Hydrogen sulfide activates calcium signaling to confer tolerance against selenium stress in Brassica rapa

Abstract

Background

Se (selenium) pollution is an emerging environmental concern. Excessive Se induces phytotoxicity. The endogenous H₂S (hydrogen sulfide) was involved in plant adaptation to Se stress, but the signaling player of H₂S remains unclear.

Methods

The study was conducted in a hydroponic system with different chemicals added to the treatment solution. Fluorescent tracking was performed to detect endogenous signaling molecules in plant tissues. Physiological changes were determined based on pharmaceutics and histochemical experiments. Gene expression was analyzed using qRT-PCR. The data were summarized using hierarchical cluster and Pearson correlation analysis.

Results

Se stress inhibited B. rapa growth (e.g. root elongation, shoot height, and seedling fresh weight and dry weight) in both dose- and time-dependent manners, with approximately 50% of root growth inhibition occurred at 20 µM Se. Se stress induced ROS (reactive oxygen species) accumulation and oxidative injury in B. rapa. Se exposure resulted in the upregulation of LCDs (L-cysteine desulfhydrase) and DCDs (D-cysteine desulfhydrase) encoding enzymes for H₂S production in B. rapa at early stage of Se exposure, followed by downregulation of these genes at late stage. This was consistent with the change of endogenous H₂S in B. rapa. Enhancing endogenous H₂S level with NaHS (H₂S donor) stimulates endogenous Ca²⁺ in B. rapa upon Se exposure, accompanied the attenuation of growth inhibition, ROS accumulation, oxidative injury, and cell death. The beneficial effects of H₂S on detoxifying Se were blocked by decreasing endogenous Ca²⁺ level with Ca²⁺ channel blocker or Ca²⁺ chelator. Finally, hierarchical cluster combined with correlation analysis revealed that Ca²⁺ might acted as downstream of H₂S to confer Se tolerance in B. rapa.

Conclusion

Ca²⁺ was an important player of H₂S in the regulation of plant physiological response upon Se stress. Such findings extend our knowledge of the mechanism for Se-induced phytotoxicity.

Graphical Abstract

Introduction

Se (selenium) contamination has been becoming an emerging environmental concern due to agricultural and industrial activities (He et al. 2018). Se is an essential nutrient for mammals. Mammals take up Se mainly through consuming agro-products containing Se. Therefore, food crops with Se biofortification has been becoming popular, requiring the increase in the application of Se-enriched fertilizers in agricultural environment (Sarwar et al. 2020). Excessive use of these fertilizers increases Se level in soil, sediments, and groundwater (Bajaj et al. 2011; Mehdi et al. 2013; Winkel et al. 2012). Mining industries (such as metals, phosphate, and coal mining) also accelerate the release of Se into the environment (Etteieb et al. 2020). Excessive Se leads to serious environmental pollution (Sakamoto et al. 2012). The limit of 10 µg/L has been commonly used for Se in drinking water in most countries (Vinceti et al. 2013). In some regions, the concentration of Se in water is up to 669.5–1400 µg/L (Bajaj et al. 2011; Kuisi & Abdel-Fattah 2010; Zelmanov & Semiat 2013). Irrigation of Se-rich soil can accelerate the mobilization of Se to enlarge the pollution (Kausch & Pallud 2013).

