Hydrogen sulfide modulates cadmium-induced physiological and biochemical responses to alleviate cadmium toxicity in rice

We investigated the physiological and biochemical mechanisms by which H2S mitigates the cadmium stress in rice. Results revealed that cadmium exposure resulted in growth inhibition and biomass reduction, which is correlated with the increased uptake of cadmium and depletion of the photosynthetic pigments, leaf water contents, essential minerals, water-soluble proteins, and enzymatic and non-enzymatic antioxidants. Excessive cadmium also potentiated its toxicity by inducing oxidative stress, as evidenced by increased levels of superoxide, hydrogen peroxide, methylglyoxal and malondialdehyde. However, elevating endogenous H2S level improved physiological and biochemical attributes, which was clearly observed in the growth and phenotypes of H2S-treated rice plants under cadmium stress. H2S reduced cadmium-induced oxidative stress, particularly by enhancing redox status and the activities of reactive oxygen species and methylglyoxal detoxifying enzymes. Notably, H2S maintained cadmium and mineral homeostases in roots and leaves of cadmium-stressed plants. By contrast, adding H2S-scavenger hypotaurine abolished the beneficial effect of H2S, further strengthening the clear role of H2S in alleviating cadmium toxicity in rice. Collectively, our findings provide an insight into H2S-induced protective mechanisms of rice exposed to cadmium stress, thus proposing H2S as a potential candidate for managing toxicity of cadmium, and perhaps other heavy metals, in rice and other crops.

Scientific RepoRts | 5:14078 | DOi: 10.1038/srep14078 Cd, a non-redox water soluble heavy metal, can be quickly taken up by plant roots, and transported to the aerial parts where it significantly impedes vital cellular processes, including respiration and CO 2 fixation 1,5 . Chlorosis, necrosis, epinasty, stunted growth, cell death and disturbance in mineral homeostasis are the common consequences of Cd toxicity in plants [5][6][7] . At the cellular level, Cd can bind to sulfhydryl and carbonyl groups of proteins and can replace essential cofactors, resulting in enzyme inactivation and production of reactive oxygen species (ROS), leading to oxidative stress induced damage 1,6 . Additionally, Cd is known to induce the production of methylglyoxal (MG), an extremely reactive aldehyde that can modify proteins, nucleic acids and carbohydrates by forming cross linkages 8,9 . Moreover, MG in excess can further intensify ROS production by inactivating antioxidant enzymes or by interfering with photosynthesis 8,10,11 .
Plants have developed several strategies to counteract Cd toxicity, among which sulfur-induced defense is of paramount importance 1,12 . Sulfur-rich small ligands, such as cysteine, glutathione (GSH) and phytochelatins, directly bind with Cd and sequestrate it into vacuoles 13,14 . In addition, plant cells employ a number of GSH-based antioxidative reactions to scavenge Cd-induced superoxide (O 2 •− ), hydrogen peroxide (H 2 O 2 ) and hydroxyl radical (OH • ) 8,15 . GSH, with the help of ascorbate (AsA), plays central role in operating the AsA-GSH cycle in which ascorbate peroxidase (APX) detoxifies H 2 O 2 in coordination to monodehydroascorbate reductase (MDHAR), dehydroascorbate reductase (DHAR) and glutathione reductase (GR) 16,17 . Glutathione peroxidase (GPX) and glutathione S-transferase (GST) also detoxify H 2 O 2 , lipid hydoperoxides and reactive aldehydes by conjugating them with GSH 18 . Likewise, plant cells eliminate MG via glyoxalase (Gly) pathway by which Gly I and Gly II enzymes act coordinately with GSH 8,9 . However, it has been reported that Cd-overloads deplete cytosolic GSH which impinges GSH/GSSG redox potential, causing enhanced Cd sensitivity in several plant species [19][20][21] .
Hydrogen sulfide (H 2 S) has traditionally been thought of a phytotoxin but recently, it has emerged as an important gasotransmitter alongside with nitric oxide and carbon dioxide with multiple functions in animals and plants 22 . H 2 S, at tiny doses, can enhance plant growth and development, leading to a sharp increase in global food supplies and plentiful stock for biofuel production 23 . Despite enormous prospects of H 2 S in sustainable plant agriculture, few studies have focused on H 2 S biology in plant systems compared with those in animals.
H 2 S plays multiple roles in regulating key physiological processes, such as root organogenesis 24 , photosynthesis 25 , seed germination 23 and stomatal apertures 26 . In fact, H 2 S has received increasing attentions due to its role in orchestrating plant responses to various environmental cues 22,27,28 . Treatment with H 2 S donor sodium hydrosulfide (NaHS) improved plant tolerance to various abiotic stresses including heavy metals in wheat 29 , cucumber 30 , barley 31 and brassica 32 . However, mechanisms regulating H 2 S functions in plant responses to heavy metal stress, especially in a major crop like rice, are still elusive.
