Structural insights into the role of the acid-alcohol pair of residues required for dioxygen activation in cytochrome P450 enzymes

Article abstract of DOI:10.1007/s00775-020-01781-4

The cytochrome P450 heme monooxygenases commonly use an acid-alcohol pair of residues, within the I-helix, to activate iron-bound dioxygen. This work aims to clarify conflicting reports on the importance of the alcohol functionality in this process. Mutants of the P450, CYP199A4 (CYP199A4_D251N and CYP199A4_T252A), were prepared, characterised and their crystal structures were solved. The acid residue of CYP199A4 is not part of a salt bridge network, a key feature of paradigmatic model system P450cam. Instead, there is a direct proton delivery network, via a chain of water molecules, extending to the surface. Nevertheless, CYP199A4_D251N dramatically reduced the activity of the enzyme consistent with a role in proton delivery. CYP199A4_T252A decreased the coupling efficiency of the enzyme with a concomitant increase in the hydrogen peroxide uncoupling pathway. However, the effect of this mutation was much less pronounced than reported with P450cam. Its crystal structures revealed fewer changes at the I-helix, compared to the P450cam system. The structural changes observed within the I-helix of P450cam during oxygen activation do not seem to be required in this P450. These differences are due to the presence of a second threonine residue at position 253, which is absent in P450cam. This threonine forms part of the hydrogen bonding network, resulting in subtle structural changes and is also present across the majority of the P450 superfamily. Overall, the results suggest that while the acid-alcohol pair is important for dioxygen activation this process and the method of proton delivery can differ across P450s. Graphic abstract[Figure not available: see fulltext.].

Full text of DOI:10.1007/s00775-020-01781-4

JBIC Journal of Biological Inorganic Chemistry
https://doi.org/10.1007/s00775-020-01781-4
ORIGINAL PAPER
Structural insights into the role of the acid‑alcohol pair of residues
required for dioxygen activation in cytochrome P450 enzymes
Tom Coleman1 · Jeanette E. Stok2 · Matthew N. Podgorski1 · John B. Bruning3 · James J. De Voss2
Stephen G. Bell1
·
Received: 10 January 2020 / Accepted: 22 March 2020
© Society for Biological Inorganic Chemistry (SBIC) 2020
Abstract
The cytochrome P450 heme monooxygenases commonly use an acid-alcohol pair of residues, within the I-helix, to acti-
vate iron-bound dioxygen. This work aims to clarify conficting reports on the importance of the alcohol functionality in
this process. Mutants of the P450, CYP199A4 (CYP199A4D251N and CYP199A4T252A), were prepared, characterised and
their crystal structures were solved. The acid residue of CYP199A4 is not part of a salt bridge network, a key feature of
paradigmatic model system P450cam. Instead, there is a direct proton delivery network, via a chain of water molecules,
extending to the surface. Nevertheless, CYP199A4D251N dramatically reduced the activity of the enzyme consistent with
a role in proton delivery. CYP199A4T252A decreased the coupling efciency of the enzyme with a concomitant increase in
the hydrogen peroxide uncoupling pathway. However, the efect of this mutation was much less pronounced than reported
with P450cam. Its crystal structures revealed fewer changes at the I-helix, compared to the P450cam system. The structural
changes observed within the I-helix of P450cam during oxygen activation do not seem to be required in this P450. These
diferences are due to the presence of a second threonine residue at position 253, which is absent in P450cam. This threonine
forms part of the hydrogen bonding network, resulting in subtle structural changes and is also present across the majority of
the P450 superfamily. Overall, the results suggest that while the acid-alcohol pair is important for dioxygen activation this
process and the method of proton delivery can difer across P450s.
Graphic abstract
Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s00775-020-01781-4) contains
supplementary material, which is available to authorized users.
Extended author information available on the last page of the article
Vol.:(0123456789)
1 3

JBIC Journal of Biological Inorganic Chemistry
Keywords Metalloenzymes · Heme · Cytochrome P450 ·
entry of two new water molecules into the active site. These
"catalytic" water molecules contribute to a hydrogen bond net-
work connecting the threonine, oxygen, and water molecules,
and have been proposed to enable proton delivery from the
proximal side of the heme. Mutations at the acid-alcohol pair
have played a key role in the development of the mechanistic
proposals of oxygen activation in P450 activity [14, 15, 17–20].
Mutation of the acidic residue of the acid-alcohol pair,
Mechanism · X-ray crystallography · Dioxygen activation
Introduction
The insertion of an oxygen atom into an unactivated C–H bond
is an extremely difcult reaction to achieve using standard
chemical methodologies. However, it is an essential biological
function and is the most common reaction catalysed by heme-
dependent cytochrome P450 (CYP) monooxygenase enzymes
[1, 2]. Moreover, in these biological systems it occurs at ambi-
ent conditions and with high selectivity. The ferryl Compound
I (Cpd I) intermediate is believed to be the oxygen-activating
species in the majority of P450 catalysed reactions [1, 3–7].
The hydroxylation reaction was proposed to occur via a radi-
cal rebound mechanism of the oxygen atom by Groves and
McClusky [8]. In the frst step of this mechanism, the highly
reactive Cpd I intermediate abstracts a hydrogen atom from the
alkyl substrate, resulting in an Fe(IV)-OH species (Compound
II, Cpd II) and a substrate radical (Scheme 1a) [4, 9]. The latter
remains bound within the active site of the enzyme where it can
rapidly recombine with a hydroxyl radical derived from Cpd II
to give the alcohol product [9, 10].
Asp251 in P450cam, to a neutral asparagine (P450camD251N
)
resulted in drastically reduced NADH oxidation activity
and camphor hydroxylation at 1% of the rate of the WT
enzyme [15]. Kinetic isotope and proton inventory experi-
ments have demonstrated that this mutation slows delivery
of the frst proton to the oxygen. Other studies, with P450cin,
P450Bm3, CYP17A1 and CYP19A1, have found analogous
reductions in the rate of NAD(P)H oxidation activity when
the conserved acidic residue is mutated, inferring a common
role for this residue [11, 21–24].
When the threonine residue is mutated (P450camT252A),
there is little change in the rate of NADH consumption. How-
ever, the enzyme becomes almost completely uncoupled with
95% of the reducing equivalents channelled into hydrogen
peroxide formation [17, 19]. It was originally suggested that
Thr252 served as a direct proton source for the hydroperoxy
intermediate (Cpd 0), but this was challenged when a Thr252-
OMe mutant, where the side chain OH was replaced with a
methoxy group, demonstrated considerable activity [25]. The
role of the conserved threonine was then proposed to act as
a hydrogen bond acceptor for Cpd 0, helping to stabilise this
intermediate and facilitate the second proton transfer step [13].
In P450camT252A, Ala252 can no longer accept a hydrogen
bond from Cpd 0 to promote O–O scission, and it is surmised
that Cpd 0 thus accumulates. Subsequent protonation on the
proximal rather than distal oxygen of Cpd 0 leads to hydro-
gen peroxide formation in preference to camphor hydroxyla-
tion (Scheme 1b). The crystal structures of T252Acam showed
that removal of the hydroxyl group induced similar structural
changes to those observed in the oxy-bound complex described
above [2, 13, 14].
The mechanism of oxygen activation, which generates the
reactive intermediates in P450s, implicates a controlled pro-
ton relay or shuttle system. The peroxo-anion, formed after
reduction of the ferrous-dioxy complex, accepts a proton to
give the hydroperoxy intermediate (Cpd 0, Scheme 1b) [2].
Protonation of this species at the distal oxygen triggers het-
erolytic O–O bond scission to give Cpd I (Scheme 1b). In the
majority of cases these proton transfer events involve a highly
conserved "acid-alcohol" amino acid pair, typically aspartate
(or glutamate) and threonine (or serine) which are found in the
oxygen-binding groove of the I-helix [2]. Exceptions include
P450 enzymes which have evolved to use diferent strategies,
for example, P450cin and P450eryF replace the alcohol group
with an asparagine [11] and alanine [12] residue, respectively.
In the former this asparagine is believed to hold a water mol-
ecule close by while in the latter a substrate hydroxyl group
is co-opted to act as the alcohol functionality.
CYP199A4, a cytochrome P450 enzyme from Rhodopseu-
domonas palustris strain HaA2, is highly substrate specifc and
has high afnity for para-substituted benzoic acids. The oxida-
tion activity of the enzyme is almost always restricted to the
substituent located para to the benzoic acid moiety [26–31]. It
is able to selectively demethylate both 4-methoxy- and 4-meth-
ylamino-benzoic acids [27, 29, 30, 32]. It can also hydroxylate
and desaturate para-substituted alkyl benzoic acids, such as
4-ethyl- and 4-isopropyl-benzoic acids, to form both alkene
and alcohol products [29]. The activities of oxidative O-dem-
ethylation, hydroxylation and desaturation of para-substituted
benzoic acids are high, and the loss of reducing equivalents
due to uncoupling side reactions is low. Crystal structures of
substrate-bound forms of the enzyme are also available [28,
Much of what we understand about the P450 catalytic
cycle and the mechanism of oxygen activation derives from
the bacterial enzyme P450cam from a Pseudomonas strain
[13–15]. Wild-type (WT) P450cam hydroxylates camphor
at>1000 min−1 with almost 100% coupling efficiency of
the NADH reducing equivalents into product formation. In
P450cam a signifcant conformational change is induced in the
region of the I-helix when oxygen binds to the ferrous iron. The
peptide carbonyl of Asp251 is fipped 90° relative to its previ-
ous orientation and this is accompanied by the repositioning of
Thr252 to hydrogen bond to the dioxygen molecule [13, 16].
These structural rearrangements open up the I-helix, allowing
1 3

