Ethyl Vinyl Ketone Activates K+ Efflux to Regulate Stomatal Closure by MRP4-Dependent eATP Accumulation Working Upstream of H2O2 Burst in Arabidopsis

Plants regulate stomatal mobility to limit water loss and improve pathogen resistance. Ethyl vinyl ketone (evk) is referred to as a reactive electrophilic substance (RES). In this paper, we found that evk can mediate stomatal closure and that evk-induced stomatal closure by increasing guard cell K+ efflux. To investigate the role of eATP, and H2O2 in evk-regulated K+ efflux, we used Arabidopsis wild-type (WT), mutant lines of mrp4, mrp5, dorn1.3 and rbohd/f. Non-invasive micro-test technology (NMT) data showed that evk-induced K+ efflux was diminished in mrp4, rbohd/f, and dorn1.3 mutant, which means eATP and H2O2 work upstream of evk-induced K+ efflux. According to the eATP content assay, evk stimulated eATP production mainly by MRP4. In mrp4 and mrp5 mutant groups and the ABC transporter inhibitor glibenclamide (Gli)-pretreated group, evk-regulated stomatal closure and eATP buildup were diminished, especially in the mrp4 group. According to qRT-PCR and eATP concentration results, evk regulates both relative gene expressions of MRP4/5 and eATP concentration in rbohd/f and WT group. According to the confocal data, evk-induced H2O2 production was lower in mrp4, mrp5 mutants, which implied that eATP works upstream of H2O2. Moreover, NADPH-dependent H2O2 burst is regulated by DORN1. A yeast two-hybrid assay, firefly luciferase complementation imaging assay, bimolecular fluorescence complementation assay, and pulldown assay showed that the interaction between DORN1 and RBOHF can be realized, which means DORN1 may control H2O2 burst by regulating RBOHF through interaction. This study reveals that evk-induced stomatal closure requires MRP4-dependent eATP accumulation and subsequent H2O2 accumulation to regulate K+ efflux.


Introduction
Plant organisms, exposed to unusual and unpredictable environmental conditions, become more vulnerable to infections and pests as a result of non-optimal growing circumstances [1]. Stomata are gas-exchange holes on leaves that affect important functions such as photosynthesis and drought tolerance and are infections' primary entry point [2].
Insect feeding, mechanical damage, and pathogen infection can cause the release of volatile organic compounds. Plant volatiles (VOCs) usually play important roles in regulating the feeding of insects and microorganisms on plants [3]. The esters in VOCs have been discovered to change the opening and closing condition of stomata in plants' resistant response to harmful bacteria. Plants demonstrated stomatal closure after treatment with (Z)-3-hexylpropionate (HP) and (Z)-3-hexylbutyrate (HB) [4].
The volatiles containing α, β-unsaturated carbonyls are collectively referred to as reactive electrophilic substances (RES) [5,6]. (E)-2-hexenal, an electrophilic substance, relative gene expressions and confocal test in mrp4, and mrp5 mutants, and evk-induced H 2 O 2 production was diminished, indicating eATP is upstream of H 2 O 2 in the process of evk-induced stomatal closure in Arabidopsis. The crosstalk between eATP, H 2 O 2 , and the K + channel during stomatal movements was also invested. Stomata maintain moisture balance and protect plants from pathogen infection. Abaxial epidermal peels were peeled off from young fully expanded rosette leaves of 4-week-old Arabidopsis plants and pretreated under light for 2 h in stomatal opening solution, then evk was added to the test solution. The apertures were photographed and measured at 0 min, 30 min, and 60 min. The findings revealed that 8 µM evk caused stomata to close in WT plants (Figure 1a). To see if ABC transporter-dependent eATP generation is involved in evk-induced stomatal closure, we explored the impact of the ABC transporter inhibitor Gli, the eATP receptor DORN1 inhibitor PPADS, and the eATP hydrolytic enzyme Apyrase on evk-induced stomatal closure. Stomatal aperture was 0.21 ± 0.06 after 1 h treatment of evk which was greatly reduced compared to 0 min (0.59 ± 0.05), while in Gli/PPADS/Apyrasetreated group, stomatal aperture was 0.49 ± 0.02, 0.47 ± 0.02, 0.46 ± 0.03, respectively. Individual administration of one of these three reagents had a considerable effect on stomatal closure, and evk-induced stomatal closure was greatly reduced (Figure 1b). According to the findings, which suggested that MRP proteins, eATP, and DORN1 may be involved in the evk-induced stomatal closure signal process. the expression of ATP-related genes in Arabidopsis leaves. Using wild-type Arabidopsis and ABC transporter mutants, we investigated the involvement of eATP in evk-regulated stomatal movements. Using mutant rbohd/f we tested eATP accumulation and MRP4/5 relative gene expressions and confocal test in mrp4, and mrp5 mutants, and evk-induced H2O2 production was diminished, indicating eATP is upstream of H2O2 in the process of evk-induced stomatal closure in Arabidopsis. The crosstalk between eATP, H2O2, and the K + channel during stomatal movements was also invested.

