H 2 O 2 , Ca 2+ , and K + in subsidiary cells of maize leaves are involved in regulatory signaling of stomatal movement

Background : The stomata of maize ( Zea mays ) contain a pair of guard cells and a pair of subsidiary cells. To determine whether H 2 O 2 , Ca 2+ , and K + in subsidiary cells were involved in stomatal movement, we treated four-week-old maize (Zhengdan 958) leaves with H 2 O 2 , diphenylene iodonium (DPI), CaCl 2 , and LaCl 3 . Changes in content and distribution of H 2 O 2 , Ca 2+ , and K + during stomatal movement were observed. Results : When exogenous H 2 O 2 was applied, Ca 2+ increased and K + decreased in guard cells, while both ions increased in subsidiary cells, leading to stomatal closure. After DPI treatment, Ca 2+ decreased and K + increased in guard cells, but both Ca 2+ and K + decreased in subsidiary cells, resulting in open stomata. Exogenous CaCl 2 increased H 2 O 2 and reduced K + in guard cells, while significantly increasing them in subsidiary cells and causing stomatal closure. After LaCl 3 treatment, decreased and K + increased in guard cells, whereas both H 2 O 2 and K + decreased in subsidiary cells and stomata became open. Conclusions : These results indicated that H 2 O 2 and Ca 2+ were correlated positively with each other and with K + in subsidiary cells during stomatal movement. Both H 2 O 2 and Ca 2+ in subsidiary cells promote an inflow of K + , indirectly regulating stomatal closure.

Background Stomata control the outward transport of water molecules and the gaseous exchange between plants and the external environment during photosynthesis and respiration [1,2]. In the stomata of maize (Zea mays),, a pair of guard cells surrounds a pair of subsidiary cells, which are thought to be donors and receptors of large amounts of water and ions [3] and assist in stomatal movement [4]. When guard cells perceive changes in the environment, they immediately produce signal molecules that regulate stomatal switches and allow the plants to adapt [5]. Besides external factors [6,7,8], these switches are controlled also by intracellular signaling substances and ions [2,9]. Due to the absence of functional plasmodesmata in mature guard cells, the change in ion content between guard cells and subsidiary cells requires ion transport through channels and transporters [10].
Hydrogen peroxide accumulation is involved in the regulation of stomatal movement [11]. Treatment with exogenous H 2 O 2 induced stomatal closure in bean epidermis [12] and in rice plants [13].
Stomatal closure resulted also from the accumulation of H 2 O 2 in subsidiary cells of maize leaves epidermis following a drought period [14]. Inhibition and scavenging of H 2 O 2 have been studied using the NADPH oxidase inhibitor diphenylene iodonium (DPI) [15][16][17] . Calcium ions have also been found to regulate stomatal movement [18], with Ca 2+ content in guard cells increasing following application of exogenous H 2 O 2 , and leading to stomatal closure [19]. Ca 2+ channel inhibitors can change the stomatal closure induced by H 2 O 2 [15]. K + accumulates in guard cells following entry through its channel, resulting in a lower osmotic potential, water absorption, cell expansion, and finally stomatal opening [20]. On the contrary, when K + is transported out of guard cells, the osmotic potential in guard cells increases and stomatal aperture decreases [21]. Subsidiary cells are storage cells for K + and Cl -1 ions [22]. During stomatal opening, depolarization in subsidiary cells depends on inactivation of the outward rectifying potassium channel protein to release K + , whereas depolarization in guard cells depends on the activation of the outward rectifying potassium channel protein to let K + in [23].
Hence, stomatal aperture changes may vary with the changes in H 2 O 2 , K + , and Ca 2+ content in guard and subsidiary cells.
Whereas the interaction between H 2 O 2 , Ca 2+ , and K + in guard cells during stomatal movement is well known, it remains unclear what their mutual relationship and role is in subsidiary cells. Some reports of H 2 O 2 and K + in subsidiary cells [14,24] have prompted the present effort to determine how H 2 O 2 , Ca 2+ , and K + in these cells affected stomatal opening and closing, and how they regulated each other. To this end, this study applied exogenous H 2 O 2 , Ca 2+ , and their inhibitors, and compared their content in subsidiary and guard cells. The results provide a new theoretical basis explaining the mechanism of stomatal movement in grasses.

