Assessment of a putative proton relay in Arabidopsis cinnamyl alcohol dehydrogenase catalysis

Article abstract of DOI:10.1039/c2ob27189c

Extended proton relay systems have been proposed for various alcohol dehydrogenases, including the Arabidopsis thaliana cinnamyl alcohol dehydrogenases (AtCADs). Following a previous structural biology investigation of AtCAD5, the potential roles of three amino acid residues in a putative proton relay system, namely Thr49, His52 and Asp57, in AtCAD5, were investigated herein. Using site-directed mutagenesis, kinetic and isothermal titration calorimetry (ITC) analyses, it was established that the Thr49 residue was essential for overall catalytic conversion, whereas His52 and Asp57 residues were not. Mutation of the Thr49 residue to Ala resulted in near abolition of catalysis, with thermodynamic data indicating a negative enthalpic change (ΔH), as well as a significant decrease in binding affinity with NADPH, in contrast to wild type AtCAD5. Mutation of His52 and Asp57 residues by Ala did not significantly change either catalytic efficiency or thermodynamic parameters. Therefore, only the Thr49 residue is demonstrably essential for catalytic function. ITC analyses also suggested that for AtCAD5 catalysis, NADPH was bound first followed by p-coumaryl aldehyde.

Full text of DOI:10.1039/c2ob27189c

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
Assessment of a putative proton relay in Arabidopsis
cinnamyl alcohol dehydrogenase catalysis
Cite this: Org. Biomol. Chem., 2013, 11,
1127
Choonseok Lee, Diana L. Bedgar, Laurence B. Davin and Norman G. Lewis*
Extended proton relay systems have been proposed for various alcohol dehydrogenases, including the
Arabidopsis thaliana cinnamyl alcohol dehydrogenases (AtCADs). Following a previous structural biology
investigation of AtCAD5, the potential roles of three amino acid residues in a putative proton relay
system, namely Thr49, His52 and Asp57, in AtCAD5, were investigated herein. Using site-directed muta-
genesis, kinetic and isothermal titration calorimetry (ITC) analyses, it was established that the Thr49
residue was essential for overall catalytic conversion, whereas His52 and Asp57 residues were not.
Mutation of the Thr49 residue to Ala resulted in near abolition of catalysis, with thermodynamic data
indicating a negative enthalpic change (ΔH), as well as a significant decrease in binding affinity with
NADPH, in contrast to wild type AtCAD5. Mutation of His52 and Asp57 residues by Ala did not signifi-
cantly change either catalytic efficiency or thermodynamic parameters. Therefore, only the Thr49 residue
is demonstrably essential for catalytic function. ITC analyses also suggested that for AtCAD5 catalysis,
NADPH was bound first followed by p-coumaryl aldehyde.
Received 9th November 2012,
Accepted 10th December 2012
DOI: 10.1039/c2ob27189c
www.rsc.org/obc
bona fide Nicotiana tabacum and Pinus taeda CADs,1,3 only
AtCAD4 and AtCAD5 proteins were CADs proper based on both
Introduction
Cinnamyl alcohol dehydrogenases (CAD, EC 1.1.1.195) from in vitro highest catalytic activity1 and in vivo studies using the
vascular plants can catalyze in vitro the reversible conversion A. thaliana double mutant cad4/cad5.4,5 The remaining 8 puta-
of p-coumaryl (1), caffeyl (2), coniferyl (3), 5-hydroxyconiferyl tive CADs had very low homology (0.9–1.6% similarity) to the
(4) and sinapyl (5) aldehydes into the corresponding alcohols N. tabacum and P. taeda CADs, and these proteins also lacked
(6–10) (Fig. 1), with the forward reaction being favoured.1 In catalytic zinc.1
Arabidopsis thaliana, the first plant species whose genome was
Cinnamyl alcohol dehydrogenases belong to the medium-
structural
sequenced,2 17 genes were provisionally annotated as CADs by chain dehydrogenase/reductase family,6 and
a
The Arabidopsis Information Resource (TAIR).3 Of the 9 puta- biology study of the AtCAD5 apo-enzyme and its binary
tive Arabidopsis thaliana CAD genes with highest homology to complex with NADP+ was reported by Youn et al.7 In terms of
the putative CAD mechanism, there are currently two possible
mechanisms for related alcohol dehydrogenases, using either
an extended proton relay (transfer)8 or a penta-coordinated
zinc system.9 The extended proton relay mechanism assumes
that a water molecule bound to tetrahedrally coordinated cata-
lytic zinc is displaced by an incoming alcohol substrate and
that this species is maintained during catalysis.8
That proposed mechanism was based on the study of the
ternary complex of horse liver alcohol dehydrogenase
(HLADH), crystallized in the presence of NAD+ and p-bromo-
benzyl alcohol, where it was reported that the catalytic zinc ion
was ligated to the alcohol oxygen via a tetrahedrally coordi-
nated complex (Fig. 2a).8,10–13 In addition, an extended hydro-
Fig. 1 Enzymatic reactions catalysed by AtCAD5 in presence of NADPH.
