ENHANCED DISEASE SUSCEPTIBILITY 1 promotes hydrogen peroxide scavenging to enhance rice thermotolerance

Abstract

Heat stress is a major factor limiting the production and geographic distribution of rice (Oryza sativa), and breeding rice varieties with tolerance to heat stress is of immense importance. Although extensive studies have revealed that reactive oxygen species (ROS) play a critical role in rice acclimation to heat stress, the molecular basis of rice controlling ROS homeostasis remains largely unclear. In this study, we discovered a novel heat-stress-responsive strategy that orchestrates ROS homeostasis centering on an immune activator, rice ENHANCED DISEASE SUSCEPTIBILITY 1 (OsEDS1). OsEDS1, which confers heat stress tolerance, promotes hydrogen peroxide (H₂O₂) scavenging by stimulating catalase activity through the OsEDS1-catalase association. The loss-of-function mutation in OsEDS1 causes increased sensitivity to heat stress, whereas the overexpression of OsEDS1 enhances thermotolerance. Furthermore, overexpression lines greatly improved rice tolerance to heat stress during the reproductive stage, which was associated with substantially increased seed setting, grain weight, and plant yield. Rice CATALASE C (OsCATC), whose activity is promoted by OsEDS1, degrades H₂O₂ to activate rice heat stress tolerance. Our findings greatly expand our understanding of heat stress responses in rice. We reveal a molecular framework that promotes heat tolerance through ROS homeostasis regulation, suggesting a theoretical basis and providing genetic resources for breeding heat-tolerant rice varieties.

A loss-of-function mutation in an immune activator gene disrupts the protein's association with catalase, reducing catalase activity and increasing rice heat sensitivity.

Introduction

Environmental temperature is one of the most crucial factors governing the seasonal growth and geographic distribution of plants, and plants have evolved sophisticated mechanisms and strategies to acclimate to seasonal temperature changes (Li et al. 2018). Heat stress caused by rapid global warming has become a serious threat to many aspects of human health, including compromising crop yields and food security (Bohra-Mishra et al. 2014; IPCC 2014). Rice (Oryza sativa), a staple food feeding over half of the world's population, is a heat-stress-sensitive plant that exhibits slow seedling growth, yellowing, stunting, withering, drying, and ultimately death under heat stress (Xu et al. 2021). Therefore, it is crucially needed to breed heat-tolerant rice varieties which can reduce grain yield losses caused by heat stress and minimalize the constraint of rice cultivation area. Rice has adapted to different environmental temperatures during natural selection and human domestication. There are only 2 species of cultivated rice, Asian rice (O. sativa) and African rice (Oryza glaberrima). Asian rice consists of 2 geographically and genetically distinct subspecies, indica/xian and japonica/geng, which are distinct in their ability to tolerate temperature stress. Indica/xian rice is more high-temperature tolerant and mainly distributed in tropical and subtropical regions (Lv et al. 2016; Xu et al. 2020). Various mechanisms have been deployed by rice to reduce heat stress-caused damage. It has been shown that the natural variations of Slender Guy 1 (SLG1) confer heat tolerance by regulating thiolated tRNA levels in indica/xian rice (Xu et al. 2020). African rice has developed delicate mechanisms for adapting to high temperatures. Recently, several major quantitative trait loci for thermotolerance in African rice have been detected. Thermo-tolerance 1 (TT1), encoding an α2 subunit of the 26S proteasome involved in efficiently eliminating cytotoxic denatured proteins, contributes to the adaptation of rice plants under distinct climatic temperatures (Li et al. 2015). TT2, encoding a Gγ subunit involved in wax retention at high temperatures, negatively regulates rice thermotolerance (Kan et al. 2022). TT3 consists of 2 genes, TT3.1 encoding a plasma membrane-localized E3 ligase which functions as a potential thermosensor, and TT3.2 encoding a chloroplast precursor protein and functioning as a chloroplast quality controller. Upon heat stress, TT3.1 translocates to the endosomes and ubiquitinates TT3.2 for vacuolar degradation which is essential for protecting chloroplasts from heat stress (Zhang et al. 2022). Thus, achieving crop productivity and food security under everlasting climate warming requires a better understanding of molecular mechanisms of heat stress resistance in crop plants.

Extensive studies show the dramatically increased generation of reactive oxygen species (ROS), including superoxide anion radical (O₂•−), hydrogen peroxide (H₂O₂), and hydroxyl radical (HO·), and their derivatives, is one of the key events that occur during heat stress (Waszczak et al. 2018; Xu et al. 2021; Mittler et al. 2022). Heat triggers ROS burst possibly owing to the upregulation of Respiratory Burst Oxidase Homolog (RBOH) gene expression and impairing the activities of antioxidant enzymes, especially superoxide dismutase and catalase (Sailaja et al. 2015; Fu et al. 2016; Zhang et al. 2018; Zhao et al. 2018a). Overaccumulation of ROS disturbs its homeostasis and leads to oxidative plant cell damage (Fang et al. 2015; Qiao et al. 2015; Suriyasak et al. 2017; Niu and Xiang 2018; Zhao et al. 2018a, 2018b). Thus, ROS homeostasis must be tightly controlled (Waszczak et al. 2018; Mittler et al. 2022). Heat shock induces a transient increase in cytosolic Ca²⁺ which mediates signaling cascades that play pivotal roles in maintaining ROS homeostasis (Saidi et al. 2009; Zheng et al. 2012; Cui et al. 2020; Mittler et al. 2022). Plasma membrane-localized nucleotide-gated ion channels (CNGCs) are known as Ca²⁺ channels (Nawaz et al. 2014). OsCNGC14 and OsCNGC16 were recently shown to modulate heat-induced Ca²⁺ influx, and both oscngc14 and oscngc16 mutants are heat sensitive and associated with disturbed heat-induced Ca²⁺ influx and ROS overaccumulation (Cui et al. 2020). Another membrane-associated Ca²⁺ channel, Annexin 1 (ANN1), also contributes to heat-induced cytosolic Ca²⁺ accumulation and plays an important role in regulating ROS levels by promoting the activities of superoxide dismutase and catalase (Qiao et al. 2015; Wang et al. 2015). OsCATB whose stability is regulated by SEMI-ROLLED LEAF 10 (SRL10) to enhance H₂O₂ scavenging confers thermotolerance (Wang et al. 2023). However, how plants keep ROS at low concentrations in response to heat stress is largely unknown.

