A single amino acid determines the catalytic efficiency of two alkenal double bond reductases produced by the liverwort Plagiochasma appendiculatum

Article abstract of DOI:10.1016/j.febslet.2013.07.051

Alkenal double bond reductases (DBRs) catalyze the NADPH-dependent reduction of the α,β-unsaturated double bond of many secondary metabolites. Two alkenal double bond reductase genes PaDBR1 and PaDBR2 were isolated from the liverwort species Plagiochasma appendiculatum. Recombinant PaDBR2 protein had a higher catalytic activity than PaDBR1 with respect to the reduction of the double bond present in hydroxycinnamyl aldehydes. The residue at position 56 appeared to be responsible for this difference in enzyme activity. The functionality of a C56 to Y56 mutation in PaDBR1 was similar to that of PaDBR2. Further site-directed mutagenesis and structural modeling suggested that the phenol ring stacking between this residue and the substrate was an important determinant of catalytic efficiency.

Full text of DOI:10.1016/j.febslet.2013.07.051

FEBS Letters 587 (2013) 3122–3128
journal homepage: www.FEBSLetters.org
A single amino acid determines the catalytic efficiency of two alkenal
double bond reductases produced by the liverwort Plagiochasma
appendiculatum
Yifeng Wu a, Yuanheng Cai b, Yi Sun a, Ruixue Xu a, Haina Yu a, Xiaojuan Han a, Hongxiang Lou a,
Aixia Cheng a,
⇑
a Key Laboratory of Chemical Biology of Natural Products, Ministry of Education, School of Pharmaceutical Sciences, Shandong University, Jinan 250012, China
b Biosciences Department, Brookhaven National Laboratory, Upton, NY 11973, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Alkenal double bond reductases (DBRs) catalyze the NADPH-dependent reduction of the a,b-unsat-
Received 1 June 2013
Revised 29 July 2013
Accepted 31 July 2013
Available online 13 August 2013
urated double bond of many secondary metabolites. Two alkenal double bond reductase genes PaD-
BR1 and PaDBR2 were isolated from the liverwort species Plagiochasma appendiculatum.
Recombinant PaDBR2 protein had a higher catalytic activity than PaDBR1 with respect to the reduc-
tion of the double bond present in hydroxycinnamyl aldehydes. The residue at position 56 appeared
to be responsible for this difference in enzyme activity. The functionality of a C56 to Y56 mutation in
PaDBR1 was similar to that of PaDBR2. Further site-directed mutagenesis and structural modeling
suggested that the phenol ring stacking between this residue and the substrate was an important
determinant of catalytic efficiency.
Edited by Miguel De la Rosa
Keywords:
Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
Alkenal double bond reductase;
Hydroxycinnamyl aldehyde;
Molecular modeling;
Site-directed mutagenesis;
Plagiochasma appendiculatum
1. Introduction
been isolated. The product of the Pinus taeda gene PtPPDBR cata-
lyzes the NADPH-dependent reduction of the a,b-unsaturated dou-
The plant phenylpropanoids phenylpropenes, coumarins, lign-
ble bond of phenylpropenal aldehydes [16]. Its Arabidopsis thaliana
homologue AtDBR1 (At5g16970) converts p-coumaryl aldehyde
and coniferyl aldehyde into their corresponding dihydrophenyl-
propanols [4]. The tobacco flavin-free double bond reductase
NtDBR shows both regional and stereo-specificity against a variety
ans and lignin are all derived from the deamination of
L-phenylal-
anine by -phenylalanine ammonia lyase. Many phenylpropanoids
L
exhibit broad spectrum antimicrobial activity [1,2] and some are
known to act as signaling molecules both in the defense response
and in the course of development [3]. Dehydrodiconiferyl deriva-
tives are well-known as potential modulators of plant cell division
[3]. The phenylpropanoid pathway can also generate many deriva-
tives with extensive medicinal/health protecting properties, such
as podophyllotoxin, matairesinol and secoisolariciresinol [4]. Dihy-
droconiferyl alcohol is a potentially useful anti-inflammatory agent
and phlorizin shows considerable promise for treatment of diatetes
mellitus [5–7]. Several oxidoreductases are involved in phenyl-
propanoid metabolism, i.e. secoisolariciresinol dehydrogenase [8–
10], phenylcoumaran benzylic ether reductase [11,12], cinnamyl
alcohol dehydrogenase [13,14] and alkenal double bond reductase
[4,15]. One of these, alkenal double bond reductase, reduces the
of a,b-unsaturated compounds with different activating group and
substitution patterns [15]. The above enzymes all belong to the
zinc-independent, medium chain dehydrogenase/reductase
(MDR) superfamily, and share a conserved GXXS motif, known to
stabilize both the adenine and nicotinamide moieties of NADPH,
along with a glycine-rich motif (either AXXGXXG or GXXGXXG),
known to participate in the enzyme’s binding with the NAD(P)+
or NAD(P)H pyrophosphate [17].