gen bonding system was reported between the hydroxyl group
of the substrate alcohol coordinated to zinc ion and Ser48,
between Ser48 and nicotinamide ribose (O2′), and between
nicotinamide ribose (O2′) and His51 in the ternary complex of
HLADH, whereas in the apo-enzyme the hydrogen bond
Institute of Biological Chemistry, Washington State University, Pullman, WA 99164-
6340, USA. E-mail: lewisn@wsu.edu; Fax: +1-509-335-8206; Tel: +1-509-335-2682
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1127–1134 | 1127

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 2 Two proposed catalytic mechanisms for alcohol dehydrogenases: (a) Proton relay system in horse liver alcohol dehydrogenase (HLADH); adapted and
redrawn from Eklund et al.8 (b) Penta-coordinated zinc system in Thermoanaerobacter brockii alcohol dehydrogenase (TbADH); adapted and redrawn from Kleifeld
et al.9 TR1: the first transient complex, TR2: the second transient complex.
system was not observed.8 In contrast to HLADH, however, a zinc ion complex, the second transient species. The catalytic
ternary complex with substrate and cofactor is not available for cycle is envisaged to be completed through dissociation of the
AtCAD5 thus far.7
water molecule and product, and re-ligation of Glu60 to regen-
In the alternative model (Fig. 2b), binding of the alcohol erate the original tetra-coordinated zinc ion species, i.e. return-
substrate putatively results in formation of a penta-coordi- ing the enzyme to its resting state.9 Site-directed mutagenesis
nated zinc complex,9 as deduced from analysis of results however, gave a Glu60Ala mutant whose catalytic efficiency
obtained with a Thermoanaerobacter brockii alcohol dehydro- was only reduced circa four fold relative to the wild type
genase (TbADH) using extended X-ray absorption fine struc- enzyme.9 Overall, this mechanism (Fig. 2b) does not require
ture, pre-steady state kinetics, and density functional theory an extended proton relay transfer as for the one described for
calculation approaches. In that study, two distinct penta- HLADH (Fig. 2a).
coordinate intermediates were postulated as being formed
In terms of developing an understanding of overall CAD
during the catalytic cycle. Specifically, the following was envi- catalysis, the first report was that describing molecular model-
saged: when NAPD+ is bound to TbADH, the tetra-coordinated ling of Eucalyptus gunnii CAD2 against HLADH.14 This study
zinc ion species is maintained, and the first penta-coordinated suggested that its putative substrate binding region in the
zinc ion transient complex is formed by addition of a water active site included residues Ile95, Tyr113, Trp119 and Phe298
molecule. The penta-coordinated zinc ion species is then con- substituted from the corresponding residues Phe93, Phe110,
verted back to a tetra-coordinated zinc ion complex by dis- Leu116 and Ile318 of HLADH as reviewed in Lewis et al.15 Both
sociation of the ligated Glu60 residue. Binding of the Trp119 and Phe298 residues14 were proposed as providing a
incoming alcohol substrate regenerates a penta-coordinated “molecular sandwich”, whereby the phenolic ring of the
1128 | Org. Biomol. Chem., 2013, 11, 1127–1134
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
substrate is stabilized, and in an orientation enabling abstrac-
In the investigation herein, site-directed mutagenesis and
tion of the pro-R hydride from NADPH during the reductive ITC were used to further evaluate and assess the validity of a
step.16 In the reverse reaction, the 9-pro-R hydrogen of coni- putative extended proton relay mechanism in AtCAD5,7 includ-
feryl alcohol (8) is removed.17 In addition, Ser212/Arg217 resi- ing study of the effects of mutating putative key amino acid
dues in E. gunnii CAD2, and Asp223/Lys228 in HLADH, were residues on substrate and cofactor binding, as well as catalytic
predicted to be involved in cofactor binding in this model,14 turnover. Additionally, this study examined the order of sub-
with the Ser212 residue further confirmed to be closely related strate/co-factor binding during AtCAD5 catalysis as an ordered
to NADPH binding through its site-directed mutant, bi–bi mechanism had been suggested by Wyrambik and
Ser212Asp.18
Grisebach.29
The putative three-dimensional molecular structure of Euca-
lyptus gunnii CAD2 also suggested that both residues, Ser49
and His52, served in a proton relay system in a comparable
way as for Ser48 and His51 residues in HLADH.14 In a some-
what analogous manner, following analysis of the binary
complex of AtCAD5, the corresponding residues, Thr49 and
His52, were also putatively involved in a proton relay, together
with Asp57 (Fig. 3), with the latter being postulated via sub-
strate modelling of the binary complex of AtCAD5.7
Isothermal titration calorimetry (ITC) is now being widely
used to investigate protein–protein, protein–nucleic acid and
protein–lipid interactions, drug design and enzyme kinetics.19
In particular, this approach is very effective for the study of
how proteins interact with either substrates, cofactors or
ligands through analysis of binding stoichiometry (N), binding
enthalpy (ΔH), dissociation constant Kd, binding entropy (ΔS)
and Gibbs free energy of binding (ΔG).19–28 Furthermore, for
site-directed mutated proteins, ITC can effectively be utilized
to compare the behaviour of native (recombinant) proteins
and their mutated forms.
