Tuning the substrate specificity by engineering the active site of cytochrome P450cam: A rational approach

Article abstract of DOI:10.1039/b922885c

Rational design of the active site of cytochrome P450cam has been carried out to catalyse oxygenation of various potentially important chemical reactions. The modeling studies showed that the distal pocket of the heme consisting of the Y96, T101, F87 and L244 residues could be suitably mutated to change the substrate specificity of the enzyme. We found that the mutant enzymes could catalyse oxygenation of indole to produce indigo. While Y96F was found to be several times better as a catalyst for conversion of indole to indigo, the double mutant Y96F/L244A showed the highest NADH oxidation rate as well as yield of indigo. The oxidative catalysis using H₂O₂ as the oxygen source was found to produce a higher purity of indigo, and lesser or no formation of indirubin was detected. The enzymatic oxygenation of aromatic hydrocarbons such as coumarin and analogues was also found to be enhanced on mutation of Y96 and L244 residues in the enzyme. The studies also showed that mutation of suitable residues can alter the regio-selectivity of hydroxylation of the aromatic hydrocarbons.

Full text of DOI:10.1039/b922885c

View Article Online / Journal Homepage / Table of Contents for this issue
This paper is published as part of a Dalton Transactions themed issue on:
Bioinspired Catalysis
Guest Editor Seiji Ogo
Kyushu University, Japan
Published in issue 12, 2010 of Dalton Transactions
Image reproduced with permission of Seiji Ogo
Articles published in this issue include:
PERSPECTIVES:
Mimicking Nitrogenase
Ian Dance
Dalton Trans., 2010, DOI: 10.1039/b922606k
Dual functions of NifEN: insights into the evolution and mechanism of nitrogenase
Yilin Hu, Aaron W. Fay, Chi Chung Lee, Jared A. Wiig and Markus W. Ribbe
Dalton Trans., 2010, DOI: 10.1039/b922555b
Quest for metal/NH bifunctional biomimetic catalysis in a dinuclear platform
Takao Ikariya and Shigeki Kuwata
Dalton Trans., 2010, DOI: 10.1039/b927357c
COMMUNICATION:
Chemiluminescence enhancement and energy transfer by the aluminium(III) complex of an
amphiphilic/bipolar and cell-penetrating corrole
Atif Mahammed and Zeev Gross
Dalton Trans., 2010, DOI: 10.1039/b916590h
Visit the Dalton Transactions website for more cutting-edge inorganic and bioinorganic research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
Tuning the substrate specificity by engineering the active site of cytochrome
P450cam: A rational approach
Soumen Kanti Manna and Shyamalava Mazumdar*
Received 2nd November 2009, Accepted 8th February 2010
First published as an Advance Article on the web 16th February 2010
DOI: 10.1039/b922885c
Rational design of the active site of cytochrome P450cam has been carried out to catalyse oxygenation
of various potentially important chemical reactions. The modeling studies showed that the distal pocket
of the heme consisting of the Y96, T101, F87 and L244 residues could be suitably mutated to change
the substrate specificity of the enzyme. We found that the mutant enzymes could catalyse oxygenation
of indole to produce indigo. While Y96F was found to be several times better as a catalyst for
conversion of indole to indigo, the double mutant Y96F/L244A showed the highest NADH oxidation
rate as well as yield of indigo. The oxidative catalysis using H2O2 as the oxygen source was found to
produce a higher purity of indigo, and lesser or no formation of indirubin was detected. The enzymatic
oxygenation of aromatic hydrocarbons such as coumarin and analogues was also found to be enhanced
on mutation of Y96 and L244 residues in the enzyme. The studies also showed that mutation of suitable
residues can alter the regio-selectivity of hydroxylation of the aromatic hydrocarbons.
oxidize almost any organic compound. The most common func-
Introduction
tion of these enzymes is to carry out hydroxylation of organic
Functionalization of the C–H bond is at the heart of synthetic
compounds. However, they are also known to catalyze various
organic chemistry that leads to the production of the huge
other types of reactions such as epoxidation, dehydrogenation,
dealkylation, dehalogenation etc.9,10
arsenal of organic molecules which are essential components of
modern life.1,2 The synthetic strategy to introduce a functional
Apart from the fact that all of these reactions catalyzed by the
group generally involves the attachment of a heteroatom, often
enzyme take place under ambient conditions, they do away with
oxygen through oxidation, in place of the hydrogen atom of the
any environmentally hazardous heavy metal or organic solvent.
C–H bond. In the traditional organic synthesis, the oxidation
Thus the cytochrome P450s offer a potentially ‘green’ alternative
of the inert organic compounds are often requires high-valent
solution to the chemistry of oxidation of organic compounds.
heavy metal compounds like chromates, dichromates, chromyl
Cytochrome P450cam (P450cam) has been one of the most ex-
ploited P450s for such a biotechnological application.11 The native
enzyme has long been shown to bind and catalyze the oxygenation
of camphor analogues albeit with lower efficiency than that for
camphor.12 Lefever and Wacket reported the potential use of this
enzyme for the degradation of the organochlorine compounds for
the first time, where they showed that the wild-type enzyme can
oxidize low molecular weight chloroalkanes.13 The active site of the
enzyme has been modified extensively by L.-L. Wong’s group14-21
at Oxford and others22,23 to accommodate a variety of substrates,
which are entirely different in size and structure compared to the
natural substrate, 1R-camphor.
The P450cam variants have been shown to catalyze oxidation of
both open chain as well as cyclic aliphatic compounds of different
sizes such as, propane to heptane and substituted heptanes.17
Several studies have demonstrated that they can catalyze the oxi-
dation of different aromatic compounds, including polyaromatic
hydrocarbons (PAH) as well.15 The engineered P450cam variants
have also been explored for fine chemicals synthesis.16
chloride, ammonium cerium nitrate, bismuthate, vanadate, Ru-,
Os-, Ir-, Pb-containing inorganic and organometallic compounds
etc.3,4 Many of these reagents show poor product selectivity and
overall activity.3 These heavy metal compounds and complexes
are generally toxic and, therefore, biohazardous in nature.5,6
For example, chromium(VI) compounds have been shown to
be carcinogenic and mutagenic in nature.7 In addition these
reactions are almost invariably carried out in organic solvents
and many commonly used organic solvents (e.g., CH2Cl2, ben-
zene, toluene, etc.) are suspected carcinogens and environmen-
tal pollutants (www.oehha.org/risk/ChemicalDB/index.asp and
http://potency.berkeley.edu/index.html). Moreover, many of the
oxidation reactions are carried out at extremes of pH, under higher
pressure and at elevated temperature.8 Thus it is essential to find
out efficient but safe alternatives for carrying out organic synthesis
in a more environmentally friendly manner.