JBIC Journal of Biological Inorganic Chemistry
Scheme 1 The radical rebound
and dynamically coupled ultra-
fast rebound mechanisms of
cytochrome P450 along with the
product formation and uncou-
pling pathways. a The hydrogen
abstraction and radical rebound
steps (or dynamically coupled
ultrafast rebound) in the oxida-
tive demethylation of substrate
(R-OCH3) by compound I (Cpd
I). b The oxygen activation
process which leads to the for-
mation of Cpd I including the
successive controlled protona-
tion steps of the peroxo-anion
and Cpd 0 (H+ in black). Also
shown in red is the potential
alternate protonation of Cpd 0,
leading to hydrogen peroxide
uncoupling
a
b
29, 33]. The high activity and selectivity of oxidation of the
CYP199A4 enzyme for multiple P450 catalysed reactions, in
combination with obtainable structural information, make it
an ideal candidate for mechanistic studies of P450 enzymes.
In CYP199A4, the residues of the acid-alcohol pair
are Asp251 and Thr252 [28], which are coincidentally at
the same position in the sequence as those in P450cam
(P450camD251 and P450camT252). Here we report the
characterisation of the equivalent threonine and aspartate
mutants of CYP199A4 to probe their role in oxygen activa-
tion in this enzyme. We investigate the activity of 4-meth-
oxy- and 4-ethylbenzoic acid with CYP199A4T252A and
CYP199A4D251N. The determination of the substrate-bound
crystal structures of CYP199A4D251N and CYP199A4T252A
with the WT enzyme provide new insight into roles of the
highly conserved I-helix acid-alcohol pair of residues in the
P450 catalytic cycle.
UV/Vis spectra and spectroscopic activity assays were
recorded at 30 0.5 °C on an Agilent CARY-60 spectro-
photometer. Analytical high performance liquid chro-
matography (HPLC) was performed on an Agilent 1260
Infinity Pump equipped with an autoinjector connected
using a Phenomenex Kinetex 5 µm, XB-C18, 100 Å col-
umn (250 mm × 4.6 mm, 5 μm). The products were sepa-
rated using a gradient between 20 and 95% acetonitrile in
water (0.1% trifuoroacetic acid at a fow rate of 1 mL min−1
over 30 min. Gas Chromatography-Mass Spectrometry
(GC–MS) data were collected on a Shimadzu GC-17A using
a QP5050A GC–MS detector and a DB-5 MS fused silica
column (30 m×0.25 mm, 0.25 μm). The injector and inter-
face were maintained at a constant temperature of 250 °C
and 280 °C. The oven temperature was held at 100 ºC for
1 min and then increased at 15 ºC min−1 up to 220 ºC. Chiral
chromatography was performed on a Shimadzu Prominence
HPLC system [32].
The expression, purification and characterisation of
WT CYP199A4, HaPux and HaPuR (HaPux and HaPuR
are the [2Fe-2S] ferredoxin and FAD dependent ferredoxin
reductase electron transfer partner of the CYP199A4 sys-
tem) were performed as described elsewhere [27–29]. The
CYP199A4 protein concentration was calculated using
ε419 = 119 mM−1 cm−1 as described previously [27, 28].
Both the CYP199A4 mutants (D251N and T252A) were
constructed by employing two separate PCRs that introduced
Experimental
General
General reagents and organic substrates were from Sigma-
Aldrich, TCI, Fluorochem or VWR. Bufer components,
NADH and isopropyl-β-d-thiogalactopyranoside (IPTG)
were from Astral Scientifc, Australia.
1 3