Enhanced disease susceptibility 1 (EDS1), a positive regulator of plant immunity, was originally characterized in Arabidopsis thaliana (Parker et al. 1996). As a conserved triacylglycerol-like protein, EDS1 association with its partners to transduce immune signals has been well documented (Liu et al. 2002; Peart et al. 2002; Hu et al. 2005; Gao et al. 2010; Wagner et al. 2013; Wang et al. 2014; Zhang et al. 2016; Chakraborty et al. 2018; Chen et al. 2018; Ke et al. 2019). EDS1-node-mediated signal transduction plays a role in abiotic stress (Wituszyńska et al. 2013, 2015), but the role of EDS1 regulatory in heat stress remains unknown. In this study, we characterized OsEDS1's role in rice thermotolerance by combining genetic, molecular, and physiological analyses. These analyses suggest that OsEDS1 positively regulates rice thermotolerance by interacting with and promoting catalase-mediated H₂O₂ scavenging activities to control ROS homeostasis. Our findings greatly expand our understanding of heat stress responses in rice and might provide genetic resources for breeding heat-tolerant rice varieties.

Result

Mutation in OsEDS1 increased rice high-temperature sensitivity

When grown in a paddy field in the rice growing season under natural high-temperature conditions, oseds1 plants (Ke et al. 2019) showed heat-sensitive like phenotypes: leaves and tillers in oseds1 plants were withered, yellow, drying, and almost dead, whereas those of the wild type remained green and alive (Supplemental Fig. S1A; Xu et al. 2021; Zhang et al. 2022). In line with the visible phenotypes, oseds1 leaves accumulated a lower level of chlorophyll (Supplemental Fig. S1B). To further verify the role of OsEDS1 in heat stress response, we first assessed its transcriptional patterns in the shoots of 2-wk-old seedlings in response to heat treatment in detail. The result showed OsEDS1 expression was induced to a maximum of 5.21-fold at 24 h of heat treatment (Fig. 1E). Then, the oseds1 seedlings were subjected to heat treatment and showed increased heat sensitivity (Fig. 1, A and F). The survival rate of oseds1 plants was 33.6%, which was significantly lower than that of the wild type (73.4%). In addition, since ion leakage is an indicator of plasma membrane damage caused by various stresses (Xu et al. 2021), we examined changes in ion leakage. After heat stress, the ion leakage of oseds1 seedlings was significantly increased compared with the wild type (Fig. 1J). Similar results were obtained in an independent biological repeat (Supplemental Fig. S2). These results suggested that oseds1 plants show increased sensitivity to heat stress.

Figure 1.

OsEDS1 positively regulated rice heat tolerance. Data represent mean ± SE (n = 3 to 4). **P < 0.01 and *P < 0.05 analyzed by Student's t-test. Different letters above the bars indicate differences by a multiple-range test at P < 0.05. ZH11, the wild type. Bars = 5 cm. A–D) Phenotypes of ZH11, oseds1(A), E^WT(B), E^S143L(C), and OsEDS1-oe(D) seedlings before and after heat treatment and recovery at the seedling stage. E)OsEDS1 expression analysis response to heat treatment. The expression level at ck was set to 1. ck, before treatment. F–I) Statistical analysis of ZH11, oseds1(F), E^WT(G), E^S143L(H), and OsEDS1-oe(I) seedlings survival rate after heat treatment and recovery. J–M) Ion leakage analysis of ZH11, oseds1(J), E^WT(K), E^S143L(L), and OsEDS1-oe(M) seedling leaves after heat treatment.

To further verify that the increased sensitivity to heat stress was caused by the mutation of OsEDS1, the oseds1 mutant was complemented by OsEDS1 cDNA driven by its native promoter, referred to as E^WT (Ke et al. 2019). After treatment, the sensitivity of E^WT to heat stress was comparable to that of the wild type but was less than that of oseds1 seedlings in terms of survival rate and ion leakage (Fig. 1, B, G, and K and Supplemental Fig. S3, A and C).

EDS1 is a putative triacylglycerol lipase. The serine residue at position 143 of the OsEDS1 protein is embedded in the motif similar to the GXSXG motif (X representing any amino acid) of eukaryotic lipases catalytic triad, and the conserved S143 residue is not required for the OsEDS1-mediated rice–pathogens interactions (Brady et al. 1990; Ke et al. 2019). To verify the role of S143 in rice heat stress response, we treated E^S143L with high temperature, which was the oseds1 mutant complemented by the mutated OsEDS1^S143L cDNA (E^S143L) driven by its native promoter (Ke et al. 2019). Similar to E^WT, the sensitivity of E^S143L was comparable to that of the wild type but was less than that of oseds1 (Fig. 1, C, H, and L and Supplemental Fig. S3, B and C). Put together, these results demonstrated that OsEDS1 was the gene responsible for the phenotype of oseds1 plants, and the conserved S143 residue appeared not to be required for the OsEDS1-mediated heat tolerance.

