d-Cysteine-Induced Rapid Root Abscission in the Water Fern Azolla Pinnata: Implications for the Linkage between d-Amino Acid and Reactive Sulfur Species (RSS) in Plant Environmental Responses

Reactive Oxygen Species (ROS) and Reactive Nitrogen Species (RNS) have been proposed as universal signaling molecules in plant stress responses. There are a growing number of studies suggesting that hydrogen sulfide (H2S) or Reactive Sulfur Species (RSS) are also involved in plant abiotic as well as biotic stress responses. However, it is still a matter of debate as to how plants utilize those RSS in their signaling cascades. Here, we demonstrate that d-cysteine is a novel candidate for bridging our gap in understanding. In the genus of the tiny water-floating fern Azolla, a rapid root abscission occurs in response to a wide variety of environmental stimuli as well as chemical inducers. We tested five H2S chemical donors, Na2S, GYY4137, 5a, 8l, and 8o, and found that 5a showed a significant abscission activity. Root abscission also occurred with the polysulfides Na2S2, Na2S3, and Na2S4. Rapid root abscission comparable to other known chemical inducers was observed in the presence of d-cysteine, whereas l-cysteine showed no effect. We suggest that d-cysteine is a physiologically relevant substrate to induce root abscission in the water fern Azolla.


Introduction
Plants sense environmental conditions and transmit the environmental signals to regulate their growth and development. It is a long standing question as to how plants sense and transmit a variety of environmental signals without nervous systems [1]. Stomatal movement is one of the rapid responses of plants, both opening and closure movements finish within an hour [2]. Because of this advantage, the stomatal movement has been a model system to be investigated for many years to answer this question [3].
It has been revealed that hydrogen peroxide (H 2 O 2 ) is involved in the stomatal closing response [4]. H 2 O 2 belongs to Reactive Oxygen Species (ROS) which can be produced in abiotic as well as biotic stress conditions [5]. Later, the Reactive Nitrogen Species (RNS) nitric oxide (NO) was found to induce stomatal movements [6,7]. The potential for cross talk between ROS and RNS has become a subject of debate regarding the signaling mechanism in guard cells [8,9]. Early in this century, Reactive Sulfur Species (RSS) was hypothesized as the third group of redox-active molecular species that may be associated with oxidative stress [1,10,11]. In mammalian studies, there is an increasing number of reports suggesting regulatory functions of H 2 S or RSS in cellular signaling mechanisms [12]. Accordingly, it has recently been reported that H 2 S also influences stomatal movements [13]. To date, however, there seems to be contradictory results for the functions of H 2 S in stomatal movements [14].

Researchers may experience technical difficulties in conducting H 2 S experiments with land plants.
In whole plant experiments, H 2 S has been delivered by fumigation of the gas [15], spraying of an NaHS solution [16,17], or inclusion of NaHS into a culture medium [18]. Pharmacological comparative studies may be difficult to conduct with those methods. To overcome such technical problems in plant H 2 S study, we suggest here the application of the water-floating fern Azolla, a good model plant to explore the physiological functions of chemical compounds [19][20][21].
Plants of the genus Azolla have been used for agriculture in East Asia as green manure [22]. Biological research on Azolla has a long history in botany [23], cell biology [24,25], and phytoremediation [26,27]. It has long been known that the roots of Azolla pinnata are deciduous [28]. In 1993, Kitoh and co-workers found that nitrite (NO 2 − ) and volatile organic acids such as acetate or propionate contained in swine waste water cause root shedding in Azolla filiculoides [29]. Uheda and Kitoh experimentally reproduced the root shedding with a variety of inhibitors, such as the hemeprotein inhibitors sodium azide (NaN 3 ) and sodium cyanide (NaCN), the uncouplers 2,4-dinitrophenol (DNP) and carbonyl cyanide m-chlorophenyhydrazone (CCCP), and the F 0 -F 1 ATPase inhibitor N ,N -dicyclohexylcarbodiimide (DCCD) [30]. The root shedding, caused by cell expansion in the abscission zone, was finished within an hour, a speed much faster than conventional abscission phenomena, which usually requires days. Since the cycloheximide treatment did not inhibit the root abscission induced by NaN 3 , they suggested that rapid root abscission does not need new protein synthesis, offering a novel type of plant abscission phenomenon [30].
Taking progress in NO studies into account, we speculated that NO production in cells treated with NO 2 − , NaN 3 , and DNP might have a role in the rapid abscission phenomenon. Thus, we reinvestigated their observations in terms of NO and RNS. In fact, A. pinnata was found to emit NO in the presence of NaN 3 and NO 2 − [21]. Nitrite is now appreciated as an endogenous NO substrate for plants [31] as well as animals [32,33]. Moreover, H 2 O 2 was found to exhibit a bimodal effect on rapid root abscission [34].
We have proposed a model involving an interplay between RNS and ROS in initiating free radical attack of polysaccharides in the apoplast to account for the rapid root abscission mechanism in Azolla plants [35]. Here, we reported the effects of novel H 2 S donors and polysulfides in induction of rapid root abscission that suggests RSS also have a role in the abscission mechanism.

