Molecular Regulation of Antioxidant Melatonin Biosynthesis by Brassinosteroid Acting as an Endogenous Elicitor of Melatonin Induction in Rice Seedlings

Gibberellic acid (GA) was recently shown to induce melatonin synthesis in rice. Here, we examined whether brassinosteroids (BRs) also induce melatonin synthesis because BRs and GA show redundancy in many functions. Among several plant hormones, exogenous BR treatment induced melatonin synthesis by twofold compared to control treatment, whereas ethylene, 6-benzylaminopurine (BA), and indole-3-acetic acid (IAA) showed negligible effects on melatonin synthesis. Correspondingly, BR treatment also induced a number of melatonin biosynthetic genes in conjunction with the suppression of melatonin catabolic gene expression. Several transgenic rice plants with downregulated BR biosynthesis-related genes, such as DWARF4, DWARF11, and RAV-Like1 (RAVL1), were generated and exhibited decreased melatonin synthesis, indicating that BRs act as endogenous elicitors of melatonin synthesis. Notably, treatment with either GA or BR fully restored melatonin synthesis in the presence of paclobutrazol, a GA biosynthesis inhibitor. Moreover, exogenous BR treatment partially restored melatonin synthesis in both RAVL1 and Gα RNAi transgenic rice plants, whereas GA treatment fully restored melatonin synthesis comparable to wild type in RAVL1 RNAi plants. Taken together, our results highlight a role of BR as an endogenous elicitor of melatonin synthesis in a GA-independent manner in rice plants.


Introduction
Melatonin is a multifunctional molecule commonly found in bacteria, archaea, plants, and animals [1,2]. In plants, melatonin is thought to be a master regulator of a diverse array of physiological functions, with roles not only in growth and development but also in defense responses against many biotic and abiotic stressors [3,4]. Many of these functions of melatonin, particularly stress tolerance responses, are derived from its intrinsic activity as a potent antioxidant that regulates redox balance in plants [1]. For instance, it is known that one molecule of melatonin scavenges up to 10 molecules of reactive oxygen species or reactive nitrogen species [5], as well as the induction of many antioxidant enzymes such as catalase and superoxide dismutase [1,3]. In addition, melatonin acts as a signaling molecule in plants to activate a mitogen-activated protein kinase (MAPK) pathway followed by the induction of many genes involved in defense against pathogens [6,7], protein quality control [8,9], stomatal closure [10], and yield increase [11,12]. Consistent with its pleiotropic roles, the mode of action of melatonin in plants is closely associated with many plant hormones by either upregulating or downregulating corresponding genes [13]. For example, the suppression of melatonin synthesis leads to decreases in levels of brassinosteroids (BR) in several RNAi transgenic rice plants with BR-deficient phenotypes, such as semi-dwarf with erect leaves [14]. Notably, melatonin-rich transgenic rice plants produce larger seeds associated with increased cytokinin synthesis [12]. In addition, exogenous melatonin treatment also promotes primary root growth in Arabidopsis thaliana in an indole-3-acetic

Quantification of Melatonin
Frozen samples (0.1 g) were pulverized to a powder in liquid nitrogen using the TissueLyser II (Qiagen, Tokyo, Japan) and then extracted with 1 mL chloroform. The chloroform extracts were centrifuged for 10 min at 12,000× g, and then the supernatants (200 µL) were completely evaporated and dissolved in 0.1 mL of 40% methanol, and 10-µL aliquots were subjected to HPLC using a fluorescence detector system (Waters, Milford, MA, USA) as described previously [30]. Melatonin was eluted at about 31 min under this HPLC condition. All measurements were performed in triplicate.
Similarly, the N-terminal 300 bp length of RAVL1 was amplified with the following primer set: RAVL1-F 5 -ACT AGT ATG GAG CAG GAG CAG GAT-3 (SpeI site underlined) and RAVL1-R 5 -GAG CTC GTT CAG CTT CCC CAC GTC-3 (SacI site underlined). These PCR products were first subcloned into the T&A cloning vector (T&A:DWARF4, T&A:DWARF11, and T&A:RAVL1; RBC Bioscience, New Taipei City, Taiwan) for further cloning experiments. From these plasmids, the antisense DWARF4, DWARF11, and RAVL1 inserts were acquired by SacI and SpeI double digestion, whereas the sense DWARF4, DWARF11, and RAVL1 inserts were prepared by KpnI and BamHI double digestion. The antisense fragments were first ligated into the pTCK303 vector followed by the sense fragments of three gene inserts. The resulting pTCK303:DWARF4, pTCK303:DWARF11, and pTCK303:RAVL1 RNAi binary vectors were independently transformed into Agrobacterium tumefaciens LBA4404 followed by Agrobacterium-mediated rice transformation with the embryogenic calli derived from rice seed (Oryza sativa cv. Dongjin) as previously described [11,32]. Transgenic rice plants suppressing Gα was generated previously and described in Hwang and Back [20].

