Rapid structural changes and acidification of guard cell vacuoles during stomatal closure require phosphatidylinositol 3,5-bisphosphate

Abstract

Rapid stomatal closure is essential for water conservation in plants and is thus critical for survival under water deficiency. To close stomata rapidly, guard cells reduce their volume by converting a large central vacuole into a highly convoluted structure. However, the molecular mechanisms underlying this change are poorly understood. In this study, we used pH-indicator dyes to demonstrate that vacuolar convolution is accompanied by acidification of the vacuole in fava bean (Vicia faba) guard cells during abscisic acid (ABA)-induced stomatal closure. Vacuolar acidification is necessary for the rapid stomatal closure induced by ABA, since a double mutant of the vacuolar H(+)-ATPase vha-a2 vha-a3 and vacuolar H(+)-PPase mutant vhp1 showed delayed stomatal closure. Furthermore, we provide evidence for the critical role of phosphatidylinositol 3,5-bisphosphate [PtdIns(3,5)P2] in changes in pH and morphology of the vacuole. Single and double Arabidopsis thaliana null mutants of phosphatidylinositol 3-phosphate 5-kinases (PI3P5Ks) exhibited slow stomatal closure upon ABA treatment compared with the wild type. Moreover, an inhibitor of PI3P5K reduced vacuolar acidification and convolution and delayed stomatal closure in response to ABA. Taken together, these results suggest that rapid ABA-induced stomatal closure requires PtdIns(3,5)P2, which is essential for vacuolar acidification and convolution.

Figure 1.

The Vacuoles of Stomatal Guard Cells Become Highly Convoluted during ABA-Induced Stomatal Closure. ABA-induced vacuolar convolution was observed in fava bean guard cells expressing VHP1-sGFP, a vacuolar membrane marker. (A) to (C) Representative mid-plane confocal images of guard cells expressing VHP1-sGFP before (A) and after ([B] and [C]) treatment with 10 μM ABA. Bars = 10 μm. (D) The extent of vacuolar convolution was quantified as the number of vacuolar compartments observed in mid-plane confocal images. The results are from four independent experiments (mean ± se, 22 ≤ n ≤ 35). [See online article for color version of this figure.]

Figure 2.

ABA Induces Vacuolar Acidification in Fava Bean Stomatal Guard Cells. (A) to (D) Vacuolar acidification as observed using acridine orange (AO). Guard cells were stained with 50 μM AO for 100 min before ABA treatment and the fluorescence ratio of red (R) to green (G) emissions was obtained using a Zeiss LSM510 Meta microscope. Representative R/G ratio images obtained at the indicated time points are shown ([A] to [C]). The R/G ratio is displayed in pseudocolor. Regions with the highest R/G ratio (more acidic) are in white, while those with the lowest R/G ratio are in dark blue. Bars = 10 μm. (D) Vacuolar acidification was quantified as the mean values of R/G ratios in vacuolar lumens. The results are from three independent experiments (mean ± se, 30 ≤ n ≤ 44). A.U., arbitrary unit. (E) to (G) Vacuolar acidification observed with LysoSensor Green DND-189 (lysosensor). Guard cells were stained with lysosensor (4 μM) for 20 min before analysis. Representative confocal images of guard cells stained with lysosensor without (0 min) or with (30 and 60 min) 10 μM ABA treatment are presented. Bars = 10 μm. (H) The extent of vacuolar acidification was quantified as the mean values of green fluorescence intensity of lysosensor in vacuolar lumens. Higher values indicate more acidic conditions. The results are from six independent experiments (mean ± se, 71 ≤ n ≤ 84).

Figure 3.

