Fungal toxin fusicoccin enhances plant growth by upregulating 14-3-3 interaction with plasma membrane H⁺-ATPase

Abstract

Fusicoccin-A (FC-A) is a diterpene glucoside produced by a pathogenic fungus. Since its discovery, FC-A has been widely recognized as a phytotoxin that induces stomatal opening and leaf wilting, eventually leading to plant death. In this study, we present the first evidence that FC-A enhances plant growth by stabilizing the protein-protein interaction between plasma membrane (PM) H⁺-ATPase and 14-3-3 in guard cells. Long-term treatment of Arabidopsis plants with FC-A resulted in ~ 30% growth enhancement. Structurally similar fusicoccin-J (FC-J) showed a similar degree of growth-promotion activity as FC-A, whereas the more hydrophilic fusicoccin-H (FC-H) exhibited no effect on plant growth, indicating that the enhancement of plant growth observed with FC-A and FC-J involves upregulation of the protein-protein interaction between PM H⁺-ATPase and 14-3-3 in guard cells, which promotes stomatal opening and photosynthesis.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-73979-4.

Introduction

Methods to increase the efficiency of food crop production are urgently needed in order to address the unprecedented demand placed on global food supplies brought about by continued population growth¹. In plant leaves, the stoma is a pore in the epidermis surrounded by a pair of guard cells that alternately swell and contract to regulate stomatal opening and closure. Stomatal opening facilitates the diffusion of CO₂ and O₂ gases as well as water vapor, which is essential for photosynthesis and transpiration². Chemical approaches for manipulation of stomatal opening may therefore hold tremendous promise for enhancing photosynthesis and plant growth, thereby increasing biomass and reducing CO₂ production. Intensive research has focused on identifying compounds that induce phenotypic changes in stomata³, stomatal closure⁴ and opening⁵, alteration of stomatal density^6,7, and perturbation of asymmetric division⁸. Nevertheless, the impact of these compounds on plant growth remains to be elucidated. It would be advantageous for agricultural applications if chemical agents that possess stomatal-opening activity and are readily available in large quantities could be identified.

Fusicoccin-A (FC-A, Fig. 1A) is a 5-8-5 tricyclic diterpene glucoside produced by the plant pathogenic fungus Phomopsis amygdali (syn. Fusicoccum amygdali)⁹. This fungal metabolite is known for its phytotoxicity, which promotes continuous stomatal opening resulting in over-transpiration, wilting of leaves, and eventual death of peach and almond trees^10–12. In the proposed molecular mechanism of action, FC-A stabilizes the protein-protein interaction (PPI) between plasma membrane (PM) H⁺-ATPase and 14-3-3 by forming a stable ternary complex in which FC-A binds to a hydrophobic cavity of the binary complex formed by the binding of 14-3-3 to the phosphorylated C-terminus of PM H⁺-ATPase¹³ (Fig. S1A). Appropriate cycling between association and dissociation of PM H⁺-ATPase and 14-3-3 in guard cells is critical for the regulation of stomatal movement in response to various signals, such as light^14,15. PM H⁺-ATPase consists of 10 transmembrane helices, with the N- and C-termini located on the cytoplasmic side of the plasma membrane¹⁶. The C-terminus of PM H⁺-ATPase possesses a 14-3-3-binding motif and acts as an autoinhibitory domain¹⁷. Photoexcitation of blue-light receptor phototropins triggers phosphorylation of the penultimate threonine (Thr) residue, and subsequent binding of 14-3-3 to the phosphorylated motif dissociates the inhibitory domain, thereby activating the enzyme^17–19. Activated PM H⁺-ATPase then releases protons and generates an inside-negative electrical potential across the plasma membrane, which in turn drives the accumulation of K⁺ through PM inward-rectifying K⁺ (K⁺ _in)-channels²⁰, resulting in swelling of guard cells and subsequent stomatal opening. As a consequence of FC-A-mediated stabilization of the PPI, PM H⁺-ATPase becomes overactivated, thus promoting abnormal stomatal opening^21–25. Note that AHA1 and AHA2 (Arabidopsis PM H⁺-ATPase 1 and 2) are known to be major PM H⁺-ATPase isoforms expressed in guard cells of Arabidopsis thaliana²⁶, and that a knockout mutant of AHA1 showed reduced light-induced stomatal opening but a knockout mutant of AHA2 showed normal light-induced stomatal opening²⁷.

Fig. 1.

Fusicoccin promotes stomatal opening in Arabidopsis thaliana and stabilizes the interaction of 14-3-3 with AHA2. (A) Chemical structures of fusicoccin A (FC-A), fusicoccin J (FC-J), and fusicoccin H (FC-H). (B) Analysis of stomatal opening promoted by FCs. Left: DIC images of guard cells in Arabidopsis cotyledons in the dark at 6 h after treatment with each compound; scale bars represent 5 μm. Right: Change in stomatal aperture (µm) over time in Arabidopsis cotyledons treated with each compound in the dark. Error bars represent SEM (n ≥ 30). The significance of differences in stomatal aperture was analyzed using one-way ANOVA Dunnett’s test (**p < 0.01, ***p < 0.001). (C) Detection of H⁺-ATPase and 14-3-3 in immunoprecipitated samples. Left: Protein samples obtained from chemically treated Arabidopsis rosette leaves (25–30 days old) were treated with Dynabeads Protein A conjugated to a polyclonal antibody against PM H⁺-ATPase or treated with pre-immune serum (Pre-)⁶⁹. Bound proteins were separated by SDS-PAGE and detected by immunoblotting using anti-PM H⁺-ATPase or anti–14-3-3 antibodies. Right: Relative intensity of 14-3-3/PM H⁺-ATPase bands ratio to control. Data were obtained from three independent experiments and averaged. Error bars represent SEM. The significance of differences in stomatal aperture was analyzed using one-way ANOVA Dunnett’s test (*p < 0.05, ***p < 0.001). Original blots are presented in Supplementary Fig. S12. (D) Detection of phosphorylated H⁺-ATPase in plants. Protein samples described in (C) were separated by SDS-PAGE and detected by immunoblotting using anti–pThr947 AHA2 and anti-PM H⁺-ATPase antibodies. Data are the means of three independent experiments. Error bars represent SEM. The significance of differences in stomatal aperture was analyzed using one-way ANOVA Dunnett’s test (*p < 0.05, **p < 0.01). Original blots are presented in Supplementary Fig. S12. (E) Fluorescence polarization titration of recombinant Arabidopsis thaliana His-14-3-3φ (0–1 µM) against FAM-labeled phosphopeptide derived from C-terminal AHA2 (FAM-TPSHYpTV, 10 nM) in the presence of each compound (4 µM). Excitation and emission wavelengths were 470 and 525 nm, respectively. Error bars represent SEM (n = 3).

