Knockout of Arabidopsis Serotonin N-Acetyltransferase-2 Reduces Melatonin Levels and Delays Flowering

Melatonin plays roles in both plant growth and defense. Serotonin N-acetyltransferase (SNAT) catalyzes formation of N-acetylserotonin (NAS) from serotonin. Plants contain two SNAT isogenes, which exhibit low-level amino acid homology. We studied the Arabidopsis thaliana SNAT2 (AtSNAT2) gene; we prepared recombinant SNAT2 protein and characterized a snat2 knockout mutant. The SNAT2 protein exhibited 27% amino acid homology with SNAT1; the Km was 232 μM and the Vmax was 2160 pmol/min/mg protein. Melatonin inhibited SNAT enzyme activity in vitro. SNAT2 mRNA was abundantly expressed in flowers; the melatonin content of flowers of the snat2 mutant was significantly less than that of wild-type flowers. The mutant exhibited delayed flowering and reductions in leaf area and biomass compared to the wild type. Delayed flowering was attributable to reductions in the expression levels of the gibberellin biosynthetic genes ent-kaurene synthase (KS) and FLOWERING LOCUS T (FT).

All plant melatonin biosynthetic genes have been cloned [25]. The genes encoding the last two steps in melatonin biosynthesis are serotonin N-acetyltransferase (SNAT) and N-acetylserotonin O-methyltransferase (ASMT). Plants contain two SNAT isogenes [25]; the rice genome contains three copies of ASMT [26]. SNAT plays important roles in both melatonin function and biosynthesis. Suppression of rice SNAT2 decreased levels of brassinosteroids (plant hormones) and induced a dwarf phenotype, but, suppression of rice SNAT1 did not result in such effects, although melatonin levels were similarly decreased [10]. The in vitro enzyme kinetics of rice SNAT1 and SNAT2 differed [25]. The SNAT1 genes of various plants (and cyanobacteria) have been cloned and characterized [27][28][29], but SNAT2 has been cloned only from rice [30] and grapevine [31].
Although many physiological functions played by melatonin have been reported, there has not been any molecular genetic study on the effect of flowering. Plant flowering is initiated by physical factors such as the photoperiod and vernalization, and by gibberellic acid (GA; a hormone; [32]).
The key photoperiod-regulated flowering gene is the night-labile CONSTANT (CO), a zinc finger protein that activates FLOWERING LOCUS T (FT), which in turn promotes floral induction. Vernalization accelerates flowering by decreasing the expression of FLOWERING LOCUS C (FLC), which encodes a MADS-box transcription factor that represses FT. In addition to the above physical factors, GA also directly or indirectly controls FT expression.
Here, we cloned Arabidopsis thaliana SNAT2 and explored the kinetics of the recombinant protein. We investigated an SNAT2 knockout mutant (snat2) in which decreased melatonin levels delayed flowering and retarded growth.

Plant Materials and Growth
Arabidopsis thaliana wild-type Col-0 and the knockout snat2 mutant line (SALK_062388) were grown at 22 • C under 60% relative humidity, a 14 h photoperiod, and a photon flux density of 50 µmol m −2 s −1 . The Arabidopsis snat2 line (with a T-DNA insert in SALK_062388) was obtained from the Arabidopsis Biological Resource Center (Ohio State University, Columbus, OH, USA). The line was genotyped using the T-DNA express: Arabidopsis gene mapping tool (http://signal.salk.edu/). Tobacco plants (Nicotiana benthamiana) were grown under a 16 h light (28 • C)/8 h dark (22 • C) photoperiod featuring white light (100 µmol photons m −2 s −1 ). Flowering times were defined by the days to bolting and the numbers of rosette leaves at bolting (n = 20). Leaf area was measured using Fuji ImageJ software [33]. Rosette leaves were photographed adjacent to a ruler; pixels were then converted to metric units. All data were processed with the aid of Microsoft Excel 2010. All experiments were performed in triplicate and were repeated at least three times.

Homology and Phylogenetic Analysis of AtSNAT2
Amino acid sequence homology was explored with the aid of the BLASTp tool; we interrogated the nonredundant, protein sequence databases of the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/). BLAST-Explorer was used to generate sequence alignments and phylogenetic trees [34].

