Danger-Associated Peptides Close Stomata by OST1-Independent Activation of Anion Channels in Guard Cells

A microbial pathogen induces endogenous plant elicitor peptides that close stomata through a PRR-dependent but OST1-independent pathway.

Abstract

The plant elicitor peptides (Peps), a family of damage/danger-associated molecular patterns (DAMPs), are perceived by two receptors, PEPR1 and PEPR2, and contribute to plant defense against pathogen attack and abiotic stress. Here, we show that the Peps-PEPR signaling pathway functions in stomatal immunity by activating guard cell anion channels in Arabidopsis thaliana. The mutant plants lacking both PEPR1 and PEPR2 (pepr1 pepr2) displayed enhanced bacterial growth after being sprayed with Pseudomonas syringae pv tomato (Pst) DC3000, but not after pathogen infiltration into leaves, implicating PEPR function in stomatal immunity. Indeed, synthetic Arabidopsis Peps (AtPeps) effectively induced stomatal closure in wild-type but not pepr1 pepr2 mutant leaves, suggesting that the AtPeps-PEPR signaling pathway triggers stomatal closure. Consistent with this finding, patch-clamp recording revealed AtPep1-induced activation of anion channels in the guard cells of wild-type but not pepr1 pepr2 mutant plants. We further identified two guard cell-expressed anion channels, SLOW ANION CHANNEL1 (SLAC1) and its homolog SLAH3, as functionally overlapping components responsible for AtPep1-induced stomatal closure. The slac1 slah3 double mutant, but not slac1 or slah3 single mutants, failed to respond to AtPep1 in stomatal closure assays. Interestingly, disruption of OPEN STOMATA1 (OST1), an essential gene for abscisic acid-triggered stomatal closure, did not affect the AtPep1-induced anion channel activity and stomatal response. Together, these results illustrate a DAMP-triggered signaling pathway that, unlike the flagellin22-FLAGELLIN-SENSITIVE2 pathway, triggers stomata immunity through an OST1-independent mechanism.

INTRODUCTION

To recognize invading microbes, plants have evolved specialized plasma membrane-localized pattern recognition receptors (PRRs) that are often receptor-like kinases (Boller and Felix, 2009). These recognition receptors can interact specifically with conserved pathogen- or microbe-associated molecular patterns (PAMPs/MAMPs) from invading microbes and induce pattern-triggered immunity defense processes. In Arabidopsis thaliana, the receptor kinase FLAGELLIN-SENSITIVE2 (FLS2) recognizes a conserved 22-amino acid peptide from the bacterial flagellin protein, flg22 (Monaghan and Zipfel, 2012). Upon perception of flg22, FLS2 recruits a leucine-rich repeat coreceptor, BRASSINOSTEROID INSENSITIVE1-associated receptor kinase 1 (BAK1), and induces phosphorylation of both proteins to form an active immunity complex (Chinchilla et al., 2007; Sun et al., 2013). A receptor-like cytoplasmic kinase, BOTRYTIS-INDUCED KINASE1 (BIK1), serves as a direct substrate of the FLS2-BAK1 complex and further interacts with and phosphorylates the NADPH oxidase Respiratory Burst Oxidase Homolog D (RBOHD) to trigger a reactive oxygen species (ROS) burst and downstream antibacterial immunity (Kadota et al., 2014; Li et al., 2014).

In addition to PAMPs, plant endogenous damage- or danger-associated molecular patterns (DAMPs), such as cell wall fragments and plant peptides, could also be perceived by specific PRRs to activate pattern-triggered immunity as a supplementary defense mechanism. In Arabidopsis, AtPeps are derived from precursor proteins called PROPEPs and are perceived by two closely related receptor kinases called PEPR1 and PEPR2, leading to immune responses (Yamaguchi et al., 2006, 2010). PROPEPs are transcriptionally activated by wounding, the defense hormone jasmonic acid, and pathogen challenge (Huffaker and Ryan, 2007; Huffaker et al., 2011, 2013; Bartels et al., 2013; Ross et al., 2014). Pretreatment of plants using Peps has been shown to improve plant resistance against bacterial or fungal infections as well as herbivore attacks (Huffaker et al., 2011, 2013; Tintor et al., 2013; Klauser et al., 2015), implicating Peps as endogenous chemical messengers for defense responses in plants. Intriguingly, like FLS2, PEPR1 also specifically interacts with BAK1 and phosphorylates BIK1 to mediate ethylene-induced defenses (Liu et al., 2013). As a result, the Peps-PEPR system is widely considered as a PAMP-triggered pathway and an amplifier of innate immunity (Bartels and Boller, 2015).

Most plant pathogenic bacteria enter plant tissues through wounds or natural openings (Hugouvieux et al., 1998; Gudesblat et al., 2009). Plant stomatal pores, formed by a pair of guard cells, are dynamic structures that modulate gas exchange in response to diverse environmental factors. Stomata enable gas exchange for photosynthesis and transpirational water loss, but they are also the major natural openings for bacterial entry. Interestingly, both human and plant pathogenic bacteria can induce stomatal closure within the first hour of application to plant epidermis, indicating that guard cells are equipped with an intrinsic mechanism to sense conserved bacterial molecules or PAMPs (Melotto et al., 2006). Indeed, this study shows that several key components of the abscisic acid (ABA) signal transduction pathway, such as ABA synthesis, NO production, and the OST1 kinase, are required for PAMP-induced stomatal closure, illustrating a mechanistic connection between PAMP-induced and ABA-dependent signaling pathway in closing stomata.

Stomatal pore aperture is largely controlled by cell turgor, which is regulated by fluxes of solutes and water in and out of guard cells. Stomatal closure is initiated by the efflux of anions through anion channels in the plasma membrane of guard cells, which depolarizes the plasma membrane and activates outward cation channels, resulting in efflux of solutes and decreased turgor in guard cells (Kim et al., 2010). The S-type anion channels are particularly important in the control of guard cell turgor pressure and are encoded by a small family of genes discovered through independent Arabidopsis mutant screens for CO₂ and ozone response mutants (Negi et al., 2008; Vahisalu et al., 2008). The founding member of this family, SLAC1, is essential for the plasma membrane S-type anion channel function in guard cells because disruption of SLAC1 expression impaired the stomatal closure induced by ozone, CO₂, ABA, and calcium (Negi et al., 2008; Vahisalu et al., 2008). Furthermore, a homolog of SLAC1, SLAC1 HOMOLOG3 (SLAH3), also contributes to S-type current in guard cells and is required for nitrate transport in the roots (Geiger et al., 2011; Zheng et al., 2015). Both SLAC1 and SLAH3 show anion channel activity in Xenopus laevis oocytes when coexpressed with their interacting protein kinases (Geiger et al., 2009, 2010, 2011; Lee et al., 2009; Brandt et al., 2012).

The SnRK2- and CDPK-induced activation of SLAC1 and SLAH3 has been considered a primary mechanism for ABA-induced stomatal closure (Murata et al., 2015). In particular, the predominant member of the SnRK2-type kinase family, OST1, appears to be the key regulator of stomatal closure (Assmann, 2003; Acharya et al., 2013). The loss-of-function mutants of OST1 display constantly open stomata even under stress conditions, causing rapid wilting of the mutant plants. Recent studies further show that OST1 is required for stomatal closure in response to bacterial pathogens and microbial elicitors (Melotto et al., 2006; Guzel Deger et al., 2015; Ye et al., 2015), solidifying the importance of OST1 (and ABA signaling) in stomatal immunity.

