Using Exogenous Melatonin, Glutathione, Proline, and Glycine Betaine Treatments to Combat Abiotic Stresses in Crops

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

Abiotic stresses, such as drought, salinity, heat, cold, and heavy metals, are associated with global climate change and hamper plant growth and development, affecting crop yields and quality. However, the negative effects of abiotic stresses can be mitigated through exogenous treatments using small biomolecules. For example, the foliar application of melatonin provides the following: it protects the photosynthetic apparatus; it increases the antioxidant defenses, osmoprotectant, and soluble sugar levels; it prevents tissue damage and reduces electrolyte leakage; it improves reactive oxygen species (ROS) scavenging; and it increases biomass, maintains the redox and ion homeostasis, and improves gaseous exchange. Glutathione spray upregulates the glyoxalase system, reduces methylglyoxal (MG) toxicity and oxidative stress, decreases hydrogen peroxide and malondialdehyde accumulation, improves the defense mechanisms, tissue repairs, and nitrogen fixation, and upregulates the phytochelatins. The exogenous application of proline enhances growth and other physiological characteristics, upregulates osmoprotection, protects the integrity of the plasma lemma, reduces lipid peroxidation, increases photosynthetic pigments, phenolic acids, flavonoids, and amino acids, and enhances stress tolerance, carbon fixation, and leaf nitrogen content. The foliar application of glycine betaine improves growth, upregulates osmoprotection and osmoregulation, increases relative water content, net photosynthetic rate, and catalase activity, decreases photorespiration, ion leakage, and lipid peroxidation, protects the oxygen-evolving complex, and prevents chlorosis. Chemical priming has various important advantages over transgenic technology as it is typically more affordable for farmers and safe for plants, people, and animals, while being considered environmentally acceptable. Chemical priming helps to improve the quality and quantity of the yield. This review summarizes and discusses how exogenous melatonin, glutathione, proline, and glycine betaine can help crops combat abiotic stresses.

Related collections

Most cited references 158

Record: found
Abstract: found
Article: found

Is Open Access

Melatonin enhances plant growth and abiotic stress tolerance in soybean plants

Wei Wei, Qing-Tian Li, Ya-Nan Chu … (2014)

Introduction Extracts of the pineal gland were shown to lighten the skin colour of tadpoles, frogs and fish. In 1958, the active molecule, isolated from bovine pineal glands, was identified as N-acetyl-5-methoxy-tryptamine, also known as melatonin (Lerner et al., 1958; Lerner et al., 1960). Melatonin is now a well-known animal hormone that has several important biological functions, including influencing circadian rhythms (Hardeland et al., 2012), mediating changes in seasonal reproduction (Barrett and Bolborea, 2012), immuno-enhancement (Calvo et al., 2013), tumour inhibition (Blask et al., 2005; Bizzarri et al., 2013), and reducing oxidative stress (Hardeland et al., 1993; Reiter et al., 2000; Gitto et al., 2001; Silva et al., 2004; Galano et al., 2011, 2013). In 1995, using HPLC (high performance liquid chromatography) and radioimmunoassay, researchers identified melatonin in plants (Dubbels et al., 1995; Hattori et al., 1995; Van Tassel et al., 1995). Later research revealed that melatonin is also present in unicellular organisms (Hardeland and Poeggeler, 2003). The biosynthesis of melatonin begins with tryptophan (Reiter, 1991). Vascular plants have similar biosynthetic pathways as that in animals (Arnao and Hernandez-Ruiz, 2006) and homologous enzymes in plants have been identified (Fujiwara et al., 2010). In 2011, the final enzyme in the melatonin biosynthesis pathway was identified in rice as N-acetylserotonin methyltransferase (ASMT; Kang et al., 2011), which has a rate-limiting role. Research in rice has also revealed some differences in melatonin synthesis from other organisms; for example, the first metabolite in rice is tryptamine, but not 5-OH Trp (Kang et al., 2007; Park et al., 2012). Melatonin may possess a variety of functions in vascular plants (Kolar and Machackova, 2005; Uchendu et al., 2013). One of the important roles of melatonin is to act as an antioxidant and protect plants against biotic/abiotic stress (Tan et al., 2012). This antioxidative effect of melatonin has been reported in several plant species (apple, rice, and grape) (Wang et al., 2012; Park et al., 2013; Vitalini et al., 2013; Yin et al., 2013). Using high-throughput sequencing technology, the important roles of melatonin in plant defence have also been revealed. Melatonin up-regulates transcript levels of many defence-related factors, including stress receptors, kinases, and transcription factors (Weeda et al., 2014). Additionally, melatonin may have the ability to regulate plant growth and to enhance crop production. For example, melatonin was reported to promote coleoptile growth in four monocot species including canary grass, wheat, barley, and oat (Hernandez-Ruiz et al., 2005). Melatonin also promotes root growth in Brassica juncea (Chen et al., 2009) and adventitious root regeneration in shoot tip explants of sweet cherry (Sarropoulou et al., 2012). Additionally, melatonin-treated corn plants had greater production than non-treated plants (Tan et al., 2012). However, melatonin’s broad functions and its molecular mechanisms in important crops remain unclear. Soybean is an important crop for oil and as a protein resource. Previous studies have shown that Alfin-like and NAC transcription factors from soybean enhance salt tolerance in transgenic Arabidopsis (Wei et al., 2009; Hao et al., 2011) and DOF, bZIP, and MYB transcription factors promote oil accumulation (Wang et al., 2007; Song et al., 2013; Liu et al., 2014). In this study, we investigated the potential roles of melatonin in regulation of soybean growth, yield-related traits, and stress tolerance. We found that melatonin promoted plant growth, increased yield, and improved abiotic stress tolerance. Transcriptome analysis revealed that melatonin may exert its functions mainly through regulation of photosynthesis, the cell cycle, DNA replication, starch/sucrose metabolism, and lipid biosynthesis. Materials and methods Melatonin application Melatonin was dissolved in 100% ethanol (EtOH) at a concentration of 30mM and stored at –20 °C. For coating seeds with melatonin, storage solution was diluted to 1mM with 100% EtOH and then further diluted to different concentrations (0 µM, 50 µM, 100 µM) with seed-coating-reagent (Bayer, Germany). Soybean seeds were coated with 300 µl per 100-seed reagent and dried in the air at room temperature. For the RNA-sequencing experiments, storage solution was diluted to 1mM with 100% EtOH and then further diluted to 100 µM with water. Growth conditions The soybean seeds (Glycine max, SuiNong 28, SN28) were sowed in pre-watered soil. The seedlings were grown in a sunlit greenhouse, with the temperature about 25 °C at night and 30–35 °C during the day. The size of the unifoliate and trifoliate was measured during their growth. Agronomic traits, including pods per plant, seeds per plant, and 100-seed weight were calculated. Thirty plants of each concentration were measured and the experiment was repeated independently. A t-test was performed to detect significant differences compared with control plants. Performance of soybean plants in field test Melatonin-coated soybean seeds were sowed in the experimental station of our institute in Beijing (located at 40°22′ N and 116°22′ E). The soil was first watered and then soybean seeds were sowed with a spacing of about 7cm. To ensure the germination rate, three seeds were sowed in one hole. If more than one seedling germinated at each site, only the healthiest seedling was kept and the others were removed within 3 weeks. Thirty plants from each row were measured for agronomic traits after harvest. Evaluation of the plants under stress Melatonin-coated soybean seeds were sowed in greenhouse. For the salt-stress test, seven-day-old seedlings were transferred to soil saturated with 1% (w/v) NaCl. The seedlings were grown at 25 °C under artificial light (about 20,000 LUX) with a photoperiod of 16-h light and 8-h dark. The phenotypes were analysed at one and three weeks later. Thirty six plants of each concentration were measured for plant height and leaf area; ten plants of each concentration were measured for biomass and five plants were measured for EL. For the drought-stress test, seven-day-old seedlings were tested for their performance. The soil used in this experiment was completely crushed and mixed with vermiculite. This mixed soil has the water capacity of 120% (w/w). The water supply was interrupted for about 12 d and the pot weight was measured every 2 d until the water content dropped to 20% of field capacity. The plants were kept under this drought condition for 10 d (with proper water supplement every day if water content was below 20%) and then the plants from above the cotyledon node were harvested. The plants were dried at 75 °C for at least 2 d and then their biomass was measured (dry weight). The value of biomass was compared with the well-watered plants and the reduction in biomass was calculated (Harb and Pereira, 2011). Ten plants of each concentration