Rice histone deacetylase 10 and Arabidopsis histone deacetylase 14 genes encode N-acetylserotonin deacetylase, which catalyzes conversion of N-acetylserotonin into serotonin, a reverse reaction for melatonin biosynthesis in plants

Article abstract of DOI:10.1111/jpi.12460

In plants, melatonin production is strictly regulated, unlike the production of its precursor, serotonin, which is highly inducible in response to stimuli, such as senescence and pathogen exposure. Exogenous serotonin treatment does not greatly induce the production of N-acetylserotonin (NAS) and melatonin in plants, which suggests the possible existence of one or more regulatory genes in the pathway for the biosynthesis of melatonin from serotonin. In this report, we found that NAS was rapidly and abundantly converted into serotonin in rice seedlings, indicating the presence of an N-acetylserotonin deacetylase (ASDAC). To clone the putative ASDAC gene, we screened 4 genes that were known as histone deacetylase (HDAC) genes, but encoded proteins targeted into chloroplasts or mitochondria rather than nuclei. Of 4 recombinant Escherichia coli strains expressing these genes, one E.?coli strain expressing the rice HDAC10 gene was found to be capable of producing serotonin in response to treatment with NAS. The recombinant purified rice HDAC10 (OsHDAC10) protein exhibited ASDAC enzyme activity toward NAS, N-acetyltyramine (NAT), N-acetyltryptamine, and melatonin, with the highest ASDAC activity for NAT. In addition, its Arabidopsis ortholog, AtHDAC14, showed similar ASDAC activity to that of OsHDAC10. Both OsHDAC10 and AtHDAC14 were found to be expressed in chloroplasts. Phylogenetic analysis indicated that ASDAC homologs were present in archaea, but not in cyanobacteria, which differs from the distribution of serotonin N-acetyltransferase (SNAT). This suggests that SNAT and ASDAC may have evolved differently from ancestral eukaryotic cells.

Full text of DOI:10.1111/jpi.12460

Rice histone deacetylase 10 and Arabidopsis histone deacetylase 14 genes
encode N-acetylserotonin deacetylase, which catalyzes conversion of N-
acetylserotonin into serotonin, a reverse reaction for melatonin biosynthesis
in plants
Kyungjin Lee | Hyoung Yool Lee | Kyoungwhan Back
Department of Biotechnology, Bioenergy Research Center, Chonnam National University,
Gwangju, Republic of Korea
Running title: N-acetylserotonin deacetylase genes
Corresponding address:
Kyoungwhan Back
Department of Biotechnology, Bioenergy Research Center, College of Agriculture and Life
Sciences, Chonnam National University, Gwangju 500-757, South Korea.
E-mail : kback@chonnam.ac.kr
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/jpi.12460
This article is protected by copyright. All rights reserved.

Keywords:
Arabidopsis, melatonin deacetylase, N-acetyltryptamine deacetylase, rice
Abstract: In plants, melatonin production is strictly regulated, unlike the production of its
precursor, serotonin, which is highly inducible in response to stimuli, such as senescence and
pathogen exposure. Exogenous serotonin treatment does not greatly induce the production of
N-acetylserotonin (NAS) and melatonin in plants, which suggests the possible existence of
one or more regulatory genes in the pathway for the biosynthesis of melatonin from serotonin.
In this report, we found that NAS was rapidly and abundantly converted into serotonin in rice
seedlings, indicating the presence of an N-acetylserotonin deacetylase (ASDAC). To clone
the putative ASDAC gene, we screened four genes that were known as histone deacetylase
(HDAC) genes, but encoded proteins targeted into chloroplasts or mitochondria rather than
nuclei. Of four recombinant Escherichia coli strains expressing these genes, one E. coli strain
expressing the rice HDAC10 gene was found to be capable of producing serotonin in
response to treatment with NAS. The recombinant purified rice HDAC10 (OsHDAC10)
protein exhibited ASDAC enzyme activity toward NAS, N-acetyltyramine (NAT), N-
acetyltryptamine, and melatonin, with the highest ASDAC activity for NAT. In addition, its
Arabidopsis ortholog, AtHDAC14, showed similar ASDAC activity to that of OsHDAC10.
Both OsHDAC10 and AtHDAC14 were found to be expressed in chloroplasts. Phylogenetic
analysis indicated that ASDAC homologs were present in archaea, but not in cyanobacteria,
This article is protected by copyright. All rights reserved.

which differs from the distribution of serotonin N-acetyltransferase (SNAT). This suggests
that SNAT and ASDAC may have evolved differently from ancestral eukaryotic cells.
1
| INTRODUCTION
Melatonin is a universal molecule found in almost all organisms at levels that vary depending
1,2
on tissue localization and diurnal changes. Melatonin also plays a wide variety of
physiological and ecological roles in animals and plants, including involvement in circadian
3
,4
rhythms, immune function, DNA repair, and defense against various stresses.
Studies on the biosynthetic pathway and physiological roles of melatonin in plants
5
have advanced rapidly, as melatonin was first identified only 20 years ago. For example,
melatonin functions in plants as a biostimulator, orchestrating an array of beneficial effects
against biotic and abiotic stresses, including exposure to cold, senescence, cadmium, nitrate,
6
-15
16-19
drought, and pathogens, via the well-known H O /NO signaling pathways.
However,
2
2
2
0
the regulatory mechanisms of melatonin biosynthesis remain elusive. In common with
animals, melatonin biosynthesis in plants requires four enzyme-mediated steps. First,
tryptophan is decarboxylated into tryptamine by tryptophan decarboxylase (TDC), which is
2
1,22
followed by serotonin synthesis via the action of tryptamine 5-hydroxylase (T5H).
The
penultimate enzyme is serotonin N-acetyltransferase (SNAT), which catalyzes the conversion
of serotonin into N-acetylserotonin (NAS). SNAT enzymes are encoded by a small gene
This article is protected by copyright. All rights reserved.

