AIM1-dependent high basal salicylic acid accumulation modulates stomatal aperture in rice
Summary
- The basal levels of salicylic acid (SA) vary dramatically among plant species. In the shoot, for example, rice contains almost 100 times higher SA levels than Arabidopsis. Despite its high basal levels, neither the biosynthetic pathway nor the biological functions of SA are well understood in rice.
- Combining with metabolite analysis, physiological, and genetic approaches, we found that the synthesis of basal SA in rice shoot is dependent on OsAIM1, which encodes a beta-oxidation enzyme in the phenylalanine ammonia-lyase (PAL) pathway.
- Compromised SA accumulation in the Osaim1 mutant led to a lower shoot temperature than wild-type plants. However, this shoot temperature defect resulted from increased transpiration due to elevated steady-state stomatal aperture in the mutant. Furthermore, the high basal SA level is required for sustained expression of OsWRKY45 to modulate the steady-state stomatal aperture and shoot temperature in rice.
- Taken together, these results provide the direct genetic evidence for the critical role of the PAL pathway in the biosynthesis of high basal level SA in rice, which plays an important role in the regulation of steady-state stomatal aperture to promote fitness under stress conditions.
Introduction
Salicylic acid (SA) is an important plant hormone that regulates growth, development, and various stress responses (Rivas-San Vicente & Plasencia, 2011). The best-established role of SA is the regulation of plant defense response against pathogen infection. The earliest report of SA in plant defense is the effect of exogenously applied SA in protecting tobacco leaves against viral infection (White, 1979). Subsequently, it was found that SA could act as an endogenous signal in plant defense (Raskin, 1992). During the past 30 yr or so, extensive progress has been made in the understanding of the signaling pathway of SA-mediated defense response in plants. In addition to its role in plant defense, SA plays important roles in root growth, leaf senescence, and heat production in some thermogenic plants (Zhang et al., 2013; Tan et al., 2020).
Endogenous SA is involved in thermogenesis in arum lily flowers for dispersing flower odor to attract pollinators (Raskin et al., 1987). The SA levels in the flowers of the thermogenic plants surge c. 100-fold upon blooming, leading to a temperature increase of c. 10°C above the ambient temperature. Increased levels of SA promote thermogenesis by inducing the expression of genes encoding alternative oxidase (AOX), a key enzyme in the alternative respiratory pathway in mitochondria that is uncoupled from ATP production and releases energy as heat (Raskin et al., 1989; Rhoads & McIntosh, 1992). In most plants, basal SA levels are very low SA (c. 100 ng g−1 fresh weight (FW)), but they can increase drastically in lily flowers during development or under certain biotic and abiotic stress conditions. Other plants such as Populus and rice have very high basal SA levels (>10 μg g−1 FW) under normal growth conditions. Whether basal SA contributes to thermogenesis in non-thermogenic plants, especially those with high basal SA levels, is unknown.
Salicylic acid is synthesized through two distinct pathways in plants: the isochorismate synthase (ICS) pathway and the phenylalanine ammonia-lyase (PAL) pathway (Dempsey et al., 2011). Chorismate is a precursor of both these pathways. Isochorismate synthase catalyzes chorismite to isochorismate in the plastid. Isochorismate is exported by ENHANCED DISEASE SUSCEPTIBILITY5 (EDS5) to the cytosol, where the cytosolic amidotransferase avrPphB SUSCEPTIBLE3 (PBS3) catalyzes the conjugation of glutamate to isochorismate to produce isochorismate-9-glutamate, which spontaneously decomposes into SA and 2-hydroxy-acryloyl-N-glutamate (Rekhter et al., 2019). On the other hand, PAL catalyzes the chorismate-derived l-phenylalanine into cinnamic acid (CA), which is then converted to SA via o-coumarate or benzoic acid (BA). β-oxidation plays a crucial role in conversion of o-coumarate to SA or CA to benzoate (Dempsey et al., 2011).