In connecting pathogen response and DAMP signaling pathways, a study showed that Arabidopsis plants lacking PEPR1 and PEPR2 displayed an increased susceptibility to Pseudomonas syringae pv tomato (Pst) DC3000 (Tintor et al., 2013). Here, we also observed enhanced bacterial growth in pepr1 pepr2 double mutant plants that were sprayed with Pst DC3000, but not in plants infiltrated with bacteria, suggesting that PEPR may play a role in stomatal immunity. We further showed that synthetic AtPeps triggered stomatal closure in a PEPR-dependent manner in Arabidopsis. The AtPep1-induced stomatal response depends on both SLAC1 and SLAH3, which function synergistically in this pathway. Although the AtPeps-PEPR system shares common components, such as the coreceptor BAK1 and downstream receptor-like cytoplasmic kinase BIK1, with the flg22-FLS signaling pathway, our results showed that OST1 was not required in AtPep1-induced stomatal response. This study demonstrates that the AtPeps-PEPR-BIK1 signaling system and SLAC1/SLAH3 channels provide an OST1-independent mechanism to modulate stomatal immunity in Arabidopsis.

RESULTS

The AtPeps-PEPR System Functions in Stomatal Immunity

As danger signals, AtPeps are transcriptionally induced by environmental stress conditions including biotic factors such as bacterial PAMPs (e.g., flg22) and mechanical or herbivore-mediated wounding, leading to pathogen defense and herbivore responses (Huffaker et al., 2013; Tintor et al., 2013). In Arabidopsis, disruption of PEPR1 and PEPR2 completely abolished the AtPep1-induced responses, including root growth inhibition and protection against bacterial growth (Krol et al., 2010; Yamaguchi et al., 2010). To examine whether the AtPeps-PEPR system functions during bacterial infection in leaves, we spray-inoculated the pepr1 pepr2 double mutant plants with a Pst DC3000 bacterial suspension. Three days after inoculation, enhanced bacterial growth was observed in pepr1 pepr2 mutant leaves (Figure 1A), consistent with the data reported previously (Tintor et al., 2013). Enhanced bacterial susceptibility of the pepr1 pepr2 mutant suggests that the AtPeps-PEPR system may function in restricting bacterial entry or propagation.

Figure 1.

AtPeps Induce Stomatal Closure in Arabidopsis.

(A) Growth of Pst DC3000 bacteria in leaves of 4-week-old plants 3 d after spray inoculation or infiltration with bacteria suspensions at 5 × 10⁸ cfu/mL and 1 × 10⁵ cfu/mL, respectively. Results are shown as means ± sd from three independent experiments (n = 6). Asterisks indicate means that are significantly different from the wild type with the same treatment at P < 0.01.

(B) Representative images of stomatal aperture in epidermal peel of 4-week-old Col-0 leaves exposed to 1 μM AtPep1, 1 μM AtPep2, or 20 μM ABA solution. Photographs were taken 30 min after incubation. Bars = 20 μm.

(C) Stomatal aperture (ratio of stomatal width:length) in epidermal peel of Col-0 leaves exposed to 1 μM AtPep1, 1 μM AtPep2, or 20 μM ABA for 30 min. Results are shown as means ± sd from three independent experiments (n = 180). Asterisks indicate means that are significantly different from the control at P < 0.01.

(D) Time-course changes in stomatal aperture with (closed squares) or without (open squares) 1 μM AtPep1 stimulation. Results are shown as means ± sd from three independent experiments (n = 60). Asterisks indicate means significantly different from the control of corresponding time point at P < 0.01.

(E) Concentration-dependent stomatal aperture changes after Col-0 leaves were incubated within 0 to 10 μM AtPep1 solution for 30 min. Results are shown as means ± sd from three independent experiments (n = 180). Asterisks indicate means that are significantly different from the control without AtPep1 treatment at P < 0.01.

To test if bacterial entry into plants was altered in the mutant, we infiltrated Pst DC3000 directly into leaves to bypass the effect of stomatal immunity and found that infiltrated Pst DC3000 showed similar growth in pepr1 pepr2 and wild-type Col-0 leaves, which was different from the results we observed when the bacteria were sprayed on the surface of leaves (Figure 1A). These results suggested that the AtPeps-PEPR system plays a role in preventing bacterial entry into plant leaves. The similarity of infiltrated bacterial growth in wild-type and pepr1 pepr2 leaves is consistent with prior research showing that, without AtPep1 pretreatment, the AtPeps-PEPR system had little effect on bacterial growth restriction when plants are inoculated with bacteria by infiltration procedure (Yamaguchi et al., 2010; Tintor et al., 2013). Together, these data indicate that the AtPeps-PEPR pathway may play a major role in stomatal immunity during bacterial infection.

Guard cells detect pathogenic bacteria as well as their conserved surface PAMPs to trigger stomatal closure (Melotto et al., 2006; Guzel Deger et al., 2015; Ye et al., 2015). However, danger-associated peptides such as AtPeps have not been previously examined in stomatal response. Our finding of PEPR function in stomatal defense suggests that the AtPeps-PEPR signaling pathway may contribute to plant immunity by inducing stomatal closure as a mechanism to restrict bacterial entry. To test this hypothesis, we used synthetic AtPeps to examine their effect on stomatal movements in the wild-type plants. Leaf samples were first incubated in stomatal opening buffer to facilitate stomatal opening, followed by application of agents such as ABA or AtPep1. Within half an hour of incubation in AtPep1 solution, like incubation in the ABA solution, dramatic decrease of stomatal aperture (as shown by width:length ratio of a stoma) was observed in the wild-type leaves (Figures 1B and 1C). In addition to AtPep1, we also used another AtPep, AtPep2, to treat the leaves and found a similar effect in the stomatal closure assay (Figures 1B and 1C), indicating that AtPeps generally induce stomatal closure. In a time-course analysis, we found that the stomatal aperture decreased rapidly within 15 to 30 min upon 1 μM AtPep1 exposure and, interestingly, recovered after 30 min of treatment to almost fully open state in 1 h (Figure 1D). The oscillatory behavior of stomatal aperture persisted throughout the 2 h assay. To determine the concentration range of AtPep1 in stomatal closure response, we performed a serial dilution and found that AtPep1 at a concentration as low as 10 nM induced stomatal closure in half an hour, suggesting that AtPep1 is highly effective in closing stomata (Figure 1E).

In order to investigate whether AtPeps-induced stomatal closure is PEPR-dependent, single and double mutants lacking PEPR1 or PEPR2 or both were examined in the stomatal closure assay described above. We found that stomatal aperture in the pepr1-2 single mutant did not respond to AtPep1 exposure, whereas the pepr2-2 single mutant responded like the wild type (Figure 2A). The pepr1 pepr2 double mutant, like the pepr1-2 single mutant, failed to respond to AtPep1 treatment (Figure 2A). Furthermore, a genomic fragment containing the PEPR1 gene, transformed into pepr1 pepr2 plants, restored the AtPep1-induced stomatal response in the double mutant (PEPR1/pepr1 pepr2) (Figure 2A; Supplemental Figure 1).

Figure 2.

Involvement of PEPR Receptors in AtPep1-Induced Stomatal Closure.

(A) Stomatal responses in leaves of Col-0, pepr1-2, pepr2-2, and pepr1 pepr2 mutant and PEPR1 complemented pepr1 pepr2 transgenic plants (PEPR1/pepr1 pepr2) to 1 μM AtPep1 solution. Stomatal apertures were measured 30 min after incubation. Results were shown as means ± sd from three independent experiments (n = 180). Asterisks indicate means that are significantly different from corresponding genotype control without AtPep1 treatment at P < 0.01.

(B) Promoter sequences of PEPR1 and PEPR2 fused to GUS reporter revealed their expression in guard cells. GUS staining of leaf epidermis from proPEPR1:GUS and proPEPR2:GUS transgenic plants, respectively. Bars = 10 μm.