The 14-3-3 proteins are a family of dimeric phospho-binding proteins that play central roles in the regulation of a diverse array of physiological processes mediated by serine/threonine kinases and phosphatases^28,29. Each 14-3-3 monomer possesses an amphiphilic binding groove ~ 25 Å in length that recognizes a phospho-consensus motif containing either phosphoserine or phosphothreonine³⁰. The 14-3-3 proteins are highly conserved and expressed in all eukaryotic cells, and hundreds of phosphorylated binding partners have been identified in plants³¹ and humans (Homo sapiens)²⁹. A study examining the crystal structure of the ternary complex (Fig. S1A) of tobacco (Nicotiana tabacum) 14-3-3c bound to FC-A and the PM H⁺-ATPase C-terminal phosphopeptide Gln-Ser-Tyr-pThr-Val (QSYpTV) demonstrated that FC-A binds in the hydrophobic cavity adjacent to the peptide, resulting in an increase in the affinity of both ligands for 14-3-3 of almost 2 orders of magnitude¹³. This cooperative binding to 14-3-3 is driven by hydrophobic interactions between the 5-8-5 fusicoccin skeleton and the isopropyl side chain of the C-terminal Val residue of the phospholigand, engaging significant van der Waals contacts. Analyses of the crystal structure revealed that FC-A forms a number of intermolecular interactions with the phospholigand and 14-3-3 (Fig. S1B): (1) the 12-hydroxy group forms a hydrogen bond with the C-terminal carboxylate of QSYpTV via a water molecule; (2) the 16-methoxy group varies in a narrow hydrophobic pocket composed of Phe126 and Met130; and (3) the 6’-pentenyl group binds to a hydrophobic surface region consisting of Leu50, Glu19, and Val45. This unique mode of PPI stabilization has prompted many researchers, including our group, to utilize FC-A not only for plant physiological studies but also for the development of new pharmaceuticals to treat various human diseases^32–34.

In 2014, Wang et al. reported that overexpression of PM H⁺-ATPase specifically in guard cells of Arabidopsis thaliana promotes light-induced stomatal opening and photosynthesis³⁵, resulting in significant enhancement of growth of the transgenic plants³⁶. In contrast, overexpression of phototropin or K⁺ _in-channels in guard cells had no effect on growth, indicating that the number of PM H⁺-ATPase molecules (i.e., the enzyme activity in guard cells) determines the efficiency of photosynthesis. Importantly, the stomata of the transgenic plants were shown to close normally in response to darkness and abscisic acid, which is essential for normal plant growth. These results prompted us to address the following question: If plants are supplied with sufficient light and water to trigger photosynthesis and avoid drought, would treatment of plants with phytotoxic FC-A, which stabilizes the PPI between PM H⁺-ATPase and 14-3-3 and thus induces stomatal opening, in turn enhance plant growth? We tested this paradoxical hypothesis by leveraging structure-activity-relationship studies using FC-A and its biosynthetic intermediates fusicoccin-J (FC-J)^37,38 and fusicoccin-H (FC-H)³⁹ (Fig. 1A). Similar to FC-A, FC-J retains critical functionalities at the 12, 16, and 6’ positions that are required for stable complex formation, whereas FC-H lacks all three functional groups. We therefore anticipated that FC-J would stabilize the PM H⁺-ATPase-14-3-3 interaction, thereby resulting in a biological effect similar to that of FC-A, whereas FC-H would be less effective because it is potentially a weaker stabilizer. Here, we report that the phytotoxin FC-A and its biosynthetic intermediate FC-J (but not FC-H) enhance plant growth by upregulating the interaction between 14-3-3 and PM H⁺-ATPase to induce stomatal opening and promote photosynthesis.

Results

Promotion of stomatal opening by fusicoccins

First, we evaluated chemical induction of stomatal opening. This was conducted by directly spraying the compound on the surface of A. thaliana leaves in light prior to harvesting cotyledons at various time intervals for measurement (Fig. 1B). As previously reported^25,40, FC-A significantly promoted stomatal opening, with openings ~ 3 times larger than those of control plants at 6 h post-treatment (Fig. 1B). Similarly, FC-J was also effective, but FC-H was apparently less potent than FC-A, indicating that the stomatal opening effect varies depending on the FC chemical structure. Infrared thermal images of Arabidopsis indicated a decrease in the temperature of the leaves after a droplet of FC-A was placed on the surface (Fig. S2), further supporting the apparent promotion of stomatal opening by FC-A in whole plants.

Stabilizing effect of fusicoccins on interaction between 14-3-3 and PM H⁺-ATPase

Next, we examined whether FCs stabilize the interaction between PM H⁺-ATPase and 14-3-3 in plants. Arabidopsis leaves were treated with each compound, and protein samples were then extracted and subjected to co-immunoprecipitation using an anti-PM H⁺-ATPase antibody; the relative amount of bound 14-3-3 was analyzed by Western blotting (Fig. 1C). The intensity of bands corresponding to 14-3-3 increased significantly in the presence of FC-A, FC-J, and FC-H, respectively, compared with the control, indicating that each FC stabilizes the complex formed by the binding of PM H⁺-ATPase to 14-3-3 (Fig. 1C). Notably, FC-H was less effective than FC-A, which was consistent with their observed effect in promoting stomatal opening. In addition, the level of PM H⁺-ATPase phosphorylation increased markedly in the presence of FC-A and FC-J, but not in the presence of FC-H (Fig. 1D), indicating that PM H⁺-ATPase activity is upregulated by stabilization of the interaction with 14-3-3, in which FC-H exhibited only a weak effect.

In order to quantitatively evaluate the effect of FCs in stabilizing the PM H⁺-ATPase-14-3-3 interaction, fluorescence polarization titration experiments in the presence/absence of FCs were performed using recombinant Arabidopsis 14-3-3φ and a fluorescein (FAM)-labeled phosphopeptide (FAM-TPSHYpTV) derived from the C-terminus of AHA2 (Fig. 1E). Significant enhancement of the affinity of the peptide for 14-3-3 (K_d = 0.89 ± 0.19 µM) was observed in the presence of FC-A (K_d = 0.15 ± 0.03 µM) and FC-J (K_d = 0.16 ± 0.03 µM), whereas only weak affinity was observed in the presence of FC-H (K_d = 0.53 ± 0.89 µM). These results clearly show that the degree of upregulation of stomatal opening is correlated with the degree of stabilization of the PM H⁺-ATPase–14-3-3 PPI.

Transgenic Arabidopsis plants expressing superfolder GFP–AHA2 were prepared and subjected to fluorescent imaging analysis using BODIPY-TR reagents (Fig. S3). When epidermal tissues were treated with a simple fluorescent BODIPY-TR derivative (1, Scheme S1), clear localization of AHA2 at the PM in guard cells was observed (Fig. S3, lane 2), whereas 1 was distributed in the cytosol and formed fluorescent puncta (Fig. S3, lane 3). This result was presumably due to the preferential accumulation of BODIPY dye in peroxisomes, as previously reported⁴¹. On the other hand, the FC-attached BODIPY-TR derivative (2, Scheme S1) showed a different cytosolic distribution than 1, forming much fewer puncta and localizing near the PM. These results support the PM localization of FC, which allows for interaction with PM H⁺-ATPase in guard cells.

Effect of fusicoccin-A on promotion of photosynthesis

We next examined whether FC-A improves the efficiency of photosynthesis. Intact leaves of four- to six-week-old Arabidopsis plants were sprayed twice with FC-A (30 µM) at 3 h and 2 h prior to irradiation with white light (600–700 µmol m⁻² s⁻¹), and changes in the CO₂ assimilation rate and stomatal conductance were monitored over time using a gas-exchange system LI-6400XT (LI-COR). Both the CO₂ assimilation rate (Fig. 2A) and stomatal conductance (Fig. 2B) increased significantly in FC-A–treated plants compared with control plants, especially at the beginning of irradiation (i.e. 5–10 min, Fig. 2C and D), indicating that FC-A enhances photosynthetic activity by reducing stomatal limitation during the induction.

Fig. 2.

Gas-exchange properties of Arabidopsis plants treated with FC-A. (A) CO₂ assimilation and (B) stomatal conductance responses to light. Arabidopsis rosette leaves were treated twice with FC-A (500 µL of 30 µM/treatment) or 0.1% EtOH, first at 3 h and second at 2 h prior to light irradiation, followed by incubation at room temperature in the dark. Measurements were carried out using an LI-6400 gas exchange system at 400 ppm CO₂; the leaf temperature and leaf chamber relative humidity were maintained at 24 °C and 40–50% (Pa/Pa), respectively. The arrow in each panel indicates the start of light irradiation (600–700 µmol m⁻² s⁻¹), and lines indicate the mean ± SD of 3 independent experiments. (C) The CO₂ assimilation rate and (D) stomatal conductance at indicated time points were extracted from the curve in (A, B), then the differences between EtOH and FC-A treatment at each time point were analyzed using one-way repeated ANOVA, followed by pairwise t test with Bonferroni adjustment (*p < 0.05; **p < 0.01).