The PAL and ICS biosynthetic routes contribute differently to the biosynthesis of SA in different plant species. Arabidopsis contains low basal levels of SA, and most of defense-related SA is synthesized via the ICS pathway (Wildermuth et al., 2001; Garcion et al., 2008). A more recent study has shown that ABNORMAL INFLOURESCENCE MERISTEM 1 (AIM1), a peroxisome β-oxidation enzyme, functions in SA biosynthesis in Arabidopsis seeds (Bussell et al., 2014), indicating a role for the AIM1-dependent β-oxidation functions in the PAL pathway in basal SA biosynthesis. In Populus, the single ICS-encoding gene primarily functions in phylloquinone biosynthesis and SA is synthesized from CA via the PAL pathway (Yuan et al., 2009; Xue et al., 2013). In soybean, the PAL and ICS pathways are equally important for pathogen-induced SA biosynthesis (Shine et al., 2016). Rice contains a very high basal level of SA, two orders of magnitude higher than Arabidopsis (Silverman et al., 1995). However, the contribution of the two biosynthesis pathways to the high basal SA level is not well understood in rice. Early feeding studies suggested that rice shoot converted CA and BA into SA (Silverman et al., 1995). In a previous study, we found that the OsAIM1-dependent PAL pathway is important for SA synthesis in rice roots (Xu et al., 2017). Rice shoots contain a much higher level of SA than the roots (Pal et al., 2014). Although a previous study has implicated ICS in rice SA production (Choi et al., 2015), there has been no genetic evidence for how the basal SA is produced in rice shoots.
Although rice is well known to contain a higher basal level of SA, its biological function is less clear. Unlike in Arabidopsis, SA levels do not greatly increase after pathogen infection in rice (Silverman et al., 1995). However, several studies have suggested that SA plays a role in defense against pathogen infection in rice (Shimono et al., 2007; Kouzai et al., 2018; Jiang et al., 2020). It has been suggested that SA is important for protecting rice from oxidative damage during pathogen infections (Yang et al., 2004). In addition, we have previously shown that SA is crucial for meristem activity in rice roots (Xu et al., 2017), but it is still largely unclear about the biological function of high basal levels of SA in rice shoots.
In this study, we have found that the high basal SA level is unaltered in a rice knockout mutant for its sole ICS gene. By contrast, the mutant for OsAIM1 is drastically reduced in basal SA levels in the rice shoots. These results provide the first genetic evidence for the PAL pathway in the biosynthesis of high basal SA levels in rice. In addition, we have also found that the high basal level of SA in rice shoots is required for maintaining the steady-state stomatal aperture and shoot temperature through a pathway that is dependent on the rice OsWRKY45 transcription factor. We have further shown that the high basal SA levels are important for tolerance to abiotic stresses in rice.
Materials and Methods
Plant materials and growth conditions
Hydroponic experiments were conducted using a modified rice (Oryza sativa L.) culture solution containing 1.425 mM NH4NO3, 0.2 mM NaH2PO4, 0.513 mM K2SO4, 0.998 mM CaCl2, 1.643 mM MgSO4, 0.009 mM MnCl2, 0.075 mM (NH4)6Mo7O24, 0.019 mM H3BO3, 0.155 mM CuSO4, and 0.152 mM ZnSO4 with 0.125 mM EDTA-Fe. pH of the solution was adjusted to 5.5. All the plants were grown in a glasshouse with a 12 h : 12 h, 30°C : 22°C, day : night photoperiod, c. 200 μmol m−2 s−1 photon density, and c. 60% humidity.
For Xanthomonas oryzae pv Oryzae (Xoo) infection, Xoo strain PXO 341 was cultured in modified Wakimoto's medium (per liter: peptone, 5 g; sucrose, 20 g; MgSO4·7H2O, 0.25 g; K2HPO4, 0.5 g; pH 7.2–7.4) at 28°C. Fourteen-day-old rice seedlings were treated with water or Xoo for 24 h, then the shoot was sampled for SA quantification and gene expression analysis.
For measurement of stomatal conductance after H2O2 treatment, 21-d-old seedlings were sprayed with 0.5 mM H2O2 for 2 h.
For measurement of H2O2 level and stomatal conductance after NaCl or polyethylene glycol (PEG) treatment, 21-d-old seedlings were treated with 100 mM NaCl/20% PEG and 100 mM NaCl/20% PEG+ 200 μM SA (SA was added before NaCl/PEG treatment for 3 h) for 6 h.
Plasmid construction and plant transformation
The sequence information for primers used to construct vectors is shown in Supporting Information Table S1. For ProOsWRKY45:GUS vector, the 2753 bp promoter before the start code was introduced into the PstI and KpnI sites on pCAMBIA 1300-GUS vector. For ProNPR1:GUS vector, the 2425 bp promoter before the start code was introduced into the SalI and KpnI sites on pCAMBIA 1300-GUS vector. For ProOsWRKY45:OsWRKY45 vector, the 4428 bp genome fragment was introduced into the KpnI and XbaI sites on pCAMBIA 1300 vector. The constructs were transformed into mature embryos developed from seeds of wild-type (WT; ‘Shishoubaimao’) or the Oswrky45 mutant via Agrobacterium tumefaciens mediated transformation. The primers used are listed in Table S1.