Cloning of AtSNAT2 and Purification of Recombinant AtSNAT2 Protein from Escherichia Coli
Total RNA of 3-week-old Arabidopsis leaves was isolated using a NucleoSpin RNA plant kit (Macherey-Nagel, Duren, Germany) and 1 µg thereof was reverse-transcribed using RevertAid transcriptase (Thermo Scientific Fermentas, St. Leon-Ro, Germany) after addition of 500 ng of oligo(dT)18 primer, at 42 • C for 1 h. The cDNA obtained (0.2 µL) served as the template for PCR-based cloning. Reverse transcription (RT)-PCR was employed to clone full-length AtSNAT2 (GenBank accession number BT005218) using the primer pair 5 -AAA AAG CAG GCT CCA TGT TTC TCG GAG GCA-3 and 5 -AGA AAG CTG GGT TTA TTT CTT GTT TCT CTG-3 . The initial AtSNAT2 PCR product was cloned into the T&A vector (RBC Bioscience, New Taipei City, Taiwan) and the integrity thereof confirmed via sequencing (Bioneer, Daejeon, Korea). Prior to AtSNAT2 expression and purification, the initial RT-PCR product was amplified using a primer pair containing attB recombination sequences (forward primer, 5 -GGG GAC AAG TTT GTA CAA AAA AGC AGG CT-3 ; reverse primer, 5 -GGG GAC CAC TTT GTA CAA GAA AGC TGG GT-3 ). The resulting AtSNAT2 PCR product was cloned via the Gateway recombination reaction into the pDONR221 vector (Invitrogen, Carlsbad, CA, USA), and then into the destination vector pET300/NT-DEST (Invitrogen). The pET300-AtSNAT2 plasmid was transformed into E. coli strain BL21(DE3) (Invitrogen). An aliquot (10 mL) of an overnight culture grown in the presence of ampicillin (50 mg/L) was inoculated into 100 mL of Terrific Broth [20 g/L Bacto-tryptone, 24 g/L Bacto-yeast extract, glycerol 4 mL/L, and phosphate buffer (0.017 M KH 2 PO 4 and 0.072 M K 2 HPO 4 )] containing 50 mg/L ampicillin and incubated at 37°C until the optical density at 600 nm attained 1.0. After addition of isopropyl-β-D-thiogalactopyranoside (IPTG; Sigma, St. Louis, MO, USA) to 1 mM, the culture was grown at 28°C with shaking at 180 rpm for 5 h. The protein was purified via affinity (Ni 2+ ) chromatography according to the recommendations of the kit manufacturer (Qiagen, Tokyo, Japan).

AtSNAT Enzyme Kinetics
Purified recombinant AtSNAT2 protein was incubated in a final volume of 100 µL with 0.5 mM serotonin or 5-methoxytryptamine (5-MT), and 0.5 mM acetyl-CoA, in 100 mM potassium phosphate (at pH 7.8 or other pH values) at 45 • C (or other temperatures) for 1 h. Reaction products including N-acetylserotonin (NAS) and melatonin were identified with the aid of high-performance liquid chromatography (HPLC). In brief, each reaction was terminated by addition of 20 µL acetic acid and 30 µL methanol and 10 µL aliquots were subjected to reverse-phase HPLC (Waters, Milford, MA, USA) on a SunFire C18 column (Waters; 4.6 × 150 mm) eluted using an isocratic profile of 6% (v/v) methanol for NAS or 10% (v/v) methanol (both in water with 0.3% [v/v] trifluoroacetic acid) at a flow rate of 1 mL/min. Compounds were detected via absorbance at 280 nm. The substrate affinity (K m ) and maximum reaction rate (V max ) were calculated using Lineweaver-Burk plots. Protein concentrations were determined using the Bradford method employing a protein-specific dye (Bio-Rad, Hercules, CA, USA). All analyses were performed in triplicate.

Subcellular Localization of AtSNAT2
We used the pER-mCherry vector [the kind gift of Dr. H.G. Kang (Texas State University, San Marcos, TX, USA)] to localize the AtSNAT2 protein. Full-length AtSNAT2 was PCR-amplified using a primer set containing AscI sites (GGCGCGCC). The resulting product was cloned into the T&A vector (RBC Bioscience) and the AtSNAT2 insert was digested with AscI and cloned into the AscI site of the binary vector, pER8-mCherry, containing the estrogen-inducible XVE promoter. The pER8-AtSNAT2-mCherry plasmid was transferred into Agrobacterium tumefaciens strain GV2260 using the freeze-thaw method. A confocal microscope (TCS-SP5; Leica, Wetzlar, Germany) was used to evaluate transient within-cell protein expression, as described previously [35].