were measured for biomass. Both salt and drought experiments were repeated independently and a t-test was performed to detect significant differences compared with control plants. Chlorophyll content measurement After treatment in 1% NaCl for 3 weeks, the leaves of soybean were cut for a chlorophyll assay. The fresh weight of leaves was measured (m). The leaves were ground with silica sand and 1ml of 95% EtOH. The mortar was washed with 95% EtOH and all of the EtOH was transferred to clean tubes with a final volume of 25 (V) ml. Chlorophyll was measured with spectra of 645nm and 663nm using spectrophotometer. Chlorophyll A (mg g–1)=(12.72A663–2.59A645)×V/(m×1000), chlorophyll B (mg g–1)= (22.88A663–4.67A645)×V/(m×1000). Three seedlings of each concentration were used in a chlorophyll assay. Relative electrolyte leakage assay After treatment in 1% NaCl for about 3 weeks, the first trifoliate was cut for the relative electrolyte leakage assay. The leaf was vacuumed and placed at room temperature for 2h. Conductivity (K1) was then measured. Bottles containing the leaves were also autoclaved for 15min to completely destroy the leaves. The samples were shaken at 200rpm at room temperature for 1h. Conductivity (K2) was measured again. REL (relative electrolyte leakage) was calculated as K1/K2. DAB staining Five-day-old seedlings were transferred into soil containing 1% (w/v) NaCl and maintained for about 3 weeks. The central-trifoliate was cut and soaked in 1mg ml–1 DAB (diaminobenzidine) solution (50mM Tris-HCl pH 4.0). After vacuum infiltration, the soybean leaf became translucent. Following DAB staining for one day and decolouration with absolute alcohol, the brown colour on the leaves indicated presence of hydrogen peroxide. RNA extracting and sequencing Three-week old seedlings were treated with water, 100 µM melatonin, 1% NaCl or 100 µM melatonin plus 1% NaCl. Because gene expression in response to environmental change is a relatively quick process, seed-coating-reagent is not appropriate for this experiment owing to its slow-releasing effect. Therefore, melatonin was directly supplied to soybean seedlings with aqueous solution. Total RNA was extracted using TRNzol Reagent (TIANGEN company). RNA-sequencing was performed by GENEWIZ company using Illumina HiSeq. After cutting off the adaptor sequence and deleting low-quality reads, raw reads were mapped to the soybean genome (http://www.plantgdb.org) using software BWA (Burrows-Wheeler Alignment, bwa-0.7.4). Differentially expressed genes were analysed using the RPKM method (reads per kilo bases per million reads): RPKM=109C/NL. “C” identifies a read number that uniquely mapped to a certain gene. “N” identifies a read number that uniquely mapped to the entire genome. “L” identifies the length of a certain gene. Gene ontology (GO) annotation and enrichment analyses were performed using a Blast2Go and GO-TermFinder (0.86) based on results of blastx. Up-/down-regulated transcripts (fold change ≥2) were examined for common genes using an online Venn diagram tool (http://bioinfogp.cnb.csic.es/tools/venny/index.html). Gene function was then annotated on KAAS (KEGG Automatic Annotation Server). Further detailed analysis was performed using perl program. Quantitative RT-PCR was performed to test the results of RNA-Seq using RAN extracted from independently grown and treated seedlings. The primers of qRT-PCR are found in Supplementary Table S4. Raw data of RNA-Seq was uploaded to NCBI (GEO accession number: GSE57960). Fatty acid content analysis Seeds from a field test were analysed for their fatty acids (FA) content. Soybean seeds were ground to a fine powder and FA were extracted based on a previously published method (Poirier et al., 1999) and analysed by gas chromatography (GC2014, SHIMADZU). Results Melatonin improves the growth and yield when coated onto soybean seeds During agricultural procedures, soybean seeds are usually coated with seed coating-reagent for protection. In the present study we coated soybean seeds with seed-coating-reagent (Bayer, Germany) containing different concentrations of melatonin and sowed them in a greenhouse. Coated seeds were sowed in potted soil with saturated water irrigation and germination rate was assessed every day. A higher concentration (200 µM) of melatonin had no significant effect (Supplementary Fig. S1) or even inhibitory effect (Hernandez-Ruiz et al., 2004) on seed germination. However, lower concentrations of melatonin (50 or 100 µM) promoted seed germination when compared with the control treatment (Fig. 1A). Most seeds germinated between the third to fifth day after sowing and these seedlings were used for further analysis. The seeds that germinated too early or too late were abandoned. Seedlings from melatonin-coated seeds had significantly larger leaves than seedlings from control-coated seeds (0 µM) (Fig. 1B). Because of the slow-releasing effect of coating-reagent, this phenomenon was observed two to three weeks after sowing. In the fifth week, melatonin-treated plants were taller and developed one more trifoliate leaf than the control plants (Fig. 1C, D). Before harvest, the central leaf of the third trifoliate from the top, which was fully expanded, was measured. The trifoliate leaves of melatonin-treated plants were much larger than those of the control seedlings (Fig. 1E, F). These results indicate that melatonin promotes soybean growth and development. Fig. 1. Melatonin effects on soybean growth in a greenhouse using the seed-coating method. (A) Germination rate of soybean seeds coated with different concentrations of melatonin. (B) Melatonin effects on leaf growth. Upper panel: leaf phenotype after treatment. Lower panel: measurement of leaf area. (C) Phenotype of five-week-old soybean seedlings after melatonin treatment. (D) Number of trifoliate after melatonin treatment. (E) The top third trifoliate of 11-week old seedlings after melatonin treatment. (F) Leaf area of central-trifoliate after melatonin treatment. For B, D, and F, * and ** indicate significant difference (P<0.05 and P<0.01, respectively) compared with mock coating (0 µm). Bars indicate standard deviation (n=30). Three months after germination, soybean seeds were harvested and agronomic traits were measured. Melatonin-treated soybean plants produced more pods and seeds than the controls (Fig. 2A–D). However, the 100-seed weight was not significantly influenced (Fig. 2E). These results indicate that melatonin increases yield of soybean plants grown in pots. Fig. 2. Yield-related traits from soybean plants grown in a greenhouse. (A) Phenotypes of soybean plants before harvest after treatments with different concentrations of melatonin. (B) Pods and seeds from plants with different treatments of melatonin. (C) Comparison of pod numbers after melatonin treatment. (D) Seed numbers in plants treated with melatonin. (E) Weight of 100 seeds after melatonin treatment. For C and D, ** indicate significant difference (P<0.01) compared with mock coating (0 µM). Bars indicate standard deviation (n=30). (This figure is available in colour at JXB online.) Performance of melatonin-treated soybean plants in a field test Soybean seeds coated with 0, 50, or 100 µM melatonin were sowed in four different regions of the same field in the experimental station. Melatonin-treated and untreated plants were grown in rows, one close to each other, and each row had roughly 70 holes. Melatonin-treated plants grew bigger than control seedlings (Fig. 3A, B). After harvest, yield-related traits were measured. Melatonin-treated plants produced more pods, more seeds and more yield than control plants (Fig. 3C–E). The results suggest that melatonin improves plant growth and soybean production under field conditions. An independent field test was also performed in Zhejiang Province and consistent enhancement in soybean yield was observed (data not shown). Fig. 3. Melatonin effects on soybeans grown under field conditions. (A) Comparison of three-week-old plants grown in the field. (B) Phenotypes of six-week-old plants. (C) Comparison of pod numbers from 30 plants. (D) Seed numbers from 30 plants. (E) Seed weight from 30 plants. For C, D, and E, * and ** indicate significant difference (P<0.05 and P<0.01 respectively) compared with mock coating (0 µM). Bars indicate standard deviation (n=4). Melatonin increases salt and drought tolerance of soybean We further tested whether melatonin had any effects on abiotic stress responses in soybean plants. Five-day-old seedlings from melatonin-coated seeds were grown in soil with 1% (w/v) NaCl. One week later, leaf area and plant height were measured. Melatonin-treated seedlings were taller and had larger leaves than the control plants (Fig. 4A–D). The treated plants also had a smaller reduction of biomass when compared with the control plants (Fig. 4E). During the third week, the leaves of the control seedlings turned yellow, whereas melatonin-treated seedlings were still green (Fig. 4F). Chlorophyll content was also measured and melatonin-treated plants had similar chlorophyll content as those untreated plants under normal conditions. However, these plants had higher chlorophyll contents than that of control plants after salt treatment (Fig. 4G). DAB staining documented that the control seedlings had higher H2O2 levels than the melatonin-treated seedlings as the leaves of the control seedlings had a deeper brown colour (Fig. 4H). The relative electrolyte leakage was