Effect of fusicoccins on plant growth

Based on results showing that treating plants with FC-A promotes photosynthesis, we examined the crucial question of whether a long-term FC-A treatment enhances plant growth. Arabidopsis plants were sprayed with various concentrations of FC-A (0.3–30 µM) once daily for 15 days and cultivated under conditions providing a sufficient supply of water and white light (80–100 µmol m⁻² s⁻¹). The plants were then harvested and analyzed. Strikingly, FC-A treatment resulted in an apparent enhancement of plant growth by 15 days compared with control plants (Fig. 3A). Plants treated with FC-A at 3 to 30 µM exhibited a significant enhancement in the size of rosette leaves, and the fresh and dry weight of plants treated with 30 µM FC-A was ~ 30% greater than that of control plants (Fig. 3B, p < 0.05). The observed growth enhancement was dose dependent (Fig. 3B and Fig. S4). Importantly, FC-J also promoted plant growth, but FC-H did not (Fig. 3C, Fig. S5). These results were consistent with the ability of the compounds to activate PM H⁺-ATPase and induce stomatal opening.

Fig. 3.

Fusicoccin enhances plant growth. (A) Arabidopsis plants were cultivated after treatment with various amounts of FC-A. Ten-day-old plants were sprayed with a chemical solution containing 0.3–30 µM compound with 0.1% EtOH (1 to 3 mL/12 plants) once daily. Plants were then cultivated under high-light conditions (200 µmol m⁻² s⁻¹) with sufficient water supply and harvested and measured at the age of 25 days. (B) Left: Rosette and juvenile leaves of 25-day-old Arabidopsis plants treated with FC-A at various concentrations. Scale bars represent 5 cm. Right: Relative aboveground fresh and dry weights of 25-day-old plants. Error bars indicate SEM (n = 12 plants). Mean values were obtained from 3 independent experiments. Differences were analyzed using one-way ANOVA Dunnett’s test (**p < 0.01, ***p < 0.001). (C) Left: Rosette and juvenile leaves of 25-day-old Arabidopsis plants after treatment with FC-A, FC-J, or FC-H (30 µM, 1 to 3 mL/12 plants) for 14 days. Relative aboveground fresh and dry weights of 25-day-old plants. Error bars indicate SEM (n = 12). Differences were analyzed using one-way ANOVA Dunnett’s test (***p < 0.001). (D) Carbon isotope ratio (δ¹³C) of 25-day-old Arabidopsis plants treated with 0.1% EtOH (control) or FC-A (30 µM containing 0.1% EtOH). Error bars represent SEM (n = 5–6 plants). Differences were analyzed using Student’s t test (***p < 0.001).

To determine whether the increase in biomass was truly associated with the promotion of stomatal opening and upregulation of photosynthesis triggered by FC-A, the δ¹³C value of Arabidopsis plants was determined after FC-A treatment and cultivation. Higher CO₂ assimilation efficiency is known to result in a decline in the ¹³C/¹²C (δ¹³C) accumulation ratio because plants discriminate against¹³C in favor of ¹²C during photosynthesis⁴². Indeed, the chemically treated plants showed significantly lower δ¹³C values than nontreated controls (Fig. 3D), further confirming that FC-A-induced chemical activation of PM H⁺-ATPase in guard cells enhances plant growth, leading to increased biomass.

Changes in the effect of fusicoccin-A over time

A general concern with treatments to chemically control stomatal movement is whether the agents are degraded or inactivated in plants over time, such that the plants recover normal stomatal movement after a certain period of time post-treatment. A previous study showed that the constitutive activation of PM H⁺-ATPase in Arabidopsis, induced by overexpression of FLOWERING LOCUS T (FT) in the guard cells⁴³, results in a constant opening of the stomata, but not an enhancement of growth³⁵. This suggests that stomatal closing at night is crucial when attempting to enhance plant growth by inducing stomatal opening. In order to determine whether the promotion of stomatal opening by FC-A is transient, leaves of Arabidopsis plants were treated with various amounts of FC-A, grown under normal conditions, and then detached from the plants. The change in weight of each detached leaf due to the upregulated transpiration caused by FC-A was then monitored and compared with the rate of weight loss of leaves from nontreated control plants. We anticipated that the rate of weight loss would decline if FC-A was degraded or inactivated in the leaf. When plants were incubated with FC-A for 12 h prior to detachment of leaves, the weight of chemically treated leaves declined significantly faster than that of control leaves in a dose-dependent manner (Fig. 4A and Fig. S6). However, the rate of weight loss of leaves from plants treated with FC-A for a longer period (24 h) declined, with no apparent difference compared with the control detectable by 48 h. These results indicate that FC-A remains effective and promotes transpiration on leaves for 12 to 24 h but becomes ineffective by 48 h under the conditions examined in this study. Consistent with these results, time-course analyses indicated an apparent increase in the adaxial stomatal aperture (not in the abaxial stomatal aperture) of Arabidopsis rosette leaves by 12 h after FC-A treatment, and the significant difference disappeared by 48 h (Fig. 4B and Fig. S7). Moreover, the results of the immunostaining analysis of PM H⁺-ATPase in guard cells indicated that the promotion of phosphorylation level at Thr947 by FC-A observed at 0 h declined to the control level after 48 h (Fig. 4C). These results indicate that the stomatal opening induced by FC-A is transient, presumably due to the inactivation of FC-A in the plant leaves.

Fig. 4.

Effect of FC-A decay. (A) The rate of weight loss increased in a dose-dependent manner after incubation with FC-A. Left: Rosette leaves of Arabidopsis plants were treated with various amounts of FC-A (0, 4.8, 48, and 480 pmol/leaf), incubated in the dark for 1 h, cultivated under normal conditions for 12, 24, or 48 h, and then detached from the plant. Images were recorded at 0, 12, and 24 h after detachment. Leaves were harvested from 5 individual plants (#1–#5). Scale bars represent 2 cm. Right: Relative change in weight of the rosette leaves at 0, 12, and 24 h after detachment. The difference in the rate of weight loss was no longer apparent after 48 h of incubation with FC-A. Error bars represent SEM (n = 5 plants). Differences were analyzed using one-way ANOVA Dunnett’s test (*p < 0.05, ***p < 0.001). (B) Change in FC-A-induced stomatal opening in the epidermis over time. Rosette leaves of Arabidopsis plants were treated with FC-A (48 pmol/leaf) in a similar manner to that described in (A). Epidermal peels were obtained and subjected to microscopic analysis. Error bars represent SEM (n ≥ 30). Differences were analyzed using one-way ANOVA Dunnett’s test (**p < 0.01, ***p < 0.001). (C) Amount of PM H⁺-ATPase (left) and phosphorylated PM H⁺-ATPase (right) in guard cells after 48 h of incubation with 0.1% EtOH (control, black bars) or FC-A (150 pmol/leaf, red bars). Rosette leaves of Arabidopsis plants were treated with FC-A for 1 h in the dark and harvested immediately (0 h) or cultivated under normal conditions for 48 h (48 h). Epidermal peels were isolated, fixed, and PM H⁺-ATPase or phosphorylated PM H⁺-ATPase were visualized and quantified by immunohistochemical staining using anti-H⁺-ATPase or anti-pThr antibodies, respectively⁷⁴. Differences were analyzed using Tukey-Kramer test (***p < 0.001).

