Selenium Inhibits Root Elongation by Repressing the Generation of Endogenous Hydrogen Sulfide in Brassica rapa

Selenium (Se) has been becoming an emerging pollutant causing severe phytotoxicity, which the biochemical mechanism is rarely known. Although hydrogen sulfide (H2S) has been suggested as an important exogenous regulator modulating plant physiological adaptions in response to heavy metal stress, whether and how the endogenous H2S regulates Se-induce phytotoxicity remains unclear. In this work, a self-developed specific fluorescent probe (WSP-1) was applied to track endogenous H2S in situ in the roots of Brassica rapa under Se(IV) stress. Se(IV)-induced root growth stunt was closely correlated with the inhibition of endogenous H2S generation in root tips. Se(IV) stress dampened the expression of most LCD and DCD homologues in the roots of B. rapa. By using various specific fluorescent probes for bio-imaging root tips in situ, we found that the increase in endogenous H2S by the application of H2S donor NaHS could significantly alleviate Se(IV)-induced reactive oxygen species (ROS) over-accumulation, oxidative impairment, and cell death in root tips, which further resulted in the recovery of root growth under Se(IV) stress. However, dampening the endogenous H2S could block the alleviated effect of NaHS on Se(IV)-induced phytotoxicity. Finally, the increase in endogenous H2S resulted in the enhancement of glutathione (GSH) in Se(IV)-treated roots, which may share the similar molecular mechanism for the dominant role of H2S in removing ROS by activating GSH biosynthesis in mammals. Altogether, these data provide the first direct evidences confirming the pivotal role of endogenous H2S in modulating Se(IV)-induced phytotoxicity in roots.


Introduction
Selenium (Se) contamination is a global environmental safety issue because Se is becoming an emerging health hazards due to the dramatic increase in Se concentration in the environment [1,2]. The rapid development of metal industry promotes the biogeochemical cycle of Se, which results in the remarkably anthropogenic release of Se into soil [1,3]. Se is an essential micronutrient for plants because Se-containing proteins play vital roles in regulating plant growth and plant adaption to the environment [4][5][6]. Additionally, human prefer to consume Serich foods because Se appears to have an critical role in strengthening the immune system in human body [7,8]. Thus, the importance of Se for both human and plants has driven the long-term application of Se fertilizers in farm work, which is another important factor contributing to the increasing anthropogenic release of Se into the agricultural environment [9,10].
Se with low dose often acts as a protector helping plants against various environmental stimuli [11], but the great concern has been raised about the possible adverse effects of the excessive Se in plants. Treatment with Se (8-16 ppm) significantly inhibits the growth of barley shoot [12]. Se at the concentration of 4-6 ppm show remarkable inhibitory effect on the growth of both shoot and root in bean seedlings [13]. By using image analysis of roots, the root development of lettuce and ryegrass can be completely inhibited by Se as low as 1 ppm [14]. The mechanism of Seinduced phytotoxicity is rarely reported because of the limited studies about the adverse effects of Se on plants. Several studies suggest that excessive Se can trigger oxidative stress in plants by inducing the production of reactive oxygen species (ROS) and the subsequent lipid peroxidation, which may contribute to Seinduced phytotoxicity [12,13,15]. A recent study indicated that Se-induced growth stunt of root was closely associated with the disturbance of plant hormones and endogenous nitric oxide (NO) in Arabidopsis [16], but the biochemical mechanisms for Seinduced phytotoxicity are still elusive.
Recently, H 2 S has been drawing increasing attention in plants because it shows great potential in the regulation of multiple physiological processes in plants, but the detailed studies in the biological role of H 2 S in plants are still very limited as compared to those in mammals [25,26]. The exogenous application of NaHS, a H 2 S donor, can alleviate the phytotoxicity induced by various metal species, such as copper (Cu) [27], chromium (Cr) [28], boron (B) [29], lead (Pd) [30,31], aluminum (Al) [32][33][34], and cadmium (Cd) [35][36][37]. All of these reports suggest that H 2 S may be an important player regulating plant response to heavy metal stress. Nevertheless, the specific role of endogenous H 2 S in modulating the phytotoxicity induced by heavy metals (including Se) is largely unknown because of the lack of the data of tracking endogenous H 2 S in situ in plants. Our recent study demonstrate that Washington Stat Probe 1 (WSP-1) is a very useful fluorescent probe for selectively capturing and tracking H 2 S in vivo in plant root, which provides a powerful tool for identifying the role of endogenous H 2 S as a true cellular signaling molecule in regulating plant physiology [38,39].
In this work, we investigated whether and how endogenous H 2 S responds to Se-induced toxicity in the roots of Brassica rapa. The effect of Se stress on the generation of endogenous H 2 S was studied in vivo by using fluorescent microscopy. To get deeper insights into the role of H 2 S in Se-induced toxicity, the involvement of the endogenous H 2 S in root elongation, cell death, and oxidative injury was investigated further by pharmacological experiments. These results were able to help our understanding for the role of H 2 S in plants under Se stress, which could extend our knowledge of H 2 S in plants and Se-induced phytotoxicity.

