Molecular evolution of multiple arylalkylamine N-acetyltransferase (AANAT) in fish

Article abstract of DOI:10.3390/md9050906

Arylalkylamine N-acetyltransferase (AANAT) catalyzes the transfer of an acetyl group from acetyl coenzyme A (AcCoA) to arylalkylamines, including indolethylamines and phenylethylamines. Multiple aanats are present in teleost fish as a result of whole genome and gene duplications. Fish aanat1a and aanat2 paralogs display different patterns of tissue expression and encode proteins with different substrate preference: AANAT1a is expressed in the retina, and acetylates both indolethylamines and phenylethylamines; while AANAT2 is expressed in the pineal gland, and preferentially acetylates indolethylamines. The two enzymes are therefore thought to serve different roles. Here, the molecular changes that led to their specialization were studied by investigating the structure-function relationships of AANATs in the gilthead seabream (sb, Sperus aurata). Acetylation activity of reciprocal mutated enzymes pointed to specific residues that contribute to substrate specificity of the enzymes. Inhibition tests followed by complementary analyses of the predicted three-dimensional models of the enzymes, suggested that both phenylethylamines and indolethylamines bind to the catalytic pocket of both enzymes. These results suggest that substrate selectivity of AANAT1a and AANAT2 is determined by the positioning of the substrate within the catalytic pocket, and its accessibility to catalysis. This illustrates the evolutionary process by which enzymes encoded by duplicated genes acquire different activities and play different biological roles.

Full text of DOI:10.3390/md9050906

Mar. Drugs 2011, 9, 906-921; doi:10.3390/md9050906
OPEN ACCESS
Marine Drugs
ISSN 1660-3397
www.mdpi.com/journal/marinedrugs
Article
Molecular Evolution of Multiple Arylalkylamine
N-Acetyltransferase (AANAT) in Fish
1
,
2
3
1
Bina Zilberman-Peled *, Sharron Bransburg-Zabary , David C. Klein and Yoav Gothilf
1
Department of Neurobiology, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel
Aviv, 69978, Israel; E-Mail: yoavg@tauex.tau.ac.il
2
3
Bioinformatics Unit, George S. Wise Faculty of Life Sciences, Tel Aviv University, Tel Aviv,
6
9978, Israel; E-Mail: sharron.zabary@gmail.com
Section on Neuroendocrinology, Office of the Scientific Directory, National Institute of Child
Health and Human Development, National Institutes of Health, Bethesda, MD 20892, USA;
E-Mail: kleind@mail.nih.gov
*
Author to whom correspondence should be addressed; E-Mail: binazp@gmail.com;
Tel.: +972-3-6409041; Fax: +972-3-6406329.
Received: 17 February 2011; in revised form: 16 May 2011 / Accepted: 17 May 2011 /
Published: 24 May 2011
Abstract: Arylalkylamine N-acetyltransferase (AANAT) catalyzes the transfer of an acetyl
group from acetyl coenzyme A (AcCoA) to arylalkylamines, including indolethylamines
and phenylethylamines. Multiple aanats are present in teleost fish as a result of whole
genome and gene duplications. Fish aanat1a and aanat2 paralogs display different patterns
of tissue expression and encode proteins with different substrate preference: AANAT1a is
expressed in the retina, and acetylates both indolethylamines and phenylethylamines; while
AANAT2 is expressed in the pineal gland, and preferentially acetylates indolethylamines.
The two enzymes are therefore thought to serve different roles. Here, the molecular
changes that led to their specialization were studied by investigating the structure-function
relationships of AANATs in the gilthead seabream (sb, Sperus aurata). Acetylation activity
of reciprocal mutated enzymes pointed to specific residues that contribute to substrate
specificity of the enzymes. Inhibition tests followed by complementary analyses of the
predicted three-dimensional models of the enzymes, suggested that both phenylethylamines
and indolethylamines bind to the catalytic pocket of both enzymes. These results suggest
that substrate selectivity of AANAT1a and AANAT2 is determined by the positioning of
the substrate within the catalytic pocket, and its accessibility to catalysis. This illustrates the

Mar. Drugs 2011, 9
evolutionary process by which enzymes encoded by duplicated genes acquire different
907
activities and play different biological roles.
Keywords: seabream; serotonin; dopamine; gene duplication
1
. Introduction
Arylalkylamine N-acetyltransferase (AANAT) belongs to the large GNAT acetyltransferases
superfamily, members of which are found in all kingdoms of life [1], reflecting the fundamental
importance of acetyl transfer. Members of the GNAT superfamily acetylate a wide range of substrates,
ranging from small molecules to large proteins [2]. AANATs selectively acetylate arylalkylamines,
including indolethylamines and phenylethylamines [3–7].
Members of the AANAT family are expressed in the vertebrate pineal gland and retina; homologs
are found in amphioxus, fungi, algae and bacteria [8,9]. In the vertebrate pineal the gland, AANAT
functions in the melatonin synthesis pathway by catalyzing acetylation of serotonin, resulting in
N-acetylserotonin; the latter is O-methylated by hydroxyindole-O-methyltransferase (HIOMT) to form
the hormone melatonin. Melatonin is produced in a rhythmic manner with high levels during the night
and low levels during the day. These temporal changes are controlled by the activity of AANAT [10,11].
Accordingly, AANAT is considered the key enzyme in the generation of the melatonin rhythm.
AANAT is a globular protein (20–25 kDa) comprised of a catalytic domain flanked by regulatory
domains. The catalytic domain includes the AcCoA binding region found in other members of the
GNAT superfamily and three AANAT-defining loops that form the arylalkylamine binding pocket.
The kinetic mechanism of AANAT obeys an ordered BiBi ternary complex (sequential) where the
binding of AcCoA precedes the binding of the amine substrate [12]. Crystallographic analysis of the
sheep AANAT indicated that the reactive groups (i.e., the primary amine and the acetyl group) are
positioned in close proximity within the catalytic funnel [13–17]. Catalysis is initiated by
1
20,122
deprotonation of the protonated amine, indirectly facilitated by neighboring histidines (His
). The
two histidines are part of a proton wire that channels protons to the surface of the enzyme. This
deprotonation leads to nucleophilic attack on the thioester bond of AcCoA. Protonation of the CoA
1
68
thiolate-leaving group is facilitated by a conserved tyrosine (Tyr ).
In mammals and chicken, only a single aanat gene has been identified, which is expressed in the
pineal gland and in the retina. In teleost fish, multiple aanat genes, aanat1a, aanat1b and aanat2, have
been identified and characterized [8,18–23]. The existence of multiple aanats in fish is likely to be the
result of a genome and gene duplications that occurred after the divergence of teleosts and the tetrapod
lineage [8,24,25]. Genome duplications are important molecular evolutionary forces that allow the
duplicate genes to acquire new or related functions, or to alter their expression pattern, a process
known also as co-option [24,26]. Such an evolutionary process is apparent from studies on aanat1a
and aanat2. These genes display different patterns of tissue expression and encode proteins that have
different kinetic characteristics and substrate preference, consistent with different functions.
AANAT1a is expressed mainly in the retina and to some extent also in the brain, and, like the
mammalian and avian AANATs, acetylates both indolethylamines and phenylethylamines. AANAT2