Evk-Induced Stomatal Closure Was Suppressed by Gli, PPADS, and Apyrase in Arabidopsis
Stomata maintain moisture balance and protect plants from pathogen infection. Abaxial epidermal peels were peeled off from young fully expanded rosette leaves of 4-weekold Arabidopsis plants and pretreated under light for 2 h in stomatal opening solution, then evk was added to the test solution. The apertures were photographed and measured at 0 min, 30 min, and 60 min. The findings revealed that 8 μM evk caused stomata to close in WT plants (Figure 1a). To see if ABC transporter-dependent eATP generation is involved in evk-induced stomatal closure, we explored the impact of the ABC transporter inhibitor Gli, the eATP receptor DORN1 inhibitor PPADS, and the eATP hydrolytic enzyme Apyrase on evk-induced stomatal closure. Stomatal aperture was 0.21 ± 0.06 after 1 h treatment of evk which was greatly reduced compared to 0 min (0.59 ± 0.05), while in Gli/PPADS/Apyrase-treated group, stomatal aperture was 0.49 ± 0.02, 0.47 ± 0.02, 0.46 ± 0.03, respectively. Individual administration of one of these three reagents had a considerable effect on stomatal closure, and evk-induced stomatal closure was greatly reduced (Figure 1b). According to the findings, which suggested that MRP proteins, eATP, and DORN1 may be involved in the evk-induced stomatal closure signal process.

eATP Accumulation Was Induced by Evk and Reversed by Gli
We assessed the change in eATP concentration in leaves treated with evk to confirm that eATP transported by ABC transporter is involved in evk-induced stomatal closure. The results revealed that evk considerably increased the eATP level in Arabidopsis seedling leaves with a peak of 523.93 ± 37.13 nM at 6 min. In the inhibitor Gli group, the eATP concentration reached a peak of 97.17 ± 7.62 nM at 6 min. The ABC transporter inhibitor Gli has the potential to reverse the evk-induced eATP increase (Figure 2a). These findings imply that evk may promote eATP and that ABC transporters may play a role in this process.
ABC transporter), PPADS (the inhibiter of eATP receptor DORN1), DPI (the inhibiter of NADPHase), and the Apyrase (eATP degradation) partially impaired the function of evk-induced stomatal closure. The point before evk treatment was treated as 0 min. (b) Histogram of stomatal aperture after 60 min evk treatment. The vertical scale represents the ratio of lateral length and longitudinal length. Data were obtained from at least 60 guard cells from different plants. Bars represent the standard error of the mean.

eATP Accumulation Was Induced by Evk and Reversed by Gli
We assessed the change in eATP concentration in leaves treated with evk to confirm that eATP transported by ABC transporter is involved in evk-induced stomatal closure. The results revealed that evk considerably increased the eATP level in Arabidopsis seedling leaves with a peak of 523.93 ± 37.13 nM at 6 min. In the inhibitor Gli group, the eATP concentration reached a peak of 97.17 ± 7.62 nM at 6 min. The ABC transporter inhibitor Gli has the potential to reverse the evk-induced eATP increase (Figure 2a). These findings imply that evk may promote eATP and that ABC transporters may play a role in this process. There was no significant difference in the size of stomatal openings between the mrp4, mrp5 mutants and the WT Arabidopsis under control conditions. The stomata of WT plants closed after treatment with evk; however, this phenomenon was impaired in mrp4 and mrp5 mutants (Figure 1a,b). After soaking in the evk solution, the eATP concentration of WT keep rising at 1 min, 3 min, and 6 min and reached a peak of 523.93 ± 37.13 nM at 6 min, then slowly declined. When MRP4 is mutated, evk can hardly mediate eATP accumulation with a peak of 138.7 ± 5.4 nM, which is similar to the Gli-pretreated group (Figure 2a,b). When MRP5 is mutated, evk can mediate eATP accumulation with a peak of 334.32 ± 7.97 nM, which is lower than the WT + evk group (Figure 2b). When mutating MRP4, evk can hardly mediate stomatal closure, but in mutant mpr5, stomas can also There was no significant difference in the size of stomatal openings between the mrp4, mrp5 mutants and the WT Arabidopsis under control conditions. The stomata of WT plants closed after treatment with evk; however, this phenomenon was impaired in mrp4 and mrp5 mutants (Figure 1a,b). After soaking in the evk solution, the eATP concentration of WT keep rising at 1 min, 3 min, and 6 min and reached a peak of 523.93 ± 37.13 nM at 6 min, then slowly declined. When MRP4 is mutated, evk can hardly mediate eATP accumulation with a peak of 138.7 ± 5.4 nM, which is similar to the Gli-pretreated group (Figure 2a,b). When MRP5 is mutated, evk can mediate eATP accumulation with a peak of 334.32 ± 7.97 nM, which is lower than the WT + evk group (Figure 2b). When mutating MRP4, evk can hardly mediate stomatal closure, but in mutant mpr5, stomas can also slightly close. These findings revealed that MRP4 and MRP5 are engaged in the evk-induced stomatal closure process, especially MRP4 plays a more important role in evk-induced eATP buildup.

Evk Treatment Enhanced the Expression of ATP-Related Genes in Arabidopsis Leaves
In order to test whether eATP is secreted via ABC transporters in evk-induced eATP accumulation, the effect of evk on the gene expression of MRP4 and MRP5 was examined. Primer sequences are as follows ( Table 1). The qRT-PCR results showed that the gene expression of MRP4 and MRP5 was highly induced. Their expression peak was a 3.80-fold change, and 2.98-fold change, respectively, which occurred 10 min after evk treatment ( Figure 3a). It suggested that evk induced the gene expression of MRP4 and MRP5 in Arabidopsis leaves, thereafter, more ATP was transported outside. Injury, pathogen infection, or osmotic stress can cause cells to release ATP, causing Apyrase1 (APY1) and Apyrase2 (APY2) gene expression to increase [50], showing that their expression is linked to eATP concentration regulation. The concentration of eATP is regulated in part by an enzyme called Apyrase (ecto-nucleoside triphosphate diphosphohydrolase). We confirmed Apyrase expression levels using qRT-PCR, the results showed that: APY1 and APY2 gene expression was significantly upregulated with a peak of a 3.22-fold change, and 2.54-fold change, respectively, after 30 min evk soaking ( Figure 3b). Especially, APY1 may play a major role in the degradation of eATP. DORN1 (DOESN'T RESPOND TO NUCLEOTIDES), extracellular binding domain binds to ATP, reached the highest expression value (3.93-fold change) at 30 min (Figure 3b). Plants respond to eATP with many of the same responses seen in animals, including the production of reactive oxygen species1 (ROS). The results showed that: RBOHF and RBOHD gene expression was upregulated after evk treatment, especially RBOHF (6.80-fold change) ( Figure 3b). The upregulated expression of MRP4/5, APY1/2, RBOHD/F, and DORN1 genes suggested that evk stimulated the participation of eATP and the activation of downstream H 2 O 2 .
TCCAGCTAAGGGTGCC GGTGGGTACTCGGAGA slightly close. These findings revealed that MRP4 and MRP5 are engaged in the evk-induced stomatal closure process, especially MRP4 plays a more important role in evk-induced eATP buildup.