Results and discussion
Residues Thr49, His52 and Asp57 of AtCAD5 were individually
mutated into Ala through PCR (see Table 1 for each primer set
used). It was envisioned that these mutations would establish
which of the three amino acid residues were potentially
involved in a proton relay (Fig. 3). After removal of parental
methylated and hemi-methylated DNA with Dpn I restriction
enzyme, each mutated DNA was individually transformed into
One shot® Top 10 Escherichia coli cells.30 After selecting each
mutated clone, mutations at each position were confirmed by
sequencing, with each construct next transformed into E. coli
cells for protein expression. Except for use of the Fast Protein
Liquid Chromatography system, the purification method
employed was as described in Kim et al.1 Each recombinant
protein was obtained from an imidazole-containing eluent
(from ∼120 to 180 mM in 20 mM Tris-HCl, 500 mM NaCl, pH
7.9) followed by concentration/buffer exchange in 20 mM Tris
buffer (pH 7.5). Each enzyme was purified to apparent homo-
geneity (data not shown) and used to obtain kinetic and ther-
modynamic parameters.
Kinetic analyses of AtCAD5 and its site-directed mutants
Kinetic parameters for AtCAD5 and its site-directed mutants
were initially obtained using the Michaelis–Menten equation31
with p-coumaryl aldehyde (1) employed as
a substrate
(Table 2).
In our hands, the native (recombinant) AtCAD5 gave a Km
13 μM, Vmax 88.1 pkat μg−1, and catalytic turnover of 6.8 s−1
.
That is, the substrate, p-coumaryl aldehyde (1), was bound very
tightly and was efficiently converted into the product, p-cou-
maryl alcohol (6). These data were thus in good agreement
with our previous study of AtCADs.1
As indicated earlier, in our previous study of the binary
complex with AtCAD5, Thr49 was proposed to be hydrogen-
Fig. 3 Putative proton shuttle mechanism for AtCAD5. The dotted lines indi-
cate the putative hydrogen bonds/proton relay. [Adapted and redrawn from
Youn et al.7] For R1, R2 explanation, see Fig. 1.
Table 1 Sequences of forward (F) and reverse (R) primers used for site-directed mutagenesis
Primers
Sequence (5′ → 3′)
AtCAD5 T49A F
AtCAD5 T49A R
AtCAD5 H52A F
AtCAD5 H52A R
AtCAD5 D57A F
AtCAD5 D57A R
GCTGTGGAATCTGCCACGCGGATCTTCATCAAACTA
TAGTTTGATGAAGATCCGCGTGGCAGATTCCACAGC
GAATCTGCCACACCGATCTTGCGCAAACTAAAAATGATCTTG
CAAGATCATTTTTAGTTTGCGCAAGATCGGTGTGGCAGATTC
CAAACTAAAAATGCGCTTGGCATGTCT
AGACATGCCAAGCGCATTTTTAGTTTG
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1127–1134 | 1129

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 2 Kinetic parameters for AtCAD5 and its site-directed mutants with p-coumaryl aldehyde (1) as substrate in presence of NADPH
Enzyme
Km (μM)
Vmax (pkat μg−1
)
kcat (s−1
)
kcat/Km (M−1 s−1
)
AtCAD5
T49A
H52A
D57A
13 1.9
114 11.3
117 6.9
36 3.1
88.1 3.5
0.33 0.01
518.9 16.5
957.7 31.0
6.8 0.3
0.026 0.001
40.2 1.3
523 000 ( 46 000)
230 ( 18)
345 000 ( 8700)
2 061 000 ( 102 000)
74.2 2.4
bonded to the O2′ of the nicotinamide ribose moiety and thus
was considered a potential residue in the putative proton
transfer mechanism.7 Participation of this residue in the
overall enzymatic reaction was established, via analysis of the
Thr49Ala mutant which was found to be essentially catalyti-
cally inactive. The affinity of the substrate, p-coumaryl alde-
hyde (1), was though still quite high (Km 114 μM), whereas the
Vmax was 0.33 pkat μg−1 and the catalytic turnover number was
0.026 s−1. Thus, the Thr49Ala mutation specifically caused a
near complete loss in catalytic efficiency.
