Activities of Arabidopsis sinapoylglucose:malate sinapoyltransferase shed light on functional diversification of serine carboxypeptidase-like acyltransferases

Article abstract of DOI:10.1016/j.phytochem.2008.03.021

Analysis of the catalytic properties of the serine carboxypeptidase-like (SCPL) 1-O-sinapoyl-β-glucose:l-malate sinapoyltransferase (SMT) from Arabidopsis showed that the enzyme exhibits besides its primary sinapoylation of l-malate, minor hydrolytic and disproportionation activities, producing free sinapic acid and 1,2-di-O-sinapoyl-β-glucose, respectively. The ability of the enzyme to liberate sinapic acid from the donor molecule 1-O-sinapoyl-β-glucose indicates the existence of a short-lived acylenzyme intermediate in the proposed random sequential bi-bi mechanism of catalysis. SMT-catalyzed formation of disinapoylglucose has been corroborated by docking studies with an established homology structure model that illustrates the possible binding of two 1-O-sinapoyl-β-glucose molecules in the active site and the intermolecular reaction of the two glucose esters. The SMT gene is embedded in a tandem cluster of five SCPL sinapoyltransferase genes, which encode enzymes with high amino acid sequence identities and partially overlapping substrate specificities. We assume that in recent duplications of genes encoding SCPL proteins, neofunctionalization of the duplicates to accept 1-O-sinapoyl-β-glucose as acyl donor was gained first, followed by subfunctionalization leading to different acyl acceptor specificities.

Full text of DOI:10.1016/j.phytochem.2008.03.021

Phytochemistry 69 (2008) 1826–1831
Contents lists available at ScienceDirect
Phytochemistry
journal homepage: www.elsevier.com/locate/phytochem
Activities of Arabidopsis sinapoylglucose:malate sinapoyltransferase shed light
on functional diversification of serine carboxypeptidase-like acyltransferases
Felix Stehle , Wolfgang Brandt , Jürgen Schmidt , Carsten Milkowski a,1, Dieter Strack a,
a
b
b
*
a
Department of Secondary Metabolism, Leibniz Institute of Plant Biochemistry (IPB), Weinberg 3, D-06120 Halle (Saale), Germany
Department of Bioorganic Chemistry, Leibniz Institute of Plant Biochemistry (IPB), Weinberg 3, D-06120 Halle (Saale), Germany
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Analysis of the catalytic properties of the serine carboxypeptidase-like (SCPL) 1-O-sinapoyl-b-glucose:
malate sinapoyltransferase (SMT) from Arabidopsis showed that the enzyme exhibits besides its primary
sinapoylation of -malate, minor hydrolytic and disproportionation activities, producing free sinapic acid
L-
Received 14 January 2008
Received in revised form 25 February 2008
Available online 17 May 2008
L
and 1,2-di-O-sinapoyl-b-glucose, respectively. The ability of the enzyme to liberate sinapic acid from the
donor molecule 1-O-sinapoyl-b-glucose indicates the existence of a short-lived acylenzyme intermediate
in the proposed random sequential bi–bi mechanism of catalysis. SMT-catalyzed formation of disinapoyl-
glucose has been corroborated by docking studies with an established homology structure model that
illustrates the possible binding of two 1-O-sinapoyl-b-glucose molecules in the active site and the inter-
molecular reaction of the two glucose esters. The SMT gene is embedded in a tandem cluster of five SCPL
sinapoyltransferase genes, which encode enzymes with high amino acid sequence identities and partially
overlapping substrate specificities. We assume that in recent duplications of genes encoding SCPL pro-
teins, neofunctionalization of the duplicates to accept 1-O-sinapoyl-b-glucose as acyl donor was gained
first, followed by subfunctionalization leading to different acyl acceptor specificities.
Keywords:
Serine carboxypeptidase-like
acyltransferase
Arabidopsis
Brassicaceae
Gene cluster
Molecular evolution
Homology modeling
Ó 2008 Elsevier Ltd. All rights reserved.
1
. Introduction
Decisive steps in the evolution of genetic systems are gene
1993). An isobutyroyltransferase from wild tomato (Li and Stef-
fens, 2000) and sinapoyltransferases of members of the Brassica-
ceae, i.e. 1-O-sinapoyl-b-glucose:L-malate sinapoyltransferase
duplication and subsequent neo- and subfunctionalization of the
enzymes encoded by gene duplicates (Ohno, 1970; Roth et al.,
(SMT) and 1-O-sinapoyl-b-glucose:choline sinapoyltransferase
(SCT) (Lehfeldt et al., 2000; Shirley et al., 2001; Fraser et al.,
2007; Weier et al., 2008), use the b-acetal esters as acyl donors.