JBIC Journal of Biological Inorganic Chemistry
the desired mutation via a mutagenic primers using pET28.
CYP199A4 as the template (see Supplementary material
for primers and the experimental details). The purifcation
of CYP199A4T252A and CYP199A4D251N were performed
using the same protocol as the WT enzyme. Prior to use,
glycerol was removed from CYP199A4 using a 5 mL gel
fltration column (PD-10, GE Healthcare). To assess the
viability of the CYP199A4 mutants they were reduced with
sodium dithionite followed by gently bubbling CO through
the sample before recording the spectrum [34].
was calculated by monitoring the absorbance at 340 nm and
using ε340 = 6.22 mM–1 cm–1. The product formation rate
and coupling efciency were calculated by quantitating the
amount of product in the turnover (see below).
The concentration of hydrogen peroxide generated during
the NADH turnovers was determined using a horseradish
peroxidase (HRP)/phenol/4-aminoantipyrine (4-AP) assay
[38]. The turnover incubation (400 µL) was mixed with 200
µL of 50 mM phenol (pH 7.4, Tris bufer) and 200 µL of
5 mM 4-AP (pH 7.4, Tris bufer). The absorbance at 510 nm
was set to zero, and 1 µL of a 20 mg mL−1 solution of HRP
was added. The absorbance at 510 nm was then recorded
and used to calculate the concentration of hydrogen peroxide
(ε510 =6580 M−1 cm−1).
Substrate binding: spin state determination
and binding titrations
The high-spin heme content was determined using enzyme
in the concentration range 1.2–3.8 µM in 50 mM Tris, pH
7.4, with the addition of 0.5–1 µL aliquots of substrate (from
a 100 mM stock) until the spectrum did not change. The per-
centage shift was estimated (to approximately 5%) by com-
parison with a set of spectra generated from the sum of the
appropriate percentages of the spectra of the substrate-free
(>95% low-spin, Soret maximum at 418 nm) and camphor-
bound (>95% high-spin, Soret maximum at 392 nm) forms
of WT CYP101A1 as described previously [35, 36].
Analysis of metabolites
The majority of the products arising from the enzyme turn-
over were identifed via co-elution of authentic standards
using HPLC analysis. After the NADH had been consumed
in substrate oxidation incubations, 150 μL of the reaction
mixture was mixed with 2 μL of an internal standard solu-
tion (20 mM 9-hydroxyfourene in ethanol). This was mixed
with 50 μL of acetonitrile before analysis by HPLC. The
HPLC retention times for the substrates and products are
given in the supporting information.
For dissociation constant determination CYP199A4 was
diluted to 1.3–1.7 µM in 50 mM Tris, pH 7.4, in a total vol-
ume of 2.5 mL. Aliquots of the substrate (0.5–2 µL) were
added using a Hamilton syringe from a 1, 10 or 100 mM
stock solution in ethanol. The maximum Soret band peak-to-
trough diference (ΔA) in absorbance was recorded between
700 and 250 nm. Further aliquots of substrate were added
until the peak-to-trough diference of the Soret band did not
change. The dissociation constants, Kd, were obtained by ft-
ting ΔA against total substrate concentration [S] to the tight
binding quadratic equation [37]:
Protein crystallisation and X‑ray crystallography
For crystallisation, the CYP199A4 mutants isolated as
described above were further purifed by size exclusion
chromatography (Enrich SEC Column, 650 10×300 mm)
and were concentrated to approx. 30–35 mg mL−1 in 50 mM
Tris, pH 7.4. Substrate was added to the protein from a
100 mM stock solution in EtOH to a fnal concentration
of 1 mM. Crystals were obtained using the hanging-drop
vapour difusion method at 16 °C using 1 µL of protein with
1 µL of reservoir solution and equilibrated with 500 µL of
the same reservoir solution. Rectangular plate-shaped crys-
tals of approx. 300 μm by 150 μm by 20 μm were obtained
in 1 week from 30 to 35 mg mL−1 CYP199A4 (with 1 mM
substrate) with a reservoir solution containing 0.2 M mag-
nesium acetate tetrahydrate, and 20–23% w/v PEG-3,350
and 0.1 M Bis–Tris pH 5.25–5.5. Crystals were harvested
using a Microloop or Micromount (MiTeGen), then cryopro-
tected by immersion in Parabar 10312 (Paratone-N, Hamp-
ton Research) and fash cooled in liq. N2. X-ray difraction
data were collected at 100 K on the MX1 beamline at the
Australian Synchrotron [39, 40].
ꢀ
ꢁ
}
(
)
(
)
[E] + [S] + Kd
−
[E] + [S] + Kd 2 − 4[E][S]
ΔA
=
,
ΔAmax
2[E]
ΔAmax is the maximum absorbance diference and [E] is the
enzyme concentration.
Activity assays
In vitro NADH turnover rate assays were performed with
mixtures (1.2 mL) containing 50 mM Tris, pH 7.4, 0.5 μM
CYP199A4, 5 μM HaPux and 0.5 μM HaPuR. The bufer
solution was oxygenated before addition of the enzymes
and the mixtures equilibrated at 30 °C for 2 min. Sub-
strates were added as a 100 mM stock solution in ethanol
to a fnal concentration of 1 mM. NADH was added to ca.
320 μM (fnal A340 = 2.00). The rate of NADH oxidation
All difraction data were indexed and integrated using
iMosfm [41]; scaled, merged and R-free fags were added
using Aimless [42], both available in the CCP4 suite of
programmes [43]. The phase problem was solved using the
1 3

JBIC Journal of Biological Inorganic Chemistry
molecular replacement method with Phaser in CCP4 [44],
using 4-methoxybenzoic acid-bound CYP199A4 (PDB:
4DO1, with the 4-methoxybenzoic acid and heme ligands
removed) as the initial search model. Electron density maps
were obtained after initial model building and the model was
rebuilt using Coot [45]. Structural refnements were per-
formed over multiple iterations using Phenix Refne avail-
able in the Phenix suite of programmes [46]. Composite
omit maps were generated using the composite omit maps
programme in Phenix. Detailed data collection and struc-
tural refnement statistics for the data sets are summarised
in Tables S1.
frst electron and commencement of the catalytic cycle [29,
31, 35]. Substrate binding can be conveniently measured
by monitoring the spin state of the enzyme upon addition
of the compound of interest [47–49]. The spin-state shifts
of CYP199A4T252A and CYP199A4D251N in the presence of
4-methoxy- and 4-ethyl-benzoic acid were similar to those
observed previously for CYP199A4WT (≥95%, Table 1, Fig.
S2) [29]. The dissociation constants of these substrates for
CYP199A4T252A and CYP199A4D251N were also similar to
CYP199A4WT with both binding with slightly higher afnity
to the variants (Table 1, Fig. S3).
The coordinates for the crystal structures of the substrate-
bound CYP199A4 mutant combinations studied have been
deposited in the Protein Data Bank (accession codes pre-
sented in Table S1).
Substrate turnover with the T252A and D251N
mutants of CYP199A4
The T252ACYP199A4 variant displayed a reduced rate
of NADH oxidation with 4-methoxybenzoic acid;
833 nmol nmol-CYP–1 min–1 (henceforth abbreviated
to min−1) compared to 1340 min−1 for the WT enzyme
(Table 1, Fig. S4). A single product, 4-hydroxybenzoic acid,
was observed in the turnovers of both the mutant and the
WT enzyme. There was a signifcant, but reduced, level of
product formation in the mutant enzyme (Fig. S5). The cou-
pling efciency of the mutant turnover was 55% compared
to 91% in the WT enzyme. The rate of productive oxida-
tive demethylation was therefore reduced to less than half of
that of CYP199A4WT at 472 min−1 (Table 1). Almost all the
unproductive reducing equivalents (~42%) were channelled
into the production of hydrogen peroxide.
Results
Substrate binding
The CYP199A4T252A and CYP199A4D251N enzymes were
generated using standard methods [27, 28]. Their viability
as functional P450 monooxygenase enzymes was initially
evaluated by measuring the UV/Vis spectra of the ferric,
ferrous and ferrous-CO forms (Fig. S1). To assess the efect
of the CYP199A4T252A and CYP199A4D251N mutations on
this enzyme we next evaluated the binding and oxidation of
4-methoxybenzoic acid. The frst step of the P450 catalytic
cycle is generally substrate binding. As with P450cam the
presence of a substrate in the active site of CYP199A4 has
been shown to be a pre-requisite for efcient transfer of the
The most notable change in the turnover assays of
CYP199A4D251N was the considerably lower (near leak
rate) activity of NADH oxidation compared with the WT
enzyme (Table 1). The NADH oxidation rate was 7.3 min−1
Table 1 Substrate binding
and in vitro turnover data for
the WT and mutant forms of
CYP199A4
Substrate
HS %
Kd (μM)
NADHb
PFR
Coupling (%)
H2O2 (%)
1.7 0.2
a
4-MethoxyBA WT
4-MethoxyBA T252A
4-MethoxyBA D251N
4-EthylBA WT
4-EthylBA T252A
4-EthylBA D251N
>95
>95
>95
>95
>95
>95
0.28 0.01
0.22 0.02
0.17 0.01
0.34 0.02
0.15 0.01
0.08 0.03
1340 28
1220 120
472 56
3.5 0.9
515 88
269 23
3.5 0.7
91
55
59
2
5
3
833
7.3 0.8
812
8
42
–
3
c
a
7
63 10
2.0 0.1
38 0.9
774 12
9.4 1.1
35
32
3
5
c
−
The NADH oxidation rate is the frequency of NADH oxidation, including product formation and
uncoupling reactions. PFR the product formation rate and the coupling efciency, is defned as the
percentage of NADH utilised for the formation of products. The turnovers were measured using a
HaPuR:HaPux:CYP199A4 concentration ratio of 1:10:1 (0.5 μM CYP enzyme, 50 mM Tris, pH 7.4).
Rates are reported as mean S.D. (n≥3) and given in nmol nmol-CYP–1 min–1
aThe WT data has been reported previously [28, 29]
bThe leak rates of the three enzymes in the absence of substrate were comparable (WT, 9.0;
CYP199A4T252A A, 8.7 and CYP199A4D251N, 7.5 nmol nmol-CYP–1 min–1). The data are reported as meas-
ured and has not been corrected for the leak rate
cThe slow rate of the CYP199A4D251N mutant means that the hydrogen peroxide concentration could not be
determined accurately [50]
1 3