On the other hand, kinetic analysis of the AtCAD5 H52A
recombinant protein gave very different results. The data
obtained established that this residue was not essential for
catalysis overall, even though it was provisionally envisioned
earlier as being involved in the proton relay via hydrogen
bonding to O3′ of the nicotinamide ribose. In our hands, sub-
strate affinity was still relatively high (Km 117 μM) for p-cou-
maryl aldehyde (1), the Vmax increased ∼6 fold (518.9 pkat
μg−1) with a catalytic turnover of 40.2 s−1. In an analogous
manner, the D57A mutant gave a catalytically active enzyme
with Km 36 μM for p-coumaryl aldehyde (1). However, the Vmax
also apparently increased ∼11 fold with the overall catalytic
turnover larger by ∼4 fold over the native (recombinant)
protein. Therefore, the kinetic analysis of AtCAD5 D57A
demonstrated that the Asp57 residue was also not essential for
overall catalysis in AtCAD5. Taken together, only the Thr49
residue can be considered essential for the overall catalytic
mechanism of AtCAD5.
There is an earlier precedent for this observation with the
β1β1 isoenzyme of human liver alcohol dehydrogenase
(HuLADH).32 Its site-directed mutant His51Gln (corresponding
to His52 in AtCAD5) also had 2 (at pH 7, 8 and 9) or 3 (at pH
10) times higher Vmax in the mutant, relative to wild type
enzyme, this being rationalized as due to less tight binding of
NADH thereby enabling it to be more readily dissociated.32–34
Accordingly, the mutation of His52 or Asp57 may facilitate
release of NADP+ during catalysis.
Fig. 4 Isothermal titration calorimetry analyses of AtCAD5 (a), AtCAD5 T49A
(b), AtCAD5 H52A (c) and AtCAD5 D57A (d) with NADPH. The top panels show
raw data of the heat pulses resulting from each titration, whereas the bottom
panels show the integrated heat normalized per mole of NADPH as a function
of the molar ratio (NADPH concentration/enzyme homodimer concentration).
from ΔG = −RT ln Ka, where R is the gas constant and T the
temperature. The binding entropy was obtained from ΔS =
(ΔH − ΔG)/T.19
Binding of p-coumaryl aldehyde (1) to AtCAD5 and its
mutants, however, yielded inconclusive results due to very low
heat generation for all enzymes tested, suggesting that p-cou-
maryl aldehyde (1) either binds very weakly or cannot bind to
the free enzyme (data not shown).
In contrast, significant heat changes were observed upon
binding of NADPH to AtCAD5 or its mutants (Fig. 4). Specifi-
cally, the NADPH binding data obtained were fitted best to a
model that specifies one set of sites, with N values from 0.7 to
0.8 obtained (Table 3), these being indicative of ∼1 : 1 stoichio-
metry per enzyme dimer. These data thus indicate that one
Isothermal titration calorimetry analyses of AtCAD5 and its
site-directed mutants
ITC measurements at constant temperature and pressure were
carried out to further investigate the binding properties of
AtCAD5 and its site-directed mutants using either p-coumaryl
aldehyde (1) or NADPH. Thermodynamic parameters which
included dissociation constant Kd (Kd = 1/Ka; Ka, binding
association constant) and enthalpy of binding (ΔH) were thus
obtained. The Gibbs free energy of binding was calculated
1130 | Org. Biomol. Chem., 2013, 11, 1127–1134
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Table 3 Thermodynamic parameters for NADPH binding to AtCAD5 and its mutants
Protein
Na
Kd (μM)
ΔG (kJ mol−1
)
ΔH (kJ mol−1
)
TΔS (kJ mol−1
)
AtCAD5
0.7 ( 0.02)
0.8 ( 0.002)
0.8 ( 0.02)
0.8 ( 0.02)
3.5 ( 0.1)
37.2 ( 11.6)
7.1 ( 1.1)
8.2 ( 0.5)
−31.7 ( 0.1)
−25.8 ( 0.9)
−29.9 ( 0.4)
−29.5 ( 0.1)
26.0 ( 0.3)
−12.0 ( 5.7)
17.1 ( 0.7)
15.6 ( 0.2)
57.7 ( 0.3)
13.8 ( 6.3)
47.0 ( 0.3)
45.1 ( 0.1)
AtCAD5 T49A
AtCAD5 H52A
AtCAD5 D57A
a N indicates binding stoichiometry of NADPH binding per enzyme homodimer.
molecule of NADPH is bound to AtCAD5, a homodimer,7 or its
site-directed mutants. Valencia et al. also reported that a puta-
tive CAD-like homolog from Saccharomyces cerevisiae binds one
NADPH molecule per homodimer, this being rationalized to
only one functional subunit probably being available for cataly-
sis.35 Similarly, Pseudomonas putida tartrate dehydrogenase
(TDH), a homodimer, also reportedly showed a 1 : 1 stoichio-
metry of NADH binding per enzyme dimer.36
The ITC data for binding of NADPH to AtCAD5 is summar-
ized in Table 3, with a binding isotherm and the fitted data
shown in Fig. 4A. NADPH was found to bind to AtCAD5 with a
dissociation constant of ∼3.5 μM. Formation of the binary
complex AtCAD5·NADPH is endothermic with
a ΔH of
26.0 kJ mol−1 (Table 3, Fig. 4a).