Total RNA Isolation and Reverse Transcription/Polymerase Chain Reaction
Total plant RNA was isolated using a NucleoSpin RNA plant kit (Macherey-Nagel, Duren, Germany). Semi-quantitative reverse transcription (RT)-PCR analyses were performed as described previously [21]. The primer sequences used for PCR amplification were as follows: SNAT1 (forward

Melatonin Quantification
Arabidopsis leaves (0.2 g) were ground in liquid nitrogen and thoroughly homogenized in 2 mL chloroform. After centrifugation at 12,000 × g for 10 min, the supernatants were decanted and evaporated to dryness, and the residues were dissolved in 1 mL amounts of methanol; 10 µL aliquots were used for melatonin quantitation via liquid chromatography/dual mass spectrometry (LC-MS/MS; Shimadzu, Kyoto, Japan) as described previously [36]. A mass transition from m/z 233.1 to m/z 174.0 identified melatonin; the retention time was 3.65 min. AB Sciex Analyst software (version 1.5.2) was used for data integration.

Data Analysis
We used the Student's t-test of SigmaPlot ver. 10 (Systat Software, Point Richmond, CA, USA). p < 0.05 was considered to indicate statistical significance.

Cloning and Phylogeny of AtSNAT2
SNAT is the penultimate enzyme of the melatonin biosynthetic pathway; plants contain two SNAT isogenes [30]. Plant SNAT1 genes have been relatively well studied [25], but the SNAT2 genes of only rice [10] and grapevine [31] have been evaluated. No plant SNAT2 knockout mutant has yet been investigated. Here, we cloned the rice SNAT2 homolog of Arabidopsis (AtSNAT2). The amino acid sequences of the rice SNAT2 and AtSNAT2 were 48% identical ( Figure 1A). Plant SNAT1 and SNAT2 share an amino acid identity level <27%. The two AtSNAT isogenes belong to different clades, implying that they arose via independent evolution rather than gene duplication ( Figure 1B).

Subcellular Location of AtSNAT2
To evaluate whether AtSNAT2 is also expressed in chloroplasts, we transiently expressed the AtSNAT2-mCherry fusion protein in tobacco leaves. Figure 2 shows that the protein localized to chloroplasts; the chlorophyll and mCherry fluorescences overlapped precisely. Thus, AtSNAT2 is expressed in chloroplasts, as are other plant SNAT proteins [30,31,35].

