Selenium Promotes Sunower Resistance to Sclerotinia Sclerotiorum by Regulating Redox Homeostasis and Hormonal Signaling Pathways

Sclerotinia wilt of sunower caused by Sclerotinia sclerotiorum is a devastating disease causing serious loss. Selenium (Se) has a benet effect to plant in stress tolerance. In this study, sunower leaves treated by foliar application of Se were inoculated with S. sclerotiorum. Pathogenesis on the inoculated leaves and transcript levels of plant genes involved in redox homeostasis and hormonal signaling pathways were examined. Se could be detected after the foliar application and was mainly transformed to selenomethionine in sunower. Consequently, Se pretreatment delayed the necrosis development caused by S. sclerotiorum and alleviated the adverse effects derived from pathogen infection by differentially balancing the regulation of enzymes involved in the redox homeostasis. Specially, the cat expression increased to alleviate its downregulation responded to pathogen infection at the earlier infection stage (12 hour post inoculation, hpi) while the pod, gpx, apx, and nox expressions decreased to alleviate their responsive upregulation at the later infection stages (24 and 36 hpi). Se pretreatment enhanced the regulation of genes involved in hormonal signaling pathways, in which the AOC and PAL expressions increased to enhance its upregulation induced by pathogen infection for improving resistant responses at the earlier infection stage (12 hpi), as well as the AOC and PDF expressions increased at the later infection stages (24 hpi). Besides, the EIN2 expression increased to alleviate its downregulation at all of infection stages. Our results suggested that Se plays the benecial effect on the resistant responses to S. sclerotiorum infection. This study provided a clue to improve the sustainable management of Sclerotinia wilt on sunower by Se foliar application. pathways under Se pretreatment during pathogen infection, the roles of Se pretreatment were elucidated.


