Selective hydroxylation of 1,8- and 1,4-cineole using bacterial P450 variants

Article abstract of DOI:10.1016/j.abb.2018.12.025

This study has evaluated the use of the P450 metalloenzymes CYP176A1, CYP101A1 and CYP102A1, together with engineered protein variants of CYP101A1 and CYP102A1, to alter the regioselectivity of 1,8- and 1,4-cineole hydroxylation. CYP176A1 was less selective for 1,4-cineole oxidation when compared to its preferred substrate, 1,8-cineole. The CYP102A1 variants significantly improved the activity over the WT enzyme for oxidation of 1,4- and 1,8-cineole. The CYP102A1 R47L/Y51F/A74G/F87V/L188Q mutant generated predominantly (1S)-6α-hydroxy-1,8-cineole (78% e.e.) from 1,8-cineole. Oxidation of 1,4-cineole by the CYP102A1 R47L/Y51F/F87A/I401P variant generated the 3α product in >90% yield. WT CYP101A1 formed a mixture metabolites with 1,8-cineole and very little product was generated with 1,4-cineole. In contrast the F87W/Y96F/L244A/V247L and F87W/Y96F/L244A variants of CYP101A1 favoured formation of 5α-hydroxy-1,8-cineole (>88%, 1S 86% e.e.) while the F87V/Y96F/L244A variant generated (1S)-6α-hydroxy-1,8-cineole in excess (90% regioselective, >99% e.e.). The CYP101A1 F87W/Y96F/L244A/V247L and F87W/Y96F/L244A mutants improved the oxidation of 1,4-cineole generating an excess of the 3α metabolite (1S > 99% e.e. with the latter). The CYP101A1 F87L/Y96F variant also improved the oxidation of this substrate but shifted the site of oxidation to the isopropyl group, (8-hydroxy-1,4-cineole). When this 8-hydroxy metabolite was generated in significant quantities desaturation of C8–C9 to the corresponding alkene was also detected.

Full text of DOI:10.1016/j.abb.2018.12.025

Accepted Manuscript
Selective hydroxylation of 1,8- and 1,4-cineole using bacterial P450 variants
Joel H.Z. Lee, Siew Hoon Wong, Jeanette E. Stok, Sarah A. Bagster, James Beckett,
Jack K. Clegg, Aidan Brock, James J. De Voss, Stephen G. Bell
PII:
S0003-9861(18)30673-8
https://doi.org/10.1016/j.abb.2018.12.025
YABBI 7900
DOI:
Reference:
To appear in: Archives of Biochemistry and Biophysics
Received Date: 29 August 2018
Revised Date: 24 October 2018
Accepted Date: 21 December 2018
Please cite this article as: J.H.Z. Lee, S.H. Wong, J.E. Stok, S.A. Bagster, J. Beckett, J.K. Clegg, A.
Brock, J.J. De Voss, S.G. Bell, Selective hydroxylation of 1,8- and 1,4-cineole using bacterial P450
variants, Archives of Biochemistry and Biophysics (2019), doi: https://doi.org/10.1016/j.abb.2018.12.025.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Selective hydroxylation of 1,8- and 1,4-cineole using bacterial P450 variants
Joel H. Z. Leea, Siew Hoon Wongb, Jeanette E. Stokb, Sarah A. Bagstera, James Beckettb,
Jack K. Cleggb, Aidan Brockb, James J. De Vossb and Stephen G. Bella
a Department of Chemistry, School of Physical Sciences, University of Adelaide, Adelaide,
SA 5005, Australia
b Department of Chemistry, School of Chemistry and Molecular Biosciences, University of
Queensland, St. Lucia, Brisbane QLD 4072, Australia
Dr Stephen G. Bell
E-mail: stephen.bell@adelaide.edu.au
Prof. James J. De Voss
E-mail: j.devoss@uq.edu.au
1

ACCEPTED MANUSCRIPT
Abstract
This study has evaluated the use of the P450 metalloenzymes CYP176A1, CYP101A1 and
CYP102A1, together with engineered protein variants of CYP101A1 and CYP102A1, to
alter the regioselectivity of 1,8- and 1,4-cineole hydroxylation. CYP176A1 was less
selective for 1,4-cineole oxidation when compared to its preferred substrate, 1,8-cineole.
The CYP102A1 variants significantly improved the activity over the WT enzyme for
oxidation of 1,4- and 1,8-cineole. The CYP102A1 R47L/Y51F/A74G/F87V/L188Q mutant
generated predominantly (1S)-6α-hydroxy-1,8-cineole (78% e.e.) from 1,8-cineole.
Oxidation of 1,4-cineole by the CYP102A1 R47L/Y51F/F87A/I401P variant generated the
3α product in >90% yield. WT CYP101A1 formed a mixture metabolites with 1,8-cineole
and very little product was generated with 1,4-cineole. In contrast the
F87W/Y96F/L244A/V247L and F87W/Y96F/L244A variants of CYP101A1 favoured
formation of 5α-hydroxy-1,8-cineole (>88%, 1S 86% e.e.) while the F87V/Y96F/L244A
variant generated (1S)-6α-hydroxy-1,8-cineole in excess (90% regioselective, >99% e.e.).
The CYP101A1 F87W/Y96F/L244A/V247L and F87W/Y96F/L244A mutants improved the
oxidation of 1,4-cineole generating an excess of the 3α metabolite (1S >99% e.e. with the
latter). The CYP101A1 F87L/Y96F variant also improved the oxidation of this substrate but
shifted the site of oxidation to the isopropyl group, (8-hydroxy-1,4-cineole). When this 8-
hydroxy metabolite was generated in significant quantities desaturation of C8-C9 to the
corresponding alkene was also detected.
2