Se at low level is beneficial for plants’ adaptation to stress conditions, but excessive Se inhibits plant growth by inducing physiological disorders. Se-induced phytotoxicity depends on different Se forms and different plant species. In plants, SeO₃²⁻ is more toxic than SeO₄²⁻ because SeO₃²⁻ is easier to be incorporated into the Se-amino acids in plant cells. This process causes protein dysfunction, further resulting in phytotoxicity and growth inhibition (Lyons et al. 2005). SeO₃²⁻ at 50–100 µM remarkably inhibits the biomass of pea, maize, Indica mustard, and Arabidopsis (Hawrylak-Nowak 2008; Lehotai et al. 2016; Molnár et al. 2018). Some plant species are even more sensitive to Se stress. SeO₃²⁻ at more than 20 µM reduces the growth and productivity of cucumber and lettuce (Hawrylak-Nowak 2013; Hawrylak-Nowak et al. 2015). Se-induced phytotoxicity includes oxidative injury, nutrient deficiency, and phytohormones disturbance, etc. (Hasanuzzaman et al. 2020b). Oxidative stress is one of the typical consequences of Se-induced phytotoxicity. Se stress inhibits root growth by disturbing the homeostasis of ROS (reactive oxygen species) and RNS (reactive nitrogen species) in Arabidopsis thaliana (Kolbert et al. 2016). Se-induced increase in ROS content and lipid peroxidation have been found in variable plant species, such as Pisum sativum (Lehotai et al. 2016), Brassica rapa (Chen et al. 2014a), Triticum aestivum (Łabanowska et al. 2012), and Hordeum vulgare (Akbulut & Cakir 2010). The protective role of Se in plants have been extensively studied in detail, but Se-induced phytotoxicity has not been fully understood.

The endogenous gaseotransmitter H₂S (hydrogen sulfide) plays vital roles in regulating plant growth and development. In plant cells, H₂S can be generated by LCD (_L-cysteine desulfhydrase, EC4.4.1.1) and DCD (_D-cysteine desulfhydrase, EC4.4.1.15) (Arif et al. 2021). H₂S has been characterized as a defensive signaling molecule combating stress conditions (Zhang et al. 2021). Our previous study demonstrated that Se stress inhibited the root growth of Brassica rapa (Chinese cabbage) by suppressing endogenous H₂S (Chen et al. 2014b), while other players related to H₂S signal pathway are still missing in plants upon Se stress. H₂S can interact with various signaling molecules to regulate plant intrinsic physiology (Wang et al. 2021a). Ca²⁺ is one of the important player of H₂S. Ca²⁺ is an important second messenger for plant stress adaptation (Tong et al. 2021). Ca²⁺ can act both upstream and downstream of H₂S in plants in response to variable abiotic stimuli (Li 2019). CaM (calmodulin), one of the core transducers of Ca²⁺ signaling, is also involved in the interplay between H₂S and Ca²⁺ (Fang et al. 2014). However, it is unclear whether and how H₂S-Ca²⁺ interaction regulates Se-induced phytotoxicity.

In this work, we analyzed Se stress-inhibited growth of B. rapa seedlings. The role of H₂S-Ca²⁺ interaction in the regulation of B. rapa tolerance against Se stress was studied. The results of this work may help understand the mechanism for plant physiological adaptation to Se stress.

Materials and methods

Plant culture and treatment

The seeds of B. rapa were sterilized with NaClO (1%) for 5 min, followed by washing with distilled water and germinating at 25 °C for 24 h. Then the germinated seeds were transferred to a floating net, culturing with 1/4 strength Hoagland solution in a light chamber with active radiation of 200 μmol/(m² s), photoperiod of 12 h, and temperature at 25 °C, based on our previous study (Cheng et al. 2021). About 30 seedlings with root length at 1.5 cm were cultured at 700 mL nutrient solution with Na₂SeO₃ added at different concentration (0–80 µM). Various chemicals were added to the nutrient solution for different treatment. NaHS (10 µM) and HT (hypotaurine) (20 µM) were applied as H₂S donor and H₂S scavenger, respectively. EGTA (ethylene glycol-bis(2-aminoethylether)-N,N,N′,N′-tetraacetic acid) (0.5 mM) and LaCl₃ (50 µM) were applied as Ca²⁺ chelator and Ca²⁺ channel blocker, respectively (Li et al. 2014). Plant tissues were harvested after treatment, respectively, followed by physiological measurement.