Statistical Analysis
The data were analyzed using analysis of variance using IBM SPSS Statistics 23 software (IBM Corp. Armonk, NY, USA). Means with asterisks and different letters indicate significantly different values at p < 0.05, according to least significance difference (LSD) test. All data are presented as mean ± standard deviations.

Effects of Various Hormone Treatments on Melatonin Synthesis in Rice Seedlings
GA acts as an endogenous plant hormone to induce melatonin synthesis in rice [20]. Here, to determine whether other plant hormones can also induce melatonin synthesis, the roots of 7-day-old rice seedlings were independently treated with varying levels of four different hormones for 24 h, followed by cadmium treatment for induction of melatonin to facilitate the detection of melatonin by HPLC. As shown in Figure 1, ethylene was unable to induce melatonin synthesis, whereas BR showed a strong capacity to induce melatonin synthesis with peak melatonin accumulation at a concentration of 1 µM. The effect of BR was further confirmed by the application of brassinazole (BZ), a BR biosynthesis inhibitor. BZ treatment significantly inhibited melatonin synthesis, indicating that BR elicits melatonin synthesis, similar to GA [20]. In addition, cytokinin (BA) and IAA treatments slightly increased melatonin synthesis, but its levels were far lower than those of BR. It is puzzling that 10 µM ethylene treatment decreased melatonin synthesis that was not observed in either 1 µM or 100 µM treatment. The reason for this is unclear, but similar results were also observed in cytokinin (0.01 µM) and IAA (0.01 µM) treatments. In summary, these results clearly show that BR acts as a potent inducer of melatonin synthesis in rice seedlings.

Characterization of Genes Involved in Melatonin Biosynthesis and Catabolism in Response to BR Treatment
To examine whether melatonin synthesis is induced by BR treatment in the absence of cadmium treatment, 7-day-old rice seedlings were challenged rhizosperically with varying concentrations of BR for 24 h, followed by quantification of melatonin using the upper parts of leaves and stems. As shown in Figure 2, melatonin synthesis increased with BR treatment at a concentration of 0.1 µM and reached a peak at 1 µM with a 40% increase compared to controls. Treatment with 10 µM BR failed to induce melatonin production, in contrast to cadmium treatment (Figures 1 and 2). To determine whether the increase in melatonin synthesis by BR treatment is associated with genes involved in melatonin synthesis and catabolism, the shoot and stem tissues of 7-day-old rice seedlings challenged with 1 µM BR for 24 h were separated and harvested for total RNA extraction followed by quantitative real-time polymerase chain reaction (qRT-PCR) analysis using UBQ5 as a reference gene. Transcript levels of melatonin biosynthetic genes, such as tryptophan decarboxylase 1 (TDC1), TDC3, tryptamine 5-hydroxylase (T5H), and N-acetylserotonin O-methyltransferase 1 (ASMT1), were significantly elevated by BR treatment, while the expression levels of the biosynthetic genes TDC2 and caffeic acid O-methyltransferase (COMT) were downregulated ( Figure 2C). Among catabolic genes, BR treatment inhibited the expression of N-acetylserotonin deacetylase (ASDAC), whereas the expression of melatonin 3-hydroxylase (M3H) was upregulated. The main difference in the effect on the increase in melatonin between GA and BR treatment was seen in TDC1 expression, which was downregulated by GA treatment but upregulated by BR. In summary, the BR-mediated increase in melatonin is primarily related to the induction of melatonin biosynthetic gene transcription rather than the suppression of catabolic genes.