The increased sensitivity of the oseds1 mutant was associated with reduced catalase activity and increased H₂O₂ accumulation

Heat stress-caused impairment of plant cells is frequently associated with disturbed physiological and metabolic processes, including membrane damage and ROS accumulation (Xu et al. 2021). To assess whether the oseds1 seedlings also had excessive levels of ROS, we stained the leaves of oseds1 and the wild-type seedlings with 3,3′-diaminobenzidine (DAB). Before treatment, the mutant had a lower accumulation of H₂O₂, whereas a higher accumulation of H₂O₂ relative to the wild type was observed after heat treatment (Fig. 2A). We further quantified H₂O₂ content response to heat stress in detail. Consistent with the staining assays, oseds1 accumulated less H₂O₂ before treatment, and high temperature promoted H₂O₂ accumulation in oseds1 and the wild-type seedlings, but H₂O₂ accumulated to higher levels in oseds1 plants than in the wild type (Fig. 2B). The E^WT and E^S143L plants accumulated similar levels of H₂O₂ as the wild type, but less than the oseds1 plants (Fig. 2C).

Figure 2.

OsEDS1-mediated heat stress response was associated with catalase activity and H₂O₂ accumulation. Data represent mean ± SE (n = 3). **P < 0.01 analyzed by Student's t-test. Different letters above the bars indicate differences by a multiple-range test at P < 0.05. ZH11, the wild type. FW, fresh weight. A) Visible detection of H₂O₂ in ZH11 and oseds1 leaves by DAB staining before and after heat treatment. Bars = 1 cm. B) Measurement of H₂O₂ in ZH11 and oseds1 plants before and of heat treatment. C and D) Measurement of H₂O₂ levels in ZH11, oseds1, E^WT and E^S143L(C), and OsEDS1-oe(D) plants at 6 h of heat treatment. E) Measurement of catalase activity of ZH11 and oseds1 plants before and of heat treatment. F and G) Measurement in vivo catalase activity of ZH11, oseds1, E^WT and E^S143L(F), and OsEDS1-oe(G) plants at 6 h of heat treatment.

Of the ROS-scavenging enzymes, catalase is a highly conserved enzyme catalyzing the breakdown of H₂O₂ into water and oxygen and plays a key role in removing excessive amounts of H₂O₂ (Mhamdi et al. 2010). To investigate whether increased H₂O₂ accumulation was associated with catalase activity reduction in oseds1 plants, we detected catalase activity in vivo. Results showed that oseds1 plants had higher catalase activity before treatment, but the catalase activity of oseds1 plants was lower than that of the wild type after heat treatment (Fig. 2E). E^WT and E^S143L plants had similar in vivo catalase activity as the wild type of heat treatment (Fig. 2F). Collectively, these results demonstrate that mutation in OsEDS1 reduces catalase activity resulting in high levels of H₂O₂ accumulation which leads to increased heat sensitivity.

OsEDS1 interacted with OsCATB and OsCATC

There are 3 catalase genes OsCATA, OsCATB, and OsCATC in the rice genome, and OsCATB and OsCATC have been proven to play key roles in eliminating H₂O₂ (Gao et al. 2021; You et al. 2022; Wang et al. 2023). To test how OsEDS1 regulates catalase activity, we first measured the expression patterns of catalase gene (OsCATs) response to heat stress. As shown in Supplemental Fig. S4, these OsCATs showed similar induced expression patterns that reached a maximum at about 4 h of heat treatment, which is consistent with our observations that heat stress promotes catalase activities in the wild type (Fig. 2E). We then compared the heat-induced expression patterns of OsCATs in oseds1 plants and the wild type and found that they only showed a slight difference at 6 h for OsCATA and at 3 h for OsCATC; overall, there was no major difference between the wild type and oseds1 under heat treatment (Supplemental Fig. S5). These results suggested that OsEDS1 effects on catalase activity may not be through modulating OsCATs expression.

We then checked whether OsEDS1 could physically interact with OsCATs. We first carried out yeast two-hybrid assays. Results showed that OsEDS1 and OsEDS1^S143L interacted with OsCATB and OsCATC (Fig. 3A and Supplemental Fig. S6). We verified OsEDS1–OsCATB/C interaction in vivo by split-luciferase complementation assays (Fig. 3C). We further tested whether OsEDS1 could directly interact with OsCATB/C by performing in vitro pull-down assays to validate OsEDS1–OsCATC interaction. The analysis showed that maltose-binding protein (MBP)-tag OsEDS1 (MBP-OsEDS1) directly interacted with glutathione-S-transferase (GST)-tag OsCATC (GST-OsCATC) in vitro (Fig. 3B). The interactions between EDS1 and its signaling partners were found in multiple cellular compartments (Feys et al. 2005; Rietz et al. 2011; Czarnocka et al. 2017; Ke et al. 2019). OsCATB is mainly localized in the peroxisomes (Gao et al. 2021). OsCATC localized in peroxisomes where it interacted with its partner (You et al. 2022). To test where the interaction between OsEDS1 and OsCATB/C happened, the bimolecular fluorescence complementation assays were conducted, and results showed that the interaction occurred at the region where peroxisomes were located (Fig. 3D). Taken together, these results allow us to conclude that OsEDS1 directly interacts with OsCATB/C.

Figure 3.

OsEDS1 interacted with OsCATB/C. A) Yeast two-hybrid assays showed OsEDS1 interacted with OsCATB/C. Yeast cells that contain both pGBKT7-p53 and pGADT7-T and should grow on SD-LWHA were set as a positive control, and those containing pGBKT7-Lam and pGADT7-T and should not grow on SD-LWHA were set as a negative control. SD-LW, SD-Leu/Trp. SD-LWHA, SD-Leu/Trp/His/Ade. B) Protein pull-down assays for detection of MBP-OsEDS1 and GST-OsCATC interaction. An anti-GST antibody was used to detect the output protein. An anti-MBP antibody was used to detect the input protein. IB, immunoblotting. C and D) Split-luciferase complementation (C) and bimolecular fluorescence complementation (D) assays in N. benthamiana for detecting the interaction between OsEDS1 and OsCATB/C. PEX, OsPEX14-mCherry was used as a peroxisome marker. Bar = 1 cm for (C), and Bars = 10 μm for (D).