Effect on growth of Japanese mustard spinach plants, Komatsuna

Finally, the potential of FC-A to enhance the growth of another dicot was investigated. Japanese mustard spinach plants (Komatsuna; Brassica rapa var. perviridis, a member of the same family as Arabidopsis) were treated with FC-A once daily and cultivated in a hydroponic system in an outfield greenhouse for 28 days prior to harvesting and analysis. Treatment with FC-A resulted in a notable enhancement of plant growth, and the biomass of both fresh and dry plants increased significantly in a concentration-dependent manner up to 48 pmol plant⁻¹ day⁻¹ (Fig. 5 and Fig. S8A), indicating that FC-A also promotes the growth of Komatsuna plants. It should be noted that the size and number of leaves decreased at a dose of 480 pmol plant⁻¹ day⁻¹ (Fig. 5), and partially wilted leaves were observed (Fig. S8B). These results suggest that FC-A is toxic to Komatsuna plants at high doses.

Fig. 5.

FC-A enhances growth of Komatsuna plants. Komatsuna plants (Japanese mustard spinach; Brassica rapa var. perviridis) were grown in a greenhouse using a hydroponic system for 10 days prior to chemical treatment. On day 11, two droplets of FC-A solution (8 µL each, 0.3–30 µM; 4.8–480 pmol plant⁻¹ day⁻¹) or 0.1% EtOH solution containing 0.05% (v/v) of a spreading agent (Approach^® BI) were applied to 2 randomly selected rosette leaves once daily for 28 days. Left: Images of rosette and juvenile leaves of a representative plant from each group. Right: Relative aboveground fresh and dry weights of 38-day-old plants. Error bars indicate SEM (n = 18 plants). Differences were analyzed using one-way ANOVA Dunnett’s test (*p < 0.05, **p < 0.01, ***p < 0.001).

Discussion

In the 60 years since its discovery, the fungal toxin FC-A has been exclusively studied as a potent phytotoxin that promotes continuous stomatal opening and over-transpiration, thereby eventually leading to plant death. Due to its unique mechanism of action, which relies on the stabilization of PPI between 14-3-3 and PM H⁺-ATPase, FC-A has been extensively utilized as a chemical tool in the field of plant physiology¹². However, there is currently a lack of studies investigating the impact of prolonged FC-A treatment on plant growth. In this study, we present the first evidence that the phytotoxin FC-A promotes plant growth, demonstrating that daily treatment of Arabidopsis plants results in an approximately 30% increase in growth under the conditions of sufficient water and light supply (Fig. 3). The detailed biochemical evaluation of FC-A, which exploits its biosynthetic intermediates FC-J and FC-H, clearly indicates that the observed plant growth effect is due to the stabilization of the PPI between 14-3-3 and PM H⁺-ATPase in guard cells (Fig. 1C-E), which upregulates stomatal opening (Fig. 1B) and photosynthesis (Fig. 2). A reduction in the carbon isotope ratio (δ¹³C) in chemically treated plants (Fig. 2) further supports the observation that the increase in biomass is due to the chemical activation of PM H⁺-ATPase in guard cells.

The reported FC-A’s stomatal opening activity²⁵ has been clearly confirmed in this study (Fig. 1B). Moreover, this work’s structure-activity-relationship study reveals that the molecular recognition between FC and 14-3-3/the C-terminal peptide of PM H⁺-ATPase is a pivotal factor influencing the degree of stomatal opening efficacy. Structurally similar FC-A and FC-J were shown to stabilize the binding of the phosphorylated C-terminal peptide of PM H⁺-ATPase to 14-3-3 by approximately 6-fold in terms of K_d value as compared with controls (Fig. 1E). Furthermore, these compounds have been demonstrated to enhance the PPI in plants (Fig. 1C) and to enhance stomatal opening by approximately 200% (Fig. 1B). In contrast, hydrophilic FC-H was apparently less effective in both enhancement of the PPI and stomatal opening compared with FC-A and FC-J, indicating that the chemical functionality at the 12, 16, and 6’ positions is important for the activity. These results are consistent with those of a previous study showing that FC-A stabilizes the binding of human 14-3-3σ to the TASK-3 C-terminal phosphopeptide 10 times more effectively than FC-H⁴⁴. The X-ray crystal structure of the ternary complex of 14-3-3σ bound to the TASK-3 phosphopeptide and FC-A showed that the 16-methoxy group of FC-A binds to a narrow hydrophobic pocket, anchoring the whole molecule to 14-3-3σ in a similar manner to that seen in the structure of tobacco 14-3-3 bound to FC-A (Fig. S1). In the case of the ternary complex of FC-H, which lacks the O-methyl group (3ux0), a highly defined water molecule was found to occupy this position⁴⁴, suggesting that binding of the O-methyl group to the pocket squeezes out the water molecule, thus achieving entropically favored binding to 14-3-3. Indeed, whole-plant evaluations demonstrated that FC-A and FC-J enhance the growth of Arabidopsis plants to a similar degree, whereas FC-H is less effective (Fig. 3C). These results confirmed that the enhancement of plant growth is caused by stabilization of the PPI between 14-3-3 and PM H⁺-ATPase by FCs in guard cells.

It remains unclear whether FC-A reversibly or irreversibly upregulates stomatal opening in whole plants. Earlier studies based on in vitro evaluations concluded that FC-A irreversibly activates PM H⁺-ATPase^19,45. In these studies, upregulation of the binding of 14-3-3 to PM H⁺-ATPase was evaluated by surface plasmon resonance spectroscopy¹⁹ or by submerging guard-cell protoplasts⁴⁵ in a buffer solution containing a relatively high concentration of FC-A (10 µM). In the present study, a sub-nmol amount of FC-A was applied to plant leaves, and at least under these conditions, the effect of FC-A on stomatal opening was found to be transient, being active up to 24 h and becoming ineffective by 48 h (Fig. 4A, B). These findings are consistent with the observation that the phosphorylation level of Thr947 in PM H⁺-ATPase is transiently upregulated by FC-A in guard cells, declining to the control level at 48 h (Fig. 4C). These results strongly suggest that inactivation of FC-A occurs in the plant leaves. At least two α-glucosidase genes and more than 40 β-glucosidase genes have been identified in Arabidopsis^46,47. Thus, metabolic degradation and inactivation involving deglycosylation are possible, as removal of the glucoside moiety from FC-A reduces its apparent affinity for 14-3-3 by approximately one order of magnitude⁴⁴. This suggests that daily treatment of FC may be necessary for optimal plant growth, which could present a challenge for practical agricultural applications. Further studies are required to determine the appropriate chemical treatment conditions, including the optimal intervals and the amount of the dose.

The pretreatment of Arabidopsis with FC-A increased the susceptibility to the bacterial pathogen Pseudomonas syringae pv. tomato DC3000, which enters the plant through stomata, although this remained at a moderate level (*p < 0.05, Fig. S13). These results may serve to highlight some potential risk associated with the agricultural application of FC-A. Many bacterial virulence factors target PM H⁺-ATPase-related plant components to induce stomatal opening⁴⁸. Some fungal pathogens may deploy fusicoccins to facilitate penetration through stomata^9,24. Thus, the negative side effect(s) of FC-A application on antimicrobial plant defenses remains to be elucidated. In this study, plants were treated with FC under the conditions of sufficient light and water supply, thereby indicating that in-house use, such as in plant factories, may be optimal for FC applications.