Mar. Drugs 2011, 9
908
is primarily expressed in the pineal gland and preferentially acetylates indolethylamines, especially
serotonin [23,27]. The activity of AANAT2 is characterized by weaker affinity and higher velocity for
its substrates as compared to those of AANAT1a [18,22,28]. The biological significance of multiple
AANATs in fish is likely to represent evolutionary responses to different selective pressures. While
AANAT2 may have acquired a higher selectivity for serotonin in order to enhance the efficiency of
melatonin production [8,29], AANAT1a preserved a broad arylalkylamine acetylation capability so
that it could serve multiple functions in the retina. Thus, retinal AANAT may play other roles in
addition to being involved in melatonin synthesis. Through the acetylation of dopamine, retinal
AANAT may contribute to the control of dopamine levels or initiate the synthesis of a novel bioactive
molecule [29]. Retinal AANAT may also play a more general role in detoxification through its action
on a broad range of arylalkylamines [8,30]. AANAT1b and AANAT1a are closely-related proteins
with high homology and identical substrate binding and catalytic sites. Although the biochemical
properties of AANAT1b are still unknown, the presence of the three AANATs in some fish species
supports the hypothesis that these enzymes play multiple roles [31,32].
The presence of multiple AANATs in fish provides an opportunity to study the effect of natural
mutations in these closely-related molecules. Here we investigate the structure-function relationship of
seabream AANAT1a and AANAT2, by attempting to identify the molecular basis of their kinetic
differences and substrate preference. Inhibition experiments were performed to assess the binding
capabilities of AANAT1a and AANAT2 to their different substrates. Site-directed mutagenesis was
used to determine the importance of residues in the arylalkylamine binding domain that differ between
the two enzymes. In addition, the predicted structures of wild type AANAT1a and AANAT2 were
compared in regard to their capacity to bind arylalkylamines. The results of these studies provide a
foundation for a molecular explanation of the functional differences between AANAT1a and AANAT2.
2
. Results and Discussion
2
.1. Kinetic Characterization of Mutated AANAT1a and AANAT2
Fish have become important models for the evolutionary significances of gene duplication since
many gene families possess more members in fish than in mammals. Two evolutionary processes can
lead to a protein with a novel function: a change in the coding sequence of the gene that alters the
biochemical properties of the protein, and a change in its regulatory sequence that alters its temporal or
spatial expression pattern [26]. The duplicate AANATs in fish illustrate these evolutionary processes
by representing differences, both in their spatio-temporal expression patterns [18,27,33,34] and in their
biochemical properties [18,22,28]. The different qualities of each enzyme are thought to serve two
melatonin generating systems, one in the retina and the other in the pineal gland [18,21,22]. It has also
been suggested that AANAT1a is involved in dopamine metabolism and amine detoxification in the
retina [8,29].
In the current study the molecular-structural base for the differences between the two enzymes was
biochemically investigated, followed by in silico experiments.

Mar. Drugs 2011, 9
909
The effect of reciprocal mutation (See ‗Experimental Section‘ for details, Figure 1, Table 1) on the
activity of the two enzymes was measured in order to study the molecular basis that underlies the
biochemical differences.
Figure 1. Sequence alignment of sheep AANAT, seabream AANAT1a and seabream
AANAT2. The alignment starts in residue 30 of the sheep AANAT, in residue 25 of the
seabream AANAT1a, and in residue 28 of the seabream AANAT2. The coloring of the
alignment follows the sequence similarity between AANAT1a or AANAT2 and the
sequence of the sheep AANAT; identical residues are shaded and colored in white, similar
residues are shaded. The secondary structure elements, as analyzed by Hickman et al. [15]
and Dyda et al. [2], are shown above the residues. The mutation sites are boxed. The
pairwise alignment was done using the Needle alignment tool.
Table 1. Mutation sites in AANAT1a and AANAT2 and the comparable sites in the
sheep AANAT.
Mutation sites
Mutation
region/name
AANAT1a
AANAT2
Sheep AANAT
β5
α2
β4
M154I
I157M
M159
-
-
-
L69R + N70H + G73T + Q74L Q71 + H72 + T75 + L76
S123A
M107L
A125
L109
αa
a
Region definition is after Hickman et al. [15] and Dyda et al. [2].
b
AANAT1a- α2, β4 and αa mutants were inactive and therefore were not analyzed.
Analysis of the acetylation activity of wild type and mutant proteins was done with serotonin and
tryptamine (indolethylamines) and dopamine and phenylethylamine (phenylethylamines). Figure 2
represents acetylation activity with serotonin and phenylethylamine, which are generally similar to the
kinetics found with tryptamine and dopamine, respectively. Table 2 represents the catalytic efficiency
in terms of Kcat/Km values. Acetylation of phenylethylamine and dopamine by AANAT2 mutants did
not obey the Michaelis-Menten equation (Figure 2), and their activities are therefore expressed relative
to wild type AANAT2 activity in the presence of 10 mM substrate (shaded values, Table 2); these