OsCATB and OsCATC play negative roles in rice immunity by scavenging H₂O₂, and both oscatb and oscatc mutants accumulate excessive amounts of H₂O₂ (Lin et al. 2012; Gao et al. 2021; You et al. 2022). OsCATB plays a positive role in rice heat stress response (Wang et al. 2023). To test whether OsCATC plays a role in rice heat stress response, we treated oscatc seedlings (You et al. 2022) with high temperature, and oscatc plants displayed heat-sensitive phenotypes compared with the wild-type plants (Fig. 4, A to C). The oscatc seedlings succumbed to heat stress (the survival rate was 0.0%), while most of the wild-type plants survived the heat treatment (the survival rate was 98.6%). The ion leakage of oscatc seedlings was significantly increased compared with the wild-type plants (Fig. 4C). Consistently, the oscatc plants accumulated a higher level of H₂O₂, but lower catalase activity than the wild type (Fig. 4, D and E), indicating that the mutation of OsCATC increased rice heat sensitivity associated with reduced catalase activity and increased H₂O₂ level. These data suggested that similar to OsCATB, OsCATC also plays a positive role in rice heat stress response.

Figure 4.

OsCATC positively regulated rice heat stress tolerance. NPB, the wild type. Data represent mean ± SE (n = 3). **P < 0.01 analyzed by Student's t-test. A) Phenotypes of NPB and oscatc seedlings before and after heat treatment and recovery at the seedling stage. Bars = 5 cm. B) Statistical analysis of NPB and oscatc seedlings’ survival rate after heat treatment and recovery. C) Ion leakage analysis of NPB and oscatc seedling leaves after heat treatment. D and E) Measurement of catalase activity D), and H₂O₂ levels E) in NPB and oscatc seedlings at 6 h of heat treatment.

OsEDS1 stabilized OsCATC and stimulated its H₂O₂ eliminating activity

OsCATB/C were stabilized by their partners (You et al. 2022; Wang et al. 2023). To investigate whether OsEDS1 stabilizes OsCATB/C, we conducted in vitro degradation assays with GST-OsCATC. There was no obvious degradation of OsCATC observed and the stability was comparable when it was incubated with buffer, MBP-OsEDS1, or MBP (Fig. 5, F and H). However, the degradation of OsCATC was faster in oseds1 proteins than in the wild-type proteins in semi-in vivo degradation assays (Fig. 5I and Supplemental Fig. S7). In vivo degradation assays also showed that the stability of OsCATC was comparable in E^WT, E^S143L, and the wild type, but was higher than in oseds1 (Fig. 5G). These results showed that OsEDS1 interacted with and stabilized OsCATC.

Figure 5.

The effect of OsEDS1 on CATC. Data represent mean ± SE (n = 3). Different letters above the bars indicate differences by a multiple-range test at P < 0.05. ZH11 and NBP, the wild type. A) Analysis of the effect of OsEDS1 on OsCATC activity; 2 μg purified GST-GFP or GST-OsCATC fusion proteins were mixed with differing amounts of MBP or MBP-OsEDS1 fusion proteins for in vitro catalase assay. B) Analysis of in vitro OsCATC activity GST-OsCATC incubated with proteins extracted from ZH11, oseds1, E^WT, and E^S143L plants at 6 h of heat treatment, and then GST-OsCATC was purified by immunoprecipitation using an anti-GST antibody. C and D) Analysis of semi-in vivo catalase activity using 3 μg total proteins extracted from ZH11 and oseds1 plants (C), and from NPB and oscatc plants (D) at 6 h of heat treatment in the presence of differing amounts of MBP or MBP-OsEDS1 fusion proteins. E) Analysis of in vivo catalase activity using total proteins extracted from protoplasts derived from NPB and oscatc seedlings transfected with OsEDS1-cYFP and cYFP, respectively. F and H) Degradation assays with GST-OsCATC incubated with MBP-OsEDS1 and MBP, respectively. ZH11 proteins were used as a positive control (F), and “*” indicates target bands (H). G) OsEDS1-promoted OsCATC stability in vivo. OsCATC-GFP was transfected into protoplasts derived from oseds1, E^WT, E^S143L, and the wild type, respectively. Samples were collected at 1 h of treatment. I) Semi-in vivo degradation assay with GST-OsCATC incubated with oseds1 and ZH11 proteins, respectively.

Increased OsCATC stability due to the OsEDS1 association may improve the OsCATC-mediated H₂O₂ scavenging ability. Thus, we investigated the effect of OsEDS1 on the activity of OsCATC. MBP-OsEDS1 and MBP-OsEDS1^S143L promoted OsCATC-mediated H₂O₂ scavenging activity in vitro, which increased gradually in accordance with the amounts of MBP-OsEDS1 (Fig. 5A). Additionally, we also tested whether the reduced catalase activity of oseds1 seedlings was caused by the absence of OsEDS1 by analyzing the catalase activity of total proteins extracted from the wild-type and oseds1 seedlings after treatment in the presence of differing amounts of MBP-OsEDS1. The catalase activity of the oseds1 mutant was rescued by exogenous MBP-OsEDS1, and it was further promoted by additional MBP-OsEDS1 in the oseds1 mutant and the wild type (Fig. 5C). Further, we incubated GST-OsCATC with proteins extracted from ZH11, oseds1, E^WT, and E^S143L plants, and purified the GST-OsCATC through immunoprecipitation with an anti-GST antibody, respectively. The H₂O₂ degradation activities of GST-OsCATC after being incubated with ZH11, E^WT, and E^S143L proteins were comparable, but higher than that of GST-OsCATC after being incubated with oseds1 proteins (Fig. 5B). However, in vivo and semi-in vivo assays showed that OsEDS1-promoted catalase activity was largely suppressed in the oscatc mutant (Fig. 5, D and E). The increased catalase activity in the oscatc mutant might be caused by OsEDS1-promoted stability and activity of OsCATB. These results suggested that OsEDS1 interacted with OsCATB/C and promoted their catalase activities both in stabilization-dependent and -independent manners.

