Glucose triggers stomatal closure mediated by basal signaling through HXK1 and PYR/RCAR receptors in Arabidopsis

Glucose-triggered stomatal closure is dependent on basal ABA signaling through PYR/RCAR receptors, CDPK6, and glucose signaling mediated by hexokinase1 (HXK1) in Arabidopsis.


Introduction
Stomata are composed of a pair of guard cells in the aerial parts of the plants and play vital roles in controlling both the intake of CO 2 for photosynthesis and transpirational water loss from the plant (Hetherington and Woodward, 2003). The stomatal pore apertures can be modulated by multiple environmental cues such as CO 2 concentration, light intensity, air humidity, drought stress, as well the plant hormone abscisic acid (ABA) (Fan et al., 2004;Acharya and Assmann, 2009). ABA is synthesized from a carotenoid precursor, which is critical in regulating plant development and stress responses (Finkelstein et al., 2002;Nambara and Marion-Poll, 2005;Israelsson et al., 2006;Yoshida et al., 2006). Recent studies have revealed the functional and structural mechanisms underlying ABA perception and its downstream signaling network (Fujii et al., 2009;Ma et al., 2009;Park et al., 2009). The central signaling module of the ABA pathway consists of three major components: the ABA receptors pyrabactin resistance1/PYR1-like/regulatory component of ABA receptor (PYR1/PYL/RCAR), type 2C protein phosphatases (PP2Cs), and subclass2 Snf1-related kinases (SnRK2s) (Ma et al., 2009;Park et al., 2009). In the absence of ABA, PP2Cs inactivate SnRK2s by dephosphorylating a key serine residue in the activation loop and developing physical complexes with SnRK2s, thereby preventing the entry of the substrates (Ma et al., 2009;Park et al., 2009). Binding of ABA to the intracellular PYR/RCAR receptors triggers a conformational change, which permits them to combine with and inactivate PP2Cs. The inactivity of PP2Cs releases its inhibitory effects on SnRK2s, and the activated SnRK2s further phosphorylate target components of their downstream signaling pathways (Ma et al., 2009;Park et al., 2009). It is well known that ABA is a key endogenous factor mediating stomatal closure in response to various environmental stresses (Nambara and Marion-Poll, 2005;Acharya and Assmann, 2009; Bari and Jones, 2009). Under stress, the accumulated endogenous ABA combines with the ABA receptors PYR1/ PYL/RCAR, which induces the interactions with a group of PP2Cs, ABI1 and ABI2, and thus relieves the inhibition of PP2Cs on open stomata 1 (OST1) (Ma et al., 2009;Park et al., 2009). Subsequently, OST1 can directly activate slow type anion channel 1 (SLAC1) responsible for anion efflux via protein phosphorylation, finally inducing stomatal closure (Fujii et al., 2009;Geiger et al., 2009;Park et al., 2009). Additionally, OST1 kinases can phosphorylate the plasma membrane NADPH oxidase, which is the most studied reactive oxygen species (ROS)-producing enzyme in mediating stomatal closure (Sirichandra et al., 2009). ROS production by NADPH oxidase has been proved to be involved in ABA-, methyl jasmonate (MeJA)-, ethylene-, ozone-, darkness-, and Chlorella-induced stomatal closure (Pei et al., 2000;Kwak et al., 2003;Desikan et al., 2004;Suhita et al., 2004;Joo et al., 2005;Bright et al., 2006;Munemasa et al., 2007;Li et al., 2014). In addition to ROS production, nitric oxide (NO) production, cytosolic free calcium concentration ([Ca 2+ ] cyt ), Ca 2+permeable channels, and Ca 2+ -dependent protein kinases (CDPKs) were also involved in the ABA-induced stomatal closure Mori et al., 2006).
