Jing Chen, Keqi Fei,Weiyng Zhng, Zhiqin Wng, Jinhu Zhng,Jinchng Yng,*

aJiangsu Key Laboratory of Crop Genetics and Physiology/Co-Innovation Center for Modern Production Technology of Grain Crops,Yangzhou University, Yangzhou 225009,Jiangsu,China

bDepartment of Biology,Hong Kong Baptist University,Hong Kong,China

cSchool of Life Sciences and State Key Laboratory of Agrobiotechnology,The Chinese University of Hong Kong,Hong Kong,China

Keywords:Rice Photo-thermosensitive genetic male-sterile (PTSGMS)High temperature Brassinosteroids Pistil activity

ABSTRACT Brassinosteroids(BRs)play critical roles in a wide range of plant developmental processes.However,it is unknown whether and how BRs mediate the effect of high temperature(HT)stress during anthesis on the pistil activity of photo-thermosensitive genetic male-sterile(PTSGMS)rice(Oryza sativa L.)lines.This study investigated the question.Three pot-grown PTSGMS rice lines were subjected to HT stress during anthesis. The contents of 24-epibrassinolide (24-EBL) and 28-homobrassinolide (28-HBL), the major forms of BR in rice plants,and levels of reactive oxygen species(ROS)or antioxidants(AOS),hydrogen peroxide(H2O2), 1-aminocylopropane-1-carboxylic acid (ACC), ascorbic acid (AsA), and catalase activity in pistils,were determined.HT stress significantly reduced the contents of both 24-EBL and 28-EBL relative to those under normal temperatures, but the reduction varied by PTSGMS line.A line with higher BR contents under HT stress showed lower contents of ACC and H2O2, higher catalase activity and AsA content in pistils, and higher fertilization rate,seed-setting rate, and seed yield when the line was crossed with a restorer line, indicating that higher levels of BRs increase HT stress resistance. Applying 24-EBL, 28-HBL or an inhibitor of BR biosynthesis confirmed the roles of BRs in response to HT stress.The results suggest that BRs mediate the effect of HT stress on pistil activity during anthesis and alleviate the harm of HT stress by increasing AOS and suppressing ROS generation.

1.Introduction

“Two-line-method”hybrid rice,a hybrid produced by crossing a photo-thermosensitive genetic male-sterile (PTSGMS) rice(Oryza sativa L.) line with its restorer line, has been widely used in China and plays a vital role in the increase of rice productivity in the country, owing to its high heterosis, the absence of a need for a maintainer rice line,and the simplicity of the breeding procedure [1-4]. PTSGMS rice lines, however,are sensitive to temperature and usually lose fertility when subjected to high temperature (HT, ≥ 35 °C) during anthesis,leading to severe reduction in hybrid seed yield[4-7].

With the increase in global temperature and changes in climate, HT stress has become a major hazard in rice production, and, when it occurs during anthesis, can result in pollen sterility and consequently a low seed-setting rate and low grain yield [7-11]. These consequences stem from imbalance between phytohormones,in particular increases in ethylene, an increase in reactive oxygen species (ROS)production, and a decrease in carbohydrate synthesis and metabolism [12-16]. Previous studies of HT stress have been based mainly on conventional male-fertile rice cultivars or lines, focusing on the response of male organs to HT stress during anthesis. In contrast to male-fertile rice cultivars or lines, PTSGMS lines show total male sterility when temperatures exceed 23 or 24°C[1-5],so that HT stress during anthesis impairs their pistil fertilization. The mechanism of this impairment is unknown.

Brassinosteroids(BRs)are a group of naturally occurring plant steroid hormones comprising brassinolide, castasterone, and their various derivatives including 24-epibrassinolide (24-EBL),28-homobrassinolide (28-HBL), and 24-epicastasterone (24-epiCS), and play critical roles in a wide range of plant developmental processes[17-21].Recent work of Zhang et al.[22-24]has shown that BRs enhance antioxidant system (AOS) and energy charge during panicle development,and consequently,promoted rice spikelet differentiation and suppress spikelet degeneration.It is not known,however,whether and how BRs regulate the pistil fertilization ability of PTSGMS lines under HT stress during anthesis.

