Melatonin has long been recognized as a positive signaling molecule and potent antioxidant in plants, which alleviates damage caused by adverse conditions such as salt, cold, and heat stress. In this study, we found a paradoxical role for melatonin in abiotic stress responses. Suppression of the serotonin N-acetyltransferase 2 (snat2) gene encoding the penultimate enzyme in melatonin biosynthesis led to simultaneous decreases in both melatonin and brassinosteroid (BR) levels, causing a semi-dwarf with erect leaf phenotype, typical of BR deficiency. Here, we further characterized snat2 rice in terms of grain morphology and abiotic stress tolerance, to determine whether snat2 rice exhibited characteristics similar to those of BR-deficient rice. As expected, the snat2 rice exhibited tolerance to multiple stress conditions including cadmium, salt, cold, and heat, as evidenced by decreased malondialdehyde (MDA) levels and increased chlorophyll levels, in contrast with SNAT2 overexpression lines, which were less tolerant to stress than wild type plants. In addition, the length and width of grain from snat2 plants were reduced relative to the wild type, which is reminiscent of BR deficiency in rice. Other melatonin-deficient mutant rice lines with suppressed BR synthesis (i.e., comt and t5h) also showed tolerance to salt and heat stress, whereas melatonin-deficient rice seedlings without decreased BR levels (i.e., tdc) failed to exhibit increased stress tolerance, suggesting that stress tolerance was increased not by melatonin deficiency alone, but by a melatonin deficiency-mediated decrease in BR.
Serotonin N-acetyltransferase (SNAT) is the penultimate enzyme in the melatonin biosynthetic pathway, in which serotonin is converted into N-acetylserotonin (NAS) in plants. To date, two SNAT isogenes with low amino acid sequence homologies have been identified. Their single suppression in rice has been reported, but their double suppression in rice has not yet been attempted. Here, we generated double-suppression transgenic rice (snat1+2) using the RNA interference technique. The snat1+2 exhibited retarded seedling growths in conjunction with severe decreases in melatonin compared to wild-types and single-suppression rice plants (snat1 or snat2). The laminar angle was decreased in the snat1+2 rice compared to that of the wild-types and snat1, but was comparable to that of snat2. The reduced germination speed in the snat1+2 was comparable to that of snat2. Seed-aging testing revealed that snat1 was the most severely deteriorated, followed by snat1+2 and snat2, suggesting that melatonin is positively involved in seed longevity.
Light is an important factor influencing melatonin synthesis in response to cadmium treatment in rice. However, the effects of light quality on, and the involvement of phytochrome light receptors in, melatonin production have not been explored. In this study, we used light-emitting diodes (LEDs) to investigate the effect of light wavelength on melatonin synthesis, and the role of phytochromes in light-dependent melatonin induction in rice. Upon cadmium treatment, peak melatonin production was observed under combined red and blue (R + B) light, followed by red (R) and blue light (B). However, both far-red (FR) LED light and dark treatment (D) failed to induce melatonin production. Similarly, rice seedlings grown under the R + B treatment showed the highest melatonin synthesis, followed by those grown under B and R. These findings were consistent with the results of our cadmium treatment experiment. To further confirm the effects of light quality on melatonin synthesis, we employed rice photoreceptor mutants lacking functional phytochrome genes. Melatonin induction was most inhibited in the phytochrome A mutant (phyA) followed by the phyB mutant under R + B treatment, whereas phyB produced the least amount of melatonin under R treatment. These results indicate that PhyB is an R light receptor. Expression analyses of genes involved in melatonin biosynthesis clearly demonstrated that tryptophan decarboxylase (TDC) played a key role in phytochrome-mediated melatonin induction when rice seedlings were challenged with cadmium.
We investigated the relationship between the blue-light photoreceptor cryptochrome (CRY) and melatonin biosynthesis by generating RNA interference (RNAi) transgenic rice plants that suppress the cryptochrome 1b gene (CRY1b). The resulting CRY1b RNAi rice lines expressed less CRY1b mRNA, but not CRY1a or CRY2 mRNA, suggesting that the suppression is specific to CRY1b. The growth of CRY1b RNAi rice seedlings was enhanced under blue light compared to wild-type growth, providing phenotypic evidence for impaired CRY function. When these CRY1b RNAi rice plants were challenged with cadmium to induce melatonin, wild-type plants produced 100 ng/g fresh weight (FW) melatonin, whereas CRY1b RNAi lines produced 60 ng/g FW melatonin on average, indicating that melatonin biosynthesis requires the CRY photoreceptor. Due to possible feedback regulation, the expression of melatonin biosynthesis genes such as T5H, SNAT1, SNAT2, and COMT was elevated in the CRY1b RNAi lines compared to the wild-type plants. In addition, laminar angles decreased in the CRY1b RNAi lines via the suppression of brassinosteroid (BR) biosynthesis genes such as DWARF. The main cause of the BR decrease in the CRY1b RNAi lines seems to be the suppression of CRY rather than decreased melatonin because the melatonin decrease suppressed DWARF4 rather than DWARF.
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.
Arylalkylamine N-acetyltransferase (AANAT) is a pivotal enzyme in melatonin biosynthesis that catalyzes the conversion of serotonin to N-acetylserotonin. Homologs of animal AANAT genes are present in animals, but not in plants. An AANAT homolog was found in Chlamydomonas reinhardtii, but not other green algae. The characteristics of C. reinhardtiiAANAT (CrAANAT) are unclear. Here, full-length CrAANAT was chemically synthesized and expressed in Escherichia coli. Recombinant CrAANAT exhibited AANAT activity with a Km of 247 μM and Vmax of 325 pmol/min/mg protein with serotonin as the substrate. CrAANAT was localized to the cytoplasm in tobacco leaf cells. Transgenic rice plants overexpressing CrAANAT (CrAANAT-OE) exhibited increased melatonin production. CrAANAT-OE plants showed a longer seed length and larger second leaf angle than wild-type plants, indicative of the involvement of brassinosteroids (BRs). As expected, BR biosynthesis- and signaling-related genes such as D2, DWARF4, DWARF11, and BZR1 were upregulated in CrAANAT-OE plants. Therefore, an increased endogenous melatonin level by ectopic overexpression of CrAANAT seems to be closely associated with BR biosynthesis, thereby influencing seed size.