Here we report the isolation from the liverwort species Plagi-
ochasma appendiculatum of two genes closely related in sequence
to NtDBR, AtDBR1 and PtPPDBR. Although their five prime (50-
UTR) and three prime (30-UTR) untranslated regions share only
26.3% and 22.7% identity, their open reading frame (ORF) se-
quences differ by just two residues. Their gene products showed
distinct substrate preferences and a non-identical catalytic effi-
ciency towards hydroxycinnamyl aldehydes. Site-directed muta-
genesis was used to demonstrate that a mutation to one of the
double bond present in a variety of
a,b-unsaturated aldehydes
and ketones. Genes encoding several reductases of this type have
⇑
Corresponding author. Fax: +86 531 88382019.
E-mail address: aixiacheng@sdu.edu.cn (A. Cheng).
0014-5793/$36.00 Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.febslet.2013.07.051

Y. Wu et al. / FEBS Letters 587 (2013) 3122–3128
3123
two coding region polymorphisms was sufficient to swap function-
ality, and it was further shown that the phenol ring stacking inter-
action between the aromatic residue and the bound p-coumaryl
aldehyde was responsible for the difference in catalytic rates be-
tween the two homologs.
(sequences lodged with GenBank under accession numbers
KF051271 and KF051272). Their deduced polypeptide sequences
were aligned with those of various plant double bond reductases
using DNAMAN software (Version 5.2.2, Lynnon Biosoft, Canada),
and a phylogenetic tree based on the neighbor-joining method
was constructed with the help of MEGA 4.0 software [19]. Homol-
ogy models of the two gene products were generated using the
Swiss-model server (http://swissmodel.expasy.org), and predicted
docking of the ligand and substrate into the active cavity was ob-
tained by applying AutoDock vina [20]. The resulting models were
visualized using PyMOL (http://www.pymol.org/citing).
2. Materials and methods
2.1. Chemicals and reagents
The synthesis of p-coumaryl aldehyde, p-coumaryl alcohol, caf-
feyl aldehyde, caffeyl alcohol, 5-hydroxyconiferyl aldehyde and 5-
hydroxyconiferyl alcohol followed a published procedure [14].
Dihydro-p-coumaryl aldehyde, dihydrocaffeyl aldehyde, dihydro-
coniferyl aldehyde, dihydro-5-hydroxyconiferyl aldehyde and
dihydrosinapyl aldehyde were all synthesized from their unsatu-
rated form by reduction with hydrogen in the presence of Pd/C.
The purity and identity of all the synthesized reagents was vali-
dated using 1H NMR. Trans-coumaryl aldehyde, trans-4-hydroxy-
cinnamic acid and 3,4-dihydroxycinnamic acid were purchased
from Alfa Aesar (Heysham, UK). All the other reagents and solvents
used were purchased from Sigma–Aldrich (St. Louis, USA).
2.3. Recombinant protein expression and purification
The PaDBR1 and PaDBR2 ORFs were amplified from the two
cDNA clones using, respectively, primer pairs PaDBR1 F/R and PaD-
BR2 F/R (Suppl. Table 1). Each of the two resulting amplicons was
digested with restriction enzymes (Takara, Japan) for ligation into
the corresponding cloning site of the pET32a vector (Novagen).
After confirmation by sequencing, the constructs were transferred
into Escherichia coli BL21 (DE3) to allow the heterologous expres-
sion of N-terminally His-tagged recombinant PaDBR1 and PaDBR2
proteins. Expression was induced by the addition of 0.5 mM iso-
propyl b- -1-thiogalactopyranoside and incubation at 18 °C for
D
2.2. cDNA cloning, sequence alignment and analysis, and protein
modelling
18 h, and the recombinant proteins were purified by passing a cell
extract through a Ni–NTA Sefinose His-bind column, according to
manufacturer’s recommendations (Bio Basic, Canada). The buffer
was exchanged by passing the eluate through an Ultrafiltration
A P. appendiculatum thallus cDNA library [18] was searched to
identify two DBR-like sequences, designated PaDBR1 and PaDBR2
PaDBR1 MAG...TEVTNRQILFKNFVVGWPQESDMEFVTTKKKLELREGQKDVIVRVLYLSCDPCQRGRMRNSTD.
PaDBR2 MAG...TEVTNRQILFKNFVVGWPQESDMEFVTTKKKLELREGQKDVIVRVLYLSCDPYQRGRMRNSTD.
AtDBR1 MTA......TNKQVILKDYVSGFPTESDFDFTTTTVELRVPEGTNSVLVKNLYLSCDPYMRIRMGKPDPS
66
66
64
65
70
NtDBR
MAE....EVSNKQVILKNYVTGYPKESDMEIKNVTIKLKVPEGSNDVVVKNLYLSCDPYMRSRMRKIEG.