Most importantly, the activity of FC-A in terms of plant growth enhancement found in this work suggests that the promotion of stomatal opening is not the sole cause of plant death and that there are other unelucidated mechanisms underlying the phytotoxicity of FC-A. Extensive studies have shown that FC-A affects a number of physiological processes in plants^12,32, including the induction of abscission^49,50, proton extrusion⁵¹, cell enlargement and cotyledon growth⁵², and seed germination⁵³. These activities are often explained by the “acid-growth theory”⁵⁴, in which overactivation of PM H⁺-ATPase by FC-A results in cell-wall acidification, followed by activation of expansion, which loosens the cellulose network for growth. The diverse activities of FC-A may also reflect the wide range of physiological roles of PM H⁺-ATPase¹⁴ as well as 14-3-3 signaling networks in plants⁵⁵. For instance, in Arabidopsis, PM H⁺-ATPase is encoded by 12 genes⁵⁶, and 13 isoforms of 14-3-3 (10 characterized and 3 putative isoforms) have been identified that are expressed in many organelles in addition to the cytoplasm, such as chloroplasts⁵⁷ and the nucleus⁵⁸. Other studies have shown that 14-3-3 plays roles in a diverse array of physiological processes, including nitrogen metabolism⁵⁹, ATP synthesis⁶⁰, potassium ion diffusion⁶¹, flowering⁶², and hormone-driven transcriptional regulation⁶³. The protein sequences of the 14-3-3 isoforms are highly conserved, and the residues involved in binding to FC-A are identical across all of the isoforms (Fig. S9); thus, FC-A could potentially stabilize any 14-3-3 interaction if the binding partner includes an appropriate consensus motif capable of forming a ternary complex with FC-A. Our present results, therefore, do not exclude any additional mechanism(s) underlying the enhancement of growth triggered by FC-A. It is worthy of note that PM H⁺-ATPase activity affects electrical signals, which influence numerous physiological processes, such as photosynthesis and respiration⁶⁴. Further research will be necessary to elucidate more details regarding the mechanism leading to growth enhancement as well as the molecular basis of the phytotoxicity of FC-A.

The growth-promoting of FC-A was also observed in Komatsuna plants, with a similar extent of enhancement as that observed for Arabidopsis plants (Fig. 5), indicating that the effect of FC-A is broad-spectrum. It should be noted that the use of FC-A in agriculture may not be cost-effective due to its high price. The biosynthetic intermediate FC-J is crystalline and can be readily isolated in gram quantities by crystallization from a culture broth⁶⁵. Given that FC-J has been demonstrated to be as effective as FC-A, it seems reasonable to suggest that the use of FC-J would be a practical option for further agricultural applications in hydroponics and plant factories from a cost-benefit perspective.

In the present work, we found that the fungal toxin FC-A and its biosynthetic intermediate FC-J enhance plant growth. This effect can be attributed to stabilization of the PPI between 14-3-3 and PM H⁺-ATPase in guard cells, which promotes stomatal opening and photosynthesis (Fig. 6). This unexpected property should stimulate further research examining potential agricultural applications of FC-A and FC-J.

Fig. 6.

Schematic representation of enhancement of plant growth by fusicoccins. The ability of FC-A and FC-J (the chemical structure is omitted in this illustration) to upregulate the protein-protein interaction (PPI) between 14-3-3 and the C-terminal phosphopeptide motif of PM H⁺-ATPase leads to the activation of PM H⁺-ATPase, upregulation of stomatal opening and photosynthesis, thereby enhancing plant growth.

Methods

Plant materials and growth conditions

Plants of Arabidopsis thaliana Col-0 were grown for 3–5 weeks and used for experiments. For experiments in stomatal aperture measurement, thermal imaging, immunoblotting, and immunoprecipitation, A. thaliana seeds were sown on rockwool (Yasaihana block, Nippon Rockwool or Grodan) and grown under a 16 -h fluorescent light (8:00 to 24:00, 80–100 µmol m^–2 s^–1)/8-h dark cycle at 22 °C in nutrient medium in a growth chamber. For biomass productivity measurements, A. thaliana seeds were sown on soil in planter (mini planter 38 type, 38 × 18.5 × 14.7 H cm, Richell) and grown under a 16-h fluorescent light (8:00 to 24:00, 50 µmol m^–2 s^–1)/8-h dark cycle at 24 ℃ in approximately > 50% (Pa/Pa) humidity in a growth room. Fusicoccin-A (FC-A), FC-J and FC-H were obtained as metabolites of Phomopsis amygdali Niigata 2⁶⁶ and its mutants⁶⁵, respectively, according to the procedure reported in literature⁶⁷. The crude products were purified either by silica gel column chromatography or recrystallization to give colorless powders, respectively (purity > 95% on NMR).

Measurement of stomatal aperture in Arabidopsis Cotyledon

Stomatal apertures were measured using cotyledons of 11-d-old plants underwent a dark adaptation overnight. Cotyledons were treated with 2 µL droplet of compound (0 ~ 30 µM of FC-A, FC-J, or FC-H in 0.1 v/v % EtOH, 0.05% Approach BI (Maruwa Biochemical), and 0.02% Sillwet L-77 in filtered water), and the plants were incubated in the dark at 22 ℃ in ~ 100% (Pa/Pa) humidity in a growth chamber. Chemically treated cotyledons were harvested, and images of abaxial stomata were recorded by a digital camera (Axiocam 503 mono; Carl Zeiss) mounted on a microscope (Axio imager A2; Carl Zeiss), and stomatal aperture was examined using image J Fiji (https://fiji.sc/). For each group treated with different concentration of compound, three cotyledons were harvested from three individual plants, and at least ten stomata per cotyledon were measured, and the resulting ≥ 30 data per group set was used for statistical analysis.

Measurement of stomatal aperture of rosette leaves of Arabidopsis

Changes in stomatal aperture using rosette leaves of Arabidopsis plants were evaluated according to a method previously described in literature⁶⁸. Briefly, rosette leaves were harvested from 25-30-d-old plants in the dark and were blended in a Warning blender (Waring Commercial) for 3 s x 2 in 35 mL of distilled water. The epidermal tissues were collected on 58-mm nylon mesh and rinsed with distilled water and were incubated in 2 mL of reaction mixture containing FC-A (30 µM, 0.1% EtOH) in a mesbistrispropane (BTP) buffer (5 mM BTP, 50 mM KCl, and 0.1 mM CaCl₂, pH 6.5) for 3 h in the dark at 24 °C. Images of stomata in the abaxial epidermis were recorded by a microscope (Axio imager A2; Carl Zeiss and ECLIPSE E100, Nikon) at 0, 12, 24, and 48 h after the chemical treatment, and the stomatal apertures were examined by image J Fiji (https://fiji.sc/). At least 30 stomata were measured for statistical analysis.

Crosslinking of Dynabeads and pre serum or anti-PM H⁺-ATPase antibody

Suspended Dynabeads (50 µL, Dynabeads Protein A Immunoprecipitation Kit, Veritas, Tokyo, Japan) was placed into a 1.5 mL tube, and supernatant was removed. To the tube was added pre serum or anti-PM H⁺-ATPase antibody⁶⁹ (20 µL) and Binding & Washing buffer (200 µL, available from the kit, Veritas, Tokyo, Japan), and the mixture was incubated for 10 min at r.t. with a gentle rotation. Supernatant was removed, and the beads was resuspended in 0.2 M boric acid (200 µL, adjusted to pH 9.0 with NaOH), and the supernatant was removed. This procedure was repeated twice. After removal of boric acid solution, to the Dynabeads was added 20 mM dimethyl pimelimidate dihydrochloride (250 µL) and gently rotated for 2 h at r.t. After removal of supernatant, to the beads was added 0.2 M ethanol amine (200 µL) to terminate crosslinking reaction. This procedure was repeated twice. The beads were treated with 0.2 M ethanol amine (250 µL) and gently rotated for 15 min at r.t, and the solution was removed. The beads were quickly suspended into 0.1 M glycine (200 µL, adjusted to pH 2.8 with HCl) and unreacted antibodies were removed immediately. This washing procedure was repeated twice. The resulting Dynabeads-antibody conjugates were washed with PBS (200 µL) and pH was confirmed at around 7.