Living organisms are outstanding examples of efficient reactors
and all these reactions are carried out under ambient conditions
by enzymes in water. Cytochrome P450s are the most versatile
biological oxygenation catalysts. These enzymes can potentially
One of the most commonly used dyes by mankind since
ancient times is indigo.24 With the advent of organic chemistry,
the industrial production of this dye has replaced the agricultural
production by adopting the synthetic routes (see Fig. 1A & B)
reported by Heumann in 1890 and later on by Pfleger in 1901.25
In both cases, the synthesis not only involves harmful corrosive
Department of Chemical Sciences, Tata Institute of Fundamental Research,
Homi Bhabha Road, Colaba, Mumbai, 400 005, India. E-mail: shyamal@
tifr.res.in; Web: http://www.tifr.res.in/~shyamal; Fax: +91 22 2280 4610;
Tel: +91 22 2278 2363
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3115–3123 | 3115
©

expected to increase the solubility and hence the bioavailability
of these compounds for degradation. Moreover, biodegradation
of many organic compounds (http://umbbd.msi.umn.edu/) has
been shown to involve the oxygenation of the parent compound or
the intermediate.29 Thus the oxidation of the aromatic compounds
by cytochrome P450cam could provide a ‘green’ solution for a
healthier environment.15
However, one of the main bottlenecks in the application of
P450cam for biotechnological purposes remains the poor affinity
of the enzyme for unnatural substrates. The catalytic cycle is
turned on by the spin-state change at the heme active site due to
expulsion of the axial water molecule on substrate binding, which
is followed by electron transfer from NADH.12 Poor substrate
binding affinity and incomplete spin-transition has been shown to
decrease the catalytic efficiency of the enzyme.12 The present work
mainly focuses on the enhancement of the binding affinity, high-
spin content and electron transfer rate through rational redesign
of the active site. The designs of the P450cam variants have been
based on the following three point strategies:
1) Increase in the active site space for bigger molecules,
2) Decrease in the active site space for smaller and flatter
molecules, and
3) The enhancement of the substrate binding affinity by
complementary non-covalent interactions between the substrate
and suitable amino acid side-chain in the active site.
Fig.
1 Schematic representation of indigo synthesis through (A)
Heumann’s method, (B) Pfleger’s method and (C) the proposed P450cam
catalyzed pathway. (D) Shows the structures of hydrocarbons studied in
the present report.
We have taken the approach of site directed mutagenesis
to achieve the goal, which if successful, would not only help
to enhance the binding of the unnatural substrates but also
provide useful information to understand the determinants of
substrate–protein interactions that can be exploited for better
design of the catalyst. Fig. 2 schematically represents the protein
engineering strategy to accommodate unnatural substrates inside
the P450cam active site. The binding affinity, electron transfer rate
and the catalytic activity of the P450cam variants towards different
substrates were measured to check the potential of these enzymes
for biotechnological applications.
chemicals like NaOH or KOH at high temperatures but they are
also carried out in organic solvents. In addition, N-phenyl glycine,
the precursor for the industrial indigo production is synthesized
from aniline using cyanide.25
The ancient as well as industrial production of this dye involves
the synthesis of 3-hydroxyindole or indoxyl, which subsequently
gets oxidized by aerial oxygen to indigo. The synthesis of indigo
by heating in alkali solutions and the hydroxylated products
of these reactions are often more toxic than the unsubstituted
parent compounds.25 We have explored the ability of the wild type
and various specifically engineered P450cam variants to produce
indigo from indole (Fig. 1C).
Thehydroxylationofaromaticcompounds involves nucleophilic
attack of the hydroxyl group on the aromatic ring. The nucleophilic
substitutions on aromatic rings are difficult and generally carried
out under harsh conditions.8 Moreover the tailor-made regio-
selectivity of aromatic hydroxylation is difficult to achieve via
conventional organic synthesis as the substitution takes place
preferably at the -ortho/-para or -meta positions depending on
the existing functional group on the aromatic ring.8 However, in
the case of the reactions by P450s the position of hydroxylation is
dictated by the orientation of the substrate inside the active site. In
view of the potential application of their hydroxyl derivatives,26,27
we have selected a few important aromatic compounds (Fig. 1D)
to test the feasibility of this hypothesis.
Fig. 2 Schematic representation of the active site engineering strategy to
catalyze the oxidation of unnatural substrates using P450cam.
The reason for the high biological persistence of PAHs
is the inability of the microorganisms to degrade them due
to their chemical inertness and poor solubility. It has been
shown that increases in the solubility increases the degradabil-
ity of environmental pollutants.28 Introduction of hydrophilic
substituents like, the hydroxyl group, into these compounds is
Experimental methods
Restriction enzymes and buffers for molecular biology works
were obtained from New England Biolabs. DEAE-Sepharose,
3116 | Dalton Trans., 2010, 39, 3115–3123
This journal is
The Royal Society of Chemistry 2010
©

Table 1 Primers used for preparing cytochrome P450cam mutants
desalted using a Sephadex G25 size exclusion column. The
desalted protein was finally purified using a Resource-Q (6 ml,
Amersham Biosciences) anion exchange chromatography column
on an AKTA FPLC system (Amersham Biosciences). The purity
of the protein was checked by both UV-visble spectroscopy
(A390/A280 > 1.5 or A404/A280 > 1.2) and SDS-PAGE. The purified
fractions were concentrated using an Amicon ultrafiltration
membrane (YM-30, Amersham Biosciences) or Centricon tubes
(3 kDa cut-off membrane, Amersham Biosciences) and stored with
40% (v/v) glycerol at -30 ◦C.