23-26
family found in both chloroplasts and mitochondria.
The last enzyme in the pathway is N-
acetylserotonin O-methyltransferase (ASMT), which converts NAS into melatonin via the
action of multiple distantly-related O-methyltransferases, including caffeic acid O-
27-30
methyltransferase (COMT).
These ASMT-encoding genes are expressed in either
cytoplasm or chloroplasts, resulting in melatonin synthesis in different subcellular sites of the
24,30
plant cells.
Exogenous serotonin treatment was not found to significantly increase NAS and
melatonin levels in transgenic rice overexpressing an animal SNAT gene or non-transgenic
31
wild-type rice expressing an endogenous SNAT gene constitutively. This result was
consistent with subsequent reports that the increase in serotonin levels as a response to
senescence or cadmium treatment was not proportionate to the increase in melatonin
32,33
levels.
These results indicate that the conversion of serotonin into NAS is a bottleneck
step in melatonin biosynthesis in plants, but the reason for the lack of NAS accumulation has
not been elucidated.
In this report, we demonstrate for the first time that NAS is rapidly converted into
serotonin upon exogenous NAS treatment in rice seedlings. We cloned the genes encoding
NAS deacetylase (ASDAC) enzymes, characterized them in rice and Arabidopsis, and
assessed their subcellular localization to chloroplasts. Based on enzyme functional analysis, it
is clear that ASDAC plays a pivotal role in regulating melatonin biosynthesis in plants.
This article is protected by copyright. All rights reserved.

2
2
| METHODS
.1 | Vector construction and expression of HDAC genes in Escherichia coli
Four full-length rice histone deacetylase (HDAC) cDNAs and one Arabidopsis thaliana
HDAC14 cDNA were provided by the National Institute of Agrobiological Sciences (NIAS,
34,35
Tsukuba, Japan).
Full-length cDNA fragments were amplified by PCR using primer sets
with each forward primer containing 14-nucleotide (nt) attB1 sequences (5'-AAA AAG CAG
GCT CC-3') and each reverse primer containing a stop codon and 12-nt attB2 sequences (5'-
2
3
AGA AAG CTG GGT-3'), as described previously. Because rice (Oryza sativa) HDAC10
and OsSIR2b were predicted to possess chloroplast transit sequences of 36 and 23 amino
acids, vectors for the expression of these enzymes were designed to express the enzymes
without the predicted transit sequences. The resulting HDAC cDNA fragments were further
amplified by a second PCR using primers harboring the attB recombination sequence
(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 second PCR
products were gel-purified and cloned into the pDONR221 Gateway vector (Invitrogen,
Carlsbad, CA, USA) via BP recombination at 25°C for 1 hour, followed by protease
treatment at 37°C for 5 min. The recombinant pDONR221 plasmids harboring various inserts
of HDAC genes were recombined with the pET300 Gateway Destination Vector (Invitrogen)
via the LR Recombination Reaction to form pET300-HDAC plasmid vectors, which were
This article is protected by copyright. All rights reserved.

transformed into BL21 (DE3) competent E. coli cells. Escherichia coli strains harboring
various pET300-HDAC plasmids were incubated in the presence of possible substrates (1
mmol/L), such as NAS and isopropyl-β-D-thiogalactopyranoside (IPTG; 1 mmol/L; Sigma)
for 16 hours at 28°C. These cultures were spun down and supernatants were subjected to high
performance liquid chromatography (HPLC) analyses.
2
.2 | Chemicals
Melatonin (N-acetyl-5-methoxytryptamine), NAS, N-acetyltryptamine, N-acetyltyramine, 5-
methoxytryptamine, serotonin, tryptamine, and tyramine were purchased from Sigma-Aldrich
(St. Louis, MO, USA). All reagents were dissolved in ethanol and further diluted into water
for analyses.
2.3 | Affinity purification of OsHDAC10 and AtHDAC14 proteins
Aliquots (10 mL) of overnight cultures of E. coli strains containing pET300-OsHDAC10 and
pET300-AtHDAC14 were inoculated into 100 mL of YEP medium (10 g/L Bacto peptone,
1
0 g/L Bacto yeast extract, 5 g/L NaCl) containing 50 mg/L ampicillin and incubated at 37°C
until the optical density of the E. coli culture at 600 nm (OD ) reached 1.0. After the
600
addition of 1 mmol/L IPTG (Sigma), the cultures were shaken for 5 h at 28°C at 250 rpm.
The cultures were centrifuged at 13 500 × g for 5 min at 4°C, and the cell pellets were used in
This article is protected by copyright. All rights reserved.

2
+
the protein purification steps. The purifications were carried out by affinity (Ni )
chromatography according to the manufacturer’s instructions (Qiagen, Tokyo, Japan).
2.4 | HPLC quantification of amines
Supernatants from each of the E. coli cultures were mixed with equal volumes of methanol,
and 20-µL aliquots of the solutions were subjected to HPLC using a fluorescence detector
system (Waters, Milford, MA, USA). In brief, the samples were separated on a Sunfire C18
column (Waters; 4.6 × 150 mm) using isocratic elution with 6% methanol in 0.3%
trifluoroacetic acid for 50 min at a flow rate of 1 mL/min. Serotonin, tyramine, tryptamine,
and 5-metrhoxytryptamine were detected at 280 nm (excitation) and 348 nm (emission). All
measurements were taken in triplicate.
2.5 | Subcellular localization of OsHDAC10 and AtHDAC14 proteins
The pER-mCherry vector, kindly donated by Dr. H.G. Kang (Texas State University, San
Marcos, TX, USA), was used for subcellular localization of two HDAC proteins including
OsHDAC10 (AK072557) and AtHDAC14 (At4g33470). Two full-length HDAC cDNAs
were amplified by PCR using a primer set containing AscI sites (GGCGCGCC). The resulting
HDAC PCR products were cloned into the TA vector (RBC Bioscience, New Taipei City,
Taiwan), and each HDAC insert was then digested using AscI and cloned into the AscI site of
This article is protected by copyright. All rights reserved.