lower in melatonin-treated seedlings compared with the control seedlings under salt stress (Fig. 4I). These findings imply that melatonin increases salt tolerance in soybean plants. Fig. 4. Performance of melatonin-treated seedlings in response to salt stress. (A) Melatonin effects on seedlings treated with 1% salt for one week. (B) Melatonin action on plant height after salt stress. (C) Phenotypes of one-week-old treated seedlings after melatonin and salt treatments. (D) Comparison of leaf area after treatments. (E) Reduction in biomass after treatments. Reduced proportion of biomass (dry weight)=[(biomass of well-watered plants)–(biomass of salt-treated plants)]/(biomass of well-watered plants). (F) Phenotypes of three-week-old treated seedlings. (G) Chlorophyll contents in soybean leaves after salt stress. Left part represents content of chlorophyll A, and right part represents content of chlorophyll B. (H) DAB staining. Brown colour indicates accumulation of H2O2. Bars=1cm. (I) Relative electrolyte leakage in treated plants. * and ** indicate significant differences (P<0.05 and P<0.01, respectively) compared with mock coating (0 µM). Bars indicate standard deviation. For leaf area and plant height, n=36; for biomass analysis, n=10; for chlorophyll test, n=3; for relative electrolyte leakage, n=5. One-week old seedlings from melatonin-coated seeds were used to test the drought response of plants, and the water supply was discontinued until the moisture content dropped to 20%. Water content dropped a bit faster in melatonin-treated seedlings than that of control seedlings (Fig. 5B). However, under this condition, melatonin-treated seedlings were larger and had less reduction of biomass compared with controls (Fig. 5A and C). The results suggest that melatonin enhances drought tolerance of soybean plants. Fig. 5. Growth of melatonin-treated seedlings in response to drought stress. (A) Performance of soybean seedlings grown in well-watered soil or in soil supplied with 20% water for one week. (B) Water content of the pot-grown plants after drought stress. (C) Reduction in biomass after drought stress. * indicates significant difference (P<0.05) compared with mock coating (0 µM). Bars indicate standard deviation (n=10). Melatonin-regulated gene expression by transcriptome analysis To investigate the possible mechanism of the promotional roles of melatonin on soybean plants, transcriptome analysis was performed. Two-week old soybean seedlings were treated with water, 100 µM melatonin, 1% NaCl or 1% NaCl plus 100 µM melatonin, and RNAs were isolated for RNA-seq analysis. Statistics of clean reads in RNA sequencing are shown in Table 1. Four comparisons were conducted, including treatments of melatonin (Mt) versus water (Mt:H2O), salt versus water (NaCl:H2O), salt plus melatonin versus salt (NaCl+Mt:NaCl), and salt plus melatonin versus melatonin (NaCl+Mt:Mt). Compared with the transcripts of non-treated samples (water), melatonin-treated samples had 5503 up-regulated genes and 2162 down-regulated genes, whereas salt-treated samples had 524 up-regulated genes and 1146 down-regulated genes. Compared with salt-treated samples, NaCl+Mt samples had 1231 up-regulated genes and 233 down-regulated genes. Compared with melatonin-treated samples, NaCl+Mt samples had 1825 up-regulated genes and 4465 down-regulated genes (Fig. 6A). The heatmap by cluster analysis also revealed that melatonin enhanced the expression level of a large number of genes compared with the other three samples (Supplementary Fig. S2). Venn diagrams were used to analyse the relationship between different treatments. Compared with water samples, there were 28 genes up-regulated by all three treatments (Fig. 6B and Supplementary Table S1), suggesting that they may respond to environmental changes. It was presumed from the experiments above that melatonin may mitigate the effects of salt (Fig. 4 and Fig. 6A), and thus the regulation of gene expressions by melatonin and salt were analysed. There were 303 (Fig 6C, left: up in Mt:H2O versus up in NaCl+Mt:NaCl) genes commonly up-regulated and 14 (Fig 6C, right: down in Mt:H2O versus down in NaCl+Mt:NaCl) genes commonly down-regulated by melatonin in the absence and presence of salt. There were 75 (Fig 6C, left: down in NaCl:H2O versus down in NaCl+Mt:Mt) genes commonly down-regulated and 46 (Fig 6C, right: up in NaCl:H2O versus NaCl+Mt:Mt) genes commonly up-regulated by salt in the absence and presence of melatonin. Four comparisons could be divided into two groups, and each group contained two contrasting comparisons (Group I: Mt:H2O and NaCl+Mt:Mt, Group II: NaCl:H2O and NaCl+Mt:NaCl) (Fig. 6C). A reciprocal analysis was also performed and much fewer common genes were found (Supplementary Fig. S3). Details of the genes in the Venn diagrams (Fig. 6C) can be found in Supplementary Table S2. Table 1. Statistics of clean reads in RNA sequencing Samples Length Total reads Total mapped Unique mapped Mapped (%) Unique mapped (%) Seq depth Melatonin 100 39 128 572 29 090 309 24 017 624 74.35 82.56 28.7 H2O 100 39 614 632 29 231 294 23 910 361 73.79 81.80 29.0 NaCl 100 49 078 288 36 222 072 29 620 331 73.80 81.77 35.9 NaCl+Mt 100 63 169 248 46 886 146 38 472 421 74.22 82.05 46.3 Fig. 6. Analysis of RNA-sequencing data. (A) Differentially expressed gene number. Mt:H2O identifies 100 µM melatonin-treated samples versus water controls, NaCl:H2O identifies 1% salt-treated samples versus water controls, NaCl+Mt:NaCl identifies salt- and melatonin-treated samples versus 1% salt- treated samples, NaCl+Mt:Mt identifies salt- and melatonin-treated samples versus melatonin-treated samples. (B) Gene numbers affected by various treatments. Up- and down-regulated genes (fold change≥2) were examined for common genes using Venn diagram. Overlapping areas represent common genes. NaCl+Mt:H2O identifies salt- and melatonin-treated samples versus water controls. (C) Comparison of gene numbers affected by different treatments using Venn diagram. Gene ontology analysis also was performed (http://www.geneontology.org/). Melatonin exhibited similar regulatory roles in both “Mt:H2O” and “NaCl+Mt:NaCl” comparisons, whereas salt stress inhibited related gene expressions. Regulated gene numbers were higher in the “Mt:H2O” comparison. Melatonin apparently increased genes related to hydrolase, oxidoreductase, primary metabolism, oxidation-reduction processes, and lipid metabolic processes, whereas salt stress had an inhibitory effect on these processes (Fig. 7). Fig. 7. Gene ontology analysis in response to different treatments. C, cellular components; F, molecular functions; P, biological processes. Under non-stress conditions, application of melatonin increased expression level of genes connected to cell cycle and DNA replication processes, including BUBR1, CDH1, CYCA, and CYCB genes. However, there was no significant change in gene expressions under salt treatment (Fig. 8, Supplementary Table S3). Fig. 8. Fold changes of gene expressions in cell cycle and DNA replication processes in three comparisons. The log2 (fold change of gene transcripts) value was analysed using cluster software. The annotation of the genes can be found in Supplementary Table S3. To confirm the results of transcriptome analysis, we extracted RNA from independently grown and treated plants and performed quantitative RT-PCR. The important genes that enriched in pathway analysis were tested and the real-time PCR results were consistent with transcriptome analysis (Figs 9B, 10B, 11B, and Supplementary Fig. S4B). Fig. 9. Melatonin enhances expressions of photosynthesis-related genes under normal and salt stress conditions. (A) Expression of genes related to photosynthesis. Red colour indicates up-regulation and green colour indicates down-regulation. Quadrangle represents a comparison of melatonin treated versus water control; circle represents a comparison of salt versus water control: triangle represents a comparison of salt plus melatonin versus salt. (B) Relative gene expression level analysed by q-PCR. A GmTubulin fragment was amplified as an internal control. Bars indicate standard deviation (n=3). Fig. 10. Changes of expression of genes involved in glucose metabolism. (A) Fold changes of gene expressions in the glucose metabolic pathway. (B) Quantitative PCR analysis of gene expression in glucose metabolism. The annotation of the genes can be found in Supplementary Table S3. Fig. 11. Altered expressions of genes involved in glycolysis, TCA cycle, ethanol metabolism, and fatty acid biosynthesis. (A) Fold change of the expression of genes in the four pathways. Dashed lines indicate omitted steps. (B) Quantitative PCR analysis of genes in glycolysis, ethanol metabolism, and fatty acid biosynthesis. Melatonin up-regulates gene expressions in photosynthesis Both the photosynthetic light reaction and dark reaction processes were up-regulated by melatonin (Fig. 9A, B). The genes, PsaA, PsaF, PsaG, PsaH, PsaK, and PsaO in photosystem I, and PsbE, PsbO, PsbP, PsbQ, PsbY, PsbZ, and Psb28 in photosystem II were up-regulated in melatonin-treated plants compared with those in non-treated plants (Fig. 9A, B). Electron transporter genes, PetF family, and an F-type ATPase gene ATPF1A were also up-regulated in the “Mt: H2O” comparison. The PetF-1 gene was down-regulated in salt-treated plants but this was reversed by melatonin application during salt stress treatment (Fig. 9A, B). In the Calvin cycle, rbcS, GAPC1, and