The wild type putidaredoxin (PdX) and putidaredoxin reductase
(PdR) were purified using a reported procedure and the concen-
tration of PdX was determined from the absorbance at 455 nm
(e455 = 5.9 mM-1 cm-1)30 and that of PdR was estimated from the
absorbance at 454 nm (e454 = 10.9 mM-1 cm-1).31
Restriction
Mutation Site
Oligonucleotide primers
Y96F
Y96V
Y96W
F87W
T101V
Bsam I
5¢ GCCGGCGAAGCaTtCGACTTCAT 3¢
5¢ ATGAAGTCGaAtGCTTCGCCGGC 3¢
5¢GCCGGCGAAGCagtgGACTTCATTC 3¢
5¢ GAATGAAGTCcactGCTTCGCCGGC 3¢
5¢ GCCGGCGAAGCtTggGACTTCAT 3¢
5¢ ATGAAGTCccAaGCTTCGCCGGC 3¢
5¢ GCGAGTGCCCaTggATCCCTCGTG 3¢
5¢ CACGAGGGATccAtGGGCACTCGC 3¢
5¢ CTTCATTCCCgtaTCGATGGATCCG 3¢
5¢ CGGATCCATCGAtacGGGAATGAAG 3¢
5¢ CTTCATTCCCgtaTCGATGGATCCG 3¢
Bts I
Hind III
Nco I
BamH I
Cla I
Y96F/
T101V
(FV)
Cla I
5¢ CGGATCCATCGAtacGGGAATGAAG 3¢
5¢ GGATGTGTGGCgccTTACTGGTCGG 3¢
Y96F/
L244A
(FA)
Sfo I
NADH consumption rate
5¢ CCGACCAGTAAggcGCCACACATCC 3¢¢¢
The NADH serves as the source of electrons during the catalytic
cycle of the P450 enzyme to activate molecular oxygen and finally
to hydroxylate camphor. The oxidation of NADH takes place
by transfer of two electrons to P450cam through putidaredoxin
reductase (PdR) and putidaredoxin (PdX). The reduced form of
NADH shows an absorption band at 340 nm (e = 6.22 mM-1 cm-1)
which disappears on oxidation. The rate of NADH oxidation
during catalysis was determined by monitoring the rate of change
of absorption at 340 nm. The reaction mixture containing 1 mM
PdR, 16 mM PdX, 0–1 mM P450cam variant, 100 mM KCl in
50 mM Tris (pH 7.4) was incubated at 25 ◦C to determine the
background aerial oxidation rate in the absence of any catalysis.
The catalytic reaction was initiated by addition of the substrate
to the reaction mixture. The NADH oxidation rate was calculated
from the slope of the initial linear part of the trace after subtraction
of the background oxidation rate.
Q-sepharose and Sephadex G-25 columns were from Pharmacia
Biotech. General Reagents, indole, coumarin, (1R)-camphor, 4-
aminoantipyrene (4-AAP), hydrogen peroxide etc., were from
Sigma, India. UV-vis absorption spectra were measured on a
Shimadzu (UV-2100) spectrophotometer coupled with a Peltier
controlled thermostated cell holder. CD spectra were measured
on a JASCO J-810 spectropolarimeter equipped with a Peltier cell
temperature controller ( 0.2 ◦C).
Mutation analysis
The pCHC1 plasmid, encoding cytochrome P450cam C334A,17
was a kind gift from Prof. L.-L. Wong (University of Oxford,
UK). The plasmids encoding wild type putidaredoxin (PdX) and
putidaredoxin Reductase (PdR) were kindly supplied by Prof. S. G.
Sligar (University of Illinois, USA). The site directed mutagenesis
of the cytochrome P450cam mutant were carried out using
the QuikchangeTM Site-Directed mutagenesis kit (Stratagene).
The plasmids were isolated from freshly transformed E. coli
(BL21DE3) cells and the purity was checked by running DNA
gel (0.7% agarose). The primers for the P450cam mutants were
designed in a way so that incorporation of the desired point
mutations also introduces appropriate restriction sites into the
P450cam gene (camC). The Table 1 shows the sequences of
the primers used. The appearance of the characteristic bands
compared to the wild type plasmid in the DNA gel of the digested
plasmid was used to analyze and confirm the incorporation of the
target mutation. The mutation of the enzyme was also confirmed
by DNA sequencing as well as by comparing the agarose gels of
the products of restriction digestion of the plasmid containing the
mutation with those of the wild type enzyme.
Indole oxidation
The oxidation of indole via the reductive cycle was performed in
the presence of 0.2 mM PdR, 2 mM PdX, 0.2 mM P450cam variant,
100 mM KCl, 1 mM NADH and 0.5 mM indole in 50 mM Tris
(pH = 7.4). The reaction mixture was kept at room temperature
for two days and extracted with CHCl3 for spectrophotometric or
ESI-MS studies.
The oxidation of indole via the peroxide dependent oxidative
cycle was performed in the presence of 4 mM P450cam variant,
100 mM KCl, 0–60 mM H2O2 and 0.5 mM indole in 50 mM
Tris (pH = 7.4). The concentration of H2O2 (e200 = 200 M-1 cm-1)
stock was determined by spectrophotometry.32 The indigo shows
a characteristic absorption peak at 600 nm in CHCl3. Thus the
relative efficiency of indigo formation via indole oxidation by
P450cam variants were expressed in terms of the absorbance
of the CHCl3 extract at 600 nm which represents the relative
concentration of indigo. The control reactions for oxidative and
reductive indole oxidation were carried out in the presence of
all the ingredients in the respective catalytic mixtures except
the P450cam variants. The purity of the indigo produced by
the oxidative and reductive cycles was checked by thin layer
chromatography (TLC) of the extracted product.
Expression and purification of P450cam variants, putidaredoxin
and putidaredoxin reductase
The cytochrome P450 variants were expressed following a slight
modification of the reported method.17 The first stage of purifi-
cation of the wild type as well as the mutant P450cam involved
passing of the crude cell lysate through a DEAE Sepharose ion
exchange column using a salt gradient. This was subsequently
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3115–3123 | 3117
©

Oxidation of aromatic compounds
The formation of phenolic compounds on hydroxylation of the
aromatic hydrocarbons by P450cam variants in the reductive cycle
was checked by the 4-aminoantipyrene (4-AAP) assay for phenolic
compounds.33 The catalytic reaction mixture containing 0.2 mM
PdR, 4 mM PdX, 0.4 mM P450cam variant, 100 mM KCl, 1.25 mM
NADH and 0.5 mM coumarin in 50 mM Tris (pH = 7.4) was
incubated at room temperature for 6 h. After the incubation, 2 ml
4-AAP (4% stock in DMSO) and 6 ml 10% ammonium persulfate
(APS) were mixed with the reaction mixture along with 40 ml
saturated NaHCO3 solution and kept at room temperature for
30 min to complete the reaction.