the binary vector pER8-mCherry containing the estrogen-inducible XVE promoter (Pxve) to
create pER8-HDAC-mCherry. The pER8-HDAC-mCherry plasmids were transferred into the
Agrobacterium tumefaciens strain GV2260 using the freeze-thaw method, and confocal
microscope (TCS-SP5; Leica, Wetzlar, Germany) transient expression analyses were
3
6,37
performed as described previously.
2.6 | Total RNA isolation and reverse transcription–polymerase chain reaction (RT-
PCR) analysis
Detached leaves of 4-week-old rice plants were treated with 0.2 mmol/L cadmium for various
lengths of time. Pulverized powders from the leaves (100 mg) were subjected to total RNA
extraction using the RNeasy Plant Mini Kit (Qiagen). The resulting total RNA (1 µg) was
reverse-transcribed using RevertAid reverse transcriptase (Thermo Scientific Fermentas, St.
Leon-Ro, Germany) and 500 ng of oligo(dT)18 primer at 42°C for 1 hour. The acquired
cDNA (0.2 µL) was used as the template for PCR amplification. The rice ubiquitin-5 gene
(UBQ5) served as a loading control. The primers (5'→ 3') used in the PCR reactions were as
follows: SNAT1 forward, 5'-GGG CTG CGG CAA CTT GGT CC-3' and SNAT1 reverse, 5'-
AGAAAG CTC GGT CTAAAA TCT GGG GTA-3'; HDAC10 forward, 5'-ACT AGT ATG
GAA CAG CTG TGG GTG-3' and HDAC10 reverse, 5'-GAG CTC ACC ACG ATG CTT
CGAAGT-3'; UBQ5 forward, 5'-CCG ACT ACAACA TCC AGAAGG AG-3' and UBQ5
reverse, 5'-AAC AGG AGC CTA CGC CTAAGC-3'. RT-PCR-amplified fragments were
This article is protected by copyright. All rights reserved.

electrophoresed on ethidium bromide gels and photographed under ultraviolet (UV) light.
2.7 | Measurement of in vitro HDAC enzyme activity in rice seedlings
Seven-day-old rice seedlings were dissected into leaves and root parts. These samples (0.1 g)
were ground in 50 mmol/L Tris-HCl buffer (pH 8.0) containing 10 mmol/L β-
mercaptoethanol. The homogenates were centrifuged for 15 min at 13,000 × g and 4°C. The
supernatants were used as a crude enzyme solution, 30 µL of which was used to assay
ASDAC enzyme activity in a total volume of 100 µL of 50 mmol/L Tris-HCl buffer (pH 8.0)
containing 1mmol/L NAS at 30°C for 90 min. The reaction was stopped by adding 20 µL of
acetic acid. After brief centrifugation, a 20-µL aliquot was subjected to HPLC as described
above. All measurements were conducted in triplicate. Protein concentrations were
determined by the Bradford method using a protein assay dye (Bio-Rad Laboratories, Inc.,
Hercules, CA, USA).
2.8 | Enzyme kinetics and phylogenetic analyses
The substrate affinity (K ) and maximum reaction rate (Vmax) were calculated from
m
Lineweaver-Burk plots based on crude E. coli extracts containing the rice or Arabidopsis
ASDAC proteins. The crude E. coli extracts (10 µL) or the purified recombinant proteins
were used to assay ASDAC enzyme activity in a total volume of 100 µL of 50 mmol/L Tris-
This article is protected by copyright. All rights reserved.

HCl buffer (pH 8.0) containing 1 mmol/L NAS (or other substrates) at 30°C for 90 min. The
reaction was stopped by adding 20 µL of acetic acid. After brief centrifugation, a 20-µL
aliquot was subjected to HPLC as described above. The network-based method (TargetP) was
used to identify chloroplast transit peptides and their cleavage sites from various N-terminal
38
sequences of ASDAC homologs. A phylogenetic tree was generated using BLAST-
39
Explorer.
3
3
| RESULTS
.1 | Substrate feeding analyses in rice seedlings
To measure the in vivo substrate conversion rate of the melatonin biosynthetic pathway
Figure 1A), roots of 7-day-old rice seedlings were treated with various melatonin
(
biosynthetic substrates for 24 hours. As shown in Figure 1B, NAS treatment (1 mmol/L) led
to serotonin production of 1514 ng/g fresh weight (FW), whereas melatonin production was
as low as 35 ng/g FW. These data suggested that the conversion rate of NAS into melatonin
by ASMT was much slower than that of NAS into serotonin, which was strongly indicative of
the presence of ASDAC. By comparison, the NAS conversion resulting from treatment with 1
mmol/L serotonin was 75 ng/g FW, whereas serotonin production resulting from treatment
with 1 mmol/L tryptamine was as high as 8281 ng/g FW. Based on the substrate conversion
rate, it was evident that rice plants express ASDAC enzyme, which catalyzes the conversion
This article is protected by copyright. All rights reserved.

of NAS into serotonin, and that the ASDAC enzyme has much higher in vivo catalytic
activity than SNAT.
3.2 | De novo molecular cloning of rice ASDAC
40
Of the 16 HDAC genes in the rice genome, three encoded proteins that were localized to
sites other than the nucleus. OsHDAC6 was exclusively localized in chloroplasts, whereas
41
OsHDAC10 was observed in both chloroplasts and mitochondria. OsSIR2b, a rice ortholog
42
of yeast sirtuin 2, was detected in mitochondria, as was its human ortholog hSIRT3. Based
41
on the subcellular localization of these rice HDAC proteins and the fact that melatonin
20,26
biosynthesis occurs in the chloroplasts and mitochondria of plants,
we predicted that these
proteins might have activity to deacetylate NAS into serotonin. To analyze the function of
these HDAC genes, we first expressed the genes in E. coli and incubated the recombinant
strains in the presence of NAS to measure possible production of serotonin via ASDAC
activity. Additionally, OsSRT1, a rice ortholog of yeast sirtuin 2, was included as an E. coli
expression control because OsSRT1 protein is a histone deacetylase that is expressed in the
43
nucleus (Figure 2A). Among four transgenic E. coli strains, an E. coli strain harboring the
Δ36:OsHDAC10 construct was found to produce serotonin in response to treatment with 1
mmol/L NAS at a concentration of 0.6 µg/mL culture, whereas other E. coli strains, including
the strain expressing OsSRT1, were unable to produce serotonin upon treatment with NAS.
This article is protected by copyright. All rights reserved.