Chemical treatment of Arabidopsis leaves

Arabidopsis rosette leaves (10 pieces, 25-30-d-old) were harvested and placed into a disposal plastic syringe (12 mL), and 0.1% EtOH water or 30 µM FC solution (5 mL) was added. Negative pressure was applied until leaves became transparent partially, and leaves were corrected, wiped, and placed into a mortar. To the ice-cold mortar was added 1x homogenizing buffer (400 µL, 50 mM MOPS-KOH, pH 7.5, 100 mM NaCl, 2.5 mM EDTA, 20 mM NaF, 1 mM ammonium molybdate, 1 mM DTT, 1x protease inhibitor cocktail), and the leaves were homogenized. Additional 1x homogenizing buffer (500 µL) was added and leaves were further homogenized. The resulting mixture was transferred into a 1.5 mL tube, 20% Triton X-100 (100 µL) was added, mixed by inverting, and then centrifuged (10,000 rpm) for 5 min at 4 °C. Supernatant (900 µL) was collected in a separated tube, and the protein concentration was measured by Bradford protein assay prior to use.

Immunoprecipitation of the PM H⁺-ATPase

Immunoprecipitation was performed according to a procedure previously described in literature⁷⁰. Briefly, the prepared Dynabeads-antibody conjugate (50 µL) was mixed with the prepared protein sample (protein concentration was adjusted to 2 mg/mL), and incubated with gentle rotation for 1 h at 4 °C. After removal of supernatant, the resin were washed twice with 1x homogenizing buffer (100 µL, 50 mM MOPS-KOH, pH 7.5, 100 mM NaCl, 2.5 mM EDTA, 20 mM NaF, 1 mM ammonium molybdate, 1 mM DTT, 1x protease inhibitor cocktail). The resulting Dynabeads-antibody-antigen complex was resuspended in 1x homogenizing buffer (50 µL) and transferred into a 1.5 mL fresh tube and kept on ice.

The coimmunoprecipitated beads were suspended in 2x sodium dodecyl sulfate sample buffer (32 µL), and the supernatant was collected into a 1.5 mL tube and mixed with 2-mercaptoethanol (6 µL). The coimmunoprecipitates were immediately applied to SDS-PAGE electrophoresis analysis within 30 min after preparation.

Immunoblot

Protein samples were separated by electrophoresis using 10% SDS-acrylamide gel and transferred to PVDF membranes in transfer buffer (25 mM Tris, 192 mM glycine, 20% methanol). The membranes were incubated in blocking buffer (TBS-T; 10 mM Tris–HCl pH 7.4, 100 mM NaCl, 0.05% Tween-20, 5% skim milk) for 30 min and then reacted with antibodies against, AHA2⁶⁹ at a dilution of 1:5000, phosphorylated Thr947 of AHA2⁶⁹ at a dilution of 1:5,000, and 14-3-3 protein¹⁸ at a dilution of 1:3,000, respectively, in blocking buffer at 4 °C overnight. The membrane was then rinsed three times for 5–10 min each in T-TBS [10 mM Tris–HCl (pH 7.4), 100 mM NaCl, 0.05% Tween-20] and reacted with a goat anti-rabbit IgG secondary antibody conjugated to horseradish peroxidase (HRP) (Santa cruz biotechnology or Cell signaling technology) at a dilution of 1:10,000 in blocking buffer for 1 h at r.t. Chemiluminescence with the HRP reaction using a chemiluminescence substrate (FUJIFILM Wako Chemicals) was detected by FUSION SOLO.7 S.EDGE (Vilber Bio Imaging).

Expression and purification of recombinant 14-3-3 protein

The plasmid pET30a-His-GF14φ⁶⁹ was transformed into E. coli strain BL21 (DE3) (Nippon Gene), which grew in LB medium (Kanamicin = 30 µg/mL) until the optical density (O. D.) reached to 0.7. Expression of 14-3-3 was induced by the addition of isopropyl-β-D-thiogalactoside (1 mM), and the cells were harvested after 3 h and lysed in lysis buffer (50 mM NaH₂PO₄, pH 8.0, 300 mM NaCl, 1% Nonidet P-40, protease inhibiter cocktail (Nacalai tesque)). The soluble fraction of the cell lysate was bound to Ni-NTA agarose resin (Qiagen) by incubation at 4 °C under rotation for 40 min. The protein was eluted with increasing concentration of imidazole (5 to 250 mM), and the fraction containing Hi-tag-14-3-3 protein were collected and dialyzed three times against dialysis buffer (1 L, 50 mM NaH₂PO₄, pH 8.0, 300 mM NaCl, 1mM DTT). The protein solution was concentrated, and the purity was checked by SDS-PAGE gels (> 90%). The resulting His-14-3-3φ were divided into aliquots and kept under − 80 °C until use.

Fluorescence polarization titration assay

Experiments were carried out using a microplate reader (Spectramax M5, Molecular Devices) in HEPES buffer (10 mM, pH 7.3, 150 mM NaCl, 0.05% (v/v) Tween-20) with 96 well microplate (Nunc F96 micro Well Black Polystyrene Plate). Wavelength for excitation and emission were set at 470 nm and 525 nm, respectively. His-14-3-3φ was titrated into FAM-TPSHYpTV (10 nM) in the presence of FC (4 µM). Binding curves were fitted using a single binding site model and dissociation constant (K_d) were calculated (Sigma Plot 14.0).

Gas-exchange measurement

Rosette leaves of four- to six-week-old Arabidopsis plants were treated twice with FC-A (500 µL of 30 µM/treat) or 0.1% EtOH, firstly 3 h and secondly 2 h prior to light irradiation, and incubated at r.t in dark. The treated leaf was then clamped in the measurement chamber of LI-6400 XT gas exchange system (LI-COR). The measurement chamber was set to 400 ppm CO₂; the leaf temperature and chamber relative humidity were maintained at 24 °C and 40–50% (Pa/Pa), respectively. The white light was illuminated to the measurement chamber and the clamped leaf with the external light source.

Thermal imaging

Infrared images of Arabidopsis plants were taken by infrared camera FLIR ONE (FLIR). 25-day-old plants were kept in the dark at 24 °C for 16–18 h prior to chemical treatment. Two leaves were randomly selected, on each of which two droplets of 8 µL FC-A solution (0.1% EtOH, 0.05% Approach BI, 0.02% Sillwet L-77) were placed, kept in the dark for 5 h and subjected to measurement in the light.

Preparation of transgenic Arabidopsis thaliana expressing sfGFP-labeled AHA2

Escherichia coli strain DH5α was used as the host for DNA manipulation. A fragment of sfGFP was amplified with the primers (5’-GCCATATGAGTAAAGGAGAAGAAC-3’ and 5’-TCCCCCGGGTTTGTATAGTTCATCCATGCC-3’). The stop codon was deleted during amplification. The amplified fragment was digested with NdeI and SmaI, and inserted into the corresponding site of pRI101-AN, a vector for protein expression under the control of the cauliflower mosaic virus (CaMV) 35 S promoter (TaKaRa Bio), resulting in pRI101-sfGFP-N. A. thaliana cDNA fragment of AHA2 (At4g30190) with N-terminal five glycine linker was amplified with primers (5’-CGGGATCCGGTGGTGGTGGTGGTTCGAGTCTCGAAGATATCAAG-3’ and 5’-GGAATTCCTACACAGTGTAGTGACTGG-3’). The amplified cDNA fragment was digested with BamHI and EcoRI and inserted into the corresponding site of pRI101-sfGFP-N, resulting pRI101-sfGFP-AHA2.