Glucose can function like a hormone and has emerged as a key signaling molecule that modulates many vital physiological processes in photosynthetic plants. Previous studies have shown that there are complicated relationships between glucose and ABA signaling pathways (Rolland et al., 2006;Zhang et al., 2008). For example, glucose can control the expressions of genes in ABA biosynthesis and signaling events during seedling development (Rolland et al., 2002(Rolland et al., , 2006. For example, Glucose insensitive 1 (GIN1) encodes cytosolic short-chain dehydrogenase/reductase, which is responsible for converting xanthoxin to ABA-aldehyde in ABA biosynthesis (W.-H. . Phenotype analysis further showed that the gin1-1 mutant was insensitive to glucose and had an increased rate of water loss, leading to symptoms of wilting and withering under low relative humidity and water stress (W.-H. . Additionally, genetic analysis revealed that GIN1 acted downstream of the glucose sensor hexokinases (HXKs) and gin1 was epistatic to HXKs in the glucose signaling pathway (Zhou et al., 1998). HXKs are the enzymes catalyzing the phosphorylation of hexose sugars in the first step of the glycolytic pathway. Previous studies have revealed that Glucose insensitive 2 (GIN2) encodes a HXK1 in the plant glucose signaling network and acts as a glucose sensor to co-ordinate light, nutrient, and hormone signaling networks in regulating plant growth and development in response to environmental changes (Moore et al., 2003). The loss-of-function HXK1 mutant gin2-1 was isolated using a two-step genetic screen in Arabidopsis. The gin2-1 mutant contains a nonsense mutation and has reduced HXK1 transcripts and truncated HXK1 protein accumulation, leading to decreased enzymatic catalytic activity (Moore et al., 2003). The mutant is specifically insensitive to glucose but sensitive to osmotic changes. Notably, compared with the wild type, the gin2-1 mutant had a higher stomatal conductance (g s ) and transpiration rate (E); however, plants overexpressing AtHKT1 in guard cells had a reduced g s and E in Arabidopsis (Kelly et al., 2013). A similar phenomenon was observed in citrus plants (Kelly et al., 2013;Lugassi et al., 2015). Kelly et al. (2013) also found that sucrose triggered guard cell-specific NO production via HXK and ABA in tomato. These results revealed that HXK1 mediated stomatal closure in Arabidopsis, tomato, and citrus plants (Kelly et al., 2013;Lugassi et al., 2015). Notably, previous work has indicated that trehalase, which specifically hydrolyzed trehalose into glucose, up-regulates stomatal closure in Arabidopsis (Van Houtte et al., 2013). Our published data further showed that glucose and mannose could induce stomatal closure mediated by ROS production mainly via NADPH oxidases, Ca 2+ , and the water channel in Vicia faba (Li et al., 2016). A new study has also revealed that G-protein signaling protein is involved in d-glucose-triggered stomatal closure (Hei et al., 2017). However, it remains largely unknown whether GIN1, GIN2, PYR/RCAR, OST1, [Ca 2+ ] cyt , the Ca 2+ channel, CDPK6, nitrate reductase (NR), and SLAC1 are required in glucose-triggered stomatal closure in Arabidopsis.
Unlike V. faba and tomato, working with certain dicotyledons (e.g. Arabidopsis thaliana) enables us to use existing mutants to explore the mechanism underlying glucose-triggered stomatal closure. Due to species specificity, differences in stomatal anatomy and responses exist among Arabidopsis, V. faba, and tomato. Therefore, we further tested whether glucose can trigger stomatal closure in Arabidopsis and, if so, (i) whether glucoseinduced stomatal closure is dependent on ROS and NO production, the [Ca 2+ ] cyt , and the Ca 2+ channel in Arabidopsis; and (ii) whether GIN1, GIN2, PYR/RCAR, OST1, CDPK6, NR, and SLAC1 are involved in glucose-triggered stomatal closure.