Plant culture and chemicals
Seeds of B. rapa (LvLing) seeds were surface-sterilized with 1% NaClO for 10 min followed by washing with distilled water. Seeds were germinated for 1 day in the dark on the floating plastic nets. Then the selected identical seedlings with radicles 0.5 cm were transferred to another Petri dish containing various treatment solutions in a chamber with a photosynthetic active radiation of 200 mmol/m 2 /s, a photoperiod of 12 h, and the temperature at 2561uC.
Seedling roots were exposed to Na 2 SeO 3 (sodium selenite, Se(IV)) with different concentrations (0.03-0.46 mM) for various treatment time (0-72 h). The 0-2.0 mM of NaHS (sodium hydrosulphide) was applied as H 2 S donor. PAG ( DL -propargylglicine) (0.05-0.2 mM) and HT (hypotaurine) (0.1-0.4 mM) are H 2 S biosynthesis inhibitor and H 2 S scavenger, respectively. The treatment solution is composed of different chemicals as mentioned above according to the experimental design. After treatments, the roots were washed with distilled water for physiological, histochemical, and biochemical analysis.

Histochemical analysis
The intracellular H 2 S was visualized using specific fluorescent probe WSP-1 [39-methoxy-3-oxo-3H-spiro[isobenzofuran-1,99xanthen]-69-yl 2-(pyridin-2-yldisulfanyl)benzoate] in situ according to our previous method [38]. The roots of seedlings after treatments were incubated at 20 mM Hepes-NaOH (pH 7.5) buffer solution containing 20 mM of WSP-1 at 25uC for 40 min. Then the roots were washed with distilled water three times and were visualized immediately by a fluorescence microscope with a 465/515 nm and an excitation/emission filter set (ECLIPSE, TE2000-S, Nikon). The relative fluorescent density of the fluorescent images was analyzed using Image-Pro Plus 6.0 (Media Cybernetics, Inc.).
Intracellular ROS was visualized using specific fluorescent probe DCFH-DA (29,79-dichlorofluorescein diacetate) in situ described by Foreman et al. [40]. The roots of seedlings were incubated in 10 mM of DCFH-DA at 25uC for 10 min. Then the roots were rinsed with distilled water for three times followed by the visualization (excitation 488 nm and emission 525 nm) with a fluorescence microscope (ECLIPSE, TE2000-S, Nikon). The relative fluorescent density of the fluorescent images was analyzed using Image-Pro Plus 6.0 (Media Cybernetics, Inc.). Intracellular superoxide radical was visualized using specific fluorescent probe DHE (dihydroethidium) in situ described by Yamamoto et al [41]. The roots of seedlings after treatment were incubated in 15 mM of DHE at 25uC for 15 min. Then the roots were rinsed with distilled water for three times and were visualized (excitation 535 nm and emission 610 nm) by a fluorescence microscope (ECLIPSE, TE2000-S, Nikon). The relative fluorescent density of the fluorescent images was analysed using Image-Pro Plus 6.0 (Media Cybernetics, Inc.).
Histochemical detection of cell death was performed by using propidium iodide (PI) in situ as described by Kellermeier et al [42]. The roots of seedlings after treatment were incubated in 20 mM of PI solution for 20 min. Then the roots were rinsed with distilled water for three times and were visualized (excitation 535 nm and emission 615 nm) by a fluorescence microscope (ECLIPSE, TE2000-S, Nikon). The relative fluorescent density of the fluorescent images was analyzed using Image-Pro Plus 6.0 (Media Cybernetics, Inc.).