ACCEPTED MANUSCRIPT
Graphical Abstract
Highlights
CYP176A1 was less selective for 1,4-cineole oxidation compared to 1,8-cineole.
CYP102A1 and CYP101A1 variants were used to generate (1S)-6α-hydroxy-1,8-cineole in
excess (>90%).
Other mutant forms of CYP101A1 were selective for 5α-hydroxy-1,8-cineole (>88%, 86%
e.e.).
Oxidation of 1,4-cineole by CYP102A1 and CYP101A1 mutants generated the 3α product
in >90% yield.
The CYP101A1 F87L/Y96F variant favoured 8-hydroxy-1,4-cineole formation.
The formation of 8-hydroxy-1,4cineole was accompanied by desaturation to a C8-C9 alkene.
Keywords: enzyme catalysis; cytochrome P450; cineole; hydroxylation, C-H bond oxidation
3

ACCEPTED MANUSCRIPT
1. Introduction
The monoterpenoid group of natural products are widespread, abundant and structurally
diverse, and these attributes correlate with a wide range of biological functions [1, 2]. They
make up the major proportion of many essential oils and have uses as flavours, fragrances
and pheromones [1, 3, 4]. They also show antimicrobial and medicinal properties and have
uses as pesticides, herbicides and components of pharmaceuticals [5, 6]. The parent
monoterpenes are often easily accessible from natural oils but the hydroxylated or
oxyfunctionalised forms usually occur in lower amounts [3, 7].
The monoterpene 1,8-cineole 1 (1,3,3-trimethyl-2-oxabicyclo[2,2,2]octane) is the
major component of eucalyptus oil, while 1,4-cineole 2 (4-methyl-1-propan-2-yl-7-
oxabicyclo[2.2.1]heptane) is found in some plant oils and is also present in lime juice (Fig.
1) [8-10]. Both cineoles are achiral but a single hydroxylation usually leads to the generation
of at least two stereogenic centres. The oxygenated derivatives of both are important chiral
synthons, intermediates in the synthesis of herbicides (e.g. cinmethylin), antimicrobials and
fragrances [7, 9, 11-20]. The chemical synthesis of hydroxylated cineole analogues is
difficult and requires the use of highly reactive, non-environmentally friendly reagents that
usually generate a mixture of racemic products. The use of metalloenzyme biocatalysts to
selectively introduce oxygen into selected C-H bonds of the parent cineoles is very attractive
as it would allow these reactions to be performed in a single step under ambient conditions
[21-23].
There has been increasing attention to the use of metalloenzymes as biocatalysts for
the discovery of new routes for the synthesis of hydroxylated terpene analogues for the
production of antimicrobial and bactericidal agents or for the fragrance industry [5, 6, 24].
Of particular interest is the ability of such enzymes, especially the cytochromes P450
4

ACCEPTED MANUSCRIPT
(P450s), to synthesise specific oxygenated molecules in a highly regio- and enantioselective
fashion.
The P450s are a superfamily of hemoprotein monooxygenases that are involved in
hydroxylation, epoxidation, heteroatom dealkylations and other, more complex reactions
[25-27]. The majority of P450 enzymes utilise a reactive high-valent iron-oxo radical cation
intermediate, compound I (Cpd I), to insert one oxygen atom from dioxygen into unactivated
carbon-hydrogen bonds. The catalytic cycle of the enzyme involves multiple steps. After
initial substrate binding, the first electron transfer step reduces the ferric iron to the ferrous
form. The electrons required for this and a subsequent reduction step are usually sourced
from a nicotinamide cofactor (NADH or NADPH) and delivered via electron transfer
proteins [28-30]. Oxygen rapidly binds to the ferrous form and is then activated by a second
electron. Protein controlled delivery of protons facilitates dioxygen bond cleavage with
generation of the reactive intermediate, Cpd I [31]. The pathway of the oxygen insertion
reaction by Cpd I is accounted for by the radical rebound mechanism [32, 33]. In general,
bacterial P450 enzymes tend to have an active site architecture that holds the physiological
substrate in such a way that carbon-hydrogen bond hydroxylation occurs with high regio-,
stereo-, and enantioselectivity. The ability of these enzymes to catalyse this reaction under
ambient conditions offers advantages over traditional multistep or unselective organic
synthesis routes [34-36]. Hence, the bacterial P450 enzymes, which are often soluble, highly
active and selective, are well suited for biocatalytic applications [34-36]. P450s such as
CYP101A1 (P450cam), CYP102A1 (P450BM3) and CYP176A1 (P450cin) have previously
demonstrated their utility in catalysing the oxidation of a number of terpenes and are
therefore, ideal candidates for exploring the oxidation products that could be generated from
1,8- and 1,4-cineole [24, 37-41].
5