Plant material and growth conditions
The A. thaliana seeds were surface sterilized in 70% ethanol for 10 min, and then sown in Petri dishes (1 × 0.15 cm) containing half-strength Murashige and Skoog (MS) solid media with 0.8% (w/v) agar and 1.5% (w/v) sucrose. The seeds were vernalized at 4 °C in the dark for 2 d and transferred into pots (6 cm×8 cm) containing a mixture of growing medium:vermiculite (3:1, v/v) after a 7 d germination. The plotted plants were put in artificial intelligence-controlled chambers with a temperature of 23 °C day/21 °C night, relative humidity of 70%, photosynthetic active radiation (PAR) of 100 μmol m -2 s -1 , and a photoperiod of 12 h light/12 h dark, and were watered daily. Four weeks later, fully expanded leaves in the plants were selected and used for further experiments.

Stomatal bioassay
Stomatal bioassay experiments were performed as described (Li et al., 2014(Li et al., , 2016 with slight modification. Briefly, the epidermis was first peeled off carefully from the abaxial surface of the youngest, fully expanded leaves of 4-week-old plants, and then cut into strips. The strips of epidermis were incubated in opening buffer (10 mM MES, 50 mM KCl, pH 6.15) for 2 h under 22-25 °C and a photon flux density of 100 μmol m -2 s -1 for stomatal opening. Once the stomata were fully open, the epidermal strips were treated with glucose solutions of different concentrations (0, 1, 10, 25, 50, 100, 150, and 200 mM) for another 2 h. In the treatment groups, where inhibitors or scavengers (DPI, SHAM, CAT, EGTA, LaCl 3 , NaN 3 , and tungstate) were needed (Potikha et al., 1999;Pei et al., 2000;Yamasaki and Sakihama, 2000;Zhang et al., 2001;Bright et al., 2006;He et al., 2013), they were added 30 min prior to the glucose treatments. To elucidate the effects of glucose on stomatal movement at the leaf level, the fully expanded leaves of Arabidopsis were kept under light (100 μmol m -2 s -1 ) for 3 h and then treated with sterile water and glucose solution (100 mM) for another 2 h. Then, the epidermal strips were excised and observed immediately by microscopy. Stomata were digitized with a Canon PowerShot G10 camera coupled to a DSZ5000X microscope (UOP, Chongqing, China). The width and length of stomatal pores were assessed using the Image-Pro plus6.0 software (Media Cybernetics, Silver Springs, MD, USA). The stomatal aperture was calculated as pore width/length . The experiments were repeated three times. A total of 50 stomata were selected as samples in each experiment. To obtain the time response course, the stomatal aperture was examined at 30 min intervals. Sixty stomata from three repetive experiments were measured in total.
Leaf stomatal conductance and transpiration rate measurements When the Arabidopsis plants were 4 weeks old, we sprayed 100 mM glucose solution and sterile water with hand sprayers onto the fully expanded leaves which had acclimated to light for 3 h in both the treatment and control groups, making both sides of the leaves uniformly wet. After a 2 h treatment, six fully expanded leaves were randomly selected in separate plants in different treatment groups and the g s and E were measured with a portable photosynthesis system Lincoln,NE,USA). During the experiment, the assimilation chamber was set to an air temperature of 25 °C, 500 ml min −1 air flow, and 380 ppm ambient CO 2 concentration. Data were recorded after 3-4 min, when photosynthesis reached the steady state. When Arabidopsis leaves were small and do not fit on the 2 × 3 cm leaf chamber, we photographed the leaves using a Canon PowerShot G10 camera and calculated the leaf areas with Image-Pro plus6.0 software (Media Cybernetics, Silver Springs, MD, USA). The raw data obtained by the gas exchange system were converted to flux rates per unit of processed leaf area (Vahisalu et al., 2008).

ROS and NO measurements in guard cells
After the treatment in the stomatal bioassay experiments described above, the epidermal strips were loaded with 50 μM H 2 DCF-DA or 10 μM DAF-FM DA in the dark at room temperature. After 30 min, the excess dye was washed three times with opening buffer. Fluorescence photographs of guard cells were taken using a Canon PowerShot G10 camera coupled to a DSZ5000X microscope (UOP, Chongqing, China) and a confocal laser scanning microscope (excitation 490 nm; emission 515 nm) (LSM 710; Zeiss, Jena, Germany). Acquired fluorescence images were analyzed with Image-Pro plus6.0 software (Media Cybernetics). Average fluorescence intensities of treated groups were normalized to the value of the control groups, which was taken as 100% Khokon et al., 2010a). The experiment was repeated three times and the epidermal strips were selected from three individual leaves on separate plants for each replicate in every treatment group.

Statistical analysis
Statistical analyses were performed in SPSS13 [analysis of covariance (ANCOVA) SPSS13, SPSS Inc., Chicago IL, USA]. Values of stomatal aperture, g s , E, ROS, and NO production were compared through the ANOVA procedure individually. Significance among treatments were based on P-values determined by the least significant difference (LSD) test (P<0.05).

Glucose-induced stomatal closure in epidermal strips of Arabidopsis
Our recent studies have revealed that glucose can induce stomatal closure in a dose-and time-dependent manner in V. faba (Li et al., 2016). To determine the effect of glucose on Arabidopsis stomatal movement, the abaxial epidermal peels of wild-type Arabidopsis (Col-0) were treated with mannitol and differing concentrations of glucose solution for 2 h under light. After the treatment with 1, 10, 25, 50, 100, 150, and 200 mM glucose solution and 200 mM mannitol (serving as an osmotic control) for 2 h, stomatal apertures were reduced by 5.1% (P<0.001), 15.6% (P<0.001), 24.1% (P<0.001), 30.5% (P<0.001), 37.1% (P<0.001), 29.3% (P<0.001), 24.9% (P<0.001), and 1.0% (P=0.2818) compared with the control, respectively ( Fig. 1A), showing that glucose caused stomatal closure in a dose-dependent manner. The threshold appeared in 100 mM glucose, while the effects at higher concentrations were less significant (Fig. 1A). There was also no significant change in stomatal aperture in the 200 mM mannitol group (Fig. 1A). In addition, Fig. 1B indicated that 100 mM glucose triggered stomatal closure in a time-dependent manner. The maximum effect appeared 2 h later after treatment, and stomatal apertures decreased by 35.0% (P<0.001) compared with the control treatment.