Oxidation of the substrates by mutant P450cam
The catalytic reaction mixture containing 1 mM PdR, 10 mM PdX,
0.1 mM P450cam variant, 0.2 mM NADH, 100 mM KCl, and
0.1 mM substrate in 50 mM Tris (pH = 7.4) was extracted in CHCl3
after completion of NADH oxidation. The reaction products were
analysed by Electrospray ionization mass spectrometry (ESIMS)
using a LCQ Deca (Thermo-Fisher) mass spectrometer. The
chloroform extract was mixed with methanol (1 : 1) and injected
into the ESI source of the instrument, at a flow rate of 5–
10 mL min-1, with 0.1% TFA to promote protonation of the
organic molecules. The source voltage was kept at +4.5 kV for
all experiments to create positively charged droplets which were
finally detected and analyzed using a Quadrupole Ion-trap mass
analyzer.
Fig. 3 Agarose gel electrophoresis of plasmid DNA and the restriction
digestion of the mutant P450cam (A) Undigested Y96F plasmid (lane 1),
Sfo I digested Y96F (lane 2) and Y96F/L244A (lane 3) plasmid along with
BenchTop 1 kb DNA Ladder marker (lane 4). (B) BenchTop 1 kb DNA
Ladder marker (lane 1) along with Bts I digested wild type (lane 2) and
Y96V (lane 3) plasmid.
heme active center of the enzyme and the UV-visible absorption
spectra of the mutant P450cam enzymes were similar to that of the
wild type enzyme with characteristic Soret absorption at 417 nm
and visible bands at ~530 nm and ~570 nm in the absence of
any substrate as shown in Fig. 4A. The overall structure of the
mutant enzyme was also identical to that of the wild type species
as indicated by the circular dichroism spectra of the enzymes in
the far-UV region (Fig. 4B). Analyses of the far-UV CD showed
the presence of ~40 1% a-helical structure which agrees with the
crystal structure of the wild type enzyme.12 The tertiary structure
around the heme was also not drastically altered on mutation and
the visible CD spectrum of the mutants as shown in Fig. 4C is
similar to that of the wild type enzyme although the relative ratios
of the visible CD bands of the mutants were slightly different than
that of the wild type enzyme owing to the small change in the
substrate binding site of the enzyme.
Results and discussion
Site specific mutation of P450cam
The substrate binding pocket of P450cam is lined with several
residues which have been shown to be involved in substrate binding
and catalysis by the enzyme.12 We designed mutations of the F87,
Y96, T101 and L244 residues to create a series of site-specific
mutants of P450cam and assayed the efficiency of these mutants.
The nature of the mutation was decided on the basis of the type of
substrate to be catalysed. Thus larger and hydrophobic amino
acids at the Y96 position were designed to accommodate flat
aromatic molecules more efficiently than the wild type enzyme.
Double mutations of the Y96F were also designed with T101V and
L244A mutations for reactions of indole as well as of coumarin.
The primers designed for mutations were such that each mutant
produced a restriction site in the gene (Table 1), e.g., the double
mutant F96F/L244A was made using the pair of primers that
introduces a new cleavage site for the restriction enzyme SfoI, while
such a restriction site was absent in the DNA of the Y96F mutant.
Agarose gel electrophoresis of the restriction digestion products
of the mutant plasmid was compared with that of the wild type
plasmid as shown in Fig. 3 and observation of the presence of the
new restriction site in the plasmid confirmed the success of the
mutation. The sequence of the mutant gene was also confirmed by
DNA sequencing wherever required.
Each of the mutant enzymes was expressed and purified for
further studies of substrate binding and enzymatic activity studies.
The mutations designed in the present study were found to have
no significant effect on the electronic environment around the
Fig. 4 Absorption spectra of the substrate-free (A) Y96F mutant of
P450cam. (B) Far-UV CD spectra of substrate-free P450cam. (C) Near-UV
and visible CD spectra of substrate-free Y96F mutant of P450cam.
3118 | Dalton Trans., 2010, 39, 3115–3123
This journal is
The Royal Society of Chemistry 2010
©

Table 2 NADH oxidation rate of P450cam variants in the presence of
indole (0.5 mM)
Variant
WT
Y96F
21
Y96F/L244A
24
Y96V
NADH oxidation rate
(nmol/nmolP450/min)
4
1
3
2
7
2
The Y96 position has been shown to form hydrogen bonds
with camphor in the substrate bound wild type enzyme.12 The
T101 residue was shown to be involved in stabilization as well
as substrate binding in the wild type P450cam.34 Extensive
site specific mutations of Y96, F87, L244 and several other
residues have been reported in the literature to alter the substrate
specificity of P450cam.12,14,15,19,21,23 L.-L Wong et al., showed that
combinations of mutations such as Y96F, F87A, F87L, F87W,
and V247L could alter the substrate recognition as well as product
selectivity of the enzyme for a-pinene, aromatic halides etc.14,17,19,20
Enzymatic reactions of 2-ethylhexanol were reported using F87W,
Y96W, T185F and L244A mutants of P450cam.22 However, the
enzymatic studies on mutation at the T101 site and combinations
of T101V with Y96 mutations have not been reported earlier. The
reactions of indole, coumarin and other aromatic hydrocarbons,
that have been studied in the present report, uses some of these
known mutants along with some new mutants of the enzyme.
Fig. 5 (A) Indigo synthesis from indole oxidation by P450cam variants
and (B) the absorption spectra of the indigo in CHCl3.
Synthesis of indigo from indole using P450cam mutants
The addition of indole to the substrate-free P450cam variants
was not found to cause any significant change in the absorption
spectrum of the enzyme. However, the P450cam variants were
found to oxidize NADH in presence of the substrate. These results
indicate that although there is no change in the steady state
spin equilibrium of the heme on binding of indole, it possibly
produces transient high-spin species, which may get reduced by
electrons from NADH and catalyze the oxygenation of indole.