Evk Treatment Enhanced the Expression of ATP-Related Genes in Arabidopsis Leaves
In order to test whether eATP is secreted via ABC transporters in evk-induced eATP accumulation, the effect of evk on the gene expression of MRP4 and MRP5 was examined. Primer sequences are as follows ( Table 1). The qRT-PCR results showed that the gene expression of MRP4 and MRP5 was highly induced. Their expression peak was a 3.80-fold change, and 2.98-fold change, respectively, which occurred 10 min after evk treatment ( Figure 3a). It suggested that evk induced the gene expression of MRP4 and MRP5 in Arabidopsis leaves, thereafter, more ATP was transported outside. Injury, pathogen infection, or osmotic stress can cause cells to release ATP, causing Apyrase1 (APY1) and Apy-rase2 (APY2) gene expression to increase [50], showing that their expression is linked to eATP concentration regulation. The concentration of eATP is regulated in part by an enzyme called Apyrase (ecto-nucleoside triphosphate diphosphohydrolase). We confirmed Apyrase expression levels using qRT-PCR, the results showed that: APY1 and APY2 gene expression was significantly upregulated with a peak of a 3.22-fold change, and 2.54-fold change, respectively, after 30 min evk soaking ( Figure 3b). Especially, APY1 may play a major role in the degradation of eATP. DORN1 (DOESN'T RESPOND TO NUCLEO-TIDES), extracellular binding domain binds to ATP, reached the highest expression value (3.93-fold change) at 30 min ( Figure 3b). Plants respond to eATP with many of the same responses seen in animals, including the production of reactive oxygen species1 (ROS). The results showed that: RBOHF and RBOHD gene expression was upregulated after evk treatment, especially RBOHF (6.80-fold change) ( Figure 3b). The upregulated expression of MRP4/5, APY1/2, RBOHD/F, and DORN1 genes suggested that evk stimulated the participation of eATP and the activation of downstream H2O2.

eATP is Up Stream of H2O2 in the Process of Evk-Induced Stomatal Closure in Arabidopsis
ATP can also cause rapid and dose-dependent increases in H2O2 generation in cultures. H2O2 is also involved in eATP-induced stomatal opening in Arabidopsis [33]. Is there any connection between eATP and H2O2 in Arabidopsis stomatal closure triggered by evk? This question was answered by further data.

eATP is Up Stream of H 2 O 2 in the Process of Evk-Induced Stomatal Closure in Arabidopsis
ATP can also cause rapid and dose-dependent increases in H 2 O 2 generation in cultures. H 2 O 2 is also involved in eATP-induced stomatal opening in Arabidopsis [33]. Is there any connection between eATP and H 2 O 2 in Arabidopsis stomatal closure triggered by evk? This question was answered by further data.

Evk-Induced H 2 O 2 Burst is NADPH-Dependent
H 2 O 2 functions as an important substance that mediates stomatal closure. We tested the variation in H 2 O 2 levels using H 2 DCF-DA, a specialized H 2 O 2 sensor, for detecting H 2 O 2 levels in guard cells by performing confocal laser scanning microscopy from evktreated versus untreated WT plants. H 2 O 2 levels were strongly increased (9.04% above control) in guard cells from evk-treated WT plants compared to the control. In contrast, in evk-treated guard cells pretreated with DPI (diphenyleneiodonium chloride) and in cells of the NADPH oxidase mutant rbohd, rbohf, and rbohd/f, H 2 O 2 enhanced levels (respectively, about 2.39%, 2.46%, 1.69% above control) were significantly lower than WT treated with evk group (Figure 4a). These results indicate that RBOHD/F functions in evk signal transduction. H2O2 functions as an important substance that mediates stomatal closure. We tested the variation in H2O2 levels using H2DCF-DA, a specialized H2O2 sensor, for detecting H2O2 levels in guard cells by performing confocal laser scanning microscopy from evktreated versus untreated WT plants. H2O2 levels were strongly increased (9.04% above control) in guard cells from evk-treated WT plants compared to the control. In contrast, in evk-treated guard cells pretreated with DPI (diphenyleneiodonium chloride) and in cells of the NADPH oxidase mutant rbohd, rbohf, and rbohd/f, H2O2 enhanced levels (respectively, about 2.39%, 2.46%, 1.69% above control) were significantly lower than WT treated with evk group (Figure 4a). These results indicate that RBOHD/F functions in evk signal transduction.