Histochemical localization of GUS expression
For GUS staining, the tissues were incubated in a solution containing 50 mM sodium phosphate buffer (pH 7.0), 5 mM K3Fe(CN)6, 5 mM K4Fe(CN)6, 0.1% Triton X-100, and 1 mM X-Gluc at 37°C. Sections (35 μm) of various plant tissues were made by vibratome (Leica VT 1000 S; Leica Biosystems, Nussloch, Germany). Images were taken by microscope (Nikon Eclipse Ni; Tokyo, Japan).
Quantitative RT-PCR
Reverse transcription was performed using 2 μg of total RNA and M-MuLV Reverse Transcriptase (NEB) according to the manufacturer's instructions. Quantitative PCR (qPCR) was performed using the Roche SYBR green I kit on the LightCycler480 machine (Roche Diagnostics, Basel, Switzerland) according to the manufacturer's instructions. Three biological replicates were performed for each gene. Rice Actin gene was used as an internal control. The primers used are listed in Table S1.
NBT and HPF staining
Rice roots were stained for 30 min in a solution of 2 mM nitroblue tetrazolium (NBT) or 5 μM 3′-(p-hydroxyphenyl) fluorescein (HPF) in 20 mM phosphate buffer (pH 6.1). The reaction was stopped by transferring the seedlings to distilled water. Root tips were imaged under bright-field illumination for NBT staining or GFP channel for HPF staining on a Nikon microscope. The intensity of NBT and HPF staining was quantified using ImageJ software.
Metabolites measurement
For metabolites measurements, seedlings were grown in a cultural solution for 2 wk. The aerial parts and roots were separately collected and freeze-dried. The freeze-dried tissues were used for metabolites measurement, following previously published methods (Chen et al., 2013). The sample extracts were analyzed using an LC–ESI–MS/MS system (HPLC, Shim-pack UFLC Shimadzu CBM20A system, www.shimadzu.com; MS, Applied Biosystems 4000 Q TRAP, www.appliedbiosystems.com).
Stomatal aperture measurement
For imaging rice stomata, leaves of 3-wk-old plants were fixed with 2.5% (v/v) glutaraldehyde, and stomatal pictures were obtained by a Hitachi TM3030Plus scanning electron microscopy. At least 60 stomata were examined in each line, and the assays were repeated three times.
Measurements of water loss
Water loss rates of the detached leaves from 3-wk-old plants were measured by monitoring the FW loss at indicated points under a constant temperature (28°C) and humidity (60%). The percentage loss of weight was calculated based on the initial weight of the plants.
Thermal imaging
Thermal images of the plant were taken with an infrared thermal camera (FLIR T400; FLIR Systems, Boston, MA, USA) and were subsequently analyzed through the FLIR R&D software.
Gas exchange measurements
The stomatal conductance (Gs) and transpiration rate (E) were measured using a LI-6800 portable photosynthesis system (PP Systems, Lincoln, NE, USA) at 28°C, 200 μmol m−2 s−1 light, 400 μmol mol−1 CO2, and 60% relative humidity.
Results
OsAIM1 but not OsICS is required for high basal SA accumulation in rice shoot
In plants, SA is synthesized through the ICS pathway and the PAL pathway (Fig. 1a). Arabidopsis contains low basal levels of SA and synthesizes defense-related SA mostly via the ICS pathway. Rice contains much higher levels of SA in shoots but the pathway responsible for rice SA synthesis has not been genetically established. Rice genome contains a single copy ICS gene (Yuan et al., 2009). To determine whether the ICS pathway participates in SA biosynthesis in rice shoots, we have isolated and analyzed an Osics1 mutant (PFG_3A-08161.R) from the T-DNA insertion mutant library (Fig. S1a). The mutant was confirmed by PCR-based DNA genotyping using primers flanking the T-DNA insertion site (Fig. S1b). The failure to detect the expression of OsICS1 suggested that Osics1 is a null mutant (Fig. S1c). Notably, the SA content in the Osics1 mutant is similar to that in WT (Fig. 1b). This result indicates that OsICS1 is not required for high basal levels of SA production in rice shoots.
It has been reported that the ICS1 is essential for phylloquinone other than the SA biosynthesis in barley. The ics1 mutant displays comparable SA content with the WT plants, while the deficiency of phylloquinone in the mutant leads to a wilting symptom (Qin et al., 2019). Likewise, Osics1 mutant displayed a pale green to yellowish pigmentation in leaf and started wilting c. 10 d after germination (Fig. S1d,e), implying that OsICS1 might be required for phylloquinone other than SA biosynthesis.
