in their conclusions, they state, ‘The experimental data presented above indicate that in certain conditions stomata on detached epidermis are able to show responses to environmental factors. However, the effects may differ in magnitude from those found on intact plants’. As a last example, in a recent study on the role of Ca2+ signalling in the CO2 response of guard cells (Young et al. 2006), time courses for stomatal closure in response to a change from 0 to 800 ppm CO2 are shown for both leaves and isolated epidermes. Although stomata responded rapidly (within 20 min) in both leaves and isolated epidermes, the response of aperture was much smaller in isolated epidermes than in leaves (Fig. 2).

Figure 1.

Stomatal aperture as a function of wavelength in isolated epidermes and leaf disks of Vicia faba (redrawn from Hsiao et al. 1973). Isolated epidermes or leaf disks were incubated for 3 h at each wavelength on 2 mm RbCl, 0.1 mm CaCl2, (isolated epidermes) or 0.1 mm CaCl2 (leaf disks). Photosynthetic light flux was 63 µmol photons m–2 s–1. Aperture values for leaf disks were calculated using the relationship between aperture and conductance supplied in fig. 1 of the study. Data are redrawn from figs 4b and 5b.

Figure 2.

Effect of changing CO2 concentration from 0 to 800 ppm in isolated epidermes and leaves (redrawn from Young et al. 2006). For leaves, stomatal conductance was determined at a temperature of 23 °C and a photon flux density of 500 µmol m–2 s–1. Isolated epidermes were incubated for 2–4 h in white light (photosynthetic light flux = 125 µmol photons m–2 s–1) in buffer (10 mm KCl/50 mm CaCl2/10 mm Mes/Tris, pH 6.15). CO2 concentration was controlled by bubbling the buffer solution with gas of known composition.

Studies with only isolated epidermes

Although few studies actually compare responses to light or CO2 in leaves and isolated epidermes, there are many studies that show responses of stomata to light and CO2 in just isolated epidermes, and it is possible to compare these responses to the typical responses of stomata in leaves. To do this, one must first identify ‘typical’ responses of stomata in leaves to light and CO2. Given the range of stomatal responses in leaves that have been reported in literature, this can only be done a general way. Firstly, stomata open as light flux is increased, and close as light flux is decreased. Typically, opening in response to an increase in light flux occurs over an hour or more, while closing in response to a decrease in light flux occurs more rapidly, typically in less than 30 min. Secondly, stomata open as CO2 is lowered, and close as it is raised. These responses are also relatively rapid, and are usually complete within 30–60 min (Meidner & Mansfield 1968).

Because both the CO2 and the light responses are rapidly reversible in leaves, most studies with leaves are performed by exposing the leaf to constant environmental conditions for about an hour, after which aperture or conductance is found to be constant. Conditions (light or CO2) are then changed, and the process is repeated so that a complete light or CO2 response curve can be obtained from a single leaf in a day. In contrast, almost all studies with isolated epidermes are performed by incubating epidermes in a buffered KCl solution at a particular light flux or CO2 concentration for several hours and then measuring apertures (Fischer 1968; Humble & Hsiao 1969; Willmer & Mansfield 1969; Humble & Raschke 1971). Stomata in illuminated epidermes open under these conditions, but epidermes incubated in the dark often open as well, although usually more slowly (Fischer 1968; Humble & Hsiao 1969; Willmer & Mansfield 1969; Lee & Bowling 1992). Also, light and CO2 effects are determined by using a different epidermis for each light or CO2 level; there are apparently no studies in which a single epidermis has been used to produce an entire response curve for light or CO2.