RiRZS1 MASGGEMQVSNKQVIFRDYVTGFPKESDMELTTRSITLKLPQGSTGLLLKNLYLSCDPYMRARMTNHHRL
*
PaDBR1 ..SYIPPYTPGKPIEGYGLGKVILSSDPAFNEGDIVSGLMGWEDYSYGF....RLVKIDDLALPISYYLG
PaDBR2 ..SYIPPYTPGKPIEGYGLGKVILSSDPAFNEGDIVSGLMGWEDYSYGF....RLVKIDDLALPISYYLG
AtDBR1 TAALAQAYTPGQPIQGYGVSRIIESGHPDYKKGDLLWGIVAWEEYSVITPMTHAHFKIQHTDVPLSYYTG
130
130
134
132
137
NtDBR
..SYVESFAPGSPITGYGVAKVLESGDPKFQKGDLVWGMTGWEEYSIIT.PTQTLFKIHDKDVPLSYYTG
RiRZS1 ..SYVDSFKPGSPIIGYGVARVLESGNPKFNPGDLVWGFTGWEEYSVIT.ATESLFKIHNTDVPLSYYTG
*
PaDBR1 ALGMAGFTAYVGFFRICEPKKGDTIYVSAASGAVGQMVGQFAKLFGCRVVGSAGSQQKVDLLTSKFGYDE
PaDBR2 ALGMAGFTAYVGFFRICEPKKGDTIYVSAASGAVGQMVGQFAKLFGCRVVGSAGSQQKVDLLTSKFGYDE
AtDBR1 LLGMPGMTAYAGFYEVCSPKEGETVYVSAASGAVGQLVGQLAKMMGCYVVGSAGSKEKVDLLKTKFGFDD
200
200
204
202
207
NtDBR
ILGMPGMTAYAGFHEVCSPKKGETVFVSAASGAVGQLVGQFAKMLGCYVVGSAGSKEKVDLLKSKFGFDE
RiRZS1 LLGMPGMTAYAGFYEICSPKKGETVYVSAASGAVGQLVGQFAKLTGCYVVGSAGSKEKVDLLKNKFGFDE
AXXGXXG
PaDBR1 AFNYKEEPDLDAALKRCCPDGIDIYFENVGGKMLDAVLVNMKVFGRIAVCGMVSQYNKEVQDPIYNLNYT
PaDBR2 AFNYKEEPDLDAALKRCCPDGIDIYFENVGGKMLDAVLVNMKVFGRIAVCGMVSQYNKEVQDPIYNLNYT
AtDBR1 AFNYKEESDLTAALKRCFPNGIDIYFENVGGKMLDAVLVNMNMHGRIAVCGMISQYNLENQEGVHNLSNI
270
270
274
272
277
NtDBR
AFNYKEEQDLSAALKRYFPDGIDIYFENVGGKMLDAVLVNMKLYGRIAVCGMISQYNLEQTEGVHNLFCL
RiRZS1 AFNYKEEADLDAALRRYFPDGIDIYFENVGGKMLDAVLPNMRPKGRIAVCGMISQYNLEQPEGVRNLMAL
GXXS
*
PaDBR1 VKRRLKIHGFLQSDHMDVQPEFFKNVIQWFKEGKLVYVEDIADGLEKGPAALIGLLAGKNVGKQSVKVAD
PaDBR2 VKRRLKIHGFLQSDHMDVQPEFFKNVIQWFKEGKLVYVEDIADGLEKGPAALIGLLAGKNVGKQSVKIAD
AtDBR1 IYKRIRIQGFVVSDFYDKYSKFLEFVLPHIREGKITYVEDVADGLEKAPEALVGLFHGKNVGKQVVVVAR
340
340
344
342
347
NtDBR
ITKRIRMEGFLVFDYYHLYPKYLEMVIPQIKAGKVVYVEDVAHGLESAPTALVGLFSGRNIGKQVVMVSR
RiRZS1 IVKQVRMEGFMVFSYYHLYGKFLETVLPYIKQGKITYVEDVVDGLDNAPAALIGLYSGRNVGKQVVVVSR
*
Fig. 1. Peptide alignment of the two PaDBRs with other double bond reductase sequences. AtDBR1 from A. thaliana (GenBank accession BT022058), NtDBR from tobacco
(BAA89423) and RiRZS1 from raspberry (JN166691). The conserved co-enzyme binding motifs AXXGXXG and GXXS are shown boxed, and active site residues indicated with
an asterisk.

3124
Y. Wu et al. / FEBS Letters 587 (2013) 3122–3128
tube (Millipore, USA) in the presence of binding buffer (20 mM
Tris–HCl, 500 mM NaCl, 5 mM imidazole, pH 8.0). Protein quantifi-
cation was achieved using the Bradford assay (Bio Rad) with bovine
serum albumin used as the standard.
(30, 60, 80, 100, 150, 200, 300, 400, 500, 600, 700, 800 lM) in reac-
tions run at the optimal pH and temperature for 10 min. The quan-
tity of reaction product present was estimated from a standard
calibration curve.
2.4. Enzyme assays
2.5. Site-directed mutagenesis
Each 100
l
l assay comprised 10
l
g purified protein, 400
l
M
Five PaDBR1 mutants were created using Stratagene Quick-
Change site-directed mutagenesis method. The PaDBR1-pET32a
vector served as the template and five pairs of complementary
primers were designed using primer X on-line software (http://
www.bioinformatics.org/primerx) to generate C56Y, C56F, C56 V,
C56A and C56S mutations (Suppl. Table 2). Following the PCR,
the wild type PaDBR1-pET32a plasmid was digested by DpnI (Ther-
substrate, 1 mM NADPH, made up in 100 mM potassium phos-
phate buffer, pH 6.0. The reactions were initiated by the addition
of the protein and terminated after a 30 min incubation at 37 °C
by the addition of 10 ll glacial acetic acid (GAA). The reaction mix-
ture was extracted twice with an equal volume of ethyl acetate, the
ethyl acetate fraction evaporated, and the residue dissolved in
100
l
l methanol. The methanol solution was subjected to HPLC
mo Scientific, USA) at 37 °C for 3.5 h, and a 10
purified restriction reaction product was transformed into E. coli
DH5 . The mutations were first validated by sequencing prior to
their heterologous expression in E. coli BL21 (DE3). An enzyme as-
say was performed using both purified PaDBR1 and extracts of the
PaDBR1 mutants, using five different hydroxycinnamyl aldehydes
as substrate, in reactions set up as described above. Catalytic effi-
ciency was measured by quantifying the amount of reaction prod-
uct based on a standard calibration curve.