The results (Table 2) also showed that the NADH consumption
rates for P450cam mutants like Y96F and Y96F/L244A (FA)
were, respectively, ~5 and 6 fold higher than that of the wild type
(WT) enzyme. Although the WT enzyme showed small NADH
oxidation activity in the presence of the indole (Table 2), there
was no detectable formation of indigo in the presence of the
WT enzyme. On the other hand, the P450cam mutants (Y96F,
Y96F/L244A and Y96V) were found to produce the blue dye
(Fig. 5A) through the reductive cycle. The absorption spectra of
the CHCl3 extract of the catalytic mixture containing these mutant
enzymes were found to consist of the characteristic peak of indigo
at 600 nm (Fig. 5B). The ESI-MS spectrum (Fig. 6) of the extract
showed peaks of unreacted indole along with that of indigo at
m/z values of 147 (indole + H+) and 263 (indigo + H+). This
also showed a small peak at m/z = 134 corresponding to indoxyl
suggesting that the formation of indigo, indeed, proceeded via
oxidation of indole at C3 as proposed in Fig. 1C. The relative
concentration of indigo was measured from the absorbance of the
CHCl3 extract at 600 nm. The results (Fig. 7) showed that the
Y96F/L244A mutant is most efficient in conversion of indole to
indigo, followed by Y96F and the Y96V mutants, through the
reductive cycle.
Fig. 6 ESI-MS spectrum of the CHCl3 extract of the reaction mixture (A)
in the absence of P450cam and (B) in the presence of Y96F. The numbers
indicate the m/z value of the respective peaks.
Fig. 7 Relative efficiency of indigo production by P450cam variants in
the reductive cycle.
Y96F mutants. This possibly increases the residence time of the
organic molecule in the pocket compared to the WT enzyme and
thus increases the NADH oxidation rate and the probability of
hydroxylation of indole. In the case of the Y96F/L244A mutation
the efficiency of indole oxidation is further enhanced possibly
due to creation of a larger space on the decrease of the chain
length of the 244th residue, which is placed on top of heme.
This might allow the planar indole molecule to access the distal
site more easily facilitating the hydroxylation of indole by the
putative oxo-ferryl species.12 However, since the binding of indole
cannot efficiently remove the water molecules from the active site,
Mutations of Y96 residue to Val and Phe increase the hy-
drophobicity of the substrate binding pocket in the Y96V and
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3115–3123 | 3119
©

uncoupling reactions may take place leading to poor efficiency
of indole oxidation through the reductive cycle. Consequently, a
significant amount of indole remains unreacted and the ESI-MS
spectrum of the CHCl3 extract of the reaction mixture, even after
reaction for 2 days, shows an intense indole peak (Fig. 6).
One of the limiting factors in the application of P450cam is the
requirement of the expensive cofactor (NADH) and the electron
transfer partners (PdR and PdX). However, the peroxide shunt
pathway12 provides a more economic and easier method for the
oxidation of organic compounds. Several P450cam mutants were
found to carry out peroxide dependent oxidation of substrates.35-37
We have studied whether the Y96F, Y96V and Y96F/L244A
mutants could carry out the catalysis of indole oxidation through
oxidative pathways using the ‘Peroxide Shunt’.12
The indole was not found to be converted to indigo by H2O2 in
the absence of the P450cam variants. The wild type enzyme as well
as the Y96V mutant was also not found to cause any significant
indigo production in the presence of H2O2. However, the Y96F and
the Y96F/L244A mutants were found to significantly enhance
the indigo formation from indole in the presence of H2O2.
The results (Fig. 8A) showed that the increase in the H2O2
concentration increases the production of indigo for both of these
mutants. However, the increase in the indigo production with H2O2
concentration was relatively more in presence of the Y96F/L244A
(Fig. 8B) compared to that in presence of the Y96F mutant. Thus
the Y96F/L244A mutant was found to have a higher efficiency
of conversion of indole to indigo in the reductive as well as the
oxidative cycle compared to the Y96F mutant.
Fig. 9 Thin layer chromatogram of the indole oxidation product via the
reductive and the oxidative cycle in the presence of the Y96F mutant.
band) along with indigo (blue band). However, the product of
the oxidative cycle was found to consist of a single blue band
corresponding to indigo. This suggests that the in vitro method of
indigo synthesis via the oxidative cycle of P450cam is associated
with a higher product selectivity compared to the reductive cycle
or to the in vivo method.40
Interaction of aromatic hydrocarbons with P450cam mutants
The addition of the aromatic compounds (Fig. 1D) was found to
have no significant effect on the UV-visible absorption spectra of
the WT P450cam. However, the absorption spectra of the mutants
were found to undergo significant change on addition of these
compounds. In contrast to that in case of indole, the addition of
indoline (NDLN) to the P450cam variants was found to induce a
red shift in the Soret maxima (Fig. 10A) and a Type II change in
the difference spectrum (Fig. 10B). This indicates the formation of
a direct Fe–N coordination in the enzyme–substrate complex. In
the case of indole the lone pair of nitrogen resides in a conjugated
‘p’ orbital as a part of the Huckel (4n+2) aromatic ‘p’ electron
system and therefore is not available for coordination to the iron
center of the enzyme. But in the case of indoline the nitrogen
lone pair resides in an ‘sp3’ orbital. Thus indoline nitrogen can
easily coordinate to the iron center leading to a Type-II shift in
the difference spectra of the enzyme. Similarly, quinoline (QNLN)
also forms a direct Fe–N coordination as indicated by a Type-II
shift in the difference spectra. However, since the nitrogen lone pair
in quinoline resides in an ‘sp2’ orbital, it coordinates less strongly
Fig. 8 (A) Variation of indole production with H2O2 concentration in
the oxidative cycle in presence of Y96F/L244A (squares), Y96F (circles)
and Y96V (triangles) mutants of P450cam. (B) Relative efficiency of
indigo production at different H2O2 concentrations in the presence of
Y96F/L244A (dense bars) and Y96V (empty bars) normalized with
respect to that in presence of the Y96F (sparse bars) mutant of P450cam.
The product selectivity is a very important issue with respect
to the viability of application of a synthetic scheme. Several
studies have shown that the synthesis of the blue dye, indigo,
is often associated with co-production of a isomeric pink dye,
indirubin.38 In fact, in vivo synthesis of indigo from indole
using P450cam variants (with mutations at the 96th and 98th
position), recombinant human CYP2E1 as well as toluene-4-
monoxygenase have been reported to be associated with co-
formation of indirubin.39,40 Hence the purity of the in vitro indole
oxidation products in the reductive as well as the oxidative cycle
was checked using TLC.
Fig. 10 The representative (A) absolute and (B) difference spectra of
quinoline (dotted line) and indoline (dash-dotted line) binding to the
substrate-free Y96F/L244A mutant (solid and dashed lines) of P450cam.