The Effects of Gli, PPADS, and Apyrase on Evk-Induced H2O2 Synthesis in Arabidopsis
The effects of Gli, PPADS, and Apyrase on evk-induced H2O2 synthesis in Arabidopsis leaf guard cells were investigated in terms of understanding the relationship between eATP and H2O2 in the process of evk-induced stomatal closure. When WT guard cells were treated with evk, H2O2 fluorescence rapidly increased (Figure 4a). However, in

In Wild Arabidopsis Guard Cells, Evk Stimulated Outward K + Currents, but This Activation Was Impaired in Mrp4 and Rbohd/f Mutants
To induce stomatal closure, does evk control K + current by influencing the K + channels in guard cells? The K + currents in wild type, mrp4 mutant, and rbohd/f mutant guard cells were measured using the NMT technique. Based on the transient processing of K + ions by evk, the test time was separated into three sections: pre-evk treatment (pre), evk-responsive peak (peak), and post-evk response (post). The results revealed that the amplitude of the basal current did not differ significantly between these lines. After evk treatment, K + flow in WT guard cells peaked at 829.54 ± 59.9 pmol cm -2 s -1 and subsequently gradually decreased, by 3 min, the K + flow had stabilized and had moved into the post-period (Figure 5a). Treatment with evk promoted the outward-rectifying K + currents of the plasma membrane in wild guard cells, but in the guard cells of mrp4 (peaked at 264.59 ± 18.65 pmol cm -2 s -1 ), rbohd/f (peaked at 397.88 ± 35.91 pmol cm -2 s -1 ) and dorn1.3 mutants (peaked at 166.71 ± 23.76 pmol cm -2 s -1 ), K + efflux decreased significantly ( Figure 5a). As shown in Figure 5b, evk caused significant net K + efflux in WT guard cells during the peak response period. Then, the net K + flux recovered to the levels before exposure to evk. In mrp4, rbohd/f, and dorn1.3 guard cells, the evk-induced net K + efflux in the peak response period was significantly lower than that in WT guard cells. The findings suggested that evk regulates stomatal movements by modulating the net content of solute and osmolarity in the cytosol by influencing the activation of K + channels in an eATP-and H 2 O 2 -dependent way.

RBOHF Interacts with DORN1
According to the confocal data, in the mutant rbohf and dorn1.3, the fluorescence of H2O2 increased by 2.45% and 2.27%, respectively, after evk was applied, the H2O2 burst caused by evk is participated by a kinase DORN1 and RBOHF. It is possible to regulate RBOHF through DORN1, so as to regulate the ROS burst. The interaction between the two proteins is illustrated by yeast two-hybrid (Y2H), firefly luciferase complementation imaging assay (LCI), pulldown, and BiFC experiments. In a Y2H experiment, DORN-BK has self-activation, which can be effectively inhibited after inhibition by 12.5 mM 3-AT, and there is interaction between DORN-BK and RBOHF-N-AD (Figure 6a). In vivo experiments showed that DORN-YNE and RBOHF-YCE could interact on the cell membrane of tobacco (Figure 6b). The results in vitro are the same. In the anti-His pulldown line, RBOHF-His was not detected in RBOHF-His with GST group, but in RBOHF-His with DORN-GST group, RBOHF-His was detected by immunoblot analysis using anti-HIS antibodies (Figure 6c). For further confirmation, DORN1-Cluc and RBOHF-Nluc were injected into tobacco leaves at the same time by Agrobacterium. We found that on leaves which were divided into four parts (Cluc/Nluc, DORN1-Cluc/Nluc, Cluc/RBOHF-Nluc, DORN1-Cluc/RBOHF-Nluc), the DORN1-Cluc/RBOHF-Nluc parts had stronger fluorescence than other three control parts (Figure 6d), indicating that the interaction can be realized in vivo and in vitro. It suggests that DORN1 may control the burst of ROS by regulating the activity of RBOHF through interaction.

RBOHF Interacts with DORN1
According to the confocal data, in the mutant rbohf and dorn1.3, the fluorescence of H 2 O 2 increased by 2.45% and 2.27%, respectively, after evk was applied, the H 2 O 2 burst caused by evk is participated by a kinase DORN1 and RBOHF. It is possible to regulate RBOHF through DORN1, so as to regulate the ROS burst. The interaction between the two proteins is illustrated by yeast two-hybrid (Y2H), firefly luciferase complementation imaging assay (LCI), pulldown, and BiFC experiments. In a Y2H experiment, DORN-BK has self-activation, which can be effectively inhibited after inhibition by 12.5 mM 3-AT, and there is interaction between DORN-BK and RBOHF-N-AD (Figure 6a). In vivo experiments showed that DORN-YNE and RBOHF-YCE could interact on the cell membrane of tobacco ( Figure 6b). The results in vitro are the same. In the anti-His pulldown line, RBOHF-His was not detected in RBOHF-His with GST group, but in RBOHF-His with DORN-GST group, RBOHF-His was detected by immunoblot analysis using anti-HIS antibodies (Figure 6c). For further confirmation, DORN1-Cluc and RBOHF-Nluc were injected into tobacco leaves at the same time by Agrobacterium. We found that on leaves which were divided into four parts (Cluc/Nluc, DORN1-Cluc/Nluc, Cluc/RBOHF-Nluc, DORN1-Cluc/RBOHF-Nluc), the DORN1-Cluc/RBOHF-Nluc parts had stronger fluorescence than other three control parts (Figure 6d), indicating that the interaction can be realized in vivo and in vitro. It suggests that DORN1 may control the burst of ROS by regulating the activity of RBOHF through interaction.