These acyltransferases have sequence similarities to serine carb-
oxypeptidases (SCPs), classifying them as serine carboxypepti-
dase-like (SCPL) proteins. The Arabidopsis genome encodes a
large family of annotated SCPL proteins with a distinguished num-
ber of proteins that cluster with proven SCPL acyltransferases (Mil-
kowski and Strack, 2004; Fraser et al., 2005).
The high overall sequence identity between the SCPL acyltrans-
ferases and their presumed origin from SCPs poses questions on
molecular mechanisms that drive the functional shift from hydro-
lase to acyltransferase activity. First studies with artificial peptides
established no concomitant peptidase activities for glucose ester-
dependent acyltransferases (Li and Steffens, 2000; Shirley and
Chapple, 2003).
The SMT gene in Arabidopsis is embedded in a tandemly ar-
ranged cluster of five SCPL acyltransferase genes, and it has been
discussed that encoded enzymes of this cluster may exhibit
overlapping substrate specificities (Fraser et al., 2007). Analysis
of accumulating sinapate esters in transgenic Arabidopsis
mutants indicated that SMT may also catalyze the formation of
1,2-di-O-sinapoyl-b-glucose (3), which is the defined activity of
2
007). Thus, new enzymes emerge during speciation, often forming
large gene families encoding enzymes with either new stringent
functions or extended substrate specificities. Genetic dynamics is
especially important in the evolution of plant secondary metabo-
lism (Hartmann, 2007). The extraordinary diversity that character-
izes plant chemistry is brought about by gene families coding for
enzymes that modify various structure skeletons of phenylpropa-
noids, isoprenoids, alkaloids or polyketides, amounting to more
than 200,000 compounds so far. Besides hydroxylation, methyla-
tion, glycosylation or prenylation, acylation particularly plays an
important role to gain versatile properties of secondary metabo-
lites. The latter includes reactions with acetate, malonate, malate
or hydroxybenzoates and hydroxycinnamates.
In various plant species, transfer of acyl groups is driven by en-
ergy-rich b-acetal esters (1-O-b-glucose esters) instead of the more
common coenzyme A thioesters (Steffens, 2000; Strack and Mock,
*
Corresponding author. Fax: +49 345 5582 1509.
E-mail address: dieter.strack@ipb-halle.de (D. Strack).
Present address: Institute of General Botany, University Jena, Am Planetarium 1,
1
D-07743 Jena, Germany.
0
031-9422/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.phytochem.2008.03.021

F. Stehle et al. / Phytochemistry 69 (2008) 1826–1831
1827
1
-O-sinapoyl-b-glucose:1-O-sinapoyl-b-glucose sinapoyltransfer-
(HPLC–DAD) two minor products. One of these products, less polar
than sinapoyl- -malate (2), with absorption characteristics of sina-
ase (SST) from Raphanus sativus (Dahlbender and Strack, 1986)
and Arabidopsis (Fraser et al., 2007).
L
pate esters (RT, 25.5 min; kmax, 330 nm), was shown to be identical
with 1,2-di-O-sinapoyl-b-glucose (3) by co-chromatography with
standard compounds and LC–ESI-MS/MS analysis. The data are in
agreement with those of the same compound isolated from R. sat-
ivus (Strack et al., 1984), Brassica napus (Baumert et al., 2005) and
Arabidopsis (Fraser et al., 2007). The second minor reaction prod-
We report herein on the variable activities of the recombinant
Arabidopsis SMT that produces not only sinapoyl-L-malate (2) but
also free sinapic acid (4) and 1,2-di-O-sinapoyl-b-glucose (3). To
substantiate the latter, we performed docking studies applying
our previously established SMT structure model (Stehle et al.,
2
006). The resulting docking arrangements clearly demonstrate
uct was identified as free sinapic acid (4) (RT, 15.4 min; kmax,
the binding of two 1-O-sinapoyl-b-glucose molecules (1) in the ac-
tive site of the enzyme and illustrate the intermolecular reaction
of the two glucose esters to form the disinapoylated product. In
addition, based on hydrolytic activity of the SMT, we postulate
the existence of a short-lived acylenzyme intermediate in a ran-
dom sequential bi–bi mechanism of enzyme catalysis. Acylenzyme
formation corresponds to that of SCPs; however, the SMT kinetics
is in contrast to their double displacement (ping–pong) mecha-
nism. We conclude that the Arabidopsis sinapoyltransferases,
classified as SCPL proteins, originated from more recent gene
duplications and are probably still at a stage of evolutionary
diversification.
324 nm), indicating hydrolytic activity of SMT towards 1-O-sina-
poyl-b-glucose (1).