Histochemical detection of glutathione (GSH) was performed by using specific molecular probe monochlorobimane in situ as described by Liso et al [43]. The endogenous GSH in root was visualized after conjugation with monochlorobimane to give fluorescent GS-bimane adduct. The roots of seedlings after treatment were incubated in 100 mM of monochlorobimane solution for 30 min. Then the roots were rinsed with distilled water for three times and were visualized (excitation 390 nm and emission 478 nm) by a fluorescence microscope (ECLIPSE, TE2000-S, Nikon). The relative fluorescent density of the fluorescent images was analyzed using Image-Pro Plus 6.0 (Media Cybernetics, Inc.). Histochemical detection of lipid peroxidation was achieved by using Schiff's regent as described by Wang and Yang [44]. The roots of seedlings after treatment were incubated in Schiff's regent for 20 min. Then the stained roots were rinsed with a solution containing 0.5% (w/v) K 2 S 2 O 5 (prepared in 0.05 M of HCl) until the root colour became light red. After that, the roots were photographed using a digital camera.
Histochemical detection of loss of plasma membrane integrity was performed by using Evans blue as described by Yamamoto et al. [45]. The roots of seedlings after treatment were incubated in 5 ml of 0.025% Evans blue solutions (w/v) for 20 min. After that, the roots were rinsed with distilled water for three times followed by photographed using a digital camera.
Screening and analysis of LCD and DCD from the genome of B. rapa The sequences of AtLCD (AT5G28030) and AtDCDs (AT3G26115 and AT2G48420) from Arabidopsis were used as baits for BLAST research in the genome of B. rapa from BRAD (http://brassicadb.org/brad/index.php). The obtained sequences were retrieved and analyzed. The phylogenetic trees were constructed using the maximum likelihood method in MEGA 5.2. The multialignment of amino acid sequences was performed with ClustalX 2.0 and DNAMAN 5.2.2. Protein structure prediction was performed on SMART (http://smart.emblheidelberg.de/).
The DNA sequences with the length of 2 kb were retrieved from the upstream of LCDs and DCDs in B. rapa for promoter analysis. The sequence between the start of target gene and the end of its upstream gene was obtained for promoter analysis if the length of this sequence was less than 2 kb. The cis-elements in the retrieved promoter regions were analyzed using online tool PLACE (http://www.dna.affrc.go.jp/PLACE/signalscan.html).

Analysis of transcripts
Total RNA was extracted from root tissues using Trizol (Invitrogen) according to the manufacturer's instructions. Reverse transcription was performed at 42uC in 25 ml reaction mixture including 3 mg of RNA, 0.5 mg of oligo (dT) primers, 12.5 nmol of dNTPs, 20 units of RANase inhibitor and 200 units of M-MLV. The first cDNA was used as a template for polymerase chain amplification and to analyse the transcripts of genes by using realtime quantitative reverse transcription-polymerase chain reaction (qRT-PCR) (Applied Biosystems 7500 Fast Real-Time PCR  System, LifeTechnologies). The primers designed for the amplification of the genes are listed in Table S1.

