Identification of Genes Preferentially Expressed in Stomatal Guard Cells of Arabidopsis thaliana and Involvement of the Aluminum-Activated Malate Transporter 6 Vacuolar Malate Channel in Stomatal Opening

Stomatal guard cells (GCs) are highly specialized cells that respond to various stimuli, such as blue light (BL) and abscisic acid, for the regulation of stomatal aperture. Many signaling components that are involved in the stomatal movement are preferentially expressed in GCs. In this study, we identified four new such genes in addition to an aluminum-activated malate transporter, ALMT6, and GDSL lipase, Occlusion of Stomatal Pore 1 (OSP1), based on the expression analysis using public resources, reverse transcription PCR, and promoter-driven β-glucuronidase assays. Some null mutants of GC-specific genes evidenced altered stomatal movement. We further investigated the role played by ALMT6, a vacuolar malate channel, in stomatal opening. Epidermal strips from an ALMT6-null mutant exhibited defective stomatal opening induced by BL and fusicoccin, a strong plasma membrane H+-ATPase activator. The deficiency was enhanced when the assay buffer [Cl–] was low, suggesting that malate and/or Cl– facilitate efficient opening. The results indicate that the GC-specific genes are frequently involved in stomatal movement. Further detailed analyses of the hitherto uncharacterized GC-specific genes will provide new insights into stomatal regulation.


INTRODUCTION
Stomata that are formed by pairs of guard cells (GCs) in the shoot epidermis of plants are key regulators of gas exchange, such as CO 2 uptake for photosynthesis and water loss during transpiration (Shimazaki et al., 2007;Munemasa et al., 2015). GCs respond to internal and external signals, such as light, CO 2 , phytohormones, and microbial elicitors, where the stomata remain either open or close (Murata et al., 2015;Inoue and Kinoshita, 2017). Many critical signaling components that are involved in GC signaling are preferentially expressed in GCs, such as Open Stomata 1 (OST1) (Mustilli et al., 2002), slow anion channel-associated 1 (SLAC1) (Negi et al., 2008;Vahisalu et al., 2008), high leaf temperature 1 (HT1) (Hashimoto et al., 2006), and aluminum-activated malate transporter 12 (ALMT12) (Meyer et al., 2010;Sasaki et al., 2010), suggesting that GCspecific genes are important candidates in hunting for new GC signaling components.
Blue light (BL) and red light (RL) are major cues for stomatal opening (Shimazaki et al., 2007;Inoue and Kinoshita, 2017). On BL perception, phototropins undergo autophosphorylation, which triggers signaling by BLUS1, BHP1, type 1 protein phosphatase (PP1), and its regulatory subunit PRSL1, in turn leading to phosphorylation of the penultimate threonine (penThr) residues of the plasma membrane (PM) H + -ATPases, and the subsequent binding of 14-3-3 proteins activates the H + -ATPases. More recently, RL was shown to induce the activation of GC PM H + -ATPases by phosphorylation (Ando and Kinoshita, 2018). PM H + -ATPases are important in terms of stomatal movement; the activation induces PM hyperpolarization, triggering a K + influx through inwardrectifying K + channels (Shimazaki et al., 2007;Inoue and Kinoshita, 2017). Together with the accumulation of K + , the increase of counter anions, such as malate, biosynthesized in GCs and/or apoplast and Cl − from apoplast, and other osmolytes, such as sucrose, lower the water potential in GCs, leading to an inflow of water, the swelling of GCs, and eventually the stomatal opening (Shimazaki et al., 2007;Santelia and Lawson, 2016). Recently, it has been shown that the activation of PM H + -ATPases occurs upstream of starch degradation associated with BL-induced stomatal opening; this, combined with CO 2 fixation in GC chloroplasts, yields the carbon skeletons required for malate synthesis (Horrer et al., 2016). Vacuoles accumulate most of the ions and water that control stomatal movement (Barbier-Brygoo et al., 2011). The electrophysiological experiments revealed that the ALMT6 and ALMT9 vacuole channels facilitated malate and Cl − import (Meyer et al., 2011); ALMT9 played a critical role in the light-induced stomatal opening .
To identify new signaling components involved in the lightinduced stomatal opening, we reasoned that GC preferentially expressed genes are good candidates and identified four new such genes by the analyses of public resources, reverse transcription PCR (RT-PCR), and promoter GUS assay in Arabidopsis thaliana. Functional analysis revealed that some of the GC-specific genes in addition to ALMT6 are critical in the light-induced stomatal opening.