The effects of CAT, DPI, and SHAM on glucose-induced stomatal closure and ROS production in Arabidopsis
To test whether ROS are involved in glucose-induced stomatal closure in Arabidopsis and which enzyme catalyzes ROS production, we assessed the effects of DPI, SHAM, and CAT on glucose-induced stomatal closure and ROS production. As is shown in Fig. 5, in contrast to the treatment with glucose alone, the glucose-induced stomatal closure was almost completely inhibited by a ROS scavenger, CAT, at 100 U ml −1 (P<0.001). Additionally, it was greatly reversed by an NADPH oxidase inhibitor, DPI, at 20 µM (P<0.001), while it was not suppressed by a peroxidase inhibitor, SHAM, at 2 mM (P=0.828) (Fig. 5). However, applying DPI, SHAM, or CAT alone caused no statistically significant alterations in stomatal aperture (Fig. 5). Furthermore, we tested the glucose-induced ROS production in guard cells of Arabidopsis by loading H 2 DCF-DA. Figure 6 shows that application of 100 mM glucose solution significantly improved ROS production compared with the control (P<0.001). ROS were almost entirely removed by 100 U ml -1 CAT (P<0.001), and significantly abolished by 20 µM DPI (P<0.001), while they were not affected by 2 mM SHAM (P=0.584) (Fig. 6). These results were in accordance with the stomatal response shown in Fig. 5, suggesting that glucose-triggered stomatal closure was mediated by ROS production mostly via DPI-sensitive NADPH oxidase but not SHAM-sensitive peroxidase in Arabidopsis. These phenomena are similar to our previous findings in V. faba that Chlorella-and glucose-elicited stomatal closure relies on ROS production mainly mediated by NADPH oxidase (Li et al., 2014(Li et al., , 2016.

[Ca 2+ ] cyt -and NR-mediated NO production are responsible for glucose-induced stomatal closure in Arabidopsis
We assessed the effects of EGTA (a Ca 2+ chelator) and LaCl 3 (a Ca 2+ channel blocker) on glucose-induced stomatal closure to clarify whether glucose-triggered stomatal closure depends on [Ca 2+ ] cyt and a Ca 2+ channel in Arabidopsis. As is shown in Fig. 7, the glucose-induced stomatal closure was greatly impaired by 2 mM EGTA (P<0.001) and 1 mM LaCl 3 (P<0.001). However, the stomatal aperture showed no statistically significant differences when EGTA and LaCl 3 were

(B) Effects of glucose on stomatal apertures in intact leaves of wild-type Arabidopsis
and ABA signaling mutants. The intact leaves of wild-type Arabidopsis and its mutants kept under light for 3 h were treated with 100 mM glucose and sterile water; after 2 h, the abaxial epidermal strips were peeled off and immediately observed under a microscope. Each bar indicates the means ±SE from three independent experiments (n=150). (C and D) Alterations in leaf g s and E of wild-type Arabidopsis and ABA signaling mutants in response to sterile water and 100 mM glucose. Each bar indicates the means ±SE (n=6). Different letters above the bars denote statistically significant differences among treatments (LSD test, P<0.05). White bar, control; gray bar, glucose.
used alone (Fig. 7). These results reveal that [Ca 2+ ] cyt and the Ca 2+ channel may be involved in glucose-induced stomatal closure in Arabidopsis. The effects of two NR inhibitors, tungstate and NaN 3 , on glucose-induced stomatal closure were examined to determine whether NO is necessary for glucose-triggered stomatal closure in Arabidopsis and what the enzyme source of NO production is. Figure 7 shows that the glucose-triggered stomatal closure was significantly inhibited by 100 µM tungstate (P<0.001) and 2 mM NaN 3 (P<0.001). However, treatment with tungstate or NaN 3 alone caused no statistically significant differences in stomatal aperture (Fig. 7). Furthermore, we monitored the glucose-induced NO production in guard cells of Arabidopsis. It was observed that with 100 mM glucose treatment, NO production was greatly increased compared with the control (P<0.001) (Fig. 8). However, the NO accumulation was largely removed by tungstate (P<0.001) and NaN 3 (P<0.001) (Fig. 8). These results were consistent with the stomatal response noted in Fig. 7. In contrast, the NO production induced by glucose was greatly inhibited in NR-null mutant nia1-1nia2-5 plants (Fig. 8), indicating that glucose-induced stomatal closure was dependent on NO production mainly mediated by NR in Arabidopsis.