JBIC Journal of Biological Inorganic Chemistry
by CYP199A4 is complicated by a competing desaturation
reaction to generate 4-vinylbenzoic acid. The turnovers of
4-ethylbenzoic acid with the mutant enzymes generated
the same two major products as the WT enzyme, which
arise from oxidation at the α-carbon of the ethyl group and
desaturation to 4-vinylbenzoic acid (Table 2, Scheme 2, Fig.
S5). Lower levels of hydroxylation at the β-carbon and fur-
ther oxidation metabolites of the two major products were
also observed (to the epoxide and 4-acetylbenzoic acid).
CYP199A4D251N produced the metabolites of 4-ethylben-
zoic acid in a similar product distribution to the WT enzyme
while the T252ACYP199A4 mutant generated a slightly lower
amount of the alkene with a compensating increase in the
amount of β-hydroxy alcohol (Table 2).
With 4-ethylbenzoic acid, the enantioselectivity of the
production of the 4-(1-hydroxyethyl)benzoic acid metabolite
also showed some variation between the CYP199A4 iso-
forms [32]. The (S)-enantiomer was formed in excess in all
cases, with CYP199A4D251N resulting in an increase in the
enantioselectivity compared to CYP199A4WT (88% versus
76% e.e.) and CYP199A4T252A showing a reduction (60%
e.e., Fig. S5c) [32].
Scheme 2 The products arising from the oxidation of 4-methoxy-
and 4-ethylbenzoic acid by CYP199A4 and its mutants investigated
in this work
Table 2 Product distribution for 4-ethylbenzoic acids with
Overall, it appears that mutation of the conserved threo-
nine has less of an efect on product formation activity in
CYP199A4 than in P450camT252A, where camphor hydroxy-
lation is efectively abolished. However, the elevated perox-
ide levels in the CYP199A4T252A turnovers suggest uncou-
pling via Cpd 0 is promoted over Cpd I formation.
CYP199A4WT, CYP199A4T252A and CYP199A4D251N
Substrate
α-OH
β-OH
Alkene
a
4-ethylBA WT
42
37 0.1
48 0.5
1
3
7
3
0.1
0.4
0.3
37 0.3
26 1.6
38 0.1
b
c
4-ethylBA T252A
4-ethylBA D251N
The amount of each product in the 4-ethylbenzoic acid turnovers were
calibrated against the three authentic product standards
Crystal structure of the 4‑methoxybenzoic
acid‑bound form of CYP199A4T252A
a1% ketone and 17% epoxide
b25% ketone and 6% epoxide
c5% ketone and 5% epoxide
To understand the effect of mutations at the conserved
acid-alcohol pair of residues of CYP199A4, crystals of
CYP199A4T252A were grown in the presence of 4-methoxy-
benzoic acid (1 mM). The crystal structures were acquired
to complement the previously solved WT enzyme struc-
ture (PDB: 4DO1) [28]. Crystals were obtained in 2 weeks
from a 500 µL reservoir solution containing 0.2 M mag-
nesium acetate tetrahydrate, 20–23% w/v PEG-3,350 and
0.1 M Bis–Tris pH 5.25. High resolution data (1.42 Å) were
obtained at 100 K on the MX1 beamline at the Australian
Synchrotron and deposited online in the wwPDB (Table S1,
Fig. S6) [39]. The structure of CYP199A4T252A contained
a single molecule in the asymmetric unit, and crystallised
with P21 cell symmetry. This was similar to other crystals
of substrate bound forms of WT CYP199A4 performed
under these conditions in our laboratory [32]. The overall
topology of the CYP199A4 mutant was identical to that of
the WT structure (PDB: 4DO1) indicating that the muta-
tions were not disruptive to the overall fold of the enzyme
(CYP199A4T252A: rmsd=0.527 Å aligned over 392 of 393
(Table 1, Fig. S4). The level of product formation was also
reduced. Due to the slow rate of turnover of the catalytic
cycle the lower coupling efciency may be due to competi-
tive air oxidation of the ferredoxin rather than a lowering of
the coupling efciency of the CYP199A4 enzyme itself. The
same single product was observed with CYPD251N as was
formed with CYP199A4WT and CYP199A4T252A (Scheme 2,
Fig. S5).
In case the easier abstraction of a methoxy C–H bond
resulted in an improved activity with the CYP199A4 T252A
variant we also assessed the activity of 4-ethylbenzoic acid.
There was a similar drop in the coupling efciency, NADH
oxidation and product formation rate of CYP199A4T252A
catalysed oxidation of 4-ethylbenzoic acid compared to
the WT enzyme (Table 1). Again there was a concomitant
increase in the amount of hydrogen peroxide formed in the
mutant turnovers. The hydroxylation of 4-ethylbenzoic acid
1 3

JBIC Journal of Biological Inorganic Chemistry
Fig. 1 a The active site environments of CYP199A4WT (PDB: 4DO1)
and b CYP199A4T252A (PDB: 5KDZ) highlighting the oxygen bind-
ing groove. 4-Methoxybenzoic acid is shown in green, the heme in
grey and water molecules are red spheres. c A comparative overlay
of the active site of the crystal structures of 4-methoxybenzoic acid-
bound forms of CYP199A4WT (cyan) and CYP199A4T252A (yellow).
d A comparative overlay of the active site region, including the oxy-
gen binding groove, of camphor-bound forms of WT P450cam (cyan,
PDB: 5CP4) and P450camT252A (yellow, PDB: 2CP4). In c and d the
water molecules are shown as spheres in the colour of their respec-
tive structures and the chloride ion (of CYP199A4) as green spheres,
with that of the WT structure labelled. Two water molecules in the
oxygen binding groove in d are numbered as W187 in their respective
structures
residues; Fig. S7). The positions of all active site residues
were functionally superimposable with the WT enzyme.
In the active site of CYP199A4T252A, electron density
corresponding to the 4-methoxybenzoic acid substrate and
changes for the mutated residue were found (Fig. S8). The
positions of the active site residues and substrate molecule
were comparable to that of CYP199A4WT with 4-methoxy-
benzoic acid bound (Fig. 1, Table S2). The chloride ligand
that caps the active site from solvent access was also found
in a nearly identical position, as was the water molecule
that is part of the substrate carboxylate hydrogen bonding
network (Fig. 1). In CYP199A4T252A the mutation breaks
the hydrogen bond observed in the WT enzyme between
the carbonyl group of A248 and the hydroxyl group of the
T252 (Fig. 1). There was an additional water molecule
bound in the loop region from alanine 248 to alanine 252 in
the CYP199A4T252A structure (Fig. 1) which is the oxygen
binding groove. The new water molecule in CYP199A4T252A
(W176) occupies the space vacated by the hydroxyl group of
the Thr252 residue. This water molecule forms a hydrogen
1 3