ll aliquot of a gel-
analysis on a reverse-phase C18 column (XDB-18, 5
l
m; Agilent),
in which the mobile phase varied linearly over 18 min from 5 parts
acetonitrile, 95 parts water containing 3% glacial acetic acid to 25
parts acetonitrile, 75 parts 3% aqueous GAA. The flow rate was
0.8 ml per min and the absorbance of the reaction product was
monitored at 280 nm. The enzymatic products were identified by
mass spectrometry (MS) using a Shimadzu LCMS-2020 system
(Shimadzu Corporation, Japan) with a multi wavelength diode ar-
ray detector and electron spray ionization mass spectrometer,
based on their m/z [MꢀH]ꢀ ion fractions.
a
3. Results and discussion
The effect on enzymatic activity of altering the solution pH over
the range 4–8 was monitored by running the reactions at 35 °C for
30 min in a range of buffers. The optimal temperature was deter-
mined in reactions formulated to a pH of 6.0. The enzymes’ kinetic
parameters were determined by altering substrate concentration
3.1. cDNA isolation and DBR sequence analysis
The length of the PaDBR1 cDNA clone was 1,371 bp, of which
1026 bp represented the ORF. Its predicted gene product was a
Artemisia annua DBR1
Rubus idaeus RZS1
Nicotiana tabacum DBR
Mentha x piperita PulR
Arabidopsis thaliana DBR
Hordeum vulgare ALH
63
64
88
53
¢,£-unsaturated
bonds reductase (DBRs)
Plagiochasma appendiculatum DBR1
Plagiochasma appendiculatum DBR2
Pinus taeda DBR1
99
100
77
100
Cavia porcellus AOR
Rattus norvegicus AOR
Escherichia coli CurA
Cucumis sativus ChlAOR
Fragaria x ananassa EO
Gallus gallus ADH6
100
100
91
66
alcohol dehydrogenase
(ADHs)
Homo sapiens ADH5
100
61
Solanum lycopersicum ADH3
Oryzias latipes ALDH1A2
Oryctolagus cuniculus ALDH1A1
Mus musculus ALDH1L1
Taeniopygia guttata ALDH3A2
Homo sapiens AFAR
98
100
aldehyde
dehydrogenase (ALDHs)
99
100
Rattus norvegicus AFAR
Arabidopsis thaliana AKR4C8
Triticum urartu AKR4C10
Oryza sativa AKR2
Aldo-keto
reductase (AKRs)
99
100
99
51 Zea mays AKR1B1
0.2
Fig. 2. Phylogenetic relationships among the DBRs, ADHs, ALDHs and AKRs. Artemisia annua DBR1 (FJ750460), Rubus idaeus RZS1 (JN16669), Nicotiana tabacum DBR
(BAA89423), Mentha x piperita PulR (AY300163), A. thaliana DBR (BT022058), Hordeum vulgare ALH (AY904340), Pinus taeda DBR1 (DQ829775), Cavia porcellus AOR
(NM_001172980), Rattus norvegicus AOR (U66322), Escherichia coli CurA (AB583756), Cucumis sativus ChlAOR (AB597302), Fragaria x ananassa EO (AY048861), Gallus gallus
ADH6 (NM_205092), Homo sapiens ADH5 (NM_000671), Solanum lycopersicum ADH3 (NM_001251867), Oryzias latipes ALDH1A2 (NM_001104821), Oryctolagus cuniculus
ALDH1A1 (NM_001082013), Mus musculus ALDH1L1 (NM_027406), Taeniopygia guttata ALDH3A2 (NM_001245686), Homo sapiens AFAR (AF026947), Rattus norvegicus AFAR
(X74673), Arabidopsis thaliana AKR4C8 (DQ837653), Triticum urartu AKR4C10 (EMS61074), Oryza sativa AKR2 (GQ227709), Zea mays AKR1B1 (NP_001105982).

Y. Wu et al. / FEBS Letters 587 (2013) 3122–3128
3125
protein consisting of 341 residues with a Mr of 37.8 kDa. The
length of the 50-UTR was 196 bp, and that of the 30-UTR
149 bp with a 30 poly (A) tail. The PaDBR2 cDNA clone was
1,401 bp long, also harboring a 1026 bp ORF which encoded a
341 residue polypeptide of Mr 37.8 kDa. The length of its 50-
UTR was 305 bp, and that of its 30-UTR 70 bp with a 30 poly
(A) tail. The sequence similarity of the two products at the pep-
tide level was 99.4%: only two residues were polymorphic,
namely C56 in PaDBR1 which was represented by Y56 in PaD-
BR2, and V338 in PaDBR1 (I338 in PaDBR2). The sequences of
both UTRs were dissimilar: the sequence identity at the nucleo-
tide level was 26.3% for 50-UTR and 22.7% for 30-UTR (Suppl.