Glucose-induced stomatal closure is mediated by functional CDPK6 and NR
The effect of glucose on stomatal aperture of two Arabidopsis mutants, cpk6-1 and nia1-1nia2-5, in the Col-0 background was determined to confirm the influence of CPDK6 and NR on glucose-induced stomatal closure. It was observed that application of 100 mM glucose caused the stomatal apertures to decrease by 48.6% (P<0.001), 15.3% (P<0.001), and 20.9%  (P<0.001) in epidermal peels of Col-0, cpk6-1, and nia1-1nia2-5 plants , respectively, after 2 h (Fig. 9A), in contrast to the control treatment. The stomatal apertures were individually reduced by 40.0% (P<0.001), 16.7% (P<0.001), and 19.8% (P<0.001) when leaves of Col-0, cpk6-1, and nia1-1nia2-5 plants were treated with 100 mM glucose for 2 h (Fig. 9B). Leaf g s and E of Col-0, cpk6-1, and nia1-1nia2-5 plants under steady-state conditions were further measured to assess the contribution of CDPK6 and NR to the regulation of stomatal movement in response to glucose treatments in Arabidopsis. Fig. 7. The effects of EGTA, LaCl 3 , tungstate, and NaN 3 on glucose-triggered stomatal closure. Abaxial epidermal strips of Arabidopsis were pre-incubated in MES-KCl buffer for 2 h under light, and then were treated with 2 mM EGTA, 1 mM LaCl 3 , 100 µM tungstate, and 2 mM NaN 3 for 30 min. They were then floated on 100 mM glucose for 2 h, and the stomatal apertures were assessed after 2 h. Each bar demonstrates the means ±SE of three biological repeats (n=150). Different letters above the bars indicate statistically significant significances among treatments as determined by ANOVA (LSD test, P<0.05). Fig. 8. Glucose induced NO production in wild-type Arabidopsis Col-0 and nia1-1nia2-5 mutants. Abaxial epidermal strips of Arabidopsis were treated with glucose alone, or glucose in the presence of tungstate and NaN 3 for 120 min. They were then loaded with 10 μM DAF-FM DA in the dark for 30 min. After a brief wash with MES-KCl buffer, the strips were observed, and representative pairs of guard cells were photographed using a confocal laser scanning microscopy (A-F). The scale bar in (F) is 20 μm, which applies to all photographs. (G and H) Quantification assay of DAF-FM DA fluorescence intensities of guard cells in images (A-F). The vertical scale represents the percentage of DAF-FM DA fluorescent levels, taking 100% as the value of control treatments. These data represent the mean ±SE of three biological replicates (n=60). Different letters above the bars represent mean values that are statistically significantly different from one another as determined by ANOVA (LSD test, P<0.05). ns, non-significant differences; ***, significant differences at P<0.001.

Glucose-induced stomatal closure in Arabidopsis
Previous studies have shown that ABA can induce stomatal closure and suppress stomatal opening by regulating a series of complex signaling pathways in guard cells (Merlot et al., 2001;Underwood et al., 2007;Acharya and Assmann, 2009). It has been further demonstrated that there are some unexpected overlaps between ABA and glucose signaling pathways (Rolland et al., 2002(Rolland et al., , 2006Zhang et al., 2008). Interestingly, our present work showed that glucose triggered stomatal closure in Arabidopsis (Fig. 1), which agreed with a previous phenomenon found in tomato (Kelly et al., 2013) and our recent findings in V. faba (Li et al., 2016).
Nevertheless, our findings were contrary to the previous theories that glucose or other carbohydrates were ineffective in inhibiting stomatal opening, which was concluded to be the case in Tulipa gesneriana L. and V. faba L. (Dittrich and Mayer, 1978). It is well known that stomatal closure and suppression of stomatal opening in response to external stimuli are two separate processes regulated by different signaling transduction pathways (Allen et al., 1999;Wang et al., 2001;Mishra et al., 2006), which may provide clarification of the inconsistency of the data. Figure 1 also showed dosage and time effects of glucose-induced stomatal closure, and the maximum effect appeared with 100 mM glucose and 2 h treatment. Our published data indicated that glucose induced stomatal closure in a dose-and timedependent manner in V. faba (Li et al., 2016). Compared with the dosage effects of glucose on stomatal aperture between V. faba and Arabidopsis, applications of 25, 50, and 100 mM glucose induced corresponding reductions in stomatal aperture. However, there were significant differences in the reductions of stomatal apertures after treatment with 1, 150, and 200 mM glucose compared with those results from V. faba and Arabidopsis. This may be attributed to different stomatal anatomy and responses between Arabidopsis Fig. 9. Glucose-induced stomatal closure was dependent on functional CDPK6 and NR in Arabidopsis. (A) Glucose-induced stomatal closure in epidermal strips of wild-type Arabidopsis Col-0, cpk6, and nia1nia2 mutants. Abaxial epidermal strips of Arabidopsis were pre-incubated in MES-KCl buffer under light for 2 h, and were immersed in 100 mM glucose solution for another 2 h. The stomatal apertures were then tested. These data indicate the means ±SE from three independent experiments (n=150 per bar). (B) Glucose-triggered stomatal closure in whole leaves of Col-0, cpk6, and nia1nia2 mutants. Intact leaves of wild-type Arabidopsis and its mutants acclimated to light for 3 h were treated with sterile water and 100 mM glucose solution. Two hours later, the abaxial epidermal strips were promptly peeled off and observed under a microscope. Each bar indicates the means ±SE from three independent experiments (n=150). (C and D) Changes in leaf g s and E of wild-type Arabidopsis and its mutants in response to sterile water and 100 mM glucose. These data represent the mean ±SE (n=6 per bar). Different letters above the bars represent significant differences among treatments as determined by ANOVA (LSD test, P<0.05). White bar, control; gray bar, glucose. and V. faba. These results were consistent with some studies showing that some stimuli such as ABA, salicylic acid (SA), hydrogen peroxide (H 2 O 2 ), ethylene, and UV-B can induce stomatal closure in both V. faba and Arabidopsis (Zhang et al., 2001;Bright et al., 2006;Desikan et al., 2006;Mori et al., 2006;He et al., 2011He et al., , 2013. Furthermore, Fig. 1 showed that the effects of glucose on stomatal closure were less significant at higher concentrations, and 200 mM mannitol had no obvious influence on stomatal aperture. This phenomenon implied that glucose-triggered stomatal closure was not due to osmotic stress caused by glucose but rather to signaling events. It was similar to the recent findings that sucrose stimulated stomatal closure mediated by HXK and ABA in an osmotic stress-independent manner in tomato and Arabidopsis (Kelly et al., 2013). To illustrate further the mechanisms underlying glucose-triggered stomatal closure, more experiments will be carried out using pharmacological methods and the stomatal system of Arabidopsis mutants.