JBIC Journal of Biological Inorganic Chemistry
Table 3 Comparison of the changes in the oxygen binding groove in
This widening of the oxygen binding groove in P450cam
has been linked to other structural changes in the back-
bone of the I-helix upon oxygen binding or mutation
of the Thr252 residue (Figs. 2, S9). A comparison of
CYP199A4WT and CYP199A4T252A variants revealed that
the secondary structure, including the peptide carbonyl of
the Asp251 residue, and hydrogen bonding network of the
I-helix does not change dramatically (except in the interac-
tions with the removed hydroxyl group of the Thr252 side
chain, Fig. 2 and Table S4). Other available structures, of
the threonine to alanine mutants (or the dioxygen or cyanide
bound forms) of CYP enzymes are limited. Like P450cam,
the closely related CYP101D1 undergoes a widening of this
groove but P450Bm3 does not (Figs. S9–S12, Table S3).
The location of this bulge in the I-helix, which makes up
the O2 binding groove, difers across these P450 enzymes
(Figs. S13–S16) [52]. Between the I-helix residues 248 and
252 of CYP199A4 there is a less pronounced deviation
from the normal α helix hydrogen bonding pattern com-
pared to P450cam. One of the diferences in the hydrogen
bonding network between CYP199A4 and P450camWT is
due to the position of the backbone carbonyl of aspartate
251. In CYP199A4WT this forms hydrogen bonding inter-
actions with the backbone and side chain amide nitrogen of
asparagine 255 (Fig. 2). Another important observation is
that the side-chain alcohol of threonine 253 participates in
the hydrogen bonding network of the I-helix of CYP199A4
(Fig. 2). This is not possible in P450cam where the equiva-
lent residue is a valine and the side chain of this residue
occupies a signifcantly diferent position (Fig. 2 and Figs.
S13 and S16). Comparing the I-helix of CYP199A4WT and
CYP199A4T252A to the other P450 enzymes reveals that they
are more similar in the vicinity of the oxygen binding groove
to P450Bm3 and P450terp than the CYP101 family enzymes
(Figs. S14 and S15). Like CYP199A4 these P450s also have
a threonine residue after the conserved acid-alcohol pair
which contributes to the I-helix hydrogen bonding network
(Fig. S16). We hypothesise that these structural changes may
be related to the modifed behaviour observed within difer-
ent members of this enzyme superfamily.
CYP199A4 and P450cam
P450
Measurement
WT
T252Aa
CYP199A4
N - - 1 - - N
O - - 2 - - N
O - - 3 - - Cβ
N - - 1 - - N
O - - 2 - - N
O - - 3 - - Cβ
5.8
4.2
3.5
5.7
3.9
3.7
5.5 (−0.3)
3.6 (−0.6)
3.5 (0)
6.0 (0.3)
4.2 (0.3)
4.6 (0.9)
P450cam
Shown is a generalised schematic of the loop region. 1, 2 and 3
denote the key measurements from one side of the groove to the
other. 1 backbone N of G249 and backbone N of T252 (or A) 2 car-
bonyl O of A248 and backbone N of T252 (or A) 3 carbonyl O of
A248 and Cβ of T252 (or A; see Table S3 for further details). For
CYP199A4, the measurements of the 4-methoxybenzoic acid-bound
forms are presented
aThe diference between the WT and the T252A mutant is shown
in brackets. Distances are in angstroms, Å. Additional measure-
ments with the CYP199A4D251N mutant and in other P450 systems
(CYP101D1 and P450Bm3) are provided in the Supporting Informa-
tion (Table S3)
bond to the carbonyl groups of both Ala248 (2.6 Å) and
Ala249 (3.0 Å) and is 3.3 Å from the alanine 252 side chain
carbon (Table S2). The water was 6.6 Å from the heme iron
and 7.8 Å from the methyl group of 4-methoxybenzoic acid.
The modifcations above resulted in very little change at
the mouth of the oxygen binding groove in the I-helix of
CYP199A4T252A, despite the smaller size of the alanine
residue. However, the remainder of the groove was slightly
narrower compared to the WT enzyme (Fig. 1, Table 3 and
Table S3).
To put these observations in context we compared the
changes observed in the crystal structure of CYP199A4T252A
to those which occurred in the equivalent variant of
P450cam. In P450camT252A, two water molecules occupy
the oxygen binding groove compared to a single water in
the P450camWT enzyme (Fig. 1, Fig. S9). The location of
these two waters (W220 and W187) in the mutant structure
are diferent to that in the P450camWT and CYP199A4T252A
structures. The closest water (W220) is held 5.9 Å from the
heme iron and 4.1 Å from the closest methyl of camphor. In
addition, the T252A mutation of P450cam results in more
signifcant structural changes to the oxygen binding groove
(Fig. 2). The backbone carbonyl of Asp251 fips, resulting in
the groove being signifcantly wider in the mutant than when
compared to the WT enzyme (Fig. 1, Table 3 and Table S3).
Of particular note is that the widening of the oxygen bind-
ing groove in the I-helix of P450camT252A is also observed
in the O2-bound crystal structure of P450camT252A (PDB:
2A1M, Fig. S9) and in the T260A mutant of the closely
related enzyme CYP101D1 from Novosphingobium aromati-
civorans (PDB: 4C9P, Fig. S10) [13, 51].
Crystal structure of the 4‑methoxybenzoic
acid‑bound form of CYP199A4D251N
In the CYP199A4D251N crystal structure (1.59 Å;
CYP199A4D251N aligned to CYP199A4WT: rmsd=0.542 Å
over 392 of 393 residues; CYP199A4D251N aligned to
CYP199A4T252A: rmsd = 0.329 Å over 393 of 393 resi-
dues; Fig. S7), there are no major changes in the active site
environment near the substrate or heme, nor are there any
additional water molecules (Fig. 3). In the I-helix the Cβ–Cγ
bond of the Asn251 residue in the mutant is rotated by 13°
compared to the same bond of the WT enzyme (Fig. 3). The
1 3