The enzyme and final substrate concentrations for the quinoline and
indoline binding were 4.5, 128 mM and 3.5, 63 mM, respectively.
The chromatogram (Fig. 9) showed that the indole oxidation
product formed in the reductive cycle consisted of indirubin (pink
3120 | Dalton Trans., 2010, 39, 3115–3123
This journal is
The Royal Society of Chemistry 2010
©

Table 3 Spectroscopic binding constant (Kb in mM-1) of aromatic hydrocarbons to P450cam variants
FAa FVa
Substrate WT
Y96F
Y96V
Y96W
F87W
T101V
CMRN (1.4 0.1) ¥ 10-3 (7.3 0.2) ¥ 10-3 (8 1) ¥ 10-3 (8.1 0.5) ¥ 10-3 (2.9 0.3) ¥ 10-3 (1.3 0.4) ¥ 10-3 (1.8 0.4) ¥ 10-3 (11 1) ¥ 10-3
NDN
QNLN
NDLN
NAa
NA
0.04 0.02
0.15 0.01
0.12 0.01
0.15 0.03
0.19 0.03
0.07 0.01
0.16 0.02
0.12 0.02
0.16 0.02
0.14 0.02
0.04 0.01
NA
0.07 0.01
NA
0.1 0.01
0.04 0.02
0.09 0.02
NA
0.02 0.01
0.23 0.03
NA
0.08 0.01
a ‘NA’ indicates no appreciable binding; FA:Y96F/L244A; FV:Y96F/T101V mutants of P450cam.
Table 4 NADH consumption rates (nmol nmol-1 P450/min) of P450cam
to the iron center compared to indoline. This was reflected by the
fact the Soret peak in the absorption spectra as well as the peak
of the difference spectra of the enzyme was slightly more red-
shifted in the presence of indoline compared to that in the presence
of quinoline. On the other hand the addition of indene (NDN)
and coumarin (CMRN) showed a blue shift in the Soret peak
(Fig. 11A) and a Type-I difference spectra (Fig. 11B) indicating
that these compounds do not form any direct coordination to
the iron center. The spectroscopic binding constants (Table 3) of
the P450cam variants were determined for these substrates. The
results showed that almost all the P450cam mutants, especially,
Y96F, Y96V, T101V, Y96F/L244A and Y96F/T101V, had higher
affinities for the aromatic hydrocarbons compared to the wild
type enzyme. Quinoline was found to have a lower binding affinity
compared to indoline in most of the cases. On the other hand
indene binding constants were considerably higher than those for
coumarin in most of the mutants. This may be attributed to the fact
that indene is a more hydrophobic moleculecompared to coumarin
and thereby prefers to reside in the hydrophobic environment of
the active site of the P450cam mutants more than coumarin.
variants in the presence of aromatic hydrocarbons (0.5 mM)
Substrate WT
Y96F FAa
30 35
FVa
Y96V Y96W F87W T101V
CMRN
NDN
QNLN
NDLN
3
1
3
3
3
1
4
1
2
1
1
1
3
1
2
1
1
12 1 194 5 188 7 29
2
1
46 4 7
11 1 14
1
1
1
2
1
5
33
1
NAa
23
3
1
2
3
1 3
14 2 40
2
3
3
26 2 18 2 22 2 24
a ‘NA’indicates no appreciable reaction; FA:Y96F/L244A; FV:Y96F/
T101V mutants of P450cam.
may find potential industrial application. The hydroxylation of this
compound would increase the molecular weight by 16 mass units
(Fig. 12). Thus the hydroxylated product is expected to show a
peak at m/z = 161 (162-H+) in the negative ion mode of the
ESI-MS spectrum. The ESI-MS spectra of the CHCl3 extract
(Fig. 12) from the reaction mixture containing the WT enzyme did
not show any significant peak intensity at m/z = 161 compared
to the control. This indicated that the wild type enzyme does
not have any measurable activity for hydroxylation of coumarin.
However, most of the mutant enzymes catalyzed the hydroxylation
of coumarin and distinct ESI-MS peak at m/z = 161 for hydroxyl
coumarin was observed (Fig. 12C). Interestingly, though the
NADH consumption rates (Table 4) for the Y96W, Y96V and
Y96F/T101V mutants in the presence of coumarin were very
small, these mutants, unlike the WT enzyme showed coumarin
hydroxylation activity (Fig. 13).
Fig. 11 The representative (A) absolute and (B) difference spectra
of coumarin (dashed line) and indene (dotted line) binding to the
substrate-free Y96F/L244A mutant (solid line) of P450cam. The final
enzyme, indene and coumarin concentrations were 4.5 mM, 105 mM and
1 mM, respectively.
Table 4 gives the rate of NADH consumption for different
P450cam mutants in the presence of various aromatic compounds.
The results in Table 4 show that the NADH consumption rates in
the presence of coumarin were significantly enhanced in the case
of Y96F and Y96F/L244A (FA) mutants.
Coumarin was selected as a representative hydrocarbon sub-
strate to evaluate the ability of the P450cam variants to oxygenate
the aromatic hydrocarbons. The selection of coumarin was based
on the facts that the hydroxycoumarin derivatives are used as
drugs, dyes/sensors, sunscreen agents and also for the assay of
microsomal P450s.41-43 Thus selective hydroxylation of coumarin
Fig. 12 The ESI-MS spectrum (negative ion mode) of the CHCl3 extract
of the catalytic mixture (A) in the absence of P450cam, (B) in the presence
of the wild type enzyme and (C) in the presence of the Y96F/L244A
mutant. The schematic representation of coumarin hydroxylation by
P450cam variants is shown at the top panel.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3115–3123 | 3121
©

Fig. 13 Relative coumarin hydroxylation activity of P450cam variants in
the reductive cycle with respect to the intensity of the ESI-MS peak at
m/z = 161.
The results (Fig. 13) showed that the coumarin hydroxylation
activity was enhanced respectively by ~ 2, 16, 52, 52 and 110
fold in the case of the Y96W, Y96F/T101V, Y96V, Y96F and
Y96F/L244A mutant compared to the wild type enzyme. These
mutants either reduce the mobility of the substrate by reducing the
height of the substrate binding pocket (Y96W) or by increasing the
hydrophobicity (Y96F/T101V, Y96F, Y96V) and the space on top
of heme (Y96F/L244A) to help the planar substrate come closer to
the putative oxo-ferryl species and, thereby, increase the substrate
hydroxylation compared to the wild type enzyme. The present
ESIMS results however, could not unambiguously identify the
position of hydroxylation of coumarin in the enzymatic reaction
(see below).