Mar. Drugs 2011, 9
910
values cannot be compared to other enzyme-substrate Kcat/Km values. The specific effect of each
mutation is detailed below.
Figure 2. Acetylation of arylalkylamines by wild type and mutant AANATs. Activities
were measured in the presence of near saturating levels of AcCoA and increasing
concentration of amine substrates. The graphs represent acetylation activities (mean ± SE)
with serotonin and phenylethylamine. A generally similar kinetics was found with
tryptamine and dopamine. The AANAT1a- α2, β4 and αa mutants did not express well and
were inactive, consistent with misfolding, and therefore were not analyzed (see
‗Experimental Section‘ for details). (A) AANAT1a and mutants with serotonin;
(B) AANAT1a and mutants with phenylethylamine; (C) AANAT2 and mutants with
serotonin; (D) AANAT2 and mutants with phenylethylamine; PEA = phenylethylamine.
Table 2. Apparent Kcat/Km values (mean ± SE) of wild type and mutant proteins for
serotonin, tryptamine, dopamine and phenylethylamine.
AANAT1a
Wild type β5
AANAT2
Wild type
β5
α2
β4
αa
Serotonin
61.36 ± 4.39 64.15 ± 3.49 6.60 ± 0.54 11.34 ± 0.93 17.65 ± 1.06 18.58 ± 1.41 7.29 ± 0.54
23.37 ± 1.59 14.84 ± 0.60 3.19 ± 0.25 8.50 ± 0.69 6.62 ± 0.33 9.06 ± 0.78 1.46 ± 0.003
Tryptamine
Dopamine
7.30 ± 0.54 1.86 ± 0.49
32.86 ± 2.74 5.05 ± 0.61
1
1
2.40 ± 0.10 3.80 ± 0.07 2.48 ± 0.18 0.65 ± 0.03
2.52 ± 0.26 4.33 ± 0.56 3.03 ± 0.09 0.61 ± 0.03
Phenylethylamine
Note: Acetylation of phenylethylamine and dopamine by AANAT2 mutants are expressed relative to wild
type and are indicated as shaded values.

Mar. Drugs 2011, 9
.1.1. β5 Mutations
911
2
1
59
Met is one of six residues that form the amine substrate binding pocket in sheep AANAT [15]. It
forms a hydrogen bond with the nitrogen of the amine substrate and is therefore crucial for
its stabilization.
The results showed that substitution of Ile to Met in AANAT2 (mutation β5 of AANAT2) increased
the ability of AANAT2 to acetylate phenylethylamines by ~2.5-fold (p < 0.05) as compared to wild
type, and also increased its affinity to indolethylamines (Km = 0.90 mM vs. 2.0 mM of wild type for
tryptamine and Km = 0.94 mM vs. 2.2 mM of wild type for serotonin), resulting in a 2- to 3-fold
increase (p < 0.05) in the Kcat/Km values (Table 2, Figure 2). The reciprocal β5 mutation of
AANAT1a (M154I) reduced AANAT1a capacity to acetylate dopamine and phenylethylamine by
3
.9- and 6.5-fold (p < 0.01), respectively, but reduced tryptamine acetylation by only 1.6-fold
(p < 0.05), while having almost no effect on its capacity to acetylate serotonin (Table 2, Figure 2),
indicating that the decrease in phenylethylamines acetylation is specifically related to the mutation.
These results suggest that the Met/Ile difference in this position contributes to the different kinetic
characteristics and the low ability of AANAT2 to acetylate phenylethylamines.
1
59
The importance of the hydrogen bond formed at this site (Met
in sheep AANAT) to the
stabilization of the amine substrate was previously emphasized [2,15]. The contribution of this site to
the catalytic mechanism of AANATs is now expanded to include determination of substrate
1
54
selectivity. It is notable that an evolutionary substitution of two neutral residues, Met in AANAT1a
1
57
by Ile in AANAT2, has a profound effect on substrate preference.
2
.1.2. α2 Mutation
The α2 region is a component of loop 1 which was recognized as the most floppy part of the
enzyme [14,15,35]. The binding of AcCoA displaces loop 1 and leads to reorganization of helices α1
and α2 at both ends of the loop. This conformational change contributes to the formation of the
arylalkylamine binding pocket [14,15]. Interestingly, this small region contains changes in five
adjacent amino acids that characterize fish AANAT2. The mutations of α2 region changed two pairs of
adjacent amino acids.
This mutation increased AANAT2 ability to acetylate dopamine and phenylethylamine by 3.80- and
4
.33- fold, (p < 0.05) respectively as compared to wild type; and increased its affinity to
indolethylamines (1.2 mM for tryptamine and 0.88mM for serotonin), as reflected in a 2- to 2.5- fold
increase (p < 0.05) in the Kcat/Km values (Table 2, Figure 2). The α2 mutation conferred upon
AANAT2 the highest increase of phenylethylamines acetylation among the four AANAT2 mutants.
The results strengthen previous studies showing the importance of this region for the catalytic
action of the AANAT enzymes. Furthermore, it suggests that evolutionary changes within the α2
region of AANAT2 underlie the structural changes that may contribute to its low capacity to
acetylate phenylethylamines.