The plant transformation plasmid vector was introduced into Agrobacterium tumefaciens strain GV3101::pMP90 by electroporation, and A. thaliana was transformed by floral dip method. Transformed plants were selected on 1/2 MS plates containing 0.8% agarose and 50 µg/mL kanamycin. F3 homozygous plants were used for the experiment.

Confocal laser-scanning microscopy

Abaxial epidermal tissues of 3 to 5-week transgenic 35 S::sfGFP-AHA2 Arabidopsis plants were collected by Perforated-tape Epidermal Detachment⁷¹. The tape-affixed abaxial epidermis was soaked in 1 mL solution of compound 1 (10 µM) in MES buffer (5 mM MES-BTP, pH 6.5, 50 mM KCl, 0.1 mM CaCl₂, 1% DMSO and 0.01% Silwet L77) and incubated at r.t. for 1.5 h, washed twice with MES buffer (1 mL, 1% DMSO), and kept in MES buffer (1 mL, 1% DMSO). Stained epidermises were observed using a confocal laser-scanning microscope (IX81; Olympus) equipped with a diode laser (405/473/635 nm), LDD559 laser (559 nm), and a 60x UPlanSApo (1.35 numerical aperture) oil-immersion objective. Images were acquired and processed using FLUOVIEW FV1000-D (Olympus) and ImageJ Fiji (https://fiji.sc/). Fluorescence of BODIPY-TR and sfGFP were detected by laser irradiation at 473 and 559 nm using dichroic mirror (DM405/473/559), Beam splitter (SDM560 for EGFP, Mirror for Alexa Fluor 594), absorbance filter (BA655-755), and detection filters CH1: EGFP and CH2: Alexa Fluor 594, respectively.

Effect of fusicoccins on the growth of Arabidopsis plants

10-d Arabidopsis plants were treated with 0.1% ethanol aqueous solution of FC-A, FC-J, or FC-H (0.3, 3, or 30 µM) once a day at 3:30 to 5 pm and were grown for further 15 days. The chemical treatment of FC-A was performed by spraying a group of 12 plants with 1 mL of compound solution for 4 days, 2 mL for 5 days, and then 3 mL for 5 days. Differences in the effect of FC-A, FC-J, and FC-H were examined by spraying plants with 2 mL of each compound solution for 15 days. Plants were harvested, roots were removed, and the plants above ground were weighed immediately for fresh weight. Dry weight was measured after plants were placed at 80 °C for 2 days.

Carbon isotope ratio analyses

Carbon isotopes were measured with an elemental analyzer (EA/IRMS, Delta V interfaced with FlashEA1112, Thermo Fisher Scientific, MA, USA). Arabidopsis plants were grown under the treatment with FC-A for 15 days prior to use. 25-d Rosette leaves were harvested and dried at 80 °C for 48 h and crashed by a tissue lyser (Retsch, Hann, Germany). The powered sample (3 mg) was subjected to measurement. Carbon isotope ration (δ13C, ‰) was obtained in δ-notation, where δ = R_sample/R_standard − 1and R_sample and R_standard are the isotope ratios of the plant sample and the PeeDee belemnite, respectively⁷².

Time-course changes in the weight of leaves

Two rosette leaves were selected and marked per a 21-d Arabidopsis plant. After dark place processing overnight, a drop of 4 µL FC-A solution (0.3, 3, and 30 µM) was placed at four different places on each leaf and were placed in the dark for 1 h. Residual liquids were wiped, and the plants were grown under normal condition. The chemically treated leaves were detached at 12, 24, and 72 h after the treatment, and each leaf was weighed at 0, 12 and 24 h after detachment. Five leaves were measured per group.

Time-course changes in stomatal opening

Two rosette leaves were selected and marked per a 21-d Arabidopsis plant and placed in the dark overnight. The selected leaves were treated with FC-A by dropping of 4 µL of solution (0.3, 3, and 30 µM) at four different places, and placed in the dark for 1 h. Residual liquids were wiped, and the plants were grown under normal condition. Three of the chemically treated leaves were harvested (a leaf from a plant) at 6, 12, 24, and 48 h after the chemical treatment and epidermal tissues were collected (see 1.3). At least five stomata were measured per a piece of tissue (six pieces were analyzed) and more than 30 stomata were examined.

Time-course changes in phosphorylation level of pThr947 in PM H⁺-ATPase

Five rosette leaves were selected and marked per a 4-week-old Arabidopsis plant. After dark place processing overnight, 50 µL of 0.3 µM FC-A was placed on abaxial side of each leaf (150 pmol FC-A/leaf) and were placed in the dark for 1 h. Residual liquids were wiped, and the plants were grown under normal condition. The chemically treated leaves were detached at 48 h after the treatment. The phosphorylation level and amount of PM H⁺-ATPase in abaxial guard cells were detected with immunohistochemical staining using specific antibodies, anti–pThr947 and anti–PM H⁺-ATPase antibodies, according to a procedure previously described in literature⁷⁰.

Effect of fusicoccin on the growth of Komatsuna plants

Komatsuna seeds (Japanese mustard spinach; Brassica rapa var. perviridis) were placed in a hydroponics sponge (Day 1), and plants were grown using a hydroponical system in the greenhouse. After 10 days, 18 plants were selected as a group and placed in a stainless-steel tray (295 × 230 mm) filled with water of 1 cm depth from the bottom. On the next day, two leaves were randomly selected per plant, and each leaf was treated with FC-A by dropping 8 µL of the aqueous solution (0.1% ethanol, 0.3, 3, and 30 µM of FC-A) in two different places of the leaf surface. This procedure was carried out during 8:00 to 8:30 am, and repeated 6 days a week, for four weeks. Plants were harvested, then roots and top were weighed immediately. Dry weight was measured after plants were placed at 80 °C for 2 days.

Bacterial inoculation

25-day-old A. thaliana Col-0 plants were treated with 0.1% ethanol water or 3 µM FC-A by placing two droplets of 8 µL solution. 4.5 h after the chemical treatment, a cell suspension of Pseudomonas syringae pv. tomato DC3000 in 10 mM MgCl₂ (OD₆₀₀ = 1.0) was spray-inoculated with 0.02% Silwet L-77 onto rosette leaves of Arabidopsis plants and incubated for 4 days. The rate of bacterial growth inside the leaves was measured by enumeration of colonies on agar plates.

Statistics and reproducibility

Statistical analysis was performed using Microsoft Excel or R (ver. 4.3.2). Sample sizes are indicated in the figure captions. The statistical significance of differences was assessed using Student’s t test, Tukey-Kramer test, one-way ANOVA Dunnett’s test, or one-way repeated ANOVA pairwise t test as indicated in the figure captions.

Synthesis and characterization

General procedures for organic synthesis

Fusicoccin A, J, and H were obtained and purified by the procedures reported in literature, respectively⁶⁵. Reagents and solvents were purchased from commercial suppliers (Tokyo chemical industry, Sigma-Aldrich, Wako Pure chemical Industry, Nacalai tesque, Funakoshi, and Kanto Chemical). Column chromatography was performed on silica gel (63–212 µM, Wako). BODIPY TR-NHS was purchased from Lumiprove. ¹H NMR and ¹³C NMR spectra were recorded on 400 MHz (AVANCE NEO 400, Bruker). Chemical shifts were recorded in δ (ppm) relative to tetramethylsilane (TMS) or the solvent residual peak as an internal standard. All coupling constants were described in hertz (Hz). High-resolution mass spectroscopy was performed using JMS-700 (JEOL).