ACCEPTED MANUSCRIPT
CYP176A (P450cin) is an example of a bacterial P450 that has been shown to
catalyse the enantiospecific hydroxylation of 1,8-cineole 1 to produce (1R)-6β-
hydroxycineole 3a (Scheme 1, Fig. 2). This P450 was initially identified in a bacterium
thought to be Citrobacter braakii along with its redox partners, which included the FMN
containing flavodoxin, cindoxin (Cdx) [42, 43]. When reconstituted with Escherichia coli
flavodoxin reductase it is an NADPH dependent P450 system that efficiently catalyses the
enantiospecific hydroxylation of 1,8-cineole [42, 44]. This is the first step in the
biodegradation of cineole by the bacterium which can use this monoterpene as it sole source
of carbon and energy [15]. Mutagenesis studies have demonstrated that a range of
hydroxycineole isomers can be obtained from a number of single amino acid mutations of
CYP176A1 around the heme containing active site (Fig. 2) [45-47]. In addition, CYP176A1
is able to catalyse the less selective oxidation of other monoterpenes, such as both
enantiomers of camphor [24, 43]. An in vivo system, comprising of P450cin and Cdx and the
endogenous E. coli flavodoxin was constructed to enable larger scale metabolite generation
using intact cells [24].
The bacterial monooxygenase CYP101A1 (P450cam), from Pseudomonas putida,
catalyses the stereospecific hydroxylation of (1R)-camphor to 5-exo-hydroxycamphor (Fig.
2) [34, 48]. This oxidation, like that of 1,8-cineole by CYP176A1, is the first step in the
utilisation of the substrate as an energy source by the bacterium. It obtains its electrons from
NADH via a class I electron transfer system, comprising an FAD containing flavoprotein,
putidaredoxin reductase, and a [2Fe-2S] ferredoxin, putidaredoxin [49, 50]. The WT enzyme
has a limited substrate range outside of camphor but can oxidise other monoterpenes
including 1,8-cineole, which is hydroxylated less selectively to (1S)-6α-hydroxycineole 3b
and both enantiomers of 5α- and 5β-hydroxycineole 3c-d (Scheme 1) [40]. Mutations at
residues in the active-site of CYP101A1 have been found to improve its affinity, activity and
6

ACCEPTED MANUSCRIPT
selectivity toward other substrates, e.g. the monoterpenes (S)-limonene and (+)-α-pinene
(Fig. 2) [34, 37, 39, 51-56]. The enzyme has also been altered to oxidise substrates as varied
as small chain alkanes (<C5) and polycyclic aromatic hydrocarbons as large as pyrene [37,
39, 52, 55, 57-61].
The P450 CYP102A1 (P450BM3), which was found in Bacillus megaterium, is a fatty
acid hydroxylase that oxidises fatty acid substrates close to the omega terminus (Fig. 2) [62-
64]. This enzyme was the first example of a P450 in which the electron transfer partner
domains are fused to the heme domain. Other members of the CYP102A subfamily have
been found to have the same secondary structure properties [65-70] CYP102A1 has a high
specificity for NADPH as the electron source,[71] and the enzyme has been shown to work
as a dimer with very high monooxygenase activity [65-68, 72, 73]. It is soluble, easy to
produce and the self-sufficient nature and high activity often make it the P450 of choice for
biocatalytic applications. As such it has been used as a template for protein engineering and
evolution studies to design efficient and selective oxidation biocatalysts [34, 39, 74-82].
CYP102A1 variants, which enhance the oxidation activity for unnatural substrates but do
not alter the product regioselectivity have been identified [83-86]. Variants in which
phenylalanine 87 and other residues in the enzyme active site have been modified, alter the
substrate binding profile and product selectivity [76, 87-90]. For example, the GVQ variant
(A74G/F87V/L188Q) of CYP102A1 has been reported to be a biocatalyst for the oxidation
of hydrophobic organic molecules including alkyl napthalenes [83, 91]
The selective metalloprotein-catalyzed oxidation of cineoles to specific
hydroxycineoles could generate new green chemistry routes to important molecules.
Recently WT CYP101B1 from Novosphingobium aromativicorans was shown to generate
predominantly (1S)-5α-hydroxy-1,8-cineole 3c with high activity [40]. Others have reported
new bacterial enzymes of the CYP101J subfamily from Sphingobium yanoikuyae B2, which
7

ACCEPTED MANUSCRIPT
can oxidise 1,8-cineole to form (1R)-6α-hydroxy-1,8-cineole 3b [92, 93]. Whole-cell
microbial oxidation of 1,4-cineole 2 by Streptomyces griseus and other bacteria yielded
mixtures of the 8- and 9-hydroxy-1,4-cineoles (5a and 5d) and isomers from hydroxylation
at the 2-position [12, 17, 18]. Here we report results of oxidation of 1,8- and 1,4-cineole
using mutant forms of CYP102A1 and CYP101A1 in order to investigate the potential use
of these metalloenzymes as selective biocatalysts in the generation of specific
hydroxycineoles. Metabolites from the oxidation of 1,4-cineole by CYP176A1 are also
identified. In addition, whole-cell oxidation of both cineoles was used to compare the
activity and selectivity of the different P450 enzymes.
8