In a previous study, we found that OsAIM1-dependent β-oxidation is required for SA biosynthesis in rice roots and OsAIM1 is also strongly expressed in rice shoots (Xu et al., 2017). Therefore, OsAIM1 might also participate in SA synthesis in rice shoots. To test this possibility, we measured the shoot SA content in the Osaim1 mutant. As shown in Fig. 1c, the SA content in the mutant shoots was only c. 10% of that in WT shoots. The introduction of the OsAIM1 gene driven by either its native promoter or the CaMV 35S promoter fully restores the SA content of the mutant to the WT levels (Fig. 1c). Thus, the OsAIM1-dependent β-oxidation is also required for the high basal levels of SA accumulation in rice shoots.
In the PAL pathway, β-oxidation is required for the shortening of the side chain of CA to produce BA, which is then converted to SA. Consistent with a critical role of OsAIM1 in the β-oxidation of CA in the PAL pathway, the BA levels were also significantly reduced in the Osaim1 mutant when compared with those in WT (Fig. 1d). To further confirm the role of OsAIM1 in SA biosynthesis, we performed chemical feeding experiments. Application with 50 μM CA resulted in a 1.4-fold increase in both BA and SA levels in WT plants (Fig. 1d,e). However, the same feeding treatment in the Osaim1 mutant only slightly increased the BA levels and had no significant effect on the SA levels (Fig. 1d,e). By contrast, feeding with BA restored the SA content of Osaim1 mutant to the WT levels (Fig. 1f). These results demonstrated that OsAIM1-mediated β-oxidation in the PAL pathway is required for the high basal levels of SA production in rice shoots.
Given that Xanthomonas oryzae pv Oryzae (Xoo) infection could increase SA production in rice (Bakade et al., 2021), we also determined whether the OsAIM1-mediated β-oxidation is required for it. We first verify the expression of OsAIM1 after Xoo (race PXO341) treatment. Similar to the positive control, OsPR1b, the expression of OsAIM1 was induced by Xoo treatment (Fig. 1g). We further measured the SA levels in WT and Osaim1 after Xoo treatment. It showed that the SA levels in WT increased 2.5-fold after Xoo infection, whereas the SA levels in the Osaim1 mutant were unchanged (Fig. 1h). Despite that the Xoo infection could not induce SA accumulation in the Osaim1 mutant, it still induced the expression of OsICS1 (Fig. 1i). It indicated that the Xoo induced SA biosynthesis was due to the OsAIM1-dependent pathway, other than the ICS pathway. Together, these results suggest that the OsAIM1-dependent pathway is not only crucial for basal SA biosynthesis, but also for pathogen-induced SA accumulation.
The high basal level of SA is required for modulating the steady-state stomatal aperture and shoot temperature
Rice shoots contain high basal levels of SA with little information on its biological function. In addition to its role in plant defense responses, SA is known to be important for heat production by inducing the expression of AOXs and promoting the alternative respiratory pathway in some thermogenic plants (Rhoads & McIntosh, 1992). To determine whether the high basal SA content influences the temperature of rice shoots, we took thermal images of WT and Osaim1 mutant shoots by the infrared thermal camera. Consistently, we observed that the shoot temperature of the Osaim1 mutant was over 1°C lower than that of WT (Fig. 2a,b). The decreased shoot temperature phenotype in Osaim1 was completely rescued by the OsAIM1 gene driven by the CaMV 35S or OsAIM1 native promoter (Fig. 2a,b). To determine whether reduced SA content is responsible for the observed shoot temperature phenotype in the Osaim1 mutant, we treated WT and Osaim1 with SA. Exogenous SA application slightly increased the WT shoot temperature and could restore the shoot temperature of the Osaim1 mutant to the WT level (Fig. 2c,d). To determine whether SA elevates rice shoot temperature through upregulating AOX genes as in some thermogenic plants, we compare the AOX gene expression between WT and the Osaim1 mutant. qRT-PCR revealed that AOX gene expression was not compromised but was actually elevated in the Osaim1 mutant when compared to that in WT (Fig. 2e). These findings indicated factors other than the AOX-based thermogenesis are responsible for reduced shoot temperature in the Osaim1 mutant.