Characterization of Genes Involved in Melatonin Biosynthesis and Catabolism in Response to BR Treatment
To examine whether melatonin synthesis is induced by BR treatment in the absence of cadmium treatment, 7-day-old rice seedlings were challenged rhizosperically with varying concentrations of BR for 24 h, followed by quantification of melatonin using the upper parts of leaves and stems. As shown in Figure 2, melatonin synthesis increased with BR treatment at a concentration of 0.1 μM and reached a peak at 1 μM with a 40% increase compared to controls. Treatment with 10 μM BR failed to induce melatonin production, in contrast to cadmium treatment (Figures 1 and 2). To determine whether the increase in melatonin synthesis by BR treatment is associated with genes involved in melatonin synthesis and catabolism, the shoot and stem tissues of 7-day-old rice seedlings challenged with 1 μM BR for 24 h were separated and harvested for total RNA extraction followed by quantitative real-time polymerase chain reaction (qRT-PCR) analysis using UBQ5 as a (D) Effects of indole-3-acetic acid (IAA) on melatonin content. Seven-day-old rice seedlings were treated rhizosperically with varying levels of hormones independently for 24 h and transferred to new 50 mL conical tubes containing 0.5 mM cadmium for 3 days to induce melatonin synthesis. Different letters denote significant differences (p < 0.05) as determined by Tukey's HSD post hoc test. C, water containing 0.1% ethanol. Ethephon that releases ethylene was used for ethylene treatment.
Antioxidants 2022, 11, 918 6 of 14 downregulated ( Figure 2C). Among catabolic genes, BR treatment inhibited the expression of N-acetylserotonin deacetylase (ASDAC), whereas the expression of melatonin 3-hydroxylase (M3H) was upregulated. The main difference in the effect on the increase in melatonin between GA and BR treatment was seen in TDC1 expression, which was downregulated by GA treatment but upregulated by BR. In summary, the BR-mediated increase in melatonin is primarily related to the induction of melatonin biosynthetic gene transcription rather than the suppression of catabolic genes.

Melatonin Biosynthesis Was Severely Compromised in Transgenic Rice Plants with Various RNAis Downregulating the BR Biosynthesis-Related Genes DWARF4, DWARF11, and RAVL1
To investigate whether a decrease in BR leads to a reduction of melatonin biosynthesis, three BR biosynthesis-related genes were downregulated using an RNAi strategy (Figure 3). The two BR biosynthetic genes, DWARF4 and DWARF11, which catalyze C-22 hydroxylation, the rate-limiting step of BR biosynthesis, have redundant functions, but show a 37% amino-acid identity [25]. The third gene, RAV-like1 (RAVL1), is a transcription factor that binds BR biosynthetic genes, such as DWARF11 and ebisu dwarf (D2), as well as the BR receptor BRI1, leading to increased BR synthesis [26,27]. All three RNAi transgenic rice Seven-day-old rice seedlings were treated rhizosperically with various concentrations of epibrassinolide (BR) for quantification of melatonin or with 1 µM epibrassinolide (BR) for qRT-PCR for 24 h. The upper parts of leaves and stems were collected for HPLC quantification of melatonin and qRT-PCR analysis of various genes involved in melatonin synthesis and catabolism. The primer sequences were previously described [20]. C, water containing 0.1% ethanol. Asterisks (*) denote significant differences (p < 0.05; ANOVA, followed by Tukey's HSD post hoc tests). C, water containing 0.1% ethanol; BR, epibrassinolide; TDC1, tryptophan decarboxylase 1 (AK069031); TDC2 (AK103253); TDC3 (Os08g0140500); T5H,

Melatonin Biosynthesis Was Severely Compromised in Transgenic Rice Plants with Various
RNAis Downregulating the BR Biosynthesis-Related Genes DWARF4, DWARF11, and RAVL1 To investigate whether a decrease in BR leads to a reduction of melatonin biosynthesis, three BR biosynthesis-related genes were downregulated using an RNAi strategy ( Figure 3). The two BR biosynthetic genes, DWARF4 and DWARF11, which catalyze C-22 hydroxylation, the rate-limiting step of BR biosynthesis, have redundant functions, but show a 37% amino-acid identity [25]. The third gene, RAV-like1 (RAVL1), is a transcription factor that binds BR biosynthetic genes, such as DWARF11 and ebisu dwarf (D2), as well as the BR receptor BRI1, leading to increased BR synthesis [26,27]. All three RNAi transgenic rice seedlings exhibited suppression of transcripts in conjunction with a common dwarf phenotype (Figures 4-6), a characteristic feature of BR deficiency. Melatonin levels were 25 ng/g fresh weight (FW) on average in both DWARF4 and DWARF11 RNAi rice seedlings, while the wild type produced melatonin at a level of 85 ng/g FW. RAVL1 RNAi rice seedlings produced melatonin at a level of 38 ng/g FW on average, which was 2.2-fold lower than the wild type. DWARF11, D2, and BRI1 transcripts were clearly suppressed in RAVL1 RNAi rice seedlings, whereas DWARF4 expression was not altered (Figure 6B), corroborating the involvement of RAVL1 in both BR biosynthesis and signaling. Taken together, these observations are consistent with the results of exogenous BR treatments, highlighting the role of BR as an elicitor of melatonin biosynthesis in vitro and in vivo, similar to GA [20].
while the wild type produced melatonin at a level of 85 ng/g FW. RAVL1 RNAi rice seedlings produced melatonin at a level of 38 ng/g FW on average, which was 2.2-fold lower than the wild type. DWARF11, D2, and BRI1 transcripts were clearly suppressed in RAVL1 RNAi rice seedlings, whereas DWARF4 expression was not altered (Figure 6B), corroborating the involvement of RAVL1 in both BR biosynthesis and signaling. Taken together, these observations are consistent with the results of exogenous BR treatments, highlighting the role of BR as an elicitor of melatonin biosynthesis in vitro and in vivo, similar to GA [20].     Seven-day-old rice seedlings were challenged with 0.5 mM cadmium for 3 days for quantification of melatonin. Asterisks (*) indicate significant differences from the wild type (p < 0.05), as determined by Tukey's HSD posthoc tests.