Measurement of Se concentration in roots
About 0.2 g of dried root sample was mixed with 10 mL of HNO 3 :HClO 4 (4:1, v/v) in a test tube and covered with Parafilm for 24 h. Then the samples were digested. The digested solution was analyzed for Se concentration by using hydride generation atomic fluorescence spectrometry (AFS-230a, Beijing Wantuo). The calibration was performed by using standard Se solution with concentration of 10-80 mg/L [9].

Statistical analysis
Each result was presented as the mean 6 standard deviation (SD) of at least three replicated measurements. The significant differences between treatments were statistically evaluated by SD and one-way analysis of variance (ANOVA) using SPSS 2.0. The data between two specific different treatments were compared statistically by ANOVA, followed by F-test if the ANOVA result is significant at P,0.05. For multiple comparison analysis, least significant difference test (LSD) was performed on all data following ANOVA tests to test for significant (P,0.05) differences among different treatments.

Se(IV) treatment inhibited root elongation of B. rapa
Treatment with Se(IV) significantly inhibited root elongation in both dose-and time-dependent manners. The roots of B. rapa were exposed to 0-0.46 mM of Se(IV) for up to 72 h. Compared to the control group, root elongation decreased by 25%, 63%, 77%, 88%, and 93% at 0.03, 0.06, 0.12, 0.23, and 0.46 mM of Se(IV) levels, respectively ( Figure 1A). In a time-course experiment, exposure of 0.06 mM of Se(IV) showed significantly inhibitory effect on root elongation. Compared to the control group, root elongation began to decreased after treatment with 0.06 mM of Se(IV) for 24 h, and continued to decrease up to 72 h ( Figure 1B).

Se(IV) treatment inhibited the generation of endogenous H 2 S in root tips
Root tip is the main expansion zone for root elongation [46]. In order to test the effect of Se(IV) stress on endogenous H 2 S in root tips, we performed in situ detection of endogenous H 2 S generation by using specific fluorescent probe WSP-1. Compared to the control, the decreased WSP-1 fluorescent density was observed in root tips in the presence of Se(IV) in a dose-dependent manner (Figure 2A and B). In a time-course experiment, the endogenous H 2 S indicated by WSP-1 fluorescence maintained stable up to 24 h in control groups. However, WSP-1 fluorescence began to decrease significantly after treatment with 0.06 mM of Se(IV) for 6 h ( Figure 2C and D). The correlation analysis suggested that the changes of endogenous H 2 S level occurred in parallel with the changes of root elongation under Se(IV) stress. Initially, WSP-1 fluorescent density decreased slowly with the light decrease in root elongation, followed by a quick decrease with the dramatic inhibition of root elongation induced by Se(IV) at high concentrations ( Figure 2E). These results suggested that the generation of endogenous H 2 S decreased significantly in root tips upon Se(IV) treatment.

Se(IV) stress differentially regulated the expression of LCD and DCD in roots
In order to understand how Se(IV) stress impacted the generation of endogenous H 2 S, we further investigated the effect of Se(IV) stress on the expression of LCD and DCD in the roots of B. rapa. According to BLAST search, sequence identity, and phylogenetic analysis, we obtained two DCD homologues (Bra025184 and Bra018726) and ten LCD homologues (Bra020605, Bra001131, Bra014529, Bra004781, Bra006115, Bra037682, Bra039708, Bra009985, Bra036910, and Bra006114) from B. rapa ( Figure S1). All of the retrieved LCDs and DCDs have typical PLP domains ( Figure S2). LCD has been well studied in Arabidopsis and Brassica napus [23,47]. The multialignment of deduced amino acid sequences revealed that the obtained LCDs from B. rapa had many typically structural features of plant LCDs, such as PLP-binding sites, the substrate binding site, and the SAT protein-interaction site ( Figure S3) [47].
The expression levels of LCDs and DCDs under 0.06 mM of Se(IV) treatment were tested by using qRT-PCR (Figure 3). The results suggested that Se(IV) stress showed extensively inhibitory effect on the expression of both LCDs and DCDs (Figure 3). Two DCDs in roots were down-regulated upon Se(IV) stress (Figure 3).