JBIC Journal of Biological Inorganic Chemistry
Fig. 2 The hydrogen bond-
ing network of the I-helix of a
CYP199A4 and b P450cam.
In each, the WT structure is
blue while the T252A mutant is
yellow. The helix representation
is a tube in the same colours.
Water molecules and hydrogen
bonds are depicted as small
spheres in the same colours of
their structures. The position of
the O2 binding groove, where
the majority of the structural
changes occur on mutation of
the Thr252 residue, is indi-
cated. In CYP199A4, minimal
changes in the helix backbone
are observed (Table S4). In
P450cam, the backbone moves
more signifcantly in the vicinity
of the D251 and T252 residues.
Note the altered position of the
side chain of CYP199A4T253
compared to P450camV253
side chain of the nearby Asn255 residue has also under-
gone a 90° rotation, and no longer interacts with the mutated
Asn251 residue.
Discussion
The mechanism of O-dealkylation is initiated via hydrogen
abstraction at a substrate carbon by Cpd I followed by radi-
cal rebound in a similar fashion to hydroxylation reactions
[6, 30, 53]. Recently C–H bond abstraction and the radical
rebound steps have been described as being dynamically
concerted (Scheme 1) [10]. Studies with the threonine to
alanine mutation within the acid-alcohol pair of P450s have
been central to the proposal of Cpd I as the oxidant involved
in C–H bond abstraction. With CYP199A4T252A there was
a drop in the productive turnover of the enzyme due to a
reduction in the coupling efciency with a concomitant
increase in hydrogen peroxide formation. This was in agree-
ment with the observations made with P450cam even if the
switch was not as dramatic. In the camphor oxidising system
there was a small change in the rate of NADH oxidation but
95% of the reducing equivalents were lost to hydrogen per-
oxide formation. Similar results to those of P450cam have
been reported with other closely related systems, for exam-
ple CYP101D1 from N. aromaticivorans which also oxidises
Compared to P450cam the D251 side chain in CYP199A4
does not interact with a salt bridge network (Fig. 3). There
is a direct interaction with the side chain of Asp251 and
one water molecule (Wat39) in the WT enzyme which is
maintained in CYP199A4D251N (Wat110, Fig. 3 and Fig.
S17). This water molecule is connected to the surface of the
enzyme through a hydrogen bonded chain of waters (Fig. 4
and Fig. S18). These observations, in conjunction with the
biochemical activity data, suggest that while the D251 resi-
due in CYP199A4 plays a similar role in proton delivery
to the equivalent residue in P450cam the precise details of
how this is achieved varies between the two systems (Figs.
S17, S19).
1 3

JBIC Journal of Biological Inorganic Chemistry
Fig. 3 a The active site of D251NCYP199A4. 4-Methoxybenzoic acid is
shown in green, the heme in grey and water molecules as red spheres.
b A comparative overlay of the active site of the crystal structures
of 4-methoxybenzoic acid-bound forms of WT CYP199A4 (cyan)
and the D251N mutant (yellow). The water molecules are shown as
spheres in the colour of their respective structures
Fig. 4 The solvent network extending from the 251 residue to the
surface of WTCYP199A4 through a solvent access channel. 4-Methoxy-
benzoic acid is green, the heme and hydrogen bonds are grey, and
water molecules are red spheres. There are more water molecules
in the solvent network of CYP199A4D251N compared to the WT. In
CYP199A4D251N, the D251-N255 hydrogen bond is broken. See Figs.
S17 and S19 for the P450cam interactions
1 3

JBIC Journal of Biological Inorganic Chemistry
camphor [51]. The large reduction in product formation in
P450cam implicated Cpd I as the active oxidant in camphor
hydroxylation. However, in other systems diferences are
reported. The P450Bm3T268A mutant displayed a moderate
reduction in both the NADPH oxidation and product forma-
tion activities compared to the WT enzyme with C12 and
C16 fatty acids [54, 55]. Later studies demonstrated that
the activity of P450Bm3T268A was maintained with C14 and
C15 fatty acids. This suggested a substrate-dependence for
the activity of P450Bm3, and a less critical role for the con-
served threonine in this enzyme. However, P450Bm3 is an
unusual fused P450 which uses a diferent electron transfer
confguration than most other bacterial P450s [56]. The oxi-
dation of the substituted benzoic acids by CYP199A4T252A
reported here is more reminiscent of fatty acid hydroxyla-
tion by P450Bm3T268A, where more moderate changes in the
activity are observed.
modifcations in the hydrogen bonding network and the fip
in the carbonyl peptide backbone of Asp251. This alters the
degree of solvent access in this region. Unfortunately there
is, as yet, no equivalent comparative substrate-bound crystal
structure data available for the P450Bm3 system but there is
very little diference between the substrate free forms of the
WT and the T268A mutants which is consistent with what
we observed here with CYP199A4. The additional water
molecule is positioned signifcantly further away from the
heme iron in CYP199A4T252A compared to P450camT252A
.
Given that the CYP199A4T252A has reduced peroxide uncou-
pling compared to P450camT252A, this additional water mol-
ecule may be bound far enough away to minimise the extent
to which it might interact and interfere with or replace the
threonine hydroxyl group in the protonation steps.
The behaviour of CYP199A4D251N is similar to those of
other analogous P450 variants. For example, in P450cam,
P450cin and P450Bm3 mutation of the equivalent residue
leads to a signifcant reduction in the NAD(P)H oxidation
rate and activity [11, 21–24]. In CYP199A4 there is an obvi-
ous direct link between the Asp251 residue and external
water molecules as a source of protons. Our structures show
that the chain of water molecules directly linked to Asp251
extends to the surface of the enzyme. There is a hydrogen
bond from this acid residue to Asn255 but in 4-methoxy-
benzoic acid-bound CYP199A4D251N structure it is broken.
This residue may possess conformational fexibility and this
may be related to its role in proton delivery. It is important
to note that no obvious source of protons is observed in the
closed structures of P450cam. However, in open conforma-
tions, for example in the putidaredoxin bound complex the
aspartate is connected via a chain of water molecules to the
surface (Fig. S19) [2].
The mechanism of hydroxylation and desaturation are
both believed to be initiated via hydrogen abstraction at
a substrate carbon by Cpd I followed by radical rebound
[29]. The branching point for P450 catalysed desaturation
(dehydrogenation) and hydroxylation of 4-ethylbenzoic acid
would occur after hydrogen abstraction by Cpd I [29]. The
product distribution for the oxidations of both 4-ethylben-
zoic acid did not change signifcantly in the mutant variants
when compared to the WT. If there was a change in the
mechanism or orientation of substrate binding, one might
expect to see larger diferences in the partition between
hydroxylation and desaturation or in the enantioselectivity
of oxidation. The small diferences in the regio- and enantio-
selectivity observed with the 4-ethylbenzoic acid could be
due to the subtle changes in the position of the ethyl sub-
stituent in the mutant enzyme but this would require further
investigation. Given that the requirement for Cpd I for P450
C–H bond hydroxylation is well established, we propose
that the smaller drop in activity for the threonine to alanine
variants of CYP199A4 and P450Bm3 is that Cpd I can form
more readily in these mutant systems, when a substrate is
present.
The crystal structure of P450cam reveals that Asp251
interacts strongly with Lys178 and Arg186 [13]. As men-
tioned above when the peptide backbone carbonyl of Asp251
in P450cam fips upon oxygen binding and in P450camT252A,
there is a widening of the oxygen binding groove enabling
the entry of additional water molecules. This conformation
change is hypothesised to release conformational strain
within the helix and is thought to be confned to only a
small region due to the interaction of Asp251 with Lys178
and Arg186. In P450camD251N the strong ionic contacts are
absent, allowing a larger degree of conformational freedom
within the helix. In this mutant the I-helix opens up to a
lesser extent than in the WT enzyme, and there is insuf-
fcient space for new water molecules to be accommodated
within the active site near Thr252 and the ferrous-dioxygen
complex to establish a Thr-OH; O=O; OH2 network [2]. The
P450camT252A/D251N double mutant has the reduced activity
of P450camD251N but generates more hydroxylated metabo-
lite and less hydrogen peroxide than P450camT252A [57].
This suggests that subtle changes in the enzyme structure in
The structure of the 4-methoxybenzoic acid-bound form
of CYP199A4T252A provides insight into why this might
be the case. There are many similarities with the changes
observed in the P450camT252A camphor bound structure,
such as the replacement of the broken T252-G248 hydrogen
bond with a water molecule. However, there are important
diferences. The widening of the oxygen binding groove
in P450camT252A, which is also observed in the P450-oxy
bound crystal structure of P450cam and in the equivalent
mutant of CYP101D1, is not observed in CYP199A4T252A
.
In fact widening is only observed at the groove entrance in
the CYP199A4 system, due to the smaller alanine residue,
and the inside of the groove is actually narrower (Table S3).
This structural change in P450cam arises through
1 3