Fluorescent detection

We used several specific fluorescent probes to perform histochemical detection in situ in roots based on our previously published methods (Li et al. 2014; Yang et al. 2022). Endogenous H₂S, Ca²⁺, total ROS, cell death were detected by using WSP-1 (Washington State Probe-1), Fluo-3, DCFH-DA (2',7'-dichlorofluorescein diacetate), and PI (propidium iodide), respectively. WSP-1 was obtained from Bejing Solarbio Science & Technology Co., Ltd (Beijing, China). Fluo-3, DCFH-DA, and PI were obtained from Beyotime Biotech. Inc (Shanghai, China). The fluorescent images were captured with a fluorescence microscope (ECLIPSE, TE2000-S, Nikon, Melville, NY, USA). Image-Pro Plus 6.0 (Media Cybernetics, Inc., Rockville, MD, USA) was applied to calculate relative fluorescent density for each image.

Histochemical detection

Endogenous hydrogen peroxide and superoxide radical in leaves were detected in vivo by using DAB (3,3-diaminobenzidine) and NBT (nitro-blue tetrazolium) staining, respectively (Zhou et al. 2008). For hydrogen peroxide detection, leaves were incubated in DAB (0.1 w/v, pH 3.8) at 25 °C for 30 min. For superoxide radical detection, leaves were incubated in 6 mM NBT (dissolved in 10 mM sodium-citrate buffer, pH 6.0) at 25 °C for 30 min. The leaves with specific staining were transferred to boiling ethanol for 30 min to remove the green background (chlorophyll), followed by photographed with a digital camera.

The lipid peroxidation in root was histochemically detected by using Schiff’s reagent. The roots were stained in Schiff’s reagent for 15 min, followed by washing with K₂S₂O₅ solution (0.5% w/v in 0.05 M HCl) for 10 min. The loss of membrane integrity in roots were histochemically detected using Evans blue. The roots were incubated at Evans blue solution (0.025%, w/v) for 15 min, followed by washing with distilled water for 10 min. For the histochemical detection in leaves, the leaves were stained with the reagent as described above. Then the stained leaves were incubated in boiling ethanol for 30 min to remove chlorophyll, followed by photographed with a digital camera (Ye et al. 2016).

Gene experssion analysis

We selected Trizol (Invitrogen, ThermoFisher Scientific, Shanghai, China) to extract total RNA from plant tissues based on manufacturer’s instructions. The reaction mixture for reverse transcription consisted of M-MLV (200 units), RNAase inhibitor (20 units), ligo (dT) primers (0.5 µg), and RNA (3 µg). The obtained first cDNA was used as template to perform quantitative RT-PCR using Applied Biosystems 7500 Fast Real-Time PCR System (LifeTechnologies™, ThermoFisher Scientific, Shanghai, China). The abundance of gene expression was quantified using 2^−ΔΔT threshold cycle method (Livak & Schmittgen 2001). The relative abundance of Actin was applied as internal standard to standardize the results. The primers used for amplifying target genes were listed in Table S1.

Data analysis

Each result were shown as mean ± SD (standard deviation) with at least three replicates. The significant difference between two designated data sets was evaluated by ANOVA (one-way analysis of variance) combined with F-test at P < 0.05. LSD (least significant difference) was performed to make multiple comparison analysis at P < 0.05. The package “corrplot” in R was used to perform Pearson correlation analysis among different parameters. The heatmaps for hierarchical cluster analysis and gene expression analysis were generated by using the package “pheatmap” in R and TBtools (Chen et al. 2020).

Results

Se stress inhibited the grwoth of B. rapa

We analyzed the phenotype of B. rapa seedlings under different concentrations of SeO₃²⁻ (0–80 µM) up to 72 h. Se stress inhibited the seedling growth in a dose-dependent manner (Fig. 1A, B). The root length significantly decreased by 5.83%, 23.2%, 50.4%, 60.5%, and 76.8% at 5, 10, 20, 40, and 80 µM Se, respectively, as compared to control (Fig. 1C). The shoot height significantly decreased by 12.8%, 25.1%, 34.9%, 42.1%, and 48.2% at 5, 10, 20, 40, and 80 µM Se, respectively, as compared to control (Fig. 1D). The FW (fresh weight) of seedlings significantly decreased by 14.7%, 20.6%, 35.3%, 58.5%, and 73.5% at 5, 10, 20, 40, and 80 µM Se, respectively, as compared to control (Fig. 1E). The DW (dry weight) of seedlings were significantly inhibited by Se at high concentrations (40–80 µM) (Fig. 1F), which may be resulted from the decrease in RWC (relative water content) (Fig. 1G).