This suggested that OsHDAC10 encoded ASDAC enzyme activity. To further verify the
ASDAC activity of OsHDAC10, various substrates were incubated with the E. coli strains.
As shown in Figure 2C–E, the E. coli strain expressing OsHDAC10 produced tyramine (42
µg/mL), tryptamine (1.8 µg/mL), and 5-methoxytryptamine (0.4 µg/mL) when incubated
with N-acetyltyramine, N-acetyltryptamine, or melatonin, respectively. Our finding therefore
demonstrated that OsHDAC10 had a potent ASDAC enzyme activity catalyzing the
conversion of a number of N-acetylamines, such as N-acetyltyramine and NAS, as well as
melatonin.
3.3 | In vitro ASDAC enzyme activity of purified recombinant OsHDAC10
3
8
TargetP analysis predicted a possible chloroplast transit sequence of 36 amino acids.
However, the results from the amino acid comparison between OsHDAC10 and its
Arabidopsis ortholog AtHDAC14 suggested that the 66 N-terminal amino acids of
OsHDAC10 might encode the chloroplast transit sequence. Therefore, we generated three
constructs of OsHDAC10: full-length OsDHAC10 (FL:HDAC10), N-terminal 36-amino-
acid-truncated OsHDAC10 (Δ36:OsHDAC10), and N-terminal 66-amino-acid-truncated
OsHDAC10 (Δ66:OsHDAC10). The resulting recombinant HDAC10 enzymes were purified
by nickel affinity chromatography (Figure 3A–C), with the highest soluble expression found
in the Δ66:OsHDAC10-producing recombinant strain. Interestingly, both the purified
This article is protected by copyright. All rights reserved.

Δ36:OsHDAC10 and Δ66:OsHDAC10 proteins did not show any in vitro ASDAC enzyme
activity, but the ASDAC enzyme activity was recovered by the addition of E. coli BL21(DE3)
crude extract (Figure 3D). Upon the addition of boiled crude E. coli extracts, no ASDAC
activity was observed, which suggests that some heat labile substances or proteins found in E.
coli were necessary for in vitro ASDAC enzyme assay. The Δ66:OsHDAC10 protein
exhibited threefold higher ASDAC activity than that of the Δ36:OsHDAC10 protein, which
indicates that the chloroplast transit sequence is about 66 amino acids rather than 36 amino
acids (Figure 3E). Given these findings, our further analyses employed the Δ66:OsHDAC10
enzyme. When the substrate specificity of the enzyme was assessed using a number of N-
acetylamines, N-acetyltyramine showed the highest ASDAC activity (18 pkat/mg protein)
followed by NAS (5 pkat/mg protein) and N-acetyltryptamine (1.6 pkat/mg protein). We
observed that Δ66:OsHDAC10 was also able to deacetylate melatonin into 5-
methoxytryptamine (Figure 3F), indicating that rice HDAC10 exhibits affinity for multiple
20
substrates, as do the SNAT enzymes of plants.
3.4 | Optimal ASDAC enzyme activity of OsHDAC10
Because NAS is a direct substrate for melatonin biosynthesis via the ASMT enzyme, we
measured ASDAC activity in the presence of NAS as a substrate. As shown in Figure 4,
ASDAC enzyme activity increased as the concentration of Δ66:OsHDAC10 increased. The
This article is protected by copyright. All rights reserved.

ASDAC enzyme demonstrated high activity at a broad range of pH values (from 6.8 to 8.8).
o
The optimum temperature for the enzyme was 30 C, but moderate ASDAC enzyme activities
o
o
were also detected at temperatures as high as 45 C and 55 C. In the presence of sodium
butyrate, a potent histone deacetylase inhibitor, no ASDAC inhibition was observed. This
indicates that the ASDAC enzyme activity of OsHDAC10 is resistant to the histone
deacetylase inhibitor.
3.5 | Chloroplast localization of OsHDAC10
To examine the ASDAC enzyme activity in rice plants, we dissected seven-day-old rice
seedlings into root and shoot, and assayed for in vitro ASDAC enzyme activity using NAS as
a substrate. ASDAC enzyme activity was observed in both tissues, but higher activity levels
were observed in the root tissue than in the shoot tissue (Figure 5A). Transcript levels of
OsHDAC10 were monitored in four-week-old detached rice leaves challenged with 0.2
mmol/L cadmium. As shown in Figure 5B, HDAC10 mRNA was constitutively expressed at
low levels in the control, but expression levels declined slightly when assessed at 12 and 24
hours after cadmium treatment, returning to the basal level at 72 hours after treatment. When
assessing the subcellular localization of the enzyme in the Nicotiana benthamiana transient
expression system, OsHDAC10 protein was found to be expressed in chloroplasts (Figure
5C). Our results partially contrast with previous reports that OsHDAC10 is expressed in
This article is protected by copyright. All rights reserved.

41
mitochondria in tobacco BY2 cells and in chloroplasts in transgenic Arabidopsis. Since no
41
mitochondrial expression of OsHDAC10 was detected in transgenic Arabidopsis, this
targeting discrepancy may be attributable to the use of different host systems. Overall, when
our data are considered in view of the prior data, it is evident that OsHDAC10 is expressed in
chloroplasts.
3.6 | Enzymatic characterization of the OsHDAC10 ortholog Arabidopsis thaliana
HDAC14 (AtHDAC14),
The amino acid sequence of Arabidopsis thaliana HDAC14 (AtHDAC14) showed 70%
sequence identity to the sequence of OsHDAC10, which suggests that it may be a functional
ortholog of OsHDAC10. To examine whether AtHDAC14 possesses ASDAC activity, the
full-length AtHDAC14 (FL:AtHDAC14) and N-terminal 45-amino-acid-truncated
AtHDAC14 (Δ45:AtHDAC14) were expressed in E. coli and serotonin biosynthesis was
measured in the presence of 1 mmol/L NAS. As shown in Figure 6A, the E. coli expressing
Δ45:AtHDAC14 produced serotonin at a concentration of 22 µg/mL culture, whereas the
FL:AtHDAC14 yielded only 0.2 µg/mL culture, which suggests that AtHDAC14 possesses
ASDAC enzyme activity. The recombinant Δ45:AtHDAC14 protein was purified using a
nickel affinity system and assessed using the ASDAC enzyme assay (Figure 6B). AtHDAC14
showed an ASDAC-specific activity of 7.3 pkat/mg protein by the addition of E. coli crude
This article is protected by copyright. All rights reserved.