Synthesis of 1

2-(4-(5,5-Difluoro-7-(thiophen-2-yl)-5 H-5l4,6l4-dipyrrolo[1,2-c:2’,1’-f][1,3,2]diazaborinin-3-yl)phenoxy)-N-isopropylacetamide (1)

To a solution of BODIPY TR-NHS (5.8 mg, 0.94 µmol), triethylamine (~ 20 µL) in dimethylformamide (500 µL) was added isopropyl amine (2 µL, 0.023 mmol) on ice. The reaction mixture was stirred at r.t. for 1 h. The reaction was quenched by adding water (1 mL), and the product was extracted with ethyl acetate (10 × 3 mL). Combined organic layers was washed with brine and dried over anhydrous sodium sulfate. Concentration, followed by purification by silica gel column chromatography dichloromethane: methanol = 40 : 1) gave 2 as a blue solid (3.5 mg, quant.). ¹H NMR (400 MHz, CDCl₃) δ: 8.10 (d, J = 3.8 Hz, 1 H), 7.97 (d, J = 9.0 Hz, 2 H), 7.47 (d, J = 5.0 Hz, 1 H), 7.26 (s, 2 H), 7.18 (s, 1 H), 7.14 (t, J = 4.4 Hz, 1 H), 7.07 (dd, J = 10.3, 4.3 Hz, 2 H), 7.02 (d, J = 8.8 Hz, 2 H), 6.82 (d, J = 4.1 Hz, 1 H), 6.65 (d, J = 4.0 Hz, 1 H), 6.37 (d, J = 5.8 Hz, 1 H), 4.53 (s, 2 H), 4.21 (m, 1 H), 1.22 (d, J = 6.5 Hz, 6 H); ¹³C-NMR (400 MHz, CDCl₃) δ: 166.9, 158.3, 158.0, 151.0, 136.8, 136.6, 133.9, 131.5, 131.4, 131.4, 131.4, 130.2, 129.8, 129.5, 129.1, 126.5, 126.2, 120.8, 120.6, 114.6, 67.4, 41.2, 22.7. HRMS: m/z calcd for C₂₄H₂₂BF₂N₃NaO₂S [M + Na]⁺ 488.1392, found 488.1380.

Synthesis of 2

2-(4-(5,5-Difluoro-7-(thiophen-2-yl)-5 H-5l4,6l4-dipyrrolo[1,2-c:2’,1’-f][1,3,2]diazaborinin-3-yl)phenoxy)-N-(((2R,3 S,4 S,5R,6 S)-6-(((1 S,4R,5R,6R,6aS,9 S,10aR, E)-1,5-dihydroxy-3-isopropyl-9-(methoxymethyl)-6,10a-dimethyl-1,2,4,5,6,6a,7,8,9,10a-decahydrodicyclopenta[a, d][8]annulen-4-yl)oxy)-3,4,5-trihydroxytetrahydro-2 H-pyran-2-yl)methyl)acetamide (2)

To a solution of 3⁷³ (5.8 mg, 0. 94 µmol) in dichloromethane (1 mL) was added trifluoroacetic acid (500 µL) on ice. The reaction mixture was stirred on ice for 30 min and concentrated in vacuo. The resulting free amine was dissolved in dichloromethane (2 mL) and dimethylformamide (100 µL) and cooled on ice. To the solution was added BODIPY TR-NHS (5.4 mg, 0.0104 mmol) and triethylamine (20 µL), and the reaction mixture was stirred at room temperature for 30 min. After concentration, the residue was directly applied to silica gel column chromatography (dichloromethane : methanol = 40 : 1) to give 2 as a blue solid (7.6 mg, 87%). ¹H NMR (400 MHz, CDCl₃) δ: 8.10 (d, J = 3.5 Hz, 1 H), 7.98 (d, J = 8.8 Hz, 2 H), 7.48 (d, J = 4.6 Hz, 1 H), 7.18 (s, 1 H), 7.15 (dd, J = 4.9, 3.9 Hz, 1 H), 7.07 (dd, J = 6.9, 4.4 Hz, 2 H), 7.02 (d, J = 9.0 Hz, 2 H), 6.94 (q, J = 4.3 Hz, 1 H), 6.82 (d, J = 4.3 Hz, 1 H), 6.66 (d, J = 4.3 Hz, 1 H), 5.33 (s, 1 H), 4.93 (d, J = 3.8 Hz, 1 H), 4.62 (dd, J = 28.1, 15.4 Hz, 2 H), 4.25 (d, J = 3.8 Hz, 1 H), 4.20 (s, 1 H), 4.15–4.08 (m, 1 H), 3.95 (dd, J = 10.0, 4.5 Hz, 1 H), 3.85–3.64 (m, 4 H), 3.67 (s, 2 H), 3.36–3.50 (m, 3 H), 3.35 (s, 3 H), 3.24 (s, 1 H), 3.16 (sep, J = 6.8, 1 H), 3.10-3.00 (m, 2 H), 2.97 (s, 1 H), 2.85–2.74 (m, 2 H), 2.49 (dd, J = 15.6, 5.9 Hz, 1 H), 2.11 (dd, J = 15.8, 4.8 Hz, 1 H), 2.07–1.97 (m, 1 H), 1.96–1.87 (m, 1 H), 1.79–1.43 (m, 7 H), 1.36–1.23 (m, 3 H), 1.18 (s, 3 H), 1.04 (d, J = 6.5 Hz, 3 H), 0.99 (d, J = 6.8 Hz, 3 H), 0.88 (d, J = 7.3 Hz, 3 H); ¹³C-NMR (100 MHz, CDCl₃) δ: 170.4, 157.9, 157.6, 151.1, 147.5, 145.7, 136.8, 136.6, 135.3, 133.9, 131.6, 131.5, 131.5, 131.4, 130.3, 129.8, 129.7, 129.1, 126.8, 126.3, 125.7, 121.0, 120.6, 114.5, 101.0, 79.5, 78.3, 77.2, 76.1, 72.5, 72.3, 70.9, 70.1, 66.8, 58.7, 56.1, 47.7, 41.6, 41.3, 39.3, 36.1, 28.3, 27.9, 24.7, 21.4, 20.4, 10.0. ESI-HRMS: m/z calcd for C₄₈H₅₈BF₂N₃NaO₁₀S [M + Na]⁺ 940.3797, found 940.3797.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

H.K. and H.I. performed measurement of stomatal aperture. H.K. conducted immunoprecipitation and immunoblot with technical supports from Y.H. and T.K. H.K. performed protein expression and purification, fluorescent titration, and thermal imaging. H.K. and H.I. conducted confocal microscopy. H.K. evaluated growth effects of Arabidopsis plants with technical support from T.K. H.K. conducted carbon isotope ratio analysis. H.K. performed time-course changes in the weight of leaves and stomatal opening. H.K. conducted organic synthesis and characterization of compounds. S.N.K. conducted gas-exchange measurement. H.I. designed transgenic Arabidopsis plants and H.K. prepared the plants. J.O. and S.K. conducted experiments using Komatsuna plants. Y.H. performed immunostaining experiment, and R.H. conducted bacterial infection experiment. T.K. and Y.H. prepared antibodies and expression vector for recombinant 14-3-3. J.O., T.K., H.I. and S.K. supervised the research. J.O. conceptualized the research and wrote the manuscript with input from all the authors. The manuscript was reviewed and edited by all the authors.

Data availability

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Declarations

Competing interests

J. Ohkanda, H. Kiriyama, S. Kasuga, H. Irieda, and T. Kinoshita (JP patent 2022-132203). All the remaining authors declare no conflict of interest.