Thus, in the current study, we have examined the regulatory role of H 2 S in rice tolerance to Cd stress. Towards this objective, a combined pharmacological, physiological, histochemical and biochemical approach was employed to identify H 2 S-controlled Cd stress tolerance mechanisms by assessing the following key components, including (i) plant growth and biomass, (ii) Cd uptake and accumulation, (iii) mineral homeostases, (iv) excessive Cd-induced oxidative stress in terms of ROS and MG levels, and lipid peroxidation, (v) photosynthesis, (vi) ascorbate, glutathione contents, and redox homeostases and (vii) the activities of the enzymes of the antioxidant defense and Gly systems in economically important rice plants under Cd stress. Additionally, this study focused on the relationship between H 2 S-induced GSH homeostases and the activities of GSH-metabolizing enzymes to evaluate their roles in alleviating Cd stress-triggered adverse effects in rice plants.

H 2 S suppresses Cd uptake and accumulation to overcome Cd toxicity. The role of H 2 S in Cd
homeostasis was investigated by measuring the levels of Cd in the roots and leaves of Cd-stressed rice plants supplemented with or without NaHS. Results in Fig. 1 showed that Cd uptake and accumulation increased in roots and leaves in a dose-dependent manner, and roots were the major organ in which most of the Cd deposited. H 2 S supplementation exhibited an inhibitory effect on Cd uptake and accumulation. Cd accumulation was decreased by 20 and 19% in roots and 23 and 41% in leaves of H 2 S + Cd1 and H 2 S + Cd2 groups, respectively, as compared with the corresponding Cd1 and Cd2 groups (Fig. 1a,b). As a consequence, Cd content was reduced by 26 and 30% in roots, and 30 and 65% in leaves of H 2 S + Cd1 and H 2 S + Cd2 groups (Fig. 1c,d). Furthermore, the H 2 S-mediated reduction in Cd uptake and accumulation was significantly minimized in NaHS and hypotaurine (HT)-treated H 2 (Table 1). Furthermore, NaHS addition significantly increased the contents of all these minerals in both roots and leaves of H 2 S group compared with control group.
Exogenous H 2 S improves phenotypic appearance. To investigate the possible effects of H 2 S in alleviating Cd toxicity, growing rice seedlings were initially exposed to three different concentrations of CdCl 2 (250, 500 and 1000 μ M) in the presence or absence of NaHS. Visual toxicity symptoms, including stunted growth, chlorosis and leaf rolling, gradually pronounced by Cd stress treatment in a Cd concentration-dependent manner, as compared with the untreated control (Fig. 2a). Exogenous NaHS almost nullified the toxicity symptoms in H 2 S + Cd1 group and those symptoms were drastically reduced in H 2 S + Cd2 group. However, the toxicity symptoms were most severe in Cd3 group and NaHS-mediated recovery was not very strong in H 2 S + Cd3 group to sustain normal growth. Therefore, the Cd3 and H 2 S + Cd3 groups were discontinued in further investigations. Application of HT, on the other hand, abolished the beneficial effects of H 2 S, as the toxicity symptoms in H 2 S + HT + Cd2 group were similar to that of Cd2 group, which implies that H 2 S released from NaHS was effective in alleviating Cd toxicity in rice plants (Fig. 2b). Moreover, no phenotypic variations were observed between control, Control, H 2 S, H 2 S + HT, Cd1, H 2 S + Cd1, Cd2, H 2 S + Cd2, and H 2 S + HT + Cd2 correspond to the group of seedlings exposed to only nutrients, 100 μ M NaHS, 100 μ M NaHS + 200 μ M hypotaurine, 250 μ M CdCl 2 , 100 μ M NaHS + 250 μ M CdCl 2 , 500 μ M CdCl 2 , 100 μ M NaHS + 500 μ M CdCl 2 and 100 μ M NaHS + 200 μ M hypotaurine + 500 μ M CdCl 2 , respectively. Bars represent standard deviation (SD) of the mean (n = 3). Different letters (a-f) indicate significant differences among the treatments at P < 0.05, according to Duncan's multiple range test. DW, dry weight.
Scientific RepoRts | 5:14078 | DOi: 10.1038/srep14078 H 2 S and H 2 S + HT groups, suggesting that NaHS itself or co-application of HT and NaHS did not exert any toxic effect on the phenotypes of the rice plants ( Fig. 2a,b).

Application of NaHS enhances endogenous level of H 2 S.
In order to ascertain the beneficial effect of exogenous NaHS, we estimated the endogenous level of H 2 S in the presence and absence of Cd stresses. Results in Fig. 2c showed that NaHS application increased H 2 S level by 55% in H 2 S group but did not affect the level in H 2 S + HT group when compared with the untreated control. Cd stress also resulted in the increment of H 2 S content by 25 and 27% in Cd1 and Cd2 groups over the control level. Adding NaHS caused a further increase in the level of H 2 S by 27 and 35% in H 2 S + Cd1 and H 2 S + Cd2 groups, as compared with Cd1 and Cd2 groups, respectively. However, adding HT with NaHS decreased H 2 S level by 22% in H 2 S + HT + Cd2 group compared with H 2 S + Cd2 group, clearly indicating the increased level of endogenous H 2 S was attributed to exogenous NaHS. This elevated level of H 2 S is correlated with the enhanced Cd tolerance as reflected in the phenotypes of the rice plants treated with Cd and NaHS (Fig. 2a,b).