Mar. Drugs 2011, 9
.1.3. β4 Mutation
The β4 region is located within the most conserved motif of the GNAT superfamily which is the
912
2
site of AcCoA binding and contains residues that participate in the catalytic reaction [2].
The results show that substitution of Ser to Ala in AANAT2 (Mutation β4 of AANAT2) increased
the ability of AANAT2 to acetylate dopamine and phenylethylamine by 2.48- (p < 0.05) and 3.03-
fold, (p = 0.079), respectively. This mutation also increased the affinity of AANAT2 for
indolethylamines (0.65 mM for tryptamine and 0.65mM for serotonin), as reflected in the ~3-fold
(p < 0.05) increase in the Kcat/Km values (Table 2, Figure 2).
These results support previous studies showing that the β4 region is important for the catalytic
1
20
123
reaction of the AANAT enzymes, and specifically suggest that Ala (in AANAT1a) and Ser (in
AANAT2) are involved in determining the kinetics and substrate preference of these enzymes.
2
.1.4. αa Mutation
αa is a subdomain of the two-turn helix that is located between the β3 and β4 sheets [14]. This helix
defines a ―third wall‖ in the amine binding site that almost entirely encloses the catalytic funnel, and
exists only in the AANAT family and not in other members of the GNAT superfamily [2].
Specifically, residue Leu-109 of the Sheep AANAT was reported to be part of the serotonin binding
site [15]. The corresponding substitution of Met to Leu in AANAT2 (mutation αa of AANAT2) did
not effect phenylethylamines or indolethylamines acetylation activities in comparison to wild type
1
04
107
(
Table 2, Figure 2). These results suggest that Leu (AANAT1a) and Met (AANAT2), which
participate in the serotonin binding site, do not contribute to the differences in the substrate preference
of these enzymes.
2
.2. Inhibition Experiments
Two reasons may account for the low ability of AANAT2 to acetylate phenylethylamines:
(1) phenylethylamines do not bind to the catalytic pocket of AANAT2; or, (2) phenylethylamines do
bind to the catalytic pocket of AANAT2 but are not acetylated by the enzyme. In the latter case, an
inhibition of indolethylamines acetylation would be expected in the presence of phenylethylamines. To
discriminate between the two options, acetylation of subsaturating tryptamine concentrations, at about
the Km value of each enzyme (0.15 mM for AANAT1a and 1.5 mM for AANAT2) was measured in
the absence or presence of increasing dopamine concentrations: 0.1 mM or 1mM for AANAT1a and
AANAT2, respectively (~1×); 1 mM or 10 mM for AANAT1a and AANAT2, respectively (~6×); and
5
mM or 50 mM for AANAT1a and AANAT2, respectively (~30×) the tryptamine concentrations
(Figure 3). Similar inhibition patterns (Pearson correlation, R = 0.87, p = 0.024) were found for both
enzymes, with minimal inhibition at the 1:1 dopamine:tryptamine ratio, ~30% inhibition at the 6:1
dopamine:tryptamine ratio, and ~ 60% inhibition at the 30:1 ratio (Figure 3). These results suggest that
dopamine binds to the catalytic pockets of AANAT1a and AANAT2 at similar affinities, which are
lower than those of tryptamine.

Mar. Drugs 2011, 9
Figure 3. Dopamine inhibits tryptamine acetylation by AANAT1a and AANAT2.
913
Dopamine inhibition of tryptamine acetylation was analyzed using the radiochemical assay.
The concentrations of tryptamine selected approximated the Km values of each enzyme
(0.15 and 1.5 mM for AANAT1a and AANAT2, respectively), in the presence of
increasing concentrations of dopamine (0, 0.1, 1, 5 mM or 0, 1, 10, 50 mM for AANAT1a
and AANAT2, respectively). Other conditions were as previously determined for wild type
AANAT1a and AANAT2 [22]. Values are given as mean ± SE.
The amine binding pocket of AANATs constitutes a small edge of the catalytic funnel that is
surrounded by hydrophobic residues. The dominant factor that determines which molecules will enter
the funnel is size; and the main forces that hold the amines are van der Waals‘ interactions and
hydrophobic packing [14,15].
The catalytic mechanism of AANAT, like other members of the GNAT superfamily, involves an
acetyl transfer from AcCoA to the acceptor through a direct nucleophilic attack. Although members of
the GNAT superfamily acetylate a wide variety of substrates and exhibit low amino acids sequence
identity, the structural conservation of the AcCoA binding site is remarkably high [2,36]. The catalytic
mechanism of AANATs depends on proper orientation of the amine and thioacetyl groups, which are
stabilized by three hydrogen bonds [2,15].
These inhibition experiments suggested that phenylethylamines bind to the AANAT2 catalytic
pocket although they are poorly acetylated by the enzyme. This may happen if their spatial and/or
electrostatic binding within the catalytic pocket does not allow efficient catalysis.
2
.3. AANAT1a and AANAT2 Have Similar Predicted Binding Capacities
Structural analysis. Since phenylethylamines appear to bind to the catalytic pocket of AANAT2, it
may be hypothesized that the kinetic differences and substrate preferences between the two enzymes
stem from differences in the spatial relationships between the bound substrate and catalytic residues.
This hypothesis was further explored by investigating the predicted 3-D structures of the seabream
AANAT1a and AANAT2, using the structure of the sheep AANAT-bisubstrate analog complex PDB
accession number: 1CJW [15] as a template. General surface and electrostatic properties were
evaluated to assess the similarities and differences between the catalytic funnels of the two enzymes.

Mar. Drugs 2011, 9
914
Figure 4 depicts the curvature analysis of the surface (upper panels), and the electrostatic properties
analysis (lower panels) of the catalytic funnel of AANAT1a and AANAT2. The overall topology of
the funnel in the AANAT1a and AANAT2 models is fairly similar, with some minor differences. One
1
81
184
stems from the presence of Leu vs. Met in AANAT1a and AANAT2, respectively. The longer
1
84
side chain of the Met (colored blue) projects to the cavity (lower part of the pocket), and covers part
1
81
181
184
of Ile of AANAT2 (colored green). It should be noted however that the Leu to Met and the
1
78
181
Val to Ile changes have been found only in seabream AANATs, not in other known AANATs.
1
15
117
Another topological difference, in the vicinity of the catalytic histidine residues (His and His in
1
18
120
AANAT1a, His and His in AANAT2), is an enlargement of the cavity in AANAT2 as compared
to that of AANAT1a (upper part of the cavity, Figure 4). In addition, the charge distribution between
1
54
157
the two histidines and in the vicinity of Met (AANAT1a) and Ile (AANAT2) is different.
As suggested by the inhibition experiments, the relatively minor chemophysical differences
between the two enzymes do not suggest limitations on the binding of phenylethylamines to the
AANAT2 pocket. However, since the catalytic event is based on proper orientation of the ternary
intermediate complex [2], these minor differences in topology and electrostatic fields may affect the
efficiency of the catalysis.
2
.4. In Silico Assessment of Binding Stability
Docking experiments are another bioinformatics tool that can be used to explore enzyme-substrate
relationships, by predicting the probability of enzyme-substrate binding. Therefore, rigid docking
experiments were performed in order to assess the binding stability of AANAT1a, AANAT2 or
β5-AANAT2, and indolethylamines (serotonin and tryptamine) or phenylethylamines (dopamine and
phenylethylamine), based on their geometric complementarity. Analysis of the spatial binding showed
that phenylethylamines bind to the same region of AANAT2 as indolethylamines. This strengthens the
hypothesis that the inhibition of tryptamine acetylation by dopamine (Figure 3) is due to competitive
binding of tryptamine and dopamine within the catalytic pocket of AANAT2. Analyses of the binding
stability values (supplementary data, Table S1) indicate that both AANAT1 and ANNAT2 have better
geometric complementarity with indolethylamines than with phenylethylamines, as reflected in the
~
13% decrease in the binding stability values for each enzyme. This is in agreement with the measured
activities of AANATs, which indicate that indolethylamines are the better substrates for both enzymes
Figure 2; [22]). Moreover, in agreement with the results of the inhibition experiments (Figure 3),
(
stable AANAT2-phenylethylamines complexes were predicted with values that were similar (>99%) to
those of AANAT1a-phenylethylamines complexes, suggesting that phenylethylamines bind AANAT1a
and AANAT2 pockets with similar efficiencies. This result supports the outcome of the inhibition
experiments and the predicted 3-D structures analyses, suggesting that the kinetic differences between
the two enzymes stem from differences in the spatial relationships between the bound substrate and
catalytic residues, which determine the proximity of the amine to the thioacetyl group and
neighboring residues.