AtSNAT2 Enzyme Kinetics
We expressed two AtSNAT2 gene constructs in E. coli; the chloroplast-targeting sequence inhibits heterologous protein expression in that bacterium [37]. We thus expressed both full-length AtSNAT2 and an N-terminal-37-amino acid-truncated protein. TargetP analysis predicted the presence of an N-terminal 37-amino acid transit sequence [38]. We purified both recombinant proteins via Ni 2+ affinity chromatography; both featured N-terminal hexa-histidine fusions ( Figure 3). IPTG enhanced the expression levels of both proteins, but the extent of inducibility and the expression levels of the N-terminal-truncated ∆37AtSNAT2 were greater than those of the full-length protein. The pH and temperature optima in terms of enzymic activity were determined. Figure 4 shows that the total SNAT activity of the full-length protein was greater than that of the truncated protein under all conditions tested. The optimal pH values in terms of activity were pH 7.8 for both proteins, followed by pH 8.8 ( Figure 4A). SNAT enzyme activity was 10-fold lower in the pH range pH 5.4-6.5 than at pH 7.8. Optimal SNAT enzyme activity at relatively high pH values is characteristic of other plant SNAT2 enzymes [30,31]. The optimal temperatures for activity were 45°C and 55°C for AtSNAT2, but only 45°C for ∆37AtSNAT2. Surprisingly, AtSNAT2 retained 30% of its peak activity even at 95°C; ∆37AtSNAT2 exhibited no activity at that temperature ( Figure 4B). Neither the truncated rice nor grape enzyme exhibited any activity at 95°C [30,31]. In contrast, rice ∆83SNAT1 exhibited enzyme activity comparable to that of AtSNAT2 at 95°C [35], implying that plant SNAT enzymes are more thermostable than other melatonin biosynthetic enzymes such as ASMT [35]. In assays performed at the optimal pH and temperature, we next determined K m and V max values ( Figure 5). The K m values for NAS and 5-MT were 232 and 630 µM, respectively. The V max values for NAS and 5-MT were 2160 and 3360 pmol/min/mg protein, respectively. The catalytic efficiency (V max /K m ) for NAS was 9.3 and that for 5-MT was 5.3, indicating that AtSNAT favors NAS synthesis from serotonin rather than melatonin synthesis from 5-MT. The effects of 5-MT on enzyme activity were assessed by measuring the conversion of serotonin to NAS in the presence of 5-MT. Figure 6 shows that NAS production (from 500 µM serotonin) was inhibited by the addition of 50 µM 5-MT, and further fell fourfold when 100 µM of 5-MT was added. The decrease in NAS production was not proportional to the 5-MT concentration. Notably, the NAS and melatonin production levels were similar when the substrate levels were 500 µM, implying that the affinity of SNAT for serotonin and 5-MT is similar when substrate levels are high.  We explored whether AtSNAT1 [28] and AtSNAT2 activity was feedback-regulated by melatonin. Recombinant AtSNAT2 activity was inhibited by addition of 1 µM melatonin to an extent twofold that of the control (Figure 7), but AtSNAT1 activity was not affected by any tested concentration of melatonin, implying that the modes of action of the SNAT isoenzymes differ. . AtSNAT1 was prepared as described previously [28]. The SNAT enzyme assay was performed in the presence of 0.5 µg AtSNAT2, 0.5 mM serotonin, and various levels of melatonin. The data are the means of three replicates. Asterisks (*) denote significant differences compared to the Col-0 controls (Student's t-test; p < 0.05).

AtSNAT2 Expression and the Phenotype of the snat2 Knockout Arabidopsis Mutant
To explore directly whether AtSNAT2 engages in melatonin synthesis and the physiological role played by melatonin, we investigated a T-DNA insertional snat2 knockout mutant (SALK_062388) ( Figure 8A). We first measured SNAT1 and SNAT2 transcript levels in various tissues via RT-PCR analysis. SNAT1 mRNA levels were highest in mature leaves; SNAT2 mRNA was abundant in flowers and young leaves; the two isogenes thus exhibited different expression patterns ( Figure 8B). RT-PCR of material from one-week-old leaves showed that SNAT2 mRNA was absent in the mutant ( Figure 8C). Melatonin levels were 50% lower in flowers of the snat2 mutant than in those of the wild-type Col-0, but the melatonin levels of leaves did not differ, implying that SNAT2 expression is tissue-specific in terms of both mRNA and melatonin synthesis ( Figure 8D). In line with this implication, the snat2 mutant exhibited delayed flowering, as evidenced by increased leaf numbers (Figure 9A-C). During bolting, the height of the snat2 mutant was less than that of the wild type, but the mutant attained the wild-type height eventually ( Figure 9D). The leaf area of the snat2 mutant was less than that of the wild type, reducing the biomass of the mutant (Figure 9E,F). The numbers in parentheses are the PCR cycle numbers. EF-1a served as a loading control, as described previously [4]. An asterisk (*) indicates a significant difference compared to the wild-type Col-0 (Student's t-test; p < 0.05). To define the flowering pathway affected by the snat2 mutation, we used RT-PCR to measure gene expression levels. As shown in Figure 10, the expression levels of photoperiod pathway-related flowering genes, including Circadian Clock Associated 1 (CCA1) and Timing of CAB expression 1 (TOC1), were similar in the mutant and wild type over the 24 h day/night cycle ( Figure 10A). However, a key gene involved in GA biosynthesis, ent-kaurene synthase (KS), was significantly suppressed, whereas GA-responsive negative transcription factor MYB33 was enhanced in the mutant compared to the wild type, implying that changes in the GA-related pathway (and not the photoperiod pathway) explained the delayed flowering of the snat2 mutant. Reductions in GA synthesis and enhanced MYB33 expression decreased levels of FT florigen, delaying flowering ( Figure 10C). We explored whether flowering-related gene expression was affected by exogenous melatonin (1 µM) that infiltrated leaves over 2 h. As shown in Figure 10B, the expression levels of melatonin-induced representative marker genes such as BIP2 and PR1 increased markedly [4], but melatonin suppressed expression of the KS, MYB33, and FT genes, consistent with a previous finding that exogenous melatonin delays flowering [39]. KS, MYB33, and FT suppression by exogenous melatonin was in sharp contrast with such suppression in the snat2 mutant, indicating that exogenous melatonin did not mimic the effects of endogenous melatonin on flowering.