Introduction
Sun ower (Helianthus annuus L.) is an annual herbal species of the Asteraceae family with medicinal and nutritional values (Guo et al. 2017). Sun ower can be used as oilseed, snack, salad garnish, livestock and pet feed, and as ornamental plants in domestic gardens (Alagawany et al. 2015;Guo et al. 2017;Toscano et al. 2017). The global yield of sun ower is approximately 56 million tons in 2019, which makes it the second largest oil crops on harvested area (FAOSAT 2020). Sun ower is threated by many plant diseases such as downy mildew, black stem, Phomopsis stem canker and Sclerotinia wilt (Debaeke et al. 2014). Particularly, Sclerotinia wilt caused by the fungal pathogen Sclerotinia sclerotiorum is considered as a devastating disease that signi cantly affects the quality and the yield of sun ower production (Liu et al. 2017;Seiler et al. 2017). This fungus generally causes rot damage on root, stem, and head of sun ower (Ekins et al. 2011;Seiler et al. 2017).
Plants establish complex defense mechanisms against pathogen infections, which are predominately dependent on the crosstalk regulation of reactive oxygen species (ROS) and plant hormone signaling pathways (Fujita et al. 2006). ROS generated in plant not only contributes in the regulation of plant growth and development, but also facilitates responses to environmental stresses (Suzuki et al. 2011).
However, the excessive accumulation of ROS has a distressing effect on the plant and ultimately leads to cell death (Sharma et al. 2019). Therefore, plants have evolved redox homeostasis, which involves various enzymes such as superoxide dismutase (SOD), catalase (CAT), peroxidase (POD), ascorbate peroxidase (APX), glutathione peroxidase (GPx), and NADPH oxidase (NOX) (Tiwari et al. 2017). SOD, CAT, POD, APX, and GPx are activated to scavenge ROS while NOX is involved in generating ROS (Sharma et al. 2019). Additionally, plant defense responses appropriately depend on the coordinated regulation between ROS and plant hormone signaling pathways in environmental abiotic stresses and pathogen infection (Fujita et al. 2006;Xia et al. 2015). Plants activate the appropriate plant hormone signaling pathways including salicylic acid (SA), jasmonic acid (JA), and ethylene (ET) according to the pathogen infection (Bari and Jones 2009;Spoel and Dong 2008).
The fungus S. sclerotiorum is a necrotrophic pathogen with a broad host range (Bolton et al. 2006). The infection process of S. sclerotiorum was proposed to have two phases, in both phases this fungus can overcome plant defense response (Kim et al. 2008;. Sclerotinia sclerotiorum suppresses plant ROS generation by interfering the host redox homeostasis in the rst phase and promotes plant ROS production to elicit the programmed cell death in the second phase (Williams et al. 2011). For example, the activities of POD, SOD, and NOX were differentially regulated while CAT activity showed unchanged during the infection course of S. sclerotiorum in oilseed rape (Liang et al. 2009). The activities of APX and GPx were also regulated in common bean and tobacco challenged by S. sclerotiorum (Fagundes-Nacarath et al. 2018;Ma et al. 2018). Therefore, ROS modulation of host plants plays critical roles in plant defense responses to S. sclerotiorum infection (Ranjan et al. 2018). Plant hormone signaling pathways were also modulated in plant defense responses to S. sclerotiorum invasion (Guo and Stotz 2007). For example, the signaling pathways of JA and ET were involved in defense against S. sclerotiorum in oilseed rape (Liang et al. 2009;Zhao et al. 2009). The genes involved in the JA and ET signaling pathways were differentially expressed in the pathosystem of S. sclerotiroum and oilseed rape (Liang et al. 2009). Additionally, the SA signaling pathway was possibly coordinated by JA regulation at the later stage of S. sclerotiorum infection in oilseed rape (Nováková et al. 2014).
Management of plant diseases can be improved by manipulating macro-and micro-nutrients such as sulphur, iron, copper, and selenium (Boyd 2007;Fagundes-Nacarath et al. 2018). Selenium (Se) can improve plant resistance to abiotic and biotic stresses by regulating ROS production and plant hormone signaling pathways (El-Ramady et al. 2014). For example, Se was used to protect Indian mustard from root infection by Fusarium species and tomato fruits from B. cinerea, respectively (Hanson et al. 2003;Wu et al. 2016). Se foliar application is a safer method of Se-bioforti cation than Se-fertilization in soil since Se uptake is affected by soil features (Bañuelos et al. 2017;Kápolna et al. 2009;Winkel et al. 2015).
However, in uence of Se foliar application on sun ower infected by S. sclerotiorum has never been reported. In this study, the effects of Se foliar application on necrotic development of sun owers infected by S. sclerotiorum were assessed and the regulation of redox homeostasis and plant hormone signaling pathways was investigated. This study provides new insights into foliar Se application for improving the sustainable management of sun ower or other plants damaged by S. sclerotiorum.

Fungal and plant materials
The pathogen S. sclerotiorum was maintained on potato dextrose agar (PDA; 200 g of potatoes, 20 g of glucose, and 18 g of agar per liter) plates in this study. The fungal strain was supplied from College of Plant Protection, Shenyang Agricultural University. Seedlings of the sun ower cultivar KWS204 (Beijing Tiankui Leader Seed Technology, Beijing, China) were grown in potting soil in a growth chamber with a photoperiod of 16-h light and 8-h dark at 22°C. Plants at the six-leaf stage (approximately four-week-old) was used for subsequent assays.

Foliar Se treatment
Sun ower leaves were sprayed with a selenite solution [i.e., sodium selenite (Sigma-Aldrich, St. Louis, MO, USA) dissolved in distilled water] at serial concentrations (20, 40, and 60 µM) with approximately 2 mL per plant until runoff. Sterile water spray was served as the control. After three days of incubation in a growth chamber, the Se-treated sun owers were thoroughly rinsed three times with sterilized water to remove the residual selenite (Farooq et al. 2019). Thereafter, each of the six leaves of each plant was inoculated with a 3-mm diameter mycelial plug excised from the active colony edge of a S. sclerotiorum culture. After inoculation, all the plants were maintained in a growth chamber with 90% relative humidity. Lesions developed on the inoculated leaves were assessed at 24 hour post inoculation (hpi) by following . The measurement was performed with ten pretreated plants and the inoculation was independently repeated three times (i.e., 60 leaves for each time), and all pretreated leaves were inoculated for each plant.