Loss-of-Function Mutants of Vacuolar Proton Pumps Are Delayed in ABA-Induced Stomatal Closure and Vacuolar Acidification. (A) ABA-induced stomatal closure in Arabidopsis mutant plants defective in vacuolar proton pump activities. Leaves from the wild type (WT) and vacuolar proton pump mutants (vha-a2 vha-a3 and vhp1) were incubated on buffer under white light for 3 h and then transferred to fresh buffer containing 4 μM ABA. Stomatal apertures were measured 30 and 60 min after the transfer. The results are from three independent experiments (mean ± se, 200 ≤ n ≤ 260). Asterisks indicate significant differences at P < 0.0001 compared with the wild type. (B) Impaired vacuolar acidification in vhp1 guard cells. Stomata of 21-d-old wild-type and vhp1 Arabidopsis plants were stained with lysosensor (4 μM) 20 min before observation. The extent of vacuolar acidification was quantified as the mean values of green fluorescence intensity of lysosensor in vacuolar lumens. The results are from four independent experiments (mean ± se, 55 ≤ n ≤ 75). Asterisks indicate significant differences at P < 0.0001 compared with the wild type. A.U., arbitrary unit.

Figure 4.

Treatment with a PI3P5K Inhibitor (PIKfyve Inhibitor) Suppresses Vacuolar Acidification during ABA-Induced Stomatal Closure. (A) PIKfyve inhibitor reduces the PI3P5K activity of Arabidopsis FAB1C in a concentration-dependent manner. Bacterially expressed GST-FAB1C (full length, 1 to 1648 amino acids) and GST-FAB1C kinase domain (856 to 1645 amino acids) were tested for PI3P5K activity in vitro. The production of radiolabeled PtdIns(3,5)P₂ from PtdIns3P and [γ-³²P]ATP was detected using thin layer chromatography and autoradiography. The results shown are representative of three replicate experiments. GST, glutathione S-transferase. (B) to (E) PIKfyve inhibitor suppresses ABA-induced vacuolar acidification, as determined by AO. Representative R/G ratio images of fava bean guard cells after treatment with 10 μM ABA in the absence ([B] and [C]) or presence ([D] and [E]) of PIKfyve inhibitor. Bars = 10 μm. (F) The extent of vacuolar acidification was quantified as the means of the R/G ratios. The results are from three independent experiments (mean ± se, 30 ≤ n ≤ 44). Asterisks indicate significant differences at P < 0.001 (**) and P < 0.0001 (***) between samples treated or not with PIKfyve inhibitor. A.U., arbitrary unit. (G) to (J) Suppression of ABA-induced vacuolar acidification by PIKfyve inhibitor, as observed with lysosensor. Representative confocal images of fava bean guard cells stained with lysosensor after treatment with 10 μM ABA in the absence ([G] and [H]) or presence ([I] and [J]) of PIKfyve inhibitor. Bars = 10 μm. (K) The extent of vacuolar acidification was quantified as the mean values of green fluorescence intensity of lysosensor measured in vacuolar lumens. The results are from six independent experiments (mean ± se, 71 ≤ n ≤ 84). Asterisks indicate significant differences at P < 0.0001 between samples treated or not with PIKfyve inhibitor.

Figure 5.

Treatment with a PIKfyve Inhibitor Suppresses ABA-Induced Stomatal Closure and Vacuolar Convolution. (A) PIKfyve inhibitor suppresses ABA-induced stomatal closure in fava bean. Leaves were incubated on 10 mM MES/10 mM KCl buffer containing 10 μM ABA with various concentrations of PIKfyve inhibitor (0 to 1 μM). The results are from four independent experiments (mean ± se, 130 ≤ n ≤ 298). Asterisks indicate significant differences at P < 0.0001 between samples treated or not with PIKfyve inhibitor. (B) to (E) PIKfyve inhibitor suppresses ABA-induced vacuolar convolution. Representative mid-plane confocal images of fava bean guard cells expressing VHP1-sGFP after treatment with ABA (10 μM) in the absence ([B] and [C]) or presence ([D] and [E]) of PIKfyve inhibitor. Bars = 10 μm. (F) The extent of vacuolar convolution was quantified as the number of vacuolar compartments observed in mid-plane confocal images. The results are from four independent experiments (mean ± se, 22 ≤ n ≤ 35). Asterisks indicate significant differences at P < 0.001 between samples treated or not with PIKfyve inhibitor. [See online article for color version of this figure.]