The purpose of this study was to test the hypothesis that BRs mediate the effect of HT during anthesis on the pistil fertilization ability of PTSGMS rice lines by regulating ROS levels. The experimental approach was to measure the contents of 24-EBL and 28-HBL, the major forms of BRs in rice [19,27]and the effects of chemical regulators of BR synthesis on pistil activity as assessed by ROS and AOS.

2.Materials and methods

2.1. Plant materials and cultivation

Experiments were conducted at the research farm of Yangzhou University, Jiangsu province, China (32°30′N, 119°25′E,21 m altitude) during the rice growing season (May-October)in 2016, and repeated in 2017. Three PTSGMS rice lines: Peiai 64S,Guangzhan 63S,and Shen 08S,and one restorer line 9311(as a normal pollen donor and a conventional rice cultivar),were used.The sowing date of the PTSGMS lines was May 15,and those of the restorer line were May 5,10,15,and 20 so that the flowering period of the restorer line would match those of the PTGMS lines. The seedlings were grown in a paddy field,and 26-day-old seedlings were transplanted to porcelain pots with three hills per pot and two seedlings per hill. Each pot(25 cm in diameter, 30 cm in height, and 14.72 L in volume)was filled with 16 kg sandy loam soil[Typic Fluvaquent,Etisol(U.S. taxonomy)]with organic matter 23.5 g kg?1, alkalihydrolyzable N 106 mg kg?1, Olsen-P 34.5 mg kg?1, and exchangeable K 76.3 mg kg?1. A 2-3 cm water layer above the soil surface was maintained in the pots and the soil pH was adjusted to 6.0 with 0.1 mol L?1HCl for 2 weeks before transplanting. At 3 days before transplanting, 1 g N as urea and 0.5 g K as KCl were mixed into the soil in each pot. N as urea was also applied at mid-tillering (0.5 g pot?1) and at panicle initiation(1 g pot?1) stages.All the plants were grown under open-air conditions except during the HT treatment period.

2.2. HT treatment

At the onset of heading (10% of the panicles were heading from the sheaths of flag leaves), plants grown in pots of the three PTSGMS lines were moved into four phytotrons (AGCMR, Zhejiang Qiushi Environment Co., Ltd., Zhejiang, China)for temperature treatments. Each phytotron was 6.0 m in length,3.2 m in width,and 2.6 m in height,and was equipped with a control system for temperature, light density, CO2concentration, and humidity. Both normal temperature (NT,the control)and high temperature(HT)treatments were set to correspond to the diurnal variation of air temperature under local weather conditions. During the period of 04:01-06:00,06:01-08:00, 08:01-10:00, 10:01-14:00, 14:01-16:00, 16:01-18:00,18:01-20:00, and 20:01-04:00 h (next day), the respective temperatures were 24, 27, 30, 33, 31, 29, 27, and 25 °C for the NT and 25,28,33,37,35,33,28,and 26°C for the HT treatment.The respective light densities for both NT and HT were 200,800, 1000, 1200, 1200, 800, 100, and 0 μmol m?2s?1. Relative humidity was maintained at 70%±10%and CO2concentration at 370 ± 20 μmol mol?1for both NT and HT during the treatment. The temperature treatment period lasted for 7 days until the end of plant heading (95% of panicles extended outside the sheaths of flag leaves). Each treatment and each PTSGMS line had 60 pots as replications.On the day before temperature treatment and the first day of temperature treatment, plants heading on the same day were chosen and tagged in 50 pots for physiological measurements. The pistils of each PTSGMS line from 10 pots under either NT or HT were manually fertilized with pollen from the restorer line grown under open-air conditions, for determination of fertilization rate,seed yield,and yield components of hybrids.

2.3. Sampling and measurement

From each PTSGMS line and each temperature treatment,100-150 tagged panicles were sampled on the day before temperature treatment(BT-1),the second day(DT-2),and the fourth day (DT-4) during temperature treatment, and on the day after the end of the treatment(AT-1).All panicles sampled for investigation received the same number of days of HT treatment. Sampled panicles were placed on ice and the pistils were quickly extracted with tweezers from each spikelet(flowering or not flowering) and weighed,so that the effect of HT stress on pistil activity could be determined for all spikelets on a panicle.The pistils from 20 panicles were dried in an oven for measurement of dry weight (DW), and other pistils were frozen in liquid nitrogen for 5 min and then stored at ?80 °C pending physiological measurements. Each physiological measurement was performed on five biological replicates.