Inoculation assays
Based on the inhibitory effects and lesion differentiation under Se pretreatment and pathogen challenge, the optimal concentration of Se pretreatment was determined. Thereafter, four treatments were applied in this study: (i) pathogen inoculation after Se treatment (SeIn), (ii) mock inoculation with PDA plugs after Se treatment (SeUn), (iii) pathogen inoculation after mock treatment with water (In), and (iv) mock inoculation with PDA plugs after mock treatment with water (Un). To elucidate the development of necrosis caused by S. sclerotiorum under the optimal Se treatment, disease severity was investigated at 12, 18, 24, 36, and 48 hpi, respectively. The leaves were photographed under a visible and high intensity UV lamp (Analytik Jena US, Upland, CA, USA) and lesion area was measured ). Furthermore, a logistic equation was constructed to differentiate necrosis development within the infection course under SeIn and In treatments using the OriginPro software (version 9.1; OriginLab, Northampton, MA, USA) as described in Wang et al. (2009). For each of the time points, four plants each with six leaves (24 leaves) were randomly selected as one repeat and the inoculation assessment was independently performed three times in this analysis.

Se transformation and speciation in pretreated leaves
Total Se content in leaves was assessed using hydride generation ame atomic uorescence spectrometry (HG-AFS; AFS-920, Beijing Jitian Instruments, Beijing, China) with three biological replicates and three technical replicates for each biological replicate (Hu et al. 2018). Brie y, 1 g of SeUn or Un sample was digested in 8 mL 15.3 M HNO 3 following the microwave assisted acid digestion. After cooling down, 2.5 mL 6 M HCl was added to each sample, and then the solution was heated to 100°C for 1 hour. The obtained solution was diluted with deionized water to a nal volume of 50 mL while 2 mL of the solution was injected into the HG-AFS system for Se analysis. Each sample was prepared with three replicates and analyzed for linear estimation of Se pretreatment based on regression analysis using Microsoft Excel. The linear regression of Se content and uorescence values was built using the standard substance of sodium selenite. Blanks and a certi ed reference material (Chinese cabbage material, GBW 10014) were included in each batch of samples for quality control.
Additionally, Se speciation in leaves was distinguished by high-performance liquid chromatography coupled to hydride generation atomic uorescence spectrometry (HPLC-AFS; SA-20, Beijing Jitian Instruments, Beijing, China) with three biological replicates (Wang et al. 2020). Brie y, 0.8 g of fresh leaves of SeUn or Un was hydrolyzed by 5 mL of 8 mg mL − 1 protease XIV (Sigma-Aldrich, St. Louis, MO, USA) at 37℃ for 12 hours. The samples were centrifuged at 12000 ×g for 5 min, and the supernatant was ltered through a 0.22-µm mixed cellulose nitrate lter. The ltered samples (300 µL) were injected into HPLC-AFS system for analysis, while the total Se content of samples was detected by HG-AFS system. The Se speciation was separated using an anion-exchange column (PRP-X100, Hamilton, Switzerland), and then eluted with 40 mM (NH 4 ) 2 HPO 4 (pH 6.0) at a ow rate of 1 mL min − 1 . Peaks were identi ed according to the retention times of standard compounds [i.e., selenocystine (SeCys 2 ), Semethyl-selenocysteine (MeSeCys), selenite, selenomethionine (SeMet), and selenate] purchased from the National Research Center for Certi ed Reference Materials, Beijing, China. The identi ed Se species were quanti ed based on the peak areas of the calibration curves using an HPLC workstation. The proportion of Se speciation in pretreated sun ower leaves was calculated using the expression: Se speciation content / total Se content of leaves × 100%. Extraction e ciency rate was calculated using the expression: total Se content of enzymatic extract / total Se content of leaves × 100%.

Effects of transformed Se on fungal growth
To further estimate the potential inhibitory effects of Se on S. sclerotiorum growth, the mycelial growth was assessed on PDA plates supplied with 226.1 µg L − 1 (Se concentration) selenite or SeMet (dissolved in distilled water) (Sigma-Aldrich, St. Louis, MO, USA), the maximal concentration that could be detected in Se-pretreated sun ower leaves. PDA plate without Se served as the control. Mycelium growth on PDA with and without Se supplement was measured at 24 and 36 hpi, which was used to differentiate mycelium growth affected by Se content in the pretreated leaves.