Functional GIN1 and GIN2 are involved in glucoseinduced stomatal closure
Recent studies have revealed that HXK1 plays significant roles in mediating stomatal closure in response to sucrose in Arabidopsis, tomato, and citrus plants (Kelly et al., 2013;Lugassi et al., 2015). The gin2-1 mutant is a nonsense mutation which has reduced HXK1 catalytic activity (Moore et al., 2003). The mutant is specific to glucose insensitivity but not to osmotic changes. Phenotype analysis revealed that highglucose repression of cotyledon expansion, chlorophyll accumulation, true-leaf development, and root elongation were impaired in the gin2-1 mutant (Moore et al., 2003). Moreover, previous studies have indicated that the gin2-1 mutant has a higher g s and E; however, AtHXK1-overexpressing plants had reduced g s and E compared with the wild type (Kelly et al., 2013). Our present work showed similar results in that the gin2-1 mutant displayed higher g s and E compared with the wild type (Fig. 2). Additionally, Fig. 2 also shows that the g s and E of the gin2-1 mutant had no obvious changes in response to glucose treatment, and glucose-induced stomatal closure was greatly restrained in epidermal peels and intact leaves of gin2-1. These results indicated that glucose-triggered stomatal closure was dependent on GIN2, as was the case for sucrose (Kelly et al., 2013;Lugassi et al., 2015). In addition, genetic analysis revealed that GIN1 acted downstream of the glucose sensor HXK1 and gin1 was epistatic to HXK1 in the glucose signaling pathway (Zhou et al., 1998). Phenotype analysis indicated that the gin1-1 mutant was insensitive to glucose and had a greater rate of water loss, which induced the symptoms of wilting and withering, especially under low relative humidity and water stress (W.-H. . In our present work, we observed that the gin1-1 mutant had higher stomatal aperture, g s , and E (Fig. 3). This may provide rational explanations for the findings by W.-H. . Glucose-induced reduction in the stomatal aperture, g s , and E was almost completely abolished in the gin1-1 mutant (Fig. 3), suggesting that GIN1 may be required in glucose-induced stomatal closure.
Glucose-induced stomatal closure is modulated by functional PYR/RCAR receptors, ABI1, OST1, and SLAC1 ABA plays vital roles in mediating stomatal closure in response to various environmental stimuli (Ma et al., 2009). The mechanisms underlying ABA perception by the ABA receptors PYL/RCAR and its downstream signaling network have been identified recently (Ma et al., 2009;Park et al., 2009). However, whether PYR/RCAR receptors and downstream signaling components, such as PP2C, OST1, and SLAC1, are required for stomatal closure induced by glucose treatments remains unknown. The present work showed that glucose-triggered stomatal closure was greatly impaired in epidermal peels and intact leaves of pyr1py-l1pyl2pyl4, abi1, ost1, and slac1-4 mutants (Fig. 4A, B). Furthermore, reductions in g s and E caused by glucose were significantly inhibited in the above mutants (Fig. 4C,  D). These results suggest that glucose-mediated stomatal responses may be dependent on ABA signaling components through PYR/RCAR receptors. It is similar to the signaling pathway where CO 2 , O 3 , Ca 2+ , H 2 O 2 , and NO induce stomatal closure (Negi et al., 2008;Vahisalu et al., 2008). The partial inhibition of glucose-triggered stomatal closure may be ascribed to genetic redundancies among PYR/RCAR, PP2C, OST1, and SLAC1 proteins (Szostkiewicz et al., 2010). Furthermore, OST1 is observed to be active in response to some stimuli independent of ABA and PYR/RCAR receptors, which explains the partial stomatal responses to glucose treatments in the ost1 mutant (Xie et al., 2006;Yoshida et al., 2006;Boudsocq et al., 2007). Recently, SLAC1 was demonstrated to be activated by CDPKs in addition to OST1 (Ye et al., 2013), which further explains the partial impairment of stomatal closure induced by glucose in the ost1 mutant. In addition to SLAC1, other anion channels such as the voltage-dependent rapid-type anion channel QUAC1 and the slow-type anion channel have been reported to be involved in stomatal closure (Meyer et al., 2010;Imes et al., 2013). This phenomenon provided an alternative explanation for the partial inhibition of glucose-induced stomatal closure in the slac1-4 mutant.