Fig. 1). Tyrosine at position 56 of PaDBR2 is highly conserved
among oxidoreductases, and has therefore been proposed to be
involved in binding with NADPH [4]. However, in PaDBR1, this
position was occupied by a Cystein residue.
(ADH), aldehyde dehydrogenase (ALDH) and aldo–keto reductase
(AKR) sequences from different organisms. The phylogenetic anal-
ysis showed that the PaDBRs were included in a clade which con-
tains HvALH, AtDBR1, MpPulR, NtDBR, RiRZS1 and AaDBR1, as well
as to the non-plant proteins CpAOR, RnAOR and EcCurA (Fig. 2).
The Rubus idaeus protein RiRZS1 [21] is responsible for the hydro-
genation of the
a,b-unsaturated double bond of phenylbutenones
in the raspberry fruit. The tobacco enzyme NtDBR [15] shows dou-
ble bond reductase activity across a number of alternative sub-
strates, while the rat liver enzyme RnAOR [22] catalyzes the
reduction of
a,b-unsaturated aldehydes and ketones. The phylog-
eny demonstrated that DBRs, ADHs, ALDHs and AKRs all map to
different clades, suggesting a possible evolutionary relationship
based on their ability to detoxify reactive carbonyls [23].
3.2. The biochemical and enzymatic properties of the PaDBRs
The two PaDBR sequences shared 59.0%, 57.8% and 55.5% iden-
tity with, respectively, NtDBR [15], AtDBR1 [4] and RiRZS1 [21]
(Fig. 1). Both PaDBRs included the conserved glycine-rich motif
AASGAVG, known to participate in the binding of the NAD(P)+ or
NAD(P)H pyrophosphate group, as well as a GXXS motif, known
to stabilize both the adenine and nicotinamide moieties of NADPH
[4]. In order to elucidate the phylogenetic relationships of the PaD-
The molecular weight of each of the two N-terminally His-
Tagged recombinant PaDBRs expressed in E. coli was about
55 kDa (Suppl. Fig. 2). When PaDBR1 was provided with p-couma-
ryl aldehyde or caffeyl aldehyde as a substrate, the major reaction
products exhibited a similar retention time and molecular ion peak
as, respectively, p-dihydro-p-coumaryl aldehyde and dihydrocaf-
feyl aldehyde (Fig. 3). There was evidence of reactivity when the
substrate was either coniferyl or 5-hydroxyconiferyl aldehyde,
BR genes, a phylogenetic tree was constructed using selected a,b-
unsaturated bonds reductase (DBR), alcohol dehydrogenase
p-Coumaryl
Coniferyl
aldehyde
5-Hydroxyconiferyl
aldehyde
Caffeyl
aldehyde
aldehyde
60
40
60
40
60
40
60
40
20
Empty pET32a
20
0
20
0
20
0
0
60
40
20
0
60
40
20
0
60
40
20
0
60
40
20
0
PaDBR1
PaDBR2
60
40
60
40
60
40
20
0
60
40
20
0
20
0
20
0
Dihydro-p-coumaryl
Dihydrocaffeyl
aldehyde
Dihydro-5-hydroxy-
coniferyl aldehyde
Dihydroconiferyl
aldehyde
aldehyde
60
40
20
0
60
40
20
0
60
40
20
0
60
40
20
0
Standard
6
10 14
2
6
10 14
5
10 15 20
5
10 15 20
2
Retention time (min)
3
3
3
3
(x10 )
(x10 )
(x10 )
(x10 )
1.0
194.9
226.9
178.9
164.8
148.9
181.1
1.0
0.5
0.0
1.5
1.0
0.0
1.0
0.5
0.0
210.9
0.5
0.0
196.9
200
200
300
100
150
100
150
100 150 200 250
100
200
m/z
Fig. 3. HPLC profiles and MS spectra of reaction products generated by recombinant PaDBRs. The activities of recombinant PaDBRs when provided with either p-coumaryl,
cafferyl, coniferyl or 5-hydroxyconiferyl aldehyde as a substrate. The MS spectra of the PaDBR2 reaction products are shown in the lower panel.

3126
Y. Wu et al. / FEBS Letters 587 (2013) 3122–3128
but not when it was sinapyl aldehyde. In contrast, PaDBR2 was able
to reduce p-coumaryl, caffeyl, coniferyl and 5-hydroxyconiferyl
aldehyde, and its catalytic efficiency was higher than that of PaD-
BR1 in each case. PaDBR2 showed trace reactivity with sinapyl
aldehyde (data not shown). We suggested the failure to accept
sinapyl aldehyde as a substrate of PaDBR1 resulted from that the
size of sinapyl aldehyde is larger and sinapyl aldehyde is more
hydrophobic than other phenylpropanoid aldehydes. As a result,
smaller size and less hydrophobic substrates such as p-coumaryl
aldehyde or caffeyl aldehyde are the preferred substrates of PaD-
BR1. The retention time and MS spectra of the major reaction prod-
ucts formed from each substrate are shown in Fig. 3. The optimal
temperature for both enzymes was 37 °C and their pH optimum
was 6.0. Their Km, Vmax and Kenz (Kcat/Km) values are given in Ta-
ble 1. Based on their Km, the preferred substrate of each enzyme
was p-coumaryl aldehyde. The graphs for kinetics and pH and tem-
perature were shown in Suppl. Figs. 3 and 4.