Reverse Transcription PCR
RNAs from gl1 GCPs, MCPs, rosette leaves, roots, petioles, stems, flowers, and etiolated seedlings were extracted using the RNeasy Plant Mini Kit (QIAGEN) according to the protocol of the manufacturer. Complementary DNA was synthesized using the PrimeScript II 1st strand cDNA Synthesis Kit (Takara). The PCR primers are listed in Supplementary Table 1.

Promoter GUS Assay
The promoter regions (3-kb upstream of the start codons) of AT5G18430, ALMT6, AT1G33811, OSP1, AT3G23840, and AT3G17070 were amplified in two PCR steps using the primers listed in Supplementary Table 2 and cloned into pCR8/GW/TOPO followed by subcloning into pGWB433 binary vector. The vectors were transformed into Agrobacterium GV3101, which were then used to transform Col-0 by floral dip. Transformants were selected using kanamycin and carbenicillin and subjected to GUS staining at various developmental stages.

Stomatal Aperture Measurement
The stomatal aperture measurement was performed as described previously (Tomiyama et al., 2014;Toh et al., 2018). Epidermal tissues and leaf disks were prepared from dark-adapted plants and subjected to light illumination and fusicoccin (FC) treatment. The apertures were measured under a microscope (Olympus).

Immunohistochemical Staining of Plasma Membrane H + -ATPase in Guard Cells
The immunohistochemical staining was performed as described previously (Hayashi et al., 2011). Epidermal tissues were prepared from dark-adapted plants and subjected to light illumination and FC treatment. RL (50 µmol m −2 s −1 ) was illuminated for 20 min (Red), and BL (10 µmol m −2 s −1 ) was illuminated with superimposed on RL for 2.5 min (Red + Blue). FC at 10 µM was applied to the epidermal tissue for 5 min in the dark (FC). PM H + -ATPases and the phosphorylation level of the penThr were detected using specific antibodies against the catalytic domain of AHA2 (anti-PM H + -ATPase antibody) and phosphorylated Thr-947 in AHA2 (anti-pThr) (Hayashi et al., 2010).

Genes Preferentially Expressed in Arabidopsis Guard Cells
We analyzed publicly available microarray data on GCPs and MCPs (Yang et al., 2008), and those of the Arabidopsis eFP browsers 1 . The inclusion criteria were as follows: (1) a microarray GCP signal unique to GCPs or at least fourfold higher than the MCP signal and (2) the "Tissue-Specific" criteria of the Arabidopsis eFP browsers indicated GC-specific expression. We retrieved 124 candidate genes and checked their expression levels by RT-PCR in various cells, tissues, and organs of A. thaliana ( Figure 1A). A total of 10 genes were strongly expressed in GCPs but not in MCPs and roots; these included SLAC1, the cation/H + exchanger-encoding AtCHX20, and genes encoding the GDSL lipases OSP1 and ALMT6 (which are preferentially 1 https://bar.utoronto.ca/efp/cgi-bin/efpWeb.cgi expressed in GCs; Padmanaban et al., 2007;Negi et al., 2008;Vahisalu et al., 2008;Tang et al., 2020) and six functionally uncharacterized genes. The RT-PCR data for OSP1, ALMT6, and the uncharacterized genes are shown in Figure 1B.
To further confirm preferential GC expression in intact plants, we constructed transgenic plants expressing the reporter β-GUSencoding gene driven by promoter regions ranging to about 3-kb upstream of the start codons. pALMT6:GUS and pOSP1:GUS exhibited the high-level GUS activity in GCs (in particular) (Figure 1C), consistent with previous findings (Meyer et al., 2011;Tang et al., 2020). Also, the AT1G33811, AT3G23840, AT3G17070, and AT5G18430 promoters drove GC-preferential GUS expression (Figures 1C, 2A). The ALMT6, AT1G33811, OSP1, AT3G17070, and AT3G23840 promoters drove GUS expression in stipules, lateral roots, and trichomes ( Figure 3A); however, the AT5G18430 promoter was active in GCs only (Figure 2A). The AT1G12030 and AT2G32830 promoters never drove GUS expression, rendering the analyses difficult. This is  probably due to the lack of or weak activity of the 3-kb upstream promoter regions of these two genes.

Stomatal Phenotypes of Null Mutants of Genes Preferentially Expressed in Guard Cells
We prepared null mutants of OSP1, AT1G33811, AT3G23840, AT3G17070, and AT5G18430. Although we failed to obtain the knockout mutants of at3g17070 (SALK_121694), the null mutants of at5g18430 (SALK_116756) and at3g23840 (GABI_180G04) were impaired in BL-induced stomatal opening ( Figure 2B) and light-induced stomatal opening (Figure 3B), respectively. Interestingly, the stomata of at1g33811 (GABI_492D11) null mutant were open even in the dark ( Figure 3B). The null mutant of OSP1, i.e., osp1-1, exhibited a normal stable-status stomatal opening in the light (Figure 3B and Supplementary Figure 1), which is consistent with a previous report (Tang et al., 2020). Thus, the previously uncharacterized GC-specific genes AT1G33811, AT3G23840, and AT5G18430 may be involved in stomatal movement.