Expression regulation analysis
To evaluate the gene expression regulation of Se-pretreated sun ower after S. sclerotiorum infection, certain crucial genes (sod, gpx, cat, apx, pod, and nox) were selected to measure the regulation of redox homeostasis. The expression regulation analysis of the key genes involved in JA (AOC and PDF), ET (EIN2) and SA (PAL) pathways were also conducted for the regulation of hormonal signaling pathways.
Brie y, total RNA was extracted using an Eastep Super Total RNA Extraction Kit (Promega, Madison, WI, USA). The extracted RNA was visualized to verify RNA quality by 1.0% agarose gel electrophoresis and quanti ed by a NanoDrop 2000 (Thermo Fisher Scienti c, Waltham, MA, USA). First-strand cDNA was synthesized from 1 µg of total RNA using a GoScript Reverse Transcription System (Promega). Primers speci c to each of those selected genes were designed using Primer Express Software (version 3.0.1; Thermo Fisher Scienti c), and actin was used as an endogenous reference gene (Table S1). Quantitative PCR (qPCR) was conducted using a StepOne Plus Real-Time PCR System (Thermo Fisher Scienti c).
Each reaction was 10 µL in volume that contained 5 µL of 2× EvaGreen qPCR Mastermix-ROX (Applied Biological Materials, Richmond, BC, Canada), 2 µL of 100× diluted cDNA templates and 3 µL of primer mixture (2.5 µM per primer). qPCR was conducted with the following program: 95°C for 10 min, 35 cycles of ampli cation at 95°C for 15 seconds and 60°C for 60 seconds. The relative expression of each selected gene was calculated according to the method (Livak and Schmittgen 2001). Three biological replicates for each treatment were performed, in which three technical replicates were conducted for each biological replicate.

Data analysis
Data were statistically analyzed between two treatments using Student's t-test or determined among multiple treatments at each of assayed time points using Duncan's multiple range test of analysis of variance (ANOVA) by the SAS software (version 9.2; SAS Institute, Cary, NC, USA). To better elucidate the regulative roles of Se pretreatment on the plant responses to pathogen infection, the expression differentiation of the assayed genes was analyzed. Brie y, the differential expression was preliminarily elucidated the regulative effects induced by pathogen infection and Se pretreatment based on the pairwise analysis of In/Un or SeUn/Un. Thereafter, the expression differentiation of the pairwise SeIn/In was further analyzed to understand the regulative effects of Se pretreatment during pathogen infection, which was excluded the regulative in uence based on the expression differentiation of the pairwise SeIn/Un.

Lesion decreased by Se pretreatment
Lesions on leaves received 20, 40, and 60 µM Se pretreatment were signi cantly (P < 0.05) decreased 27.2%, 46.8%, and 39.0% compared to the control, respectively (Fig. 1). These results indicated that Se pretreatment can impact the disease progress caused by S. sclerotiorum. The 40 µM of Se pretreatment more e ciently reduced the lesion size. Thus, this concentration was selected for future analysis of necrotic development caused by S. sclerotiorum.

Se transformation and speciation in pretreated leaves
Total Se content in the sun ower leaves after Se pretreatment (SeUn) was 226.1 µg kg − 1 , which was signi cantly (P < 0.01) higher (3.4-fold) compared to that without Se pretreatment (Un) (Fig. 2a) . 2b). In this study, only two peaks (262 and 356 of retention times) were detected in Se pretreated leaves (SeUn), which matched with those of the standard selenite and SeMet (Fig. 2c). The contents of SeMet and selenite in the leaves of SeUn were 180.7 µg kg − 1 and 13.7 µg kg − 1 , respectively. SeMet accounts for 79.9% of the total Se, which was the main Se speciation in the Se pretreated sun ower leaves (SeUn).
To evaluate the potential inhibitory effects of the detected Se speciation on fungal growth, mycelia growth of S. sclerotiorum was assessed on PDA supplied with each of detected Se speciation at their corresponding detected maximum concentration (i.e., 226.1 µg L − 1 ) based on the analysis of total Se content and speciation in SeUn leaves. Our results indicated that both detected Se speciation did not affect mycelium growth of S. sclerotiorum (Fig. 2d). Therefore, Se and its transformed speciation detected in sun ower leaves after Se pretreatment did not directly inhibit the vegetative growth of S. sclerotiorum. This suggested that the impact on necrotic development caused by S. sclerotiorum infection were associated with plant responses induced by Se pretreatment, and were not derived from the inhibitory effect of Se.