Y81
S283
Y56
p-Coumary laldehyde
NADP+
Although the two PaDBR sequences differed with respect to two
residues, the catalytic efficiency of PaDBR1 was substantially less
than that of PaDBR2, especially towards cinnamyl aldehydes carry-
ing a methoxy group (coniferyl, 5-hydroxyconiferyl and sinapyl
aldehydes). With respect to AtDBR1, it has been suggested that
Y53, Y81, Y260 and S287 (equivalent to C56/Y56, Y81, Y256 and
S283 respectively for the PaDBRs) are the key residues determining
the enzyme’s interaction with NADP+/p-coumaryl aldehyde [4].
The sequence alignment implied that the 56C/Y substitution may
therefore be responsible for the observed difference in catalytic
ability between PaDBR1 and PaDBR2. To test this hypothesis,
site-directed mutagenesis was employed to convert 56C to 56Y
in PaDBR1. The result of this substitution was that substrate selec-
tivity and catalytic efficiency became indistinguishable from those
of PaDBR2 (data not shown). Further mutation of this residue to
either F, V, A or S showed that the catalytic efficiency of PaD-
Y256
Fig. 4. Docking arrangement of PaDBR1C56Y with NADP+/p-coumaryl aldehyde.
The arrow indicates potential stacking interaction between phenol rings.
Hydrogen bonds shown as dashed lines.
a
BR1C56F and PaDBR1C56Y was similar for each substrate tested,
and was higher than that of the wild type enzyme – specifically,
about three fold higher for p-coumaryl, caffeyl aldehydes and ten
fold higher for coniferyl and 5-hydroxyconiferyl aldehydes (Ta-
ble 2). On the other hand, the activities of the PaDBR1C56V, -
Table 1
Kinetic parameters of the recombinant PaDBR1, PaDBR2, PaDBR1C56Y and PaDBR1C56V proteins. The substrates are as follows: (1) p-Coumaryl aldehyde; (2) Caffeyl aldehyde;
(3) Coniferyl aldehyde; (4) 5-Hydroxyconiferyl aldehyde.
Enzyme
PaDBR1
Substrate
Km
(
l
M)
Vmax (nmol mgꢀ1 minꢀ1
)
Kcat (sꢀ1
)
Kenz (Mꢀ1 sꢀ1
)
1
2
1
2
3
4
1
2
3
4
1
2
171.9 43.0
221.4 31.3
94.3 22.4
155.7 14.2
210.2 11.1
440.4 27.4
377.4 19.4
187.1 11.3
267.9 18.3
394.0 22.9
438.2 23.4
212.7 13.0
255.2 13.2
58.7 2.2
0.098 0.009
0.132 0.007
0.278 0.017
0.238 0.012
0.118 0.007
0.169 0.012
0.249 0.014
0.277 0.015
0.134 0.008
0.161 0.008
0.037 0.001
0.068 0.004
570.7
598.2
2949.3
1314.9
562.8
549.0
2101.9
1924.5
777.7
794.9
222.4
349.7
PaDBR2
181.2 25.6
209.9 29.4
308.1 50.1
118.3 18.9
143.7 23.3
172.6 28.5
202.6 27.4
166.3 18.0
193.3 32.7
PaDBR1C56Y
PaDBR1C56V
107.3 6.7
Table 2
The catalytic efficiency of wild-type PaDBR1 and PaDBR1 mutant proteins. The substrates are as follows: (1) p-Coumaryl aldehyde; (2) Caffeyl aldehyde; (3) Coniferyl aldehyde;
(4) 5-Hydroxyconiferyl aldehyde; (5) Sinapyl aldehyde; (6) Cinnamyl aldehyde; (7) p-Coumaric acid; (8) Caffeic acid; (9) p-Coumaryl alcohol; (10) Caffeyl alcohol; (11) Coniferyl
alcohol; (12) 5-Hydroxyconiferyl alcohol; (13) Sinapyl alcohol.
Specific activity (nmol mgꢀ1 minꢀ1
)
Subtrate
PaDBR1
C56Y mutant
C56F mutant
C56A mutant
C56S mutant
C56V mutant
1
2
3
4
23.70 0.38
17.68 0.61
1.78 0.23
2.66 0.54
NDa
74.31 4.64
49.66 1.96
23.85 1.45
23.35 1.23
Trb
66.01 1.74
45.42 0.93
27.09 0.36
22.01 2.06
Tr
4.09 0.80
7.76 0.31
2.63 0.22
4.64 0.35
ND
7.91 1.04
9.11 0.70
11.54 0.40
11.62 0.23
ND
12.43 2.06
14.95 0.91
8.35 0.49
8.43 0.64
ND
5
6–13
ND
ND
ND
–
–
–
a
No detectable activity.