Fig. 14 Colorimetric assay for aromatic hydroxylation using 4-AAP
coupling reaction. * represents the site of coupling reaction.
In order to determine the position of hydroxylation in coumarin,
we carried out the assay of the enzymatic activity of the P450cam
variants using 4-aminoantipyrene (4-AAP) for detection of hy-
droxycoumarins as described in the previous section. The reaction
of 4-aminoantipyrene (4-AAP) in the presence of ammonium
persulfate (APS) in mild alkaline solution with a phenolic
compound produces a quinoneimine dye that could be detected
spectroscopically. The 4-AAP assay of hydroxycoumarin, in the
present case would give positive results when the para-position
to the –OH group is vacant in the hydroxycoumarin to form the
quinoneimine dye, that shows an absorption maximum at 467 nm
(Fig. 14). The 4-AAP coupling reactions with coumarin in the
absence of P450cam showed that the APS itself does not cause
any significant aromatic hydroxylation under the conditions of the
assay (control experiment). The wild type enzyme too did not show
any appreciable hydroxylation of the aromatic ring of coumarin
that could be detected by the 4-AAP assay. The quinoneimine
dye, detected from the increase in absorbance at 467 nm (Fig. 14),
formed by coupling of the 4-AAP at the -para position of the
phenol was formed when mutants of P450cam were used for
enzymatic hydroxylation of coumarin.
The results of the assay showed (Fig. 15) that the aromatic
hydroxylation activity was enhanced by ~25, 100, 175 and
400 times in the presence of the Y96F/T101V, Y96F, Y96V and
Y96F/L244A mutant, respectively, compared to that of the wild
type enzyme. As the 4-AAP assay recognizes only those phenolic
compounds, which have an open -para position for coupling
(Fig. 14),33 the present assay would measure the concentration
of only 5- and 8-hydroxy coumarin products only. So the results
suggest that the P450cam variants, especially, Y96F/L244A not
Fig. 15 Relative coumarin hydroxylation activity (at C5 and/or C8)
of P450cam variants in the reductive cycle with respect to the wild
type enzyme. (The inset shows the structure of coumarin with IUPAC
numbering. The numbers followed by * show the hydroxylation sites
sensitive to detection by 4-AAP assay and the numbers in italics represent
other possible hydroxylation sites.)
only have a higher activity but also have a higher regio-selectivity
for coumarin hydroxylation at C5 and C8 positions.
The results shown in Fig. 13 and Fig. 15 suggest that the relative
yield of the coumarin hydroxylation product detected from the
mass spectra were different from those by a 4-AAP assay of
5/8-hydroxycoumarin in the catalytic reactions of coumarin with
different P450cam mutants. The ESI-MS detects a total yield of
hydroxycoumarin while the 4-AAP assay would give a measure
of only 5/8-hydroxycoumarin. The ratios of hydroxycoumarin
produced by catalytic reaction of coumarin with Y96F/T101V,
Y96V, Y96F and Y96F/L244A mutants of P450cam were
16 : 52 : 52 : 110 (~0.15 : 0.5 : 0.5 : 1). On the other hand the ratios
of 5/8-hydroxycoumarin for the same reaction, under identical
conditions, were 25 : 100 : 175 : 400 (~0.06 : 0.25 : 0.4 : 1) for the
catalytic reactions in the presence of Y96F/T101V, Y96V, Y96F
and Y96F/L244A mutants of P450cam.
Assuming that the Y96F/L244A mutant catalyzes regio-
selective hydroxylation of coumarin to give solely 5- and/or
8-hydroxycoumarin the above results suggest that the ratios of
total hydroxycoumarin to 5/8-hydroxycoumarin are 0.15 : 0.06,
0.5 : 0.25, 0.5 : 0.4 and 1 : 1, respectively for Y96F/T101V, Y96V,
Y96F and Y96F/L244A mutants as catalysts.
3122 | Dalton Trans., 2010, 39, 3115–3123
This journal is
The Royal Society of Chemistry 2010
©

Coumarin can undergo hydroxylation at C3, C4, C6 and C7 apart
from at C5 and C8 positions (see inset of Fig. 15). The products
of C3, C4, C6 and C7 hydroxylation would not give a positive 4-
AAP test. The results indicate that the Y96F/T101V mutant of the
enzyme possibly produces 1.5 times more hydroxylation at C3, C4,
C6 and C7 positions compared to that at C5 and C8 of coumarin.
Thus the above results show that the engineering of the active site
of P450cam can be successfully exploited for the enhancement of
activity as well as regio-selectivity of hydroxylation of aromatic
compounds. However, it must be mentioned that the detailed and
quantitative characterization of the enzymatic properties of the
P450cam variants and the product distributions call for further
studies.
11 V. B. Urlacher, S. Lutz-Wahl and R. D. Schmid, Appl. Microbiol.
Biotechnol., 2004, 64, 317–325.
12 E. J. Mueller, P. J. Loida and S. G. Sligar, in Cytochrome P450: Structure,
Mechanism, and Biochemistry, ed. P. R. Oritz de Montellano, Plenum
Press, New York, 2nd edn, 1995, pp. 83-124.
13 M. R. Lefever and L. P. Wackett, Biochem. Biophys. Res. Commun.,
1994, 201, 373–378.
14 F. Xu, S. G. Bell, Z. Rao and L.-L. Wong, Protein Eng., Des. Sel., 2007,
20, 473–480.
15 C. F. Harford-Cross, A. B. Carmichael, F. K. Allan, P. A. England,
D. A. Rouch and L. L. Wong, Protein Eng., Des. Sel., 2000, 13, 121–
128.
16 R. J. Sowden, S. Yasmin, N. H. Rees, S. G. Bell and L. L. Wong, Org.
Biomol. Chem., 2005, 3, 57–64.
17 J. A. Stevenson, A. C. Westlake, C. Whittock and L.-L. Wong, J. Am.
Chem. Soc., 1996, 118, 12846–12847.
18 P. J. Jonathan, J. O. H. Ellen and W. Luet-Lok, Eur. J. Biochem., 2001,
268, 1460–1467.