Mar. Drugs 2011, 9
Figure 4. 3-D surface properties of the catalytic funnel of seabream AANAT1a and
915
AANAT2. The 3-D models of AANAT1a (left panels) and AANAT2 (right panels) were
reconstructed based on the 3-D structure of the sheep AANAT complex with a bisubstrate
analog [15]. The curvature analysis of the surface is shown on the upper part of the picture,
with the innermost regions colored in grey and the outermost in green (see scale bar). The
lower part of the picture depicts the electrostatic properties analysis of the funnel, with the
positive parts colored in blue and the negative parts in red (see scale bar). Residues that
form the amine binding site and catalytic residues are colored as follows:
5
1
54
59
62
Phe (AANAT1a) and Phe (AANAT2), red; Pro (AANAT1a) and Pro (AANAT2),
1
54
157
178
181
yellow; Met (AANAT1a) and Ile (AANAT2), magenta; Val (AANAT1a) and Ile
1
81
184
115
(
AANAT2), green; Leu (AANAT1a) and Met (AANAT2), blue; His (AANAT1a)
1
18
117
120
and His
(AANAT2), cyan; His
(AANAT1a) and His
(AANAT2), grey. Grasp
software package was used for the surface property analysis.
AANAT1a
AANAT2
His120
His117
Phe51
Phe54
His115
Met154
His118
Pro59
Pro62
Ile157
Val178
Leu181
Ile181
Met184
AANAT1a
AANAT2
A possible extension of this hypothesis stems from the increased affinity of the AANAT2 mutants,
not only to the phenylethylamines but also to the indolethylamine substrates (Table 2, Figure 2). This
finding suggests that the substrate selectivity of AANAT2 is an outcome of its general low affinity,
even for its preferred substrates, which results in an almost total reduction of activity with
phenylethylamines. However, it should be noted that the reciprocal effect was not observed, i.e., the
reaction rate of β5-AANAT1a with indolethylamines was only slightly decreased as compared to the

Mar. Drugs 2011, 9
916
wild type AANAT1a. Nevertheless, it is likely that this phenomenon at least partially contributes to the
different substrate preferences of the two enzymes.
Multiple functions of fish AANATs are consistent with the hypothesis that the ancestral AANAT
evolved as a detoxification enzyme in the primitive photoreceptor [30,31]. Later in evolution, when
melatonin became an important signal, its increased production became a pineal gland-specific feature
in order to avoid toxic processes within the visual phothoreceptor cell. Consequently, current AANAT
expression in the vertebrate retina may reflect its ancestral-conserved function as a detoxification
enzyme. The expression of AANAT in extra-pineal and extra-retinal sites including the brain [37–40],
raises the hypothesis that this enzyme is involved in detoxification functions in these tissues. Of
special interest is the idea that AANAT acetylates dopamine in the brain.
The results of the current study provide a possible molecular-structural mechanism for the substrate
specificity of the two paralog AANATs, and demonstrate how changes in the coding sequence may
have led to the formation of two enzymes with different biochemical properties and possibly
different functions.
3
. Experimental Section
3
.1. Mutant Design
Candidate residues that control substrate preference of AANAT1a and AANAT2 were identified
using site-directed mutagenesis. Mutations were selected based on: (1) the uniqueness of residues that
characterize fish AANTA2 vs. fish AANAT1s and AANATs of other vertebrates (multiple sequence
alignment, ClustalW 1.82); and (2) the deduced or demonstrated importance of the region for the
catalytic activity or substrate binding [13,15,17]. AANTA1a and AANAT2 were mutated by reciprocal
1
54
substitutions. For example, in mutation β5 (Table 1), Met in AANAT1a was changed to Ile which
occupies the corresponding position in AANAT2 and vice versa. A description of each mutant enzyme
is presented in Table 1.
3
.2. Preparation of Wild Type and Mutant Proteins
Wild type and mutant proteins were expressed as GST fusion proteins. Seabream AANAT1a and
AANAT2 expression vectors, psb1ex and psb2ex, respectively [22], were used as templates to prepare
mutant protein expression vectors using the Quickchange PCR-based site-directed mutagenesis kit
(Stratagene, La Jolla, CA, USA). Mutations were performed according to manufacturer‘s instructions,
and were verified by DNA sequencing. Plasmid vectors (pGEX-4T-1, Amersham) containing the wild
type or mutant coding region were transformed into BL21 E. coli. Proteins were produced and purified
as described [22]. Purity and concentration of the purified recombinant proteins were determined by
1
0% SDS-polyacrylamide gel electrophoresis and Bradford assay, respectively. Protein aliquots
(10 µL) were stored at −80 C. The preparations of the wild type and mutant AANAT2s yielded high
expression and purity (~80%) and the proteins exhibited acetylation activities (see results). Wild type
AANAT1a and mutant β5-AANAT1a had lower expression, were isolated at ~70% purity and
exhibited acetylation activities. Mutant proteins α2-, β4- and αa-AANAT1a did not express well and
were inactive, consistent with misfolding, and therefore were not analyzed.