Trace activity.
b

Y. Wu et al. / FEBS Letters 587 (2013) 3122–3128
3127
CHO
CHO
+
NADP
NADPH
Double Bond Reductase
(DBR)
R
R
R
R
2
1
1
2
OH
OH
R1,R2= H, p-Coumaryl aldehyde
R1,R2=H, Dihydro-p-coumaryl aldehyde
R1=OH; R2=H, Caffeyl aldehyde
R1=OH; R2=H, Dihydrocaffeyl aldehyde
R1=OCH3; R2=H, Coniferyl aldehyde
R1=OCH3; R2=OH, 5-Hydroxyconiferyl aldehyde
R1,R2=OCH3, Sinapyl aldehyde
R1=OCH3; R2=H, Dihydroconiferyl aldehyde
R1=OCH3; R2=OH, Dihydro-5-hydroxyconiferyl aldehyde
R1,R2=OCH3, Dihydrosinapyl aldehyde
Fig. 5. Substrates catalyzable by PaDBRs and their corresponding reaction products.
C56S and -C56A enzymes were less effective than those in which
an aromatic amino acid was present at position 56. The Y53 resi-
due is conserved across the oxidoreductases and is located be-
References
[1] Dixon, R.A., Achnine, L., Kota, P., Liu, C.J., Reddy, M. and Wang, L. (2002) The
phenylpropanoid pathway and plant defence-a genomics perspective. Mol.
Plant Pathol. 3, 371–390.
tween a relatively well conserved short loop and a single turn
a
helix; its phenol ring in AtDBR1 forms a hydrophobic stacking
interaction with the phenol ring of the bound p-coumaryl alde-
hyde, thereby facilitating the orientation of the substrate within
the specificity pocket [4]. To verify whether Y56 has a similar role
in PaDBRs, the protein structures of PaDBR1 and PaDBR1C56Y were
modeled, based on AtDBR1’s crystal structure [4]. The model sug-
gested that, just as in AtDBR1, Y56 formed a stacking interaction
with the phenol ring of the bound p-coumaryl aldehyde (Fig. 4).
The Y81 phenolic hydroxyl group was located within a single
hydrogen bond distance of the p-coumaryl aldehyde’s phenolic hy-
droxyl group. The Y256 phenolic hydroxyl group also was located
within hydrogen bonding distance of the p-coumaryl aldehyde car-
bonyl group and the nicotinamide ribose 2’-hydroxyl group. The
assumption therefore was that Y256 established hydrogen bonding
with both. The hydroxyl groups in Y81 and S283 may also facilitate
substrate binding. The suggestion is that the PaDBRs and AtDBR1
share the same catalytic mechanism, while the hydrophobic inter-
action between Y56 and the substrate’s phenolic ring may be a uni-
versal feature of plant double bond reductases.
In summary, the PaDBRs were able to reduce hydroxycinnamyl
aldehydes (Fig. 5). While several other possible substrates (includ-
ing cinnamyl aldehyde, hydroxycinnamic acids and alcohols) were
tested using PaDBR1 and PaDBR2 (Table 2), none of these com-
pounds was converted into its corresponding dihydro form. The
lack of reactivity towards cinnamyl aldehyde indicated the impor-
tance of the phenolic hydroxyl group in the hydroxycinnamyl alde-
hydes. Both PaDBR1 and PaDBR2 lacked any hydrogenation activity
when exposed to two diterpenoids, which were purified from an
extract of the liverwort species Jungermannia fauriana (data not
shown),, indicating that the substrate versatility of the PaDBRs is
quite low, being mainly involved in the formation of dihydrophe-
nylpropanoid derivatives.
[2] VanEtten, H.D., Mansfield, J.W., Bailey, J.A. and Farmer, E.E. (1994) Two classes
of plant antibiotics: phytoalexins versus’’ phytoanticipins’’. Plant Cell 6, 1191–
1192.
[3] Teutonico, R.A., Dudley, M.W., Orr, J.D., Lynn, D.G. and Binns, A.N. (1991)
Activity and accumulation of cell division-promoting phenolics in tobacco
tissue cultures. Plant Physiol. 97, 288–297.
[4] Youn, B., Kim, S.J., Moinuddin, S.G., Lee, C., Bedgar, D.L., Harper, A.R., Davin, L.B.,
Lewis, N.G. and Kang, C. (2006) Mechanistic and structural studies of apoform,
binary, and ternary complexes of the Arabidopsis alkenal double bond
reductase At5g16970. J. Biol. Chem. 281, 40076–40088.
[5] Karonen, M., Hamalainen, M., Nieminen, R., Klika, K.D., Loponen, J.,
Ovcharenko, V.V., Moilanen, E. and Pihlaja, K. (2004) Phenolic extractives
from the bark ofPinus sylvestris L. and their effects on inflammatory mediators
nitric oxide and prostaglandin E2. J. Agric. Food Chem. 52, 7532–7540.
[6] Thompson, L.U., Seidl, M.M., Rickard, S.E., Orcheson, L.J. and Fong, H.H. (1996)
Antitumorigenic effect of a mammalian lignan precursor from flaxseed. Nutr.
Cancer 26, 159–165.
[7] Ehrenkranz, J.R., Lewis, N.G., Kahn, C.R. and Roth, J. (2005) Phlorizin: a review.
Diabetes-Metab. Res. Rev. 21, 31–38.
[8] Xia, Z.Q., Costa, M.A., Pelissier, H.C., Davin, L.B. and Lewis, N.G. (2001)
Secoisolariciresinol dehydrogenase purification, cloning, and functional
expression: implications for human health protection. J. Biol. Chem. 276,
12614–12623.