JBIC Journal of Biological Inorganic Chemistry
this region can have a signifcant efect on the enzyme activ-
ity. For example, changing the size of the residue at position
252 of P450cam has been shown to have an efect on the
activity of this enzyme. P450camT252V and P450camT252I
generated more product and less hydrogen peroxide than
P450camT252A [17, 57] However, P450camT252S and
P450camT252N mutants were more active than all mutants
containing hydrophobic substitutions [57, 58].
from the bacterial species Frankia inefcax and Mycobac-
terium marinum contain a second alcohol-containing amino
acid residue directly after the acid-alcohol pair [62, 63].
Analysis of the I-helix hydrogen bonding patterns highlights
that this threonine, when present, is involved in hydrogen
bonding interactions in the I-helix, though it interacts with
diferent residues in specifc enzymes (Fig. S16). Therefore
it appears that CYP199A4, which contains a second threo-
nine, may have more features in common with the majority
of cytochrome P450 enzymes in this region of the I-helix.
The valine residue in P450cam has been studied in mutant
forms of the enzyme (T252N/V253T) and its role in P450cin
(T243A) has been investigated (Fig. S16) [58, 64]. It was
found to alter substrate binding and activity but not the
selectivity. However, as yet the structural changes induced
in the I-helix by these mutants have not been determined.
The CYP101D1 enzyme, which is very similar to
P450cam, shows a similar reduction in the NADH oxida-
tion rate with the equivalent D259N mutant [51]. However,
it does not have such an extensive salt bridge network com-
pared to P450cam. Here the Lys178 residue of the latter
has been replaced with a glycine and this has been linked
to a less stringent efector role for the electron transfer part-
ner ferredoxin in this system [51]. In CYP101D1 similar
structural changes are also observed in the oxygen binding
groove upon mutation of the Thr260 residue and cyanide
binding (an oxygen binding mimic). In both camphor oxidis-
ing systems (P450cam and CYP101D1) the presence of new
water molecules in the oxygen-bound form of the enzyme
is believed to be important for the protonation of the oxy-
gen. The crystal structure of the oxy- or cyano-bound forms
of the WT and mutant forms of CYP199A4 could provide
valuable information on whether the same subtle structural
changes occur in this P450 system upon oxygen binding.
As the aspartate residue is tied up in salt bridge networks
in P450cam and CYP101D1 it is proposed that these inter-
actions have to break during the catalytic cycle and this is
achieved through dynamic changes in the enzyme.
Conclusion
In contrast to P450cam the T252 residue in the CYP199A4
system (and the equivalent residue in P450Bm3) has
been shown to be important for activity but not essen-
tial. CYP199A4T252A could support O-demethylation and
hydroxylation activity but the level of hydrogen peroxide
uncoupling did increase. The activity was lower than that
of the WT enzyme. Structural changes in the I-helix, which
are important for P450cam function, are not observed in
CYP199A4T252A. This was hypothesised to be due to several
factors in CYP199A4 including the presence of a second
threonine residue, which contributes to the hydrogen bond-
ing network of the I-helix near the oxygen binding groove,
and the lack of a salt bridge network with the acid residue
of the acid-alcohol pair. Both of these features are unusual
in the P450 superfamily but are found in the model CYP101
family systems. The results explain the variations reported in
other studies with diferent bacterial and mammalian P450
systems. They also provides a rationale of how the conserved
threonine residue could be altered during protein evolution
without a complete loss of activity through uncoupling reac-
tions: for example in the natural variants P450cin (Asn) [11]
and P450eryF (Ala) [12], and in directed evolution studies
with P450Bm3 (Ala) [65]. Critically it also suggests that
care should be taken when extending the results obtained
with P450cam to other systems.
The absence of salt bridges and the lack of a require-
ment to open a water channel may explain why the struc-
tural changes observed in the oxygen binding groove in
CYP199A4T252A are not as substantial as in P450camT252A
.
Another diference in CYP199A4 is that the side chain of
threonine 253 is involved in the hydrogen bonding of the
I-helix and could contribute to the diferences in the struc-
ture around the oxygen binding groove. This threonine resi-
due is conserved in P450Bm3 and P450terp which share a
similar I-helix conformation with CYP199A4. The equiva-
lent residue in P450cam and CYP101D1 (and CYP101D2)
is valine. While not a strictly conserved residue across this
superfamily of enzymes the presence of a threonine at this
position is relatively common. For example, 44 out of the
57 human P450 enzymes have a threonine (or serine) at
this position (for comparison 45 of these enzymes have the
acid residue while 50 posess the alcohol residue of the acid-
alcohol pair) [59]. This includes the CYP2B4 and CYP2E1
variants which were used to study the role of the alcohol
residue of the acid-alcohol pair by Vaz and Coon [60, 61].
Furthermore 54 out of 68 and 41 out of 47 P450 enzymes
Acknowledgements This work was supported by ARC Grant
DP140103229 (to JJDV and SGB). SGB acknowledges the ARC for a
Future Fellowship (FT140100355). The authors also acknowledge the
award of Australian Government Research Training Program Scholar-
ships (PhD to TC and MPhil to MNP). We would like to thank the
scientists at MX1 beamline for help with data collection. We acknowl-
edge ANSTO for fnancial support and in providing the facility used
in this work.
1 3