Materials and Methods
Laboratory cultures of Azolla pinnata R, Br. were established from plants collected in April 2018 from a paddy field in Ogimi, Okinawa, Japan. The plants were thoroughly washed to remove attached mud and debris. The plants were then treated with a solution of 0.12% sodium hypochlorite and 0.01% Triton X-100 for 30 min followed by repeated washings in a large volume of distilled water and finally transferred into nutrient culture medium [20]. A. pinnata was cultured in a two-fifth strength cobalt-supplemented nitrogen-source-free Hoagland's E-medium [20]. Medium pH was adjusted to 5.8 with potassium hydroxide (KOH). Plants were grown in a plant growth chamber (Type FLI-2000 H, Eyla, Tokyo, Japan) maintained at 27 • C, 16:8 h light:dark photoperiod and 50 µmol m −2 s −1 (at plant level) provided by fluorescent lamps (Type FL 40 SBR-A, NEC, Tokyo, Japan). For experiments, fronds were randomly selected from the culture stock and de-rooted manually using forceps. Rootless fronds were placed in the culture medium after rinsing in distilled water and transferred to fresh mediums every 7 days.
Abscission assays were carried out using roots of equal age (i.e., from fronds that had been de-rooted at the same time, 7 days prior to preforming the assay). Three to six fronds (with 20-30 roots) were suspended in a beaker containing 20 mL 10 mM Hepes-KOH (pH 7) or the culture medium (pH 5.8). The abscission test was carried out at room temperature (24 • C) under room light. The chemicals to be tested, the H 2 S chemical donors Na 2 S, GYY4137, 5a, 8l, and 8o (Dojindo Laboratories, Kumamoto, Japan) and the polysulfides Na 2 S 2 , Na 2 S 3 , Na 2 S 4 (Dojindo Laboratories, Kumamoto, Japan) were subsequently supplied as concentrated stock solutions according to the instruction manuals. The total number of dropped roots following addition of the chemicals was recorded. The abscission response was quantified as the ratio of the detached to the initial number of roots. H 2 S gas was measured with a handheld O 2 /H 2 S monitor (XOS-326, New Cosmos Electric, Osaka, Japan). The H 2 S that was released into the headspace (10 mL) of a beaker was monitored for 30 s. The H 2 S releasing activity of each solution was expressed as ppm/min.

Effects of H 2 S Donors on the Root Abscission in Azolla
To investigate biological functions of H 2 S, the application of H 2 S donors would be the first choice for physiological experiments. Sodium hydrosulfide (NaHS) and sodium sulfide (Na 2 S) are inorganic compounds that release H 2 S by hydrolysis. Figure 1 shows a time course of H 2 S emission from a Na 2 S solution. After resolving Na 2 S into a Hepes-KOH buffer at pH 7.0, an abrupt release of H 2 S followed by its decay was observed ( Figure 1). This rapid spontaneous reaction makes it difficult to obtain a stable H 2 S concentration for physiological purposes. To overcome this difficulty in application, the synthetic H 2 S donor GYY4137 has been used for stable delivery of H 2 S to cells in many studies. GYY4137 is a Lawesson's reagent derivative that releases H 2 S via hydrolysis both in vivo and in vitro [36]. In 2011, Xian and co-workers discovered a series of N-(benzoylthio) benzamide derivatives that can be activated by thiols to release H 2 S [37,38]. The H 2 S donors 5a, 8l, and 8o are such new tools for exploring biological function of H 2 S. Under the conditions we used, unlike with Na 2 S, we did not measure detectable H 2 S release (ppm) into the air from either GYY4137 or 5a solutions ( Figure 1). H2S gas was measured with a handheld O2/H2S monitor (XOS-326, New Cosmos Electric, Osaka, Japan). The H2S that was released into the headspace (10 mL) of a beaker was monitored for 30 s. The H2S releasing activity of each solution was expressed as ppm/min.