ACCEPTED MANUSCRIPT
2. Experimental Section
2.1 General
General reagents, 1,8-cineole 1 and 1,4-cineole 2 were from Sigma-Aldrich. Buffer
components, NADPH, and isopropyl-β-D-thiogalactopyranoside (IPTG) were from Astral
Scientific (Australia) or VWR. Production and purification of full-length CYP102A1
variants for in vitro use were carried out as described previously [83, 94]. UV/Vis
spectroscopy was performed on an Agilent Cary 60 spectrophotometer with temperature
control at 30 °C. Gas chromatography mass spectrometry (GC-MS) analyses were carried
out on a Shimadzu GC-2010 coupled to a GC-MS-QP2010S detector or a GC-17A coupled
to a QP5050A MS detector. Both systems used a DB-5 MS fused silica column (30 m x 0.25
mm, 0.25 µm) and helium as the carrier gas. The GC retention times are given in the figure
legends and the methods in the supplementary material. Additional GC analysis and
enantioselective chromatography were performed on a Shimadzu Tracera GC coupled to
Barrier discharge Ionization Detector (BID) detector using a Supelcowax column (30 m x
0.32 mm x 0.25 um) and a RT®-BDEXse chiral silica column (Restek; 30 m x 0.32 mm x
0.25 um), respectively.
Enantioselective GC analysis of the cineole metabolites was
performed on Cyclodextrin-B column (25 m × 0.25 mm) on a Shimadzu GC-17A equipped
with an FID.
2.2 Activity assays
CYP102A1 in vitro NADPH turnovers were run at 30 °C in 1200 µL of 50 mM Tris, pH 7.4
at 30 °C, containing 0.2 µM enzyme and 120 µg bovine liver catalase. The buffer was
saturated with oxygen gas just before use and the assays were allowed to equilibrate for 1
min prior to the addition of the appropriate cineole substrate (1 mM substrate from a 100
mM stock in DMSO). To initiate the reaction NADPH was added, from a 20 mg mL–1 stock,
to a final concentration of ~320 µM (equivalent to 2 AU). A period of 10 seconds was
9

ACCEPTED MANUSCRIPT
allowed to elapse after NADPH addition, to enable the rate of consumption to become
linear, before the absorbance at 340 nm was monitored. The reactions were allowed to run
until all the NADPH was consumed. The NADPH turnover rate was derived using ε340
=
6.22 mM−1 cm−1.
2.3 In vivo metabolite generation and product identification
The CYP101A1 variants were screened using a plasmid system pCWSGB++ (pCWori+
based) which contained the genes for PdR, Pdx and the relevant mutant as described
previously (Table S1) [37, 39, 51, 55, 57, 58]. CYP176A1 was screened using the
biscistronic system (also in pCWori+) containing P450 enzyme and Cdx which is supported
by the endogenous flavoprotein reductase of the E. coli [24, 95]. The plasmids transformed
in E. coli (DH5α) cells and a single colony was grown in 2 x YT medium containing
ampicillin (100 µg mL−1) and the protein was produced by induction with IPTG [53]. After
induction the cell pellet was harvested by centrifugation and resuspended in E. coli minimal
medium (EMM; K2HPO4, 7 g, KH2PO4, 3 g, Na3citrate, 0.5 g, (NH4)2SO4, 1 g, MgSO4, 0.1
g, 20% glucose, 20 mL and glycerol, 1% v/v per litre).
To isolate and identify the cineole products, and to compare the CYP102A1
turnovers a whole-cell oxidation system utilising the pET28 plasmid containing the
CYP102A1 gene of interest was used. The plasmids were transformed into competent
BL21(DE3) cells, grown in 2 x YT broth containing kanamycin (30 µg mL−1). Protein
generation was induced by addition of IPTG and the cultures were allowed to continue for
16 h before the cells were harvested by centrifugation. The cell pellet was washed and
resuspended in EMM. The cineoles (2 mM from a 100 mM stock in ethanol) were added to
the cells (50 ml in a 250 mL flask) the reactions were shaken at 150 rpm and 30 °C. A
second aliquot of substrate was added after 6 hours and the reactions were allowed to
10