The negative ion ESI mass spectrum of sinapic acid (4) showed a
À
[MÀH] ion at m/z 223, prominent key ions at m/z 208 ([MÀH–
À
À
À
2
Me] ), 193 ([MÀH–CH
2
O] ), 179 ([MÀH–CO
] ) and a base peak
À
ion at m/z 164 ([MÀH–CO
2
–Me] ). While 1-O-sinapoyl-b-glucose
À
(1) ([MÀH] at m/z 385) displayed a significant loss of 162 mass
À
units (hexose moiety) to form the ion at m/z 223 ([sinapate–H] )
and the base peak ion at m/z 205, the ESI-CID mass spectrum of
À
1,2-di-O-sinapoyl-b-glucose (3) ([MÀH] at m/z 591) is character-
À
ized by key ions at m/z 367 ([MÀH–sinapate] ) and 223.
Fig. 1 illustrates the three activities of the recombinant SMT.
Formation of 1,2-di-O-sinapoyl-b-glucose (3) is the prime activity
of SST, first identified from R. sativus (Dahlbender and Strack,
2
. Results and discussion
.1. SMT mediates SST activity and hydrolysis of sinapoylglucose
In SMT assays lacking the preferred acceptor -malate, we ob-
1
986) and recently also proposed for Arabidopsis (Fraser et al.,
2
2007). The activity of SST was shown to be highest near pH 8.0,
with residual activities of 25% at pH 6.0 (Dahlbender and Strack,
1986). When we measured SMT activity at pH 8.0, formation of
L
served by chromatographic and diode array detection analyses
sinapoyl-L-malate (2) decreased to about 1%, whereas that of
OMe
OH
OMe
O
O
-
O C
Glc
2
CO -
2
2
a
OMe
OH
OMe
OH
+
L-Malate
OMe
Glc
OMe
O
O
O
O
O
O
O
HO
HO
O
HO
HO
+ 1-O-Sinapoyl-ß-glucose (1)
b
OH
OH
OMe
OH
OH
1
+
H O
2
3
OMe
Glc
OMe
c
OH
OMe
HO C
2
4
Fig. 1. Scheme of the SMT-catalyzed reactions. (a) Primary activity of SMT in the formation of sinapoyl-
standard conditions (pH 6.0); (b) formation of 1,2-di-O-sinapoyl-b-glucose (3) in the absence or weak binding of
standard conditions to liberate sinapic acid (4), strongly increasing in the absence of -malate and at pH 8.0 (see Table 1).
L
-malate (2) with 1-O-sinapoyl-b-glucose (1) and
L-malate under
L-malate at pH 8.0; (c) minor hydrolytic SMT activity under
L

1
828
F. Stehle et al. / Phytochemistry 69 (2008) 1826–1831
1
,2-di-O-sinapoyl-b-glucose (3) increased, along with free sinapic
acid (4). Table 1 summarizes the different SMT activities at pHs
.0 and 8.0 with and without -malate, using the recombinant pro-
tein produced in Saccharomyces cerevisiae cultures.
tion possibly pave the way to experimentally access structure–
function relationships of the Arabidopsis SMT.
We performed docking studies with our homology structure
model to simulate the possible binding of a second molecule of
6
L
1
-O-sinapoyl-b-glucose (1). These analyses suggest that there are
no marked alterations in the protein structure compared to the
structure model with bound -malate (Fig. 2A). The docking
2.2. The SMT structure model supports SST activity
L
The unexpected SMT-catalyzed formation of 1,2-di-O-sinapoyl-
b-glucose (3) prompted us to examine this reaction using our pre-
viously developed Arabidopsis SMT structure model (Stehle et al.,
006). Although we recently established a high yield expression
system for heterologous production of active SMT (Stehle et al.,
008), we were not able yet to produce this protein in amounts
that would facilitate crystallographic approaches for protein struc-
ture elucidation. However, further studies on expression optimiza-
arrangement with the highest fitness-value of the second 1-O-sina-
poyl-b-glucose (1), located close to the first molecule, is shown in
Fig. 2B. The binding of the acyl donor is not affected and that of the
acceptor molecule is achieved by the formation of only two hydro-
2
gen bonds, instead of five bonds with L-malate to Lys268, His272,
2
Asp278 and Arg322. The first bond with the acceptor 1-O-sina-
poyl-b-glucose (1) is formed between Gly75 and the C3 hydroxyl
group of the glucose moiety and the second between the e-nitro-
gen of Arg322 and the C3 methoxyl group of the sinapoyl moiety.
This weak binding of the acyl acceptor compared to the strong fix-
Table 1
À1
ation of
tionation reactions of the SMT are detectable only in the absence
of -malate.