Effects of H2S Donors on the Root Abscission in Azolla
To investigate biological functions of H2S, the application of H2S donors would be the first choice for physiological experiments. Sodium hydrosulfide (NaHS) and sodium sulfide (Na2S) are inorganic compounds that release H2S by hydrolysis. Figure 1 shows a time course of H2S emission from a Na2S solution. After resolving Na2S into a Hepes-KOH buffer at pH 7.0, an abrupt release of H2S followed by its decay was observed ( Figure 1). This rapid spontaneous reaction makes it difficult to obtain a stable H2S concentration for physiological purposes. To overcome this difficulty in application, the synthetic H2S donor GYY4137 has been used for stable delivery of H2S to cells in many studies. GYY4137 is a Lawesson's reagent derivative that releases H2S via hydrolysis both in vivo and in vitro [36]. In 2011, Xian and co-workers discovered a series of N-(benzoylthio) benzamide derivatives that can be activated by thiols to release H2S [37,38]. The H2S donors 5a, 8l, and 8o are such new tools for exploring biological function of H2S. Under the conditions we used, unlike with Na2S, we did not measure detectable H2S release (ppm) into the air from either GYY4137 or 5a solutions ( Figure 1). The water fern Azolla pinnata shows a characteristic rapid root abscission phenomenon in response to certain environmental stimuli. In laboratory experiments, a range of chemicals have been reported to be effective in abscising the roots, such as the uncouplers CCCP and DNP, the hemebinding NaN3 and KCN [21], the polyamines spermine and spermidine [20], the NO precursor NaNO2, and the NO donor spermine NONOate (SNN). Also, H2O2 appears to be involved in the abscission mechanism [34]. Since the rapid root abscission phenomenon is responsive to both the RNS NO and the ROS H2O2, it was logical to speculate that H2S or RSS might similarly exert an effect. To test this hypothesis, we compared the effects of five H2S donors: Na2S, GYY4137, 5a, 8l, and 8o on root abscission. Among the compounds tested, the H2S donor 5a was found to be effective in detaching the roots ( Figure 2). Figure 2b demonstrates the root abscission induced by the H2S donor 5a at 200 µM. The detached end of a root abscised by the H2S donor 5a showed expanded cells within the abscission zone, which agrees with the morphological features of abscised roots as in previous reports [30,34,35]. The water fern Azolla pinnata shows a characteristic rapid root abscission phenomenon in response to certain environmental stimuli. In laboratory experiments, a range of chemicals have been reported to be effective in abscising the roots, such as the uncouplers CCCP and DNP, the heme-binding NaN 3 and KCN [21], the polyamines spermine and spermidine [20], the NO precursor NaNO 2 , and the NO donor spermine NONOate (SNN). Also, H 2 O 2 appears to be involved in the abscission mechanism [34]. Since the rapid root abscission phenomenon is responsive to both the RNS NO and the ROS H 2 O 2 , it was logical to speculate that H 2 S or RSS might similarly exert an effect. To test this hypothesis, we compared the effects of five H 2 S donors: Na 2 S, GYY4137, 5a, 8l, and 8o on root abscission. Among the compounds tested, the H 2 S donor 5a was found to be effective in detaching the roots ( Figure 2). Figure 2b demonstrates the root abscission induced by the H 2 S donor 5a at 200 µM. The detached end of a root abscised by the H 2 S donor 5a showed expanded cells within the abscission zone, which agrees with the morphological features of abscised roots as in previous reports [30,34,35].            (Figure 4). Up to 400 µM, the effect of 5a showed concentration dependency (Figure 4, inset). Figures 3 and 4 clearly demonstrate that the H2S donor 5a is a novel chemical compound that can induce the root abscission of Azolla.