Mar. Drugs 2011, 9
.3. Assay of AANAT Activity
A colorimetric assay was used to measure the activity of wild type and mutant proteins. This assay
917
3
is based on the quantification of CoASH generated during acetyl transfer [12,41]. Standard reactions
for wild type and mutated proteins were as previously described for AANAT1a and AANAT2 [22] in
the presence of the indicated concentrations of amine substrates. Assays were done 2–3 times in
duplicate. Apparent Km and Kcat values were calculated by multiple non-linear regressions, using the
program GOSA [42], and are presented as mean ± SE.
A radiochemical assay [43] was used to determine the activity of wild type proteins in ‗inhibition
experiments‘ which were designed to determine the effect of dopamine on the acetylation of
tryptamine. Wild type proteins (25 nM) were incubated with the indicated concentration of tryptamine
3
and dopamine and 0.5 mM H-acetyl-coenzyme A (4.1 Ci/mmol, Amersham, Buckinghamshire, UK),
in a final volume of 100 µL for 40 min, at optimal conditions of each enzyme. The product
3
N-[ H]-acetyltryptamine was extracted into chloroform; this procedure does not extract
acetyldopamine. The chloroform was evaporated and the residue was redissolved in scintillation fluid
for counting. Assays were done twice in duplicate and are presented as mean ± SE.
3
.4. Structural Analysis
Pairwise alignment of seabream AANAT1a (GenBank accession number: AY533402), AANAT2
(
GenBank accession number: AY533403), and sheep AANAT (Uniprot accession number: 002785)
Figure 1) was done using the Needle alignment tool [44,45], and was used as a basis for the homology
(
modeling of the 3-D models of the seabream AANAT1a and AANAT2.
The structure-modeling package JACKAL (Nest module, [46]) was used to calculate the AANAT1a
and AANAT2 structures, based on the template structure of sheep AANAT in complex with a
bisubstrate analog (PDB entry: 1CJW). The stereochemical quality of the models was checked using
the PROCHECK suite [47]. The PROC-AVA values were 0.35 and 0.3 for AANAT1a and AANAT2,
respectively, indicating high structural reliability. In addition, the resulting models were compared to
models calculated using the ModBase modeling server [48]. The structure models were found to be
similar in both modeling methods. Hydrogen atoms were added using H++ server [49] under the
following conditions: pH = 6.5, I = 0.15 M, ε_in = 6, ε_out = 80. The 3-D calculated models of
AANAT1a and AANTA2 are provided upon request. Grasp software package [50] was used for
surface property analysis (same conditions as above). Assessments of the binding stability between
AANAT1a, AANAT2, or β5-AANAT2 and indolethylamines (serotonin and tryptamine) or
phenylethylamines (dopamine and phenylethylamine), were obtained by rigid docking based on
geometric complementarity of the protein and substrate (PatchDock server, [51,52].
3
.5. Statistical Analysis
Differences in acetylation activity between wild type AANAT1a and its mutant were evaluated
using t-test. Differences between wild type AANAT2 and its mutants were evaluated using one sample
t-test for phenylethylamine substrates or by analysis of variance (ANOVA) followed by Tukey‘s

Mar. Drugs 2011, 9
918
post-hoc test for the indolethylamine substrates. To test similarity of dopamine inhibition effect
between AANAT1a and AANAT2, Pearson‘s correlation was carried out.
4
. Conclusions
In the current study, the bases for the biochemical properties and substrate specificity differences
between AANAT1a and AANAT2 paralogs were biochemically investigated, followed by supporting
in silico experiments. The results suggest that the kinetics of these enzymes is determined by the
accessibility of the bound substrate to catalytic residues within the catalytic pocket. This expands the
understanding of substrate selectivity and mechanism of action of the AANAT enzyme family, and
may be useful for the development of specific inhibitors of AANAT, which are of special interest for
melatonin-related clinical applications such as sleep disorders, and jet lag or mood disorders, in which
serotonin and dopamine are involved.
Acknowledgments
This research was supported by Grant No 232/00–17.2 from the Israel Science Foundation, Jerusalem.
References
1
2
3
.
.
.
Neuwald, A.F.; Landsman, D. GCN5-related histone N-acetyltransferases belong to a diverse
superfamily that includes the yeast SPT10 protein. Trends Biochem. Sci. 1997, 22, 154–155.
Dyda, F.; Klein, D.C.; Hickman, A.B. GCN5-related N-acetyltransferases: A structural overview.
Annu. Rev. Biophys. Biomol. Struct. 2000, 29, 81–103.
Ferry, G.; Ubeaud, C.; Dauly, C.; Mozo, J.; Guillard, S.; Berger, S.; Jimenez, S.; Scoul, C.;
Leclerc, G.; Yous, S.; Delagrange, P.; Boutin, J.A. Purification of the recombinant human
serotonin N-acetyltransferase (EC 2.3.1.87): Further characterization of and comparison with
AANAT from other species. Protein Expr. Purif. 2004, 38, 84–98.
4
5
.
.
Khalil, E.M.; de Angelis, J.; Cole, P.A. Indoleamine analogs as probes of the substrate selectivity
and catalytic mechanism of serotonin N-acetyltransferase. J. Biol. Chem. 1998, 273, 30321–30327.
Coon, S.L.; Roseboom, P.H.; Baler, R.; Weller, J.L.; Namboodiri, M.A.; Koonin, E.V.; Klein,
D.C. Pineal serotonin N-acetyltransferase: Expression cloning and molecular analysis. Science
1
995, 270, 1681–1683.
6
7
.
.
Voisin, P.; Namboodiri, M.A.; Klein, D.C. Arylamine N-acetyltransferase and arylalkylamine
N-acetyltransferase in the mammalian pineal gland. J. Biol. Chem. 1984, 259, 10913–10918.
Ferry, G.; Loynel, A.; Kucharczyk, N.; Bertin, S.; Rodriguez, M.; Delagrange, P.; Galizzi, J.P.;
Jacoby, E.; Volland, J.P.; Lesieur, D.; Renard, P.; Canet, E.; Fauchere, J.L.; Boutin, J.A. Substrate
specificity and inhibition studies of human serotonin N-acetyltransferase. J. Biol. Chem. 2000,
2
75, 8794–8805.
8
.
Coon, S.L.; Klein, D.C. Evolution of arylalkylamine N-acetyltransferase: Emergence and
divergence. Mol. Cell. Endocrinol. 2006, 252, 2–10.