Effect of exogenous H 2 O 2 ,Ca 2+ , and their inhibitors on stomatal aperture
To verify that H 2 O 2 , K + , and Ca 2+ in subsidiary cells were involved in stomatal movement, the lower epidermis of maize leaves was treated with exogenous H 2 O 2 , Ca 2+ , and their inhibitors prior to scanning electron microscopy observation. Results showed that the treatments promoted or inhibited stomatal opening to varying degrees (Fig. 1 (Fig. 2). DPI and LaCl 3 inhibited stomatal closure by 72% and 26%, respectively, when switching form light to dark (Fig. 2). These results showed that H 2 O 2, Ca 2+ , and their inhibitors could regulate stomatal opening and closing.

Effect of exogenous H 2 O 2, Ca 2+ , and their inhibitors on H 2 O 2 distribution in guard and subsidiary cells during stomatal movement
Under light conditions (i.e., when stomata were open), H 2 O 2 was low and distributed at both ends of guard cells, but it was negligible in subsidiary cells (Fig. 3a). Under dark conditions (i.e., when stomata were closed), H 2 O 2 content augmented in both guard and subsidiary cells, with the latter exhibiting a significantly higher increase (Fig. 3b). A low amount of exogenously added H 2 O 2 under light conditions significantly increased H 2 O 2 in guard and subsidiary cells (Fig. 3c), mimicking the control situation in the dark (Fig. 3b) and promoting stomatal closure (Fig. 1c). DPI was added after 3 h of light and samples were transferred to the dark for 3 h, which significantly reduced H 2 O 2 content in guard and subsidiary cells (Fig. 3d). Guard cells treated with exogenous H 2 O 2 under light conditions exhibited 78.5% higher level of H 2 O 2 than light control (Fig. 4a). In samples exposed to 3 h of light and then treated with DPI for 3 h in the dark, H 2 O 2 amounted to only 18.8% of that in dark control, and 39% of that in guard cells of light control (Fig. 4a). Subsidiary cells treated with exogenous H 2 O 2 under light conditions displayed a 86.7% level of H 2 O 2 in dark control, but only 10.4% of the latter when treated with DPI (Fig. 4b). The difference between treatment and control was significant, indicating that DPI effectively inhibited the production of H 2 O 2 .
H 2 O 2 content in guard cells was more than1.7 times as high following CaCl 2 treatment than in light control (Fig. 3e, Fig. 4a). Seemingly, H 2 O 2 content in subsidiary cells increased with the addition of CaCl 2 (Fig. 4b). This finding indicated a positive correlation between Ca 2+ and H 2 O 2 in both subsidiary and guard cells. To further test this relationship in stomatal cells, LaCl 3 was added for 3 h under dark conditions after 3 h in the light. Results showed a significant reduction in H 2 O 2 content in guard cells (50.9%) and subsidiary cells (72.8%) compared with dark control (Fig. 3f, Fig. 4). This finding confirmed that H 2 O 2 in subsidiary cells correlated positively with changes in Ca 2+ .