Mutant in Terms of Blue Light-Induced Stomatal Opening
The ALMT6 is a vacuolar malate channel (Meyer et al., 2011). We explored the stomatal movements of the almt6-1 mutant in FIGURE 3 | Expression patterns of GC-preferentially expressed genes and the stomatal phenotypes of knockout mutants. (A) Gene expression in tissues other than stomata as revealed by the GUS-reporter assay. (B) Upper panels show the RT-PCR of target genes. Lower graphs show BL-induced stomatal opening in at3g23840, at1g33811, and osp1-1 null mutants. Epidermal tissues (at3g23840 and at1g33811 mutants) or leaf disks (osp1-1 mutant) were incubated in 50 mM KCl, 0.1 mM CaCl 2 , and 10 mM Mes-BTP (pH 6.5). Dark bars, dark treatment for 3 h (Dk); white bars, light treatment for 3 h (RL, 50 µmol m −2 s −1 ; BL, 10 µmol m −2 s −1 ) (Lt). Average values from three independent experiments are shown. Error bars: SDs (n = 3).
detail. The expression of ALMT9 in GCs is not altered in almt6-1 (Supplementary Figure 2). As shown in Figure 4A, the stomata of almt6-1 were slightly narrower than wild type under the dark and RL condition, as well as opened but less efficiently to a similar size of those of wild type on BL illumination. Less efficiency of almt6-1 stomatal opening in response to 10 µM FC, an activator of PM H + -ATPase, was more prominent compared with the case of BL-induced stomatal opening ( Figure 4B). Stomatal apertures in the almt6-1 mutant were comparable to those in wild type when epidermal peels were treated with light or FC for more than 3 or 4 h, respectively (Figures 4A,B). Thus, ALMT6 may be required for stomatal opening induced by BL and FC. It is worthy of note that usually BL-insensitive mutants, such as phot1 phot2 double mutant, show completely insensitive phenotype to BL but open normally in response to FC (Kinoshita et al., 2001). The less efficient phenotype of stomatal opening in almt6-1 is very similar to a kincless mutant (Lebaudy et al., 2008) and aks1 aks2 mutant (Takahashi et al., 2013), which shows the low activity of PM inward K + channels in GCs, suggesting that the deficient of ion transport for stomatal opening leads to less efficiency of stomatal opening.
The ALMT6 transports (principally) malate and fumarate but Cl − to a lesser extent (Meyer et al., 2011); we thus explored how Cl − affected BL-induced stomatal opening. The usual stomatal assay buffer contains 50 mM KCl, 0.1 mM CaCl 2 , and 10 mM Mes-BTP (pH 6.5). To exclude exogenous Cl − , we evaluated stomatal opening in a buffer with 50 mM potassium gluconate, 0.1 mM CaCl 2 , and 10 mM Mes-BTP (pH 6.5). Gluconate does  Figure 4C shows that the 3-h BLinduced stomatal opening at a low [Cl − ] was impaired in the almt6-1 mutant in terms of both speed and amplitude.
Blue Light-and FC-Induced Phosphorylation of Plasma Membrane H + -ATPase in the almt6-1 Mutant Both BL and FC induce the phosphorylation of the penThr of PM H + -ATPases, in GCs, which provides a driving force for stomatal opening (Inoue and Kinoshita, 2017). Thus, we immunohistochemically investigated the effects of BL and FC on the phosphorylation status of GC PM H + -ATPase; such phosphorylation was not impaired in the almt6-1 mutant ( Figure 4D). The amount of PM H + -ATPase in almt6-1 under BL and FC was comparative to that in wild type (Supplementary  Figure 3). The almt6 mutation did not affect PM H + -ATPase phosphorylation and amount in response to BL and FC.