19 S. G. Bell, X. Chen, F. Xu, Z. Rao and L. L. Wong, Biochem. Soc.
Trans., 2003, 31, 558–562.
Conclusions
20 S. G. Bell, X. Chen, R. J. Sowden, F. Xu, J. N. Williams, L.-L. Wong
and Z. Rao, J. Am. Chem. Soc., 2003, 125, 705–714.
21 D. P. Nickerson, C. F. Harford-Cross, S. R. Fulcher and L.-L. Wong,
FEBS Lett., 1997, 405, 153–156.
22 K. J. French, D. A. Rock, J. I. Manchester, B. M. Goldstein and J. P.
Jones, Arch. Biochem. Biophys., 2002, 398, 188–197.
23 V. Reipa, M. P. Mayhew, M. J. Holden and V. L. Vilker, Chem.
Commun., 2002, 318–319.
24 E. S. Ferreira, A. N. Hulme, H. McNab and A. Quye, Chem. Soc. Rev.,
2004, 33, 329–336.
25 H. Zollinger, in Color Chemistry: Syntheses, Propertiesand Applications
of Organic Dyes and Pigments, Wiley, 3rd edn, 2003, pp. 255-354.
26 R. Balamurugan, F. J. Dekker and H. Waldmann, Mol. BioSyst., 2005,
1, 36–45.
Our study showed that the protein engineering of cytochrome
P450cam can give a possible alternative route for the synthesis
of the industrially important blue dye indigo. It also showed that
the in vitro oxidative cycle is a cleaner method compared to the
reductive cycle as well as in vivo synthesis of the dye from indole.
The rational redesigning of the P450cam active site was found
to enhance the affinity of the enzyme for unnatural substrates as
well as the NADH oxidation rate in the presence of aromatic
compounds. The engineering of the P450cam active site was
found to enhance the activity of the enzyme for regio-selective
hydroxylation of coumarin. The present studies thus demonstrate
that the cytochrome P450cam can be successfully engineered for
synthetic as well as environmental applications.
27 N. Au and A. E. Rettie, Drug Metab. Rev., 2008, 40, 355–375.
28 P. H. Lee, S. K. Ong, J. Golchin and G. L. Nelson, Water Res., 2001,
35, 3941–3949.
29 K. M. Weir, T. D. Sutherland, I. Horne, R. J. Russell and J. G.
Oakeshott, Appl. Environ. Microbiol., 2006, 72, 3524–3530.
30 V. Y. Kuznetsov, E. Blair, P. J. Farmer, T. L. Poulos, A. Pifferitti and
I. F. Sevrioukova, J. Biol. Chem., 2005, 280, 16135–16142.
31 I. F. Sevrioukova and T. L. Poulos, J. Biol. Chem., 2002, 277, 25831–
25839.
32 R. A. Crowell, R. Lian, J. Sauer, M. C., D. A. Oulianov and I. A.
Shkrob, Chem. Phys. Lett., 2004, 383, 481–485.
33 E. Emerson, J. Org. Chem., 1943, 8, 417–428.
34 S. K. Manna and S. Mazumdar, Biochemistry, 2006, 45, 12715–12722.
35 B. Munge, C. Estavillo, J. B. Schenkman and J. F. Rusling, Chem-
BioChem, 2003, 4, 82–89.
Acknowledgements
The Authors wish to thank Mr. Bharat T. Kansara for help. This
work was supported by the Indo French Centre for the Promotion
of Advanced Research (IFCPAR/CEFIPRA), New Delhi. The
work was carried out at the Tata Institute of Fundamental
Research, Mumbai.
36 S. Jin, T. M. Makris, T. A. Bryson, S. G. Sligar and J. H. Dawson, J. Am.
Chem. Soc., 2003, 125, 3406–3407.
References
1 E. J. Corey and X.-M. Cheng, The Logic of Chemical Synthesis, Wiley,
New York, 1995.
37 H. Joo, A. Arisawa, Z. Lin and F. H. Arnold, Chem. Biol., 1999, 6,
699–706.
2 K. Godula and D. Sames, Science, 2006, 312, 67–72.
3 W.-M. Cheung, H.-Y. Ng, I. D. Williams and W.-H. Leung, Inorg.
Chem., 2008, 47, 4383–4391.
4 J. M. Keith, R. J. Nielsen, J. Oxgaard and W. A. Goddard, 3rd, J. Am.
Chem. Soc., 2005, 127, 13172–13179.
5 N. S. Loumbourdis, Bull. Environ. Contam. Toxicol., 1997, 58, 945–952.
6 J. O. Duruibe, M. O. C. Ogwuegbu and J. N. Egwurugwu, Int. J. Phys.
Sci., 2007, 2, 112–118.
38 J. L. Royo, E. Moreno-Ruiz, A. Cebolla and E. Santero, J. Biotechnol.,
2005, 116, 113–124.
39 A. Celik, R. E. Speight and N. J. Turner, Chem. Commun., 2005, 3652–
3654.
40 E. M. Gillam, A. M. Aguinaldo, L. M. Notley, D. Kim, R. G.
Mundkowski, A. A. Volkov, F. H. Arnold, P. Soucek, J. J. DeVoss
and F. P. Guengerich, Biochem. Biophys. Res. Commun., 1999, 265,
469–472.
7 T. Norseth, Environ. Health Perspect., 1981, 40, 121–130.
8 F. A. Carey and R. J. Sundberg, Part B: Reaction and Synthesis, 4th
edn, Kluwer Academic/Plenum Publishers, New York, 2001.
9 F. P. Guengerich, Chem. Res. Toxicol., 2001, 14, 611–650.
10 P. R. Oritz de Montellano and J. J. De Voss, in Cytochrome P450:
Structure, Mechanism, and Biochemistry, ed. P. R. Oritz de Montellano,
Plenum Press, New York, 3rd edn, 2005, pp. 183-246.
41 A. Singh, K. Chen, S. J. Adelstein and A. I. Kassis, Radiat. Res., 2007,
168, 233–242.
42 K. Nakamura, I. H. Hanna, H. Cai, Y. Nishimura, K. M. Williams and
F. P. Guengerich, Anal. Biochem., 2001, 292, 280–286.
43 B. Thati, A. Noble, B. S. Creaven, M. Walsh, M. McCann, K.
Kavanagh, M. Devereux and D. A. Egan, Cancer Lett., 2007, 248,
321–331.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 3115–3123 | 3123
©