extracts when NAS was used as the substrate (Figure 6C). Similar to the findings for
OsHDAC10, AtHDAC14 had the highest ASDAC activity when NAT was used as the
substrate (30 pkat/mg protein), followed by the activity values obtained with NAS and N-
acetyltryptamine as substrates. Melatonin was also deacetylated into 5-methoxytryptamine by
AtHDAC14, with an ASDAC activity of 1.6 pkat/mg protein (Figure 6D). Confocal
microscopy also revealed that the AtHDAC14 protein was localized in chloroplasts, as the
mCherry fluorescence was found to be completely superimposed with the autofluorescence
signals of chlorophyll (Figure 7A–D).
The K and V values were measured using crude E. coli extracts obtained from
m
max
recombinant strains expressing the enzymes. Rice HDAC10 had a lower K value (409
m
µmol/L) than that of Arabidopsis HDAC14 (701 µmol/L) (Figure 7B), whereas the Vmax
values of the two enzymes were similar. The K and ASDAC-specific activity of OsHDAC10
m
2
3,24
were similar to those for rice SNAT1.
In summary, these data indicate that previously classified HDAC genes, such as
OsHDAC10 and AtHDAC14, encode enzymes with ASDAC activity. The functional
significance of ASDAC in vivo requires further investigation with respect to melatonin
synthesis in plants. The reversible acetylation and deacetylation of serotonin in plants may
play a pivotal regulatory role in maintaining the steady-state level of melatonin biosynthesis
in plants.
This article is protected by copyright. All rights reserved.

4
| Discussion
Deacetylase enzymes are classified into two types, depending on their substrates. One type
deacetylates organic compounds and the other, proteins. The best protein substrate of
deacetylase is histone proteins, which play a key role in modulating chromatin structure. The
activity of enzymes known as histone deacetylases (HDACs) leads to the tightness of
chromatin structure, which blocks the access of transcription factors and causes the
44
repression of corresponding genes. HDAC proteins have been identified in almost all
eukaryotes, including animals, plants, fungi, and archaebacteria, and regulate many functions
4
5,46
in transcription and the cell cycle.
HDAC proteins have also been shown to deacetylate
44
non-histone substrates, such as α-tubulin, heat shock proteins, and β-catenin. In contrast to
HDAC, many deacetylases have been reported that are responsible for deacetylation of
47
organic compounds. These include N-acetylglucosaminyl N-deacetylase, acetylpolyamine
48
49
50
amidohydrolase, polyamine deacetylase, and N-acetylornithine deacetylase.
4.1 | Significance of plant ASDAC genes
Phylogenic tree analysis of 123 HDACs from eukaryotic and prokaryotic organisms
identified five classes of the enzyme: class I is related to yeast RPD3, whereas classes II and
41
III are associated with yeast HDA1 and yeast SIR2, respectively. The annotated rice
genome was found to possess 16 HDAC genes spanning all five classes of HDACs, seven of
This article is protected by copyright. All rights reserved.

which belong to class I and one which belongs to class IV. Among the 16 rice HDAC proteins,
OsHDAC6 belongs to class IV, OsHDAC10 to class II, and OsSIR2b to class III. These three
OsHDAC proteins are targeted to chloroplasts, mitochondria, or both, rather than the
41
nucleus. The localization data suggested that these three OsHDAC proteins may possess
deacetylase activity toward organic compounds and/or other proteins, rather than histone. The
functional roles of these three OsHDACs had not previously been assessed, but AtHDA2A
and AtHDA14, the Arabidopsis orthologs of OsHDAC6 and OsHDAC10, were found to be
51,52
involved in gene silencing and tubulin deacetylation, respectively.
In this report, we uncovered for the first time that OsHDAC10 has a functional
ASDAC activity which catalyzes the conversion of NAS into serotonin, a reverse pathway for
melatonin biosynthesis (Figure 1). The ASDAC enzyme activity of 5 pkat/min/mg protein
23
was almost identical to that reported for SNAT. This reversible activity between the
production of serotonin and NAS may play an important role in regulating a steady-state level
of NAS in chloroplasts prior to the final catalysis by ASMT for the synthesis of melatonin.
Additionally, the OsHDAC10 protein showed ASDAC activity toward other related N-
acetylated amines, such as NAT, N-acetyltryptamine, and melatonin. The deacetylase activity
of rice ASDAC (OsHDAC10) for multiple substrates is consistent with the acetylase activity
25
of SNAT for multiple substrates.
This article is protected by copyright. All rights reserved.

To further characterize the ASDAC activity of OsHDAC10, we assessed the function
of AtHDAC14, an ortholog of OsHDAC10 (Figure 6). AtHDAC14 also exhibited ASDAC
enzyme activity similar to that of OsHDAC10, suggesting the existence of conserved
ASDAC proteins in plants. Genes orthologous to rice ASDAC have been widely distributed in
plants, including green algae (GenBank accession no. XP_005851005) and even
mycobacteria (GenBank accession no. EUA44530). Thus, the universal presence of
melatonin in living organisms is further supported by the widespread existence of ASDAC
genes. The functional significance and enzyme kinetics of ASDAC proteins from green algae
and mycobacteria will shed light on the regulatory mechanism of melatonin in other
organisms. Based on our results and previous reports, both OsHDAC10 and AtHDAC14
possess deacetylase activity for multiple substrates, such as N-acetylated amines and α-
tubulin. However, it is more likely that the ASDAC activity of OsHDAC10 and AtHDAC14
proteins prevails over the α-tubulin deacetylase activity inside plant cells, because
OsHDAC10 and AtHDAC14 reside in chloroplasts where α-tubulin is absent.
The cofactor requirements of rice and Arabidopsis ASDAC enzymes remain to be
elucidated. Rice ASDAC enzyme activity was undetectable in vitro both in the absence of E.
coli crude extracts and in the presence of boiled E. coli crude extracts, which suggests that
heat labile cofactors may be required for enzyme activity. In contrast, Arabidopsis ASDAC
had more than 40% activity in the absence of E. coli crude extracts, which indicates some
This article is protected by copyright. All rights reserved.