It is believed that PEPR1 and PEPR2 are functionally equivalent in AtPeps signaling (Yamaguchi et al., 2010). However, differential function of PEPR1 and PEPR2 in AtPep-induced stomatal closure suggests that these two receptors may have different roles in transmitting the signals (AtPeps). We therefore performed a root growth inhibition assay, another bioassay for AtPeps, and found that, in sharp contrast to the stomatal assay, the pepr2 mutant, but not the pepr1 mutant, became less sensitive to AtPep1 (Supplemental Figure 2) (Krol et al., 2010), suggesting that the two receptors may function in different processes. In other words, PEPR1 appears to play a more dominant role in AtPep1-induced stomatal response, whereas PEPR2 emerges as a major player in AtPep1 perception in roots.

Their differing roles may result from differential expression of the receptors in different tissues. To examine the expression patterns of PEPR1 and PEPR2, we constructed GUS reporters driven by PEPR1 and PEPR2 promoters (proPEPR1:GUS and proPEPR2:GUS) and transformed them into Arabidopsis plants. Histochemical staining of plant tissues demonstrated that GUS activity in both proPEPR1:GUS and proPEPR2:GUS transformants was detected ubiquitously in roots and leaves, consistent with results reported earlier (Bartels et al., 2013). Furthermore, we also observed GUS activity in guard cells from both proPEPR1:GUS and proPEPR2:GUS transgenic plants (Figure 2B). These data indicated that PEPR1 and PEPR2 are both expressed in the guard cells but functionally different in AtPep1-induced stomatal response.

Another possible mechanism for differential action of PEPR1 and PEPR2 is distinct binding of AtPep1 to the two receptors. We thus treated the pepr1-2 and pepr2-2 single mutants with AtPep2 and found that both mutants responded to AtPep2 by closing stomata (Figure 3A). In contrast, AtPep1 induced stomatal closure only in the pepr2 mutant (Figure 3A). In both cases, the pepr1 pepr2 double mutant failed to respond to the peptide signals. Furthermore, we examined the peptide-induced MPK3/MPK6 activation by spraying plants leaves with AtPeps solution. Our data showed that both AtPep1 and AtPep2 induced the activation of MAP3/MAP6 in the wild-type leaves (Figures 3B and 3C). However, MPK3/MPK6 activation by AtPep1 was reduced in pepr1-2 mutants, but not in pepr2-2, compared with the wild-type control (Figure 3B). By spraying AtPep2, we noted that both pepr1-2 and pepr2-2 mutants responded by MPK3/MPK6 activation, similar to the wild type (Figure 3C). MPK3/MPK6 activation in the pepr1 pepr2 double mutant failed to respond to either AtPep1 or AtPep2 (Figures 3B and 3C). These data support the idea that PEPR1 and PEPR2 appear to contribute equally to AtPep2-mediated responses but PEPR1 plays a major role in AtPep1-induced responses.

Figure 3.

Differential Function of PEPR1 and PEPR2 in AtPeps-Induced Stomatal Closure and MAPK Activation.

(A) Stomatal apertures of Col-0, pepr1-2, pepr2-2, and pepr1 pepr2 plant leaves treated with 1 μM AtPep1 or AtPep2 for 30 min. Results were shown as means ± sd from three independent experiments (n = 180). Asterisks indicate means that are significantly different from corresponding genotype control at P < 0.01.

(B) and (C) Immunoblot with total proteins extracted from leaves of 4-week-old Col-0, pepr1-2, pepr2-2, and pepr1 pepr2 plants sprayed with 1 μM AtPep1 (B) or AtPep2 (C) for 30 min. MAPKs were detected (upper panel) with the rabbit anti-phospho-p44/42 MAPK (Erk1/2) (Thr-202/Tyr-204) monoclonal antibody. Ponceau S staining of proteins on PVDF membrane was shown for loading control (lower panel).

AtPep1-PEPR Pathway Activates S-Type Anion Channels in Guard Cells

Activation of S-type anion channels plays a central role in decreasing guard cell turgor, leading to stomatal closure (Negi et al., 2008; Vahisalu et al., 2008; Kim et al., 2010). To explore the mechanism of AtPep1-induced stomatal response, we used the whole-cell patch clamp technique to measure the S-type anion channel activity across the plasma membrane of guard cells as described earlier (Geiger et al., 2011). Under the control conditions without AtPep1, all samples exhibited comparable basal levels of S-type NO₃⁻ currents around 40 pA/pF at the test voltage of −145 mV (Figure 4). Upon exposure to AtPep1, the S-type NO₃⁻ currents of wild-type guard cells increased by ∼80% over the control (Figures 4A and 4B). Under the same condition, guard cells from pepr1-2 and pepr2-2 single mutants both displayed smaller S-type NO₃⁻ currents than the wild-type plants (Figures 4A and 4B). Guard cells from pepr1 pepr2 double mutants did not show significant increase in the NO₃⁻ current after addition of AtPep1 (Figures 4A and 4B). Furthermore, AtPep1-induced activation of S-type anion channels recovered in the PEPR1/pepr1 pepr2 transgenic plant (Figures 4A and 4B), which was consistent with the stomatal aperture assay. Taken together, these results suggest that the S-type anion channel in guard cells is activated by the AtPep1-PEPR signaling pathway, leading to stomatal closure.

Figure 4.

PEPR1 and PEPR2 Are Responsible for S-Type Anion Channel Activity of Guard Cells Induced by AtPep1.

S-type anion channel activity in nitrate-based media recorded from guard cells isolated from leaves treated with or without 1 μM AtPep1 solution for 30 min.

(A) Representative whole-cell nitrate current density of Col-0, pepr1-2, pepr2-2, pepr1 pepr2, and PEPR1/pepr1 pepr2.

(B) Current density-voltage curves derived from whole-cell anion currents in guard cells isolated from Col-0, pepr1-2, pepr2-2, pepr1 pepr2, and PEPR1/pepr1 pepr2 leaves treated with or without 1 μM AtPep1 for 30 min as shown in (A). Results were shown as means ± sd from three independent experiments (Col-0, n = 7; Col-0+AtPep1, n = 6; pepr1-2, n = 6; pepr1-2+AtPep1, n = 6; pepr2-2, n = 5; pepr2-2+AtPep1, n = 6; pepr1 pepr2, n = 7; pepr1 pepr2 +AtPep1, n = 7; PEPR1/pepr1 pepr2, n = 6; PEPR1/pepr1 pepr2+AtPep1, n = 6).

(C) Recording protocol. Current density traces were recorded in response to voltage pulses ranging from –145 to +35 mV in 30-mV decrements.

The study on anion selectivity of guard cell S-type anion channels indicates that they display high permeability to nitrate (Schmidt and Schroeder, 1994). Under our recording ionic conditions (100 mM NO₃⁻ and 13.728 mM Cl⁻ in pipette; 30 mM NO₃⁻ and 9 mM Cl⁻ in bath solution), a rapid voltage ramp (60 mV⋅s⁻¹) analysis gave a reversal potential of 26 to 29 mV (Supplemental Figure 3). Using the Goldman-Hodgkin-Katz equation and activities of the ions, a relative permeability for NO₃⁻ to Cl⁻ of ∼10 was derived, which is similar to the high nitrate permeability previously reported for guard cell anion currents (Schmidt and Schroeder, 1994). Moreover, AtPep1 treatment did not change the reversal potential of anion currents, whereas the current amplitude was increased significantly by AtPep1 application in the wild type (Supplemental Figure 3), suggesting that AtPep1 activates the same channels recorded in the control.