Despite these differences in methodology, there are studies that show responses in isolated epidermes that appear similar to those in leaves. Closer examination, however, reveals that there are usually fundamental differences. For example, Tominaga et al. (2001) showed that stomata in isolated epidermes opened in response to red light (photosynthetic light flux = 1000 µmol photons m–2 s–1) and concluded that guard cell chloroplasts provide ATP for H+ pumping in guard cells. However, stomata in their isolated epidermes did not open in the presence of red light unless fusiccocin was also present. Similarly, Olsen et al. (2002) concluded that red light (photosynthetic flux = 125–150 µmol m–2 s–1) activates ion uptake in guard cells in isolated epidermes, but their data show that stomata did not open in the presence of red light when the CO2 was approximately 400 ppm (assuming a ci/ca ratio of 0.75, this would be equivalent to a ca value of about 600 ppm in an intact illuminated leaf), and at lower CO2 concentrations they opened only very slowly over a period of 4 h. Kuiper (1964) found that the action spectrum for maintaining apertures of already-open stomata in isolated epidermes of Senecio odoris showed distinctive red and blue peaks at very low light intensities (photosynthetic light flux = 11–19 µmol m–2 s–1). He was, however, unable to cause stomata in these isolated epidermes to open in response to light at these intensities. Schwartz & Zeiger (1984) showed a red light stimulation of opening in isolated epidermes of V. faba, but the effect saturated at about 50 µmol photons m–2 s–1, far below the level at which stomata saturate with red light in leaves. Brearley et al. (1997) showed rapid stomatal closure in response to an increase in CO2 using isolated epidermes of V. faba in darkness. They increased CO2 from 0 to either 1000 or 10 000 µL L–1 and found decreases in aperture of approximately 20 and 30%, respectively. Although there are very few measurements of stomatal responses to CO2 in leaves in darkness, those available show much larger changes in aperture or conductance for similar changes in CO2 (Heath & Russell 1954; Messinger et al. 2006).

The data of Travis & Mansfield (1981) are unusual because they show large responses of stomata in isolated epidermes to light and CO2. Their data are from epidermes incubated at constant conditions for several hours, though, and it is not clear how quickly stomata reacted to a change in CO2 or light. Humble & Hsiao (1969) is one of the few studies that shows large, rapid responses of stomata in isolated epidermes to changes in light flux. Experiments were performed with V. faba epidermes in CO2-free air to eliminate responses to light that were caused by changes in CO2 associated with the changes in light flux. They show that apertures increase from about 1 µm to nearly 10 µm within 3 h of illumination. More remarkably, they show that apertures decline to about 3 µm after only 1 h in the dark, then back to 9 or so in the light. They state, however, that ‘with some plant material, stomata opened partially on buffer and did not close readily in the dark. Data obtained with such material were disregarded’.

EVIDENCE FOR A MESOPHYLL ROLE IN THE LIGHT RESPONSE

It is possible that many of the differences in stomatal behaviour between isolated epidermes and leaves discussed earlier are because of damage to the guard cells or epidermal cells during isolation. There are, however, data suggesting that these differences and the difficulty in identifying the mechanisms for stomatal responses red light and CO2 are because the mesophyll is somehow involved in these responses. These data are discussed below.

The simplest possibility for a mesophyll effect is that there is something produced by the mesophyll that renders guard cells responsive to light and CO2. This argument presupposes that guard cells perceive red light and CO2, but are incapable of responding to them without the correct signal from the mesophyll. This idea is supported by several studies that have shown that the magnitude or existence of stomatal responses to CO2 or light in isolated epidermes is dependent on the composition of the external bathing solution (Humble & Hsiao 1969; Willmer & Mansfield 1969; Travis & Mansfield 1979). Nevertheless, there are apparently no studies that have succeeded in finding a solution that supports rapid and reversible responses of stomata to light and CO2 in isolated epidermes.

Alternatively, it is possible that a signal is produced by the mesophyll in response to changes in light and CO2, and that the corresponding changes in stomatal aperture are primarily a response to this signal rather than a direct response to light or CO2. This idea presupposes that guard cells have only a small (or no) direct response to light and CO2, and that most of the response is the result of a signal from the mesophyll. The idea that the mesophyll might be somehow involved in stomatal behaviour has been suggested several times based on the correlation between stomatal conductance and photosynthetic capacity (Wong et al. 1979) when photosynthetic capacity was varied using growth conditions (Wong et al. 1985a), short-term change in light flux (Wong et al. 1985b) or water stress and photoinhibition (Wong et al. 1985c). In addition, there have been a few studies showing that something from the mesophyll is involved in stomatal responses to light and/or CO2. For example, Mouravieff (1956, 1957) noted that stomata in isolated epidermes responded more readily to CO2 and light when they were placed back in close contact with the mesophyll. Subsequently, Lee & Bowling (1992) showed that although stomata in isolated epidermes did not respond to CO2 and light, responses to light and CO2 could be restored by floating isolated epidermes on solutions containing illuminated mesophyll cells or chloroplasts. Responses could also be restored by floating epidermes on the supernatant from the solution containing illuminated mesophyll cells. In a separate study (Lee & Bowling 1993), they showed that this unknown factor from the mesophyll could also stimulate swelling guard cell protoplasts, but only if the mesophyll cells were illuminated with red or white light; blue light had no effect. Conversely, they found a direct effect of white and blue light on the swelling of guard cell protoplasts, but no effect of red light. They concluded that blue light was perceived directly by the guard cells, but that red light was perceived by the mesophyll and its effect transmitted to the guard cells by an unknown factor. Although this factor has never been identified, Lee & Bowling (1992, 1995) proposed the name ‘stomatin’.