Positive effects of H 2 S on growth and biomass. The effect of H 2 S on the growth parameters of rice plants under Cd stress were investigated in terms of plant height and biomass (FW and DW). As expected, plant height was reduced by 23 and 39% in Cd1 and Cd2 groups, respectively, as compared with control group (Fig. 2d). Application of NaHS with Cd reduced the negative effect of Cd stress on plant height, as it was recovered by 21 and 24% in H 2 S + Cd1 and H 2 S + Cd2 groups compared with Cd1 and Cd2 groups respectively. However, H 2 S-mediated plant height recovery was reduced by 11% in H 2 S + HT + Cd2 group when compared with H 2 S + Cd2 group (Fig. 2d). In comparison with control group, plant FW was decreased by 10 and 24%, and DW by 11 and 20% in Cd1 and Cd2 groups respectively. On the other hand, adding NaHS with Cd diminished the negative impact of Cd stress on plant biomass where FW was increased by 15 and 21%, and DW by 19 and 21% in H 2 S + Cd1 and H 2 S + Cd2 groups, as compared with Cd1 and Cd2 groups, respectively. Again, H 2 S-mediated plant biomass recovery in H 2 S + HT + Cd2 group was reduced by 17% (FW) and 13% (DW), as compared with H 2 S + Cd2 group (Fig. 2e, on the losses of photosynthetic pigments and soluble protein contents in H 2 S + HT + Cd2 group was significantly reduced due to simultaneous application of NaHS and HT (Table 2). Water soluble proteins significantly declined in H 2 S + HT group but remained unchanged in H 2 S group compared with control group.

H 2 S maintains water status and inhibits endogenous Pro accumulation. Leaf RWC was
decreased by 17 and 26% in Cd1 and Cd2 groups, respectively, as compared with the untreated control (Table 2). However, the observed 17 and 19% increase in leaf RWC in H 2 S + Cd1 and H 2 S + Cd2 groups relative to that of Cd1 and Cd2 groups, respectively, indicated that exogenous NaHS significantly prevented the reduction of RWC in rice plants under Cd stress conditions. On the other hand, H 2 S-mediated restoration of leaf RWC significantly reversed in H 2 S + HT + Cd2 group (Table 2). Being an osmoprotectant, Pro accumulates in response to water shortage in plants under abiotic stresses, which might help plant adapt better to water deficit 33 . Cd stress resulted in a significant increase in Pro content in the leaves of Cd-stressed plants, which reached the maximum in Cd2 group (768%), when compared with the untreated control. NaHS addition decreased Pro content by 8 and 73% in H 2 S + Cd1 and H 2 S + Cd2 groups in comparison with Cd1 and Cd2 groups, respectively. However, simultaneous application of HT
and NaHS increased Pro content by 159% in H 2 S + HT + Cd2 compared with H 2 S + Cd2 group, which coincided with the decrease in the level of RWC as well (  16 . A drastic rise in the levels of H 2 O 2 was recorded in Cd1 (40%) and Cd2 (75%) groups compared with the untreated control (Table 3). LOX, as an oxidative enzyme, often participates in lipid peroxidation contributing to oxidative stress 34 . LOX activity was also found to be significantly increased in Cd1 and Cd2 groups compared with control group. As a consequence, MDA level was markedly increased by 49 and 204% in Cd1 and Cd2 groups, respectively, when compared with control group (Table 3). On the other hand, adding NaHS reduced Cd-induced oxidative stress as evident by 26 (Fig. 3a,b). Adding NaHS alone or in combination with HT in the absence of Cd stress did not alter the generation of O 2 •− and H 2 O 2 compared with the untreated control.
Evidence of H 2 S-mediated protection of membrane integrity. Increased ROS production is supposed to disturb membrane integrity through lipid peroxidation 35 . Schiff 's test was employed to assess the ability of H 2 S to protect membrane from Cd-induced lipid peroxidation. Lipid peroxidation was considerably increased, as observed by intense red color, in the roots of rice plants under Cd stress when compared with the untreated control (Fig. 3c). However, NaHS application was found to reduce the lipid peroxidation in the roots of Cd-stressed rice plants, which is comparable to that of the untreated control.
A parallel experiment about the staining of the roots with Evan's blue was carried out to confirm the loss of membrane integrity due to lipid peroxidation. We observed that Cd stress resulted in an increased uptake of Evan's blue, as indicated by deep blue color, by the roots of the rice plants under Cd stress alone (Fig. 3d). In contrast, adding NaHS with Cd considerably decreased the root uptake of Evan's blue when compared with Cd stressed roots only. Moreover, application of NaHS and HT with Cd showed a severe disturbance in plasma membrane integrity as evidenced by higher stained intensity of roots (Fig. 3c,d). Collectively, these results suggest a healing effect of H 2 S for plasma membrane of rice roots under Cd stress conditions. H 2 S efficiently elevates AsA and GSH levels, and redox status. A noticeable reduction of AsA level (23 and 47% in Cd1 and Cd2 groups, respectively) and increment of DHA (17% in Cd2 group) were observed, as compared with the untreated control (Table 4). However, AsA/DHA ratio significantly decreased in these stress groups (Cd1 and Cd2). Adding NaHS increased AsA level by 22 and 38% and AsA/DHA ratio by 19 and 136% in H 2 S + Cd1 and H 2 S + Cd2 groups in comparison with Cd1 and Cd2 groups, respectively. However, AsA level decreased by 30%, DHA content increased by 103%, and AsA/ DHA ratio decreased by 66% in H 2 S + HT + Cd2 group, as compared with H 2 S + Cd2 group. Although GSH content significantly increased in Cd1 group, it was drastically decreased by 32% in Cd2 group in comparison with the untreated control (Table 4). GSSG content increased by 23% and GSH/GSSG ratio decreased by 45% in Cd2 group compared with control group. NaHS addition did not affect GSH content in H 2 S + Cd1 group but it increased GSH content by 124% and GSH/GSSG ratio by 304% in H 2 S + Cd2 group as compared with Cd2 group (Table 4). Additionally, H 2 S-mediated increase in the level of GSH and GSH/GSSG ratio significantly reversed in H 2 S + HT + Cd2 group when compared with H 2 S + Cd2 group. Thus, H 2 S elevated the levels of AsA and GSH, and thus the redox status, to boost the antioxidant capacity of Cd-stressed rice plants.