24-EBL and 28-HBL were extracted and purified following Ding et al. [25]with minor modifications. Each sample containing 0.6-0.8 g fresh pistils was ground with a tissue grinder(MM400,Retsch Corp,Haan,Germany),and then 0.3 g of the ground powder was transferred into a 5-mL centrifuge tube and extracted with 2 mL acetonitrile overnight at 20 °C.The extraction, dehydration and double-layered solid-phase extraction followed Chen et al. [26]. Quantification of 24-EBL and 28-HBL was performed with a high-performance liquid chromatograph electrospray ionization-tandem mass spectrometry (HPLC-ESI-MS/MS, Thermo Fisher Scientific, Waltham,MA,USA)system following Ding et al.[25].The BR level was expressed as pmol g?1DW.

ACC levels in pistils were determined following Cheng and Lur [28]: ethylene evolved from ACC was measured in a gas chromatograph (HP5890 Series II, Hewlett Packard Corp., Palo Alto, CA, USA), and then ACC content was calculated using a transformation rate from ethylene of 92.5%±4.6%[29,30].ACC content was expressed as nmol g?1DW.The contents of H2O2and AsA in pistils were measured following Brennan and Frenkel [31]and Wang et al. [32], respectively, and expressed as μmol g?1DW. The activity of catalase in pistils was determined by monitoring the consumption of H2O2as described previously[33],and was expressed as mmol g?1DW h?1.

Fig.1- Effects of high temperature stress during anthesis on contents of 24-epibrassinolide(24-EBL,A-F)and 28-homobrassinolide(28-HBL,G-L)in pistils of photo-thermosensitive genetic male-sterile rice lines.NT and HT represent normal and high temperature, respectively.BT-1,DT-2,DT-4,and AT-1 represent one day before temperature treatment,the second day during treatment,the fourth day during treatment,and one day after the end of treatment,respectively.Vertical bars represent±standard error of the mean(n=5).Different letters above bars indicate statistical significance at P=0.05 within the same rice line and in the same year.

Fig.2-Effects of high temperature stress during anthesis on catalase activity(A-F)and ascorbic acid content(AsA,G-L)in the pistils of photo-thermosensitive genetic male-sterile rice lines.NT and HT represent normal and high temperature,respectively.BT-1,DT-2,DT-4,and AT-1 represent one day before temperature treatment,the second day during treatment,the fourth day during treatment,and one day after the end of treatment,respectively.Vertical bars represent±standard error of the mean(n=5).Different letters above the column indicate statistical significance at P=0.05 within the same rice line and in the same year.

The hybrids of PTSGMS lines crossed with 9311 (Peiai 64S/9311, Guangzhan 63S/9311, and Shen 08S/9311) were harvested at maturity (October 25-28) for determination of fertilization rate, seed yield, and yield components. The fertilization rate was defined as the percentage of fertilized spikelets relative to the total number of spikelets per pot,and the seed-setting rate was defined as the percentage of filled kernels (with specific gravity ≥1.0 g cm?3) relative to the total number of spikelets.Each measurement was of five replicates(pots).

2.4. Chemical application

The PTSGMS line Peiai 64S was pot-grown and subjected to HT stress during anthesis as described above. At the onset of anthesis, 1 μmol L?124-EBL (T1), 1 μmol L?128-HBL (T2),1 μmol L?1brassinazole (BRz, an inhibitor of BR biosynthesis)(T3),or 1 μmol L?124-EBL+1 μmol L?1BRz(T4)(all from Sigma Chemical Co.,St Louis,MO,USA)were applied to the panicles using a paintbrush dipped in the solutions or by careful injection from above into the boot with a 10-mL syringe if a panicle was in the flag leaf sheath, at the rate of 4 mL per panicle. This method reduces the effect of chemicals on leaves compared to that of spraying chemicals onto panicles.All the solutions contained ethanol at a final content of 0.05%(v/v). Control plants were treated with the same volume of deionized water containing the same content of ethanol.Each chemical treatment had 10 pots as replicates. Plants of five pots from each treatment were used for measuring the contents of 24-EBL, 28-HBL, AsA, ACC, and H2O2and catalase activity in the pistils on the first day and the second day after chemical application. Pistils of another five pots for each treatment were fertilized with pollen from the restorer line 9311 for the determination of fertilization rate,seed yield,and yield components. The methods used for these determinations were the same as described above.