Fig. 1

Growth changes of B. rapa seedlings upon Se exposure. A Photos of seedlings grown in hydroponics with different concentrations of Se for 72 h. B Phenotype of seedlings after treated with Se for 72 h. Bar = 1 cm. C The root length of seedlings. D The shoot height of seedlings. E The fresh weight of seedlings. F The dry weight of seedlings. G The RWC of seedlings. Different lowercase letters in (C-G) indicated significant difference among different treatments (n = 3–10; P < 0.05; LSD, ANOVA)

As the root length under 20 µM Se treatment was about half of control, we monitored the time-course changes of seedling growth upon 20 µM Se. Compared to the control, Se treatment began to significantly prohibit root elongation after exposure of 12 h, followed by decreased growth speed of root with the prolong of treatment time (Fig. 2A). The root FW showed similar changes with root length upon Se exposure (Fig. 2B). The shoot FW and whole plant FW were decreased after 24 h of 20 µM Se treatment (Fig. 2C, D).

Fig. 2

Time-course observation of the growth of B. rapa seedlings under 20 µM Se. A Root length. B Root fresh weight. C Shoot fresh weight. D Whole plant fresh weight. Asterisk indicated significant difference between control and treatment of 20 µM Se at each time point (n = 3–10; P < 0.05; ANOVA)

Se stress induced oxidative stress in B. rapa

We detected total ROS level in roots using specific fluorescent probe DCFH-DA. Se stress led to ROS accumulation in roots in a dose-dependent manner (Fig. 3A). Compared to the control, the DCF fluorescent density significantly increased by 44.5%, 69.3%, 181.2%, 284.6%, and 377.7% at 5, 10, 20, 40, and 80 µM Se, respectively (Fig. 3B). In leaves, we evaluated two typical ROS (superoxide radical and hydrogen peroxide) in vivo by using histochemical analysis. The leaves showed intensified staining with the increase in Se concentration (Fig. 3C, D), indicating that Se treatment resulted in the accumulation of superoxide radical and hydrogen peroxide in leaves.

Fig. 3

ROS accumulation in B. rapa seedlings under Se stress. A Total ROS in roots indicated by DCF fluorescence. B DCF fluorescent density in roots. Different lowercase letters in indicated significant difference among different treatments (n = 3; P < 0.05; LSD). C Superoxide radical in leaves indicated by NBT staining. D Hydrogen peroxide in leaves indicated by DAB staining

As over-accumulated ROS always attack protein and lipid to cause oxidative stress, lipid peroxidation and loss of membrane integrity were detected histochemically to evaluate oxidative injury in B. rapa seedlings upon Se exposure. We observed aggravated lipid peroxidation in both leaves and roots with the increase in Se concentration (Fig. 4A, B). Se stress also induced the loss of membrane integrity in leaves and roots (Fig. 4C, D). The roots showed more severe oxidative injury than that of leaves because of direct Se exposure of roots. These results suggested that Se stress induced oxidative injure in B. rapa.

Fig. 4

Oxidative injury in B. rapa seedlings under Se stress. A Lipid peroxidation in leaves indicated by Shiff’s reagent. B Lipid peroxidation in roots indicated by Shiff’s reagent. C Loss of membrane integrity in leaves indicated by Evans blue. D Loss of membrane integrity in roots indicated by Evans blue

Se stress disturbed endogenous H₂S level in B. rapa

We previously identified the gene family of BrLCD and BrDCD from the genome of B. rapa, which included 10 BrLCDs and 2 BrDCDs (Chen et al. 2014b). Here we detected the expression file of these genes in roots upon the exposure of Se at 20 µM that led to moderate inhibition of root elongation. Se stress induced the upregulation of most of these genes after treatment for 3–48 h followed by downregulation with prolonged treatment (72 h) (Fig. 5).