Effects of Polysulfides on the Root Abscission
The H 2 S donors tested in this study released H 2 S by different mechanisms. Both Na 2 S and GYY4137 produce H 2 S by a spontaneous hydrolysis reaction. Although the final amount of H 2 S release may depend on pH, those types of H 2 S donors do not require specific conditions other than the presence of H 2 O. In contrast, the new types of H 2 S donors 5a, 8l, and 8o require the presence of thiols such as cysteine or reduced form of glutathione (GSH) [37]. In fact, no H 2 S emission from 5a in buffered solution was measured, as shown in Figure 1. The mechanisms for H 2 S release from 5a, 8l, and 8o are considered to be:

Effects of Polysulfides on the Root Abscission
The H2S donors tested in this study released H2S by different mechanisms. Both Na2S and GYY4137 produce H2S by a spontaneous hydrolysis reaction. Although the final amount of H2S release may depend on pH, those types of H2S donors do not require specific conditions other than the presence of H2O. In contrast, the new types of H2S donors 5a, 8l, and 8o require the presence of thiols such as cysteine or reduced form of glutathione (GSH) [37]. In fact, no H2S emission from 5a in buffered solution was measured, as shown in Figure 1. The mechanisms for H2S release from 5a, 8l, and 8o are considered to be: It is important to note that the H2S donor 5a (Equation (1)) as well as 8l and 8o (Equation (2)) decomposes with multiple steps. For each reaction, the presence of thiol (-SH) is required [37]. The difference in structure between 8l and 8o is the R moiety (8l: R = -CH3, 8o: R = -C(CH3)3). If the compound 5a, in fact, induced root abscission by its chemical reaction, the question arose as to whether the effect relied on H2S or on interactions with thiols within the cells. The minimal effect of the spontaneous H2S donor Na2S and GYY4137 on root abscission led us to check the actions of polysulfides. Figure 5 shows effects of sodium polysulfides (Na2Sn) on the root abscission. We compared sodium sulfide (Na2S), sodium disulfide (Na2S2), sodium trisulfide (Na2S3), and sodium tetrasulfide (Na2S4). All of these tested compounds showed some abscission-inducing effects, with the polysulfide Na2S4 being much more effective than Na2S in detaching the roots ( Figure 5).

Effects of Polysulfides on the Root Abscission
The H2S donors tested in this study released H2S by different mechanisms. Both Na2S and GYY4137 produce H2S by a spontaneous hydrolysis reaction. Although the final amount of H2S release may depend on pH, those types of H2S donors do not require specific conditions other than the presence of H2O. In contrast, the new types of H2S donors 5a, 8l, and 8o require the presence of thiols such as cysteine or reduced form of glutathione (GSH) [37]. In fact, no H2S emission from 5a in buffered solution was measured, as shown in Figure 1. The mechanisms for H2S release from 5a, 8l, and 8o are considered to be: It is important to note that the H2S donor 5a (Equation (1)) as well as 8l and 8o (Equation (2)) decomposes with multiple steps. For each reaction, the presence of thiol (-SH) is required [37]. The difference in structure between 8l and 8o is the R moiety (8l: R = -CH3, 8o: R = -C(CH3)3). If the compound 5a, in fact, induced root abscission by its chemical reaction, the question arose as to whether the effect relied on H2S or on interactions with thiols within the cells. The minimal effect of the spontaneous H2S donor Na2S and GYY4137 on root abscission led us to check the actions of polysulfides. Figure 5 shows effects of sodium polysulfides (Na2Sn) on the root abscission. We compared sodium sulfide (Na2S), sodium disulfide (Na2S2), sodium trisulfide (Na2S3), and sodium tetrasulfide (Na2S4). All of these tested compounds showed some abscission-inducing effects, with the polysulfide Na2S4 being much more effective than Na2S in detaching the roots ( Figure 5).  It is important to note that the H 2 S donor 5a (Equation (1)) as well as 8l and 8o (Equation (2)) decomposes with multiple steps. For each reaction, the presence of thiol (-SH) is required [37]. The difference in structure between 8l and 8o is the R moiety (8l: R = -CH 3, 8o: R = -C(CH 3 ) 3 ). If the compound 5a, in fact, induced root abscission by its chemical reaction, the question arose as to whether the effect relied on H 2 S or on interactions with thiols within the cells. The minimal effect of the spontaneous H 2 S donor Na 2 S and GYY4137 on root abscission led us to check the actions of polysulfides. Figure 5 shows effects of sodium polysulfides (Na 2 S n ) on the root abscission. We compared sodium sulfide (Na 2 S), sodium disulfide (Na 2 S 2 ), sodium trisulfide (Na 2 S 3 ), and sodium tetrasulfide (Na 2 S 4 ). All of these tested compounds showed some abscission-inducing effects, with the polysulfide Na 2 S 4 being much more effective than Na 2 S in detaching the roots ( Figure 5).