ACCEPTED MANUSCRIPT
proceed for 16 hours. Aliquots of the turnover (1 mL) were taken and extracted with ethyl
acetate for analysis by GC-MS.
The CYP101A1 and CYP102A1 variants were transformed into the E. coli strain
BL21(DE3) and grown in LB with the relevant antibiotics (see above). The cells were
harvested, the biomass yield was recorded as the cell wet weight, before they were
resuspended in EMM (50 mL in 250 mL flask). The reactions were performed in duplicate
and analysed as described below. The concentration of P450 was determined by lysing the
cells (via sonication) and recording the CO difference spectrum of the supernatant [96].
2.4 Product analysis
After the NADPH consumption assays were completed with CYP102A1 variants or when
analysing the whole-cell oxidation systems, 990 µL of the reaction mixture (including the
cells in the case of whole-cell turnovers) was mixed with 10 µL of an internal standard
solution (p-cresol, 20 mM stock solution in DMSO). The mixture was extracted with 400 µL
of ethyl acetate and the organic extracts were used directly for GC-MS or GC analysis.
Products were initially identified by matching the GC-MS mass spectra to those expected for
the products (see supplementary material). To obtain the coupling efficiency products were
calibrated against 1,8-cineole and 1.4-cineole using the assumption that isomeric products
would give comparable responses.
The supernatant (50 mL) from an in vivo turnover of the cineoles was extracted with
ethyl acetate (3 x 50 mL), washed with brine (50 mL). The organic extracts were pooled,
dried with magnesium sulphate, filtered and the solvent was removed by vacuum distillation
and then under a stream of nitrogen. The products were purified by silica gel
chromatography using a hexane/ethyl acetate stepwise gradient. The composition of the
fractions was assessed by TLC and GC-MS and those containing a single major product
were combined for characterisation. The purified product was dissolved in CDCl3 and the
11

ACCEPTED MANUSCRIPT
organics characterised by NMR spectroscopy and GC-MS. When separation was not
possible fractions were analysed by NMR and distinctive signals were matched to those of
the metabolites reported in the literature. These NMR spectra were acquired on a Bruker AV
at 500 MHz.
2.5 NMR and MS data for 1,4-cineole products
The distinctive NMR signals and MS fragmentation patterns used to characterise the 1,4-
cineole derivatives are shown below. These were purified from in vivo turnovers with
enzymes discussed in this work or synthesised from such compounds as described. The
oxidised 1,8-cineole derivatives have been identified previously and were characterised by
comparison to previous literature or authentic standards using methods which have been
previously described [24, 40, 42, 46, 95].
8-hydroxy-1,4-cineole 5a [12, 97] (oil) lit [98].
1H NMR (CDCl3, 400 MHz): δ 1.25 (s, 6H, 9-CH3 & 10-CH3), 1.43 (s, 3H, 7-CH3).
13C NMR (CDCl3, 100 MHz): δ 21.1, 25.3 (2 × CH3), 31.9 (2 × CH2), 37.5 (2 × CH2), 71.6,
83.9, 91.7.
GC-MS (m/z): 170 (M+, 7), 155 (4), 137 (7), 112 (13), 111 (39), 110 (18), 109 (12), 97 (10),
95 (9), 93 (15), 85 (13), 83 (8), 79 (9), 69 (17), 67 (12), 59 (100), 55 (19), 43 (96).
3β-hydroxy-1,4-cineole 5b [12] (oil, major component in a mixture).
1H NMR (CDCl3, 400 MHz): δ 0.91 & 1.00 (2 × dd, J = 6.9 & 0.7 Hz, 2 × 3H, 9-CH3 & 10-
CH3), 1.45 (s, 3H, 7-CH3), 1.70 (td, J = 12.2 & 3.4 Hz, 1H), 2.18 (dd, J = 13.3 & 6.9 Hz,
1H), 2.36 (septet, J = 6.9 Hz, 1H), 3.89 (ddd, J = 9.3, 6.8 & 2.0 Hz, 1H).
GC-MS (m/z): 170 (M+, 7), 155 (1), 139 (1), 137 (3), 127 (8), 125 (6), 111 (8), 109 (8), 86
(54), 84 (43), 71 (58), 43 (100).
2β-hydroxy-1,4-cineole 5c [12] [13] (solid, major component in a mixture).
12