H 2 S upregulates various antioxidant enzymes to counteract oxidative stress.
To explore the regulatory role of H 2 S in alleviation of Cd-induced oxidative stress, the activities of various antioxidant enzymes were investigated (Fig. 4a-h). SOD activity was increased by 18 and 43% in Cd1 and Cd2 groups, respectively, whereas CAT activity increased by 16% in Cd1 group and then declined by 17% under higher concentration of Cd (Cd2 group) as compared with control group. Applying H 2 S showed an increased in SOD (16 and 18%) and CAT (28 and 70%) activities in H 2 S + Cd1 and H 2 S + Cd2 groups, respectively, when compared with their respective Cd1 and Cd2 groups, which corroborates the levels of O 2 •− and H 2 O 2 (Figs 3a,b and 4a,b). As expected, combined application of NaHS and HT significantly decreased the activities of SOD and CAT in H 2 S + HT + Cd2 group when compared with H 2 S + Cd2 group.
The ascorbate-glutathione cycle contributes to reduce oxidative stress by detoxifying H 2 O 2 through multistep enzymatic reactions 35 . APX and MDHAR activities increased in an intensity-dependent manner in the leaves of Cd-stressed rice plants (Fig. 4c,d). In comparison with the untreated control, DHAR and GR activities were increased by 45 and 26%, respectively, in Cd1 group but their activities did not increase further upon increasing the level of Cd concentration (Cd2 group) (Fig. 4e,f). Applying NaHS did not show boosting effect on APX and MDHAR activities but further increased DHAR (11%) and GR (27%) activities in H 2 S + Cd2 group relative to Cd2 group. Meanwhile, co-treatment of NaHS and HT caused a similar pattern of APX and MDHAR activities in Cd2 group but significantly curtailed   the activities of DHAR and GR in H 2 S + HT + Cd2 group when compared with H 2 S + Cd2 group. No significant differences in the activities of APX, MDHAR, DHAR and GR were observed in NaHS and H 2 S + HT groups when compared with the untreated control ( Fig. 4c-f). GSH metabolizing enzymes GPX and GST showed differential responses under Cd stress; that is, GPX activity increased in a Cd concentration-dependent manner, whereas GST activity drastically decreased by 43% in Cd2 group in comparison with untreated control (Fig. 4g,h). Exogenously applied H 2 S exhibited a further increase in GPX activity by 32 and 18% in H 2 S + Cd1 and H 2 S + Cd2 groups, respectively, while it reduced the loss of GST activity in H 2 S + Cd2 group. Co-application of NaHS and HT significantly decreased the activities of GPX and GST in H 2 S + HT + Cd2 group when compared with H 2 S + Cd2 group. Following NaHS addition alone, GPX activity showed significant enhancement but GST activity remained unchanged in H 2 S group as compared with respective control groups. H 2 S attenuates MG toxicity by enhancing the activities of Gly enzymes. The reactive aldehyde MG is detoxified by the maintenance of GSH homeostasis via Gly enzymes 8,9 . MG level increased gradually as the Cd concentration increased; that is 38% in Cd1 group and 84% in Cd2 group, as compared with control group (Fig. 5a). A significant inhibition in MG production was observed in H 2 S + Cd1 and H 2 S + Cd2 groups when compared with Cd1 and Cd2 groups, respectively. Moreover, co-supplementation of NaHS and HT increased the level of MG by 25% in H 2 S + HT + Cd2 group compared with H 2 S + Cd2 group. In the presence or absence of NaHS, MG detoxifying enzymes exhibited differential responses in Cd-stressed seedlings (Fig. 5b,c). Gly I activity was increased by 25 and 76% in Cd1 and Cd2 groups, respectively, as compared with the untreated control. Applying NaHS resulted in an increase in Gly I activity by 13% in H 2 S + Cd1 group and a decrease by 6% in H 2 S + Cd2 group, as compared with Cd1 and Cd2 groups, respectively. Insignificant change in Gly II activity was observed in Cd1 group but the activity was reduced by 21% in Cd2 group relative to control group (Fig. 5c). NaHS addition showed noticeable increase in Gly II activity by 43 and 64% in H 2 S + Cd1 and H 2 S + Cd2 groups when compared with Cd1 and Cd2 groups, respectively. On the other hand, adding H 2 S scavenger HT displayed insignificant change in Gly I activity and a decrease in Gly II activity by 32% in H 2 S + HT + Cd2 group compared with H 2 S + Cd2 group (Fig. 5b,c).