Transpiration also negatively influences leaf temperature (Lin et al., 2017). Therefore, we examined the potential difference in transpiration between WT and the Osaim1 mutant. Indeed, the Osaim1 mutant displayed a transpiration rate c. 30% higher than WT (Fig. 2f). Consistent with increased transpiration, the detached leaves of the Osaim1 mutant lost water more quickly than WT plants (Fig. S2a). Transpiration is mainly determined by stomata density, size, and aperture. However, we could not detect any significant difference in the density and size of stomata between WT and Osaim1 mutant (Fig. S2b,c). Therefore, we further checked the stomatal apertures of the WT and Osaim1 mutant using the scanning electron microscope. As shown in Fig. 2(g), there were higher percentages of completely open stomata in the Osaim1 mutant (38%) than in WT (27%). The percentage of partially open stomata was c. 54% in WT and 58% in the Osaim1 mutant (Fig. 2g). Consistent with the higher percentage of open stomata, the Osaim1 mutant displayed higher stomatal conductance than WT (Fig. 2h).
Taken together, these results strongly suggest that the OsAIM1-dependent PAL pathway is responsible for the production of the high basal SA levels in rice shoots, which, in turn, is required for modulating stomatal aperture and transpiration to influence shoot temperature and other physiological traits.
SA regulates the stomatal aperture and shoot temperature through an OsWRKY45-dependent pathway
To determine how the high basal SA level regulates stomatal aperture and shoot temperature in rice, we analyzed the factors that are involved in SA signaling. Contrary to the central role of NPR1 in SA signaling in Arabidopsis, SA signaling in rice is mediated by both the OsNPR1 and OsWRKY45 sub-pathways (Shimono et al., 2007; Nakayama et al., 2013). To investigate the roles of OsNPR1 and OsWRKY45 in SA-regulated stomatal aperture and shoot temperature, we generated rice Oswrky45 and Osnpr1 loss-of-function mutants by CRISPR-Cas9 (Fig. S3). Like Osaim1 mutant, the Oswrky45 mutant displayed significantly lower shoot temperature than WT (Fig. 3a,b). Consistent with reduced shoot temperature, the Oswrky45 mutant had an increased transpiration rate than WT plants (Fig. 3c). The detached leaves of the Oswrky45 mutant also lost water more quickly than WT plants (Fig. S4). On the other hand, the shoot temperature, transpiration, and water loss rate of the Osnpr1 mutant were all similar to those of WT (Figs 3a–c, S4). These results suggest that OsWRKY45, but not OsNPR1, mediates SA-regulated stomatal aperture and shoot temperatures. Consistent with this hypothesis, GUS staining of the POsWRKY45:GUS transgenic plant showed that OsWRKY45 is highly expressed in the stomata apparatus (Fig. S5a), while expression of OsNPR1 is absent in the stomata apparatus based on the histochemical staining of POsNPR1:GUS transgenic plants (Fig. S5b).
We further analyzed the stomatal apertures of the Oswrky45 mutant and WT plants using scanning electron microscopy. As shown in Fig. 3(d), the percentage of completely open stomata in the Oswrky45 mutant (29%) was significantly higher than that of WT (25%). The percentage of partially open stomata in the Oswrky45 mutant (61%) was also significantly higher than that of WT (56%) (Fig. 3d). As a result, the percentage of completely closed stomata in the Oswrky45 mutant (10%) was substantially lower than that of WT (19%; Fig. 3d). Consistent with the higher percentage of open stomata, the Oswrky45 mutant also displayed higher stomatal conductance than WT (Fig. 3e). The decreased shoot temperature and increased stomatal conductance in Oswrky45 were completely suppressed in the transgenic Oswrky45 complementation lines containing the OsWRKY45 genomic fragment (Fig. S6). These corroborate the role of OsWRKY45 in the modulation of stomatal aperture and shoot temperate in rice.
To further determine whether OsWRKY45 acts downstream of SA to regulate stomatal aperture and shoot temperatures, we analyzed the effect of SA on OsWRKY45 expression. Both the qRT-PCR analysis and GUS staining analysis showed that exogenous SA application strongly induced the expression of OsWRKY45 in rice shoot and guard cells (Fig. 3f,g). To determine the effect of endogenous SA on OsWRKY45 expression, we further analyzed the expression of OsWRKY45 in the Osaim1 mutant and found it to be significantly repressed in the mutant when compared with that in WT (Fig. 3f). The exogenous application of SA restored the expression of OsWRKY45 in the Osaim1 mutant to the WT level (Fig. 3f). By contrast, there was no significant alteration of OsNPR1 expression in the Osaim1 mutant relative to that in WT (Fig. S7). Given that SA treatment could rescue both OsWRKY45 expression and the defects of stomatal apertures and shoot temperature in the Osaim1 mutant, it is possible that the high basal SA level in rice shoots promotes OsWRKY45 expression to regulate stomatal aperture, thereby affecting shoot temperature in rice. Reduced SA levels in the Osaim1 mutant lead to increased stomatal aperture and reduced shoot temperature due to the reduction of OsWRKY45 expression.