Effects of Polysulfides on the Root Abscission
The H2S donors tested in this study released H2S by different mechanisms. Both Na2S and GYY4137 produce H2S by a spontaneous hydrolysis reaction. Although the final amount of H2S release may depend on pH, those types of H2S donors do not require specific conditions other than the presence of H2O. In contrast, the new types of H2S donors 5a, 8l, and 8o require the presence of thiols such as cysteine or reduced form of glutathione (GSH) [37]. In fact, no H2S emission from 5a in buffered solution was measured, as shown in Figure 1. The mechanisms for H2S release from 5a, 8l, and 8o are considered to be: It is important to note that the H2S donor 5a (Equation (1)) as well as 8l and 8o (Equation (2)) decomposes with multiple steps. For each reaction, the presence of thiol (-SH) is required [37]. The difference in structure between 8l and 8o is the R moiety (8l: R = -CH3, 8o: R = -C(CH3)3). If the compound 5a, in fact, induced root abscission by its chemical reaction, the question arose as to whether the effect relied on H2S or on interactions with thiols within the cells. The minimal effect of the spontaneous H2S donor Na2S and GYY4137 on root abscission led us to check the actions of polysulfides. Figure 5 shows effects of sodium polysulfides (Na2Sn) on the root abscission. We compared sodium sulfide (Na2S), sodium disulfide (Na2S2), sodium trisulfide (Na2S3), and sodium tetrasulfide (Na2S4). All of these tested compounds showed some abscission-inducing effects, with the polysulfide Na2S4 being much more effective than Na2S in detaching the roots ( Figure 5).  Effects of polysulfides (Na 2 S n ) on root abscission. Root abscission is represented as % of the total number of the roots. Azolla was incubated for 24 h with a 10 mM Hepes-KOH (pH 7.0) buffer containing 500 µM polysulfide (Na 2 S n ). The values are means ± SE (n = 3). A significant difference is indicated as * (p < 0.05).

d-Cysteine-Induced Root Abscission
Although root abscission was induced by the H 2 S donor 5a (Figures 2 and 3) and polysulfides (Figure 4), the abscission proceeded over the course of hours, much slower than for other previously reported chemical inducers [20,30,34]. There are increasing numbers of reports suggesting that l/d-cysteine could be an endogenous substrate for H 2 S synthesis in redox signaling mechanisms [39]. Thus, we tested the effects of l-and d-cysteine on root abscission in Azolla. Interestingly, d-cysteine, but not l-cysteine, was found to be efficient in inducing rapid root abscission ( Figure 6). Unlike the case of the H 2 S donor 5a or polysulfides, the abscission finished less in than an hour, which is comparable to the effects of other chemical inducers. The effect of d-cysteine showed concentration dependence (Figure 7). The initial speed of the abscission increased with the d-cysteine concentration.

D-Cysteine-Induced Root Abscission
Although root abscission was induced by the H2S donor 5a (Figures 2 and 3) and polysulfides (Figure 4), the abscission proceeded over the course of hours, much slower than for other previously reported chemical inducers [20,30,34]. There are increasing numbers of reports suggesting that L/Dcysteine could be an endogenous substrate for H2S synthesis in redox signaling mechanisms [39]. Thus, we tested the effects of L-and D-cysteine on root abscission in Azolla. Interestingly, D-cysteine, but not L-cysteine, was found to be efficient in inducing rapid root abscission ( Figure 6). Unlike the case of the H2S donor 5a or polysulfides, the abscission finished less in than an hour, which is comparable to the effects of other chemical inducers. The effect of D-cysteine showed concentration dependence (Figure 7). The initial speed of the abscission increased with the D-cysteine concentration.