Further support for a signal from the mesophyll was provided by a recent study (Mott et al. 2008) showing that stomata in isolated epidermes that were opened on KCl showed essentially no response to light and CO2, but epidermes that were placed back on an exposed mesophyll responded to CO2 and light in a manner that was indistinguishable from an intact leaf. These data show that the difference in stomatal response between isolated epidermes and leaves was not caused by damage to the guard or epidermal cells during isolation. Epidermes from one species could even be placed on the mesophyll of another species and still regain their sensitivity to CO2 and light. That study also showed that stomata in leaves do not open if only the guard cells are illuminated with white light. Only when the nearby mesophyll was illuminated did the illuminated stoma open, and nearby stomata with unilluminated guard cells opened as well (Mott et al. 2008).

There are also several studies that suggest a role for the mesophyll in the red light effect, but in each case the authors interpreted their data to mean that the red light effect was mediated by changes in ci. Roelfsema & Hedrich (2002) used narrow beams of light to illuminate a single guard cell in an intact leaf. When a guard cell was illuminated with blue light in this manner, it resulted in hyperpolarization of the plasma membrane. Red light did not result in hyperpolarization. However, when a larger diameter beam of red light was used such that the guard cell and the surrounding mesophyll were illuminated, hyperpolarization was observed in response to red light. In another study, Nelson & Mayo (1975) showed that guard cells of Paphiopedilum spp. that did not have chloroplasts responded to CO2, and to red and blue light. This result suggests that chloroplasts in the guard cells were not involved in the red light response. This conclusion was supported by a later study showing that guard cells in isolated epidermes of this orchid do not respond to red light, although they do respond to blue light (Zeiger & Assmann 1983). Finally, Roelfsema et al. (2006) studied the variegated leaves of Chlorophytum comosum in which albino areas have guard cells with functional chloroplasts. Stomata in green areas responded to both red and blue light, but stomata in albino areas did not respond to red light, although they did respond to blue light. The authors of each of these studies interpreted their results to mean that the red light response of stomata was mediated via a reduction in ci. However, the CO2 response is not large enough to account for stomatal responses to red light (discussed earlier), and stomata respond to red light even when ci is artificially maintained constant (Messinger et al. 2006). It therefore seems more plausible to explain these results by a signal from the mesophyll (other than ci), which is produced in response to light.

The nature of this putative signal is, as yet, unknown. One possibility is a mesophyll-induced change in the composition of the apoplastic solution in response to light and CO2. Roelfsema & Hedrich (2002) discussed changes in inorganics, organics and hormones that occur in the apoplastic solution following a dark to light transition. They concluded that these changes could alter the response to light, but that they were not large enough or in the correct direction to explain the differences in stomatal behaviour between isolated epidermes and leaves. The presence of an unknown compound in the apoplastic solution remains a possibility, though. Another option for communication between the mesophyll and the guard cells is electrical. This idea is supported by studies that show electrical signals in response to heat that cause rapid decreases in photosynthetic rate and stomatal conductance (Koziolek et al. 2004; Lautner et al. 2005). There are also studies that show electrical currents along the epidermis that are produced only when the epidermis is in contact with the mesophyll (Lee 2006). As yet, though, there are no known effects of electrical signals on guard cells.

CONCLUSION (OPINION)

The mechanisms by which stomata respond to red light and CO2 are unknown, and most literature data indicate that stomatal responses to these two parameters are smaller in isolated epidermes than in leaves. In addition, there are several lines of evidence suggesting that stomatal responses to red light and CO2 in leaves depend on the presence of the mesophyll. Taken together, these observations suggest that guard cells may have small direct responses to red light and CO2, but the larger responses observed in leaves are a result of an unknown signal from the mesophyll.

ACKNOWLEDGMENTS

I thank Erik Sibbernsen and Jesse Spinner for help in preparing this manuscript.

Ancillary