difference in the cofactor requirements of the two enzymes. The requirement of cofactors was
also observed in glucosaminyl N-deacetylase, which was found to require E. coli crude
52
extracts or a polycation polybrene for deacetylase activity. However, the addition of
polybrene was not effective to restore the activity of rice ASDAC, suggesting that other
cofactors or proteins are required for rice ASDAC enzyme activity (data not shown).
4.2 | Phylogenic significance of ASDAC genes in plants
The amino acid sequence of the OsHDAC10 protein possesses more than 60% identity to
proteins of green algae, such as Ostrecoccus tauri (65%), Micromonas pusilla (65%), and
Chlorella variabilis (64%). Also, the OsHDAC10 protein has 43% identity to that of the
archeon Methanosaeta thermophila (Figure 8A). Although the functional analysis of the
putative M. thermophila ASDAC gene is required, it is highly likely that the M. thermophila
ASDAC gene possesses ASDAC enzyme activity, as the cyanobacteria SNAT homolog
5
4
showed SNAT enzyme activity. It is intriguing to note that rice ASDAC homolog genes were
absent in cyanobacteria, but present in Archaea, while the rice SNAT homolog was present in
cyanobacteria, not in Archaea. These data may suggest that plant melatonin synthesis
originates from cyanobacteria via an endosymbiotic event, whereas the ASDAC gene evolved
later via another endosymbiotic event, which supports the theory of multiple endosymbiotic
55
events in plants. Both SNAT and ASDAC genes are present in organisms from green algae to
This article is protected by copyright. All rights reserved.

higher plants, suggesting that both genes are necessary for plants to survive and thrive during
normal plant growth and reproduction, as well as in response to various adverse
56
environmental stresses, including cold, pathogens, and high temperature.
Predicted transit sequences for subcellular localization, such as in chloroplasts, were
found in O. tauri and C. variabilis, but not in M. pusilla (Figure 8B). These data suggest that
the chloroplast targeting sequences of ASDAC genes were also acquired in the era of green
57
algae, as were the SNAT genes. The predicted transit sequence targeting mitochondria in C.
variabilis is of interest and needs to be examined further.
Rice HDAC10 showed 35% identity with human HDAC10 (HsHDAC10). Whether
HsHDAC10 has functional ASDAC activity or not remains to be investigated, but it is
unlikely because it has a low sequence identity and a different gene structure. The gene for
HsHDAC10 encodes 673 amino acids and consists of two catalytic domains: the N-terminal
catalytic domain spanning amino acids 1–354, which is complete, and the C-terminal
5
8,59
catalytic domain spanning amino acids 355–673, which has a partial catalytic domain.
Human HDAC10 is expressed in both the nucleus and cytoplasm with HDAC activity.
Recently, HsHDAC10 was found to possess polyamine deacetylase (PADA) activity, which
catalyzes the conversion of N-acetylspermidine into spermidine with a high K value (0.28
cat
–
1 60
S ).
This article is protected by copyright. All rights reserved.

In summary, plants possess two genes encoding SNAT and ASDAC proteins, which
act in the reversible biosynthesis of melatonin. SNAT favors melatonin synthesis, whereas
ASDAC restricts melatonin synthesis. The simultaneous presence of two enzymes in
chloroplasts suggests that melatonin synthesis is tightly controlled so that it is maintained at
an optimal level that is essential for plant growth and development. These ideas are consistent
with previous reports that the overexpression of melatonin biosynthetic genes, such as
61,62
9,29
31,63,64
tryptophan decarboxylase,
SNAT, and N-acetylserotonin methyltransferase,
did
not abundantly lead to the accumulation of melatonin in plants and thus led to almost normal
phenotypes. Further genetic alteration of plant ASDAC genes will shed light on the novel
functions of melatonin that have not been previously identified in plants.
ACKNOWLEDGMENTS
This research was supported by grants (2010–0020141 and 2017R1A2A2A05069253) from
the Basic Science Research Program of the National Research Foundation of Korea, funded
by the Ministry of Education, Republic of Korea. We express sincere thanks to Dr. Ok Jin
Hwang for technical assistance.
This article is protected by copyright. All rights reserved.

REFERENCES
1
2
3
4
5
6
7
8
9
. Reiter RJ, Mayo JC, Tan DX, et al. Melatonin as an antioxidant: under promises but over
delivers. J Pineal Res. 2016; 61:253-278.
. Hardeland R. Melatonin in plants – diversity of levels and multiplicity of functions. Front
Plant Sci. 2016;7:198.
. Ren W, Liu G, Chen S, et al. Melatonin signaling in T cells: Functions and application. J
Pineal Res. 2017;62:e12394.
. Majidinia M, Sadeghpour A, Mehrzadi S, et al. Melatonin: a pleiotropic molecule that
modulates DNA damage response and repair pathways. J Pineal Res. 2017;63:e12416.
. Tan DX, Manchester LC, Esteban-Zubero E, et al. Melatonin as a potent and inducible
endogenous antioxidant: synthesis and metabolism. Molecules. 2015;20:18886-18906.
. Arnao MB, Hernández-Ruiz J. Melatonin: plant growth regulator and/or biostimulator
during stress?. Trends Plant Sci. 2014;19:789-797.
. Li H, Chang J, Zheng J, et al. Local melatonin application induces cold tolerance in
distant organs of Citrullus lanatus L. via long distance transport. Sci Rep. 2017;7:40858.
. Zhang R, Sun Y, Liu Z, et al. Effects of melatonin on seedling growth, mineral nutrition,
and nitrogen metabolism in cucumber under nitrate stress. J Pineal Res. 2017;62:e12403.
. Lee K, Back K. Overexpression of rice serotonin N-acetyltransferase 1 in transgenic rice
plants confers resistance to cadmium and senescence and increases grain yield. J Pineal
This article is protected by copyright. All rights reserved.