Application of Chemical H2S Donors for Inducing Root Abscission
To explore biological functions of H2S, physiological experiments with H2S chemical donors have been conducted both in plants [13] and mammals [40]. Since Na2S and NaHS are relatively cheap, these inorganic H2S donors can be applied for large-scale experiments using whole plants [18]. As shown in Figure 1, however, the H2S releasing activity of Na2S quickly decays such that it is virtually impossible to maintain a stable concentration within a physiological range. Moreover, the

D-Cysteine-Induced Root Abscission
Although root abscission was induced by the H2S donor 5a (Figures 2 and 3) and polysulfides (Figure 4), the abscission proceeded over the course of hours, much slower than for other previously reported chemical inducers [20,30,34]. There are increasing numbers of reports suggesting that L/Dcysteine could be an endogenous substrate for H2S synthesis in redox signaling mechanisms [39]. Thus, we tested the effects of L-and D-cysteine on root abscission in Azolla. Interestingly, D-cysteine, but not L-cysteine, was found to be efficient in inducing rapid root abscission ( Figure 6). Unlike the case of the H2S donor 5a or polysulfides, the abscission finished less in than an hour, which is comparable to the effects of other chemical inducers. The effect of D-cysteine showed concentration dependence (Figure 7). The initial speed of the abscission increased with the D-cysteine concentration.

Application of Chemical H2S Donors for Inducing Root Abscission
To explore biological functions of H2S, physiological experiments with H2S chemical donors have been conducted both in plants [13] and mammals [40]. Since Na2S and NaHS are relatively cheap, these inorganic H2S donors can be applied for large-scale experiments using whole plants [18]. As shown in Figure 1, however, the H2S releasing activity of Na2S quickly decays such that it is virtually impossible to maintain a stable concentration within a physiological range. Moreover, the

Application of Chemical H 2 S Donors for Inducing Root Abscission
To explore biological functions of H 2 S, physiological experiments with H 2 S chemical donors have been conducted both in plants [13] and mammals [40]. Since Na 2 S and NaHS are relatively cheap, these inorganic H 2 S donors can be applied for large-scale experiments using whole plants [18]. As shown in Figure 1, however, the H 2 S releasing activity of Na 2 S quickly decays such that it is virtually impossible to maintain a stable concentration within a physiological range. Moreover, the inorganic H 2 S donors are extremely moisture sensitive and they are easily oxidized in the presence of O 2 . To compensate for these technical difficulties, many synthetic new H 2 S donors have been developed. In plant science, GYY4137 has been employed to demonstrate physiological functions of H 2 S [13]. We expected that GYY4137 should induce root abscission, but found it does not have a significant abscission inducing activity even at mM concentrations (data not shown). The root abscission activity of Na 2 S was also weak. The novel H 2 S donors 5a, 8l, and 8o need cysteine or GSH for releasing H 2 S into an aqueous phase [38]. Because of this nature, we had not expected the abscission inducing activity of such thiol-activated type of H 2 S donors. However, the H 2 S donor 5a showed reasonably good abscission activity at sub mM concentrations (Figure 4).
Although recent studies have highlighted the "positive" regulatory functions of H 2 S, the molecule is yet cytotoxic and potentially inactivates metalloenzymes. The application of high concentrations of H 2 S from outside of the plants may disturb many enzymatic reactions or metabolisms that are required for the initiation of specific physiological events. We speculate that the structure of 5a bearing aromatic rings facilitates an efficient delivery of the compound to the target due to its lipophilic nature. The requirement of thiols (cysteine or GSH) could further localize the H 2 S production by the compound, thereby minimizing the negative impact of H 2 S. The difference in the effect between 8o and 8l could be also explained by difference in the polarity of the R moiety ( Figure 1).