Discussion
Among the various heavy metals, Cd is considered non-essential and highly toxic, adversely affecting growth, development and quality of lives of all living organisms, including plants 2,13 . Nowadays, rapid industrialization and excess use of fertilizers and pesticides in agriculture greatly contribute to the deposition of Cd in water and soils 2,4,14 . Once contamination occurred, Cd is readily accumulated in the aboveground parts of plants, thereby entering food chains and threatening human health worldwide 1,2 . Plant cells have evolved several defense strategies in restricting Cd toxicity, including (i) inhibition of uptake by exudates like organic acids, (ii) exclusion by sugar alcohols, (iii) chelation, conjugation and sequestration into vacuoles by strong ligands like GSH and phytochelatins and (iv) expression of anti-stress genes associated with thiol metabolism and antioxidant defense systems 1,8,14 . In this study, we have provided evidence that NaHS addition increased the level of endogenous H 2 S, which in turn modulates physiological and biochemical mechanisms associated with Cd stress tolerance in hydroponically raised rice plants. Furthermore, the alleviating effects of H 2 S were confirmed by adding the H 2 S scavenger HT, which dismounted the level of H 2 S and subsequently negated all the positive effect of H 2 S (Tables 1-4 and Figs 1-6).
Results indicated that Cd contents in the Cd-challenged rice plants increased in a Cd-concentration dependent manner, which was also correlated with the severity of the damage induced by excessive Cd (Figs 1 and 2a,b). Roots are the primary armaments that plants deploy to accumulate most of the heavy metals exposed [36][37][38][39] , as was also observed in this study. However, excessive Cd can have detrimental effects on root architecture, which affects plants' capacity to absorb water and minerals 7 . Our results also showed that excessive Cd accumulation considerably reduced the uptake of several essential minerals in the roots and leaves of rice plants with the exception of Ca and Fe in the roots, whose content was increased upon increasing Cd concentrations (Table 1). More importantly, Cd stress significantly decreased the leaf contents of these minerals, indicating that the complex interactions between toxic and nutrient elements interfere with the mechanism of root-to-shoot transfer of these elements in the plants. In line with the finding of Ali et al. 32 , this result indicated that Cd-induced disturbance in mineral homeostasis hampered mineral-driven biochemical events in Cd-stressed rice plants. However, NaHS application significantly restricted the uptake and accumulation of Cd, thereby minimizing antagonistic effects of Cd on essential mineral transportation in rice plants ( Fig. 1 and Table 1). It is plausible that co-application of CdCl 2 and NaHS might produce non-toxic cadmium sulfide (CdS) to some extent, which partly contributed in decreasing cellular levels of Cd. This pattern of Cd homeostasis explained a specific mechanism that restored the capability of plants to accumulate essential nutrients for normal metabolic functions, and at the same time avoid Cd toxicity by keeping its level below the toxicity threshold (Fig. 6). This finding is supported by Dawood et al. 31 , who investigated mineral homeostases in Hordeum vulgare under the combined treatment of aluminum and NaHS.
In relation to the increased levels of Cd in the roots and leaves, growth and biomass of Cd-challenged rice plants in terms of plant height, FW and DW were greatly suppressed (Fig. 2d-f), as was reported in other plant species 38,39 . Cd-caused reduction in the level of the photosynthetic pigments and water soluble proteins also accorded with the reduced plant growth and biomass (Table 2 and Fig. 2), suggesting that Cd impaired photosynthetic ability by disrupting chloroplast structure, Chl-protein complexes and perhaps by deactivating the enzymes of Calvin cycle 5,40 . H 2 S, on the other hand, alleviated Cd-induced decline in Chl a, Chl b, carotenoids and proteins contents, thereby improving overall growth and biomass of Cd-stressed rice plants. The possible biochemical mechanism of H 2 S in improving rice growth could partly be attributed to the enhancement of Chl biosynthesis by delivering Mg, Fe, Zn and Mn ions and protection of Chl pigments from ROS by enhancing the level of carotenoids. Additionally, H 2 S, as a signaling molecule, can improve photosynthetic performance by promoting biogenesis of chloroplast and increasing the ability of CO 2 fixation as observed in Spinacea oleracea 25 . Plant-water imbalance and accumulation of osmolytes like Pro are general consequences of heavy metal toxicity, including Cd 33,38 . In this study, we found that leaf RWC and Pro levels were inversely related under Cd stress, whereas H 2 S addition restored RWC without accumulating much Pro (Table 2). These data suggested that H 2 S likely employed other metabolic adjustment(s) in which high level of Pro accumulation was not needed. Importantly, the observed enhancement of RWC (Table 2) might lead to an enhanced stomatal conductance and transpiration in H 2 S-treated Cd-stressed rice plants which in turn contributes to the improvement of photosynthesis 41,42 , thereby enhancing the overall performance of the rice plants.