DISCUSSION
In this study, we identified 10 genes including SLAC1, Cation/H + Exchanger AtCHX20, GDSL lipases OSP1 and ALMT6, preferentially expressed in GCs based on the analyses of public resources, RT-PCR, and promoter GUS assay (Figures 1-3). Of these, AT1G33811, AT3G17070, AT3G23840, and AT5G18430 have not been functionally characterized in stomata. Among these four genes, three genes, namely, AT1G33811, AT3G23840, and AT5G18430, were found involved in the stomatal movement (Figures 2, 3). Remarkably, two of them, AT1G33811 and AT5G18430, are members of the GDSL family of serine esterases/lipases (Akoh et al., 2004), indicating the importance of this family in regulating stomatal movement. Tang et al. (2020) found that a GDSL lipase, i.e., OSP1, is preferentially expressed in GCs, and osp1 mutants showed low stomatal conductance and high leaf temperature due to a high percentage of occluded stomata. Detail biological and/or biochemical analyses revealed that OSP1 is required for wax biosynthesis and proper formation of the stomatal outer cuticular ledge (Tang et al., 2020). Interestingly, osp1 mutants were also impaired in abscisic acid (ABA)-induced stomatal closure, indicating a potential role of OSP1 in stomatal movement (Tang et al., 2020). It would be very interesting to investigate whether AT1G33811 and AT5G18430 have similar functions as OSP1. AT3G23840, previously named as CER26-like, is probably related to very long-chain fatty acid metabolism (Pascal et al., 2013), and a knockout mutant, at3g23840, showed reduced light-induced stomatal opening ( Figure 3B). Future study is needed to provide a detailed mechanism mediated by those genes in stomatal movement.
The detailed analyses of expression pattern by GUS-reporter assay of ALMT6, OSP1, AT1G33811, AT3G17070, AT3G23840, and AT5G18430 revealed that promoter regions, i.e., 3-kb upstream of the start codon, drive GC-preferential gene expression with AT5G18430 promoter showing the most GCselective property (Figures 1-3). The 1-kb upstream region of the AT3G23840 start codon was promoter-active in flowers (Pascal et al., 2013). The 3-kb promoter region of AT3G23840 studied here showed strong GC signals in addition to flower ( Figure 1B). The 3-kb promoter region of ALMT6 studied here was more GCspecific than a 1.8-kb region used previously, which exhibited strong activities in floral tissues, such as sepals, petals, and anthers (i.e., not only GCs; Meyer et al., 2011) (Figures 1C, 3A). These results suggest that specific elements and their patterns determine the GC-specific activity of promoters. So far, several GC-preferentially expressed genes, such as GC1 (Yang et al., 2008) and MYB60 (Cominelli et al., 2005), were reported. However, these genes were out of our strict criteria, indicating that the genes reported in this study are more specific in GCs (Supplementary Figure 4). Cell type-specific promoters are important tools to study gene function and of great application potential (Imlau et al., 1999;Yang et al., 2008;Wang et al., 2014). The promoters identified in this study thus add the options for promoter engineering for GC-specific gene regulation.
Both ALMT6 and ALMT9 were initially identified as channels mediating malate accumulation in vacuoles (Kovermann et al., 2007;Meyer et al., 2011). Later, ALMT9 was shown to be a Cl − channel regulated by cytosol malate and to be required for the light-induced stomatal opening . As shown in Figures 4A,B, almt6-1 mutant opened stomata with less efficiency in response to BL illumination and FC treatment, suggesting that ALMT6 is also required for BLand FC-induced stomatal opening. Notably, almt6-1 mutant showed significant impairment in BL-induced stomatal opening under low Cl − condition (Figure 4C), suggesting that ALMT6, such as ALMT9, also contributes Cl − influx to vacuole during stomatal opening. Since both almt6 and almt9 single mutants are impaired in the light-induced stomatal opening and there is no compensation of ALMT9 expression in almt6 (Supplementary  Figure 2), it is possible that ALMT6 and ALMT9 function in an additive and/or cooperative manner. In the cooperative mode, a bold hypothesis is that ALMT6 and ALMT9 form heteromeric channels mediating anion accumulation in GC vacuoles as they were shown to form tetramer channels in heterosystems . Future electrophysiological and genetic studies such as phenotyping using almt6 almt9 double mutant are needed to clarify the contribution of ALMT6 and ALMT9 in stomatal opening.
In this study, we identified preferentially expressed genes in GCs and found that some uncharacterized genes are involved in stomatal movement. Especially, to our knowledge, AT5G18430 shows the most specific expression in GCs. In addition, we showed evidence that ALMT6 is important for BL-and FC-induced stomatal opening. Further detailed analyses of the uncharacterized GC-specific genes will provide novel understandings for stomatal movement.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author/s.

AUTHOR CONTRIBUTIONS
WY, SK, and TK designed the experiments. WY, SK, YH, HJ, TO, KK, and TK performed the experiments. WY, SK, YH, and TK wrote the manuscript. All authors contributed to the article and approved the submitted version.

FUNDING
This study was supported by the Grants-in-Aid for Scientific Research from MEXT (Nos. 15H05956, 20H05687, and 20H05910 to TK).