Assays showed OsEDS1 increased catalase activity (Fig. 5, A to E); this prompted us to hypothesize that the upregulation of OsEDS1 expression can increase catalase activity and reduce H₂O₂ accumulation, thus enhancing rice heat tolerance. To test this hypothesis, we generated OsEDS1-overexpression lines with OsEDS1 cDNA controlled by the Cauliflower mosaic virus 35S promoter in Zhonghua 11 (OsEDS1-oe1 and OsEDS1-oe2; Supplemental Fig. S8). The OsEDS1-oe seedlings displayed reduced heat-sensitive phenotypes (Fig. 1, D, I, and M). The wild-type seedlings had a survival rate of 35.4%, which is significantly lower than those of 60.9% and 67.2% for OsEDS1-oe seedlings, respectively. The ion leakage of OsEDS1-oe seedlings was significantly less than that of the wild type. Consistently, the OsEDS1-oe seedlings accumulated lower levels of H₂O₂, but higher catalase activity (Fig. 2, D and G), indicating overexpressing OsEDS1 increased rice heat stress tolerance associated with increased catalase activity and reduced H₂O₂ accumulation. Collectively, these data may suggest positive roles of OsEDS1 in rice thermotolerance potentially through interacting with OsCATB/C and promoting their stability and activities.

OsEDS1 enhanced heat stress tolerance at the reproductive stage

During the entire growth, the rice plant is highly sensitive to heat stress and hyper-sensitive at the reproductive stage (Xu et al. 2021). Thus, improving heat tolerance at the reproductive stage is critical for reducing heat-stress-caused yield losses. We field-tested whether OsEDS1 confers heat tolerance at the reproductive stage. Under normal growth conditions, the oseds1 and OsEDS1-oe plants showed similar gross morphologies with the wild type (Supplemental Fig. S9). Under heat stress, leaves and tillers in the wild-type and OsEDS1-oe plants remained green and alive, while those of the oseds1 plants were drying and almost dead (Fig. 6, A and B). The oseds1 plants exhibited defects in seed setting, grain weight, and plant yield (Fig. 6, C to E). The seed setting was 32.7%, 50-grain weight was 1.0 g, and plant yield was 10.7 g for oseds1, while these for the wild type were 59.0%, 1.2 g, and 21.6 g, respectively, suggesting the oseds1 mutant was also sensitive to natural heat stress at the reproductive stage. More importantly, the OsEDS1-oe plants showed some advantages in seed setting, grain weight, and plant yield (Fig. 6, C to E). Compared with the wild type, OsEDS1-oe plants showed increased seed setting (>35%), grain weight (>7%), and plant yield (>17%), suggesting overexpressing OsEDS1 reduced sensitivity to heat stress at the reproductive stage. We also performed high-temperature treatment at the adult stage in the growth chamber and the findings were consistent with the results obtained under natural high-temperature conditions (Supplemental Fig. S10). These results showed OsEDS1-promoted heat tolerance at the reproductive stage. Together with these findings, OsEDS1 could be a candidate genetic resource for improving rice heat tolerance at both the seedling and the reproductive stages.

Figure 6.

OsEDS1-promoted rice heat stress tolerance at the adult stage in the field. Data represent mean ± SE (n = 31 to 43 for C; n = 6 for D and E). **P < 0.01 analyzed by Student's t-test. ZH11, the wild type. A) Gross morphology of ZH11, oseds1, and OsEDS1-oe plants under natural high-temperature conditions at the adult stage. Bars = 20 cm. B) Daily highest temperature in Wuhan (29°58′–31°22′N, 113°41′–115°05′E) from June 01 2022 to August 31 2022. Data obtained from http://www.tianqihoubao.com/lishi/wuhan.html. C–E) Comparison of seed setting (C), grain weight (D), and plant yield (E) among ZH11, oseds1, and OsEDS1-oe plants. F) A proposed working model of the OsEDS1-mediated ROS homeostasis in heat stress response. Under heat stress, the transcripts of OsEDS1 and OsCATB/C are induced, leading to the accumulation of OsEDS1–OsCATB/C protein complexes, which then scavenge heat-triggered H₂O₂ accumulation, and in turn, control ROS at low levels to confer thermotolerance in rice. ROS, reactive oxygen species. Arrows and T lines indicate promoting and inhibiting effects, respectively. Solid and dotted lines indicate direct action and indirect actions, respectively.