In our study, the rice plants treated with Cd exhibited a severe oxidative stress in leaf tissues as evident by increased level of H 2 O 2 and overproduction of ROS like O 2 •− and H 2 O 2 (Table 3 and Fig. 3a,b). This Cd-induced ROS together with increased LOX activity were correlated with the significant increase in MDA level (Table 3), indicating that Cd potentiated oxidative injuries by peroxidizing membrane lipids. Such effects were further confirmed by in vivo staining of peroxidation of lipid and injury of plasma membrane integrity in root tissues (Fig. 3c,d). Although Cd is unable to generate ROS via Haber-Weiss Figure 6. A Schematic diagram representing Cd-induced toxicity and protective mechanism of H 2 S underlying rice tolerance to Cd stress. Cd stress exerted harmful effects by inhibiting photosynthesis and affecting water status and ionic balance. Excessive Cd can induce reactive oxygen species (ROS) production by depleting antioxidant levels. Cd can also overaccumulate methylglyoxal (MG) that is highly toxic to DNA, RNA and proteins. ROS and MG contribute to oxidative stress, resulting in lipid peroxidation, DNA and protein oxidation, all of which cause growth inhibition, biomass loss and ultimate yield reduction. In contrast, elevated H 2 S level following sodium hydrosulfide (NaHS) application showed protective mechanism against Cd-induced toxicity by re-establishing Cd homeostasis, ionic balance and water balance, and by enhancing photosynthesis. H 2 S enhanced the detoxification of ROS by stimulating the activities of SOD and CAT. H 2 S also replenished the Cd-induced reduction in glutathione (GSH) level, which in turn enhanced the ROS and MG detoxifications as well as redox homeostases, leading to alleviation of oxidative stress. As a consequence, H 2 S enhanced Cd stress tolerance by retaining better growth of Cd-stressed rice plants. or Fenton reactions, it can induce ROS production, and thus oxidative stress, by increasing the activity of plasma membrane NADPH oxidases, depleting the non-enzymatic antioxidants, inhibiting/down-regulating ROS-detoxifying enzymes and disrupting electron transport chain 43,44 . Plants have evolutionally equipped with ROS-detoxifying techniques and the enhanced productions and synchronized actions of various antioxidants govern the balance between ROS generation and detoxification in plant cells under stress conditions 16,17,36,45 . The current study revealed that Cd, particularly at higher concentrations, seriously impaired ROS detoxification mechanism by decreasing the contents of AsA and GSH, as well as the redox states in terms of AsA/DHA and GSH/GSSG ratios (Table 4 and Fig. 6). Moreover, unsynchronized activities of SOD and CAT (Fig. 4a,b), which constitute the frontline enzymatic network by converting O 2 •− and H 2 O 2 consecutively into H 2 O 16 , further accelerated the accumulation of ROS in Cd-stressed rice plants (Fig. 3a,b). Importantly, we observed that CAT activity increased at lower concentration, but significantly decreased at higher concentration of Cd. Similar results were obtained in different plant species, including rice under Cd stress 38,41,[46][47][48] . The possible mechanism might be that at low concentration, Cd stimulates CAT activity to enhance the basal antioxidant capacity to overcome oxidative stress. However, at higher concentration, overaccumulation of Cd inhibits CAT activity, possibly by replacing the Fe from the active center of CAT 7 . It was also suggested that Cd applied to plants at higher doses and/ or for longer exposure period inhibits the synthesis of CAT or downregulates CAT activity or inactivates CAT in plants, mainly by disturbing the assembly of different subunits of CAT 49 . On the other hand, H 2 S-induced amendment of AsA and GSH levels and redox states as well as enhancement of SOD and CAT activities (Table 4 and Figs 3a-d and 4a,b) augmented the detoxification of ROS, thereby demonstrating an efficient role of H 2 S in tailoring major antioxidant defense system to protect the cells from oxidative damage.
The ascorbate-glutathione cycle involves a series of reactions with four enzymes (APX, MDHAR, DHAR and GR) that act coordinately in removing H 2 O 2 and maintaining cellular redox balance, especially under stress conditions 16,17,35 . In this study, despite the stimulation of the activities of all four enzymes upon Cd stress, the H 2 O 2 level remained significantly higher (Table 3 and Fig. 4c-f), which indicated that their stimulation was not up to the requisite level in lowering excessive H 2 O 2 . In consistent with our findings, Dixit et al. 19 also observed an accumulation of H 2 O 2 even after increasing the activity of H 2 O 2 scavenging enzymes in roots and leaves of Pisum sativum, suggesting that production of H 2 O 2 exceeded ROS scavenging capacity. In contrast, H 2 S, by intensifying DHAR and GR activities as well as maintaining APX and MDHAR activities above the control level, coordinates the activities of these four enzymes in the presence of enhanced SOD and CAT activities, thereby contributing well to the regulation of ROS level (Figs 4a-f, 3a,b, 6 and Table 3). This antioxidative role of H 2 S supported by the observation that H 2 S increased the activity of antioxidant enzymes, including SOD, CAT, DHAR and GR and the abundance of their corresponding transcripts, which were responsible for enhanced salt tolerance in Medicago sativa 28 . Interestingly, we observed that NaHS addition increased H 2 O 2 content by 12% under non-stress condition; however, MDA content remained unchanged. This finding indicated that the increase in H 2 S level did not exert oxidative stress (Table 3). Thus, we assumed that this slight increase in the amount of H 2 O 2 might stimulate the signaling role of H 2 S, as H 2 O 2 is a well-known signaling molecule with positive regulatory role at lower cellular concentration 50 .