Fig. 5

Relative expression of BrLCDs and BrDCDs in the roots of B. rapa seedlings upon treatment of 20 µM Se. The heatmap data were shown as the expression of Se treatment with respect to control. Red, white, and blue indicated upregulation, unchanged, and downregulation, respectively

As 20 µM Se induced strong upregulation and downregulation of H₂S-producing genes at 48 and 72 h, respectively (Fig. 5), we detected endogenous H₂S level in roots with specific fluorescent probe WSP-1 at these two time points. Se treatment for 48 h resulted in significant increase in H₂S level as compared to control, with remarkably decreased H₂S level at 72 h (Fig. 6A, B). This was consistent with the expression pattern of BrLCDs and BrDCDs. Endogenous H₂S was decreased at the end of Se treatment (72 h). H₂S was supplied by adding NaHS (H₂S donor) in the treatment solution. As expected, adding NaHS significantly enhanced endogenous H₂S level in Se-treated roots, accompanied the recovery of root elongation. The effect of NaHS was counteracted by adding H₂S scavenger HT (Fig. 6C, D). These results suggested that H₂S generation in roots was triggered upon the early exposure of Se. However, prolonged Se stress decreased H₂S generation, which was associated with the final inhibition of root growth.

Fig. 6

The change of endogenous H₂S in the roots of B. rapa upon Se and NaHS. A Root endogenous H₂S fluorescence detected by WSP-1 upon Se (20 µM) exposure at 48 h and 72 h. B Calculated WSP-1 fluorescent density according to (A). C WSP-1-based endogenous H₂S level in roots under treatment of Se (20 µM), NaHS (10 µM), and HT (20 µM) for 72 h. D Root length under treatment of Se (20 µM), NaHS (10 µM), and HT (20 µM) for 72 h. Asterisk in (B) indicated significant difference between control and Se treatment (n = 3; P < 0.05; ANOVA). Different lowercase letters in (C) and (D) indicated significant difference among different treatments (n = 3–10; P < 0.05; LSD)

Ca²⁺ was involved in H₂S-regulated growth of B. rapa upon Se stress

We selected specific fluorescent probe Fluo-3 to detect endogenous Ca²⁺ in the roots (Fig. 7A). Se (20 µM) exposure for 72 h significantly decreased endogenous Ca²⁺ in the roots compared to control. Adding NaHS significantly enhanced endogenous Ca²⁺ in Se-treated roots, an effect reversed by further adding Ca²⁺ chelator (EGTA) or Ca²⁺ influx channel blocker (La³⁺) (Fig. 7B). The change of relative expression of BrCaM1 showed similar pattern with endogenous Ca²⁺ upon the same treatment (Fig. 7C).

Fig. 7

H₂S activated endogenous Ca²⁺ signaling in the roots of B. rapa upon Se and NaHS. A Root endogenous Ca²⁺ fluorescence detected by Fluo-3 upon NaHS, EGTA (0.5 mM), or La³⁺ (50 µM) in the presence of Se (20 µM) for 72 h. B Calculated Fluo-3 fluorescent density. C Relative expression level of BrCaM1 upon the same treatment. Different lowercase letters in (B) and (C) indicated significant difference among different treatments (n = 3; P < 0.05; LSD)

Then we detected the growth response of B. rapa seedlings upon the same treatment. NaHS attenuated root growth inhibition was counteracted by adding EGTA or La³⁺ under Se exposure (Fig. 8A). The fresh weight of root, shoot, and the whole seedlings also showed similar changing patterns with root length (Fig. 8B-D). The above results suggested that H₂S activated intracellular Ca²⁺ signal to facilitate B. rapa growth under Se stress.