Figure 6.

Arabidopsis Null Mutants of FAB1s Exhibit Delayed ABA-Induced Stomatal Closure and Accelerated Water Loss. (A) Structure of Arabidopsis FAB1 PI3P5Ks. aa, amino acids. (B) Transcript levels of FAB1 PI3P5Ks in mature Arabidopsis guard cells. Expression was analyzed using quantitative RT-PCR. β-Tubulin was used as an internal control. Phosphoenolpyruvate carboxylase is a marker gene of mesophyll cells and its expression was not detected (N.D.) in our guard cell–enriched epidermal sample. These results are from three independent experiments (mean ± se). (C) Stomatal closure of fab1 mutants is delayed in response to ABA. Wild-type (WT) and fab1b, fab1c, and fab1b fab1c mutant Arabidopsis leaves were incubated under white light for 3 h and then transferred to fresh buffer containing 4 μM ABA. The results are from three independent experiments (mean ± se, 150 ≤ n ≤ 255). (D) fab1 mutant leaves exhibit increased water loss. Water loss is presented as the decrease in weight relative to the initial value after exposure to dry air. The results are from seven independent experiments (mean ± se). FW, fresh weight.

Figure 7.

PtdIns(3,5)P₂ Binds to V-PPase. A PIP strip, a hydrophobic membrane spotted with various phospholipids, was overlaid with purified V-PPase and immunoblotted with anti-V-PPase antibody. V-PPase interacted with PtdIns(3,5)P₂, PtdIns(4,5)P₂, and PtdIns(3,4,5)P₃. LPA, lysophosphatidic acid; LPC, lysophosphocholine; PtdIns, phosphatidylinositol; PtdIns4P, phosphatidylinositol 4-phosphate; PtdIns5P, phosphatidylinositol 5-phosphate; PE, phosphatidylethanolamine; PC, phosphatidylcholine; S1P, sphingosine 1-phosphate; PtdIns(3,4)P₂, phosphatidylinositol 3,4-bisphosphate; PA, phosphatidic acid; PS, phosphatidylserine.

Figure 8.

Treatment with a PIKfyve Inhibitor Suppresses Cytosolic Alkalization during ABA-Induced Stomatal Closure. Cytosolic alkalization was observed using BCECF-AM. Fava bean guard cells were incubated with 20 μM BCECF-AM for 30 min before observation and the emission ratio with cyan (488 nm) to blue (458 nm) excitation was obtained using a Zeiss LSM510 Meta microscope. Cyan-to-blue ratios of fava bean guard cells after treatment with 10 μM ABA in the absence or presence of PIKfyve inhibitor were compared. The results are from five independent experiments (mean ± se, 59 ≤ n ≤ 87). Asterisks indicate significant differences at P < 0.05 (**) and P < 0.001 (***) between samples with or without PIKfyve inhibitor. A.U., arbitrary unit.

Figure 9.

Model Showing the Mechanism by Which PtdIns(3,5)P₂ May Regulate Stomatal Closure. A stomatal closure signal (e.g., ABA) may induce the activation of the PI3P5Ks (e.g., FAB1B and FAB1C), causing an increase in (PtdIns[3,5]P₂). PtdIns(3,5)P₂ may activate vacuole membrane-localized proton pumps and thereby acidify the vacuolar lumen. The acidified vacuole undergoes structural changes and releases osmolytes, allowing rapid stomatal closure. The vacuolar acidification likely contributes to the maintenance of alkalized cytosolic pH and vice versa. Solid lines, relationships based on experimental data; dashed lines, hypothesized relationships. [pH]vac, vacuolar pH; [pH]_cyt, cytosolic pH.