GAPCP-2, which encoded glyceraldehyde-3-phosphate dehydrogenase, were up-regulated by melatonin under normal and salt stress conditions (Fig. 9A). These results show that melatonin improves photosynthesis-related processes under normal and salt stress conditions. Gene expression changes in starch and sucrose metabolism The synthesis genes for sucrose and trehalose were up-regulated by melatonin but down-regulated by salt treatment. Melatonin also activated both the synthesis and degradation of cellulose, pectin, and xylan, whereas salt inhibited these processes for the first two components (Fig. 10A, B). Genes related to ascorbate synthesis and metabolism, including the UDP-glucuronosidase gene, VTC4, and APX4 were also up-regulated by melatonin (Fig. 10A). Melatonin also enhanced some of the above gene expressions during salt stress (Fig. 10A, B). Gene expression changes in glycolysis and downstream processes Under non-stress conditions, gene expression of enzymes that catalyse reactions from glucose to fructose-6P, including HK, ALDEP and GLUPE, were increased by melatonin. The genes PGK and PK connected with pyruvate biosynthesis were also up-regulated by melatonin. When melatonin was combined with salt treatment, PFK, GAPC1, GAPCP-2, PGAM, and PKP2 were up-regulated. The downstream processes for pyruvate metabolism were also changed by melatonin and salt. For ethanol synthesis and metabolic processes, GroES-like genes and ALDH3 were up-regulated by melatonin. ADH1 and GroES-like genes showed the opposite expression in “NaCl+Mt:NaCl” comparisons. Pyruvate can be catalysed to acetyl-CoA, which further participates in the tricarboxylic acid (TCA) cycle and fatty acid biosynthesis. In the TCA cycle, ACLA, MDH, and FUM2 were up-regulated by melatonin. In fatty acid biosynthesis, the KAS I and KCS gene family were up-regulated by melatonin (Fig. 11A, B). Some of the gene expressions were also further confirmed by quantitative PCR (Fig. 11B). Thus, melatonin promotes glycolysis and facilitates processes involving pyruvate and acetyl-CoA. As acetyl-CoA is the substrate of de novo fatty acid biosynthesis, we further measured fatty acid content in soybean seeds from field-grown plants using gas chromatography. Total FA contents were increased by 1.58% and 2.37% with 50 and 100 µM melatonin treatment, respectively (Fig. 12). These increases were probably due to the major rises of C18:2 composition (Fig. 12). Fig. 12. Fatty acid analysis in melatonin-treated plants. Seeds from field-grown plants were analysed for their FA content (% w/w). Bars indicate standard deviation (n=3). * indicates significant difference (P<0.05) compared with mock coating (0 µM). Discussion We have examined the effects of melatonin on soybean plants and found that melatonin, when coated onto seeds, promotes plant growth, development, and yield. It also improved salt and drought stress tolerance. These roles are most likely achieved through enhancement of processes involved in photosynthesis and sugar metabolism. These results provide a novel approach for improving yield of soybeans and possibly of other crops commonly used in agriculture. Previous studies reported that melatonin enhances root growth in other plants (Arnao and Hernandez-Ruiz, 2007; Chen et al., 2009; Sarropoulou et al., 2012). The present study proved that melatonin also improves soybean growth at both the vegetative stage (Fig. 1) and the reproductive stage (Fig. 2). It also increases abiotic stress tolerance (Figs 4 and 5) and the accumulation of fatty acids in soybean (Fig. 12). Melatonin has functions in plants that differ from those in animals; one of these is growth improvement (Tan et al., 2012). Melatonin not only enhanced the size of soybean seedlings, but also improved their growth rate (Fig. 1). New trifoliates developed faster when treated with melatonin (Fig. 1C, D). Moreover, melatonin also increased yield of soybean both in greenhouse and in the field (Fig. 2 and 3), suggesting its potential application in agriculture. Like plant hormones, melatonin displayed weak effects at higher concentrations (Supplementary Fig. S1), or even had inhibitory actions (Hernandez-Ruiz et al., 2004). In the field test, 50 µM melatonin-treated seedlings seemed to be heathier than control seedlings or 100 µM-treated seedlings (Fig. 3A, B). However, 100 µM melatonin-treated plants had much higher seed number than control plants and had slightly higher seed number than 50 µM melatonin-treated plants (Fig. 3D). This fact indicates that high concentrations of melatonin may allow the effects to persist for a long period, thereby more significantly enhancing the yield. Melatonin improves salt and drought tolerance in soybean plants as observed from the increased height and leaf area, and less biomass reduction when subjected to these stresses (Figs 4, 5). These effects are probably a result of the increased antioxidative ability and more stable membrane systems, as judged from the DAB staining and electrolyte leakage (Fig. 4H, I). Transcriptome analysis was performed to investigate the possible mechanisms by which melatonin promotes plant growth and stress tolerance. From the observed results, we propose that melatonin may act as an activator of many genes. In the current study, salt stress suppressed many genes, whereas melatonin by yet undefined mechanisms was able to overcome the inhibitory effects of salt stress and reactivated many of the suppressed genes (Fig. 6A and Supplementary Fig. S2). Gene ontology analysis also showed that melatonin promoted the expression of many genes and inhibited the effects of salt stress (Fig. 7). The promotional effects of melatonin on plant growth may be achieved through activation of DNA replication and cell division as many related genes are up-regulated (Fig. 8). In vascular plants, the photosystem consists of two parts, photosystem I and photosystem II. Two subunits (PsaK and PsaG) in photosystem I can influence plant size because the deletion mutants of them, psak-1 and psag-1.4, had smaller plant size (Varotto et al., 2002). Melatonin enhanced expression levels of PsaK and PsaG (Fig. 9A), which may further enhance plant size of soybean plants. In photosystem II, water is converted to oxygen and protons in a cluster of oxygen-evolving complexes (OEC) (Cady et al., 2008). PsbO (oxygen-evolving enhancer protein 1/OEE1) is essential for the stabilization of the cluster; and PsbP (OEE2) is required for the oxygen-evolving activity (Mayfield et al., 1987). The expression levels of PsbO and PsbP may influence the activity of OEC and thus influence plant growth. It has been found that mutation of the PsbO gene caused growth retardation in Arabidopsis (Murakami et al., 2005). Melatonin up-regulated expression levels of PsbO and PsbP (Fig. 9A), hence leading to the larger size of soybean plants. Melatonin enhances ferredoxin gene PetF and suppresses salt inhibition of this gene (Fig. 9). Ferredoxin regulates the amount of reduced ascorbate and protects chlorophyll from degradation (Lin et al., 2013). Low expression of PetF may affect the scavenging of reactive oxygen species (ROS) generated during photosynthesis or as a result of salt stress, consistent with the growth retardation and H2O2 accumulation in salt stress (Fig. 4). Melatonin may promote PetF expression under salt stress and, hence, reduce H2O2 accumulation (Fig. 4H and 9). We also found that genes involved in ascorbate metabolism, including VTC4 and APX4, are up-regulated by melatonin under normal and/or salt stress conditions (Fig. 10A). The former gene is involved in biosynthesis of ascorbate (Torabinejad et al., 2009), and the latter gene functions in reducing H2O2 levels (Panchuk et al., 2005). These findings indicate that melatonin probably also has a role in the promotion of the antioxidative capacity of soybeans. Sugar metabolism-related genes were also altered by melatonin. Both the synthesis- and degradation-related genes were up-regulated in the melatonin-treated plant, suggesting an active metabolism of sugars possibly for the activated cell division/cell cycle process during enhanced plant growth. These results agree well with GO analysis that primary metabolism was enhanced by melatonin (Fig. 7). Melatonin also promoted the expression of the trehalose synthesis gene. Trehalose is an important carbohydrate that helps plants preserve their cellular integrity under various stresses (Jain and Roy, 2009). Melatonin enhanced the genes involved in glycolysis under both normal and salt stress conditions (unidirectional arrows in Fig. 11A). Glycolysis is responsible for glucose conversion into pyruvate, which is further converted into acetyl-CoA required for the biosynthesis of fatty acids (Jeoung et al., 2014). Additionally, melatonin raised the expression of a number of genes in fatty acid biosynthesis (Okuley et al., 1994; Millar and Kunst, 1997; Wu and Xue, 2010) (Fig. 11), which accounted for the fatty acid accumulation in soybean seeds (Fig. 12). Recently, we identified a transcription factor, GmbZIP172, which binds and activates the expression of two sucrose transporter genes and three cell-wall invertase genes. In GmbZIIP172-overexpressing Arabidopsis plants, sucrose and glucose contents are increased in young seeds, leading to elevated level of oil accumulation in mature seeds of transgenic plants (Song et al., 2013). Gene expression related to changes in amino acid metabolism were also detected (Supplementary Fig. S4A). Tryptophan is the precursor of melatonin and ASMT is the last enzyme of melatonin biosynthesis (Kang et al., 2011). The up-regulation of ASMT by exogenous melatonin application suggests the possibility of a positive feedback control of melatonin synthesis (Supplementary Fig. S4B). This mechanism may be the basis for the observation that low amounts of melatonin induced huge and long-lasting promotional effects on plant growth. The results show that melatonin increases plant growth, seed production, and abiotic stress tolerance in soybean plants, possibly through enhancement of photosynthesis, carbohydrate metabolism, and antioxidative actions. This agent may have great potential for improving crop yield. Further study should examine the molecular mechanisms of melatonin’s functions in plants. Supplementary data Supplementary data are available at JXB online Figure S1. Germination rate of soybean seeds coated with different concentrations of melatonin. Figure S2. Cluster analysis of the four samples. Figure S3. Venn diagram analysis of the four comparisons. Figure S4. Pathways for biosynthesis and metabolism of amino acids. Table S1. Common genes up-regulated in all treatments compared with H2O samples. Table S2. Details of Venn diagram in Fig. 6C. Table S3. Gene annotation. Table S4. Realtime PCR primers. Supplementary Data