To test this possibility, we further generated the Osaim1 Oswrky45 double mutant and compared the effects of SA on the single and double mutants. First, the Osaim1 Oswrky45 double mutant had a shoot temperature similarly lower than WT as the Osaim1 or Oswrky45 single mutant (Fig. 3h,i). Thus, the effects of the Osaim1 and Oswrky45 mutations are not additive, supporting that they act in the same pathway. Second, unlike the Osaim1 mutant, whose reduced shoot temperature could be restored by SA treatment, the Oswrky45 single and Osaim1 Oswrky45 double mutants were insensitive to SA application for restoration of their reduced shoot temperature (Fig. 3h,i). Likewise, while SA fully suppressed the elevated stomatal conductance of the Osaim1 mutant, it failed to do so with the Oswrky45 single and Osaim1 Oswrky45 double mutants (Fig. 3e). The non-additive nature of the effects of the Osaim1 and Oswrky45 mutations and their differential responses to SA strongly support that OsWRKY45 acts downstream of the high basal shoot SA levels in the regulation of stomatal apertures and shoot temperatures in rice.
The high rice shoot basal SA maintains OsWRKY45 expression to affect H2O2 accumulation in the stomatal apparatus and promote abiotic stress tolerance
H2O2 is an important signaling molecule involved in the regulation of stomatal aperture (Waszczak et al., 2018). Given the critical role of OsAIM1 and OsWRKY45 in the regulation of stomatal apertures, we further investigated the H2O2 levels in the Osaim1 and Oswrky45 mutants using 3′−ρ-HPF staining. Indeed, the H2O2 levels in the guard cells of the Osaim1, Oswrky45, and Osaim1 Oswrky45 double mutants were decreased when compared to those in WT. To determine whether H2O2 is involved in OsWRKY45-regulated stomatal aperture, we also determined the effect of SA on the H2O2 levels in the guard cells of WT, Osaim1, Oswrky45, and Osaim1 Oswrky45. SA treatment stimulated H2O2 production in WT and could rescue the defect of reduced H2O2 accumulation in the Osaim1 mutant, but not in the Oswrky45 single and Osaim1 Oswrky45 double mutants (Fig. 4a,b). Furthermore, the elevated stomatal conductance of the Osaim1 and Oswrky45 mutant was suppressed by H2O2 treatment (Fig. S8). These results indicate that the high basal SA level enhances H2O2 accumulation to regulate stomatal aperture in an OsWRKY45-dependent manner.
In rice, the DST zinc finger protein transcription factor interacts with another zinc finger protein DCA1 as a co-activator to affect drought and salt tolerance in rice via stomatal aperture control by regulating H2O2 homeostasis in rice (Huang et al., 2009; Cui et al., 2015). We analyzed the expression of OsDST and OsDCA1 in both the Osaim1 and Oswrky45 mutants and found expression levels of both genes to be increased in the Osaim1 and Oswrky45 mutants (Fig. 4c). These results suggest that OsDCA1 and OsDST might participate in OsWRKY45-regulated H2O2 accumulation and stomatal aperture in rice.
Steady-state stomatal aperture is crucial for plant adaptation to abiotic stress (Cui et al., 2015; Murata et al., 2015; Sukiran et al., 2020) and, therefore, the positive regulation of OsWRKY45 expression by the high basal shoot SA content could play a role in the plant tolerance to abiotic stress in rice. To test this, we first assessed the responses of WT and the Osaim1 to salt and PEG treatments. Fourteen-day-old seedlings were first treated with 100 mM NaCl for 7 d or 20% PEG 6000 for 10 d and then recovered in the normal growth medium for 10 d. As shown in Fig. 4(d), the Osaim1 mutant was substantially more sensitive to the salt and PEG treatments than WT. Approximately 40% and 70% of WT plants survived after the salt and PEG treatments, respectively (Fig. 4d–g). For comparison, only 8% and 24% of the Osaim1 mutant plants survived under the salt and PEG treatments, respectively (Fig. 4d–g). Furthermore, exogenous application of SA could significantly increase the survival rate of Osaim1 under the salt and PEG treatments. Thus, compromised SA synthesis resulted in hypersensitivity to salt and PEG treatments in the Osaim1 mutant. Similar to the Osaim1 mutant, the Oswrky45 mutant was less tolerant to both the salt and PEG treatments. However, SA application failed to restore the tolerance to salt and PEG treatments in the Oswrky45 mutant (Fig. 4d–g). After salt and PEG treatments, the Osaim1 and Oswrky45 mutants still displayed lower H2O2 levels and higher stomatal conductance compared with WT. However, SA application could rescue the defects of H2O2 accumulation and stomatal conductance of Osaim1, but not Oswrky45 under salt and PEG treatments (Fig. S9).