The essentially catalytically inactive T49A mutant was next
examined (Table 3 and Fig. 4b). The thermodynamics of
NADPH binding were affected by the T49A mutation: the Kd of
NADPH increased by >10 fold, indicating a significant decrease
in binding affinity, with formation of the binary complex now
being exothermic (ΔH of −12.0 kJ mol−1) and driven by a less
favourable −TΔS. As shown in the putative modelled ternary
complex of AtCAD5 (Fig. 3),7 the Thr49 is proposed to interact
with the substrate and NAD(P)H and is also located near to the
catalytic Zn2+. Replacement of Thr49 with Ala thus disrupts the
interaction between this amino acid residue, catalytic Zn2+,
and putatively NADPH and the substrate, the net effect being
decreased NADPH binding affinity (Table 3) and a near com-
plete abolition of catalytic activity (Table 2).
In agreement with the kinetic data, the thermodynamic
properties of NADPH binding for H52A and D57A mutant pro-
teins were very similar to that of WT (Table 3 and Fig. 4c,d).
Specifically, NADPH binding yielded a Kd of 7.1 μM, ΔH of
17.1 kJ mol−1 and TΔS of 47.0 kJ mol−1 for the H52A mutant,
and Kd of 8.2 μM, ΔH of 15.6 kJ mol−1 and TΔS of 45.1 kJ
mol−1 for the D57A mutant. Thus, this implies that mutation
of His52 or Asp57 to alanine did not very significantly affect
binding of NADPH to the corresponding protein in contrast to
mutation of Thr49 to Ala.
Fig. 5 Isothermal titration calorimetry (ITC) analysis of AtCAD5. (a) Titration at
25 °C of AtCAD5 and NADPH with p-coumaryl aldehyde (1). (b) Titration at
25 °C of AtCAD5 and p-coumaryl aldehyde (1) with NADPH. The top panel
show raw data of the heat pulses resulting for each titration, whereas the
bottom panels show the integrated heat normalized per mole of injectant as a
function of the molar ratio (injectant concentration/enzyme homodimer
concentration).
which was endothermic (Fig. 4a). However, when NADPH was
injected to the cell containing both AtCAD5 and p-coumaryl
aldehyde (1), a very different profile was observed (Fig. 5b).
First, an endothermic reaction occurred corresponding to
binding of NADPH to AtCAD followed by an exothermic
response due to the catalytic conversion of 1 into 6. Together,
these results further confirm that the cofactor binds first to
the enzyme followed by the aldehyde substrate. Similar results
were observed with the mutant enzymes (data not shown).
Conclusions
In the putative proton relay system for AtCAD5,7 three residues,
Thr49, His52 and Asp57, were provisionally implicated in its
catalysis as well as for other dehydrogenases.37,38 However,
from the kinetic and ITC data of these site-directed mutations
in AtCAD5, all three residues were apparently not essential for
catalysis. Notably, only mutation of Thr49 residue caused a
loss in catalytic activity for AtCAD5, this being further verified
by a significant decrease in binding affinity, a negative ΔH
from a positive ΔH in AtCAD5 and a decrease of TΔS in
NADPH binding.
In sum, NADPH binding to AtCAD5 and its mutants is a
thermodynamically favourable processes as indicated by the
negative ΔG values (between −25.8 and −31.7 kJ mol−1).
Titration of the binary complex AtCAD5·NADPH with p-cou-
maryl aldehyde (1), followed by the enzymatic conversion into
the corresponding alcohol (6), was next investigated at 25 °C.
The overall process was found to be exothermic (Fig. 5a), in
contrast to formation of the binary complex AtCAD5·NADPH
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1127–1134 | 1131

View Article Online
Paper
Organic & Biomolecular Chemistry
However, mutation of His52 and Asp57 residues did not DNA individually transformed into One Shot TOP 10 compe-
affect catalytic efficiency, but, in contrast, the turnover tent E. coli cells. Transformants were selected on LB medium
numbers and maximum velocities of the His52Ala and containing carbenicillin (100 μg ml−1). Positive clones were
Asp57Ala mutants increased. Moreover, binding properties of next subjected to sequencing to confirm single mutations.
AtCAD5 H52A and D57A remained largely unchanged.