When the acylenzyme is formed, the second 1-O-sinapoyl-b-
L-malate may explain why measurable in vitro dispropor-
SMT activities (pkat mg protein; mean ± s.d.) at pHs 6.0 and 8.0 with and without
L-malate, catalyzing the formation of sinapoyl-L-malate (2), sinapic acid (4) and
,2-di-O-sinapoyl-b-glucose (3)
1
L
Substrates
Products
glucose molecule (1) slightly changes the docking arrangement in
the active site due to more available space (Fig. 2B). Then the
acceptor molecule is stabilized by p–p-interactions between the
hydrophobic benzyl ring and the imidazol ring of His272. Further
stabilization of the sinapoyl moiety is achieved by hydrogen bonds
between the hydroxyl and methoxyl groups and the amino acid
residues Thr412 and Glu414, respectively. Arg322 forms hydrogen
bonds to the carbonyl oxygen, while the glucose moiety is still
fixed at Gly75. This arrangement enables the catalytic His411 to
abstract a proton from the C2 hydroxyl group. As a result, the
activated acceptor 1-O-sinapoyl-b-glucose (1) is in an optimal po-
Sinapoylglucose
Malate Sinapoylmalate
Sinapic acid
(4)
Disinapoylglucose
(3)
(1)
(2)
pH 6.0
+
+
+
À
23,069 ± 129
–
189 ± 37
2268 ± 18
–
27 ± 4
pH 8.0
+
+
+
À
185 ± 14
–
1934 ± 5
1903 ± 40
173 ± 4
186 ± 1
Presence and absence of compounds is marked by ‘‘+” and ‘‘À”, respectively.
Minor amounts of non-enzymatically liberated sinapic acid (4) in control assays
containing heat-denatured enzyme was taken into consideration.
Fig. 2. Homology-based structure model of the active site of the SMT with the docked acyl donor (orange) and acyl acceptors (yellow). The residues of the catalytic triad
Ser173, His411 and Asp358 are highlighted in green. Dotted lines indicate hydrogen bonds. (A) SMT with the two docked substrates 1-O-sinapoyl-b-glucose (1) and -malate
left) and the corresponding acylenzyme complex (right) (after Stehle et al., 2006); (B) SMT with the two docked 1-O-sinapoyl-b-glucose molecules (1) (left) and the
corresponding acylenzyme complex (right). For better visualization of hydrogen bonds, a different view is shown for the disproportionation reaction.
L
(

F. Stehle et al. / Phytochemistry 69 (2008) 1826–1831
1829
sition to attack the acylenzyme complex to form 1,2-di-O-sinapoyl-
b-glucose (3).
the enzyme, demanding the presence of both substrates in the ac-
tive site that is not accomplished when water displaces the acyl
acceptor. However, the observation of the hydrolytic reaction of
SMT most likely indicates the existence of a short-lived acylen-
zyme intermediate that can be attacked by a water molecule.
2.3. SMT employs a short-lived acylenzyme complex
It was previously shown that the SMT reaction follows a ran-
We achieved probably weak binding of L-malate at pH 8.0, the
dom sequential bi–bi mechanism (Stehle et al., 2008). Both sub-
strates have to be present in the active site before the reaction is
initiated. However, in the SMT-mediated SST reaction the displace-
ment of the weakly bound second 1-O-sinapoyl-b-glucose mole-
cule (1) by water is facilitated. This leads to hydrolysis resulting
in an increased liberation of sinapic acid (4) (Table 1) that points
to a pivotal question concerning our model of the SMT kinetic
mechanism. So far, based on modeling studies and kinetic analysis
pH optimum determined for SST activity from R. sativus (Dahlb-
ender and Strack, 1986), and observed the formation of 1,2-di-O-
sinapoyl-b-glucose (3) and sinapic acid (4) in appreciable amounts,
along with sinapoyl-L-malate (2). At this pH value L-malate is
mostly deprotonated and therefore His272 in the structure with
the two docked substrates and Asp172 in the acylenzyme complex
are no longer able to stabilize the acceptor by hydrogen bonds
(Fig. 2A). This assumption was supported by an enzymatic activity
À1
of sinapoyl-L-malate (2) formation (Stehle et al., 2006, 2008), we
of 185 pkat mg protein for sinapoyl-
L-malate (2) formation, indi-
were not able to distinguish clearly between two conceivable
modes for the SMT kinetic mechanism – including an acylenzyme
complex (sinapoylseryl-SMT), characteristic for the serine carboxy-
peptidases (Douglas et al., 1976), or the catalytic serine acting as a
base, described for a C–C hydrolase (Fleming et al., 2000). In stan-
cating a strong decrease to a residual activity value below 1% com-
pared to the activity at pH 6.0, whereas the hydrolytic activity
increased about 12-fold (Table 1). Additionally, the SMT catalyzed
the disproportionation reaction of two 1-O-sinapoyl-b-glucose
À1
molecules (1) resulting in an activity of 170 pkat mg protein.