Effects on necrotic development under Se pretreatment
Symptomatic lesions were observed at 12 hpi and gradually enlarged on the leaves challenged by S. sclerotiorum with (SeIn) and without Se pretreatment (In) (Fig. 3a). The distinguished necrosis on the inoculated leaves under SeIn and In was also validated by UV light observation (Fig. 3b). The lesion of the SeIn treatment was not signi cantly different to that of In at 12 hpi, while the lesion of SeIn was 38.5% less than that of In at 18 hpi. Subsequently, the lesion of SeIn was 52.1% smaller than that of In at 24 hpi.
Thereafter, the lesions derived from SeIn signi cantly declined 54.4% and 51.3% compared to those of In at 36 and 48 hpi, respectively (Fig. 3c). According the necrosis development and lesion differentiation, three infection stages (i.e., 12, 24, and 36 hpi) were determined to further investigate the responses to S. sclerotiorum infection in this study. In addition, the necrosis development on sun ower leaves of SeIn and In was further analyzed based on a logistic equation of time course of pathogenic development (Fig. 3d).
Results indicated that the time course to reach the fastest rate of lesion expansion (50% of the nal value) derived from SeIn was 11 hours later than that of In (Table 1). The values of the slopes, including slope1 at the point of in exion of the logistic curve and slope2 of the linear regression [i.e., y'=-k(x-x c )], indicated that the rate of lesion development of SeIn was slower than that of In. These results indicated that Se pretreatment reduced necrotic development caused by S. sclerotiorum infection on sun ower.