ACCEPTED MANUSCRIPT
1H NMR (CDCl3, 400 MHz): δ 0.94 & 0.95 (2 × d, J = 6.9 Hz, 2 × 3H, 9-CH3 & 10-CH3),
1.40 (s, 3H, 7-CH3), 2.04 (m, 2H), 3.73 (t, J = 7.5 Hz, 1H).
13C NMR (CDCl3, 100 MHz): δ 16.4, 18.18, 18.21, 32.60, 32.63, 32.9, 45.2, 76.9, 85.6, 88.8
GC-MS (m/z): 170 (M+, 4), 153 (4), 137 (2), 127 (6), 125 (9), 112 (17), 109 (10), 97 (13), 95
(10), 86 (10), 83 (19), 71 (23), 43 (100).
9-hydroxy-1,4-cineole 5d [12, 97] (oil, mixture) [97].
1H NMR (CDCl3, 400 MHz): δ 0.83 (d, J = 7.0 Hz, 10-CH3), 1.43 (s, 3H, 7-CH3), 3.48 (dd,
J = 11.2 & 4.2 Hz, 1H), 3.70 (dd, J = 11.1 & 9.6 Hz, 1H).
GC-MS (m/z): 170 (M+, 14), 155 (1), 141 (11), 139 (12), 137 (3), 123 (13), 111 (38), 97
(10), 95 (18), 93 (17), 91 (7), 87 (35), 83 (15), 81 (13), 79 (19), 69 (29), 67 (28), 55 (52), 43
(100).
2α-hydroxy-1,4-cineole 5e [99] (oil, major component in a mixture from whole-cell oxidation
system)
1H NMR (CDCl3, 400 MHz): δ 0.901 & 0.905 (2 × d, J = 6.9 Hz, 2 × 3H, 9-CH3 & 10-CH3),
1.37 (s, 3H, 7-CH3), 1.93 (sept, J = 6.8 Hz, 1H), 2.05 (m, 1H), 2.28 (ddd, J = 12.3, 8.6, & 5.7 Hz,
1H), 3.90 (m, 1H)
GC-MS (m/z): 170 (M+, 4), 153 (3), 140 (1), 137 (2), 127 (5), 125 (10), 112 (18), 111 (7), 97
(12), 95 (9), 83 (18), 81 (14), 71 (23), 58 (27), 43 (100)
23
3α-hydroxy-1,4-cineole 5f mp. 81 – 82 °C (lit. [100] mp. 56 °C); [α]D 17° (c 0.14,
CHCl3), lit. [100] [α]D 2.5° (c 0.5, CHCl3).
1H NMR (CDCl3, 500 MHz): δ 0.98 & 0.99 (2 × d, J = 6.9 Hz, 2 × 3H, 9-CH3 & 10-CH3),
1.34 (s, 3H, 7-CH3), 1.38 (dd, J = 12.5 & 3.1 Hz, 1H, 2-CH), 1.46 – 1.59 (m, 3H, OH, 5-CH
& 6-CH), 1.75 (m, 1H, 6-CH), 1.98 – 2.06 (m, 2H, 8-CH & 2-CH), 2.25 (ddd, J = 11.8, 9.4
& 4.3 Hz, 1H, 5-CH), 4.17 (dddd, J = 9.6, 4.6, 3.1 & 1.6 Hz, 1H, 3-CH).
13C NMR (CDCl3, 125 MHz): δ 17.9, 18.3, 21.7, 25.3, 32.3, 37.3, 47.9, 74.9, 83.0, 90.4
13

ACCEPTED MANUSCRIPT
GC-MS (m/z): 170 (M+, 8), 137 (2), 128 (1), 127 (8), 125 (9), 111 (9), 109 (7), 97 (6), 95
(4), 87 (22), 86 (55), 71 (55), 43 (100).
HR-MS (C10H19O2): Found 171.1374 (Calculated 171.1385).
Crystallography of 3α-hydroxy-1,4-cineole, 5f, was performed as described in the
supplementary material.
2.6 Synthesis of 1,4-cineole derivatives
3-keto-1,4-cineole 7b
A solution of tetrapropylammonium perruthnate (1.3 mg, 0.004 mmol) and 4-methylmorpholine
N-oxide (10.1 mg, 0.086 mmol) in dry methylene chloride (0.25 ml) was added dropwise to a
solution of 3-hydroxy-1,4-cineole (5 mg, 0.018 mmol) dissolved in dry methylene chloride (0.25
ml). The resulting mixture was stirred at room temperature for 1 hr before filtering through a
silica plug and concentrating under a gentle stream of N2 to give a clear oil (4.8 mg, 96%).
1H NMR (500 MHz, CDCl3): δ 2.22 (m, 2H, 6-CH2), 2.18-2.12 (sept, J = 7 Hz, 1H, 8-CH), 1.88
(td, J = 12.4, 4.5 Hz, 1H, ax 3-CH), 1.80-1.75 (m, 1H, ax 2-CH), 1.74 – 1.65 (m, 1H, eq 2-CH),
1.64 – 1.57 (m, 1H, eq 3-CH), 1.55 (s, 3H, 7-CH3), 1.06 (d, J = 6.9 Hz, 1H, 10-CH3), 1.05 (d, J =
7.0 Hz, 1H, 9-CH3).
13C NMR (125 MHz, CDCl3): δ 214.18, 91.85, 81.42, 51.24, 35.54, 28.29, 28.23, 21.54, 17.57,
17.35.
GC-MS (m/z): 168 (M+, 1), 140 (30), 126 (18), 97 (18), 71 (100), 69 (64), 55 (25), 43 (88), 41
(44).
2-keto-1,4-cineole 7a
A solution of tetrapropylammonium perruthnate (1 mg, 0.003 mmol) and 4-methylmorpholine
N-oxide (9.4 mg, 0.08 mmol) in dry methylene chloride (0.25 ml) was added dropwise to a
solution of 3-hydroxy-1,4-cineole (1 mg, 0.006 mmol) dissolved in dry methylene chloride (0.25
14

ACCEPTED MANUSCRIPT
ml). The resulting mixture was stirred at room temperature for 1 hr before filtering through a
silica plug and concentrating under a gentle stream of N2 to give a clear oil (1 mg, 100%).
GC-MS (m/z): 168 (M+, 1), 140 (49), 107 (35), 97 (39), 83 (26), 69 (29), 55 (59), 43 (100), 41
(27)
8-dehydro-1,4-cineole 6
Freshly distilled POCl3 (10 µL, 0.11 mmol) was added to a solution of 8-hydroxy-1,4-cineole
(10 mg, 0.059 mmol) dissolved in pyridine (0.2 ml) before stirring for 24 hours at RT. The
reaction as quenched with H2O (1 ml) and extracted with ethyl acetate (2 x 1 ml). The organic
layer was then washed with copper sulfate (2 x 1 ml), and brine (2 x 1 ml) before drying with
magnesium sulfate and concentrating under a gentle stream of N2 to afford the product as a clear
oil (3 mg, 30%).
1H NMR (500 MHz, CDCl3): δ 4.98 (dq, J = 1.8, 0.9 Hz, 1H, 10-CH2), 4.83 – 4.80 (m, 1H, 10-
CH2), 1.82 (dd, J = 1.5, 0.9 Hz, 3H, 9-CH3), 1.81-1.66 (m, 9H, 2,3,5,6-CH2) 1.49 (s, 3H, 7-CH3).
GC-MS (m/z): 152 (M+, 60), 123 (87), 109 (79), 79 (59), 69 (100), 67 (41), 43 (93), 41 (63).
15