Taken all these results together, OsAIM1-dependent PAL pathway is essential for the accumulation of high basal level SA in rice shoots. In addition, the high basal level of SA is required for the regulation of the steady-state stomatal aperture through an OsWRKY45-dependent pathway in rice, which is crucial for abiotic stress tolerance in the important crop (Fig. 4h).
Discussion
Salicylic acid acts as a signaling molecule in plant inducible responses against various forms of environmental stresses and its biosynthesis is stress responsive in many plant species. Salicylic acid can be synthesized through two distinct pathways in plants, the ICS pathway and the PAL pathway (Ding & Ding, 2020). Isochorismate synthase plays an important role in SA production in some plant species (Rekhter et al., 2019). In Arabidopsis thaliana, AtICS1 is essential for stress-induced SA production. Likewise, a Nicotiana benthamiana ICS gene (NbICS) is transcriptionally induced by a pathogen elicitor and is required for induced SA production in response to biotic and abiotic stress conditions (Catinot et al., 2008; Shibata et al., 2010). By contrast, ICS in tobacco (Nicotiana tabacum; NtICS) does not appear to be involved in stress-induced SA production. The transcript levels of NtICS were not induced, and ICS activity was undetectable when SA production was induced by tobacco mosaic virus inoculation or ozone exposure (Ogawa et al., 2005). Rice constitutively contains two to three orders of magnitude higher levels of SA than Arabidopsis, N. benthamiana, and tobacco. If OsICS is required for the production of such higher levels of SA in rice, its activity would be expected to be much higher than that of AtICS1, NbICS, and NtICS. However, a recent study has shown that the in vitro and in planta ICS activities of OsICS were much lower than those of AtICS1 (Yokoo et al., 2018). Previously, we have found that OsAIM1-dependent β-oxidation is required for SA biosynthesis in rice roots (Xu et al., 2017). A recent study found that AIM1-dependent β-oxidation is required for the biosynthesis of SA in chlorophyte algae (Jia et al., 2023). In this study, we have further found that OsAIM1 is crucial for the biosynthesis of high levels of basal SA in rice shoots. The basal SA level in the shoots of rice Osaim1 mutant was reduced by almost 90% when compared with that in WT shoots. Thus, OsAIM1-dependent β-oxidation in the PAL pathway is responsible for the synthesis of a majority of basal SA in rice. Meanwhile, OsAIM1 is also required for pathogen induced SA biosynthesis. Although the ICS1 is present in the rice, our results indicate that it does not participate in the biosynthesis of high levels of basal SA. Therefore, the biosynthesis pathway of SA may vary in different plants. However, we could not exclude the possibility that ICS is responsible for SA biosynthesis in certain stress conditions in rice.
In Arabidopsis, most pathogen-induced SA is derived from isochorismate, which is generated from chorismate by ICS1 in the plastid (Rekhter et al., 2019). Isochorismate generated by ICS is also an early precursor for phylloquinone, which can be synthesized via 1,4-dihydroxy-2-naphthoic acid (DHNA) in a multistep pathway (Gross et al., 2006). As a result, Arabidopsis ics1 ics2 double mutant is deficient not only in SA but also in phylloquinone biosynthesis. Because phylloquinone is essential for electron transfer in photosystem I, the ics1 ics2 double mutants are smaller than WT and display chlorosis (Garcion et al., 2008). Like rice, barley contains a single ICS gene. Phylloquinone in the barley ics1 mutant is undetectable and the mutant displays wilting symptoms, both of these phenotypes could be rescued by DHNA application, indicating that the wilting phenotype of barley ics1 mutant is caused by phylloquinone deficiency (Qin et al., 2019). However, the SA level in barley ics1 mutant was comparable to that in WT. Similarly, we have found that the SA content in the Osics1 mutant was similar to that of WT. These results indicate that OsICS1 might be required for the production of phylloquinone but not SA in rice.