Discussion

Heat stress is one of the most crucial abiotic factors adversely affecting plant growth and crop production (Li et al. 2018). Tremendous progress has been made in the understanding of the genes, pathways, and regulatory networks underlying heat stress response, revealing the complexities of mechanisms and strategies deployed by rice (Li et al. 2015; Lv et al. 2016; Xu et al. 2020; Kan et al. 2022; Zhang et al. 2022). ROS burst, triggered by biotic and abiotic stresses, leads to oxidative plant cell damage, thus ROS homeostasis must be tightly controlled (Waszczak et al. 2018; Xu et al. 2021; Mittler et al. 2022). Ca²⁺ signaling plays a pivotal role in maintaining heat-induced ROS homeostasis (Saidi et al. 2009; Zheng et al. 2012; Cui et al. 2020; Mittler et al. 2022). Recent studies demonstrated that OsCATB and OsCATC eliminate ROS to repress rice immunity. The Magnaporthe oryzae effector AvrPiz-t and susceptibility protein RESISTANCE OF RICE TO DISEASES1 (ROD1) interact with and raise the activity of OsCATB to degenerate ROS (Gao et al. 2021); E3 ubiquitin ligase AvrPiz-t Interacting Protein 6 (APIP6) interacts with OsCATC and reduces its activity by ubiquitinating it for degradation (You et al. 2022). OsCATB whose stability is maintained by SRL10 through interaction confers thermotolerance by scavenging H₂O₂ (Wang et al. 2023). However, little is known about the function of OsCATC in controlling ROS homeostasis or whether there is a protein interacting with and regulating its stability and activity in rice heat stress response. In this study, we revealed OsCATC degenerating H₂O₂ and triggering rice thermotolerance. First, the transcripts of catalase genes and catalase activity were upregulated by heat stress. Second, OsCATC could positively regulate heat stress in rice, and its loss-of-function mutant oscatc displayed heat-sensitive phenotypes. Third, under heat stress, the catalase activity was reduced in the oscatc mutant, while the H₂O₂ level in the oscatc mutant was higher than in the wild type.

There is growing evidence that for a comprehensive insight into the function of genes, it is crucial to assess their functionalities under a wide range of stress conditions. EDS1 is a pleiotropic gene playing critical roles in multiple aspects of the plant stress response, which encodes a conserved triacylglycerol-like protein with unknown biochemical activities (Wagner et al. 2013). EDS1 is an important positive regulator of immunity in plants. The loss-of-function mutations in EDS1 suppress plant immune response (Liu et al. 2002; Peart et al. 2002; Hu et al. 2005; Gao et al. 2010; Wagner et al. 2013; Wang et al. 2014; Zhang et al. 2016; Chakraborty et al. 2018; Chen et al. 2018; Ke et al. 2019). EDS1-node-mediated signal transduction playing roles in abiotic stress has also been reported in Arabidopsis (Wituszyńska et al. 2013, 2015). Previous analysis showed OsEDS1 was selected in high-temperature tolerance indica/xian rice subspecies during natural selection and human domestication (Lv et al. 2016; Xu et al. 2020; Chen et al. 2022). In this study, we characterized OsEDS1 for its role in rice thermotolerance. Heat stress upregulated OsEDS1 expression. The genetic evidence exhibited that OsEDS1 could positively regulate heat stress in rice seedlings and the reproductive stages. OsEDS1 loss-of-function mutant oseds1 displayed heat-sensitive phenotypes, while the overexpression of OsEDS1 increased the thermotolerance of rice. Biochemical and physiological analyses showed OsEDS1 conferred thermotolerance was associated with its function in modulating catalase activity. Further assays illustrated that OsEDS1 interacted with and stabilized OsCATC in vivo and facilitated their H₂O₂ scavenging activities in vivo and in vitro. Thus, OsEDS1-modulated OsCATC H₂O₂ scavenging activity was partially independent of its stabilization function. The conserved S143 residue embedded in the motif similar to the GXSXG motif of eukaryotic lipases catalytic triad was not required for the OsEDS1-mediated rice thermotolerance. We did not have sufficient genetic material required for the overexpression of OsEDS1 in oscatb/c mutant background; therefore, the genetic relationship between OsEDS1 and OsCATB/C requires further investigation.

This report associates biochemical activity of EDS1 with enzyme activity, although the molecular base underlying how OsEDS1 promotes catalase stability and activity remains elusive. Our findings greatly expand our understanding of gene function. Further illustration of how OsEDS1 translocates into the peroxisomes for interacting with and facilitating the functions of OsCATB/C will provide valuable knowledge in understanding the strategies deployed by rice to keep from oxidative damage under heat stress.

Based on the results obtained in this study and results from a previous study (Wang et al. 2023), we propose a working model to illustrate the functions of the OsEDS1–OsCATB/C hub in rice thermotolerance (Fig. 6F). Upon heat stress, OsEDS1–OsCATB/C complexes are activated through the induction of OsEDS1, OsCATB, and OsCATC expression to buffer the heat-triggered ROS burst. When OsEDS1, OsCATB, or OsCATC functions are deprived, rice plants accumulate excessive ROS that cause plant cell oxidative damage and increase sensitivity to heat stress, revealing a previously unrecognized mechanism of the plant's response to heat stress. Moreover, upregulation of OsEDS1 expression promoted rice thermotolerance and plant yield, which might provide a candidate gene that could be exploited for breeding heat-tolerant rice varieties.

Materials and methods

Plant materials and growth conditions

The oseds1, E^WT, and E^S143L lines in the Oryza sativa japonica/geng variety Zhonghua 11 background, and the oscatc mutant in the japonica/geng variety Nipponbare background were generated previously (Lin et al. 2012; Ke et al. 2019; You et al. 2022). The OsEDS1-overexpression lines were generated in Zhonghua 11 with full-length coding sequences of OsEDS1 inserted into pCAMBIA-35S-cYFP using Agrobacterium-mediated transformation of rice calli as previously described (You et al. 2022). Primers used for plasmid construction are listed in Supplemental Table S1.

All rice plants were cultivated in experimental paddy fields of Huazhong Agricultural University (30.27°N, 114.36°E) during the normal growing season in Wuhan, China, from May to October. For experiments using seedlings, plants were grown in a growth chamber under the conditions of 14 h day, 28 °C, and 60 ± 5% relative humidity followed by a 10 h night, 25 °C, and 60 ± 5% relative humidity.