ACCEPTED MANUSCRIPT
3. Results
3.1 1,4-Cineole
Identification of the products of P450 catalysed oxidation of 1,4-cineole 2 was achieved by
isolation and characterisation (NMR and mass spectrometry). In vivo oxidation reactions
were used to prepare pure or highly enriched samples of 5a-f (Scheme 2). These samples
allowed identification of the products of in vitro turnover by GC-MS analysis via retention
time and mass spectral fragmentation comparisons. In the case of 5f, an X-ray structure was
obtained for the product isolated from 1,4-cineole oxidation with F87W/Y96F/L244A
(WFA) CYP101A1 and allowed the determination of its absolute configuration as (1S).
3.1.1 CYP176A1 Oxidation: 1,4-Cineole 2
CYP176A1 catalyses the selective oxidation of 1,8-cineole 1 to (1R)-6β-hydroxycineole 3a
and possesses a high binding affinity for 1,8-cineole (Kd = 0.7 µM) which also has the
ability to induce a complete shift of the ferric heme from the low spin to the high spin form
[42, 43]. Based on these properties of CYP176A1, 1,4-cineole was examined as a potential
substrate. Binding studies were not possible with 1,4-cineole as it contains a small amount of
1,8-cineole that binds preferentially to CYP176A1 hindering quantitative analysis. Catalytic
oxidation of 1,4-cineole by CYP176A1 generated two major products that were identified as
2β-hydroxy-1,4-cineole 5c and 8-hydroxy-1,4-cineole 5a (Fig. 3 and Scheme 2). Low levels
of a third metabolite, which was assigned as 3β-hydroxy-1,4-cineole 5b, was also detected
(Fig. 3 and Scheme 2). Very low levels of a turnover product were also observed in the GC-
MS and this was subsequently identified as 6 (vide infra).
3.1.2 CYP101A1 Oxidation: 1,4-Cineole 2
Two other enzymes, CYP101A1 and CYP102A1, and their mutants were then assessed to
determine if these P450s could selectively oxidise 1,8- and 1,4-cineole. Firstly, WT
16

ACCEPTED MANUSCRIPT
CYP101A1 oxidation of 1,4-cineole using a whole-cell system resulted in low yields and
non-selective oxidation (Fig. 4). The two major products observed were subsequently
identified (vide supra) as 8-hydroxy-1,4-cineole 5a and 3α-1,4-hydroxycineole 5f. However,
mutants of CYP101A have been used to improve the yield and selectivity of monoterpene
oxidation as was demonstrated with pinene and limonene [37, 51]. A library of eighteen
members of CYP101A1 mutant enzymes (Table S1), member of which have been screened
for the oxidation of terpenes and other hydrophobic substrates and have been cloned into a
whole-cell oxidation system along with the physiological electron transfer partners, Pdx and
PdR, was screened for selective catalysts for 1,4-cineole oxidation [37, 51, 52, 55, 101, 102].
Oxidation of 1,4-cineole with select CYP101A1 mutants improved the product yields
and several showed high levels of regioselectivity. The CYP101A1 F87W/Y96F/L244A
(WFA) mutant was greater than 90% selective for oxidation at the 3α position 5f (Fig. 4,
Scheme 2). This product was isolated in pure form and an X-ray structure of a single crystal
not only confirmed its relative configuration but also revealed its absolute configuration as
(1S). The optical rotation of the isolated product indicated that the bulk material was indeed
significantly optically enriched ([α]D23 17°, lit. [100] [α]D 2.5°; see supplementary material
for more details). Several minor products were identified based on their MS fragmentation
pattern, NMR spectra, and GC-MS retention time comparison with authentic standards (vide
supra). These included 2α- 5e, 2β- 5b, 3β- 5c, 8-hydroxy- 5a, 9-hydroxy- 5d and 3-keto-
1,4-cineole 7b (the latter only from large scale whole-cell turnovers) (Fig. 4, Experimental
and Supplementary material).
The F87L/Y96F (LF) variant switched the selectivity of 1,4-cineole oxidation to
favour (<85%) hydroxylation on the isopropyl substituent to give 8-hydroxy-1,4-cineole 5a
(Fig. 4, Scheme 2). Minor products were assigned as 2β- 5b, 2α- 5e, 9-hydroxy- 5d and 2-
keto-1,4-cineole 7a (the latter only from large scale whole-cell turnovers) (Fig. 4,
17