Regulation of redox homeostasis
The expression regulation of the crucial enzymes involved in redox homeostasis was investigated in a time course (e.g., 12, 24, and 36 hpi) of S. sclerotiorum infection (Fig. 4). Results indicated that the cat and sod expressions of In showed the decreased expression induced by pathogen infection (Fig. 4a and  4b). The cat expression of In were consistently downregulated 68.3%, 36.6%, and 64.0% compared to those of Un at 12, 24, and 36 hpi, while the cat expressions of SeUn were signi cantly (P < 0.05) upregulated 1.5-and 2.0-fold than those of Un at 24 and 36 hpi (Fig. 4a). The cat expression of SeIn was induced to be similar to the Un at 12 hpi, which showed a 2.5-fold increase compared to that of In (Fig. 4a). Our results indicated that the cat expression was decreased under pathogen infection, but was increased under Se pretreatment. Thus, Se pretreatment contributed to maintain the normal level (Un) of the cat expression at the earlier infection stage by upregulation of the cat expression. The sod expressions of In were consistently downregulated 72.3%, 59.4%, and 61.4% compared with those of Un at 12, 24, and 36 hpi, while those of SeUn were also signi cantly (P < 0.05) downregulated 66.3% and 45.5% than those of Un at 12 and 24 hpi (Fig. 4b). However, the sod expression showed no differentially regulation between SeIn and In at any assayed time points (Fig. 4b). Such results indicated that the sod expression was downregulated by either pathogen infection or Se pretreatment, but Se pretreatment was not involved in the sod expression regulation derived from pathogen infection in this study. These results indicated that the cat and sod expressions were decreased due to pathogen infection. Besides, cat expression was decreased due to Se pretreatment while sod expression was increased. Furthermore, Se pretreatment potentially increased the downregulation of cat expression caused by pathogen infection at the earlier infection stage.
The expression patterns of three genes (e.g., apx, gpx and pod) encoding ROS-scavenging enzymes were analyzed in this study. The apx, gpx, and pod expressions of In were signi cantly (P < 0.05) upregulated than those of Un, except for the apx expression at 36 hpi. For example, the apx expressions of In were signi cantly (P < 0.05) upregulated 2.1-and 2.0-fold than those of Un at 12 and 24 hpi, while the apx expression was not differentially regulated between SeUn and Un at any assayed time points (Fig. 4c). The apx expression of SeIn was regulated to be similar to the Un at 24 hpi, which was 36.3% decreased compared to that of In (Fig. 4c). Similarly, the gpx expressions of In were consistently upregulated 1.9-, 6.7-, and 3.7-fold compared to those of Un at 12, 24, and 36 hpi. However, the gpx expressions of SeUn were not differentially regulated compared to those of Un at any assayed time points (Fig. 4d). The gpx expressions of SeIn were signi cantly (P < 0.05) upregulated than those of Un at 24 and 36 hpi, but downregulated 45.7% and 42.5% compared to those of In (Fig. 4d). Furthermore, the pod expression patterns were similar to those of gpx. The pod expressions of In were signi cantly (P < 0.05) upregulated 3.0-, 4.3-, and 1.5-fold compared with those of Un at 12, 24, and 36 hpi. However, the pod expressions of SeUn were similar to those of Un at 12 and 24 hpi, which was downregulated 52.0% compared with that of Un at 36 hpi (Fig. 4e). The pod expressions of SeIn were signi cantly (P < 0.05) upregulated than those of Un at 12 and 24 hpi, but 64.5% and 40.8% downregulation compared with those of In at 24 and 36 hpi (Fig. 4e). These results indicated that the apx, gpx, and pod expressions were increased due to pathogen infection, but not affected by Se pretreatment. However, Se pretreatment potentially played a regulative role to alleviate the increased expression caused by pathogen infection by downregulating the apx, gpx, and pod expressions at the later infection stage.
In addition, the nox expression of In was signi cantly (P < 0.05) upregulated 1.3-fold than that of Un at 36 hpi, while the nox expressions of SeUn were signi cantly (P < 0.05) downregulated 47.5% and 36.6% compared to those of Un at 24 and 36 hpi, respectively (Fig. 4f). The nox expression levels of SeIn were consistently downregulated than those of Un at any assayed time points, and signi cantly (P < 0.05) downregulated 51.1% and 56.5% compared to those of In at 24 and 36 hpi (Fig. 4f). These results indicated that the nox expressions were increased under pathogen infection, but were decreased under Se pretreatment. Thus, Se pretreatment downregulated the nox expressions, and played the regulative roles to alleviatory the upregulation induced by pathogen infection at the later infection stage. Based on the expression analysis of these genes involved in redox homeostasis under Se pretreatment during pathogen infection, the regulative roles to alleviate the effects induced by Se pretreatment were elucidated based on the gene expression differentiation during pathogen infection, especially the upregulation of cat at the earlier infection stage (12 hpi), the downregulation of apx, gpx, and pod at the developing infection stage (24 hpi), and the downregulation of gpx and nox at the later infection stages (36 hpi).