SLAC1 and SLAH3 Function Additively in AtPep1-Induced Stomatal Closure

The S-type anion channel activity in guard cells is produced predominantly by SLAC1, which constitutes an active channel in a heterologous expression system such as frog oocytes, when coexpressed with the proper regulators (Geiger et al., 2009, 2010; Lee et al., 2009). Previous studies show that mutant plants lacking SLAC1 are compromised in ABA-, ozone-, Ca²⁺- and high CO₂-induced stomatal closing responses, implying that SLAC1 is the major component of S-type channel in these responses (Negi et al., 2008; Vahisalu et al., 2008). We thus analyzed the stomatal movement of slac1-5 loss-of-function plants in response to AtPep1. Surprisingly, stomata of slac1-5 plants showed similar sensitivity to AtPep1 as wild-type plants (Figure 5A). Patch-clamp experiments with guard cells from slac1-5 plants revealed that the NO₃⁻-based S-type currents were halved comparing to the wild type (Figure 5B). With AtPep1 treatment, however, the S-type anion currents in slac1-5 guard cells increased more than 100% over the control (Figures 5B and 5C). These results suggested that an anion channel other than SLAC1 counts for AtPep1-activated S-type anion current in guard cells and such a current is sufficient to support AtPep1-induced stomatal closure in the slac1-5 mutant.

Figure 5.

Both SLAC1 and SLAH3 Are Responsible for Slow Anion Channel Activity Induced by AtPep1.

(A) Stomatal responses of Col-0, slac1-5, slah3-4, and slac1 slah3 double mutant leaves to 30 min treatments with 1 μM AtPep1 solution. Results were shown as means ± sd from three independent experiments (n = 180). Asterisks indicate means that are significantly different from corresponding genotype control without AtPep1 treatment at P < 0.01.

(B) Representative whole-cell nitrate current density in guard cells isolated from Col-0, slac1-5, slah3-4, and slac1 slah3 double mutant leaves treated with or without 1 μM AtPep1 solution for 30 min.

(C) Current density-voltage curves derived from whole-cell anion currents as in (B). Results were shown as means ± sd from three independent experiments (Col-0, n = 6; Col-0+AtPep1, n = 6; slac1-5, n = 6; slac1-5+AtPep1, n = 7; slah3-4, n = 5; slah3-4+AtPep1, n = 7; slac1 slah3, n = 7; slac1 slah3+AtPep1, n = 7).

(D) Promoter sequences of SLAC1 and SLAH3 fused to GUS reporter revealed their expression patterns. GUS staining of leaves and roots from proSLAC1:GUS and proSLAH3:GUS transgenic plants, respectively. Rt, root; Lf, leaf. Bars = 50 μm.

In addition to SLAC1, four genes homologous to SLAC1 (SLAHs) are present in the Arabidopsis genome. The members of this gene family exhibit distinctive expression patterns but have similar transport properties (Negi et al., 2008). Like SLAC1, SLAH3 is also expressed in guard cells and mediates anion efflux (Figure 5D) (Geiger et al., 2011; Zheng et al., 2015). To determine whether SLAH3 is involved in AtPep1-induced stomatal closure, we performed a stomatal assay using slah3-4 mutant plants and found a normal response to AtPep1 (Figure 5A). However, when the S-type anion channel activity was recorded from the mutant guard cells by using NO₃⁻ as carrier, we observed significantly reduced currents compared with wild-type plants (Figure 5B), indicating that SLAH3 contributed to the S-type anion channel activity in guard cells. Nevertheless, the magnitude of these currents significantly increased after AtPep1 treatment (Figures 5B and 5C), suggesting presence of another AtPep1-inducible channel, in addition to SLAH3, sufficiently supporting AtPep1-induced stomatal closure.

As both slac1 and slah3 single mutants exhibited reduced guard cells S-type anion (NO₃⁻) current but normal stomatal responses to AtPep1, we speculated that both SLAC1 and SLAH3 contribute to the S-type anion channel activity in guard cells and each of them is sufficient to support AtPep1-induced stomatal closure. To test this hypothesis, we generated slac1 slah3 double mutant by crossing slac1-5 and slah3-4 single mutants (Supplemental Figure 4). In the stomatal assay, we found that stomata of slac1 slah3 plants failed to respond to AtPep1 treatment (Figure 5A). When S-type anion channel activity was analyzed, guard cells of slac1 slah3 plants displayed much smaller whole-cell NO₃⁻ currents compared with the wild type or their single mutant plants. Furthermore, the NO₃⁻ currents did not show significant difference before or after AtPep1 treatment (Figures 5B and 5C). These results support the conclusion that guard cell-expressed S-type anion channels SLAC1 and SLAH3 both contribute to the total S-type anion channel activity responsible for AtPep1-induced stomatal closure. Interestingly, either one of them is sufficient to support AtPep1-induced stomatal response because both slac1 and slah3 single mutant responded to AtPep1 normally in the stomatal closure assay.

OST1 Is Not Essential for AtPep1-Induced Stomatal Closure or Activation of S-Type Anion Channels

So far, we have established that both SLAC1 and SLAH3 play a role in stomatal responses to AtPep1-PEPR signaling. To explore the signaling mechanism for the activation of these channels, we examined the possible function of several signaling molecules including a SnRK2-type kinase, OST1, which has been shown to be essential for ABA-induced activation of S-type anion channels in guard cells and ABA-induced stomatal closure (Geiger et al., 2009; Lee et al., 2009; Joshi-Saha et al., 2011; Brandt et al., 2012). In particular, more work also connects OST1 with PAMP-induced stomatal immunity responses (Mustilli et al., 2002; Melotto et al., 2006). Further study shows that OST1 is required for flg22-induced stomatal closure and activation of S-type anion channel in guard cell (Guzel Deger et al., 2015). Because the AtPep1-PEPR signaling shares several common components with the flg22-FLS2 signaling, we expected that AtPep1 signaling may also requires OST1 to initiate stomatal closure. Therefore, we challenged the leaves of two independent ost1 mutant alleles, ost1-3 and ost1-4, with ABA, flg22, or AtPep1, respectively. Both ost1 mutants failed to respond to ABA or flg22, but were, surprisingly, fully responsive to AtPep1 (Figure 6A), indicating that OST1 is essential for stomatal closure induced by ABA and flg22 but not by AtPep1. Furthermore, guard cells of both ost1 mutants exhibited significantly reduced basal S-type nitrate currents as compared with the wild type under control conditions. In the presence of ABA and flg22, guard cells from wild-type plants showed increased currents, but there was no significant increase in the currents of guard cells from the two ost1 mutants (Figures 6B and 6C). In response to AtPep1 exposure, however, the S-type anion currents increased 54% and 80% in guard cells of ost1-3 and ost1-4, respectively (Figures 6B and 6C). Moreover, we constructed the ost1 slah3 double mutant (Supplemental Figure 5) and found that its stomata remained responsive to AtPep1 (Figure 7A), further supporting the conclusion that OST1 is not required for SLAC1 activation in the AtPep1-PEPR signaling pathway.

Figure 6.

OST1 Is Essential for ABA- and flg22- but Not AtPep1-Induced Stomatal Closing Responses.

(A) Stomatal responses of Col-0 and ost1 mutant leaves to treatments with 20 μM ABA, 2 μM flg22, or 1 μM AtPep1 for 30 min. Results were shown as means ± sd from three independent experiments (n = 180). Asterisks indicate means that are significantly different from corresponding genotype control at P < 0.01.

(B) Representative whole-cell nitrate current density in guard cells isolated from Col-0, ost1-3, and ost1-4 leaves treated with 20 μM ABA, 2 μM flg22, or 1 μM AtPep1 for 30 min.

(C) Current density-voltage curves derived from whole-cell anion currents in guard cells isolated from Col-0, ost1-3, and ost1-4 leaves treated with 20 μM ABA, 2 μM flg22, or 1 μM AtPep1 as shown in (B). Results were shown as means ± sd from three independent experiments (Col-0, n = 5; Col-0+ABA, n = 7; Col-0+flg22, n = 6; Col-0+ AtPep1, n = 7; ost1-3, n = 6; ost1-3+ABA, n = 6; ost1-3+flg22, n = 7; ost1-3+AtPep1, n = 8; ost1-4, n = 5; ost1-4+ABA, n = 6; ost1-4+flg22, n = 6; ost1-4+AtPep1, n = 7).