dard enzyme assays (pH 6.0), containing
L-malate as second sub-
In enzyme assays lacking L-malate, hydrolase activity was re-
strate, only minute amounts of liberated sinapic acid (4) are
detectable. Thus, water is probably not recognized as acyl acceptor
and unable to initiate the reaction. Hence it follows that if the SMT
activity would depend on a general base mechanism, hydrolysis
should not proceed, given the sequential reaction mechanism of
mained at the same level at both pH values, whereas the formation
of 1,2-di-O-sinapoyl-b-glucose (3) at pH 8.0 increased about seven-
fold, comparable to the activity of sinapoyl-
at pH 8.0. These results show that SMT specificity towards
is favoured by more acidic pH values (optimum at pH 6.0), which is
L-malate (2) formation
L-malate
Fig. 3. Alignment of amino acid sequences of the Arabidopsis SCPL acyltransferases, encoded by the SCPL gene cluster on chromosome 2. The cDNA-encoding amino acid
sequence of SMT was aligned with the sequences of the identified SAT and SST (Fraser et al., 2007) and the two putative acyltransferases encoded by At2g22970 and
At2g22980 by CLUSTAL W using the BLOSUM 62 matrix (Henikoff and Henikoff, 1992, 1993). The N-terminal leader sequences have been removed. Fully conserved residues
are shaded in black and those in grey indicate conservation of at least 50%. Residues shaded in different colours are part of the reaction mechanism and substrate recognition
that was found out for the SMT (Stehle et al., 2006); the catalytic triad in red, the oxyanion hole in green, recognition of the sinapoyl moiety of the donor molecule in blue, the
hydrogen bond network in yellow and the primary L-malate recognition in the SMT in magenta.

1
830
F. Stehle et al. / Phytochemistry 69 (2008) 1826–1831
consistent with SMT localization to the vacuoles of Arabidopsis
leaves (Hause et al., 2002). Nevertheless, SMT does not exhibit
strict substrate specificity in planta, which was shown in analysis
of an sng1-6 transgenic line, harbouring the SMT gene (Fraser
et al., 2007), that still accumulated 1,2-di-O-sinapoyl-b-glucose (3).
members. Nevertheless, the present results establish the molecular
basis required for gaining further insights into acyl donor and
acceptor specificities for SCPL acyltransferases.
3. Concluding remarks
2
.4. SCPL sinapoyltransferases are still on evolutionary diversification
So far the ability to catalyze hydrolysis is only proven for the
SMT, but it is most likely, that other SCPL acyltransferases retained
functionally redundant hydrolytic activities. Interestingly, en-
zymes with bifunctional activities, transacylation and hydrolysis,
have also been described from other systems. Kowalczyk et al.
(2003) showed that indole-3-acetyl transferase from Zea mays cat-
alyzes both synthesis and hydrolysis of indole-3-acetylinositol. In
S. cerevisiae two acyl-coenzyme A:ethanol O-acyltransferases,
Eht1 and Eeb1, with an a/b-hydrolase fold and a catalytic triad con-
sisting of the amino acid residues Ser–Asp–His exhibit acyltrans-
ferase as well as esterase activities (Saerens et al., 2006).
Therefore, the change from a hydrolytic a/b-hydrolase enzyme to
an acyltransferase seems to be a common phenomenon.
The SMT gene is one of five SCPL acyltransferase genes on Ara-
bidopsis chromosome 2. Of these genes, three have been consid-
ered as potentially mediating SST activity in planta (Fraser et al.,
2
tein encoded by At2g23010, minor SST activities have also been as-
cribed to the At2g22980 gene product and to SMT (At2g22990). For
the latter, in vitro activity has been proven in a recent biochemical
approach (Stehle et al., 2008).
In a homology-based SMT structure model, amino acid residues
of the active site could be identified. By site-directed mutagensis
these were proven to play an important role in enzyme catalysis
007). While a major SST activity has been postulated for the pro-
(
1
Stehle et al., 2006). Particularly, the residues for recognition of
-O-sinapoyl-b-glucose (1) as well as those forming the hydrogen
The different enzymatic activities of SMT suggest that evolution
of this enzyme has not yet reached strict substrate specificity and
loss of hydrolytic properties. We assume that SMT and the other
four SCPL acyltransferase genes on Arabidopsis chromosome 2
originated from more recent gene duplications and might be still
at a stage of evolutionary diversification. Because of the high se-
quence similarities of all five proteins (Fig. 3), encoded by tan-
demly arranged genes, and the fact that the enzymes all use 1-O-
sinapoyl-b-glucose (1) and exhibit partially overlapping acyl
acceptor specificities, it is likely that neofunctionalization of the
duplicates towards acyl donor specificity was gained first, followed
by subfunctionalization resulting in specificities for different acyl
acceptors.
bond network and the oxyanion hole are conserved in the SMT-re-
lated enzymes SST and SAT (1-O-sinapoyl-b-glucose:anthocyanin
acyltransferase) (Fig. 3). Most interestingly, these enzymes display
at a conserved position the functionally important residue Asn73
(
SMT), involved in binding and approximation of both 1-O-sina-
poyl-b-glucose (1) and -malate. Among the functionally important
amino acid residues, only the SMT-Asn73 homolog, part of the
hydrogen bond network, has changed to Asp in the SAT enzyme.