Regulation of plant hormone signaling pathways
The expression regulation of the critical genes involved to signaling pathway was also analyzed at 12, 24, and 36 hpi (Fig. 5). The AOC expression of In were signi cantly (P < 0.05) upregulated 13.5-, 9.0-, and 12.9-fold compared to those of Un at 12, 24, and 36 hpi, while AOC expression was not differentially regulated between SeUn and Un at any assayed time points (Fig. 5a). AOC expressions of SeIn were signi cantly upregulated than those of Un, and increased 2.0-and 2.7-fold than those of In at 12 and 24 hpi, respectively (Fig. 5a). The PDF expressions of In were also signi cantly (P < 0.05) upregulated 2.7-, 3.0-, and 5.9-fold than those of Un at 12, 24, and 36 hpi, respectively, while the PDF expression of SeUn was signi cantly (P < 0.05) upregulated 4.4-fold than that of Un at 12 hpi (Fig. 5b). The PDF expression of SeIn was signi cantly upregulated than those of Un, and increased 1.8-fold than that of In at 24 hpi ( Fig. 5b). The PAL expressions of In were consistently upregulated 1.9-, 3.3-, and 4.5-fold than those of Un at 12, 24, and 36 hpi, while the PAL expression of SeUn was not differentially regulated compared to that of Un (Fig. 5c). The PAL expressions of SeIn were signi cantly (P < 0.05) upregulated than those of Un at any assayed time points, while increased 1.5-fold at 12 hpi and decreased at 36 hpi than that of In (Fig. 5c). These results indicated that the AOC, PDF, and PAL expressions were increased by pathogen infection, and were not signi cantly regulated by Se pretreatment. However, Se pretreatment potentially played regulative roles to alleviate the upregulation of the AOC, PDF, and PAL expressions induced by pathogen infection during all of infection stage.
The EIN2 expressions of In were consistently downregulated 42.0%, 56.4%, and 36.0% compared with those of Un at 12, 24, and 36 hpi, while the EIN2 expressions of SeUn were signi cantly (P < 0.05) upregulated 1.4-and 2.1-fold than those of Un at 12 and 24 hpi (Fig. 5d). The EIN2 expressions of SeIn were consistently not changed compared to those of Un at 12, 24, and 36 hpi, and upregulated 1.8-, 2.0-, and 1.5-fold than those of In (Fig. 5d). These results indicated that the EIN2 expression was decreased under pathogen infection, and was increased under Se pretreatment. Thus, Se pretreatment contributed to maintain the normal level (Un) of the EIN2 expression at all of infection stage by upregulation of EIN2 expression. According to the expression analysis of the curial genes involved in hormonal signaling pathways under Se pretreatment during pathogen infection, the roles of Se pretreatment were elucidated.
Se pretreatment upregulated AOC, PAL, and EIN2 at the earlier infection stage (12 hpi) and upregulated AOC, PDF, and EIN2 at the developing infection stage (24 hpi), while downregulated PDF and PAL and upregulated EIN2 at the later infection stage (36 hpi).