Analysis of nitric oxide-, auxin-, and metal-responsive cis-elements in the promoter region of LCDs and DCDs in roots
According to the identification of NO-responsive cis-element (NRE) from higher plants [48,49], several NREs (e.g. ACGT Box, MYCL, and W-BOX) could be found in the promoter region of all the LCDs and DCDs obtained from B. rapa (Table S2). In addition, the auxin-responsive cis-element (ARE) could be found in most LCDs and DCDs except for Bra018726 (Table S2).

Application of H 2 S donor NaHS alleviated Se(IV)-induced root inhibition by enhancing endogenous H 2 S level
To obtain more evidence for the involvement of H 2 S in the regulation of root elongation under Se(IV) stress, the H 2 S donor NaHS was added to the treatment solution. A preliminary experiment with NaHS at 0.06-2.0 mM was carried out to determine the point where NaHS showed the most significant effect. Treatment with NaHS at 0.5 mM had the greatest effect on the alleviation of Se(IV)-induced inhibition of root elongation ( Figure 4A). The root elongation increased by 90% in seedlings treated with 0.5 mM NaHS+0.06 mM Se(IV) as compared to 0.06 mM Se(IV) treatment alone ( Figure 4A). In a time-course experiment, Se(IV)-induced reduction in root elongation was significantly recovered when roots were incubated in the treatment solution containing both Se(IV) and 0.5 mM of NaHS ( Figure 4B), which may resulted from the enhancement of endogenous H 2 S level.
To verify the alleviated effect of NaHS on root elongation under Se(IV), we tested the root elongation treated with PAG (endogenous H 2 S biosynthesis inhibitor) and HT (H 2 S scavenger), respectively. Compared to the control, PAG and HT resulted in the significant decreases in root elongation ( Figure 5A), respectively, suggesting that the endogenous H 2 S is essential for root elongation. Furthermore, the addition of PAG or HT could partially blocked the alleviated effect of NaHS on Se(IV)-induced root inhibition ( Figure 5B), which may resulted from the decrease in endogenous H 2 S level.
Subsequently, we test the effect of NaHS application on the endogenous H 2 S level in roots under Se(IV) stress. In Se(IV)-free roots, NaHS could enhance the level of endogenous H 2 S while both PAG and HT were able to decrease endogenous H 2 S level ( Figure 6). Additionally, the addition of NaHS could recover the decrease in endogenous H 2 S level in Se(IV)-treated roots. However, in both Se(IV)-free and Se(IV)-treated roots, the enhancement of endogenous H 2 S level by NaHS supplement could be blocked by the addition of PAG and HT, respectively ( Figure 6). All of these results suggested that the enhancement of endogenous H 2 S could alleviate Se(IV)-induced inhibition in root elongation.

Treatment with NaHS attenuated Se(IV)-induced ROS generation, cell death, and oxidative injury in roots
Compared to the control, treatment with 0.06 mM of Se(IV) resulted in the over-generation of total endogenous ROS in root tips indicated by staining with specific fluorescent probe DCFH-DA. However, the addition of NaHS significantly decreased the accumulation of total ROS induced by Se(IV) ( Figure 7A and B). Superoxide radical, one of the most important ROS, was detected with specific fluorescent probe DHE. The addition of NaHS significantly inhibited the increase in superoxide radical level in root tips under Se(IV) stress ( Figure 7C and D).
Cell death in root tips were fluorescently detected with PI. The application of NaHS was able to significantly alleviate Se(IV)induced cell death in root tips ( Figure 7E and F). Because the overgeneration of ROS is closely related to the oxidative injury to plant cells, we further determined the peroxidation of membrane lipids and the loss of membrane integrity by using histochemical staining with Shiff's regent and Evans blue, respectively [44]. Compared to the control, the roots treated with Se(IV) showed more extensive staining. However, the roots treated with Se(IV)+NaHS had only light staining as compared to Se(IV) treatment alone (Figure 8). These results indicated that the enhancement of endogenous H 2 S by applying NaHS could alleviate Se(IV)-induced cell injury in root tips.
Treatment with NaHS enhanced endogenous GSH level in Se(IV)-treated roots By using specific molecular probe for detecting endogenous GSH in situ, treatment with 0.06 mM of Se(IV) significantly decreased the endogenous GSH level in root as compared to the control. However, the addition of NaHS could remarkabley enhance the GSH level ( Figure 9A). The relative GS-bimane fluorescent density indicated that treatment with Se(IV)+NaHS increased the endogenous GSH by 82.6% as compared to Se(IV) treatment alone ( Figure 9B).