Discussion
Given the many available Arabidopsis T-DNA knockout mutants, it is simple to explore the roles played by genes of interest [40], including those involved in melatonin metabolism. Melatonin is biosynthesized from tryptophan in four steps catalyzed by tryptophan decarboxylase (TDC), tryptamine 5-hydroxylase (T5H), SNAT, and ASMT [25]. The SNAT-catalyzed reaction is the penultimate step; the SNAT gene family plays pivotal roles in terms of melatonin synthesis regulation in both plants and animals [41]. Rice SNAT1 was first cloned in 2013 [37], followed by SNAT2 [30]; orthologs exist in all plant taxa explored [27,30]. Both SNAT1 and SNAT2 are nucleus-encoded chloroplast proteins involved in melatonin synthesis. It is known that all plant SNAT proteins explored to date were located in chloroplasts [27,30,31,35]. However, single transgenic rice plants differ in terms of phenotype and stress response. Rice plants overexpressing SNAT1 are resistant to abiotic stressors such as senescence [21], while SNAT1 suppression increased the susceptibility of rice plants to cold [42]. However, SNAT2 suppression conferred tolerance to several abiotic stressors, including both cold and heat [43], and was associated with a novel dwarf-like phenotype combined with erect leaves [10]. Thus, the two SNAT isogenes mediate different physiological responses in terms of melatonin synthesis in plants.
Recently, the grape SNAT2 gene was cloned and characterized [31]. The grapevine SNAT2 is 55% identical (in terms of their amino acid sequences) to rice SNAT2. The optimal pH and temperature for grape and rice SNAT2 activity are similar, but the catalytic efficiency (V max /K m ) of the grape enzyme is less than that of the rice enzyme [30]. Transgenic Arabidopsis overexpressing grape SNAT2 exhibit enhanced melatonin synthesis and increased resistance to powdery mildew (compared to the wild type) [31]. However, no knockout mutant of grape SNAT2 has yet been reported. Our Arabidopsis SNAT2 knockout mutant (snat2) clearly exhibited delayed flowering (Figure 9), as confirmed via molecular genetic analysis. Thus, melatonin affects plant flowering to some extent, but does not delay flowering markedly.
In contrast to the effects of the snat2 mutation, exogenous melatonin retarded Arabidopsis flowering by elevating the level of DELLA protein, the master regulator of GA production, in turn suppressing the expression of flowering genes, including SUPPRESSOR OF OVEREXPRESSION OF CONSTANT 1 (SOC1) and FLOWERING LOCUS T (FT) [39]. These contradictory results are explained by the fact that high-dose melatonin (0.5 mM) retards the growth of Arabidopsis seedlings. The melatonin biosynthetic mutant snat1 also exhibited delayed flowering (data not shown). We are currently studying the snat1 mutant in detail.
We found, for the first time, that SNAT2 was involved in melatonin synthesis in Arabidopsis and was abundantly expressed in flowers. The snat2 mutant produced less melatonin than did the wild type, in turn affecting GA biosynthesis and FT expression, both of which regulate flowering. We suggest that the reduced melatonin levels are associated with less GA synthesis, and compromise FT expression, in turn inhibiting flowering. The mechanism by which melatonin regulates GA synthesis remains to be investigated.

Conclusions
Melatonin plays roles in both plant growth and defense. Serotonin N-acetyltransferase (SNAT) catalyzes the formation of N-acetylserotonin (NAS) from serotonin; NAS is then converted to melatonin. Plants contain two SNAT isogenes, which exhibit low-level amino acid homology. SNAT is the penultimate enzyme of melatonin biosynthesis. This is the first study to characterize a snat2 mutant genetically. The mutant exhibited delayed flowering; this is the first genetic indication that melatonin is involved in flowering.

Conflicts of Interest:
The authors declare no conflict of interest.