Fig. 8

H₂S-Ca²⁺ interaction promoted the growth of B. rapa upon Se stress. A Root length upon NaHS, EGTA (0.5 mM), or La.³⁺ (50 µM) in the presence of Se (20 µM) for 72 h. B Root fresh weigh. C Shoot fresh weight. D Seedling fresh weight. Different lowercase letters indicated significant difference among different treatments (n = 3–10; P < 0.05; LSD)

Ca²⁺ was involved in H₂S-attenuated oxidative injury in B. rapa upon Se stress

NaHS prohibited Se-induced ROS accumulation in roots and leaves, which could be reversed by further adding EGTA or La³⁺ (Fig. 9). NaHS attenuated Se-induced oxidative injury in roots and leaves, an effect that could be diminished by further adding EGTA or La³⁺ (Fig. 10A-D). Root cell death was detected by using Trypan blue staining and PI fluorescence, respectively. NaHS attenuated Se-induced root cell death, which could be reversed by further applying EGTA or La³⁺ (Fig. 10E-G). These results suggested that Ca²⁺ acted downstream of H₂S to ameliorate ROS accumulation, oxidative injury, and cell death in B. rapa upon Se stress.

Fig. 9

H₂S-Ca²⁺ interaction prohibited ROS accumulation B. rapa upon Se stress. A Root total ROS fluorescence detected by DCF upon NaHS, EGTA (0.5 mM), or La.³⁺ (50 µM) in the presence of Se (20 µM) for 72 h. B Calculated DCF fluorescent density. C NBT-stained superoxide radial in leaves upon the same treatment. D DAB-stained hydrogen peroxide in leaves upon the same treatment. Different lowercase letters in B indicated significant difference among different treatments (n = 3; P < 0.05; LSD)

Fig. 10

H₂S-Ca²⁺ interaction attenuated oxidative injury and cell death in B. rapa upon Se stress. A Loss of membrane integrity in roots upon NaHS, EGTA (0.5 mM), or La.³⁺ (50 µM) in the presence of Se (20 µM) for 72 h. B Lipid peroxidation in roots. C Loss of membrane integrity in leaves. D Lipid peroxidation in leaves. E Root cell death stained by Trypan blue. F Root cell death indicated by PI fluorescence. G Calculated PI fluorescent density. Different lowercase letters in (G) indicated significant difference among different treatments (n = 3; P < 0.05; LSD)

Cluster analysis of H₂S-Ca²⁺ interaction in B. rapa upon Se stress

We performed hierarchical cluster analysis to summarize the relationship among different treatments based on the changes of physiological parameters obtained above (Fig. 11A). The control and Se + NaHS were clustered together, suggesting that NaHS was able to counteract the effect of Se. The Se + NaHS + La³⁺, Se + NaHS + EGTA, and Se were clustered together, suggesting that either La³⁺ or EGTA partially diminished the effect of NaHS under Se stress. Therefore, enhancing H₂S level (provided by NaHS) mitigated Se-induced physiological disorders, an effect that was partially blocked by decreasing intracellular Ca²⁺ (by adding La³⁺ or EGTA).

Fig. 11

Cluster analysis of H₂S-Ca²⁺-facilitated Se tolerance in B. rapa. A Hierarchical cluster analysis among different treatments based on obtained physiological parameters. The legend showed the value of Log₂(Treatment/Control) for each parameter. Red, white, and blue indicated upregulation, unchanged, and downregulation, respectively, for each treatment as compared to control. B Pearson correlation analysis among different physiological parameters under the same treatment setup (Control, Se, Se + NaHS, Se + NaHS + EGTA, and Se + NaHS + La.³⁺)

Then we performed Pearson correlation analysis to investigate the relationship among different physiological parameters under same treatment setup (control, Se, Se + NaHS, Se + NaHS + EGTA, and Se + NaHS + La³⁺) (Fig. 11B). ROS was positively correlated to cell death, indicating that ROS accumulation caused cell death in B. rapa under Se stress. The growth parameters (e.g. root FW, shoot FW, total FW, and root length) were negatively correlated to ROS and cell death, respectively, indicating that ROS-triggered cell death caused growth inhibition. Ca²⁺ was positively correlated to CaM, suggesting the synchronous activation of Ca²⁺ signaling by H₂S. Ca²⁺ and CaM were positively correlated to growth parameters but negatively correlated to ROS and cell death. This indicated that H₂S activated Ca²⁺ signaling to suppress ROS and cell death in order to promote growth under Se stress.