L
In contrast, the residues that potentially play a role for L-malate
recognition of SMT (Stehle et al., 2006) are neither completely
conserved in the two related acyltransferases, SAT and SST, nor in
the two putative acyltransferases, encoded by At2g22970 and
At2g22980 (Fraser et al., 2007). At2g22980 was assumed to encode
an enzyme with minor SST activity, but the function of At2g22970
protein is unknown. The amino acid sequence derived from the
At2g22970 ORF carried the conserved residues Lys268 (SMT),
4. Experimental
4.1. Enzyme assays
Asp278 (SMT) and Arg322 (SMT) involved in
L
-malate recognition
The SMT protein used in this study has been produced in S. cere-
visiae cultures, transformed with the Arabidopsis SMT cDNA that
has been optimized for high expression levels (Stehle et al.,
2008). The enzyme assays with the recombinant protein contained
1 mM 1-O-sinapoyl-b-glucose (1) in the presence or absence of
by SMT. Moreover, the At2g22970 sequence contains the conserved
residues Trp71 (SMT) and Asp172 (SMT) that form together with
Asn73 (SMT) the hydrogen bond network in SMT. In At2g22970,
however, the Asn73 (SMT) homolog has changed to a Ser residue.
In a current study on molecular evolution of the SCPL acyltrans-
ferase genes in Arabidopsis, we cloned the At2g22970 cDNA and
10 mM L-malate in a total volume of 100 ll 100 mM MES buffer
(pH 6.0) or 100 mM Tris–HCl (pH 8.0), both containing 5% (v/v)
DMSO. After incubation at 30 °C at varying time intervals, the reac-
tion was terminated by adding TFA to a final concentration of 10%
(v/v). Product formation was analyzed by HPLC. After centrifuga-
tion, assay aliquots were injected onto a Nucleosil C18 column
(5 lm; 250 Â 4 mm i.d.; Macherey-Nagel, Düren, Germany). Sepa-
were able to prove the formation of sinapoyl-
L-malate (2) from
1
-O-sinapoyl-b-glucose (1) and -malate in activity assays contain-
L
ing the recombinant At2g22970 protein, when we specifically re-
placed some amino acids with those implicated in the SMT-like
reaction (not documented). The mutant variants were generated
based on the alignment of this protein with the SMT amino acid se-
quence (Fig. 3) and a model structure using SMT as template pro-
tein (Stehle et al., 2006). These results clearly classify this protein
as another SCPL sinapoyltransferase, but a targeted metabolomics
approach on Arabidopsis mutants and detailed biochemical work
is needed to understand the nature of this enzyme.
The multiple alignments in Fig. 3 suggests that the most diverse
sequence positions are located near the residues involved in acyl
acceptor recognition, indicating that this region may be a target
for evolution-driven structural alterations without changing the
chemical identity of the functional residues. Interestingly, these
amino acid residues are also highly conserved in the related
SCT – except of those for acyl acceptor recognition (Stehle et al.,
ration was achieved using a 40-min linear gradient at a flow rate
À1
of 1 ml min from 10% to 40% MeCN in 1.5% aq. H
3
PO
4
. Com-
pounds were photometrically detected (maxplot between 210
and 500 nm) by a Waters 2996 photodiode array detector. Co-chro-
matography was achieved by adding standard compounds (Baum-
ert et al., 2005) to the terminated enzyme assay before HPLC
analysis. The quantitative data, calculated as equivalents of 1-O-
sinapoyl-b-glucose (1), represent mean values (±s.d.) from three
independent measurements.
4.2. LC–ESI-MS/MS
The negative ion ESI mass spectra of sinapic acid (4), 1-O-sina-
poyl-b-glucose (1) and 1,2-di-O-sinapoyl-b-glucose (3) were ob-
tained from a Finnigan MAT TSQ Quantum Ultra AM system
equipped with a hot ESI source (HESI), electrospray voltage
3.0 kV, sheath gas: nitrogen; vaporizer temperature: 50 °C;
2
006). This is remarkable, since the SCT clusters in the SCPL acyl-
transferase family (Milkowski and Strack, 2004; Fraser et al.,
005) but a post-translational endoproteolysis during enzyme
maturation distinguished this protein from the other group
2

F. Stehle et al. / Phytochemistry 69 (2008) 1826–1831
1831
capillary temperature: 250 °C. The MS system is coupled with a
Surveyor Plus micro-HPLC (Thermo Electron), equipped with a
Ultrasep ES RP18E column (5 lm, 1 Â 100 mm, SepServ). HPLC sep-
aration was achieved by using a 15-min linear gradient at a flow
Douglas, K.T., Nakagawa, Y., Kaiser, E.T., 1976. Mechanistic studies of
carboxypeptidase Y. Kinetic detection of an acyl-enzyme intermediate in
trimethylacetate esterase action. J. Am. Chem. Soc. 98, 8231–8236.