BR Induced Melatonin Synthesis in a GA-Independent Manner
The 7-day-old rice seedlings were challenged with 10 μM paclobutrazol (PB), a GA biosynthesis inhibitor, together with either BR or GA alone or BR + GA to determine whether PB-mediated melatonin decrease could be reversed. As shown in Figure 7, the PB-mediated reduction of melatonin was fully reversed by either BR or GA alone and in combination. These data suggest that BR-induced melatonin induction is not dependent on GA-induced melatonin synthesis. The synergistic increase in melatonin level by combined treatment with BR and GA compared to either individual treatment alone has not been observed. Unlike cell elongation in which BR acts as an upstream regulator of GA

BR Induced Melatonin Synthesis in a GA-Independent Manner
The 7-day-old rice seedlings were challenged with 10 µM paclobutrazol (PB), a GA biosynthesis inhibitor, together with either BR or GA alone or BR + GA to determine whether PB-mediated melatonin decrease could be reversed. As shown in Figure 7, the PB-mediated reduction of melatonin was fully reversed by either BR or GA alone and in combination. These data suggest that BR-induced melatonin induction is not dependent on GA-induced melatonin synthesis. The synergistic increase in melatonin level by combined treatment with BR and GA compared to either individual treatment alone has not been observed. Unlike cell elongation in which BR acts as an upstream regulator of GA biosynthesis [21], the increase in melatonin level associated with BR treatment did not require the downstream GA regulator. These pharmacological complementation experiments were further verified in RAVL1 and Gα RNAi transgenic rice plants. Melatonin decreases in these transgenic plants were partially ameliorated by exogenous BR treatment because these genes are associated with the BR signaling pathway (Figure 8). However, the decrease in melatonin level in the RAVL1 RNAi transgenic rice was fully reversed by exogenous GA treatment, but only partially reversed in the Gα RNAi transgenic rice plants. These observations clearly indicate that BR and GA act independently to induce melatonin synthesis.

Discussion
The first report of crosstalk between BR and melatonin came from transgenic rice plants with downregulation of serotonin N-acetyltransferase 2 (SNAT2), the penultimate gene in the melatonin biosynthesis pathway [33]. The SNAT2-suppressed RNAi rice plants showed a BR-deficient phenotype exhibiting semi-dwarf with erect leaves in conjunction with BR decrease. This indicated that melatonin decrease is related to BR decrease, while SNAT2 overexpression did not result in an increase in BR, suggesting an indirect effect of melatonin on BR synthesis, similar to the GA decrease in SNAT1 knockout A. thaliana mutant [8,34]. In addition to the effect of melatonin on the levels of various plant hormones [13], we recently reported that GA acts as an endogenous elicitor to induce melatonin synthesis [20]. One possible explanation for GA-induced melatonin synthesis is that GAmediated growth promotion is susceptible to many stresses, including H 2 O 2 [35], which may be counteracted by an increase in melatonin, as melatonin is a potent antioxidant as well as a signaling molecule with roles in protecting plants from a diverse array of abiotic and biotic stresses by way of the induction of many antioxidant enzymes and their defense-related genes [1,3,[36][37][38][39].
Interestingly, it was reported that BR regulates cell elongation by modulating GA biosynthesis in rice [21] and A. thaliana [22], suggesting that BR treatment may also induce melatonin synthesis indirectly via an increase in GA. In this study, BR treatment induced melatonin synthesis in a dose-dependent manner, similar to GA, while the other hormones, ethylene, BA, and IAA, failed to induce melatonin synthesis (Figure 1). To gain direct genetic evidence, we used RNAi for three BR synthesis-and signaling-related genes, i.e., DWARF4, DWARF11, and RAVL1. Transgenic rice plants with RNAi-mediated downregulation of these genes showed decreased synthesis of melatonin, indicating that BR acts as an endogenous elicitor in plants. The greatest decline in melatonin synthesis was observed in the DWARF4 RNAi rice followed by DWARF11 RNAi and RAVL1 RNAi lines. Melatonin synthesis was markedly decreased in the presence of paclobutrazol, a GA biosynthesis inhibitor, while either exogenous BR or GA treatment fully recovered melatonin synthesis. These observations suggested that BR-mediated melatonin induction is independent of GA, with both acting as an independent elicitor of melatonin synthesis (Figure 9). These data were further confirmed by the observation that GA treatment of the RAVL1 RNAi rice seedlings fully recovered melatonin synthesis comparable to wild type, but only partially recovered melatonin synthesis in Gα RNAi rice seedlings. The partial recovery of melatonin synthesis in the Gα RNAi line by either GA or BR indicated that Gα is involved in the signaling processes of both GA and BR. GA and BR induce melatonin production independently, whereas abscisic acid inhibits melatonin production. The induction patterns of genes responsible for melatonin biosynthesis and metabolism were slightly different between hormone treatments. BR treatment induced several genes involved in melatonin biosynthesis, including TDC1, TDC3, T5H, and ASMT1, whereas GA induced TDC3, T5H, and ASMT1. ASDAC and M2H were downregulated by GA treatment, while only ASDAC was downregulated by BR treatment. Unlike BR and GA, abscisic acid inhibits melatonin synthesis [20].