Effects of Polysulfides (Na 2 S n ) on the Root Abscission
Recent progress in mammalian H 2 S studies has strongly suggested that potential chemical entities regulating biological functions of RSS are polysulfur species (H 2 S n ), rather than H 2 S itself [40]. A growing number of reports have supported the participation of polysulfidation of cysteine thiols [41][42][43]. To test the involvement of polysulfidation, in this study we applied sodium polysulfides (Na 2 S n ) for the Azolla root abscission experiments. As shown in Figure 4, all polysulfides induced root abscission. The results may imply that polysulfidation of thiol(s) could be involved in the root abscission mechanism. In a good agreement with this aspect, we observed that S-methyl methane thiosulfonate (MMTS), which covalently sulfenylates the thiol of cysteine residues, inhibited nitrite-induced root abscission [34]. Moreover, excessive GSH exogenously added was reported to be suppressive to the nitrite-induced root abscission in A. pinnata [34]. It should be noted that the polysulfide effects required relatively higher concentrations and longer incubation time compared with the H 2 S donor 5a (Figures 3 and 5). Since H 2 S emission (ppm) into the air was observed even in Na 2 S 2 , Na 2 S 3 , and Na 2 S 4 solutions (data not shown), we suspect that effects of H 2 S might overlay the actions of the sodium polysulfides. Obviously, further confirmation is necessary to conclude the effects of Na 2 S n have physiological relevance. Like with the H 2 S donor 5a, we may need to wait for new tools of synthetic polysulfides that can effectively mediate a local polysulfidation without spontaneous H 2 S production.

d-Cysteine is a Novel Inducer of Rapid Root Abscsision in Azolla
The present study has demonstrated that d-cysteine is a good inducer of rapid root abscission in A. pinnata (Figures 6 and 7), a novel finding that provides an important clue to reveal the root abscission mechanism. d-amino acids had long been thought as a laboratory artifact. Only recently, their biological functions have come to be recognized [39]. In mammals, d-cysteine has been found to be the substrate for H 2 S synthesis catalyzed by 3-mercaptopyruvate sulfurtransferase (3MST) along with d-amino acid oxidase (DAO), namely, the 3MST/DAO pathway [44]. In plants, d-cysteine desulfhydrase (d-CDes) has been suggested to produce H 2 S [45]. The recently sequenced Azolla filiculoides genome encodes for a single d-CDes homolog [46]. d-cysteine may induce root abscission via a localized production of H 2 S that may react with an oxidized protein cysteine to form a hydropersulfide [35], or potentially by serving as a substrate for direct enzymatic formation of a persulfide on a cysteine thiol [40]. Interestingly, l-cysteine showed no abscission inducing activity ( Figure 6). In Arabidopsis plants l-cysteine desulfhydrase activity has been shown to produce H 2 S in the plant cells [47]. It is of a great interest to speculate why Azolla does not respond to l-cysteine: is it either due to evolutional [48] or symbiotic [49] reasons?

Conclusions
Mammalian cells are surrounded by abundant hemeproteins such as blood hemoglobin and muscle myoglobin. Those proteins could maintain low local NO or H 2 S concentrations in the tissues, which is a necessary condition for NO or H 2 S acting as a signaling molecule [1]. In contrast, O 2 -evolving photosynthetic organisms, such as plants, produce H 2 O 2 , NO and H 2 S as the byproducts of the assimilation metabolisms under the light, particularly in stress conditions [39,50]. The lower hemeprotein content in plant tissue enables NO and H 2 S diffuse both in and out of the tissue, thereby permitting sensing of environmental conditions directly through those gases. Presumably, the ROS, RNS and RSS signals would be integrated at the functional thiol groups that are potentially modulated by those reactive molecular species in different forms [40]. In fact, the interplay among ROS, RNS, and RSS produce a variety of reactive products derived from thiols, such as nitrosothiol (SNO), thionitrous acid (HSNO), nitroxyl (HNO), nitropersulfide (SSNO − ), and polysulfides H 2 S n [40]. This study has, for the first time, demonstrated that d-cysteine is a strong inducer of rapid root abscission in A. pinnata. To reveal the molecular mechanism for the d-cysteine-induced root abscission, controlling the chemical redox reactions while monitoring those key molecules will be essential. As we experienced in NO studies [51], physiological experiments with gaseous H 2 S are difficult to handle, and it is sometimes hard to obtain good reproducibility. We suggest that Azolla is a good model system to explore RSS-mediated signaling mechanisms in plants because of its tiny size (advantageous for culture [29]), water floating nature (advantageous for pharmacological experiments [20,21]), and rapid response comparable to stomatal responses (advantageous for analysis [35]).