Effect of exogenous H 2 O 2 ,Ca 2+ and their inhibitors on K + distribution in guard and subsidiary cells during stomatal movement
The amount of K + in guard cells regulates cell turgor, which is very important for stomatal opening and closing. As a K + reservoir, subsidiary cells play an important role in stomatal movement. Under light conditions, K + could be found mainly in guard cells (Fig. 5a), which absorbed water and expanded to open stomata (Fig. 2). Under dark conditions, K + flowed out from guard cells and into subsidiary cells (Fig. 5b), which caused the former to lose water and stomata to close (Fig. 2). Under light conditions, K + content in subsidiary cells was only 26.8% of that in guard cells, whereas in guard cells it was only 19.5% of that in subsidiary cells under dark conditions (Fig. 6). When the epidermis of maize leaves epidermis was treated with H 2 O 2 under light conditions, K + was distributed mainly in subsidiary cells and less in guard cells (Fig. 5c); with the former accounting for 1.7 times as much K + as the latter. However, upon DPI treatment for 3 h in the dark following 3 h in the light, K + was found mainly in guard cells and less in subsidiary cells (Fig. 5d); with the latter accounting for only 36.2% as much K + as the former (Fig. 6). Even in the presence of light, K + content in guard cells decreased with addition of exogenous H 2 O 2 , while it increased in subsidiary cells, reflecting the situation under dark conditions (Fig. 5b, c). H 2 O 2 inhibition following DPI addition was similar under light settings (Fig. 5a, d). Therefore, H 2 O 2 content correlated negatively with K + in guard cells, but positively in subsidiary cells.
K + content in guard cells treated with CaCl 2 under light conditions was 71% lower compared to light control, whereas in subsidiary cells it increased by 4.3 times over the control (Fig. 5e, Fig. 6). In addition, when samples were treated with LaCl 3 for 3 h in the dark after 3 h in the light, K + was detected mainly in guard cells, and less in subsidiary cells ( Fig. 5f), where it was only half as much as in the former (Fig. 6). These results indicated that Ca 2+ content correlated positively with K + content in subsidiary cells, but negatively in guard cells.

Effect of exogenous H 2 O 2 ,Ca 2+ and their inhibitors on Ca 2+ distribution in guard and subsidiary cells during stomatal movement
Ca 2+ content in guard cells was 52.4% higher in the dark than under light conditions ( Fig. 7a, b, Fig.   8). The change in Ca 2+ content was similar to that observed for H 2 O 2 in subsidiary cells, where it was almost undetectable in the light (Fig. 7a), but increased significantly in the dark (Fig. 7b). When exogenous Ca 2+ was applied under light conditions, Ca 2+ content increased significantly in both cell types (Fig. 7c), becoming about twice as high as in the dark control (Fig. 8). Incubation with LaCl 3 for 3 h in the dark after 3 h in the light resulted in only a small amount of Ca 2+ was being left in guard cells and almost none in subsidiary cells, which was similar to the light control (Fig. 7d). The difference in fluorescence intensity was only 4% following LaCl 3 treatment (Fig. 8), which indicated that exogenous Ca 2+ and LaCl 3 augmented and reduced, respectively, Ca 2+ content in stomata.
H 2 O 2 treatment under light conditions promoted a significant increase in Ca 2+ content in both cell types compared with light controls (Fig. 7e, Fig. 8). The amount of Ca 2+ in guard cells treated with DPI for 3 h in the dark after 3 h in the light was essentially analogous to that under light control conditions ( Fig. 7a, f) with only a 1.9% difference in fluorescence intensity between them (Fig. 8a). In contrast, in subsidiary cells there was no difference at all compared to the light control (Fig. 7a, f, Fig. 8b). These findings indicated that DPI had an inhibitory effect on the increase of Ca 2+ in both guard and subsidiary cells. Taken together with the changes in H 2 O 2 , the results were point to a positive correlation between Ca 2+ and H 2 O 2 .