0 comments Cited 233 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: found

Is Open Access

Comparative physiological, metabolomic, and transcriptomic analyses reveal mechanisms of improved abiotic stress resistance in bermudagrass [Cynodon dactylon (L). Pers.] by exogenous melatonin

Haitao Shi, Chuan Jiang, Tiantian Ye … (2014)

Introduction Melatonin (N-acetyl-5-methoxytryptamine) is a well-known animal hormone that is involved in many biological processes including sleep, mood, circadian rhythms, retinal physiology, seasonal reproductive physiology, temperature homeostasis, sexual behaviour, antioxidative activity, and immunological enhancement (Galano et al., 2011; Venegas et al., 2012; Calvo et al., 2013). However, melatonin is not only found exclusively in animals, but is ubiquitously present in almost all forms of life including protists, prokaryotes, eukaryotic unicells, algae, fungi, and plants (Dubbels et al., 1995; Hattori et al., 1995; Kolář and Macháčkova, 2005; Arnao and Hernández-Ruiz, 2006, 2007; Tan et al., 2007a, b, 2012). In 1995, two reports first identified melatonin in higher plants (Dubbels et al., 1995; Hattori et al., 1995). In the last 20 years, additional research found that melatonin is universally distributed in leaves, roots, stems, fruits, and seeds of all plant species examined (Manchester et al., 2000; Reiter et al., 2001; Kolář and Macháčkova, 2005; Hernández-Ruiz and Arnao, 2008; Murch et al., 2009; Zhao et al., 2013). Interestingly, remarkably high concentrations of melatonin have been identified and quantified in popular beverages (beer, tea, coffee, and wine), crops (barley, corn, grape, wheat, rice, tobacco, and oats), and Arabidopsis in comparison with those in animals (Manchester et al., 2000; Kolář and Macháčkova, 2005; Arnao and Hernández-Ruiz, 2006, 2009b , 2013a , b; Tan et al., 2007a, 2012; Ramakrishna et al., 2012). Additionally, the melatonin content of tomato and rice has been modified by genetic engineering (Okazaki and Ezura, 2009; Okazaki et al., 2009, 2010; Byeon et al., 2012, 2013, 2014; Byeon and Back, 2014a , b). The well-known beneficial effects of melatonin on human health and the abundance of melatonin in popular beverages and crops may encourage the daily consumption of these products (Tan et al., 2012). To date, melatonin has also been found to be a ubiquitous modulator in multiple plant developmental processes and various stress responses (Kolář and Macháčkova, 2005; Arnao and Hernández-Ruiz, 2006; Tan et al., 2007a, 2012). Changes in melatonin in plants may be involved in circadian rhythms, flowering, promotion of photosynthesis, preservation of chlorophyll (Arnao and Hernández-Ruiz, 2009a ; Tan et al., 2012), stimulation and regeneration of root system architecture (Hernández-Ruiz et al., 2005; Pelagio-Flores et al., 2012; Zhang et al., 2014), delayed senescence of leaves (Byeon et al., 2012; Wang et al., 2012, 2013 a, b), and alleviation of oxidative damage mediated by reactive oxygen species (ROS) and reactive nitrogen species (RNS) burst (Tan et al., 2012) Moreover, melatonin protects against multiple abiotic processes such as cold stress (Posmyk et al., 2009a ; Kang et al., 2010; Bajwa et al., 2014), copper stress (Posmyk et al., 2008, 2009b), high temperature (Byeon and Back 2014b ), salt stress (Li et al., 2012), osmotic stress (Zhang et al., 2013), drought stress (Wang et al., 2014), and pathogen infection (Yin et al., 2013). The mechanisms were partially characterized after the direct exogenous application of melatonin to plants (Posmyk et al., 2008, 2009a, b; Zhao et al., 2011; Li et al., 2012; Pelagio-Flores et al., 2012; Wang et al., 2012, 2013a , b; Yin et al., 2013; Bajwa et al., 2014) or the creation of transgenic plants that produced either more or less melatonin through modulating its metabolic pathway (Kang et al., 2010; Byeon et al., 2013, 2014; Park et al., 2013; Byeon and Back, 2014a ; Wang et al., 2014). Finally, the recent studies which showed the protective roles of melatonin in response to abiotic stress indicate that this indole might be a potentially ideal target for future genetic engineering technology to improve abiotic stress resistance in plants. Thus, transgenic plants with higher melatonin concentration might lead to breakthroughs to improve crop production in agriculture as well as the general health of humans (Tan et al., 2012). Bermudagrass [Cynodon dactylon (L). Pers.] is a warm-season turfgrass for lawns, parks, and sport fields cultivated worldwide (Shi et al., 2012, 2013a, b , 2014 b, c, d). In response to global changed environmental stresses, improvement of abiotic stress resistance is very important for grass engineering (Shi et al., 2012, 2013a, b , 2014 b, c, d). As mentioned above, melatonin might be an ideal target for future genetic engineering of some plant species. However, the endogenous melatonin concentration and the possible role of melatonin in response to abiotic stress in bermudagrass is largely unknown. In this study, endogenous melatonin was examined after abiotic stress treatments in bermudagrass plants, and exogenous melatonin treatment was applied to investigate the in vivo role of melatonin in the response of bermudagrass to abiotic stress. In addition, the effects of exogenous melatonin treatment on ROS accumulation and cell damage, as well as underlying antioxidant responses, were determined. Moreover, comparative metabolomic and transcriptomic analyses were performed to identify differentially expressed metabolites and genes after exogenous melatonin treatment. This study provided some insights into the physiological and molecular mechanisms of melatonin in bermudagrass responses to multiple abiotic stresses. Materials and methods Plant materials and growth conditions Newly harvested common bermudagrass seeds were used in this study. After stratification in deionized water at 4 °C for 4 d in darkness, the bermudagrass seeds were sown in soil in the growth room, which was controlled at 28 °C, with an irradiance of ~150 μmol quanta m–2 s–1, 16h light and 8h dark cycles, and ~65% relative humidity. Plant abiotic stress treatment To test the effect of exogenous melatonin on plant physiological responses and abiotic stress resistance, 21-day-old bermudagrass plants were irrigated with water or with different concentrations of melatonin solutions for 7 d, respectively. After melatonin pre-treatment, all control and melatonin-pre-treated 28-day-old bermudagrass plants were subjected to subsequent salt, drought, or cold stress treatments. For salt stress treatment, 28-day-old bermudagrass plants were irrigated with NaCl solutions for 25 d; the NaCl concentration was increased stepwise by 50mM every 5 d to a final concentration of 300mM. For drought stress treatment, 28-day-old plants were subjected to a drought condition by withholding water for 21 d and then re-watered for 4 d. For cold stress treatment, 28-day-old bermudagrass plants were subjected to 4 °C treatment for 21 d, and then transferred to –10 °C for 8h. The freezing stress-treated plants were then recovered overnight at 4 °C and transferred to a standard growth room (28 °C) for 4 d. In each independent experiment, three pots with ~40 plants in each pot were used for each treatment in one concentration of melatonin, and at least three independent experiments were performed to obtain the results. The survival rate of the salt-, drought-, or freezing-stressed plants was recorded at 25 d after stress treatments. The plant leaf samples from melatonin-pre-treated 28-day-old plants were collected at the indicated time points after salt, drought, or cold treatment for the assays of multiple of physiological parameters. Quantification of melatonin by enzyme-linked immunosorbent assay (ELISA) Melatonin from plant leaves was extracted using the acetone–methanol method as described by Pape and Lüning (2006). Briefly, 1g of plant leaf samples was ground in liquid nitrogen, and then transferred to 5ml of extraction mixture (acetone:methanol:water=89:10:1) and homogenized extensively on ice, and the homogenate was centrifuged at 4500 g for 