Discussion

Se is an emerging environmental pollutant impeding plant growth. Se can cause phytotoxicity at a wide range of concentration (15–100 µM), which depends on plant species (Hasanuzzaman et al. 2020b). Se at 0.21–4.08 mg/kg (about 2.66–51.7 µM) in agricultural soil resulted in the toxic accumulation of Se in paddy rice and Chinese cabbage (Huang et al. 2009). We found that Se at concentration more than 5 µM inhibited the growth of B. rapa seedlings, with 20 µM of Se resulted in moderate inhibition of seedling growth. Se-induced phytotoxicity not only depends on plant species, but also depends on the growth stage of plant. Further studies are needed to identify the toxic dose of Se on adult plants of B. rapa.

We previously found that endogenous H₂S is vital for the survival of B. rapa seedlings under Se stress (Chen et al. 2014b). In this study, four lines of evidence indicate that H₂S-Ca²⁺ interaction plays a role in combating Se stress in B. rapa. First, Se stress inhibited the growth of B. rapa by inducing oxidative stress. Second, the biosynthesis of endogenous H₂S in B. rapa was suppressed at the end of Se exposure. Enhancing endogenous H₂S alleviated growth inhibition of B. rapa under Se stress, an effect reversed by decreasing endogenous H₂S. Third, Se stress decreased endogenous Ca²⁺ level and BrCaM1 expression in B. rapa, which were elevated by enhancing endogenous H₂S. Fourth, H₂S-ameliorated oxidative stress and growth inhibition were blocked by decreasing endogenous Ca²⁺ in B. rapa under Se stress. These results suggested that H₂S conferred Se tolerance by regulating Ca²⁺ signaling in B. rapa.

Excessive Se frequently induce oxidative injury in plants (Hasanuzzaman et al. 2020a; Van Hoewyk 2013). The antioxidative role of H₂S has been widely identified in plants (Liu et al. 2021; Zhang et al. 2021). We found increased endogenous H₂S in B. rapa at early stage of Se exposure, suggesting that Se stress triggered H₂S-mediated defensive responses. However, prolonged exposure of Se led to decreased endogenous H₂S, accompanying the occurrence of oxidative injury and growth inhibition. Se stress for 72 h finally dampens the defensive role of H₂S, leading to phytotoxicity. NaHS is a reliable donor of H₂S for suppressing ROS in roots (Chen et al. 2014b). NaHS-provided H₂S enhanced endogenous H₂S level in Se-treated roots up to 72 h, leading to the detoxification of excessive Se by alleviating oxidative stress in B. rapa. These results propose a defensive role of endogenous H₂S against Se stress.

The expression patterns of BrLCDs and BrDCDs were similar to the changes of endogenous H₂S level in B. rapa under Se stress. This suggested that Se stress disturbed the biosynthesis of endogenous H₂S. The H₂S biosynthesis can be differentially regulated by different environmental stimuli. Cadmium stress induces H₂S biosynthesis by activating both LCD and DCD in alfalfa (Cui et al. 2014). Drought stress triggers the expression of LCD but not DCD in Arabidopsis thaliana (Shen et al. 2013). The gene expression is mainly controlled by the cis-elements in the promoter region of the gene. Se stress may regulate gene promoters to modulate gene expression in plants (Chen et al. 2014a). Further studies are needed to identify the cis-elements in the promoters of LCDs and DCDs. This may help understand the differential regulation of plant H₂S biosynthesis by Se stress and other environmental stimuli.