2.5. Statistical analysis

Analysis of variance was performed using the SAS/STAT 9.2 statistical analysis package (SAS Institute, Cary, NC, USA).Data from each sampling date were analyzed separately.Means were tested by least significant difference at the P =0.05 level (LSD0.05). Correlations were calculated to evaluate the relationships of the levels of 24-EBL and 28-HBL with those of AsA, ACC, and H2O2and catalase activity, and the relationships of these physiological parameters with the fertilization rate and seed-setting rate of the hybrids produced by the PTSGMS lines crossed with the restorer line.

3. Results

3.1. Changes in BR contents

As shown in Fig.1,the contents of BRs(24-EBL and 28-HBL)in the pistils of PTSGMS lines showed no significant difference between HT and NT one day before temperature treatment(BT-1). HT significantly reduced BR levels in pistils at the second day (DT-2) and the fourth day (DT-4) of treatment,with more reduction at DT-4 in comparison with the corresponding NT. One day after the end of the temperature treatment (AT-1), BR contents were still lower under HT than under NT, but were higher than at DT-4 under the HT treatment.The three PTSGMS lines showed the same tendencies in both years.However,the content of 24-EBL or 28-HBL in pistils was the highest for Peiai 64S, followed by Guangzhan 63S, and the lowest for Shen 08S at the same measurement time, especially under the HT treatment (Fig. 1A-L), suggesting that BR levels in pistils vary with PTSGMS line. Under the same treatment and at the same measurement time, the content of 28-HBL in a PTSGMS line was approximately 1.5-1.7 fold that of 24-EBL(Fig.1A-L).

Fig.3- Effects of high temperature stress during anthesis on contents of hydrogen peroxide(H2O2,A-F)and 1-aminocylopropane-1-carboxylic acid(ACC,G-L)in pistils of photo-thermosensitive genetic male-sterile rice lines.NT and HT represent normal and high temperature,respectively.BT-1,DT-2,DT-4,and AT-1 represent one day before temperature treatment,the second day during treatment,the fourth day during treatment,and one day after the end of treatment,respectively.Vertical bars represent±standard error of the mean(n=5).Different letters above the column indicate statistical significance at P=0.05 within the same rice line and in the same year.

Changes in the content of 24-EBL and 28-HBL were also significant as growth proceeded(DT-2, DT-4, and AT-1) at NT(Fig. 1A-L). These changes were more marked than the changes among lines, indicating that both 24-EBL and 28-HBL levels in pistils varied by growth period.

3.2.Changes in catalase activity and levels of AsA,ACC,and H2O2

Similar to BR contents,catalase activity and AsA content in pistils of PTSGMS lines were markedly decreased under HT relative to those under NT,with more reduction at DT-4(Fig.2A-L). They were increased at AT-1 in comparison with those at DT-4 under the HT treatment. Among the three PTSGMS lines, Peiai 64S had the highest and Shen 08S the lowest catalase activity and AsA content in pistils at the same measurement time and under the HT treatment (Fig.2A-L).

Fig.4- Correlations of contents of 24-epibrassinolide(24-EBL)and 28-homobrassinolide (28-HBL) with catalase activity and contents of ascorbic acid(AsA), H2O2,and 1-aminocylopropane-1-carboxylic acid(ACC)in pistils of photo-thermosensitive genetic male-sterile rice lines.Values used for calculations are from Figs.1,2,and 3.**indicate correlation significance at P=0.01(n =48,i.e.,3 rice lines×2 temperature treatments×4 measurement times×2 study years).

?