Heat stress condition

To examine heat tolerance at the seedling stage, seeds were germinated and grown in a growth chamber (Ruihua HP1000GS-LED/H11, 28 °C/25 °C, day/night, 60 ± 5% relative humidity). Two weeks after germination, seedlings were treated at 44 °C, day/night (80 ± 5% relative humidity) for heat treatment (for convenience, they were collectively referred to as 2-wk-old rice seedlings treated at 44 °C for 48 h in this study, the exact number varied with the growing conditions and the rice cultivar). After 10 d of recovery, survival rates were estimated, and survival phenotypes were photographed. For identifying rice thermotolerance at the reproductive stage in laboratory conditions, rice plants were transferred from experimental paddy fields to a growth chamber (40 °C, 14 h day/35 °C, 10 h night, 80 ± 5% relative humidity) at the spikelet development stage and treated until mature.

RNA extraction and reverse transcription quantitative PCR

For heat-stress-induced expression patterns, 2-wk-old seedlings were treated at 44 °C for the indicated periods. Total RNA was extracted from shoots using TRIzol reagent (Invitrogen). cDNA was synthesized with HiScript III RT SuperMix (Vazyme, China) according to the manufacturer's protocol. For gene expression analysis, reverse transcription quantitative PCR (RT-qPCR) was performed using LightCycler 480 SYBR Green I Master (Roche, Switzerland) in the ABI 7500 Real-Time PCR System (Applied Biosystems, USA) with gene-specific primers (Supplemental Table S2). The rice actin gene was used to standardize relative RNA measures as inner control.

Ion leakage assay

Leaves from different plants for each sample were cut into 1 cm fragments and incubated in ddH₂O at room temperature with gentle shaking for 3 h before the initial conductivity of the solution was measured with a conductivity meter (DDS-307A, Shanghai Leici, China). The total conductivity of the solution was measured after incubation at 95 °C for 20 min and gentle shaking for 30 min at room temperature. The ion leakage rate was expressed as the percentage of the ratio of initial conductivity to total conductivity.

DAB staining and measurement of H₂O₂ content

Two-week-old seedlings were used for physiological measurements. Leaves before and after high-temperature treatment were collected and stained with DAB and the H₂O₂ content of the leaves was determined using an H₂O₂ assay kit (Sangon, China) according to the manufacturer's instructions.

Recombinant protein expression and purification

The full-length coding sequences of OsEDS1 and OsCATC were inserted into the pMAL or pGEXT-6p-1 vectors, respectively, and transformed into Escherichia coli strain BL21 to express MBP-OsEDS1 and GST-OsCATC. These recombinant proteins were induced by 1 mM isopropyl β-D-1-thiogalactopyranoside at 16 °C for 20 h. MBP-OsEDS1 was purified with Amylose Resin (New England Biolabs, USA) and GST-OsCATC was purified with Glutathione Agarose Resin (Thermo Scientific, USA) according to the manufacturer's instructions. Primers used for plasmid construction are listed in Supplemental Table S1.

Catalase activity assay

To analyze catalase activity, leaves from different plants for each sample were homogenized in liquid nitrogen and were extracted in 1 mL extraction buffer (150 mM NaCl, 50 mM phosphate buffer, pH7.5, 0.5% [v/v] Triton X-100, 10% [v/v] glycerol, and 1% [w/v] protease inhibitor cocktail). The supernatant was collected after being centrifuged at 12,000 rpm for 10 min at 4 °C, and protein concentration was estimated by the Bradford method according to the manufacturer's protocol and standardized using a BSA standard curve (0.25 to 4 μg μL⁻¹). The supernatant was used for catalase activity analysis with the Catalase Assay Kit (Beyotime, China) according to the manufacturer's instructions.

The activity of purified E. coli expressed GST-OsCATC was measured at 240 nm in 50 mM phosphate buffer and 10 mM H₂O₂ as the absorbance decreased and was shown as units/mg (Weydert and Cullen 2010). One unit represents the amount of enzyme that catalyzes the decomposition of 1 mM H₂O₂ per minute at 30 °C. To assess the effect of OsEDS1 on OsCATC activity, the purified MBP and MBP-OsEDS1 proteins were mixed with OsCATC for 1 h at 30 °C, and the catalase activity was then determined.

Chlorophyll measurement

About 100 mg of leaves were sampled for chlorophyll measurement. Rice leaves were homogenized and incubated in extraction solution (acetone:absolute ethanol:water, 4.5:4.5:1, v/v/v) for 12 h at 4 °C in the dark with violent shaking, and the extraction was measured spectrophotometrically at 645 and 663 nm using a SparkTMM Multimode Microplate Reader (Tecan). The total chlorophyll (mg/g) = (20.29×A₆₄₅ + 8.05×A₆₆₃) × v/m × 1000, where v is the volume of extraction solution and m is the mass of leaves.

Yeast two-hybrid assay

The full-length coding sequence of OsEDS1 was amplified from rice leaf total cDNA and was inserted into the pGBKT7. The full-length coding sequences of OsCATA, OsCATB, and OsCATC were amplified from rice leaf total cDNA and were inserted into the pGADT7, respectively. The resulting constructs and the corresponding empty vectors were co-transformed into AH109 Yeast, and the interaction was screened on SD-Trp/Leu/His/Ade medium. Primers used for plasmid construction are listed in Supplemental Table S1.