The concentration of Se in roots were not affected effectively by the application of NaHS
In order to test the effect of NaHS treatment on the uptake of Se by the roots, the Se concentration in roots were measured and compared between Se(IV) treatment and Se(IV)+NaHS treatment. In a time-course experiment, treatment with 0.6 mM of Se(IV) resulted in the continuous increase in concentration of Se in roots as compared to the control groups ( Figure S4). However, the addition of 0.5 mM of NaHS didn't affect the concentration of Se in roots significantly as compared to the treatment of Se(IV) alone ( Figure S4).

Discussion
In comparison with other heavy metals (e.g. Cd, Zn, Al, Pb, and Hg), whose phytotoxicity have been well documented [50][51][52][53], the biochemical mechanisms for plant responses to Se are rarely known. It has been demonstrated that the exogenous application of H 2 S can modify plant physiology in response to heavy metal stress [25]. However, whether and how the endogenous H 2 S influences plant growth under heavy metal stress remains unclear. In the present study, by using in situ fluorescent tracking of endogenous H 2 S, we found that the inhibition of endogenous H 2 S generation underlay Se(IV)-induced inhibition of root elongation in B. rapa, which could be supported by four lines of evidence. First, Se(IV)-induced inhibition of root elongation was closely correlated with the decrease in endogenous H 2 S in root tips. Second, Se(IV)-induced inhibition of endogenous H 2 S generation may result from the down-regulation of LCDs and DCDs. Third, the application of H 2 S donor NaHS could enhance endogenous H 2 S level in root tips, which further resulted in the recovery of root elongation under Se(IV) stress. The decrease in endogenous H 2 S level by the addition of PAG and HT could block the recoverable effect of NaHS on root elongation under Se(IV) stress. Fourth, the enhancement of endogenous H 2 S by NaHS resulted in the alleviation of Se(IV)-induced ROS accumulation, cell death, and oxidative injury in root tips.
It has been reported that H 2 S is required for the organogenesis of lateral root and adventitious root [38,[54][55][56]. In the present study, decreasing the endogenous H 2 S level using PAG or HT could inhibit the root elongation of B. rapa ( Figure 5A, suggesting that endogenous H 2 S is indispensable for root elongation. Se(IV)induced inhibition of endogenous H 2 S generation in root tips may probably contributed to the depression of root elongation (Figure 1 and 2). Several LCDs and DCDs responsible for the endogenous generation of H 2 S have been identified from plants [23,24,47], but the regulation of the expression of these genes by environmental stimuli is rarely reported. In this study, we detected the genome-wide expression pattern of LCDs and DCDs in B. rapa under Se(IV) stress. We found an effective inhibitory action of Se(IV) to the expression of LCDs and DCDs in the roots of B. rapa (Figure 3), which may contributed to the significant decrease in endogenous H 2 S level ( Figure 2). The similarly depressed mode of LCD expression was also observed in B. napus under Cd stress [47]. Interestingly, we also found some typical NO-or auxin-responsive elements in the promoter region of LCDs (Table S2), which may explain the previous observation that exogenous NO and auxin could stimulate the expression of LCD in B. napus [47]. Supported by genetic evidences it was proposed that the inhibition of endogenous NO and auxin contributed to Se(IV)-induced root growth inhibition in Arabidopsis [16]. Thus, it can an indication that NO and auxin may act upstream of H 2 S by manipulating the expression of LCDs and DCDs during Se(IV)-induced inhibition of root growth. In addition, the transcriptional regulation of LCDs and DCDs in B. rapa by Se(IV) may also results from the presence of MREs in their promoter region.