Evk Reduced Stomatal Conductance
After fumigation with evk, there was no significant difference in the dry/fresh ratio between evk and the control group, indicating that evk did not cause stress on plant growth (Figure 7a). Compared with the control group, the stomatal conductance of the evk treatment group decreased significantly, and the stomatal conductance of the mutant also decreased, but the degree of decline was less than that of the wild type (Figure 7b). Measurements of stomatal conductance showed similar results to the stomatal aperture experiments (Figure 1a). It indicates that MRP4, RBOHD/F, and DORN1 are involved in the process of evk-mediated stomatal closure.  , BiFC assay, LCI assays, and in vitro pulldown, respectively, show that RBOHF interacts with DORN1. (a) In the Y2H assays, the DORN1 was used as the bait vector. DORN1 has self-activation and can be inhibited by 12.5 mM 3-AT. Yeast clones were grown on SD/-Leu/-Trp and SD/-Ade/-His/-Leu/-Trp and SD/-Ade/-His/-Leu/-Trp with X-α-gal plate. When the self-activation is inhibited, the colony grows well on these plates. (b) In BiFC assay confirms the interactions between RBOHF and DORN1 on the plasma membrane. DAPI (blue) was applied to mark the nucleus. Bar = 25 µm. (c) GST-DORN1 can pull down the RBOHF with His tag, while GST-4T-1 cannot pull down the RBOHF. (d) In the LCI assay, only DORN1-Cluc/RBOHF-Nluc zone has stronger fluorescence than the other three control parts: Cluc/Nluc, DORN1-Cluc/Nluc, Cluc/RBOHF-Nluc. between evk and the control group, indicating that evk did not cause stress on plant growth (Figure 7a). Compared with the control group, the stomatal conductance of the evk treatment group decreased significantly, and the stomatal conductance of the mutant also decreased, but the degree of decline was less than that of the wild type (Figure 7b). Measurements of stomatal conductance showed similar results to the stomatal aperture experiments (Figure 1a). It indicates that MRP4, RBOHD/F, and DORN1 are involved in the process of evk-mediated stomatal closure.

Discussion
α, β-unsaturated carbonyl is an important structure of long-distance signal transduction in plants. The long-distance signal transduction mediated by GLVs may also contain α, β-unsaturated carbonyl [6]. Evk as a plant volatile, as well as RES, would be induced to release under stresses such as Pieris rapae feeding in Arabidopsis, the freeze-thaw damaged soybean leaves [17,18]. Evk can reduce the incidence of flower end rot [19]. However, little is known about whether and how evk can cause stomata closure in Arabidopsis. Evk contains α, β-unsaturated carbonyl structure, has the ability to move vast distances, and controls stomatal closure (Figure 1a).
Firstly, K + efflux will lead to stomatal closure. The study found that evk can mediate K + efflux, thereby mediating stomatal closure. So, whether eATP and H2O2 are involved in the stomatal closure caused by evk? We first found through biological experiments that inhibition of MRP protein to accumulate eATP (Gli treatment), inhibition of eATP receptor DORN1 (PPADS treatment), degradation of eATP (Apyrase treatment), inhibition of NADPHase to produce H2O2 (DPI treatment) could weaken evk-mediated stomatal closure (Figure 1a,b), and evk mediated upregulation of ATP related genes and H2O2 related genes (Figure 3a,b), which indicates that ATP and H2O2 are involved in evk mediated stomatal closure.
Evk can mediate the accumulation of eATP through MRP4 and trigger MRP4 and MRP5 relative gene expression indicating that evk can activate rather than inhibit MRP4