Another important mechanism in regulating heavy metal-induced toxicity is associated with the GSH-dependent conjugation of lipid hydroperoxides and endobiotic substrates by GPX and GST 8,18 . Our findings indicated that higher DHAR and GPX activities together with inefficient recycling of GSSG by GR might have contributed to the Cd-induced depletion of GSH and GSH/GSSG ratio, which consequently distressed the activity of GST (Fig. 4e-h and Table 4). However, a careful analysis of GSH-related defense mechanism under H 2 S addition revealed that H 2 S replenished the depleted GSH level, possibly by inducing its biosynthesis or by upregulating GR activity under Cd stress, which then participates in redox homeostasis as well as GPX and GST-mediated efficient scavenging of peroxides and other electrophiles (Fig. 6). This finding together with reduced LOX activity (Table 3) suggested a membrane protecting role of H 2 S under Cd stress in rice plants. Our results are consistent with those reported by Cui et al. 51 , who found that H 2 S-induced GSH and ROS homeostases were involved in enhancing alfalfa tolerance to Cd stress.
Additionally, the current study showed encouraging results on plant defense against MG toxicity under combined application of NaHS and Cd. Several reports 25,28,29,31 , including this study, proved the regulatory role of H 2 S on antioxidant defense; however, its role in GSH-based MG detoxification has remained to be determined in plant systems. We noticed a correlation between increased MG levels and inefficient Gly system in Cd-stressed plants, as the activity of Gly II decreased upon increasing the concentration of Cd (Fig. 5a-c), thereby intensifying Cd toxicity in rice plants. On the other hand, the observed lower contents of MG in H 2 S-and Cd-treated plants were associated with enhanced activities of Gly I and Gly II, which indicated a positive effect of H 2 S in amplifying MG detoxification (Fig. 6). This finding also suggested that enhanced Gly II activity efficiently recycled GSH into the system, which facilitated GSH homeostasis and higher activities of DHAR, GPX and GST (Table 4 and Figs 4f-h, 6) in preventing oxidative stress. It has also been reported that the enhanced activities of Gly enzymes and antioxidant enzymes (APX, GST and GPX) provided higher tolerance to Zn toxicity in Brassica oleracea 34 . Thus, our results suggested that H 2 S coordinates the biochemical actions of the antioxidant defense and Gly systems to mitigate the Cd-induced ROS and MG toxicity in rice plants (Fig. 6).
Based on all the above observations, a schematic diagram was depicted in Fig. 6, which illustrates the physiological and biochemical mechanisms involving Cd-induced toxicity and the H 2 S-mediated Cd stress tolerance in rice plants. Our results provided strong evidence that H 2 S successfully alleviated Cd-induced growth inhibition and biomass loss, which was mainly attributed to (i) decreased Cd uptake and accumulation, (ii) enhanced mineral homeostases and photosynthetic pigments and (iii) reduced oxidative stress through inducing ROS and MG detoxifications by maintaining redox homeostases. Taking into account that heavy metal tolerance requires a coordination of complex physiological and biochemical processes, this study, therefore, advances our understanding of the complex system integrating rice responses to Cd stress, and perhaps other environmental stresses. However, molecular approaches using H 2 S-biosynthetic mutants should be considered in future studies to decipher the exact role of H 2 S in modulating tolerance mechanism in crop plants under Cd stress.

Methods
Plant materials, growth conditions and treatments. The germination and cultivation of rice (Oryza sativa L. cv. BRRI dhan52) in a hydroponic condition were carried out according to the method described by Mostofa et al. 52,53 and Supplementary Fig. S1 online. The hydroponic culture consisted of a hyponex solution, which contains 8% N, 6.43% P, 20.94% K, 8% S, 11.8% Ca, 3.08% Mg, 0.07% B, 0.24% Fe, 0.03% Mn, 0.0014% Mo, 0.008% Zn, and 0.003% Cu. The hyponex solution was diluted to 5,000 fold and renewed every three days. Uniformly grown 13-d-old seedlings were exposed to nutrient solution supplemented with 250, 500 and 1000 μ M of CdCl 2 in the presence and absence of H 2 S donor sodium hydrosulfide (NaHS, 100 μ M). These environmentally relevant CdCl 2 concentrations were selected based on previous reports 4, 46 . To verify the role of H 2 S released from NaHS, we applied H 2 S scavenger hypotaurine (HT, 200 μ M) 28,47 with 500 μ M CdCl 2 and 100 μ M NaHS. Based on literature 32,47,48 and our preliminary experiments with a range of NaHS concentrations (25, 50, 100, 150 and 200 μ M), we observed that 100 μ M NaHS was optimally effective in alleviating Cd-induced toxic symptoms. We also observed that 100 μ M NaHS was not effective in alleviating the toxic effects of high dose of CdCl 2 (1000 μ M CdCl 2 ) (Fig. 1a). Therefore, our experimental design consisted of eight treatments as follows: (i) Control, (ii) 100 μ M NaHS (H 2 S), (iii) 100 μ M NaHS + 200 μ M hypotaurine (H 2 S + HT), (iv) 250 μ M CdCl 2 (Cd1), (v) 100 μ M NaHS + 250 μ M CdCl 2 (H 2 S + Cd1), (vi) 500 μ M CdCl 2 (Cd2), (vii) 100 μ M NaHS + 500 μ M CdCl 2 (H 2 S + Cd2) and (viii) 100 μ M NaHS + 200 μ M hypotaurine + 500 μ M CdCl 2 (H 2 S + HT + Cd2) (see Supplementary Table S1 online). The seedlings were harvested after 3 days of Cd stress treatments to investigate the H 2 S-induced mechanisms modulating physiological and biochemical responses. Each treatment was replicated three times under the same experimental conditions. Growth parameters. The growth of rice seedlings was assessed by measuring plant height, fresh weight (FW) and dry weight (DW). Plant height was determined by measuring the length from the bottom of the main stem to the end of the emerging third leaf. To determine FW and DW, 10 seedlings from each treatment were weighed and oven-dried at 80 °C for 48 h, then expressed as g seedling −1 .