ROS has been suggested as the main inducer of plant cell death [57]. In the present study, Se(IV)-induced over-accumulation of ROS may contribute to the cell death in root tips, which was also accompanied with the decrease in endogenous H 2 S (Figure 2 and  7). The enhancement of endogenous H 2 S by supplement with NaHS could reverse the inducible effect of Se(IV) on ROS accumulation, cell death, and oxidative injury (Figure 7), suggesting that H 2 S has an important role in the plant protection from Se(IV) stress by scavenging the over-accumulated of ROS. In mammals, the effective stimulation of GSH biosynthesis induced by low level of H 2 S contribute to the suppression of oxidative stress more efficiently than the scavenging of ROS by H 2 S itself [58].  Cystine is indispensable for the biosynthesis of glutathione. H 2 S can enhance the activity of cysteine/glutamate antiporter, leading to the increase in the transport of cystine into cells [59]. Cystine is subsequently reduced to cysteine in cells and incorporated into glutathione. Additionally, H 2 S can directly interact with glutamylcysteine synthase, a limiting enzyme for glutathione biosynthesis, thereby increasing the production of glutathione [59,60]. Our present results demonstrated that the increase in endogenous H 2 S by applying NaHS significantly enhance GSH level in roots under Se(IV) stress (Figure 9). Glutathione plays important role in protecting plants from metal toxicity by scavenging ROS or chelating metals [61]. Plants share similar mechanism with mammals for glutathione biosynthesis [62]. Therefore, in Se(IV)treated plants, whether H 2 S depressed the generation of ROS through the similar mechanism mentioned above remains to be elucidated.
The biology of H 2 S in mammals has been significantly advanced, but evaluating the role of endogenous H 2 S in plants is just beginning. By using in situ fluorescent detection of endogenous H 2 S in plant, we provide the direct evidence that Se(IV) stress can inhibits the generation of endogenous H 2 S in the roots of B. rapa. The enhancement of endogenous H 2 S can alleviate Se(IV)-induced root inhibition by depressing ROS generation and decreasing cell death. These data support the fact that Se(IV) induces phytotoxicity by hijacking the generation of endogenous H 2 S in B. rapa. Despite of the observation in this study, H 2 S-mediated signaling components upon Se stress is still elusive. Thus, a more precise understanding of this question will accelerate the investigation on the mechanism of Se-induced phytotoxicity, which in turn will help the improvement of crop production in Se-polluted environment.  Figure S2 The location of PLP-dependent domain in LCDs and DCDs from B. rapa. The protein structure of two LCDs and two DCDs were analyzed by online tool SMART. The typical PLP-dependent domains were indicated as orange box. Bar indicated 100 amino acids (aa). (TIF) Figure S3 Alignment of the predicted amino acid sequences of LCDs in A. thaliana, B. napus, and B. rapa. Dark shading with white letters and gray shading with black letters reveal 100% and 75% sequence similarity, respectively. Database accession numbers are the same as described in Figure  S1. The PLP-binding sites are shown by red box. The substrate binding site is indicated by blue box. The SAT protein-interaction site is indicated by red box.

Supporting Information
(TIF) Figure S4 The concentration of Se in the roots of B. rapa exposed to Se(IV) or Se(IV)+NaHS. The roots were exposed to Se(IV) (0.06 mM) or Se(IV) (0.06 mM)+NaHS (0.5 mM) for 6, 12, 24, 48, 72 h, respectively. The roots were harvested at each point of treatment time for Se analysis. Each value was presented as the mean of three replicates with SD. (TIF)