Discussion
α, β-unsaturated carbonyl is an important structure of long-distance signal transduction in plants. The long-distance signal transduction mediated by GLVs may also contain α, β-unsaturated carbonyl [6]. Evk as a plant volatile, as well as RES, would be induced to release under stresses such as Pieris rapae feeding in Arabidopsis, the freeze-thaw damaged soybean leaves [17,18]. Evk can reduce the incidence of flower end rot [19]. However, little is known about whether and how evk can cause stomata closure in Arabidopsis. Evk contains α, β-unsaturated carbonyl structure, has the ability to move vast distances, and controls stomatal closure (Figure 1a).
Firstly, K + efflux will lead to stomatal closure. The study found that evk can mediate K + efflux, thereby mediating stomatal closure. So, whether eATP and H 2 O 2 are involved in the stomatal closure caused by evk? We first found through biological experiments that inhibition of MRP protein to accumulate eATP (Gli treatment), inhibition of eATP receptor DORN1 (PPADS treatment), degradation of eATP (Apyrase treatment), inhibition of NADPHase to produce H 2 O 2 (DPI treatment) could weaken evk-mediated stomatal closure (Figure 1a,b), and evk mediated upregulation of ATP related genes and H 2 O 2 related genes (Figure 3a,b), which indicates that ATP and H 2 O 2 are involved in evk mediated stomatal closure.
Evk can mediate the accumulation of eATP through MRP4 and trigger MRP4 and MRP5 relative gene expression indicating that evk can activate rather than inhibit MRP4 (Figure 3a). Curcumin, which contains two electrophilic compounds α, β-unsaturated carbonyl group, inhibits MRP by curcumin and MRP protein interaction [51]. It has also been confirmed that curcumin significantly inhibits MRP1 and MRP2 mediated transport in the isolated membrane capsule of Sf9 cells expressing MRP1 and MRP2. Curcumin also inhibits MRP1 mediated activity in the intact MRP1 overexpressed Madin Darby canine kidney (mdckii-mrp1) cell monolayer. Trans-2-hexanal, which contains one electrophilic compound α, β-unsaturated carbonyl group did not inhibit MRP1 [51]. Drugs containing this group may not all inhibit or activate MRP proteins. Whether evk really activates MRP4 needs more professional experiments. Secondly, eATP and H 2 O 2 are involved in evk-mediated stomatal closure. What is their upstream and downstream relationship? Through qRT-PCR experiment and confocal experiment, it was found that when RBOHD/F was mutated, it did not affect evk mediated eATP accumulation or MRP protein expression (Figure 2a,b). However, the treatment of Gli, PPADS, and Apyrase prevented the activation of evk on H 2 O 2 synthesis in Arabidopsis guard cells (Figure 4a-d). In summary, eATP is upstream of H 2 O 2 in the process of evkinduced stomatal closure in Arabidopsis. This result is consistent with previous studies [52].
Mutation of DORN1 phosphorylation sites on RBOHD eliminates the ability of ATP to induce stomatal closure [20]. Evk also upregulated DORN1, RBOHF gene (Figure 3b), according to confocal data, mutation of DORN1, RBOHF showed a decrease in H2O2 burst compared to WT treated with evk group (Figure 4a,c), which indicated that DORN1 and RBOHF may be involved in evk-induced H2O2 burst. Moreover, the interaction between these two is illustrated by yeast two-hybrid (Y2H), firefly luciferase complementation imaging assay (LCI), pulldown, and BiFC experiments (Figure 6a-d), indicating they can be realized in vivo and in vitro. The expression of heterologous RBOHD in HEK293T cells shows that direct phosphorylation or Ca 2+ binding to EF chiral domain can synergistically activate RBOHD and produce ROS [34]. In addition to the regulation of Ca 2+ and phosphorylation, more mechanisms have been found, such as 14-3-3, ROP regulation and sulfhydryl nitrite modification [53][54][55][56]. Some studies have shown that RBOHF can be directly regulated by Ca 2+ or by CBL1/9-CIPK26 calcium dependent phosphorylation complex [57]. Both CIPK11 and CIPK26 can phosphorylate RBOHF-N in vitro and activate RBOHF in HEK293T cells [58]. SRC2 can activate the production of ROS mediated by Ca 2+ -dependent RBOHF in Arabidopsis [59]. OST1 is located upstream of ROS and can phosphorylate Ser13 and Ser174 sites of RBOHF [60]. In this study, we demonstrated that DORN1 and RBOHF can be realized in vivo and in vitro. RBOHF can be regulated by DORN1 through interaction.
Finally, are H 2 O 2 and ATP involved in evk mediated K ion efflux? The rapid accumulation and release of K + and of organic and inorganic anions by guard cells control the opening and closing of stomata and thereby gas exchange and transpiration of plants. A recent study identified an insertional T-DNA disruption mutant in the Arabidopsis guard cell K + out channel gene, GORK [41], they provide direct genetic evidence that outward rectifying K + out channels in guard cells contribute to stomatal closing in leaves. We also conducted NMT experiments. Evk promoted the outward-rectifying K + currents of the plasma membrane in wild guard cells, but in the guard cells of mrp4, rbohd/f, and dorn1.3 mutants K + efflux decreased significantly, which implies that evk activation of outward K + channels in an eATP-and H 2 O 2 -dependent way (Figure 5a,b). The findings suggested that evk regulates stomatal movements by modulating the net content of solute and osmolarity in the cytosol by influencing the activation of K + channels in an eATP-and H 2 O 2 -dependent way.
The activity of GORK channels may be modulated by cyclic nucleotides, gammaaminobutyric acid, G-proteins, protein phosphatases, inositol, ATP, reactive oxygen species, and nitric oxide [61][62][63][64]. We guess GORK can be activated by ATP, DORN1 kinase, or H 2 O 2 during evk-mediated K + efflux. More research into how to be activated is recommended. Because GORK is regulated by both ATP and H 2 O 2 , we believe GORK is involved in the evkmediated K + efflux pathway. This hypothesis will require more research to be confirmed.
In terms of experimental methods, three improved experimental methods are needed in future research. Firstly, a previous study demonstrated that DPI at subpicomolar concentrations (10 −13 to 10 −14 M) exhibits specific effects against NOX2 in primary midbrain neuron-glia cultures [65]. However, what is the dosage of DPI in plants? Through consulting the literature, we found that the concentration of NADPHase inhibition in plants is 100 µM [66], 10 µM [67], and 50 µM [68]. After WT was pretreated with DPI at different concentrations, it was treated with 8 µM evk for 1 h to see the stomatal aperture. It was found that the effect of evk on stomatal closure could be inhibited at the DPI concentration of 50 µM. Will the concentration in plants be too high? Will it lead to a toxic effect rather than a real inhibitory effect? This part needs further verification. Secondly, for the determination of ROS, it is necessary to complete the H 2 DCF-DA concentration-specific reaction curve, as well as the dyeing process and the de-esterification process.
Thirdly, determine the effect of evk on plant eATP concentration. Planting plants in 96-well plates will have an impact on plant growth, which is not conducive to plant growth. The nutrients obtained by the roots are very limited. Due to the space problem, only 180 uL of culture medium is added, so it will be subject to nutrient stress, and the overall growth of plants is relatively small. The advantage of this method is that it can detect the release of eATP concentration of seedlings by detecting the eATP in the living state, but its limitations cannot be ignored, and it needs to be improved here. However, the problem of the growth restriction of plants from the narrow space of the 96-well plates cannot be ignored, and I need to propose better research methods.
in future research. Firstly, a previous study demonstrated that DPI at subpicomolar concentrations (10 −13 to 10 −14 M) exhibits specific effects against NOX2 in primary midbrain neuron-glia cultures [65]. However, what is the dosage of DPI in plants? Through consulting the literature, we found that the concentration of NADPHase inhibition in plants is 100 μM [66], 10 μM [67], and 50 μM [68]. After WT was pretreated with DPI at different concentrations, it was treated with 8 μM evk for 1 h to see the stomatal aperture. It was found that the effect of evk on stomatal closure could be inhibited at the DPI concentration of 50 μM. Will the concentration in plants be too high? Will it lead to a toxic effect rather than a real inhibitory effect? This part needs further verification. Secondly, for the determination of ROS, it is necessary to complete the H2DCF-DA concentration-specific reaction curve, as well as the dyeing process and the de-esterification process.
Thirdly, determine the effect of evk on plant eATP concentration. Planting plants in 96-well plates will have an impact on plant growth, which is not conducive to plant growth. The nutrients obtained by the roots are very limited. Due to the space problem, only 180 uL of culture medium is added, so it will be subject to nutrient stress, and the overall growth of plants is relatively small. The advantage of this method is that it can detect the release of eATP concentration of seedlings by detecting the eATP in the living state, but its limitations cannot be ignored, and it needs to be improved here. However, the problem of the growth restriction of plants from the narrow space of the 96-well plates cannot be ignored, and I need to propose better research methods.
Detection of qRT-PCR, stomatal aperture measurements, K + efflux was performed on 18-day-old Arabidopsis WT or mutant plants, which were planted in soil. Detection of eATP content and reactive oxygen species (ROS) in guard cells was analyzed on 12-dayold Arabidopsis WT or mutant plants. Seeds were surface sterilized and sown on Petri plates with 1/2 Murashige and Skoog (MS) solid medium (2.2 g L −1 MS, 10 g L −1 sucrose, 0.5 g L −1 MES, and 1.2% agar, pH 5.8 adjusted using KOH). The concentration of evk used in all experiments was 8 µM.
Nicotiana culture conditions: take the dry stored tobacco seeds into the centrifuge tube, add an appropriate amount of deionized water to clean the surface 3 times, and add 1ml of deionized water to soak in the tube for 24 h to make the seeds absorb water and prepare for germination. Prepare nutritious soil and vermiculite, mix them with water in the proportion of 3:1, and then put them into a plastic flowerpot with a diameter of 7 cm and a height of 8 cm. Absorb the soaked tobacco seeds with a rubber-tipped dropper and directly sow them in the flowerpot. The surface is covered with fresh-keeping film to reduce water evaporation during germination. Put it in the light incubator and cultivate it according to the temperature of 25°C, humidity of 60%, light intensity of 100 µ mol m −2 s −1 , light cycle of 16 h light/8 h darkness. After a week, the seeds grow two leaves, remove the plastic wrap and continue to culture for a week. Select the seedlings with the same growth trend and transfer them to the plastic flowerpot with nutrient soil: vermiculite of 3:1 and continue to culture in the incubator under the same conditions for 3-week-old Nicotiana benthamiana used for firefly luciferase complementation imaging assay.