Discussion
Changes in H 2 O 2 content in guard and subsidiary cells H 2 O 2 has a dual function in plants, acting both as a signaling substance to regulate growth and development when its level is low [25], or as an inhibitor when its concentration is high [26]. Figure 3 shows that external factors could increase H 2 O 2 content in plants [27], whereas DPI could inhibit the production of H 2 O 2 [16,17]. In the latter case, abscisic acid-induced stomatal closure is inhibited [28].
H 2 O 2 content in guard cells increased compared with light control after treatment with H 2 O 2 , and decreased compared with dark control after treatment with DPI (Fig. 4a). In contrast, H 2 O 2 content in subsidiary cells upon H 2 O 2 treatment was higher than in dark control, but only 16.5% of dark control treated with DPI (Fig. 4b). This indicated that exogenous H 2 O 2 or DPI could effectively promote or reduce H 2 O 2 content in stomatal cells [12,17]. At the same time, it also indicated that during stomatal switching, H 2 O 2 in guard and subsidiary cells arose mainly from the action of NADPH oxidase [16,29].
Generally, variations in H 2 O 2 content in guard cells are believed to affect the intracellular concentration of K + and thus regulate stomatal switching [29,30]. Ca 2+ is an important link in the regulation of K + by H 2 O 2 in guard cells [31]. External environmental factors affect H 2 O 2 content in guard cells, which impacts on the activity of intracellular Ca 2+ channels [19] to produce a signal transmission system, and cause inward or outward flow of K + to regulate stomatal movement [28].
According to the present results and previous studies, Ca 2+ seems to promote the production of H 2 O 2 in guard and subsidiary cells. Thus, even in the presence of light, exogenous Ca 2+ treatment augments H 2 O 2 content in guard cells by up to 212.3% of that in the light control, while also increasing significantly that in subsidiary cells (Fig. 3e, Fig. 4). When going form light to dark conditions, the content of H 2 O 2 in guard and subsidiary cells was significantly reduced by LaCl 3 compared with the dark control (Fig. 3f). Importantly, it also indicated that Ca 2+ in guard and subsidiary cells promoted the production of H 2 O 2 , confirming a positive correlation between these two signaling molecules [32].
Changes in K + content in guard and subsidiary cells The flow of K + in guard cells occurs in the opposite direction to that in subsidiary cells [33]. For example, when stomata are closed, K + flows out from guard cells and into subsidiary cells (Fig. 5b).
Ca 2+ and H + control K + channels to regulate stomatal movement [31]. The mutual regulatory relationship between K + and H 2 O 2 was further confirmed via H 2 O 2 and DPI treatments. Specifically, treatment with H 2 O 2 under light conditions and DPI under light-to-dark transition, led to K + becoming concentrated in subsidiary cells and guard cells, respectively (Fig. 5c, d). This indicated that H 2 O 2 activates K + outflow in guard cells [30] while promotes its inflow in subsidiary cells.
When Ca 2+ content in guard cells increases to a micromolar concentration, it significantly reduces the activity of inward K + channels but increases the activity of outward K + channels [31,34]. The calcium-dependent protein kinase GORK is an outward-rectifying K + -channel, playing an important role in stomatal closure [35]. In the presence of light, K + content in guard cells was reduced whereas that in subsidiary cells was increased following treatment of maize leaf epidermis with CaCl 2 (Fig. 5e).

This suggested that Ca 2+ promoted K + release in guard cells and absorption in subsidiary cells. LaCl 3 addition for 3 h in the dark after 3 h in the light resulted in more K + in guard cells and less in
subsidiary cells (Fig. 5f), indicating that Ca 2+ promotes K + influx in subsidiary cells. Increased Ca 2+ concentration in the cytoplasm activates the plasma membrane binding anion channel, resulting in membrane depolarization and inhibition of the inward K + channel [31]. Guard cells plasma membrane depolarized to close stomata, whereas transient hyperpolarization occurs in subsidiary [3]. Hence, the increase in Ca 2+ concentration in subsidiary cells may result in hyperpolarization of the plasma membrane and activation of the introverted K + channel.

Change in Ca 2+ content in guard and subsidiary cells
The activation of hyperpolarization-dependent Ca 2+ channels on the membrane of guard cells under dark conditions promoted an increase in intracellular Ca 2+ [36]. Figure 7a, b reveal that the changes in Ca 2+ content in guard cells are consistent with the above results. However, no one had previously studied changes in Ca 2+ content in subsidiary cells of maize leaf epidermis stomata under dark and light conditions. In this experiment, Ca 2+ content in subsidiary cells also increased significantly with the closure of stomata. It has been reported that during maize stomatal closure, hyperpolarization and cytoplasmic acidification of subsidiary cells differed from those of guard cells or were actually opposite to them [3]. Changes in H 2 O 2 and Ca 2+ content in subsidiary cells were akin to those in guard cells, but plasma membrane polarization and cytoplasmic acidification were the opposite, indicating that the two signaling molecules employed different transmission modes between subsidiary and guard cells.
have an important effect on the activation of Ca 2+ channels in cytoplasmic membrane [19].
After addition of H 2 O 2 and DPI, Ca 2+ content in guard and subsidiary cells increased and decreased in a corresponding manner (Fig. 7e, f). Indeed, patch-clamp experiments had shown that Ca 2+ channels were sensitive to H 2 O 2 , confirming the hypothesis that reactive oxygen species and Ca 2+ acted synergistically [37]. This finding indicates that H 2 O 2 can promote an increase of Ca 2+ in both guard and subsidiary cells.