5min at 4 °C. The supernatant was moved to a new centrifuge tube containing 0.5ml of 1% trichloric acid for protein precipitation. After centrifugation at 12 000 g for 10min at 4 °C, the extract was used for quantification of melatonin using the Melatonin ELISA Kit (EK-DSM; Buhlmann Laboratories AG, Schonenbuch, Switzerland) according to the manufacturer’s instruction as described in Shi and Chan (2014a ). Quantifications of chlorophyll content Plant leaf chlorophyll was extracted using 80% (v/v) acetone for 6h with shaking in the dark. The concentration of chlorophyll was then calculated by examining the absorbance at 645nm and 663nm of the centrifuged supernatant. Quantification of electrolyte leakage (EL) The EL of plant leaves under control and abiotic stress conditions was assayed using a conductivity meter (Leici-DDS-307A, Shanghai, China) as previously described (Shi et al., 2012, 2013a, b , 2014 b, c, d). The relative EL was expressed as the ratio of initial conductivity to the conductivity after boiling. Determination of malondialdehyde (MDA) content The MDA content was extracted using chilled thiobarbituric acid (TBA) reagent, and was quantified via determining the absorbance of the supernatant at 450, 532, and 600nm as previously described (Shi et al., 2012, 2013a, b , 2014 b, d). Determination of ROS accumulation and antioxidants As two major indicators of ROS accumulation, hydrogen peroxide (H2O2) and superoxide radical (O2·–) contents were quantified using the titanium sulphate method and the Plant O2·– ELISA Kit (10-40-488, Bejing Dingguo, Beijing, China) as previously described (Shi et al., 2012, 2013a, b, 2014 b, d). The activities of three antioxidant enzymes, namely superoxide dismutase (SOD; EC 1.15.1.1), catalase (CAT; EC 1.11.1.6). and peroxidase (POD; EC 1.11.1.7), were assayed using a Total SOD Assay Kit (S0102; Haimen Beyotime, Haimen city, China), a CAT Assay Kit (S0051; Haimen Beyotime), and a Plant POD Assay Kit (A084-3; Nanjing Jiancheng, Nanjing city, China), respectively, as described by Shi et al., 2012, 2013a, b , 2014 b, d). The concentrations of reduced glutathione (GSH) and oxidized glutathione (GSSG) were determined using the GSH and GSSG Assay Kit (S0053; Haimen Beyotime) as described by Shi et al. (2014b , d ), and the GSH redox state was calculated as the ratio of GSH concentration to the concentration of GSH plus GSSG. Extraction, identification, and quantification of metabolites Extraction, identification, and quantification of metabolites from plant leaves were performed as in Shi et al. (2014d ). Briefly, the metabolite extraction and sample derivatization were performed as in Lisec et al. (2006), then the derivatizated extract was injected into a DB-5MS capillary cloumn (30 m×0.25 mm×0.25 μm; Agilent J&W GC column, California, USA) using gas chromatography time-of-flight mass spectrometry (GC-TOF-MS) (Agilent 7890A/5975C) according to the protocol described by Shi et al. (2014d ). After the GC-TOF-MS assay, the various metabolites were identified via comparing every retention time index-specific mass with reference spectra in mass spectral libraries (NIST 2005, Wiley 7.0). The numerous metabolites were then quantified based on the pre-added internal standard (ribitol) in the process of metabolite extraction. Hierarchical cluster analysis The hierarchical cluster analysis of several metabolites was performed using the CLUSTER program (http://bonsai.ims.u-tokyo.ac.jp/~mdehoon/software/cluster/) and Java Treeview (http://jtreeview.sourceforge.net/) as in Shi et al., 2012, 2013a, b , 2014 b, d). For cluster analysis, all metabolites were quantified as fold change relative to the wild-type bermudagrass plants under control conditions, which was set as 1.0. RNA extraction, library construction, and sequencing For RNA extraction, 28-day-old bermuagrass plants in pots that were irrigated with water or 20 μM melatonin for 7 d were used. Each treatment was represented by two replicate leaf samples, and each sample contained leaves from at least 30 seedlings. Total RNA was extracted with TRIzol (Invitrogen) and was quantified as previously described (Shi et al., 2013c ). RNA quality was determined using a 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) according to the manufacturer’s protocol. The cDNA libraries were constructed with the mRNA-Seq Sample Preparation Kit™ (Illumina, San Diego, CA, USA) and the DNA yield and fragment insert size distribution of each library were determined on the Agilent Bioanalyzer. The cDNA libraries were then sequenced on an Illumina HiSeq2500 sequencing instrument using the 100bp single end protocol. Quantitative real-time PCR The above RNA samples were also used for synthesis of first-strand cDNA with reverse transcriptase (BIO-RAD, Hercules, CA, USA), and the cDNAs were used for quantitative real-time PCR using a CFX96™ Real Time System (BIO-RAD) as previously described (Shi et al., 2013c ). The specific primers of the analysed genes for real-time PCR are listed in Supplementary Table S1 available at JXB online, and the housekeeping genes have been described in Hu et al. (2012). Bioinformatics analyses of RNA-Seq data Raw RNA-Seq reads were first trimmed for low quality regions using clean reads with length longer than 25bp, obtained after trimming low quality bases (Q 1e-5). GO annotation was performed using the Blast2GO pipeline, and 18 701 (66%) transcripts were assigned with at least one GO term. Among the three GO categories, 13 402 transcripts were annotated in Biological Process, 14 052 transcripts in Molecular Function, and 14 685 in Cellular Component. Using fold change >2 and false discovery rate (FDR) 16-fold after melatonin treatment (Table 2). GO enrichment analysis in the biological process domain suggested that genes related to the cysteine biosynthetic process, response to light signal, and the photosynthetic process were down-regulated. In particular, the studies of Wang et al. (2012) showed that melatonin can lower ROS damage of many photosynthetic components. Therefore, the expression of genes involved in the photosystem might been suppressed through a negative feedback mechanism. The up-regulated genes were greatly enriched with the GO terms involved in gene expression regulatory process, such as protein phosphorylation, DNA-dependent transcription, regulation of circadian rhythm, etc. (Fig. 7). Table 1. Pathway enrichment analysis of genes whose expression was significantly affected by exogenous melatonin pre-treatment in bermudagrass Group MapMan pathway Up-regulation Down-regulation NF P-value NF P-value I N metabolism 5.71 0.0000 4.00 0.0140 Major CHO metabolism 3.30 0.0000 2.31 0.0100 TCA/org transformation 2.58 0.0023 2.63 0.0076 Transport 2.05 0.0000 2.42 0.0000 Hormone metabolism 2.02 0.0000 1.44 0.0110 Metal handling 1.78 0.0440 2.19 0.0260 Redox 1.59 0.0160 2.60 0.0000 Secondary metabolism 1.42 0.0094 2.75 0.0000 II Gluconeogenesis/ Glyoxylate cycle – – 24.02 0.0000 Glycolysis – – 2.30 0.0210 Photosystem – – 8.21 0.0000 Tetrapyrrole synthesis – – 4.88 0.0001 Fermentation 6.63 0.0005 – – Minor CHO metabolism 2.99 0.0000 – – Signalling 2.53 0.0000 0.95 0.0550 C1 metabolism 2.32 0.0430 – – III RNA 1.84 0.0000 1.04 0.0340 Cofactor and vitamin metabolism 1.83 0.0400 – – Amino acid metabolism 1.71 0.0036 1.70 0.0120 Lipid metabolism 1.64 0.0009 1.69 0.0023 Nucleotide metabolism 1.54 0.0280 1.87 0.0120 Stress 1.46 0.0000 – – Development 1.41 0.0016 – – Miscellaneous 1.27 0.0015 1.71 0.0000 Protein 1.11 0.0031 0.73 0.0000 IV Cell 0.80 0.0260 – – Not assigned 0.45 0.0000 0.76 0.0000 Mitochondrial electron transport/ATP synthesis 0.24 0.0085 – – DNA 0.12 0.0000 0.18 0.0000 MicroRNA, natural antisense, etc 0.00 0.0000 0.00 0.0000 Twenty-one-day-old bermuagrass plants in pots were irrigated with water or 20 μM melatonin for 7 d, then the 28-day-old plants (without and with 7 d of pre-treatment) were used for transcriptomic analysis. Differentially expressed genes (i.e. with P-value ≤0.05 and log2 fold change ≥1 or log2 fold change ≤ –1) were annotated using the Classification SuperViewer Tool and with MapMan. MapMan was used as the classification source. Group I indicates highly enriched pathways of both up- and down-regulated genes; group II indicates highly enriched pathways of either up- or down-regulated genes; group III indicates slightly enriched pathways; and group IV indictes under-represented pathways. Scales of normalized frequency (NF) are as follows: ≥4 2–4 1–20. 