From the ITC analyses, it was also concluded that NADPH
binds first to AtCAD5 followed by the substrate.
Taken together, the data provide no evidence for an
extended proton relay system.
Expression and purification of AtCAD5 and its mutants
Clones of AtCAD5 and its mutants were individually inoculated
in 5 ml of LB medium containing carbenicillin (100 μg ml−1
)
and incubated at 37 °C overnight. Next, 250 ml of LB medium
containing carbenicillin (100 μg ml−1) was inoculated with
2.5 ml of the corresponding 5 ml culture. The E. coli cells were
incubated at 37 °C at 230 rpm until an OD600 ∼ 0.8 was
reached, with induction then initiated by addition of isopropyl
β-D-1-thiogalactopyranoside (IPTG, 0.2 mM final concen-
tration). After incubation at 20 °C for 24 h, the cells were har-
vested by centrifugation, and stored at −20 °C until needed.
Frozen pellets were individually thawed, resuspended in
10 ml of BugBuster protein extraction reagent containing Ben-
zonase® nuclease (250 units) and rLysozyme™ solution
(10 000 units) at room temperature for 20 min. Each lysed solu-
tion was centrifuged (4300 × g, 25 min), with the supernatant
filtered through a 0.2 μm syringe filter (Pall).
Recombinant protein purification was carried out at 4 °C
on a POROS 20 metal chelate column (60 × 10 mm) pre-equili-
brated in binding buffer (20 mM Tris-HCl, pH 7.9, 500 mM
NaCl, 5 mM imidazole) at a flow rate of 4 ml min−1. Aliquots
(∼10 ml) of each recombinant protein preparations were next
loaded onto the affinity column, with the latter first washed
with 10% elution buffer (20 mM Tris-HCl, pH 7.9, 500 mM
NaCl, 500 mM imidazole) in binding buffer, after which a gra-
dient from 10 to 100% was applied in 25 min. Fractions
eluting between ∼120 and ∼180 mM imidazole were combined
and concentrated with Centricon Plus-70 (Amicon), diluted
with 20 mM Tris-HCl buffer (pH 7.5), and concentrated to
10–20 mg ml−1 (Centricon Plus-70, Amicon). Final purity was
confirmed by SDS-PAGE (4–15% acrylamide, BioRad) with
silver staining (Amersham Bioscience Application Note).
Protein concentrations were determined by the Bradford
method39 using γ-globulin as standard.
Experimental
Materials
p-Coumaryl (1) aldehyde and p-coumaryl alcohol (6) were syn-
thesized as in Kim et al.1 NADPH (tetrasodium salt) was
obtained from Sigma. Water (Optima® LC/MS) and acetonitrile
(Optima® LC/MS) were obtained from Fisher Scientific, and
glacial AcOH (HPLC grade) was procured from Mallinckrodt
Baker.
The QuikChange® XL Site-Directed Mutagenesis Kit was
purchased from Stratagene, whereas pTrcHis2-TOPO vector
and One Shot TOP 10 competent E. coli cells were from Invitro-
gen. BugBuster® protein extraction reagent, Benzonase® nucle-
ase and rLysozyme™ solution, were obtained from Novagen.
The POROS 20 metal chelate resin was from Applied
BioSystems.
Instrumentation
Polymerase chain reactions (PCR) were carried out using a
PTC-0220 DNA engine dyad Peltier thermal cycler (MJ
Research). Recombinant protein purification was carried out
on a Fast Protein Liquid Chromatography (FPLC, Amersham
Pharmacia Biotech) system using a column packed with
POROS 20 metal chelate resin.
Reversed-phase chromatography employed an Ultra Per-
formance Liquid Chromatography® system (UPLC, Waters)
equipped with a BEH shield RP 18 column (Waters); flow rate
of 0.3 ml min−1; detection at 280 nm. The solvent system con-
sisted of a concave gradient (Waters curve #8) of CH3CN : 3%
AcOH (v/v) in H2O from 5 : 95 to 40 : 60 between 0 and
8.30 min.
Assay of AtCAD5 and its mutants
Standard assays were as described in Kim et al.1 with initial
velocity kinetics data carried out in presence of 50 ng ml−1
AtCAD5, 105 ng ml−1 AtCAD5 H52A, 100 ng ml−1 AtCAD5
D57A, and 14.3 μg ml−1 AtCAD5 T49A. Final p-coumaryl alde-
Isothermal titration calorimetry measurements used a
VP-ITC microcalorimeter (MicroCal Inc.).
