R. Zhang et al. / Archives of Biochemistry and Biophysics 596 (2016) 1e9
3
transition state was generated using nudged elastic band (NEB)
approach [41] implemented in NWChem. The NEB approach opti-
mizes the trial reaction pathway and a total of 10 beads were used
for pathway representation.
A0 ꢀ At
RateðmM=sÞ ¼
ꢁ 1000
(2)
20:7 ꢁ ðt0 ꢀ ttÞ
Assuming that the reaction follows the MichaeliseMenten ki-
netics, Michaelis constant (Km) and maximum rate (Vmax) were
obtained from LineweavereBurk plot Equation (3), where S is the
concentration of orange G, V is the initial reaction rate measured in
the linear range of degradation (5e10 s).
2.5. Molecular docking
Molecular docking of orange G was performed with AutoDock
[42], version 4.2.3 to predict the possible binding sites. The struc-
ture of CPO for docking was taken from the high resolution X-ray
structure (PDB entry 2CIW). Minor modification of X-ray structure
was introduced to simplify docking as described previously [43],
i.e., glycosylation sites, manganese ion, and crystallographic water
in the PDB file were removed.
The orange G structures was built by MarvinSketch, version 6.2,
in the JChem software package (ChemAxon, Ltd.) and further
optimized with ORCA [44], version 2.9, using density function
theory (DFT) with B3LYP//6-31G*. AutoDockTools [45], version
1.5.4, was used to add Gasteiger charges to CPO (þ1.00 was added
manually on Fe) and orange G. During simulation, CPO structure
was kept rigid. Orange G was simulated in a box centered at the
1
Km
1
1
¼
,
þ
(3)
V
Vmax ½Sꢂ Vmax
The turnover number, kcat, was calculated according to Equation
(4), where E0 is the concentration of CPO.
Vmax ¼ kcat½E0ꢂ
(4)
2.3. NMR characterization of degradation products from orange G
The NMR experiments were carried out on a Bruker 600 MHz
NMR spectrometer operating at a proton frequency of 599.73 MHz.
All spectra were recorded at 300 K and two-dimensional data were
processed using NMRPipe [33]. The COSY experiment was per-
formed using the standard Bruker pulse sequence (cosyprqf) with
4096 data points in the F2 dimension, and 512 increments in F1. The
data was processed to give a matrix of 2048 ꢁ 512 points, and a sine
bell apodization was applied before Fourier transformation.
Chemical shift values were referenced to the residual HDO signal at
4.76 ppm.
heme iron, which was confined using
a
grid size of
30 Å ꢁ 30 Å ꢁ 30 Å with 0.375 Å spacing. Docking consisting of 60
separate simulation runs was performed with 25 million energy
evaluations per run.
3. Results and discussion
3.1. UV-Vis study of CPO-catalyzed degradation of orange G
UV-Vis spectra of orange G (0.01 mM) were recorded in the
presence of 0.01 m
M CPO, 20 mM chloride ion (Clꢀ), and 0.5 mM
2.4. QM/MM study on the formation of HClO by compound X
hydrogen peroxide at pH 2.7. Fig. 1A illustrates the typical UV-Vis
spectra of orange G obtained during the degradation process.
Before addition of hydrogen peroxide (H2O2), the spectrum of or-
ange G showed a strong absorption at 477 nm (lmax) with a
shoulder at 419 nm (Fig. 1A, black line). These absorption bands
originate from the azo bond (eN]Ne) of orange G that undergoes
azo-hydrazone tautomerization [46]. The weak absorption at
The starting structure for CPO was taken from the high resolu-
tion X-ray structure (PDB entry 2CIW). The cyanide ligand in the
heme center was manually replaced by ClOꢀ ligand. The structure
was solvated in a truncated octahedron box (82 Å) of extended
simple point charge (SPC/E) waters. A total of 7624 waters were
contained in the system. The protonation states of the titratable
amino acid residues of CPO were assigned as follows: þ1 for His and
Lys, and neutral for Tyr, Glu and Asp. Therefore, the protein has a
net charge of þ20, thus 20 Clꢀ counter-ions were added to create a
neutral system for simulation.
The hybrid quantum mechanics/molecular mechanics (QM/
MM) calculations were performed using software NWChem [34],
version 6.3. The overall algorithm involves alternating optimiza-
tions of QM and MM regions until convergence is achieved. The QM
region consisted of heme ligated with sulfur atom (representing
cysteine 29) as well as the side chain functional groups of His 105
and Glu 183. The substituents of the heme were not included to
simplify calculation. This simplification used in DFT calculations is
made as the compound I geometries do not vary noticeably with
varying QM region sizes [35]. The covalent bonds crossing to the
MM region were capped with hydrogen atoms. The DFT geometry
optimizations were performed at the spin-unrestricted B3LYP
[36,37] level with LACVP basis sets (LANL2DZ [38] effective core on
iron and 6-31G [39] on other atoms). The single point energy
calculation was performed at the spin-unrestricted B3LYP with
LANL2DZ on iron atom and 6-31 þ G* on other atoms. The
remainder of the protein (MM region) was described at the mo-
lecular mechanics level using an AMBER95 force field [40]. During
optimization of MM regions, the electrostatic potential (ESP)
charges for QM region was calculated and used to compute elec-
trostatic interactions with the MM regions. The initial guess for
329 nm (Fig. 1A, black line) is due to the
p
to p* transition in the
naphthalene structure [47]. Five minutes after addition of H2O2, the
absorption at 477 nm significantly decreased (Fig. 1A, red line),
indicating the cleavage of orange G. Essentially identical spectral
changes were observed after hypochlorous acid (HClO) alone was
added to orange G (Fig. 1A, blue line), indicating the formation of
similar products as in the CPO-H2O2eClꢀ system. It is worth
mentioning that degradation of orange G is negligible in the pres-
ence of CPO and H2O2 without Clꢀ (Fig. 1A, yellow line), suggesting
the imperative role of Clꢀ in CPO-catalyzed degradation of azo dyes.
Fig. 1B shows the kinetic behavior of orange G at 477 nm during
CPO-catalyzed degradation. After addition of H2O2, the absorbance
of orange G decreased continuously and approached baseline in
approximately 70 s (Fig. 1B), indicating a fast and efficient degra-
dation. The degradation rate was calculated to be 21 in the first 10 s,
30 between 10 and 30 s, and 4 mmole substrate/s per micro-mole of
CPO between 30 and 70 s. The decrease of the absorbance was
linear within a short period after reaction is initiated (10e30 s)
(Fig. 1B, inset). Thus, the reaction rate was calculated within this
linear range and discussed below unless otherwise specified.
Table 1 lists the degradation efficiency of a variety of synthetic
dyes including orange G, methyl orange, nuclear fast red, gentian
violet, and azure B in the CPO-H2O2eClꢀ system. The degradation of
all dyes (0.01 mM) proceeded with high efficiency (70.4e99%
within 180 s). This result demonstrates that the CPO-H2O2eClꢀ
system is highly efficient to degrade different types of synthetic