Figure 7.

AtPep1-Induced Stomatal Closure Is Independent of ROS Production.

(A) Stomatal responses of Col-0, ost1 slah3, and rbohd rbodf plant leaves to treatments with 1 μM AtPep1 for 30 min. Results were shown as means ± sd from three independent experiments (n = 180). Asterisks indicate means that are significantly different from corresponding genotype control without AtPep1 treatment at P < 0.01.

(B) DAB staining of Col-0 and rbohd rbodf leaves 30 min after spraying with water (mock) or 1 μM AtPep1 solution.

(C) Quantification of ROS production in guard cells of the wild type and rbohd rbodf mutant 30 min after 1 μM AtPep1 treatment. Fluorescence intensities were quantified as average pixel intensities of each guard cell using Image J. The relative ROS production of each treatment was normalized to control wild type (100%). Results were shown as means ± sd from three independent experiments (n = 15). Asterisks indicate means that are significantly different from corresponding genotype control without AtPep1 treatment at P < 0.01.

ROS play an essential role in ABA signaling (Kim et al., 2010). The ABA-induced ROS production appears to be defective in the ost1 guard cells, suggesting that OST1 plays a positive role in ABA-induced ROS generation (Mustilli et al., 2002). The plant NADPH oxidases, also known as respiratory burst oxidase homologs (RBOHs), are extensively studied (Marino et al., 2012). OST1 interacts with both RBOHD and RBOHF and phosphorylate the N terminus of RBOHF (Sirichandra et al., 2009; Acharya et al., 2013). Interestingly, RBOHD has also been identified as required for PAMP-induced stomatal closure (Kadota et al., 2014; Li et al., 2014) and for AtPep3-induced immune defense response (Ma et al., 2013). In our study, stomata of rbohd rbohf double mutants, like those in the wild type, showed similar response to AtPep1 stimulus (Figure 7A), suggesting that RBOHD and RBOHF are not essential for AtPep1-induced stomatal response. By DAB (3,3′-diaminobenzidine) staining assay, we observed AtPep1-induced ROS production in wild-type leaves but not in the rbohd rbohf double mutant (Figure 7B). We further analyzed the accumulation of ROS using H₂DCF-DA (2′,7′-dichlorodihydrofluorescein diacetate) and found that AtPep1-induced ROS was almost eliminated in rbohd rbohf mutant compared with the wild-type plants (Figure 7C; Supplemental Figure 6). These data suggest that AtPep1 activates RBOH-dependent ROS production in plant tissues including guard cells, but RBOHD and RBOHF are not required for AtPep1-induced stomatal closure.

BIK1 Is Required for AtPep1-Induced Stomatal Closure

The fact that OST1 is not essential for AtPep1-induced stomatal closure suggested that the AtPep1-PEPR pathway utilizes a downstream component that differs from the ABA response pathway. Further, the AtPep1-PEPR pathway apparently differ also from the flg22-FLS2 pathway despite sharing several early signaling molecules including BAK1 and BIK1 (Liu et al., 2013). We examined if BIK1, a common downstream component of AtPep1-PEPR and flg22-FLS2 pathways, plays a role in AtPep1-induced stomatal closure. After the detached leaves of bik1 mutants were incubated in ABA or AtPep1 solution, stomatal aperture was reduced in response to ABA but not AtPep1, whereas bik1 mutant expressing BIK1 (35S:BIK1/bik1) became sensitive to AtPep1 (Figure 8A). Consistent with the results from the stomatal assay, ABA, but not AtPep1 treatment, induced activation of in the bik1 guard cells (Figures 8B and 8C). Slow-type nitrate currents in guard cells of the transgenic bik1 plants harboring 35S:BIK1 construct, like the wild-type plants, responded to AtPep1 (Figures 8B and C). These results suggest that BIK1 is an essential component of the AtPep1 signaling pathway, leading to activation of the S-type anion channels and stomatal closure.

Figure 8.

BIK1 Is Required for AtPep1-Induced Stomatal Response.

(A) Stomatal aperture of Col-0, bik1, and 35S:BIK1/bik1 transgenic plant leaves treated with 20 μM ABA or 1 μM AtPep1 for 30 min. Results were shown as means ± sd from three independent experiments (n = 180). Asterisks indicate means that are significantly different from corresponding genotype control at P < 0.01.

(B) Typical whole-cell nitrate current density in guard cells isolated from Col-0, bik1, and 35S:BIK1/bik1 leaves treated with 20 μM ABA or 1 μM AtPep1 solution.

(C) Current density-voltage curves derived from whole-cell anion currents in guard cells isolated from Col-0, bik1, and 35S:BIK1/bik1 leaves treated with or without 1 μM AtPep1 for 30 min. Results were shown as means ± sd from three independent experiments (Col-0, n = 6; Col-0+ABA, n = 5; Col-0+ AtPep1, n = 5; bik1, n = 6; bik1+ABA, n = 7; bik1+AtPep1, n = 7; 35S:BIK1/bik1, n = 6; 35S:BIK1/bik1+ABA, n = 5; 35S:BIK1/bik1+ AtPep1, n = 6).

DISCUSSION

PAMP perception by host PRRs constitutes elaborate signaling networks for plants to detect and defend against diverse invading pathogens and herbivores. On top of this first layer of defense, plant endogenous elicitors such as Peps trigger further signaling mechanisms that contribute to both local and systemic immunity. In our report, we have uncovered a function for AtPeps-PEPR in stomatal immunity. Through activation of two related anion channels, SLAC1 and SLAH3, the AtPeps-PEPR pathway initiates stomatal closure in an OST1-independent manner, representing a unique mechanism for stomatal regulation that has not been found in other signaling pathways.

In Arabidopsis, AtPeps are perceived by PEPR1 and PEPR2 that have been shown to be redundant in defense responses (Yamaguchi et al., 2010). In our study, however, we found differential contribution of PEPR1 and PEPR2 in AtPep1-induced stomatal closure. The pepr1 mutant, like the double mutant pepr1 pepr2, is insensitive to AtPep1-induced stomatal closure, whereas the pepr2 mutant is as sensitive as the wild type in the same assay. Further analysis of anion channel activity, however, reveals contribution of both PEPR1 and PEPR2 in AtPep1 perception. But, importantly, AtPep1-induced activation of guard cell S-type current is reduced more in pepr1 mutant than in pepr2 mutant. The AtPep1-induced MAPK activation was also more affected in the pepr1 mutant than in pepr2 mutant. These results are consistent with substrate saturation analysis showing that PEPR1 displays a higher affinity for AtPep1 than PEPR2 (Yamaguchi et al., 2010). However, in addition to binding affinity difference of PEPR1 and PEPR2, other factors including differential gene expression pattern and protein abundance could also contribute to the differential function of these receptors toward AtPep1-triggered responses.

As targets of the AtPeps-PEPR signaling pathway, two guard cell-expressed S-type anion channels, SLAC1 and SLAH3, are identified to be responsible for the AtPep1-induced stomatal closure. Disruption of both SLAC1 and SLAH3, but not any one of them alone, renders stomata AtPep1-insensitive, which is reminiscent to the result on flg22-induced stomatal closure (Guzel Deger et al., 2015). The S-type anion channels are elaborately regulated by various sets of kinase and phosphatase pairs in guard cells to control stomatal movements. In particular, SnRK2-type protein kinases (e.g., OST1) and calcium-dependent protein kinases (CPKs) have been shown to activate either SLAC1 and/or SLAH3 in X. laevis oocytes and possibly in planta as well (Geiger et al., 2009, 2010, 2011; Lee et al., 2009; Brandt et al., 2012, 2015). The plant PP2C family of protein phosphatases are central negative regulators of ABA signaling (Munemasa et al., 2015). Studies have shown that PP2Cs could directly interact with and dephosphorylate SLAC1 and SLAH3 channels and their regulatory kinases, including OST1 and CPK21 (Umezawa et al., 2009; Vlad et al., 2009; Geiger et al., 2011; Brandt et al., 2012, 2015; Maierhofer et al., 2014).

