Aquaporins Contribute to ABA-Triggered Stomatal Closure through OST1-Mediated Phosphorylation

Abstract

Stomatal movements in response to environmental stimuli critically control the plant water status. Although these movements are governed by osmotically driven changes in guard cell volume, the role of membrane water channels (aquaporins) has remained hypothetical. Assays in epidermal peels showed that knockout Arabidopsis thaliana plants lacking the Plasma membrane Intrinsic Protein 2;1 (PIP2;1) aquaporin have a defect in stomatal closure, specifically in response to abscisic acid (ABA). ABA induced a 2-fold increase in osmotic water permeability (Pf) of guard cell protoplasts and an accumulation of reactive oxygen species in guard cells, which were both abrogated in pip2;1 plants. Open stomata 1 (OST1)/Snf1-related protein kinase 2.6 (SnRK2.6), a protein kinase involved in guard cell ABA signaling, was able to phosphorylate a cytosolic PIP2;1 peptide at Ser-121. OST1 enhanced PIP2;1 water transport activity when coexpressed in Xenopus laevis oocytes. Upon expression in pip2;1 plants, a phosphomimetic form (Ser121Asp) but not a phosphodeficient form (Ser121Ala) of PIP2;1 constitutively enhanced the Pf of guard cell protoplasts while suppressing its ABA-dependent activation and was able to restore ABA-dependent stomatal closure in pip2;1. This work supports a model whereby ABA-triggered stomatal closure requires an increase in guard cell permeability to water and possibly hydrogen peroxide, through OST1-dependent phosphorylation of PIP2;1 at Ser-121.

Figure 1.

Stomatal Response of Col-0, pip2;1, and PIP2;1-Complemented pip2;1 Plants to Light, ABA, and High CO₂. Peeled epidermal strips from Col-0 (black circles), pip2;1-1 (green squares), pip2;1-2 (red squares), or pip2;1-1 PIP2;1 (blue triangles) were maintained in a bathing solution at ambient air and transferred at t = 0 from darkness to light (300 μE m⁻² s⁻¹). After 180 min, the bathing solution was equilibrated with 10 μM ABA at ambient air (A) or with air containing 800 ppm CO₂ (B). Mean stomatal aperture was measured at the indicated time points. Averaged data (±se; n > 60) from n = 1 to 3 (A) and n = 2 (B) independent experiments, each with ∼60 aperture measurements per time point.

Figure 2.

Effects of ABA on Water Permeability of Guard Cell Protoplasts from Col-0 and pip2;1 Mutant Plants. (A) Representative swelling kinetics of individual Col-0 guard cell protoplasts. The initial rate of swelling together with protoplast size allows determination of P_f. The protoplasts were prepared in the presence of light (open circles: P_f = 60 μm s⁻¹) or light plus 10 μM ABA (closed circles: P_f = 143 μm s⁻¹). (B) Averaged P_f (±se) of protoplasts prepared in the presence of light (gray bars) or light plus 10 μM ABA (black bars) from the indicated genotype. Data from n = 14 to 19 protoplasts and at least three independent plant cultures (t test; P < 0.01).

Figure 3.

ABA-Dependent Accumulation of ROS in Stomata of Col-0 and pip2;1 Mutant Plants. Peeled epidermal strips from Col-0 (blue circles), pip2;1-1 (green squares), or pip2;1-2 (red squares) were maintained in a bathing solution under light for 120 min to induce stomatal aperture. They were then incubated in the presence of 50 μM H₂DCFDA for 20 min, and extracellular H₂DCFDA was removed by four successive washings, prior to addition (t = 0) of 50 μM ABA or an equivalent volume of ethanol (mock). Mean stomatal DCF fluorescence intensity was measured at the indicated time and normalized to the initial fluorescence (t = 4 min) (see data in Supplemental Figure 7). The graph shows the relative difference in fluorescence (%) between ABA- and mock-treated stomata. Averaged data (±se) from n = 7 independent experiments and two independent plant cultures, each experiment with 10 to 20 stomata per time point.

Figure 4.

In Vitro Phosphorylation of PIP2;1 Peptides by OST1. (A) Peptides from PIP2;1 (blue circles) or rbohF (red triangles) and carrying a putative phosphorylation site at Ser-121 or Ser-174, respectively, were incubated at the indicated concentration, in the presence of labeled ATP and purified OST1. Incorporated ATP from n = 4 independent experiments. Error bars (±se) fall into symbols. (B) Phosphorylation by purified OST1 of native or mutated peptides from the loop B and C-terminal region of PIP2;1. Incorporated ATP (±se) from n = 2 to 8 independent experiments was normalized to the signal observed in a native loop B peptide.

Figure 5.

Functional Coexpression in Xenopus Oocytes of Wild-Type or Mutated Forms PIP2;1 with OST1. Oocytes were injected with the indicated amount (nanograms; in parentheses) of cRNAs encoding wild-type (PIP2;1) or mutated (S121A) PIP2;1. When indicated, oocytes were also injected with 5 ng of OST1 cRNA. Uninjected oocytes were used as controls. Osmotic water permeability values (P_f ±se) from n = 10 to 96 oocytes. Different letters indicate statistically different values (one-way ANOVA; Newman-Keuls, P < 0.05).

Figure 6.

Effects of ABA on P_f of Guard Cell Protoplasts from Plants Expressing PIP2;1 Phosphorylation Mutants. P_f was measured as exemplified in Figure 2 in guard cell protoplasts of the indicated genotypes prepared in the absence (gray bars) or presence (black bars) of 10 μM ABA. Data (±se) from n = 7 to 25 protoplasts and at least three independent plant cultures. Asterisks indicate significant effects of ABA on P_f (t test; P < 0.01).

Figure 7.

Stomatal Opening and Closing Responses of Plants Expressing PIP2;1 Phosphorylation Mutants. Stomatal aperture was monitored in Col-0 (black circles), pip2;1-2 (red squares), or pip2;1-2 plants expressing S121A (S121A-1, dark-blue squares; S121A-2, light-blue squares) or S121D (S121D-1, dark-green squares; S121D-2, light-green squares) as described in Figure 1A, except that stomatal opening (t = 0 to 180 min) was induced in the light and in a solution depleted in CO₂. From t = 180 min, epidermal peels were incubated in the same solution but containing 10 μM ABA. Averaged data (±se; n > 150) from n = 3 independent experiments, each with ∼60 aperture measurements per time point.