Measurement of Stomatal Aperture
In order to ensure that before the measurement of stomatal aperture, the stomata of plants were unified and completely open, tear off the lower epidermis and lay it on the bottom of the dish. Fix the lower epidermis of 4-5-week-old leaves blade with adhesive tape on a 35 mm dish incubated in 5 mL stomatal opening solution (30 mM KCl, 1 mM CaCl 2 , 10 mM Tris, pH 5.8) to expose to white light for at least 2 h, room temperature. After this incubation, evk was added to the buffer. The stomatal aperture was measured at 0 min, 60 min, and 90 min. The longitudinal length and lateral length were measured using Image-Pro Plus (Media Cybernetics Co., Silver Spring, MD, USA). At least 60 stomas were measured per epidermal peel. The ratio of transverse length and longitudinal length was calculated and used as an index of stomatal aperture.

Detection of eATP Content in Arabidopsis
All seedings planted in the 96-well plate (one seedling per well) germinate in 3 days and then grow for 7 days. Seedlings that grow for 10 days are actually only 7 days old. The seedlings are in a relatively small state, so the whole plant can be covered with 100uL of test solution. For comparison, seedlings were simply submerged in 100 µL of test solution (0.1 mM KCl, 0.1 mM CaCl 2 , 0.1 mM MgCl 2 , 0.5 mM NaCl, 0.3 mM MES, 0.2 mM Na 2 SO 4 , pH 6.0) at the onset of the treatment. Wild-type seedlings were incubated in 100 µL of test solution, evk solutions, or solutions with Gli, respectively, and the mutant seedings were incubated in the test solution or evk solutions, and samples were immersed in different solutions for 1 min, 3 min, 6 min, 9 min, and 12 min. The final concentration of Gli and evk is 100 µM, 8 µM. For analysis of the eATP concentration, 50 µL of sample was mixed with 50 µL of Luciferase/Luciferin reagent (ENLITEN ® ATP Assay System, Promega, Madison, WI, USA), and the light emission at 420 nm was measured in a GloMaxTM 96 Luminometer, and subsequent luciferin-luciferase treatment and luminescence measurements were performed at the luminometer using a 5 s integration time [21]. The amount of eATP was calculated from a standard curve. Experiments were performed in triplicate.

qRT-PCR
Total RNA was isolated from leaves of Arabidopsis using an RNA extraction kit, and cDNA was produced using a reverse transcriptase kit for quantitative RT-PCR (Takara). ACTIN2 and EF1α were used as references [69][70][71]. RT-PCR was performed in a 7500 fast real-time PCR system (Applied Biosystems, Foster City, California, USA) using a Power SYBR Green PCR Master Mix kit (Applied Biosystems, Foster City, CA, USA), and the 2 -∆∆Ct method was used to calculate relative gene expression levels. The expression levels of several important eATP-related genes were detected, including in Table 1.

H 2 O 2 Detection in Guard Cells
Epidermal peels isolated from 4-5-week-old plants were fixed on the bottom of a Petri dish and then incubated in 30 mM KCl and 10 mM MES (pH 6.5) for 2 h in light (150 mmol m -2 s -1 ). To observe transient evk responses, the leaves of two-month-old seedlings (WT, WT pretreatment with DPI, mrp4, mrp5, dorn1.3, rbohf, rbohd, rbohd/f ) were immerged in 50 µM H 2 DCF-DA for 12 min at 25 • C in the dark. Then, the leaves were steeped in liquid 1/2 MS and treated with 8 µM evk. After 30 min of evk processing, the sample will be placed on the microscope and observed samples under a microscope with excitation at 490 nm and emission at 530 nm. Dye-loaded guard cells were examined using a laser scanning confocal microscope (Leica TCS SP8, Leica Microsystems, Wetzlar, Germany). Each datum was obtained from at least 60 guard cells. The amount of H 2 DCF-DA fluorescence in guard cells was measured with Image-J.