In contrast to catalase activity and AsA content, contents of ACC and H2O2in pistils were significantly increased under HT, with higher increases at DT-4, in comparison with those under NT (Fig. 3A-L). At the same measurement time and under the HT treatment, Peiai 64S had the lowest and Shen 08S the highest ACC and H2O2contents in the pistils among the three PTSGMS lines(Fig.3A-L), suggesting again that ROS production in the pistils under HT stress during anthesis varies by PTSGMS line.

Changes of catalase activity and contents of AsA, ACC and H2O2were closely correlated with contents of BRs(Figs.1, 2, and 3). The contents of both 24-EBL and 28-HBL were highly significantly and positively correlated with catalase activity and AsA content with r = 0.89**to 0.97**(Fig. 4A-D),and highly significantly and negatively correlated with ACC and H2O2contents with r= ?0.88**to ?0.96**(Fig.4E-H).

3.3. Fertilization rate and seed yield of hybrids

Table 1 shows fertilization rate, seed yield, and its components in hybrids produced by crosses of PTSGMS lines with the restorer line 9311. As it was applied during anthesis, the HT treatment did not affect spikelet number per pot for the same hybrid (Table 1). HT significantly reduced fertilization rate, and correspondingly, significantly decreased seedsetting rate and seed yield in comparison with the NT treatment. However,these reductions varied with hybrid.For example,in comparison with the corresponding NT,the seed yield was reduced by 46.1%-47.4% for Peiai 64S × 9311, by 49.3%-50.5% for Guangzhan 63S × 9311, and by 54.6%-55.7%for Shen 08S × 9311, and these reductions were attributed mainly to differences in the reduction of fertilization rate or seed-setting rate among the hybrids(Table 1).Relative to that under NT, 1000-kernel weight was significantly increased under HT,as a result of the substantial reduction in fertilized spikelets(Table 1).

As shown in Fig.5,both fertilization rate and seed-seeding rate of the hybrids were highly significantly and positively correlated with the contents of 24-EBL, 28-HBL, and AsA and catalase activity in pistils of PTSGMS lines (r = 0.91** to 0.96**,Fig. 5A-D, G-J), and highly significantly and negatively correlated with ACC and H2O2contents in pistils (r = ?0.96**to ?0.97**,Fig.5E,F,K,L),suggesting that decreases in levels of BRs and AOS and the increase in ROS production in the pistils of PTSGMS lines result in the reduction of fertilization rate and seed-setting rate of their hybrids subjected to HT stress during anthesis.

3.4. Effects of chemical regulators

Fig.5- Correlations of fertilization rate and seed-setting rate of hybrids with contents of 24-epibrassinolide(24-EBL)and 28-homobrassinolide(28-HBL),catalase activity,and contents of ascorbic acid(AsA),1-aminocylopropane-1-carboxylic acid(ACC),and H2O2 in pistils of photo-thermosensitive genetic male-sterile rice lines.Values used for calculations are from Table 1 and Figs.1-3,and physiological parameter values are averaged over the four measurement times(BT-1,DT-2,DT-4,and AT-1,refer to Fig.1)within the same rice line, under the same temperature treatment,and in the same year. **indicate correlation significance at P =0.01 (n= 12,i.e., 3 rice lines×2 temperature treatments× 2 study years).

Compared with control (application of deionized water),application of 24-EBL (T1) or 28-HBL (T2) significantly increased the contents of 24-EBL, 28-HBL, and AsA and catalase activity, but markedly decreased ACC and H2O2contents in the pistils of Peiai 64S(Table 2),and significantly increased fertilization rate,seed-setting rate and seed yield of the hybrid (Table 3). The opposite effects were observed when BRz (T3) was applied to these plants. However, the inhibitory effects of BRz on the pistil activity of the PTSGMS line and on the fertilization ability of the hybrid were eliminated when BRz was accompanied with 24-EBL (T4)(Tables 2 and 3).

4. Discussion

Prior to this study, it was unknown whether BRs in pistils of PTSGMS lines respond to HT stress during anthesis and thereby regulate pistil fertilization ability. The results herein demonstrate that BRs (24-EBL and 28-HBL) in pistils mediate the effect of HT during anthesis on pistil fertilization ability of PTSGMS lines, and that a higher level of BRs in the pistil mitigates this harm and thereby reduces the seed yield loss of their hybrids in response to HT stress.