Discussion
Nutrient elements (e.g., Se) are frequently applied on plant to improve growth and disease resistance (Dordas 2008;Feng et al. 2013). In previous studies, Se was used as a protective chemical on mustard against Fusarium speices as well as B. cinerea and Alternaria solani infection on tomato (Hanson et al. 2003;Quiterio-Gutiérrez et al. 2019;Wu et al. 2016). Our results showed that Se pretreatment delayed the lesion occurrence and slowed the necrosis expansion on sun ower leaves. Se content in the Se-pretreated sun ower leaves was detected mainly as the form of SeMet. SeMet is more active than selenite and can incorporate into proteins nonspeci cally through the metabolic pathways of sulfur analogues by replacing Met to the form of Se-containing proteins (Wu et al. 2020;Zhang et al. 2020). In plants, Secontaining proteins not only carry enzymatic functions, but also carry antioxidant activity by scavenging free radicals directly (Liu et al. 2015;Zhang et al. 2020).
Activation of plant resistance to pathogen infection has been found to be a result of complex mechanisms, with the involvement of redox homeostasis regulation and plant hormone signaling pathways (Bari and Jones 2009;Torres et al. 2006). For example, sun ower seeds pre-soaked with Se weakened the oxidative damage caused by cadmium (Saidi et al. 2014). Thereby, the regulations of genes associated with redox homeostasis and plant hormone signaling pathways were assessed in Se pretreated sun ower after challenge by S. sclerotiorum. Expression analysis indicated the coordinated regulation between Se and the pathogen. Interacting with the regulatory effects caused by S. sclerotiorum infection, Se pretreatment increased the downregulation of cat and decreased the upregulation of apx, gpx, pod, and nox at the different infection stages. In a previous study, upregulation of cat in dun ower was found to be able toimprove resistance to S. sclerotiorum (Na et al. 2018). In another study, the downregulation of pod and apx in tomato by the treatment of manganese phosphit could reduce the disease severity caused by S. sclerotiorum (Chaves et al. 2021). The downregulation of sod expression was previously found to improve soybean resistance to S. sclerotiorum after sprayed with calcium (Arfaoui et al. 2018). However, in this study, Se pretreatment did not affect the downregulation of sod expression in sun ower infected by S. sclerotiorum. Thus, the mechanism of Se pretreatment affecting sod expression might be different. In addition, the activated response dependent on nox regulation perturbed the disease progress of S. sclerotiorum in Arabidopsis (Zhou et al. 2013). Our results suggested that Se pretreatment contributed to the coordinated regulation of redox homeostasis by alleviating ROS during the disease development of S. sclerotiorum in sun owers.
Plant signaling pathways are involved in the regulation of plant defense responses, particularly the JA, ET, and SA pathways (Yang et al. 2015). For example, melatonin was found to promote tomato resistance to B. cinerea through regulating H 2 O 2 accumulation and JA signaling pathway (Liu et al. 2019). In this study, AOC and PDF in the JA signaling pathway, EIN2 in the ET signaling pathway, and PAL in the SA signaling pathway were signi cantly upregulated in the Se pretreated sun ower inoculated by S. sclerotiorum. In a previous study, upregulated expressions of AOC and PDF in the JA pathway was found in oilseed rape after infected by S. sclerotiorum (Zhao et al. 2009). Similarly, the AOC expression was signi cantly enhanced in mustard infected by S. sclerotiorum, while PDF was able to improve Arabidopsis resistance against A. brassicicola (Penninckx et al. 1998;Yang et al. 2010). In addition, PAL expression in tomato was upregulated as a response of S. sclerotiorum infection (Farzand et al. 2019). EIN2 expression was upregulated as the defense responses to S. sclerotiorum in Arabidopsis (Guo and Stotz 2007). Therefore, Se pretreatment contributed to the resistant improvement by coordinately regulating plant hormone signaling pathways during the infection process of S. sclerotiorum.
A model that Se pretreatment modulated the responses to S. sclerotiorum infection was proposed for crosstalk between redox homeostasis and plant hormone signaling pathways (Fig. 6). Based on the evidence of our analysis, this model illuminates the contribution by Se on the improvement of sun ower resistance to S. sclerotiorum and perhaps in other plant pathosystems. According to the regulation of gene expressions, the redox homeostasis in sun ower responded to S. sclerotiorum infection was alleviated by the Se application. In other studies, Se was also found to mitigate the damage of ROS under cadmium stress in rice . Iron oxide nanoparticles regulated the antioxidant defense to maintain redox homeostasis in Moldavian dragonhead under salinity stress (Moradbeygi et al. 2020).
Plant signaling pathways were also associated with the regulation of redox homeostasis (Overmyer et al. 2003). Overall, we proposed that Se coordinated the regulation of redox homeostasis and signaling pathways to improve resistance to S. sclerotiorum. This study provided a new clue for the sustainable management of Sclerotinia wilt on sun ower by Se foliar application.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analyzed during this study are included in this published article and its supplementary information les.

Competing interests
No conflict of interest declared.  Tables   Table 1 Logistic  Lesion size on sun ower leaves after pre-treatment with Se and then inoculated with S. sclerotiorum.
Data is shown as mean ± standard error. Means in the plot topped by the same letter do not differ based on Duncan's multiple range test at P < 0.05 (n = 3).  Appearance of sun ower leaves inoculated with S. sclerotiorum with Se treatment (SeIn) and the inoculated plants without Se treatment (In). Photographs were taken at 12, 18, 24, 36, and 48 hpi under visible (a) and UV light (b). Lesion size on sun ower leaves (c). Data are means ± standard error. Lesion caused by the pathogen inoculation with (SeIn) and without (In) Se treatment among the assayed time points were statistically analyzed by ANOVA with Duncan's multiple range test (P < 0.05) while those at a speci c time point between the SeIn and In were evaluated by the Student's t-test (*, P < 0.05 and **, P < 0.01). For each treatment, means in the plot topped by the different letters indicate signi cant differences (the capital for the In and the lower case for the SeIn). (d) Logistic function was tted to the data of lesion size against time and lesion size (y) was regressed on time (x) from inoculation.