Rice shoot accumulates an extremely high basal level of SA, almost 100-fold higher than that in Arabidopsis. A previous report has found that SA plays an important role in protecting rice from oxidative damage (Yang et al., 2004). Here, we found that the high basal SA level in rice shoot is required for the maintenance of shoot temperatures by modulating the stomatal aperture. The Osaim1 mutant had c. 90% less SA than WT in shoot and displayed increased stomatal aperture and reduced shoot temperatures relative to those in WT. NPR1 is the master regulator of SA-mediated responses (Dong, 2004), and SA induces stomatal closure in an NPR1-dependent manner in Arabidopsis (Ou et al., 2022). In rice, however, both OsNPR1 and OsWRKY45 play a crucial role in SA signaling (Shimono et al., 2007). Our results demonstrate that an OsWRKY45, rather than OsNPR1, plays a critical role in SA-regulated stomatal aperture. H2O2 is an important signal molecule that induces stomatal closure (Waszczak et al., 2018). We have found that OsWRKY45-dependent SA signaling pathway may control stomata aperture by regulating H2O2 accumulation. H2O2 level was decreased in the Osaim1, Oswrky45 single and Osaim1 Oswrky45 double mutants. Salicylic acid treatment could rescue the H2O2 accumulation defect of Osaim1, but not the Oswrky45 and Osaim1 Oswrky45 mutants, suggesting that OsWRKY45 acts downstream of SA to regulate H2O2 accumulation in rice. Furthermore, we found that the expression of OsDCA1 and OsDST, negative regulators of stomatal closure via regulating H2O2 homeostasis, increased in the Osaim1 and Oswrky45 mutants. These results suggest that the high basal SA level in rice shoot may maintain the steady-state stomatal aperture through OsWRKY45-H2O2 signaling pathway.
In addition to its crucial role in plant biotic stress response, SA plays an important role in plant response to abiotic stresses, such as salt, drought, and cold (Rivas-San Vicente & Plasencia, 2011). Although many studies have analyzed the effect of SA on abiotic stresses in plants, its role and molecular mechanism in abiotic stress tolerance in rice are still poorly understood. Steady-state stomatal aperture is a key component in plant adaption to abiotic stress (Cui et al., 2015; Murata et al., 2015; Sukiran et al., 2020). Therefore, the high basal SA level in rice shoots may maintain the steady-state stomatal aperture to increase the abiotic stress tolerance. Indeed, we have found that Osaim1 and Oswrky45 are more sensitive to NaCl and PEG treatment, and exogenous SA could restore this phenotype of Osaim1, but not Oswrky45. These results support that steady-state stomatal aperture maintained by the OsWRKY45-dependent SA signaling pathway is crucial for abiotic stress tolerance in rice. As sessile organisms constantly exposed to diverse environmental challenges, including biotic and abiotic stress, plants have evolved complex molecular mechanisms to increase their adaptation and fitness. In this study, we have established that OsWRKY45 is a key player in SA-mediated abiotic stress adaptation. Previous studies have found that OsWRKY45 plays a crucial role in SA-mediated defense signaling by activating redox-related genes in rice (Shimono et al., 2007). Regulation of stomatal aperture also contributes to plant defense by restricting pathogen entry into plants (Melotto et al., 2008). Thus, OsWRKY45 is a critical regulator of stomatal closure in the SA signaling pathway in plant responses to both biotic and abiotic stresses.
In summary, we have shown here that the OsAIM1-dependent PAL pathway is essential for the biosynthesis of high basal levels of SA in rice shoot. We have also found that the high basal SA level in rice shoot is required for maintaining the steady-state stomatal aperture through an OsWRKY45-dependent pathway, which is important for enhanced plant tolerance to abiotic stresses.
Acknowledgements
This work was supported by the National Natural Science Foundation (32130096, 31800246), the Youth innovation Program of Chinese Academy of Agricultural Sciences (no. Y2022QC14), and Natural Science Foundation of Tianjin (20JCYBJC00260). KY was supported by the Agricultural Science and Technology Innovation Program of Chinese Academy of Agricultural Sciences.
Competing interests
None declared.
Author contributions
KY, LX, XX and JL conceived the project and designed the research. LX, HZ, JW, XW, XJ, LW, ZX and RL performed the experiments. LX, HZ, KJ, ZC, JL, XX and KY were involved in analyzing the data and preparing the manuscript. LX, HZ and JW contributed equally to this work.
Open Research
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