Res. 2017;62:e12392.
1
0. Zhang J, Shi Y, Zhang X, et al. Melatonin suppression of heat-induced leaf senescence
involves changes in abscisic acid and cytokinin biosynthesis and signaling pathways in
perennial ryegrass (Lolium perenne L.). Environ Exp Bot. 2017;138:36-45.
11. Ding F, Wang M, Liu B, et al. Exogenous melatonin mitigates photoinhibition by
accelerating non-photochemical quenching in tomato seedlings exposed to moderate light
during chilling. Front Plant Sci. 2017;8:244.
1
1
2. Antoniou C, Chatzimichail G, Xenofontos R, et al. Melatonin systemically ameliorates
drought stress-induced damage in Medicago sativa plants by modulating nitro-oxidative
homeostasis and proline metabolism. J Pineal Res. 2017;62:e12401.
3. Wei Y, Hu W, Wang Q, et al. Identification, transcriptional and functional analysis of
heat-shock protein 90s in banana (Musa acuminate L.) highlight their novel role in
melatonin-mediated plant response to Fusarium wilt. J Pineal Res. 2016;62:e12367.
4. Liang C, Li A, Yu H, et al. Melatonin regulates root architecture by modulating auxin
response in rice. Front Plant Sci. 2017;8:134.
1
1
5. Gu Q, Chen Z, Yu X, et al. Melatonin confers plant tolerance against cadmium stress via
the decrease of cadmium accumulation and reestablishment of microRNA-mediated
redox homeostasis. Plant Sci. 2017;261:28-37.
This article is protected by copyright. All rights reserved.

1
1
6. Gong B, Yan Y, Wen D, et al. Hydrogen peroxide produced by NADPH oxidase: a novel
downstream signaling pathway in melatonin-induced stress tolerance in Solanum
lycopersicum. Physiol Plant. 2017;160:396-409.
7. Chen Z, Xie Y, Gu Q, et al. The AtrbohF-dependent regulation of ROS signaling is
required for melatonin-induced salinity tolerance in Arabidopsis. Free Radic Biol Med.
2017;108:465-477.
1
1
8. Lee HY, Back K. Melatonin is required for H O -and NO-mediated defense signaling
2
2
through MAPKKK3 and OXI1 in Arabidopsis thaliana. J Pineal Res. 2017;e12379.
9. Kaur H, Bhatla SC. Melatonin and nitric oxide modulate glutathione content and
glutathione reductase activity in sunflower seedling cotyledons accompanying salt stress.
Nitric Oxide. 2016;59:42-53.
2
2
0. Back K, Tan DX, Reiter RJ. Melatonin biosynthesis in plants: Multiple pathways catalyze
tryptophan to melatonin in the cytoplasm or chloroplasts. J Pineal Res. 2016;61:426-437.
1. Fujiwara T, Maisonneuve S, Isshiki M, et al. Sekiguchi lesion gene encodes a cytochrome
P450 monooxygenase that catalyzes conversion of tryptamine to serotonin in rice. J Biol
Chem. 2010;285:11308-11313.
2
2. Byeon Y, Back K. Melatonin production in Escherichia coli by dual expression of
serotonin N-acetyltransferase and caffeic acid O-methyltransferase. Appl Microbiol
Biotechnol. 2016;100:6683-6691.
This article is protected by copyright. All rights reserved.

2
2
2
2
2
3. Kang K, Lee K, Park S, et al. Molecular cloning of rice serotonin N-acetyltransferase, the
penultimate gene in plant melatonin biosynthesis. J Pineal Res. 2013; 55:7-13.
4. Byeon Y, Lee HY, Lee K, et al. Cellular localization and kinetics of the rice melatonin
biosynthetic enzymes SNAT and ASMT. J Pineal Res. 2014;56:107-114.
5. Byeon Y, Lee HY, Back K. Cloning and characterization of the serotonin N-
acetyltransferase-2 gene (SNAT2) in rice (Oryza sativa). J Pineal Res. 2016;61:198-207.
6. Wang L, Feng C, Zheng X, et al. Plant mitochondria synthesize melatonin and enhance
the tolerance of plants to drought stress. J Pineal Res. 2017;63:e12429.
7. Kang K, Kong K, Park S, et al. Molecular cloning of a plant N-acetylserotonin
methyltransferase and its expression characteristics in rice. J Pineal Res. 2011;50:304-
309.
2
2
3
8. Byeon Y, Lee HY, Lee K, et al. Caffeic acid O-methyltransferase is involved in the
synthesis of melatonin by methylating N-acetylserotonin in Arabidopsis. J Pineal Res.
2014;57:219-227.
9. Byeon Y, Lee HJ, Lee HY, et al. Cloning and functional characterization of the
Arabidopsis N-acetylserotonin O-methyltransferase responsible for melatonin synthesis. J
Pineal Res. 2016;60:65-73.
0. Zheng X, Tan DX, Allan AC et al. Chloroplastic biosynthesis of melatonin and its
involvement in protection of plants from salt stress. Sci Rep. 2017;7:41236.
This article is protected by copyright. All rights reserved.