5–1 ≤0.5 Table 2. List of getnes highly induced (>16-fold) by exogenous melatonin pre-treatment in bermudagrass Seq_ID logFC P-value FDR Putative annoation in Arabdopsis E-value a Putative annoation in rice E-value a comp58508_c2_seq12 8.47 5.35E-247 2.50E-245 CBF4, DREB1D 2e-07 DRE 1e-35 comp58508_c2_seq7 7.49 0 0 DREB1A, CBF3 9e-15 DRE 2e-25 comp57631_c0_seq7 2.98 0 0 DREB2A 2e-23 AP2 domain containing protein 2e-62 comp55697_c1_seq4 4.92 0 0 DNA-binding protein 1e-23 AP2 domain containing protein 6e-34 comp49674_c0_seq1 4.49 0 0 HRE1 1e-14 AP2 1e-44 comp52672_c0_seq8 5.57 0 0 ERD15 5e-17 Expressed protein 2e-46 comp51935_c1_seq1 5.89 0 0 Cold-regulated gene 27 3e-13 Expressed protein 4e-74 comp59530_c3_seq1 5.50 4.31E-51 4.22E-50 HSFA6B 5e-61 HSF 4e-133 comp51719_c2_seq14 4.61 0 0 HSFC1 1e-57 HSF 1e-113 comp56606_c0_seq9 5.48 9.30E-52 9.25E-51 HSP20-like 2e-17 hsp20 3e-27 comp60859_c0_seq13 5.13 1.51E-14 5.65E-14 SLT1 | HSP20-like 4e-172 SLT1 protein 0 comp53718_c0_seq3 6.51 0 0 DnaJ 2e-36 Heat shock protein DnaJ 1e-78 comp51393_c1_seq14 5.36 2.40E-58 2.67E-57 DnaJ 6e-16 Heat shock protein DnaJ 1e-36 comp23443_c0_seq1 7.89 6.84E-42 5.67E-41 DNAJ heat shock protein 1e-35 Expressed protein 4e-61 comp53904_c0_seq1 4.32 0 0 WRKY25 4e-35 WRKY53 4e-51 comp55751_c0_seq1 6.18 0 0 WRKY40 2e-53 WRKY71 1e-132 comp55960_c0_seq3 4.57 1.84E-212 7.05E-211 WRKY46 4e-35 WRKY74 1e-93 comp54790_c0_seq12 5.79 1.72E-246 7.99E-245 WRKY51 1e-31 WRKY67 6e-37 comp57307_c2_seq3 5.57 0 0 AZF1 8e-12 C2H2 zinc finger protein 1e-23 comp54697_c5_seq1 4.22 1.63E-163 4.59E-162 ATL6 3e-42 Zinc finger, C3HC4 type 1e-55 comp57307_c1_seq2 5.61 0 0 STZ, ZAT10 4e-25 C2H2 zinc finger protein 3e-36 comp55694_c0_seq3 5.27 0 0 NIP2 1e-51 RING-H2 finger protein ATL2B 2e-90 comp57199_c1_seq6 4.07 1.01E-155 2.69E-154 AtMYB78 6e-14 MYB 2e-14 comp60805_c1_seq6 6.20 0 0 RVE1 5e-36 MYB 7e-97 comp60805_c1_seq4 5.70 0 0 LHY 2e-13 MYB 2e-92 comp58066_c0_seq2 4.59 8.75E-312 5.22E-310 BHLH 3e-38 Ethylene-responsive protein 1e-95 comp54812_c1_seq1 6.33 0 0 BHLH92 3e-19 BHLH 2e-79 comp56171_c3_seq1 5.02 0 0 anac032, NAC032 6e-07 NAC 3e-54 comp54298_c0_seq8 4.56 0 0 AF2 7e-73 NAC 2e-145 comp54697_c3_seq1 4.60 2.23E-132 5.10E-131 RING/U-box protein 1e-28 Zinc finger 4e-49 comp53888_c2_seq1 4.15 9.66E-50 9.21E-49 CMPG2 1e-15 U-box 7e-20 comp53888_c0_seq1 4.74 1.02E-122 2.15E-121 PUB29 2e-31 U-box 9e-90 comp48740_c0_seq1 6.07 0 0 Calcium-binding EF-hand protein 3e-35 EF hand family protein 8e-68 comp48880_c0_seq1 5.33 0 0 Calcium-binding EF-hand protein 1e-28 OsCML31 4e-60 comp61173_c2_seq4 5.64 0 0 Calcium-binding protein 3e-34 OsCML10 1e-57 comp42867_c0_seq1 5.25 6.59E-310 3.93E-308 Calcium-binding protein 1e-43 OsCML14 7e-72 comp50275_c0_seq5 4.36 1.47E-284 7.97E-283 Calcium-binding protein 5e-25 EF hand 1e-40 comp60797_c0_seq1 5.12 0 0 Calmodulin-binding protein 2e-160 Calmodulin-binding protein 0 comp55169_c0_seq2 4.50 0 0 Calmodulin-binding protein 8e-85 Calmodulin-binding protein 3e-145 comp51845_c0_seq1 4.67 0 0 CML43 3e-39 OsCML27 6e-78 comp55740_c0_seq1 4.20 0 0 TCH2, CML24 2e-38 OsCML16 1e-80 comp57565_c0_seq8 9.08 7.50E-220 3.05E-218 CKA1 0 Casein kinase II 0 comp54827_c0_seq1 6.78 6.62E-48 6.09E-47 Kinase 2e-20 Kinase 3e-26 comp55548_c0_seq1 4.39 0 0 Kinase 6e-148 Phosphotransferase 0 comp55703_c1_seq1 4.29 7.18E-159 1.96E-157 Kinase 7e-65 Tyrosine protein kinase 2e-124 comp60546_c0_seq1 4.11 1.75E-75 2.41E-74 Kinase 0 Leucine-rich repeat protein 0 comp54489_c2_seq1 5.52 3.60E-185 1.19E-183 JMJD5 8e-139 jmjC protein 5 9e-146 comp48478_c1_seq1 4.45 1.09E-76 1.52E-75 HDA18 2e-64 Histone deacetylase 2e-71 comp57103_c1_seq1 4.45 0 0 ERF-1 3e-22 ERF 1e-43 comp51949_c0_seq2 4.49 0 0 JAZ2, TIFY10B 2e-10 ZIM protein 8e-31 comp59913_c1_seq8 7.52 0 0 CYP707A1 8e-50 Cytochrome P450 1e-64 comp50455_c0_seq1 3.42 0 0 PYL5, RCAR8 7e-55 Cyclase/dehydrase 1e-87 comp55459_c2_seq1 4.20 0 0 PP2C 6e-101 PP2C 1e-164 comp55059_c0_seq4 4.33 2.77E-269 1.43E-267 SHY2, IAA3 2e-47 OsIAA24 2e-64 comp56859_c0_seq6 5.51 3.14E-169 9.21E-168 Alcohol dehydrogenase 8e-100 Dehydrogenase 9e-100 comp55475_c1_seq1 4.41 0 0 WCRKC thioredoxin 1 4e-45 Thioredoxin 4e-80 comp49522_c0_seq2 4.02 9.96E-36 7.38E-35 Oxidoreductase 2e-88 Dehydrogenase 4e-136 comp57222_c0_seq2 5.09 0 0 FMO1 2e-91 Monooxygenase 0 comp56296_c0_seq6 5.02 2.03E-49 1.92E-48 Copper transport protein 2e-09 Heavy metal-associated protein 3e-26 comp46851_c0_seq1 4.92 5.38E-117 1.07E-115 Heavy metal transport 1e-06 Expressed protein 4e-21 comp51430_c0_seq1 4.45 7.94E-278 4.22E-276 Heavy metal transport 1e-15 Heavy metal-associated protein 3e-41 Twenty-one-day-old bermuagrass plants in pots were irrigated with water or 20 μM melatonin for 7 d, then the 28-day-old plants (with and without 7 d of pre-treatment) were used for transcriptomic analysis. logFC, log2 fold change; FDR, false discovery rate. e E-value, expected value for putative annotation in Arabidopsis or rice. Fig. 7. The Biological Process GO terms enrichment of down-regulated (A) and up-regulated (B) genes between control and melatonin-pre-treated bermudagrass. The horizontal axis shows –log10 of the P-value. Twenty-one-day-old bermuagrass plants in pots were irrigated with water or 20 μM melatonin for 7 d, then the 28-day-old plants (with and without 7 d of pre-treatment) were used for transcriptomic analysis. (This figure is available in colour at JXB online.) To confirm the reliability of the RNA-Seq data, the expression of 18 genes (nine up-regulated and nine down-regulated by exogenous melatonin treatment) that were differentially expressed between control and melatonin-treated plants was assessed via quantitative real-time PCR. Consistently, the results of the real-time PCR assay exhibited the same trend and were correlated with the RNA-Seq data (Supplementary Fig. S1 at JXB online), confirming the reproducibility of RNA-Seq data. Pathway and GO term enrichment analyses The transcriptome data were submitted to the Mercator web tool to align with the public protein database, and 14 288 transcripts were located in at least one point in plant biological pathways. As shown in Table 1, pathway analysis revealed that melatonin affected the expression of many genes involved in N metabolism, minor carbohydrate metabolism, TCA/org transformation, transport, hormone metabolism, metal handling, redox, and secondary metabolism (Table 1, group I). Other transcripts involved in stress response and metabolism were extensively changed after melatonin treatment (Fig. 6B; Supplementary Fig. S2 at JXB online). These results indicated that melatonin treatment might induce a stress response in bermudagrass. The pathway analysis results were consistent with the study carried out by Weeda et al. (2014), which showed that melatonin altered many genes involved in plant defence (Supplementary Fig. S2), and these changes might contribute to the enrichment of stress-related GO terms (Fig. 7). Discussion As sessile organisms, plants have developed sophisticated strategies to respond to diverse environmental stresses. The stress signals are perceived by several receptors at the cell membrane level, followed by their transduction to multiple second messengers such as abscisic acid (ABA), H2O2, nitric oxide (NO), etc. These activate downstream stress-responsive genes and physiological responses, eventually leading to protective responses at the whole-plant level (Shi et al., 2012, 2013a, b , 2014 b, c, d; Shi and Chan, 2014a , b). Although no direct evidence had indicated that melatonin could serve as a second messenger, the induction of melatonin by multiple stress treatments (Fig. 1) indicates an in vivo role for melatonin in bermudagrass response to abiotic stress (Arnao and Hernández-Ruiz, 2006, 2009b , 2013a , b; Tan et al., 2007a, 2012). In this study, the protective role of melatonin on the response of bermudagrass to abiotic stress was revealed. Under control conditions, melatonin had no significant effect on bermudagrass growth or its physiological responses (Fig. 2). Under abiotic stress conditions, however, melatonin-pre-treated plants exhibited significantly higher chlorophyll content, lower EL, and higher survival rate than did non-treated bermudagrass plants (Fig. 2C–E). After recovery from abiotic stress treatments, melatonin-pre-treated plants exhibited better growth status than non-treated plants, with higher biomass (plant height and weight) (Fig. 2F, G). These results indicate that exogenous melatonin application improved salt, drought, or freezing stress resistance in bermudagrass, in accordance with the enhanced resistance to cold stress (Posmyk et al., 2009a ; Kang et al., 2010; Bajwa et al., 2014), copper stress (Posmyk et al., 2008, 2009b), high temperature (Byeon and Back, 2014b ), salt stress (Li et al., 2012), osmotic stress (Zhang et al., 2013), drought stress (Wang et al., 2014), and pathogen infection (Yin et al., 2013) due to melatonin in various plant species. As an antioxidant in animals, melatonin scavenges free radicals directly, stimulates the activities of antioxidant enzymes including CAT, SOD (both MnSOD and CuSOD), glutathione reductase (GR), and glutathione peroxidase (GPX), and increases the efficiency of mitochondrial oxidative phosphorylation (Tan et al., 1993, 1999, 2003, 2007a, b; Reiter et al., 2000; Kolář and Macháčkova, 2005). In plants, melatonin is also an important antioxidant and a radical scavenger (Arnao et al., 1996, 2001; Cano et al., 2003, 2006). Li et al. (2012) also found that exogenous melatonin modulates salinity-induced oxidative damage by directly scavenging H2O2 and enhancing the activities of antioxidative enzymes in Malus hupehensis. Consistently, exogenous application of melatonin significantly activated ROS detoxification of antioxidants, including enzymatic antioxidant enzymes (SOD, CAT, and POD) and non-enzymatic glutathione (GSH redox state), to maintain cellular ROS (mainly including H2O2 and O2·–) at a relatively low level. This results in the alleviation of abiotic stress-induced oxidative damage and further conferred improved abiotic stress resistance (Figs 3, 4). Consistently, RNA-Seq found that many genes involved in redox, many POD genes and glutathione S-transferases (GST) genes were significantly modulated by exogenous melatonin treatment (Table 1; Supplementary Fig. S2 at JXB online). In summary, the positive modulation by exogenous melatonin of the ROS detoxification system might contribute greatly to enhanced abiotic