Conclusions
Melatonin is a pleiotropic molecule with various functions in plant growth and development as well as in defense responses to abiotic and biotic stresses. Although there have been many reports of changes in plant hormones by melatonin treatment, there have been few studies of the effects of plant hormones on melatonin biosynthesis. Here, we demonstrated that BRs play essential roles in the induction of melatonin synthesis, similar to GA. Our findings indicate that melatonin seems to function in concert with BR and GA during plant growth and development. As both BR and GA are commonly involved in cell elongation and growth, which are accompanied by increased susceptibility to various abiotic stresses [35,40,41], the concomitant increases in melatonin level by BR and GA may counteract these adverse effects due to increases in either GA [40] or BR [41] by increasing protein quality control and antioxidant activity exerted by melatonin [1,3,5,9]. Further indepth studies of the precise underlying mechanisms and interplay between plant hormones and melatonin will facilitate the discovery of new functions of melatonin in plant growth and development as well as in defense responses to stressors. GA and BR induce melatonin production independently, whereas abscisic acid inhibits melatonin production. The induction patterns of genes responsible for melatonin biosynthesis and metabolism were slightly different between hormone treatments. BR treatment induced several genes involved in melatonin biosynthesis, including TDC1, TDC3, T5H, and ASMT1, whereas GA induced TDC3, T5H, and ASMT1. ASDAC and M2H were downregulated by GA treatment, while only ASDAC was downregulated by BR treatment. Unlike BR and GA, abscisic acid inhibits melatonin synthesis [20].
Commensurate with the induction of melatonin biosynthetic genes in response to GA, BR treatment induced many genes responsible for melatonin biosynthesis, including TDC3 and T5H. In addition, M2H was suppressed in response to BR, similar to GA. However, differential gene expression between BR and GA treatments was also observed in TDC1, TDC2, SNAT1, SNAT2, and M3H. Taken together, these results indicate that BR acts as an elicitor for induction of melatonin synthesis, similar to GA [23]. Both BR and GA seem to play independently roles in melatonin synthesis ( Figure 9A) unlike the BR-mediated GA increase pathway in growth and leaf angle ( Figure 9B). The results presented here suggest that melatonin synthesis is positively correlated with endogenous levels of either GA or BR in plant cells.

Conclusions
Melatonin is a pleiotropic molecule with various functions in plant growth and development as well as in defense responses to abiotic and biotic stresses. Although there have been many reports of changes in plant hormones by melatonin treatment, there have been few studies of the effects of plant hormones on melatonin biosynthesis. Here, we demonstrated that BRs play essential roles in the induction of melatonin synthesis, similar to GA. Our findings indicate that melatonin seems to function in concert with BR and GA during plant growth and development. As both BR and GA are commonly involved in cell elongation and growth, which are accompanied by increased susceptibility to various abiotic stresses [35,40,41], the concomitant increases in melatonin level by BR and GA may counteract these adverse effects due to increases in either GA [40] or BR [41] by increasing protein quality control and antioxidant activity exerted by melatonin [1,3,5,9]. Further indepth studies of the precise underlying mechanisms and interplay between plant hormones and melatonin will facilitate the discovery of new functions of melatonin in plant growth and development as well as in defense responses to stressors.