3
1. Kang K, Lee K, Park S, et al. Enhanced production of melatonin by ectopic
overexpression of human serotonin N-acetyltransferase plays a role in cold resistance in
transgenic rice seedlings. J Pineal Res. 2010;49:176-182.
3
2. Byeon Y, Park S, Kim YS, et al. Light-regulated melatonin biosynthesis in rice during the
senescence process in detached leaves. J Pineal Res. 2012;53:107-111.
3. Lee K, Choi GH, Back K. Cadmium-induced melatonin synthesis in rice requires light,
hydrogen peroxide, and nitric oxide: key regulatory roles for tryptophan decarboxylase
and caffeic acid O-methyltransferase. J Pineal Res. 2017;63:e12441.
3
3
3
3
4. Kikuchi S, Satoh K, Nagata T, et al. Collection, mapping, and annotation of over 28,000
cDNA clones from japonica rice. Science. 2003;301:376-379.
5. Seki M, Narusaka M, Kamiya A, et al. Functional annotation of a full-length Arabidopsis
cDNA collection. Science. 2002;296:141-145.
6. Voinnet O, Rivas S, Mestre P, et al. An enhanced transient expression system in plants
based on suppression of gene silencing by the p19 protein of tomato bushy stunt virus.
Plant J. 2003;33:949-956.
3
3
7. Byeon Y, Lee HY, Hwang OJ, et al. Coordinated regulation of melatonin synthesis and
degradation genes in rice leaves in response to cadmium treatment. J Pineal Res.
2015;58:470-478.
8. Emanuelsson O, Nielsen H, Brunak S, et al. Predicting subcellular localization of proteins
This article is protected by copyright. All rights reserved.

based on their N-terminal amino acid sequence. J Mol Biol. 2000; 300:1005-1016.
9. Dereeper A, Audic S, Claverie JM, et al. 2010. BLAST-EXPLORER helps you building
datasets for phylogenetic analysis. BMC Evol Biol. 10, 8.
3
4
4
4
0. Fu W, Wu K, Duan J. Sequence and expression analysis of histone deacetylase in rice.
Biochem Biophys Res Commun. 2007;356:843-850.
1. Chung PJ, Kim YS, Park SH, et al. Subcellular localization of rice histone deacetylases in
organelles. FEBS Lett. 2009;583:2249-2254.
2. Schwer B, Bunkenborg J, Verdin RO, et al. Reversible lysine acetylation controlsteh
activity of the mitochondrial enzyme acetyl-CoA synthetase 2. Proc Natl Acad Sci USA.
2006;103:10224-10229.
4
3. Huang L, Sun Q, Qin F, et al. Down-regulation of a SILENT INFORMATION
REGULATOR2-related histone deacetylase gene, OsSRT1, induces DNA fragmentation
and cell death in rice. Plant Physiol. 2007;144:1508-1519.
4
4. Ma X, Lv S, Zhang C, Yang C. Histone deacetylases and their functions in plants. Plant
Cell Rep. 2013;32:465-478.
4
5. Haberland M, Montgomery RL, Olson EN. The many roles of histone deacetylases in
development and physiology: implications for disease and therapy. Nat Rev Genet.
2009;10:32-42.
4
6. Mayo JC, Sainz RM, Menéndez PG, et al. Melatonin and sirtuins: A “not-so unexpected”
This article is protected by copyright. All rights reserved.

relationship. J Pineal Res. 2017;62:e12391.
4
4
4
5
5
5
7. Orellana A, Hirschberg CB. Molecular cloning and expression of a glycosaminoglycan N-
acetylglucosaminyl N-deacetylase/N-sulfotransferase from a heparin-producing cell line.
J Biol Chem. 1994;269:2270-2276.
8. Sakurada K, Ohta T, Fujishiro K, et al. Acetylpolyamine amidohydrolase from
Mycoplana ramose: gene cloning and characterization of the metal-substituted enzyme. J
Bacteriol. 1996;178:5781-5786.
9. Lombardi PM, Angell HD, Whittington DA, et al. Structure of prokaryotic polyamine
deacetylase reveals evolutionary functional relationships with eukaryotic histone
deacetylases. Biochemistry. 2011;50:1808-1817.
0. Molesini B, Zanzoni S, Mennella G, et al. The Arabidopsis N-acetylornithine deacetylase
controls ornithine biosynthesis via a linear pathway with downstream effects on
polyamine levels. Plant Cell Physiol. 2017;58:130-144.
1. Song Y, Wu K, Dhaubhadel S, et al. Arabidopsis DNA methyltransferase AtDNMT2
associates with histone decaetylase AtHD2s activity. Biochem Biophys Res Commun.
2010; 396:187-192.
2. Tran HT, Nimick M, Uhrig G, et al. Arabidopsis thaliana histone deacetylase 14 (Hda14)
is an alpha-tubulin deacetylase that associates with Pp2a and enriches in the microtubule
fraction with the putative histone acetyltransferase Elp3. Plant J. 2012;71:263-272.
This article is protected by copyright. All rights reserved.

5
5
5
3. Eriksson I, Sandback D, Ek B, et al. cDNA cloning and sequencing of mouse
mastocytoma glucosaminyl N-deacetylase/N-sulfotransferase, an enzyme involved in the
biosynthesis of heparin. J Biol Chem. 1994;269:10438-10443.
4. Byeon Y, Lee K, Park YI, et al. Molecular cloning and functional analysis of serotonin N-
acetyltransferase from the cyanobacterium Synechocystis sp. PCC 6803. J Pineal Res.
2013;55:371-376.
5. Reiter RJ, Rosales-Corral S, Tan DX, et al. Melatonin as a mitochondria-targeted
antioxidant: one of evolution’s best ideas. Cell Mol Life Sci. 2017; DOI 10.1007/s00018-
017-2609-7.
5
5
6. Reiter RJ, Tan DX, Zhou Z, et al. Phytomelatonin: assisting plants to survive and thrive.
Molecules. 2015;20:7396-7437.
7. Byeon Y, Lee HY, Choi DW, et al. Chloroplast encoded serotonin N-acetyltransferase in
the red alga Pyropia yezoensis: gene transition to the nucleus from chloroplasts. J Exp Bot.
2015;66:709-717.
5
5
6
8. Fischer DD, Cai R, Bhatia U, et al. Isolation and characterization of a novel class II
histone deacetylase, HDAC10. J Biol Chem. 2002;277:6656-6666.
9. Guardiola AR, Yao TP. Molecular cloning and characterization of a novel histone
deacetylase HDAC10. J Biol Chem. 2002; 277:3350-3356.
0. Hai Y, Shinsky SA, Porter NJ, et al. Histone deacetylase 10 structure and molecular
This article is protected by copyright. All rights reserved.