Ca²⁺ is a universal messenger regulating plant stress responses (Pirayesh et al. 2021). Little is known about the role of Ca²⁺ in the regulation of Se-induced phytotoxicity. In this study, H₂S-promoted Ca²⁺ was associated with the growth promotion of B. rapa under Se stress. H₂S triggered intracellular Ca²⁺ signaling that further attenuated oxidative stress and growth inhibition in B. rapa under Se stress. The cross-talk between H₂S and Ca²⁺ play a role in alleviating oxidative damage in plants upon abiotic stresses. Ca²⁺ interacts with H₂S can ameliorate oxidative injury to confer salt tolerance in mung bean roots (Khan et al. 2021). The interplay between H₂S and Ca²⁺/CaM facilitates acclimation of zucchini to nickel toxicity by suppressing oxidative stress (Valivand et al. 2019). Having link Ca²⁺ and H₂S into a signaling cassette provides new signaling events of Se tolerance in plants. In mammalian cells, H₂S interacts with the sulfhydryl group of Ca²⁺ channel protein to regulate its activity, leading to the modulation of Ca²⁺ homeostasis (Munaron et al. 2013; Yong et al. 2010). The Ca²⁺ channel blocker compromised H₂S-conferred tolerance against Se stress, suggesting the possible role of Ca²⁺ channel in H₂S signaling. Further studies are needed to identify the possible target Ca²⁺ channel that can be regualted H₂S in plants upon Se stress.

The interaction between H₂S and Ca²⁺/CaM has been found in both plants and fungi. H₂S improves heat tolerance of tobacco by triggering the influx of extracellular Ca²⁺ into cytosol, working with CaM (Li et al. 2012). H₂S induces betulin accumulation by elevating endogenous Ca²⁺ and CaM in the mycelia of Phellinus linteus (Fan et al. 2016). Ca²⁺ can also work upstream of H₂S. In Arabidopsis against chromium stress, Ca²⁺/CaM2 interacts with transcription factor TGA3 to facilitate the binding of TGA3 to the promoter of LCD, enhancing the transcription of LCD to promote H₂S generation (Fang et al. 2017). In the present study, Ca²⁺/CaM acts downstream of H₂S to improve Se tolerance in B. rapa. Whether and how Ca²⁺ modulates H₂S signaling through a possible feedback regulation need further investigation.

Se is an essential micronutrient for both crops and humans. It is important to avoid Se pollution during Se biofortification performance to achieve the development of sustainable agriculture. In Se-deficient areas, efforts are made to enhance the uptake and assimilation of Se in crops. This can be achieved by supplying organic Se fertilizers at low level because crops prefer to accumulate organic forms of Se. This would help minimize the overaccumulation of exogenous Se in agricultural environment. Another possible approach is the genetic engineering of crops with enhanced Se uptake ability (Malagoli et al. 2015). In Se-polluted areas, phytoremediation is a promising approach to decrease Se level in soil. The byproducts of Se phytoremediation need to be disposed carefully (Hasanuzzaman et al. 2020b; Zhu et al. 2009). It is also possible to promote crop tolerance against Se stress in Se-polluted soil. In the present study, exogenous application of H₂S donor was able to trigger Se tolerance in B. rapa, but it is difficult to apply H₂S in field because of the inevitable loss of gaseous H₂S. The promising approach is to construct nano-emulsion particles to package donors, achieving the sustainable supply of gaseous molecules for a long time (Wang et al. 2021b).

Conclusion

Understanding plant physiological adaptation to Se stress is important for the management of excessive Se pollution. This study revealed a new signaling module (H₂S-Ca²⁺) involved in plant response to Se stress. Ca²⁺/CaM acts as a downstream player of H₂S to facilitate plant tolerance against Se stress. H₂S, Ca²⁺ and CaM worked as a liner signaling cassette to suppress ROS accumulation followed by the alleviation of oxidative injury and cell death, promoting root growth under Se stress (Fig. 12). More evidences are needed to identify the detailed biochemical mechanisms for H₂S-Ca²⁺ interaction in Se-treated plants, but our current results would help understand the adaptation of plants to Se contamination.

Fig. 12

Schematic model of the role of H₂S-Ca²⁺ interaction in the regulation Se tolerance in B. rapa. Se exposure decreases endogenous H₂S level by downregulating the expression of LCDs and DCDs. NaHS-provided H₂S enhances endogenous H₂S level upon Se stress. Elevated H₂S triggers Ca²⁺/CaM signaling to alleviate ROS accumulation, oxidative injury and cell death, further promoting root growth. The beneficial effect of NaHS can be abolished by adding HT, EGTA, or La³⁺. NaHS, H₂S donor; HT, H₂S scavenger; EGTA, Ca²⁺ chelator; La³⁺, influx channel blocker