Fleming, S.M., Robertson, T.A., Langley, G.J., Bugg, T.D., 2000. Catalytic mechanism of
a C-C hydrolase enzyme: evidence for a gem-diol intermediate, not an acyl
enzyme. Biochemistry 39, 1522–1531.
Fraser, C.M., Rider, L.W., Chapple, C., 2005. An expression and bioinformatics
analysis of the Arabidopsis serine carboxypeptidase-like gene family. Plant
Physiol. 138, 1136–1148.
Fraser, C.M., Thompson, M.G., Shirley, A.M., Ralph, J., Schoenherr, J.A., Sinlapadech,
T., Hall, M.C., Chapple, C., 2007. Related Arabidopsis serine carboxypeptidase-
like sinapoylglucose acyltransferases display distinct but overlapping substrate
specificities. Plant Physiol. 144, 1986–1999.
À1
rate of 50 ll min from 10% to 95% MeCN in 0.2 % aq. HOAc, with
the latter held at 95% MeCN for another 15 min. The collision-in-
duced dissociation (CID) mass spectra of sinapic acid (4) and 1-
O-sinapoyl-b-glucose (1) were recorded during the HPLC run with
À
a collision energy of 15 eV for the [MÀH] -ions at m/z 223 (4) and
3
85 (1) by using a skimmer voltage of 25 eV, respectively. For 1,2-
GOLD, Cambridge Crystallographic Data Centre (CCDC), 1998. Available from:
di-O-sinapoyl-b-glucose (3) a collision energy of 20 eV was used
<
http://www.ccdc.cam.ac.uk/>.
(
collision gas: argon; collision pressure: 1.5 mTorr).
Hartmann, T., 2007. From waste products to ecochemicals: fifty years research of
plant secondary metabolism. Phytochemistry 68, 2831–2846.
Hause, B., Meyer, K., Viitanen, P.V., Chapple, C., Strack, D., 2002. Immunolocalization
of 1-O-sinapoylglucose:malate sinapoyltransferase in Arabidopsis thaliana.
Planta 215, 26–32.
Henikoff, S., Henikoff, J.G., 1992. Amino acid substitution matrices from protein
blocks. Proc. Natl. Acad. Sci. USA 89, 10915–10919.
Henikoff, S., Henikoff, J.G., 1993. Performance evaluation of amino acid substitution
matrices. Proteins 17, 49–61.
Sinapic acid (4): RTHPLC = 11.7 min, ESI-CID mass spectrum (m/z,
À
rel. int. (%)): 223 ([MÀH] , 41), 208 (97), 193 (39), 179 (12), 164
(
100), 149 (77), 121 (15).
1
-O-Sinapoyl-b-glucose (1): RTHPLC = 4.6 min, ESI-CID mass spec-
À
trum (m/z, rel. int. (%)): 385 ([MÀH] , 18), 265 (3), 247 (10), 223
(
24), 205 (100), 190 (22).
1
,2-Di-O-sinapoyl-b-glucose (3): RTHPLC = 15.2, ESI-CID mass
Kowalczyk, S., Jakubowska, A., Zielínska, E., Bandurski, R.S., 2003. Bifunctional
indole-3-acetyl transferase catalyses synthesis and hydrolysis of indole-3-
acetyl-myo-inositol in immature endosperm of Zea mays. Physiol. Plant. 119,
À
spectrum (m/z, rel. int. (%)): 591 ([MÀH] , 14), 367 ([MÀH–sina-
À
À
pate] , 84), 223 ([sinapate–H] , 100), 205 (6).
165–174.
Lehfeldt, C., Shirley, A.M., Meyer, K., Ruegger, M.O., Cusumano, J.C., Viitanen, P.V.,
4.3. Sequence analysis
Strack, D., Chapple, C., 2000. Cloning of the SNG1 gene of Arabidopsis reveals a
role for
a serine carboxypeptidase-like protein as an acyltransferase in
secondary metabolism. Plant Cell 12, 1295–1306.
The multiple sequence alignment was generated with the pro-
gram CLUSTAL W (Thompson et al., 1994) using the BLOSUM 62
matrix (Henikoff and Henikoff, 1992, 1993).
Li, A.X., Steffens, J.C., 2000. An acyltransferase catalyzing the formation of
diacylglucose is a serine carboxypeptidase-like protein. Proc. Natl. Acad. Sci.