[9] Youn, B., Moinuddin, S.G., Davin, L.B., Lewis, N.G. and Kang, C. (2005) Crystal
structures of apo-form and binary/ternary complexes of Podophyllum
secoisolariciresinol dehydrogenase, an enzyme involved in formation of
health-protecting and plant defense lignans. J. Biol. Chem. 280, 12917–12926.
[10] Moinuddin, S.G., Youn, B., Bedgar, D.L., Costa, M.A., Helms, G.L., Kang, C., Davin,
L.B. and Lewis, N.G. (2006) Secoisolariciresinol dehydrogenase: mode of
catalysis and stereospecificity of hydride transfer inPodophyllum peltatum.
Org. Biomol. Chem. 4, 808–816.
[11] Min, T., Kasahara, H., Bedgar, D.L., Youn, B., Lawrence, P.K., Gang, D.R., Halls,
S.C., Park, H., Hilsenbeck, J.L. and Davin, L.B. (2003) Crystal structures of
pinoresinol-lariciresinol and phenylcoumaran benzylic ether reductases and
their relationship to isoflavone reductases. J. Biol. Chem. 278, 50714–50723.
[12] Gang, D.R., Kasahara, H., Xia, Z.-Q., Vander Mijnsbrugge, K., Bauw, G., Boerjan,
W., Van Montagu, M., Davin, L.B. and Lewis, N.G. (1999) Evolution of plant
defense mechanisms relationships of phenylcoumaran benzylic ether
reductases to pinoresinol-lariciresinol and isoflavone reductases. J. Biol.
Chem. 274, 7516–7527.
[13] Ma, Q.-H. (2010) Functional analysis of a cinnamyl alcohol dehydrogenase
involved in lignin biosynthesis in wheat. J. Exp. Bot. 61, 2735–2744.
[14] Kim, S.-J., Kim, M.-R., Bedgar, D.L., Moinuddin, S.G., Cardenas, C.L., Davin, L.B.,
Kang, C. and Lewis, N.G. (2004) Functional reclassification of the putative
cinnamyl alcohol dehydrogenase multigene family in Arabidopsis. Proc. Natl.
Acad. Sci. USA 101, 1455–1460.
[15] Mansell, D.J., Toogood, H., Waller, J., Hughes, J.M., Levy, C., Gardiner, J.M. and
Scrutton, N.S. (2013) Biocatalytic asymmetric alkene reduction: crystal
structure and characterization of a double bond reductase from Nicotiana
tabacum. ACS Catal. 3, 370–379.
Conflict of interest
The authors declare no conflict of interest in the present
investigation.
Acknowledgments
[16] Kasahara, H., Jiao, Y., Bedgar, D.L., Kim, S.-J., Patten, A.M., Xia, Z.-Q., Davin, L.B.
and Lewis, N.G. (2006) Pinus taeda phenylpropenal double-bond reductase:
Purification, cDNA cloning, heterologous expression in Escherichia coli, and
subcellular localization in P. taeda. Phytochemistry 67, 1765–1780.
[17] Pereira, P.J.B., Macedo-Ribeiro, S., Párraga, A., Pérez-Luque, R., Cunningham, O.,
Darcy, K., Mantle, T.J. and Coll, M. (2001) Structure of human biliverdin IXb
reductase, an early fetal bilirubin IXb producing enzyme. Nat. Struct. Mol. Biol.
8, 215–220.
We are grateful for the support of the National Natural Science
Foundation of China (No. 31170280), Young Scientists Foundation
Grant of Shandong Province (No. BS2010SW019).
Appendix A. Supplementary data
[18] Cheng, A., Wang, L., Sun, Y. and Lou, H. (2013) Identification and expression
analysis of key enzymes of the terpenoids biosynthesis pathway of a liverwort
Plagiochasma appendiculatum by EST analysis. Acta. Physiol. Plant. 35, 107–
118.
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.febslet.2013.
07.051.

3128
Y. Wu et al. / FEBS Letters 587 (2013) 3122–3128
[19] Tamura, K., Dudley, J., Nei, M. and Kumar, S. (2007) MEGA4: molecular
evolutionary genetics analysis (MEGA) software version 4.0. Mol. Biol. Evol.
24, 1596–1599.
[20] Trott, O. and Olson, A.J. (2010) AutoDock Vina: improving the speed and
accuracy of docking with a new scoring function, efficient optimization, and
multithreading. J. Comput. Chem. 31, 455–461.
the alpha, beta-hydrogenation of phenylbutenones in raspberry fruits.
Biochem. Biophys. Res. Co. 412, 104–108.
[22] Dick, R.A. and Kensler, T.W. (2004) The catalytic and kinetic mechanisms of
NADPH-dependent alkenal/one oxidoreductase. J. Biol. Chem. 279, 17269–
17277.
[23] Mano, J.I., Belles-Boix, E., Babiychuk, E., Inzé, D., Torii, Y., Hiraoka, E., Takimoto,
K., Slooten, L., Asada, K. and Kushnir, S. (2005) Protection against
photooxidative injury of tobacco leaves by 2-alkenal reductase. Detoxication
of lipid peroxide-derived reactive carbonyls. Plant Physiol. 139, 1773–1783.
[21] Koeduka, T., Watanabe, B., Suzuki, S., Hiratake, J., Mano, J.i and Yazaki, K.
(2011) Characterization of raspberry ketone/zingerone synthase, catalyzing