Conclusions
The present study shows that the interaction between H 2 O 2 , Ca 2+ , and K + in guard cells of maize during stomatal movement is consistent with existing reports. As shown in Figure 9 Cytochemical localization of K + The K + staining methods proposed by Raschke and Fellows [22] and [38] were combined and improved by replacing Na 3 Co(NO 2 ) 6 with NaPbCo(NO 2 ) 6 [39]. The epidermis was incubated in MES-KCl buffer, rinsed with 20 mM CaCl 2 , immediately transferred to 15% NaPbCo(NO 2 ) 6 fixing solution acidified with acetic acid, for incubated 15 min, rinsed three times for 15 min with 50% alcohol, stained with 5% (NH 4 ) 2 S color solution for 1 min, rinsed with distilled water, and finally observed and photographed under a fluorescence microscope. The black-yellow crystals were CoS formed by the action of reaction between K-NaPbCo(NO 2 ) 6 and (NH 4 ) 2 S. The above steps were carried out at 4 °C or on ice with a pre-cooled reagent.

Cytochemical localization of Ca 2+
Cytochemical localization of Ca 2+ was performed as described by Qu et al. [40]. The epidermis was transferred to MES-KCl buffer containing 4 µM Fluo-3 AM (Sigma), incubated in the dark for 2 h at 4°C to prevent esterase hydrolysis in cell walls, incubated in the dark for 2 h at 25 °C to allow the fluorescent dye to fully mark intracellular Ca 2+ following esterase hydrolysis, and washed with MES-KCl buffer. A fluorescence microscope and Image-Pro Plus software were used to observe and analyze the content and distribution of Ca 2+ in guard and subsidiary cells.

Pharmacological treatment of epidermis of maize leaves
For H 2 O 2 and DPI treatment, epidermis exposed to light for 3 h and then treated with 10 μM H 2 O 2 [41] for 3 h under light or exposed to light for 3 h and then treated with 10 μM DPI [42] for 3 h under darkness. For CaCl 2 and LaCl 3 treatment, epidermis exposed to light for 3 h and then treated with 10 mM CaCl 2 [43] for 3 h under light or exposed to light for 3 h and then treated with 5 mM LaCl 3 [44] for 3 h under darkness. In all cases, changes in H 2 O 2 , K + , and Ca 2+ content and distribution in guard and subsidiary cells were observed as described in the previous sections. The transitions from light to dark involved in the above treatment are carried out at room temperature in a black paper sleeve or exposed to fluorescent lamp.

Determination of stomatal aperture in maize leaf epidermis
Stomatal aperture was determined according to Yao et al. [14]. Maize leaves at the six-leaf stage were cut into pieces of 5 mm 2 and rapidly fixed with 4% (v/v) glutaraldehyde solution. The pieces were completely submerged by vacuum and stored at 4 °C overnight. They were then rinsed three times for 10 min with 0.1 M phosphate buffer (pH 7.2), dehydrated in an alcohol gradient, ethanol was replaced by isoamyl acetate, dried at the CO 2 critical point of CO 2, gold-plated, and finally observed and recorded under a scanning electron microscope (Hitachi S-3400N, Hitachi Koki Co., Ltd., Japan).

Availability of data and materials
All data generated or analyzed during this study are included in this manuscript.
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Competing interests
The authors declare that they have no competing interests.  State of the stomatal apparatus from the epidermis of maize leaves following different treatments: a epidermis exposed to light for 6 h (light control); b epidermis exposed to light for 3 h and then darkness for 3 h (dark control); c epidermis exposed to light for 3 h and then treated with H2O2 for 3 h under light; d epidermis exposed to light for 3 h and then treated with DPI for 3 h under darkness; e epidermis exposed to light for 3 h and then treated with CaCl2 for 3 h under light; f epidermis exposed to light for 3 h and then treated with LaCl3 for 3 h under darkness.