USA 97, 6902–6907.
Milkowski, C., Strack, D., 2004. Serine carboxypeptidase-like acyltransferases.
Phytochemistry 65, 517–524.
Ohno, S., 1970. Evolution by Gene Duplication. Springer-Verlag, New York.
Roth, C., Rastogi, S., Arvestad, L., Dittmar, K., Light, S., Ekman, D., Liberles, D.A., 2007.
Evolution after gene duplication: models, mechanisms, sequences, systems, and
organisms. J. Exp. Zool. B: Mol. Dev. Evol. 308, 58–73.
Saerens, S.M., Verstrepen, K.J., Van Laere, S.D., Voet, A.R., Van Dijck, P., Delvaux, F.R.,
Thevelein, J.M., 2006. The Saccharomyces cerevisiae EHT1 and EEB1 genes encode
novel enzymes with medium-chain fatty acid ethyl ester synthesis and
hydrolysis capacity. J. Biol. Chem. 281, 4446–4456.
4
.4. Docking studies
Based on the recently published SMT homology model
(
pdb = 2drf; Stehle et al., 2006), docking studies were preformed
to predict putative binding of a second 1-O-sinapoyl-b-glucose
molecule (1). For this purpose, -malate at the active site has been
L
replaced by 1-O-sinapoyl-b-glucose (1). Subsequently, two differ-
Shirley, A.M., Chapple, C., 2003. Biochemical characterization of sinapoylglucose:
ent docking investigations using the automatic docking function
choline sinapoyltransferase,
functions as an acyltransferase in plant secondary metabolism. J. Biol. Chem.
78, 19870–19877.
a serine carboxypeptidase-like protein that
of the program GOLD (GGOLDÓ Genetic Optimized Ligand Docking,
2
Cambridge Crystallographic Data Center, 1998, Cambridge, UK)
were carried out. The first one was done with 1-O-sinapoyl-b-glu-
cose (1) as acyl donor bound to the enzyme as in the original mod-
el. The second docking run was performed with the acylenzyme
intermediate, where the donor sinapoyl moiety is covalently linked
to the catalytic active Ser173. In each case, 30 different docking
arrangements were produced.
Shirley, A.M., McMichael, C.M., Chapple, C., 2001. The sng2 mutant of Arabidopsis is
defective in the gene encoding the serine carboxypeptidase-like protein
sinapoylglucose:choline sinapoyltransferase. Plant J. 28, 83–94.
Steffens, J.C., 2000. Acyltransferases in protease’s clothing. Plant Cell 12, 1253–
1256.
Stehle, F., Brandt, W., Milkowski, C., Strack, D., 2006. Structure determinants and
substrate recognition of serine carboxypeptidase-like acyltransferases from
plant secondary metabolism. FEBS Lett. 580, 6366–6374.
Stehle, F., Stubbs, M.T., Strack, D., Milkowski, C., 2008. Heterologous expression of a
serine carboxypeptidase-like acyltransferase and characterization of the kinetic
mechanism. FEBS J. 275, 775–787.
Strack, D., Dahlbender, B., Grotjahn, L., Wray, V., 1984. 1,2-Disinapoylglucose
accumulated in cotyledons of dark-grown Raphanus sativus seedlings.
Phytochemistry 23, 657–659.
Strack, D., Mock, H.-P., 1993. Hydroxycinnamic acids and lignins. In: Dey, P.M.,
Harborne, J.B. (Eds.), Methods in Plant Biochemistry, vol. 9. Academic Press,
New York, NY, pp. 45–97.
Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W: improving the
sensitivity of progressive multiple sequence alignment through sequence
weighting, position-specific gap penalties and weight matrix choice. Nucleic
Acid Res. 22, 4673–4680.
Weier, D., Mittasch, J., Strack, D., Milkowski, C., 2008. The genes BnSCT1 and
BnSCT2 from Brassica napus encoding the final enzyme of sinapine
Acknowledgement
Research on SCPL acyltransferases is supported by the DFG pri-
ority program 1152, ‘‘Evolution of Metabolic Diversity”.
References
Baumert, A., Milkowski, C., Schmidt, J., Nimtz, M., Wray, V., Strack, D., 2005.
Formation of a complex pattern of sinapate esters in Brassica napus seeds,
catalyzed by enzymes of a serine carboxypeptidase-like acyltransferase family?
Phytochemistry 66, 1334–1345.
Dahlbender, B., Strack, D., 1986. Purification and properties of 1-(hydroxy-
biosynthesis
375–385.
– molecular characterization and suppression. Planta 227,
cinnamoyl)-glucose:1-(hydroxycinnamoyl)-glucose
hydroxycinnamoyl-
transferase from radish seedlings. Phytochemistry 25, 1043–1046.