Determination of H 2 S content. H 2 S content in rice leaves was determined following the method described by Christou et al. 54 with modifications. Rice leaves (0.25 g) were homogenized in 1 mL of 100 mM K-P buffer (pH 7.0) containing 10 mM EDTA and centrifuged at 11,200 × g for 15 min. The supernatant (100 μ L) was mixed with 1,880 μ L extraction buffer and 20 μ L of 20 mM 5,5′ -dithiobis (2-nitrobenzoic acid) and incubated at 25 °C for 5 min. The absorbance was read at 412 nm, and H 2 S was quantified using a standard curve developed with known concentrations of NaHS.

Measurement of Cd and mineral nutrient contents in roots and leaves.
To determine Cd and mineral nutrient contents (Ca, Mg, Fe, Zn and Mn), the root and leaf samples were harvested separately and oven-dried at 80 °C for 48 h. Dried samples (0.1 g) were ground and digested with a HNO 3 :HClO 4 (5:1 v/v) mixture at 80 °C until the yellow color disappeared. The Cd, Ca, Mg, Fe, Zn and Mn contents were measured by using flame atomic absorption spectrophotometry.
Relative water content and proline content. Relative water content (RWC) was determined as described by Mostofa et al. 52 . Pro content was determined according to the method of Bates et al. 55 with minor modifications as reported in Mostofa and Fujita 36 .
Photosynthetic pigments. After extracting 0.5 g leaves in 80% chilled acetone, the absorbance of the supernatants was recorded at 663, 645 and 470 nm for determining the contents of Chl and carotenoids according to the formulas suggested by Arnon 56  Histochemical analyses. For detecting lipid peroxidation and the loss of membrane integrity in roots, Schiff 's reagent test and Evan's blue test were performed, respectively, following the method of Lai et al. 28 . Histochemical analyses of two major ROS molecules O 2 •− and H 2 O 2 were carried out in rice leaves following the method of Mostofa and Fujita 36 . Extraction and estimation of non-enzymatic antioxidants. Fresh leaves (0.5 g) were homogenized in 3 mL of ice-cold 5% meta-phosphoric acid containing 1 mM EDTA and centrifuged at 11,500 × g for 15 min. Reduced and total AsA contents were assayed at 265 nm in 100 mM K-phosphate buffer (pH 7.0) with 1.0 U of ascorbate oxidase (AO) following the method of Mostofa and Fujita 36 . Oxidized ascorbate (DHA; dehydroascorbic acid) content was calculated by deducting reduced AsA amount from total AsA content. Based on enzymatic recycling, total GSH and oxidized glutathione (GSSG) contents were determined according to the method of Griffiths 59 . Reduced GSH content was measured after subtracting the value of GSSG from total GSH level.
Extraction and assay of enzymes. Extraction of enzymes was carried out following the method of Mostofa et al. 52,53 . Activities of antioxidant and glyoxalase enzymes were determined by the standard methods reported in Doderer et al. 60  Estimation of MG content. Following the method of Wild et al. 65 , leaves (0.5 g) were homogenized in 5 mL of 0.5 M perchloric acid and incubated for 15 min on ice. The extract was centrifuged for 10 min at 11,200 × g and 1 mL supernatant was transferred to a microcentrifuge tube containing 10 mg charcoal and kept at room temperature for 15 min. After centrifugation for 10 min at 11,200 × g, the supernatant was neutralized by saturated K 2 CO 3 and then centrifuged at 11,200 × g for 10 min. The neutralized supernatant (650 μ L), 100 mM potassium phosphate (K-P) buffer (pH 7.0, 330 μ L) and 0.5 M N-acetyl-L-cysteine (20 μ L) were added together and incubated for 15 min. Formation of N-α -acetyl-S-(1-hydroxy-2-oxo-prop-1-yl) cysteine was recorded at a wavelength of 288 nm and MG content was calculated by using a standard curve of known concentrations of MG.
Statistical analysis. The data were subjected to one-way analysis of variance (ANOVA) and different letters indicate significant differences between treatments at P < 0.05, according to Duncan's multiple range test (DMRT) using IRRISTAT version 3 (IRRI, Biometrics Unit, Manila, Philippines). Data represented in the Tables and Figures are means ± standard deviations (SD) of three replicates for each treatment.