Site-directed mutagenesis
AtCAD5, cloned into the pTrcHis2-TOPO vector, was used as hyde (1) concentrations were: 0.76–80 μM for AtCAD5,
previously described.1 Primers were designed and synthesized 0.76–200 μM for AtCAD5 H52A and D57A, and finally,
(Invitrogen) to individually change Thr49, His52 and Asp57 7.6–480 μM for AtCAD5 T49A, respectively. Assays, carried out
into Ala (Table 1). Site-directed mutagenesis was carried out in quadruplicate, were initiated by enzyme addition, incubated
using a QuikChange® XL Site-Directed Mutagenesis Kit follow- at 30 °C for 1 min (2 min for AtCAD5 T49A) and stopped by
ing the manufacturer’s instructions with PCR performed as addition of glacial AcOH (10 μl). Aliquots (10 μl) from each
follows: 95 °C for 1 min, 18 cycles at 95 °C for 50 s, 60 °C for assay mixture were subjected to reversed phase UPLC as
50 s, 68 °C for 4 min, and 68 °C for 7 min. Parental plasmid described in the Instrumentation section. For each enzyme
DNA was next digested with Dpn
I restriction enzyme preparation, kinetic data was calculated using Origin 7.5
(10 units), at 37 °C for 1 h, with the resulting mutated plasmid (OriginLab).
1132 | Org. Biomol. Chem., 2013, 11, 1127–1134
This journal is © The Royal Society of Chemistry 2013

View Article Online
Organic & Biomolecular Chemistry
Paper
Isothermal titration calorimetry (ITC)
7 B. Youn, R. Camacho, S. G. A. Moinuddin, C. Lee,
L. B. Davin, N. G. Lewis and C. Kang, Org. Biomol. Chem.,
2006, 4, 1687–1697.
8 H. Eklund, B. V. Plapp, J. P. Samama and C. I. Brändén,
J. Biol. Chem., 1982, 257, 14349–14358.
9 O. Kleifeld, A. Frenkel, J. M. L. Martin and I. Sagi, Nat.
Struct. Mol. Biol., 2003, 10, 98–103.
10 J. P. Klinman, Crit. Rev. Biochem. Mol. Biol., 1981, 10,
39–78.
11 R. Ladenstein, J. O. Winberg and J. Benach, Cell. Mol. Life
Sci., 2008, 65, 3918–3935.
12 L. A. LeBrun, D.-H. Park, S. Ramaswamy and B. V. Plapp,
Biochemistry, 2004, 43, 3014–3026.
13 G. Pettersson, Crit. Rev. Biochem. Mol. Biol., 1987, 21,
349–389.
14 J. H. McKie, R. Jaouhari, K. T. Douglas, D. Goffner,
C. Feuillet, J. Grima-Pettenati, A. M. Boudet, M. Baltas
and L. Gorrichon, Biochim. Biophys. Acta, 1993, 1202,
61–69.
Titrations were carried at constant pressure and at 30 °C (for
cofactor/substrate binding) or at 25 °C (for order of binding
study). Twenty nine injections (10 μl each) at 200 s intervals
were performed, with the first injection not used for data
analysis. The stirring speed was 300 rpm.
For cofactor/substrate binding, the sample cell (1.4 ml) was
filled with either AtCAD5 or its mutants at a final concen-
tration of 50 μM (of homodimer) in HEPES buffer (20 mM, pH
6.8) containing 100 mM NaCl (Buffer A). The injection syringe
(250 μl) contained NADPH or p-coumaryl aldehyde (1) (1 mM)
in Buffer A. Titrations were carried out in triplicate. A control
titration was carried out by injecting ligand in Buffer A in
order to determine the dilution/mixing heat which was then
subtracted for each experiment prior to data analysis.
To study the order of binding, p-coumaryl aldehyde (1,
1 mM) was titrated into a cell containing AtCAD5 (50 μM,
homodimer) and NADPH (75 μM) in Buffer A. In another
experiment, NADPH (1 mM) was titrated into a cell containing
AtCAD5 (50 μM, homodimer) and p-coumaryl aldehyde (1,
75 μM). ITC analyses were then carried out as above.
15 N. G. Lewis, L. B. Davin and S. Sarkanen, in Comprehensive
Natural Products Chemistry, ed. Sir D. H. R. Barton, K. Naka-
nishi and O. Meth-Cohn, 1999, Vol 3, pp. 617–745.
16 R. L. Mansell, G. G. Gross, J. Stöckigt, H. Franke and
M. H. Zenk, Phytochemistry, 1974, 13, 2427–2435.
17 M. Klischies, J. Stöckigt and M. H. Zenk, Phytochemistry,
1978, 17, 1523–1525.
The Origin 5.0 software (MicroCal) was utilized to obtain
dissociation constant Kd, and binding enthalpy (ΔH) for each.
18 V. Lauvergeat, K. Kennedy, C. Feuillet, J. H. McKie,
L. Gorrichon, M. Baltas, A. M. Boudet, J. Grima-
Pettenati and K. T. Douglas, Biochemistry, 1995, 34,
12426–12434.