Pull-down assay

For the protein pull-down assay, MBP-OsEDS1 was incubated with Amylose Resin at 4 °C for 1 h, and then the beads were treated with 10% skimmed milk for 3 h at 4 °C to block them. The beads were washed 5 times, and GST-OsCATC was added. The incubation continued for 3 h, and the beads were washed 5 times. The beads were boiled in 1× SDS loading buffer and separated by 8% (W/V) SDS-PAGE. The anti-MBP antibody (E8032L, NEB) was used to detect the MBP-tagged protein. Anti-GST antibody (AE001, ABclonal) was used to detect pull-down protein.

Degradation assay

For in vitro degradation assays, purified GST-OsCATC recombinant protein was used and incubated with buffer, MBP-OsEDS1, and MBP, respectively. For semi-in vivo degradation assays, GST-OsCATC was incubated with total proteins from oseds1 and ZH11, respectively. For in vivo degradation assays, the Ubi-OsCATC-GFP plasmid was transfected into protoplasts derived from oseds1, E^WT, E^S143L, and ZH11, respectively. Assays were carried out as previously described (Lv et al. 2014). Reactions were terminated at indicated time points. The OsCATC protein abundance was visualized via immune detection against anti-GST antibody and anti-GFP antibody (Clonetech, JL-8), respectively.

Split-luciferase complementation assay

The full-length coding sequence of OsEDS1 was inserted into pCAMBIA-35S-nLUC. The full-length coding sequences of OsCATB and OsCATC were inserted into pCAMBIA-35S-cLUC, respectively. Constructs and the corresponding empty vectors were transferred into Agrobacterium tumefaciens strain GV1301 and grown at 28 °C overnight. The Agrobacterium cells were collected and resuspended in infiltration buffer (10 mM MgCl₂, 10 mM MES pH 5.6, 150 mM acetosyringone) to the final concentration of OD₆₀₀ = 0.5 and incubated for 2 to 3 h at 30 °C. The suspensions of a combination of corresponding empty vectors were co-infiltrated into the abaxial side of 4-wk-old Nicotiana benthamiana leaves. D-luciferin (Promega) was used as a substrate, and the LUC luminescence intensity was captured by the Tanon-5200 image system (Tanon, China). Primers used for plasmid construction are listed in Supplemental Table S1.

Bimolecular fluorescence complementation assay

The full-length coding sequence of OsEDS1 was inserted into pCAMBIA-35S-cYFP. The full-length coding sequence of OsCATB and OsCATC were inserted into pCAMBIA-35S-nYFP, respectively. Constructs and the corresponding empty vectors were transferred into A. tumefaciens strain GV1301 and grown at 28 °C overnight. The Agrobacterium cells were collected and resuspended in infiltration buffer (10 mM MgCl₂, 10 mM MES pH 5.6, 150 mM acetosyringone) to the final concentration of OD₆₀₀ = 0.5 and incubated for 2 to 3 h at 30 °C. The suspensions of a combination of corresponding empty vectors were co-infiltrated into the abaxial side of 4-wk-old N. benthamiana leaves. The fluorescence was observed by confocal microscopy (TCS SP2; Leica) with the following parameters (lasers: YFP: 488 nm, mCherry: 561 nm; intensity: YFP: 723, mCherry: 780; collection bandwidth: 111 μm; gains: 1). Primers used for plasmid construction are listed in Supplemental Table S1.

Statistical analysis

Data were processed in GraphPad Prism 8 (GraphPad Software Corporation). Comparisons between group means were performed with Student's t-test. The statistical significance of differences between control and sample treatments was assessed using the pair-wise t-test installed in the Microsoft Office Excel program (Excel version 2019, Microsoft Corporation). The multiple samples were analyzed by one-way ANOVA using Tukey's multiple comparison test in software R (the R project for Statistical Computing; https://www.r-project.org) as described by Ke et al. (2019).

Accession numbers

Sequence data were obtained from the Rice Annotation Project Database (http://rice.plantbiology.msu.edu/index.shtml) according to the following accession numbers: LOC_Os09g22450 (OsEDS1), LOC_Os02g02400 (OsCATA), LOC_Os06g51150 (OsCATB), and LOC_Os03g03910 (OsCATC).

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. oseds1 plants showing heat-sensitive like phenotypes.

Supplemental Figure S2. OsEDS1 positively regulates rice heat tolerance.

Supplemental Figure S3. Complementation assays of oseds1 mutant.

Supplemental Figure S4. OsCAT expression analysis response to heat treatment.

Supplemental Figure S5. Expression levels of OsCATA, OsCATB, and OsCATC in ZH11 and oseds1 plants under ck or heat treatment.

Supplemental Figure S6. OsEDS1^S143L interaction with OsCATB/C in yeast cells.

Supplemental Figure S7. Semi-in vivo degradation assay of GST-OsCATC.

Supplemental Figure S8. Analysis of OsEDS1 expression levels in OsEDS1-oe plants.

Supplemental Figure S9. Gross morphology of ZH11, oseds1, and OsEDS1-oe plants under normal growth conditions.

Supplemental Figure S10. OsEDS1-promoted rice heat stress tolerance at the adult stage in the growth chamber.

Supplemental Table S1. PCR primers used for the construction of vectors.

Supplemental Table S2. Primers used for RT-qPCR for gene expression analysis.

Author contributions

PF.Y. and YG.K. supervised the project, analyzed the data, and revised the manuscript. PF.Y., YG.K., M.L., and ZM.M. conceived and designed the research plans. M.L. and ZM.M. performed most of the experiments, analyzed the data, and drafted the manuscript. YR.K. and BM.Z. performed protein–protein interaction in plants. XL.G. and F.Y. performed the reverse transcription PCR and gene expression assays. All authors revised the manuscript.

Funding

This work was supported by the Foundation of Hubei Hongshan Laboratory.

Data availability

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/General-Instructions) is: Yinggen Ke (ygke@hubu.edu.cn).