Glucose-induced stomatal closure is mediated by ROS production in guard cells of Arabidopsis
ROS have been well established as vital second messengers in regulating stomatal closure in response to diverse stimuli (Mustilli et al., 2002;Munemasa et al., 2007). In the present study, we found that optimal concentrations of glucose could noticeably increase the level of ROS in guard cells of Arabidopsis (Fig. 6B), just as ABA and MeJA did (Kwak et al., 2003;Desikan et al., 2004;Suhita et al., 2004). Pharmacological experiments indicated that glucose-induced ROS production was almost totally removed by a membraneimpermeable ROS scavenger, CAT, in Arabidopsis (Fig. 6E). These results imply that glucose-induced ROS production may function outside the plasma membrane of guard cells, due to the high permeability of the plasma membrane to H 2 O 2 (Lee et al., 1999;Zhang et al., 2001;Munemasa et al., 2007). In plants, ROS can be induced via different enzymes in response to various stimuli (Pei et al., 2000;Munemasa et al., 2007). These enzymes include NADPH oxidases, cell walllocalized peroxidases, xanthine oxidases, oxalate oxidases, and amine oxidases (Luis et al., 2002;Vranová et al., 2002;Kwak et al., 2003;Cuevas et al., 2004;Cona et al., 2006). Among these enzymes, NADPH oxidase and peroxidase have been the most studied. For instance, NADPH oxidases have been shown to mediate stomatal closure induced by ABA, MeJA, ozone, darkness, ethylene, allyl isothiocyanate, a low dose of UV-B, flg22, lipopolysaccharide (LPS), elf18, and Chlorella (Kwak et al., 2003;Desikan et al., 2004;Suhita et al., 2004;Joo et al., 2005;Bright et al., 2006;Melotto et al., 2006;Munemasa et al., 2007;Khokon et al., 2011;Sawinski et al., 2013;Li et al., 2014). Peroxidases are proved to modulate stomatal closure triggered by SA, a high dose of UV-B, chitosan, yeast elicitor (YEL), methylglyoxal, and yeast (Mori et al., 2001;Khokon et al., 2010b;He et al., 2011;Hoque et al., 2012). The present work indicated that glucose-induced ROS production was greatly suppressed by an NADPH oxidase inhibitor, DPI, while it was not affected by a peroxidase inhibitor, SHAM (Fig. 6). These results suggest that ROS production induced by glucose is mainly mediated by NADPH oxidases in Arabidopsis. It is similar to the phenomena observed in cultured vascular cells (Inoguchi et al., 2000(Inoguchi et al., , 2003. In addition, our results showed that glucosetriggered stomatal closure was almost completely restored by CAT, strongly inhibited by DPI, and not impaired by SHAM (Fig. 5). It coincided with the effects of CAT, DPI, and SHAM on ROS production stimulated by glucose, which was shown in Fig. 6. These results indicate that glucose-triggered stomatal closure is mainly mediated by ROS production via DPI-sensitive plasma membrane NADPH oxidases but not SHAM-sensitive peroxidases in Arabidopsis. This mechanism is consistent with our recent findings in V. faba (Li et al., 2016). Previous studies have revealed that OST1 kinase can phosphorylate the plasma membrane NADPH oxidase (Sirichandra et al., 2009). As is shown in Fig. 4, reductions in stomatal aperture, g s , and E caused by glucose were restored in the ost1 mutant, implying that OST1 is involved in the glucose-triggered stomatal closure. Based on the previous and present results, we speculate that glucose-induced ROS production and stomatal closure are dependent on OST1activated NADPH oxidase.