stress resistance in bermudagrass. Moreover, comparative metabolomic analysis showed the actions of melatonin treatment on carbon metabolites and amino acid metabolism under abiotic stress conditions. Notably, melatonin-pre-treated plants exhibited higher concentrations of 54 metabolites compared with non-melatonin-treated plants (Fig. 5; Supplementary Table S2 at JXB online). Among these metabolites, proline and some carbohydrates (fructose, sucrose, glucose, maltose, cellobiose, trehalose, galactose, and galactinol) are important compatible solutes to respond to abiotic stress for osmotic adaptation (Krasensky and Jonak, 2012). Thus, higher levels of proline and carbohydrates (glucose, maltose, fructose, sucrose, and trehalose) in melatonin-pre-treated plants provided beneficial effects under abiotic stress conditions. In addition, higher levels of other metabolites including multiple amino acids, organic acids, and sugars in melatonin-pre-treated plants indicate the beneficial physiological processes in melatonin-pre-treated plants during abiotic stress treatment; the data further confirm the protective role of melatonin in response to abiotic stress. Notably, 18 metabolites comprising 10 amino acids, six sugars, and two sugar alcohols of the carbon metabolic pathway exhibited significantly higher levels in melatonin-pre-treated plants under abiotic stress conditions (Fig. 6A). These results indicate the widespread effects of melatonin treatment in carbon metabolism and amino acid metabolism; these metabolites might play some role in melatonin-mediated abiotic stress resistance in bermudagrass. Additionally, comparative transcriptomic analysis identified 2361 up-regulated and 1572 down-regulated transcripts as a consequence of exogenous melatonin treatment. Quantitative real-time PCR of 18 genes supported the reliability of the RNA-Seq data (Supplementary Fig. S1 at JXB online). Pathway enrichment analysis indicated that eight pathways were over-represented among differentially expressed genes between control and melatonin-treated bermudagrass plants, including N metabolism, major carbohydrate metabolism, TCA/org transformation, transport, hormone metabolism, metal handling, redox, and secondary metabolism (Table 1, group I). The enrichment of redox-related genes affected by melatonin (Table 1, group I) was consistent with the effects of exogenous melatonin on ROS detoxification in bermudagrass (Figs 4, 5). In animals, melatonin is known to be involved in circadian rhythms (Tal et al., 2011). Interestingly, the GO enrichment analysis also showed that transcripts that function in regulation of rhythms and flowering were over-represented (Fig. 7). Notably, the pathway analysis results were consistent with those of the study carried out in Arabidopsis by Weeda et al. (2014). Thus, melatonin altered many genes involved in plant defence in bermudagrass (Supplementary Fig. S2); these changes probably contributed to the enrichment of stress-related GO terms (Fig. 7). Additionally, exogenous melatonin treatment had significant effects on various signalling pathways including primary metabolism, secondary metabolism, photosynthesis, large enzyme families, receptor-like kinases, proteolysis, and autophagy pathways in bermudagrass as determined using MapMan software (Fig. 6B; Supplementary Fig. S2). Those genes modulated by exogenous melatonin treatment might also contribute to melatonin-enhanced abiotic stress resistance in bermudagrass. Asparagine accumulation shows that nitrogen re-distribution and mobilization were important features of the melatonin response (Lea et al., 2007; Maaroufi-Dguimi et al., 2011). Jia et al. (2001) also suggested that asparagine may be a signalling molecule involved in sensing the nitrogen status. In addition, asparagine is an amino group donor for the synthesis of the photorespiratory intermediate glycine. Nagy et al. (2013) and Shi et al. (2014) found that this is also a good indicator of drought stress in drought-tolerant and sensitive wheat and bermudagrass cultivars. At the same time, some carbohydrate metabolism compounds increased: these included fructose, glucose, and trehalose, but not sucrose. Such differential dynamics of carbohydrates could reflect modifications of carbon balance and carbon utilization. Moreover, asparagine synthetase genes that are involved in asparagine synthesis are regulated by the level of carbohydrates (Lam et al., 1998; Foito et al., 2009). TCA/org transformation is important for the Calvin cycle for CO2 assimilation and separation of initial carbon fixation by contact with air and secondary carbon fixation into sugars (Selwood and Jaffe, 2011). Glycolysis is an important metabolic pathway in carbohydrate metabolism, and the central role of glycolysis in the plant metabolic pathway is to provide energy such as ATP and generates precursors for anabolism such as fatty acids and amino acids (Plaxton, 1996). In accordance with the metabolic profiles, transcriptomic analysis found that many genes involved in TCA/org transformation and N metabolism were modulated by melatonin treatment (Table 1; Fig. 6B). Genes which functioned in sucrose and amino acid metabolism were also greatly changed after melatonin treatment (Fig. 6B), leading to altered sucrose and amino acid contents revealed by metabolite analysis (Figs 5, 6). These results showed that the underlying mechanisms related to melatonin may involve major reorientation of photorespiratory and carbohydrate and nitrogen metabolism. To date, various TFs have been shown to be involved in plant stress responses via activating stress-responsive gene expression, such as BASIC LEUCINE ZIPPER PROTEINS (bZIPs), CBF/DREBs, ETHYLENE-RESPONSIVE ELEMENT-BINDING FACTORS (ERFs), MYBs, WRKYs, and zinc finger proteins (ZFPs) (Shi et al., 2014a , e ). In the current study, many TFs were significantly regulated by exogenous melatonin treatment (Table 2; Supplementary Tables S4, S5 at JXB online), and these TFs might contribute to the enhanced stress tolerance of melatonin-treated plants, thus indicating that melatonin might pre-condition to be resistant to abiotic stresses. Some protein kinases [such as mitogen-activated protein kinase (MAPK)] and calcium signalling kinases [including calcium-dependent protein kinases (CDPKs), calcineurin B-like (CBL)-interacting protein kinases (CIPKs), and calcium-related protein kinases (CRKs)] were also transcriptionally regulated by exogenous melatonin treatment (Table 1; Supplementary Tables S4, S5). This suggests that kinase signalling might play critical roles in melatonin-mediated stress responses. As sessile organisms, plants cannot avoid unfavourable stress conditions by adjusting their location; thus, they have evolved complex strategies to perceive stress signals and further translate the perception into effective responses, which might largely depend on various protein kinases and TFs (Shi et al., 2014a , e ). Recently, it was found that one cysteine2/histidine2-type zinc finger TF, zinc finger of Arabidopsis thaliana 6 (ZAT6), is involved in melatonin-mediated freezing stress response, and the AtZAT6-activated CBF pathway was essential for melatonin-mediated freezing stress response in Arabidopsis (Shi and Chan, 2014a ). This study together with others in sunflower (Helianthus annuus) (Mukherjee et al., 2014) and in Arabidopsis (Shi and Chan, 2014a ; Weeda et al., 2014) indicate that melatonin is involved in long-distance signal transduction in plants. Moreover, some important genes in plant hormone signalling [RCAR/PYR/PYL, SNF1-related protein kinases 2 (SnRK2), and nine-cis-epoxycarotenoid dioxygenase (NCED) genes in ABA signalling, and jasmonate (JA)-JIM-domain proteins (JAZs) in JA signalling] that were significantly regulated by exogenous melatonin treatment (Table 1; Supplementary Tables S4, S5 at JXB online) might also have some function in melatonin-mediated cross-talk among plant hormones, as well as in stress responses. Kolář and Macháčkova (2005) also found that melatonin might function as an auxin to promote vegetative growth. These results suggested that melatonin might serve as a plant hormone that cross-talks with other plant hormones. Thus, melatonin triggered extensive transcriptional reprogramming and pre-conditioned resistance to multiple abiotic stresses. Further investigation of the in vivo roles of these genes will shed additional light on melatonin-mediated stress responses in bermudagrass. Taken together, this study provides the first evidence of the protective roles of exogenous melatonin in bermudagrass response to multiple abiotic stresses. This involved the activation of antioxidants, modulation of metabolic homeostasis, and extensive transcriptional reprogramming. Supplementary data Supplementary data are available at JXB online. Figure S1. Validation of differentially expressed genes by quantitative real-time PCR. Figure S2. Effect of exogenous melatonin treatment on stress-related pathways in bermudagrass. Table S1. The specific primers used for real-time PCR. Table S2. Concentrations of 54 metabolites in 28-day-old bermudagrass plants under control conditions and different treatments [20 μM melatonin, 300mM NaCl, drought, and cold (4 °C)] stress conditions for 14 d. Table S3. Summary of RNA-Seq data. Table S4. List of up-regulated genes in melatonin-treated bermudagrass plants. Table S5. List of down-regulated genes in melatonin-treated bermudagrass plants. Supplementary Data