19 R. J. Falconer, A. Penkova, I. Jelesarov and B. M. Collins,
J. Mol. Recognit., 2010, 23, 395–413.
20 C. T. Morgan, R. Tsivkovskii, Y. A. Kosinsky,
R. G. Efremov and S. Lutsenko, J. Biol. Chem., 2004, 279,
36363–36371.
21 A. Grunau, M. J. Paine, J. E. Ladbury and A. Gutierrez, Bio-
chemistry, 2006, 45, 1421–1434.
Acknowledgements
This research was supported in part by a grant from the
Chemical Sciences, Geosciences and Biosciences Division,
Office of Basic Energy Sciences (DE-FG-0397ER20259), and the
G. Thomas and Anita Hargrove Center for Plant Genomic
Research. The latter support partially funded the graduate
student stipend of C. Lee. The authors thanks Dr Verna Frasca
(G.E. Healthcare Life Sciences) for valuable discussions on ITC
analyses.
22 V. Gonzalez-Covarrubias, D. Ghosh, S. S. Lakhman,
L. Pendyala and J. G. Blanco, Drug Metab. Dispos., 2007, 35,
973–980.
23 R. Perozzo, G. Folkers and L. Scapozza, J. Recept. Signal
Transduction, 2004, 24, 1–52.
Notes and references
1 S.-J. Kim, M.-R. Kim, D. L. Bedgar, S. G. A. Moinuddin,
C. L. Cardenas, L. B. Davin, C. Kang and N. G. Lewis, Proc. 24 A. Ciulli, D. Y. Chirgadze, A. G. Smith, T. L. Blundell and
Natl. Acad. Sci. U. S. A., 2004, 101, 1455–1460. C. Abell, J. Biol. Chem., 2007, 282, 8487–8497.
2 The Arabidopsis Genome Initiative, Nature, 2000, 408, 25 S. Dutta and K. Rittinger, PLoS One, 2010, 5, e10478.
796–815.
26 H. Li, S. Bai, J. Y. Wei, S. A. Berkowitz and M. L. Brader,
3 M. A. Costa, R. E. Collins, A. M. Anterola, F. C. Cochrane,
J. Pharm. Sci., 2011, 100, 4597–4606.
L. B. Davin and N. G. Lewis, Phytochemistry, 2003, 64, 27 X. Li, D.-C. Chow and S.-C. Tu, Biochemistry, 2006, 45,
1097–1112. 14781–14787.
4 R. Sibout, A. Eudes, G. Mouille, B. Pollet, C. Lapierre, 28 Q. Wan, Md F. Ahmad, J. Fairman, B. Gorzelle, M. de la
L. Jouanin and A. Séguin, Plant Cell, 2005, 17, 2059–2076.
5 M. Jourdes, C. L. Cardenas, D. D. Laskar,
Fuente, C. Dealwis and M. E. Maguire, Structure, 2011, 19,
700–710.
S. G. A. Moinuddin, L. B. Davin and N. G. Lewis, Phytochem- 29 D. Wyrambik and H. Grisebach, Eur. J. Biochem., 1979, 97,
istry, 2007, 68, 1932–1956. 503–509.
6 B. Persson, J. Hedlund and H. Jörnvall, Cell. Mol. Life Sci., 30 P. E. Carrigan, P. Ballar and S. Tuzmen, Methods Mol. Biol.,
2008, 65, 3879–3894.
2011, 700, 107–124.
This journal is © The Royal Society of Chemistry 2013
Org. Biomol. Chem., 2013, 11, 1127–1134 | 1133

View Article Online
Paper
Organic & Biomolecular Chemistry
31 L. Michaelis and M. L. Menten, Biochem. Z., 1913, 49, 333– 35 E. Valencia, C. Larroy, W. F. Ochoa, X. Parés, I. Fita and
369. J. A. Biosca, J. Mol. Biol., 2004, 341, 1049–1062.
32 T. Ehrig, T. D. Hurley, H. J. Edenberg and W. F. Bosron, Bio- 36 W. E. Karsten and P. F. Cook, Biochemistry, 2006, 45, 9000–9006.
chemistry, 1991, 30, 1062–1068.
33 W. F. Bosron, L. J. Magnes and T. K. Li, Biochemistry, 1983,
22, 1852–1857.
34 T. D. Hurley, H. J. Edenberg and W. F. Bosron, J. Biol.
Chem., 1990, 265, 16366–16372.
37 J.-O. Höög, H. Eklund and H. Jörnvall, Eur. J. Biochem.,
1992, 205, 519–526.
38 M. Estonius, J.-O. Höög, O. Danielsson and H. Jörnvall, Bio-
chemistry, 1994, 33, 15080–15085.
39 M. Bradford, Anal. Biochem., 1976, 72, 248–254.
1134 | Org. Biomol. Chem., 2013, 11, 1127–1134
This journal is © The Royal Society of Chemistry 2013