Figure 2
Stomatal aperture under different treatments: LL epidermis exposed to light for 6 h (light control); LD epidermis exposed to darkness for 3 h after (dark control); LL+H2O2 epidermis exposed to light for 3 h and then treated with H2O2 for 3 h under light; LD+DPI epidermis exposed to light for 3 h and then treated with DPI for 3 h under darkness; LL+CaCl2 epidermis exposed to light for 3 h and then treated with CaCl2 for 3 h under light; LD+LaCl3 epidermis exposed to light for 3 h and then treated with LaCl3 for 3 h under darkness. Data represent the means of 15 replicates and the bars depict the SD. Distribution of H2O2 in stomata under different treatments: a epidermis exposed to light for 3 h (light control); b epidermis exposed to darkness for 3 h (dark control); c epidermis exposed to light for 3 h and then treated with H2O2 for 3 h under light, d epidermis exposed to light for 3 h and then treated with DPI for 3 h under darkness; e epidermis exposed to light for 3 h and then treated with CaCl2 for 3 h under light; f epidermis exposed to light for 3 h and then treated with LaCl3 for 3 h under darkness.

Figure 4
Fluorescence intensity of H2O2 in guard and subsidiary cells under different treatments: LL epidermis exposed to light for 6 h (light control); LD epidermis exposed to darkness for 3 h after (dark control); LL+H2O2 epidermis exposed to light for 3 h and then treated with H2O2 for 3 h under light; LD+DPI epidermis exposed to light for 3 h and then treated with DPI for 3 h under darkness; LL+CaCl2 epidermis exposed to light for 3 h and then treated with CaCl2 for 3 h under light; LD+LaCl3 epidermis exposed to light for 3 h and then treated with  Distribution of K+ in stomata under different treatments: a epidermis exposed to light for 3 h (light control); b epidermis exposed to darkness for 3 h (dark control); c epidermis exposed to light for 3 h and then treated with H2O2 for 3 h under light conditions, d epidermis exposed to light for 3 h and then treated with DPI for 3 h under darkness conditions; e epidermis exposed to light for 3 h and then treated with CaCl2 for 3 h under light conditions; f epidermis exposed to light for 3 h and then treated with LaCl3 for 3 h under darkness conditions.

Figure 6
Concentration of K+ in guard and subsidiary cells under different treatments: LL epidermis exposed to light for 6 h (light control); LD epidermis exposed to darkness for 3 h after (dark control); LL+H2O2 epidermis exposed to light for 3 h and then treated with H2O2 for 3 h under light; LD+DPI epidermis exposed to light for 3 h and then treated with DPI for 3 h under darkness; LL+CaCl2 epidermis exposed to light for 3 h and then treated with CaCl2 for 3 h under light; LD+LaCl3 epidermis exposed to light for 3 h and then treated with LaCl3 for 3 h under darkness. Data represent the means of 15 replicates and the bars depict the SD.

Figure 7
Distribution of Ca2+ in stomata under different treatments: a epidermis exposed to light for 3 h (light control); b epidermis exposed to darkness for 3 h (dark control); c epidermis exposed to light for 3 h and then treated with CaCl2 for 3 h under light; d epidermis exposed to light for 3 h and then treated with LaCl3 for 3 h under darkness; e epidermis exposed to light for 3 h and then treated with H2O2 for 3 h under light; f epidermis exposed to light for 3 h and then treated with DPI for 3 h under darkness.

Figure 8
Fluorescence intensity of Ca2+ in guard and subsidiary cells under different treatments: LL epidermis exposed to light for 6 h (light control); LD epidermis exposed to darkness for 3 h after (dark control); LL+H2O2 epidermis exposed to light for 3 h and then treated with H2O2 for 3 h under light; LD+DPI epidermis exposed to light for 3 h and then treated with DPI for 3 h under darkness; LL+CaCl2 epidermis exposed to light for 3 h and then treated with CaCl2 for 3 h under light; LD+LaCl3 epidermis exposed to light for 3 h and then treated with