The protein kinase OST1 is a central regulator for initiating stomatal closure. The function of OST1 was most intensively studied in ABA-induced stomatal response, and it is also essential for stomatal closure induced by other stimuli, such as CO₂ and ozone (Xue et al., 2011; Merilo et al., 2013; Tian et al., 2015). It is noteworthy that several studies have investigated the interplay of PAMP- and abiotic stress-mediated stomatal closure through the OST1 kinase. Upon exposure to PAMPs, such as flg22 and yeast elicitors, the guard cell S-type anion channels are activated to induce stomatal closure, a process that requires OST1 (Melotto et al., 2006; Koers et al., 2011; Guzel Deger et al., 2015; Ye et al., 2015). However, the mechanism of OST1 action in PAMP signaling remains elusive, except for a recent study that identifies BAK1 as an upstream activator of OST1 (Shang et al., 2016). In our study here, stomata of two independent OST1 mutant alleles (ost1-3 and ost1-4), surprisingly, remain responsive to both AtPep1 and AtPep2. Guard cells of these mutants, like those from wild-type plants, retained activation of S-type channel activity after AtPep1 treatment, indicating that OST1 is dispensable in the Peps-PEPR pathway.

Several signaling pathways for stomatal regulation involve ROS accumulation (Song et al., 2014). Upon PAMP (flg22 and elf18) binding, BIK1 interacts physically with and phosphorylates RBOHD, and the BIK1-mediated RBOHD phosphorylation is essential for PAMP-induced stomatal immunity (Kadota et al., 2014; Li et al., 2014). In guard cell ABA response, OST1 acts as upstream activator of ROS production. Studies have shown that OST1 physically interacts not only with RBOHD but also RBOHF, and phosphorylates the N terminus of RBOHF (Sirichandra et al., 2009; Acharya et al., 2013). Both BIK1 and OST1 seem to be upstream regulators of ROS to control stomatal closure. However, a study shows that function loss of RBOHD and RBOHF mutants are still responsive to either ABA or flg22 in stomatal assays (Guzel Deger et al., 2015). In this study, BIK1, but not RBOHs, is required for AtPep1-induced stomatal closure, implying that BIK1, either directly or through an ROS-independent pathway, activates anion channels in response to AtPep1-PEPR. Furthermore, we performed bimolecular fluorescence complementation (BiFC) assay by cotransforming pairs of SLAC1-nVenus, SLAH3-nVenus, and BIK1-cCFP constructs transiently into mesophyll protoplasts. The combination of BIK1-cCFP and SLAC1-nVenus or SLAH3-nVenus but not nVenus produced green fluorescent signals on plasma membrane (Supplemental Figure 7A), implying that SLAC1 and SLAH3 may interact with BIK1 and serve as regulating targets of BIK1. However, little activation of S-type anion channel currents was recorded when phosphomimetic BIK1 (BIK1D) and SLAC1 (or SLAH3) were coexpressed in X. laevis oocyte (Supplemental Figures 7B and 7C). Compared with the activation of SLAC1/SLAH3 by CBL1-CIPK23, activation of the channels by BIK1 is insignificant.

Much remains to be dissected in the future concerning steps from AtPep1 perception to activation of SLAC1 and SLAH3 channels. Based on available data so far, it appears that the flg22 pathway and AtPep1 pathway start to differ downstream of BIK1 at least in stomatal immunity. In the working model (Figure 9), AtPeps are perceived by PEPR receptor complex in the plasma membrane where PEPR1 interacts with BAK1 and BIK1. AtPeps binding activates the receptor complex, leading to activation and release of BIK1. BIK1 may directly or indirectly activate SLAC1 and SLAH3, resulting in efflux of anions and loss of turgor in guard cells. In the same model, ABA-receptor complex represses PP2C activity and thereby activating SnRK2-type kinases (such as OST1) that trigger SLAC1 activation and stomatal closure. The flg22 signal is perceived by its receptor complex consisting of FLS2, BAK1, and BIK1 that transmits the signal to OST1, leading to activation of anion channels and stomatal closure. While ABA and flg22-FLS2 pathways appear to converge at the point of OST1, the Peps-PEPR pathway may take a distinct signaling path to target the same anion channels. It is possible that BIK1 directly interacts with, phosphorylates, and activates the anion channels, which requires extensive further work to test the physical interaction and biochemical processes and their functional relevance. Alternatively, intermediate components downstream of BIK1 may be involved in activation of anion channels. It is widely accepted that such anion channels are activated by both calcium-independent (SnRK2-type kinases such as OST1) and calcium-dependent kinases, such as CDPKs and CBL-CIPK (Geiger et al., 2009, 2010, 2011; Lee et al., 2009; Brandt et al., 2012, 2015; Maierhofer et al., 2014). Because OST1 is not required for Peps-PEPR-mediated anion channel activation, we suspect that calcium-dependent kinases could function as signaling components between BIK1 and anion channels in this pathway. Further work will be directed to identify calcium channels and calcium-dependent protein kinases that may be involved in DAMP-induced signaling.

Figure 9.

The AtPeps-PEPR Signaling Pathway Activates S-Type Anion Channels in Guard Cell.

AtPeps bind to the receptor kinases PEPR1 or PEPR2, which recruit the coreceptor BAK1 and BIK1 to form a receptor complex. The interaction between these receptors activates BIK1 by phosphorylation. BIK1 is released from the receptor complex after activation, and it may directly or indirectly activate the S-type anion channels SLAC1 and SLAH3, leading to anion efflux in guard cells. ABA interacts with its receptor PYR/PYL/RCAR to inhibit PP2Cs activity and activate OST1. The flg22 peptide is perceived by the receptor complex consisting of FLS2, BAK1, and BIK1. The flg22-FLS2 pathway merges with the ABA pathway at the point of OST1. OST1 directly phosphorylates and activates SLAC1-mediated anion efflux.

METHODS

Plant Materials and Growth Conditions

All Arabidopsis thaliana lines used in this study were Columbia (Col-0) ecotype background. Surface-sterilized seeds were plated on 0.5× strength Murashige and Skoog (MS) medium (Caisson; pH 5.8) supplemented with 1% sucrose and 0.7% phytogel (Sigma-Aldrich). The plated seeds were kept at 4°C for 3 d for stratification before incubation in a growth chamber at 22°C, 65 to 80% humidity, and a 16-h-light/8-h-dark regime with white fluorescent tubes (Philips) at a light intensity of 90 mmol m⁻² s⁻¹. One-week-old seedlings were transferred to pots (8.5 × 8.5 cm) containing mixed soil (Sun Gro Horticulture) and grown in a growth chamber at 22°C, 65 to 80% humidity, and a 16-h-light/8-h-dark regime at 90 mmol m⁻² s⁻¹. The mutants pepr1-2 (SALK_014538), pepr2-2 (SALK_004447), pepr1 pepr2 (Ma et al., 2012), bik1, 35S:BIK1/bik1, rbohd rbohf (Li et al., 2014), ost1-3 (SALK_008068), and ost1-4 (GK-516B05) (Ding et al., 2015) were described in previous studies. The mutants slac1-5 (SALK_099145), slah3-4 (SALK_111623), proSLAC1:GUS, and proSLAH3:GUS transgenic plants were screened in our previous study (Zheng et al., 2015). The slac1 slah3 double mutants were obtained by genetic crosses of slac1-5 and slah3-4, while ost1 slah3 double mutants were obtained by crossing ost1-3 and slah3-4. Homozygous mutant plants were screened and corresponding genes were checked by RT-PCR using specific primers (Supplemental Table 1).