Evk Fumigation
Arabidopsis plants were fumigated in glass bell jars with height of 12.5 cm and diameter of 15 cm with evk (≥97 percent, obtained from Sigma-Aldrich, Shanghai, China, CAS NO: 1629-58-9). Cotton balls with a diameter of 1 cm were soaked in evk dissolved in ethyl alcohol (evk at a final concentration of 8 µM) and hung in the bell jars; control cotton balls were soaked in the same volume of ethanol. The bell jars were immediately sealed with Vaseline petroleum jelly after putting the cotton balls [66].

K + Flux Measurements in Guard Cells
Non-invasive micro-test technology (NMT) was used to detect K + flow in guard cells (BIO-001A, Younger LLC. Amherst, MA, USA). In the test, epidermal peels isolated from 18-day-old plants were fixed on the bottom of a 35 mm dish incubated in the test buffer (0.1 mM KCl, 0.1 mM CaCl 2 , 0.1 mM MgCl 2 , 0.5 mM NaCl, 0.3 mM MES, 0.2 mM Na 2 SO 4 , pH 6.0), for 20 min in light. Silanized glass micropipettes (2-4 m aperture) were filled with electrolyte solution (100 mM KCl) and then front-filled to about 20 µm with a selective liquid ion exchange (LIX) cocktail. For electrode calibrating, use a 0.05 mM, 0.1 mM, or 0.5 mM KCl solution. The glass micropipettes were connected to the NMT system with a silver chloride wire, and electrodes with Nernstian slopes of 58 ± 5 mV log -1 were used.
Then, the K + flux of individual guard cells on the epidermal peels were measured using NMT [72]. The sensor was located above the guard cells (1-2 µm) and was controlled by stepper motors that oscillated along the Z axis above the guard cells at a distance of 10 µm, implying that the sensor oscillated between 1-12 mm above the guard cells. After about 5 min of measurement of the basic ion flux, evk was quickly added to the test solution to a final concentration of 8 µM, data were collected for approximately 2.5 min as the evk response peak (peak) group. Data were then collected from the post-evk response (post) group to indicate the end of the reaction. The final flux values are reported as the mean of eight individual plants per treatment. The K + flux was calculated using Fick's law: J is the flux of K + (pmol cm -2 s -1 ); D is the diffusion coefficient (1.96 × 10 -5 cm 2 s -1 ); ∆C is the difference between the concentrations near and far from the cells; ∆X is 10 µm. Each group contained 3-4 replicates.

Firefly Luciferase Complementation Imaging (LCI) Assay
Coding sequences of DORN1 were cloned into the Cluc plasmid. Coding sequences of RBOHF were cloned into the Nluc plasmid, and the constructed plasmids were transformed into Agrobacterium strain GV3101. A single colony was put into the resistant YEB liquid medium (beef extract 10 g, yeast extract 10 g, NaCl 5 g, pH 7.0) and cultivated overnight at 28 • C, 200 rpm. After 2 min of centrifugation at 12,000 rpm, the medium was discarded. The bacterium was washed five times in tobacco transformation solution (10 mM MES, 10 mM MgCl 2 , pH 5.6), resuspended in tobacco transformation buffer containing 0.1 mM acetosyringone (AS), and infiltrated into the leaves of 3-week-old Nicotiana benthamiana leaves. Fluorescence from luciferase in Nicotiana tabacum leaves infected with Agrobacterium was imaged with a molecular imaging system (LB983, Berthold Technologies, Bad Wildbad, Germany). Each leaf was divided into four quadrants prior to injection with the following combinations: Cluc/Nluc, DORN1-Cluc/Nluc, Cluc/RBOHF-Nluc, and DORN1-Cluc/RBOHF-Nluc. The experiments were repeated at least three times.

In Vitro Pulldown Assay
The genes encoding 344-616aa DORN1 were cloned into pGEX-4T-1 vector via EcoRI/XhoI, and the genes encoding 1-387aa RBOHF were cloned into pET28A vector via EcoRI/XhoI. DORN1-GST and RBOHF-HIS were transformed into E. coli Rosetta (DE3) cells for protein expression. DORN1-GST was used as the bait protein. Protein-His was as the capture protein.
Glutathione beads containing 50 µg DORN1-GST or GST were incubated with 50 µg RBOHF-His in pulldown buffer (1% NP40, 150 mM NaCl, 50 mM Tris-HCl, 1 mM EDTA, pH 7.5, Add protease inhibitor before use) at 4 • C for 2 h, respectively. The protein beads complexes were washed with pulldown buffer, then centrifuged 500× g for 5 min, 5 times; then, add 150 µL elution buffer (250 µL 1 M Tris-HCl, 30 mg Glutathione, pH 8.0). Elute at room temperature for 30 min. Centrifuge 500× g for 5 min, suck the supernatant and add 5× boiling in loading buffer for 5 min and centrifuging at 12,000 rpm for 5 min, at last, the SDS sample buffer was boiled for 10 min. The binding of DORN1-GST with RBOHF-His was detected by immunoblot analysis using anti-GST and anti-His antibodies.

Bimolecular Fluorescence Complementation Assay
DORN1 was cloned into the pSPYNE vector via BamHI/XhoI, RBOHF was cloned into the pSPYCE vector via BamHI/XhoI, and the plasmids were transformed into Agrobacterium strain GV3101. Both constructs (DORN1-YNE and RBOHF-YCE) were transformed to Nicotiana tabacum leaves infected with Agrobacterium. YFP fluorescence was imaged with confocal laser scanning microscopy (Leica TCS SP8, Leica Microsystems, Wetzlar, Germany) at an excitation wavelength of 510 nm and an emission wavelength of 510-530 nm.