?

Usually,HT stress substantially increases ROS and reduces AOS in plant organs, and excessive production and accumulation of ROS damages cell membranes and causes programmed cell death [34-36]. In the present study, decreases in 24-EBL and 28-HBL levels in pistils were closely associated with increases in ROS level (H2O2content) and decreases in AOS levels (catalase activity and AsA content) there (Figs. 1 and 2). The contents of both 24-EBL and 28-HBL were positively correlated with catalase activity and AsA content and negatively correlated with H2O2content (Fig. 4A-F).Increasing 24-EBL or 28-HBL contents by applying 24-EBL or 28-HBL significantly increased catalase activity and AsA content and reduced H2O2content in the pistils of a PTDGMS line under HT during anthesis.These roles of 24-EBL or 28-HBL in regulating ROS and AOS levels were further confirmed by application of an inhibitor of BR biosynthesis(Table 2).These results suggest that BRs strengthen AOS and suppress ROS production in rice pistils, and accordingly increase pistil fertilization ability under HT stress during anthesis.

ACC and ethylene levels are much increased in plants under abiotic stress,and promote H2O2generation and thereby inhibit plant growth and development [37-40]. In the present study,decreases in BR contents were closely associated with increases in ACC content in pistils under HT stress(Figs.1 and 2),and both 24-EBL or 28-HBL contents were negatively correlated with ACC content(Fig.4G and H).Increases or decreases in BR(24-EBL or 28-HBL)levels were accompanied by decreases or increases in ACC levels in pistils when an exogenous BR or an inhibitor of BR biosynthesis was applied to panicles (Table 2). We accordingly speculate that a potential metabolic interaction between BRs and ethylene influences the effects of HT stress on pistil fertilization ability by regulating ROS,such as via H2O2generation,under HT stress.

?

Proper application of N fertilizer or an irrigation regime with alternate wetting and moderate drying during panicle development increased BR biosynthesis in rice plants,thereby promoting spikelet differentiation and suppressing spikelet degeneration [22-24]. In the present study, application of 24-EBL or 28-HBL significantly increased fertilization rate, seedsetting rate, and seed yield of a hybrid from a PTSGMS line hybridized with a restorer line under HT stress (Table 3),suggesting that either 24-EBL or 28-HBL could be used as a chemical regulator in rice production to alleviate the impairment of pistil fertility when HT stress occurs during anthesis.Appropriate water and N management practices for reducing HT stress during anthesis by elevating BR levels in rice plants merit investigation.

The variation observed in both 24-EBl and 28-HBL contents in pistils among PTSGMS lines, especially under HT stress (Fig. 1)suggests that a PTSGMS line with higher BR levels could be developed so that a higher fertilization rate and a higher seed yield of its hybrid could be obtained under HT stress.The genetic and molecular mechanisms underlying the differences in BR levels in pistils among PTSGMS rice lines await further study.

5.Conclusions

The response of BRs (24-EBL and 28-HBL) in rice pistils to HT stress during anthesis varied among PTSGMS lines. A line with higher BR content showed lower contents of ACC and H2O2,higher catalase activity and higher AsA content in pistils, and higher fertilization rate, seed-setting rate, and seed yield when the line was crossed with a restorer line.Both 24-EBL and 28-HBL in pistils may mediate the effect of HT stress during anthesis on pistil fertilization ability and alleviate the harm of HT stress to pistil fertility by increasing catalase activity and AsA content and suppressing ACC and H2O2generation.

Declaration of competing interest

The authors declare there are no conflicts of interest.

CRediT authorship contribution statement

J.C. Yang and J.H. Zhang designed the research. J. Chen, K.Q.Fei, W.Y. Zhang, and Z.Q. Wang performed the research and analyzed the data. J. Chen and J.C. Yang wrote the paper. All authors reviewed the manuscript.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (31771710, 31901445), the National Key Research and Development Program of China(2016YFD0300206-4,2018YFD0300800),the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD), and the Top Talent Supporting Program of Yangzhou University(2015-01).