JBIC Journal of Biological Inorganic Chemistry
33. Bell SG, Xu F, Forward I, Bartlam M, Rao Z, Wong L-L (2008)
J Mol Biol 383:561–574
Compliance with ethical standards
34. Omura T, Sato R (1964) J Biol Chem 239:2370–2378
35. Coleman T, Chao RR, De Voss J, Bell SG (2016) Biochim Bio-
phys Acta Proteins Proteom 1864:667–675
Conflict of interest None of the authors have any conficts of interest
to declare.
36. Maddigan NK, Bell SG (2017) Arch Biochem Biophys 615:15–21
37. Williams JW, Morrison JF (1979) Methods Enzymol 63:437–467
38. Xu F, Bell SG, Rao Z, Wong LL (2007) Protein Eng Des Sel
20:473–480
References
39. Cowieson NP, Aragao D, Clift M, Ericsson DJ, Gee C, Harrop SJ,
Mudie N, Panjikar S, Price JR, Riboldi-Tunniclife A, Williamson
R, Caradoc-Davies T (2015) J Synchrotron Radiat 22:187–190
40. McPhillips TM, McPhillips SE, Chiu HJ, Cohen AE, Deacon AM,
Ellis PJ, Garman E, Gonzalez A, Sauter NK, Phizackerley RP,
Soltis SM, Kuhn P (2002) J Synchrotron Radiat 9:401–406
41. Battye TG, Kontogiannis L, Johnson O, Powell HR, Leslie AG
(2011) Acta Crystallogr D Biol Crystallogr 67:271–281
42. Evans PR, Murshudov GN (2013) Acta Crystallogr D Biol Crys-
tallogr 67:1204–1214
1. Ortiz de Montellano PR (2010) Chem Rev 110:932–948
2. Poulos TL (2014) Chem Rev 114:3919–3962
3. Ogliaro F, de Visser SP, Cohen S, Sharma PK, Shaik S (2002) J
Am Chem Soc 124:2806–2817
4. Rittle J, Green MT (2010) Science 330:933–937
5. Sharma PK, De Visser SP, Shaik S (2003) J Am Chem Soc
125:8698–8699
6. Meunier B, de Visser SP, Shaik S (2004) Chem Rev
104:3947–3980
7. Shaik S, Cohen S, Wang Y, Chen H, Kumar D, Thiel W (2010)
Chem Rev 110:949–1017
43. Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley P, Evans
PR, Keegan RM, Krissinel EB, Leslie AG, McCoy A, McNicholas
SJ, Murshudov GN, Pannu NS, Potterton EA, Powell HR, Read
RJ, Vagin A, Wilson KS (2011) Acta Crystallogr D Biol Crystal-
logr 67:235–242
8. Groves JT, McClusky GA (1976) J Am Chem Soc 98:859–861
9. Huang X, Groves JT (2017) J Biol Inorg Chem 22:185–207
10. Sarkar MR, Houston SD, Savage GP, Williams CM, Krenske EH,
Bell SG, De Voss JJ (2019) J Am Chem Soc 141:19688-19699
11. Stok JE, Yamada S, Farlow AJ, Slessor KE, De Voss JJ (2013)
Biochim Biophys Acta 1834:688–696
44. McCoy AJ, Grosse-Kunstleve RW, Adams PD, Winn MD, Storoni
LC, Read RJ (2007) J Appl Crystallogr 40:658–674
45. Emsley P, Lohkamp B, Scott WG, Cowtan K (2010) Acta Crystal-
logr D Biol Crystallogr 66:486–501
12. Cupp-Vickery JR, Han O, Hutchinson CR, Poulos TL (1996) Nat
Struct Biol 3:632–637
46. Adams PD, Afonine GBPV, Chen VB, Davis IW, Echols N, Headd
JJ, Hung L-W, Kapral GJ, Grosse-Kunstleve RW, McCoy AJ,
Moriarty NW, Oefner R, Read RJ, Richardson DC, Richardson
JS, Terwilliger TC, Zwart PH (2010) Acta Crystallogr D Biol
Crystallogr 66:213–221
13. Nagano S, Poulos TL (2005) J Biol Chem 280:31659–31663
14. Raag R, Martinis SA, Sligar SG, Poulos TL (1991) Biochemistry
30:11420–11429
15. Vidakovic M, Sligar SG, Li H, Poulos TL (1998) Biochemistry
37:9211–9219
47. Jefcoate CR (1978) Methods Enzymol 52:258–279
48. Dawson JH, Andersson LA, Sono M (1982) J Biol Chem
257:3606–3617
16. Schlichting I, Berendzen J, Chu K, Stock AM, Maves SA, Benson
DE, Sweet RM, Ringe D, Petsko GA, Sligar SG (2000) Science
287:1615–1622
49. Dawson JH, Andersson LA, Sono M (1983) J Biol Chem
258:13637–13645
17. Imai M, Shimada H, Watanabe Y, Matsushima-Hibiya Y, Makino
R, Koga H, Horiuchi T, Ishimura Y (1989) Proc Natl Acad Sci
USA 86:7823–7827
50. Podgorski MN, Coleman T, Chao RR, De Voss JJ, Bruning JB,
Bell SG (2020) J Inorg Biochem 203:110913
18. Gerber NC, Sligar SG (1994) J Biol Chem 269:4260–4266
19. Martinis SA, Atkins WM, Stayton PS, Sligar SG (1989) J Am
Chem Soc 111:9252–9253
51. Batabyal D, Poulos TL (2013) Biochemistry 52:8898–8906
52. Hasemann CA, Kurumbail RG, Boddupalli SS, Peterson JA,
Deisenhofer J (1995) Structure 3:41–62
20. Gerber NC, Sligar SG (1992) J Am Chem Soc 114:8742–8743
21. Ishigooka M, Shimizu T, Hiroya K, Hatano M (1992) Biochem-
istry 31:1528–1531
53. Guengerich FP (2001) Chem Res Toxicol 14:611–650
54. Yeom H, Sligar SG, Li H, Poulos TL, Fulco AJ (1995) Biochem-
istry 34:14733–14740
22. Yeom H, Sligar SG (1997) Arch Biochem Biophys 337:209–216
23. Lee-Robichaud P, Akhtar ME, Akhtar M (1998) Biochem J 330(Pt
2):967–974
55. Truan G, Peterson JA (1998) Arch Biochem Biophys 349:53–64
56. Cryle MJ, De Voss JJ (2008) ChemBioChem 9:261–266
57. Hishiki T, Shimada H, Nagano S, Egawa T, Kanamori Y, Makino
R, Park SY, Adachi S, Shiro Y, Ishimura Y (2000) J Biochem
128:965–974
24. Zhou DJ, Korzekwa KR, Poulos T, Chen SA (1992) J Biol Chem
267:762–768
25. Kimata Y, Shimada H, Hirose T, Ishimura Y (1995) Biochem
Biophys Res Commun 208:96–102
58. Kim D, Heo YS, Ortiz de Montellano PR (2008) Arch Biochem
Biophys 474:150–156
26. Bell SG, Hoskins N, Xu F, Caprotti D, Rao Z, Wong LL (2006)
Biochem Biophys Res Commun 342:191–196
59. Dhers L, Ducassou L, Boucher JL, Mansuy D (2017) Cell Mol
Life Sci 74:1859–1869
27. Bell SG, Tan AB, Johnson EO, Wong LL (2010) Mol Biosyst
6:206–214
60. Vaz AD, Pernecky SJ, Raner GM, Coon MJ (1996) Proc Natl Acad
Sci USA 93:4644–4648
28. Bell SG, Yang W, Tan AB, Zhou R, Johnson EO, Zhang A, Zhou
W, Rao Z, Wong LL (2012) Dalton Trans 41:8703–8714
29. Bell SG, Zhou R, Yang W, Tan AB, Gentleman AS, Wong LL,
Zhou W (2012) Chem Eur J 18:16677–16688
61. Vaz AD, McGinnity DF, Coon MJ (1998) Proc Natl Acad Sci USA
95:3555–3560
62. Lau ICK, Feyereisen R, Nelson DR, Bell SG (2019) Arch Bio-
chem Biophys 669:11–21
30. Coleman T, Chao RR, Bruning JB, De Voss J, Bell SG (2015)
RSC Adv 5:52007–52018
63. Child SA, Bradley JM, Pukala TL, Svistunenko DA, Le Brun NE,
Bell SG (2018) Chem Sci 9:7948–7957
31. Chao RR, De Voss JJ, Bell SG (2016) RSC Adv 6:55286–55297
32. Coleman T, Wong SH, Podgorski MN, Bruning JB, De Voss JJ,
Bell SG (2018) ACS Catal 8:5915–5927
64. Slessor KE, Farlow AJ, Cavaignac SM, Stok JE, De Voss JJ (2011)
Arch Biochem Biophys 507:154–162
1 3

JBIC Journal of Biological Inorganic Chemistry
65. Coelho PS, Brustad EM, Kannan A, Arnold FH (2013) Science
339:307–310
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional afliations.
Afliations
Tom Coleman1 · Jeanette E. Stok2 · Matthew N. Podgorski1 · John B. Bruning3 · James J. De Voss2
·
Stephen G. Bell1
2
3
* Stephen G. Bell
School of Chemistry and Molecular Bioscience, University
of Queensland, Brisbane, QLD 4072, Australia
stephen.bell@adelaide.edu.au
School of Biological Sciences, University of Adelaide,
Adelaide, SA 5005, Australia
1
Department of Chemistry, University of Adelaide, Adelaide,
SA 5005, Australia
1 3