Glucose-induced stomatal closure is mediated by [Ca 2+ ] cyt and CDPK6
Previous studies have documented that [Ca 2+ ] cyt functions in ABA and glucose signaling transduction that controls plant growth, development, and stress responses (S.H. . [Ca 2+ ] cyt is a vital secondary messenger in ABAdependent stomatal closure (Mori et al., 2006). In the present work, we found that glucose-induced stomatal closure was significantly repressed by LaCl 3 (a Ca 2+ channel blocker) and EGTA (a Ca 2+ chelator) (Fig. 7). These results reveal that glucose-induced stomatal closure is dependent on [Ca 2+ ] cyt in Arabidopsis, which is similar to our findings in V. faba (Li et al., 2016). This pathway is the same as that by which ABA triggers stomatal closure mediated by [Ca 2+ ] cyt (Srivastava et al., 2009;Ye et al., 2013). CDPKs are confirmed to play vital roles in the Ca 2+ -dependent signaling pathway (Mori et al., 2006;Zhu et al., 2007;Geiger et al., 2009;Brandt et al., 2012). In guard cells, CDPKs mediate stomatal closure mainly via the activation of S-type and Ca 2+ channels, as well as via the inhibition of inwardly rectifying potassium (K in ) channels (Mori et al., 2006;Zou et al., 2010;Munemasa et al., 2011). For instance, CDPK3 and CDPK6 can activate Ca 2+ -permeable channels and S-type channels in response to ABA and Ca 2+ , inducing stomatal closure (Mori et al., 2006). In our present work, we observed that glucose-triggered stomatal closure was impaired in epidermal peels and intact leaves of the cpk6-1 mutant (Fig. 9A, B). In addition, the reductions in g s and E were restored in the cpk6-1 mutant in response to glucose (Fig. 9C, D). These findings imply that CDPK6 is required for glucose-induced stomatal closure, consistent with previous findings that CDPK6 participates in Ca 2+ -, ABA-, MeJA-, and YEL-triggered stomatal closure (Mori et al., 2006;Munemasa et al., 2011;Ye et al., 2013). Nevertheless, the partial impairments of stomatal responses to glucose in the cpk6-1 mutant are likely to be ascribed to genetic redundancies of CDPKs. An alternative explanation is that a Ca 2+ -independent parallel signaling pathway may be involved in glucose-induced stomatal closure.

Glucose-induced stomatal closure is dependent on NO and NR
NO has been recognized as a key secondary messenger in ABA-and microbe-associated molecular pattern (MAMP)mediated stomatal closure and functions downstream of ROS production (Neill et al., 2002;Garcia-Mata and Lamattina, 2007;Khokon et al., 2010b). It has also been reported that sucrose can elicit stomatal closure by stimulating guard cell-located NO production via HXK (Kelly et al., 2013). However, the source of NO in plants and the role of NO in stomatal closure are still controversial (Lozano-Juste and León, 2010;Yu et al., 2012). Notably, our studies showed that glucose induced NO production in guard cells of Arabidopsis (Fig. 8). This result was accompanied by stomatal closure (Fig. 7), as also caused by yeast, LPS, and sucrose (Melotto et al., 2006;Gao et al., 2013;Kelly et al., 2013;Sawinski et al., 2013). In addition, glucose-induced NO production was greatly reduced by two NR inhibitors, tungstate and NaN 3 , followed by reversal of the corresponding stomatal closure (Figs 7, 8). These pharmacological data suggest that glucoseinduced NO production and stomatal closure are mainly mediated by NR. Moreover, glucose-induced NO accumulation and stomatal closure were significantly abolished in NR-null mutant nia1-1nia2-5 plants (Fig. 8). Unlike the wild type, the nia1-1nia2-5 mutant greatly reversed the reductions in stomatal aperture, g s , and E in response to glucose (Fig. 9). Genetic evidence further reveals that NR is the main source of NO production responsible for glucose-triggered stomatal closure. Recently, it has been revealed that ABA-triggered stomatal closure and inhibition of opening are not affected in the NO-deficient mutant nia1 nia2 noa1-2 (Lozano-Juste and León, 2010). As is noted in Fig. 9, the stomatal responses to glucose were partially inhibited in the nia1-1nia2-5 mutant, implying that NR-independent NO or an NO-independent pathway might be involved in glucose-triggered stomatal closure. Whether a nitric oxide synthase (NOS)-like enzyme is another source of NO required for glucose-induced stomatal closure remains an open question and needs to be answered in the future.
In conclusion, glucose could induce stomatal closure, showing dosage and time effects in epidermal strips of Arabidopsis. ROS production via NADPH oxidases, [Ca 2+ ] cyt , a Ca 2+ channel, and NR-mediated NO production were responsible for glucose-induced stomatal closure in Arabidopsis. Glucose-induced stomatal closure was mediated by GIN1, GIN2, PYR/RCAR, PP2C, OST1, CDPK6, NR, and SLAC1 in Arabidopsis. This study may provide new evidence for the involvement of ABA and sugar signaling in glucose-triggered stomatal closure.