Mar. Drugs 2011, 9
919
9
1
.
Iyer, L.M.; Aravind, L.; Coon, S.L.; Klein, D.C.; Koonin, E.V. Evolution of cell-cell signaling in
animals: Did late horizontal gene transfer from bacteria have a role? Trends Genet. 2004, 20,
92–299.
2
0. Klein, D.C.; Coon, S.L.; Roseboom, P.H.; Weller, J.L.; Bernard, M.; Gastel, J.A.; Zatz, M.;
Iuvone, P.M.; Rodriguez, I.R.; Begay, V.; Falcon, J.; Cahill, G.M.; Cassone, V.M.; Baler, R. The
melatonin rhythm-generating enzyme: Molecular regulation of serotonin N-acetyltransferase in
the pineal gland. Recent Prog. Horm. Res. 1997, 52, 307–357; discussion 357–308.
1
1
1
1
1
1. Ganguly, S.; Coon, S.L.; Klein, D.C. Control of melatonin synthesis in the mammalian pineal
gland: The critical role of serotonin acetylation. Cell Tissue Res. 2002, 309, 127–137.
2. De Angelis, J.; Gastel, J.; Klein, D.C.; Cole, P.A. Kinetic analysis of the catalytic mechanism of
serotonin N-acetyltransferase (EC 2.3.1.87). J. Biol. Chem. 1998, 273, 3045–3050.
3. Scheibner, K.A.; de Angelis, J.; Burley, S.K.; Cole, P.A. Investigation of the roles of catalytic
residues in serotonin N-acetyltransferase. J. Biol. Chem. 2002, 277, 18118–18126.
4. Hickman, A.B.; Klein, D.C.; Dyda, F. Melatonin biosynthesis: The structure of serotonin
N-acetyltransferase at 2.5 A resolution suggests a catalytic mechanism. Mol. Cell 1999, 3, 23–32.
5. Hickman, A.B.; Namboodiri, M.A.; Klein, D.C.; Dyda, F. The structural basis of ordered substrate
binding by serotonin N-acetyltransferase: Enzyme complex at 1.8 A resolution with a bisubstrate
analog. Cell 1999, 97, 361–369.
1
6. Obsil, T.; Ghirlando, R.; Klein, D.C.; Ganguly, S.; Dyda, F. Crystal structure of the
1
4-3-3zeta:serotonin N-acetyltransferase complex. A role for scaffolding in enzyme regulation.
Cell 2001, 105, 257–267.
1
1
1
7. Wolf, E.; De Angelis, J.; Khalil, E.M.; Cole, P.A.; Burley, S.K. X-ray crystallographic studies of
serotonin N-acetyltransferase catalysis and inhibition. J. Mol. Biol. 2002, 317, 215–224.
8. Coon, S.L.; Begay, V.; Deurloo, D.; Falcon, J.; Klein, D.C. Two arylalkylamine N-acetyltransferase
genes mediate melatonin synthesis in fish. J. Biol. Chem. 1999, 274, 9076–9082.
9. Gothilf, Y.; Coon, S.L.; Toyama, R.; Chitnis, A.; Namboodiri, M.A.; Klein, D.C. Zebrafish
serotonin N-acetyltransferase-2: Marker for development of pineal photoreceptors and circadian
clock function. Endocrinology 1999, 140, 4895–4903.
2
2
0. Mizusawa, K.; Iigo, M.; Masuda, T.; Aida, K. Photic regulation of arylalkylamine
N-acetyltransferase 1 mRNA in trout retina. Neuroreport 2000, 11, 3473–3477.
1. Falcon, J.; Gothilf, Y.; Coon, S.L.; Boeuf, G.; Klein, D.C. Genetic, temporal and developmental
differences between melatonin rhythm generating systems in the teleost fish pineal organ and
retina. J. Neuroendocrinol. 2003, 15, 378–382.
2
2
2. Zilberman-Peled, B.; Benhar, I.; Coon, S.L.; Ron, B.; Gothilf, Y. Duality of serotonin-N-
acetyltransferase in the gilthead seabream (Sparus aurata): Molecular cloning and characterization
of recombinant enzymes. Gen. Comp. Endocrinol. 2004, 138, 139–147.
3. Falcon, J.; Besseau, L.; Fuentes, M.; Sauzet, S.; Magnanou, E.; Boeuf, G. Structural and
functional evolution of the pineal melatonin system in vertebrates. Ann. N.Y. Acad. Sci. 2009,
1
163, 101–111.
2
2
4. Aparicio, S. Vertebrate evolution: Recent perspectives from fish. Trends Genet. 2000, 16, 54–56.
5. Volff, J.N. Genome evolution and biodiversity in teleost fish. Heredity 2005, 94, 280–294.