In the root growth inhibition assay, surface-sterilized seeds were plated on 0.5× MS medium solidified by 0.7% phytogel (Sigma-Aldrich). Plates were kept at 4°C for 2 d and put vertically in growth chamber with a 16-h-light/8-h-dark regime at 22°C. Three-day-old seedlings were transferred to 0.5× MS solid medium supplemented with or without 0.1 μM AtPep1 as indicated. Photographs were taken after plants were grown for another 6 d.

Peptide Synthesis

Peptides used in this study were synthesized by GL Biochem. Their sequences are shown from N terminus to C terminus as follows: AtPep1, ATKVKAKQRGKEKVSSGRPGQHN; AtPep2, DNKAKSKKRDKEKPSSGRPGQTNSVPNAAIQVYKED; and flg22, QRLSTGSRINSAKDDAAGLQIA.

RT-PCR

Total RNA was extracted from 2-week-old Col-0 or mutant seedlings grown on 0.5× MS solid medium using TRIzol reagent (Invitrogen) following the manufacturer’s instructions. Two micrograms of total RNA was subjected to reverse transcription reaction using the M-MLV Reverse Transcriptase (Promega). The resulting cDNA was used for PCR amplification to clone related genes or verify absence of transcript in mutants with the gene-specific primers (Supplemental Table 1) on a T100 thermal cycler (Bio-Rad). PCR products were separated on a 1% (w/v) agarose gel stained with ethidium bromide.

Stomatal Aperture Measurements

Fully expanded rosette leaves from at least four different 4-week-old plants of each genotype were adopted for stomatal aperture measurement as previously described with modification (Vahisalu et al., 2008). Generally, leaves were detached and floated on stomatal opening buffer (SOS buffer: 10 mM MES-KOH, pH 6.15, 10 mM KCl, and 0.1 mM CaCl₂) under light condition (120 mmol m⁻² s⁻¹) at least for 2 h to induce stomatal opening. Then, leaves were treated with AtPeps, ABA, or flg22 for the indicated period of time. The treated leaves were dried on a filter paper immediately and their abaxial side was covered by clear sealing tape. The leaves were peeled away manually by tweezers. The stomatal epidermis stuck to the tape was placed on a microscope slide for examination and measurements. To avoid bias results, blind experiments were performed so that experimenters in charge of stomatal measurements were not informed of the genotypes or treatments. The stomatal apertures were recorded by a digital camera attached to a light microscope (Olympus). Width and height of stomatal aperture were measured using Image J software (Rawak Software).

Generation of Transgenic Plants

For generation of proPEPR1:GUS and proPEPR2:GUS constructs, 1500- and 1414-bp genomic sequences upstream of PEPR1 and PEPR2 starting codon, respectively, were PCR amplified from wild-type plant genomic DNA and cloned into pCAMBIA1300+GUS. For genetic complementation, full-length genomic DNA of PEPR1 was amplified and subcloned into pCAMBIA1300 vector. The plasmids were transformed into Agrobacterium tumefaciens strain GV3101 followed by transformation of plants using the floral-dip method (Clough and Bent, 1998). Seedlings were selected on 0.5× MS solid medium containing 25 μM Hygromycin B, and the T2 transgenic plants were used for further experiments.

Histochemical GUS Activity

GUS staining was performed according to the method described by Jefferson et al. (1987). Two-week-old seedlings and leaves from 4-week-old T2 transgenic plants were vacuum-infiltrated for 20 min in the staining solution [1 mM X-Gluc (5-bromo-4-chloro-3-indolyl-β-d-glucuronide), 0.5 mM K₄Fe(CN)₆, 0.5 mM K₃Fe(CN)₆, 1% Triton X-100, 10 mM EDTA, and 100 mM NaPO₄, pH 7.0] followed by incubation at 37°C for 3 h. Pictures were taken with a digital camera attached to a light microscope (Olympus).

Detection of ROS

To detect ROS in situ, DAB (Sigma-Aldrich) staining was performed on Arabidopsis leaves as described by Daudi et al. (2012). Specifically, leaves of 4-week-old plants were sprayed with 1 μM AtPep1. Six plants were used as replicates, and three rosette leaves from each plant were sampled. Thirty minutes after spraying, leaves were sampled and immersed in 1 mg/mL DAB solution supplemented with 0.05% (v/v) Tween 20 and 10 mM Na₂HPO₄. Samples were placed on a shaker at 80 rpm and incubated for 5 h at room temperature. Following the incubation, DAB staining solution was replaced with bleaching solution (ethanol:acetic acid:glycerol = 3:1:1). Samples were kept in a water bath at 95°C for 15 min. Leaves were reimmersed in fresh bleaching solution and then photographed on a white background.

ROS detection in guard cells was performed using H₂DCF-DA (Sigma-Aldrich) as described (Murata et al., 2001). Briefly, leaf epidermis strips were prepared from 4-week-old plants and incubated for 3 h in a buffer solution containing 30 mM KCl and 10 mM MES-KOH, pH 6.15. Then, 30 μM H₂DCF-DA was added to incubation buffer. After 20 min of incubation, the dye was removed by three washes with distilled water. The epidermis strips were then incubated with or without 1 μM AtPep1 for 30 min. Guard cell fluorescence was detected by a confocal microscopy (Zeiss 710) operated with LSM Image Browser software using an excitation wavelength of 488 nm. Fluorescence intensities of guard cells were quantified as average pixel intensities using Image J (Rawak Software).

Bacterial Growth Assay

Bacterial growth assays were performed as described previously (Zipfel et al., 2004). Pseudomonas syringae pv tomato DC3000 was cultured in liquid Luria-Bertani (LB) medium containing 100 mg/L rifampicin at 28°C to an OD₆₀₀ of 1.0. Bacteria were collected and resuspended in 10 mM MgCl₂. For spray inoculation, 4-week-old plants were sprayed with Pst DC3000 bacterial suspension containing 5 × 10⁸ colony-forming units per mL (cfu/mL) with 0.02% Silwet L-77. For syringe inoculation, Pst DC3000 suspension at 1 × 10⁵ cfu/mL was pressure-infiltrated into three leaves per plants with a needleless syringe. After bacterial treatments, plants were covered with a clear plastic lid to keep high humidity and grown in a growth chamber. Three days after inoculation, leaves were surface sterilized in 70% ethanol and rinsed in sterile water. One leaf disc was excised using a cork borer from each infected leaf. One replicate consists of three leaf disks and 6 replicates were included in each experiment. Leaf discs were grounded in 10 mM MgCl₂. Bacterial quantification was performed by plating serial dilutions on LB agar medium containing 100 mg/L rifampicin.

Immunoblotting

Rosette leaves of 4-week-old plants were sprayed with 1 μM AtPep1 or AtPep2 solution for 30 min, and leaves were harvested and frozen in liquid nitrogen. Ground leaves were homogenized at least for 1 h in protein extraction buffer containing 50 mM HEPES-KOH (pH 7.5), 150 mM KCl, 1 mM EDTA, 0.2% (v/v) Triton X-100, 1 mM DTT, 0.1 mg/mL protease inhibitors (Roche), and 0.1 mg/mL phosphatase inhibitors (Roche), and centrifuged at 15,000 rpm for 20 min. Supernatants containing total protein extracts, were separated on 10% SDS-PAGE and transferred to PVDF membranes. Immunoblot analysis was conducted using anti-phospho-p44/42 MAPK (Erk1/2) (Thr-202/Tyr-204) monoclonal antibody (Cell Signaling) as the primary antibody at 1:3000 dilution in 5% nonfat milk TBST buffer and horseradish peroxidase-conjugated anti-rabbit antibody (Sigma-Aldrich) as the secondary antibody at 1:10,000 dilution in the same buffer. The immunoblot signal was captured with film using Super Signal West Pico Chemiluminescent Substrate (Thermo Scientific). Ponceau S staining of blots was performed for protein visualization as loading control.