Mar. Drugs 2011, 9
920
2
2
2
6. True, J.R.; Carroll, S.B. Gene co-option in physiological and morphological evolution. Annu. Rev.
Cell Dev. Biol. 2002, 18, 53–80.
7. Falcon, J.; Migaud, H.; Munoz-Cueto, J.A.; Carrillo, M. Current knowledge on the melatonin
system in teleost fish. Gen. Comp. Endocrinol. 2010, 165, 469–482.
8. Benyassi, A.; Schwartz, C.; Coon, S.L.; Klein, D.C.; Falcon, J. Melatonin synthesis:
Arylalkylamine N-acetyltransferases in trout retina and pineal organ are different. Neuroreport
2
000, 11, 255–258.
2
9. Zilberman-Peled, B.; Ron, B.; Gross, A.; Finberg, J.P.; Gothilf, Y. A possible new role for fish
retinal serotonin-N-acetyltransferase-1 (AANAT1): Dopamine metabolism. Brain Res. 2006,
1
073–1074, 220–228.
3
3
3
3
0. Klein, D.C. The 2004 Aschoff/Pittendrigh lecture: Theory of the origin of the pineal gland—A
tale of conflict and resolution. J. Biol. Rhythms. 2004, 19, 264–279.
1. Klein, D.C. Evolution of the vertebrate pineal gland: The AANAT hypothesis. Chronobiol. Int.
2
006, 23, 5–20.
2. Klein, D.C. Arylalkylamine N-acetyltransferase: ―The Timezyme‖. J. Biol. Chem. 2007, 282,
233–4237.
4
3. Appelbaum, L.; Vallone, D.; Anzulovich, A.; Ziv, L.; Tom, M.; Foulkes, N.S.; Gothilf, Y.
Zebrafish arylalkylamine-N-acetyltransferase genes—Targets for regulation of the circadian
clock. J. Mol. Endocrinol. 2006, 36, 337–347.
3
3
3
3
3
3
4
4. Vuilleumier, R.; Boeuf, G.; Fuentes, M.; Gehring, W.J.; Falcon, J. Cloning and early expression
pattern of two melatonin biosynthesis enzymes in the turbot (Scophthalmus maximus). Eur. J.
Neurosci. 2007, 25, 3047–3057.
5. Pavlicek, J.; Coon, S.L.; Ganguly, S.; Weller, J.L.; Hassan, S.A.; Sackett, D.L.; Klein, D.C.
Evidence that proline focuses movement of the floppy loop of arylalkylamine N-acetyltransferase
(EC 2.3.1.87). J. Biol. Chem. 2008, 283, 14552–14558.
6. Vetting, M.W.; de Carvalho, L.P.S.; Yu, M.; Hegde, S.S.; Magnet, S.; Roderick, S.L.; Blanchard,
J.S. Structure and functions of the GNAT superfamily of acetyltransferases. Arch. Biochem.
Biophys. 2005, 433, 212–226.
7. Isorna, E.; El M‘rabet, A.; Confente, F.; Falcon, J.; Munoz-Cueto, J.A. Cloning and expression of
arylalkylamine N-acetyltranferase-2 during early development and metamorphosis in the sole
Solea senegalensis. Gen. Comp. Endocrinol. 2009, 161, 97–102.
8. Shi, Q.; Ando, H.; Coon, S.L.; Sato, S.; Ban, M.; Urano, A. Embryonic and post-embryonic
expression of arylalkylamine N-acetyltransferase and melatonin receptor genes in the eye and
brain of chum salmon (Oncorhynchus keta). Gen. Comp. Endocrinol. 2004, 136, 311–321.
9. Yu, E.Z.; Hallenbeck, J.M.; Cai, D.; McCarron, R.M. Elevated arylalkylamine-N-acetyltransferase
(AA-NAT) gene expression in medial habenular and suprachiasmatic nuclei of hibernating ground
squirrels. Brain Res. Mol. Brain Res. 2002, 102, 9–17.
0. Velarde, E.; Cerda-Reverter, J.M.; Alonso-Gomez, A.L.; Sanchez, E.; Isorna, E.; Delgado, M.J.
Melatonin-synthesizing enzymes in pineal, retina, liver, and gut of the goldfish (Carassius):
mRNA expression pattern and regulation of daily rhythms by lighting conditions. Chronobiol. Int.
2
010, 27, 1178–1201.

Mar. Drugs 2011, 9
921
4
4
1. Riddles, P.W.; Blakeley, R.L.; Zerner, B. Reassessment of Ellman‘s reagent. Meth. Enzymol.
983, 91, 49–60.
1
2. Czaplicki, J.; Cornelissen, G.; Halberg, F. GOSA, a simulated annealing-based program for global
optimization of nonlinear problems, also reveals transyears. J. Appl. Biomed. 2006, 4,
8
7–94.
4
3. Sugden, D.; Namboodiri, M.A.; Weller, J.L.; klein, D.C. Melatonin synthesizing enzymes:
Serotonin N-acetyltransferase and hydroxyindol-O-methyltransferase. In Methods in Biogenic
Amine Research; Parvez, H., Parvez, S., Nagatsu, T., Eds.; Elsevier/North Holland Biomedical
Press: Amsterdam, The Netherlands, 1983; pp. 567–572.
4
4
4. Needleman, S.B.; Wunsch, C.D. A general method applicable to the search for similarities in the
amino acid sequence of two proteins. J. Mol. Biol. 1970, 48, 443–453.
5. Karuskal, J.B. An overview of sequence comparison. In Time Warps, String Edits and
Macromolecules: The Theory and Practice of Sequence Comparison; Sankoff, D., Kruskal, J.B.,
Eds.; Addison Wesley: Boston, MA, USA, 1983; pp. 1–44.
4
6. Petrey, D.; Xiang, Z.; Tang, C.L.; Xie, L.; Gimpelev, M.; Mitros, T.; Soto, C.S.;
Goldsmith-Fischman, S.; Kernytsky, A.; Schlessinger, A.; Koh, I.Y.; Alexov, E.; Honig, B. Using
multiple structure alignments, fast model building, and energetic analysis in fold recognition and
homology modeling. Proteins 2003, 53 (Suppl 6), 430–435.
4
4
7. Laskowski, R.A.; MacArthur, M.W.; Moss, D.S.; Thornton, J.M. PROTHECK: A program to
check the steriochemical quality of protein structures. J. Appl. Cryst. 1993, 26, 283–291.
8. Pieper, U.; Eswar, N.; Braberg, H.; Madhusudhan, M.S.; Davis, F.P.; Stuart, A.C.; Mirkovic, N.;
Rossi, A.; Marti-Renom, M.A.; Fiser, A.; Webb, B.; Greenblatt, D.; Huang, C.C.; Ferrin, T.E.;
Sali, A. MODBASE, a database of annotated comparative protein structure models, and
associated resources. Nucl. Acids Res. 2004, 32, D217–D222.
4
9. Gordon, J.C.; Myers, J.B.; Folta, T.; Shoja, V.; Heath, L.S.; Onufriev, A. H++: A server for
estimating pKas and adding missing hydrogens to macromolecules. Nucleic Acids Res. 2005, 33,
W368–371.
5
5
0. Petrey, D.; Honig, B., GRASP2: Visualization, surface properties, and electrostatics of
macromolecular structures and sequences. Meth. Enzymol. 2003, 374, 492–509.
1. Duhovny, D.; Nussinov, R.; Wolfson, H.J. Efficient unbound docking of rigid molecules. In the
Proceedings of the 2nd Workshop on Algorithms in Bioinformatics (WABI), Rome, Italy, 17–21
September 2002; Guigo, R., Gusfield, D., Eds.; Springer: Berlin, Germany, 2002; Volume 2452.
2. Schneidman-Duhovny, D.; Inbar, Y.; Polak, V.; Shatsky, M.; Halperin, I.; Benyamini, H.;
Barzilai, A.; Dror, O.; Haspel, N.; Nussinov, R.; Wolfson, H.J. Taking geometry to its edge: Fast
unbound rigid (and hinge-bent) docking. Proteins 2003, 52, 107–112.
5
Samples Availability: Available from the authors.
©
2011 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license
http://creativecommons.org/licenses/by/3.0/).
(