Patch-Clamp Whole-Cell Recording of Guard Cell Protoplast

Rosette leaves of 4-week-old plants were used to isolate guard cell protoplasts. Epidermal strips were manually peeled from the leaves using a pair of tweezers and were enzymatically digested for 30 min at 28°C in a shaker with a shaking speed of 110 rpm. The enzyme solution contained 1.4% (w/v) cellulase Onozuka-RS (Serva), 0.4% (w/v) macerozyme Onozuka-R10 (Serva), 0.1% (w/v) BSA (Sigma-Aldrich), 0.1 mM CaCl₂, 0.1 mM KCl, and 5 mM ascorbic acid and adjusted to pH 5.6/MES and to an osmolality of 400 mOsmol⋅kg⁻¹ with d-mannitol. The suspension was then filtered through a 30-μm nylon mesh, washed with wash solution (400 mM d-sorbitol and 1 mM CaCl₂) and centrifuged at 100g for 10 min. The enriched protoplasts were stored on ice until aliquots were used for whole-cell patch clamp recordings of S-type anion currents.

Whole-cell S-type anion currents were recorded by applying the patch-clamp procedure similar to those described previously (Vahisalu et al., 2008; Geiger et al., 2011; Tian et al., 2015), using the Axon Multiclamp 700B Amplifier (Molecular Devices). Patch pipettes were pulled from borosilicate glass (Sutter Instrument) on a pipette puller (P-97; Sutter Instrument) and fire-polished to a tip resistance of 25 MΩ in NO₃⁻-based solutions. Only those patches with seal resistance higher than 5 GΩ were used to perform whole-cell recordings. To equilibrate the cytoplasm with pipette solution, we waited at least 15 min after the establishment of the whole-cell configuration before recording was initiated. Digital low-pass Bessel filtering of currents elicited by positive voltage pulses was performed at a cutoff frequency of 2.9 kHz. Data were analyzed with the software Clampex and Clampfit (version 10.3). For analyzing anion currents in nitrate-based media, the pipette solution contained 100 mM NaNO₃, 1 mM MgCl₂, 5 mM Mg-ATP, 5 mM Tris-GTP, 6.7 mM EGTA, 5.864 mM CaCl₂ (free Ca²⁺ was 2 μM), and 10 mM HEPES-Tris to pH 7.1; the bath solution comprised 30 mM NaNO₃, 2 mM MgCl₂, 1 mM CaCl₂, 1 mM LaCl₃, and 10 mM MES adjusted with Tris to pH 5.6. The osmolality of solutions were adjusted to 400 mOsmol kg⁻¹ for the bath solution and 500 mOsmol.kg⁻¹ for the pipette solution, by supplementing d-sorbitol. The membrane voltage was stepped from +35 to –145 mV with 30-mV decrements, and the holding potential was at 0 mV. Raw currents were normalized into current densities (pA/pF) by taking the capacitance of each cell into consideration. Current-voltage curves were obtained by plotting current densities against the applied test voltages.

BiFC Assay

For the BiFC assay, the SLAC1-nVenus, SLAH3-nVenus, and BIK1-cCFP were cloned into the vectors pE3308 and pE3449, respectively, using primers listed in Supplemental Table 1. Well-expanded leaves from 4-week-old Arabidopsis plants were used for protoplast isolation. Isolation and transient protoplast transformation was performed as described previously (Abel and Theologis, 1994). Briefly, leaves were sliced on a filter paper and immediately transferred into enzyme solution (1% [w/v] cellulase Onozuka-RS [Serva], 0.4% [w/v] macerozyme Onozuka-R10 [Serva], 0.1% [w/v] BSA, 10 mM CaCl₂, 20 mM KCl, 20 mM MES/KOH, pH 5.6), the digestion was continued for ∼3 h with gentle shaking in the dark. After incubation, the enzyme solution containing protoplasts was filtered, then washing solution was added and protoplasts were collected by centrifugation and resuspended in MMG solution. The indicated pairs of expression constructs were cotransformed into protoplasts by PEG-mediated transfection. Protoplast images were taken by a confocal microscopy (Zeiss 710) operated with LSM Image Browser software using an excitation wavelength of 488 nm.

Oocyte Recording

The cDNAs of SLAH3 and BIK1 were cloned into the pGEMHE vector. The phosphomimetic BIK1D mutant was produced using PCR-based mutagenesis (Supplemental Table 1). The SLAC1, CBL1, and CIPK23 constructs for expression in Xenopus laevis oocytes were described in previous studies (Li et al., 2006; Lee et al., 2009). The capped RNA (cRNA) was prepared with the mMessage mMachine in vitro transcription Kit (Ambion) according to the manufacturer’s protocol. The quality was checked by agarose gel electrophoresis, and the concentration was determined by A260/A280 and adjusted to final concentration of 1 μg⋅μL⁻¹. A total 11.5 ng of each tested cRNA, in a total volume of 46 nL, was injected into each oocyte. Injected oocytes were incubated in ND96 at 18°C for 2 d prior to electrophysiological recording. Oocytes were voltage-clamped using a TEV 200 amplifier (Dagan) and monitored by computer through Digidata 1440 A/D converter and pCLAMP 10.2 software (Molecular Devices). The pipette solution contained 3 M KCl. For experiments with SLAC1, the perfusion buffer contain 30 mM CsCl, 1 mM Ca(gluconate)₂, 1 mM Mg(gluconate)₂, 1 mM K(gluconate), and 10 mM Tris/MES (pH 5.6). In experiments with SLAH3, CsCl was replaced by 30 mM CsNO₃. Osmolalities were adjusted to 220 mOsmol⋅kg⁻¹ using d-mannitol. The membrane voltage was stepped to potentials starting at +40 to −140 mV for 7 s with 20-mV decrements and the holding potential was 0 mV. Steady state currents were extracted at 6 s.

Statistical Analysis

For all experiments, three independent repetitions were performed. One-way ANOVA Tukey’s test was used for statistical analysis (Supplemental Table 2).

Accession Numbers

Sequence data from this article can be found in GenBank data library under the following accession numbers: PEPR1, AT1G73080; PEPR2, AT1G17750; SLAC1, AT1G12480; SLAH3, AT5G24030; OST1, AT4G33950; BIK1, AT2G39660; RBOHD, AT5G47910; RBOHF, AT1G64060; MPK3, At3g45640; MPK6, At2g43790; EF1α, At5g60390; CBL1, AT4G17615; and CIPK23, AT1G30270.

Supplemental Data

• Supplemental Figure 1. Construction of PEPR1 complementary transgenic plants in pepr1 pepr2.

• Supplemental Figure 2. Mutation of PEPRs impairs AtPep1 sensitivity during seedling growth assay.

• Supplemental Figure 3. Reversal potentials in guard cell.

• Supplemental Figure 4. Construction of slac1 slah3 double mutant plant.

• Supplemental Figure 5. Construction of ost1 slah3 double mutant plant.

• Supplemental Figure 6. Representative images of AtPep1-induced ROS accumulation in guard cells.

• Supplemental Figure 7. Interaction between BIK1 and SLAC1/SLAH3.

• Supplemental Figure 8. Confirmation of stomatal aperture assay.

• Supplemental Table 1. Primers used in this study.

• Supplemental Table 2. ANOVA table for statistical analysis.