ACCEPTED MANUSCRIPT
Experimental and Supplementary material). In addition, there was also a significant increase
in the minor metabolite at 4.15 min which possesses a MS consistent with a desaturation
product (Fig. 4). Given that the increased level of this metabolite correlated with an increase
in 8-hydroxy-1,4-cineole 5a production across the range of mutants tested, we tentatively
assigned this as the alkene between the C8 and C9 of 1,4-cineole 6 (Scheme 2). A standard
sample of 6 was synthesised from 5a by dehydration with POCl3/pyridine and used to
confirm the structure of this metabolite by comparison of GC-MS retention time and
fragmentation pattern.
3.1.3 CYP102A1 Oxidation: 1,4-Cineole 2
The second enzyme that was examined for specific oxidation of 1,4- and 1,8-cineole was
CYP102A1. CYP102A1 and different variants of CYP102A1 (6 variants in total, details in
Table S1) tested for 1,4- and 1,8-cineole oxidation [83, 94]. The addition of polyfluorinated
carboxylic acid decoy molecules (PFCs) as a method to improve the activity was also
investigated [66, 94, 103-105]. The activity of WT CYP102A1 for both cineoles was low
(Product Formation Rate (PFR); < 0.2 nmol(nmol-CYP)−1min−1; henceforth abbreviated to
min−1). The rate accelerating mutants and the use of the PFC decoy molecules did not
significantly enhance the oxidation of either cineole (data not shown; these combination had
previously been shown to improve the activity of alkylbenzenes and cycloalkanes) [66, 94].
Only variants that contained a mutation at the F87 active site residue showed significant
product formation for either cineole (Fig. 5).
The CYP102A1 R47L/Y51F/A74G/F87V/L188Q (RLYFGVQ) and the
A74G/F87V/L188Q (GVQ) variants improved the oxidation of 1,4-cineole, as compared to
wild type, but two products were formed in roughly equal amounts and arose from
hydroxylation at the 2α 5e and the 3α 5f positions (Fig. 5 and Scheme 2). The CYP102A1
R47L/Y51F/F87A/I401P (RLYFAIP) variant essentially generated 3α-hydroxy-1,4-cineole
18

ACCEPTED MANUSCRIPT
5f as a single product (>90%). The PFR of 529 ± 48 min−1 (Table S2) was significantly
faster than with any of the other CYP102A1 variants that were tested with 1,4-cineole. This
presumably arises as the extra space created by the F87A mutation creates more space for
the cineole to bind closer to the heme. Two minor products were formed in roughly equal
amounts and these arose from hydroxylation at the 3β 5c and 2α 5e positions.
3.2 1,8-Cineole 1
3.2.1 CYP101A1 Oxidation
Previously we have shown that CYP101A1 from a P. putida was able to oxidise 1,8-cineole
1 to generate three major products; (1S)-6α- 3b and 5α- 3c and 5β-hydroxycineole 3d in a
18:68:14 ratio. The latter two metabolites were also enriched in the (1S) enantiomer. A small
amount of the 5-ketocineole 4 product was also generated (Fig. 6). In this study, the WT
CYP101A1 oxidation of 1,8-cineole was found to be consistent with that reported previously
(Fig. 6) [40]. The results obtained from screening the mutant libraries revealed that the
selectivity of 1,8-cineole oxidation was altered in different CYP101A1 variants. For
example the F87V/Y96F/L244A (VFA) variant showed a strong preference for oxidation at
the 6α position (90%) to give (1S)-6α-hydroxycineole 3b in essentially >99% e.e.
Hydroxylation at the 5α and β positions made up the remainder of the observed products
(8% and 2%, respectively, Fig. 6, Scheme 3). The F87W/Y96F/L244A/V247A (WFAL)
variant hydroxylated 1,8-cineole to generate 5α-hydroxy-1,8-cineole 3c as the major product
(88%) with the minor products consisting of 6α-hydroxy-1,8-cineole 3b (10%) and 5β-
hydroxy-1,8-cineole 3d (2%, Fig. 6, Scheme 2). The WFA variant was again more selective
for oxidation at the 5α position (3c, 90%) with only 5β 3d being detected as the minor
product but the overall yield was lower (Fig. 6, Scheme 2). The optical purity of the 5-
hydroxycineoles could not be determined directly by enantioselective GC [9, 40]. However,
oxidation to the 5-ketocineole 4 allowed analysis by enantioselective GC and showed that
19

ACCEPTED MANUSCRIPT
the major compound from the WFAL variant produced was the (1S) enantiomer in 86% e.e.
The other mutants tested either resulted in lower yields or poorer selectivity after GC/GC-
MS analysis (data not shown).
3.2.2 CYP102A1 Oxidation
The catalytic turnover of 1,8-cineole with WT CYP102A1 produced extremely small
quantities of oxidation products. However, the CYP102A1 RLYFGVQ mutant with 1,8-
cineole generated 6α-hydroxy-1,8-cineole 3b as a single product (>98%), as did the GVQ
mutant), with a PFR of 103 ± 25 min−1 (Fig. 5, Scheme 3 and Table S2). Enantioselective
GC analysis indicated that this was again the (1S) isomer in 78% e.e.. The RLYFAIP variant
was less selective for oxidation of 1,8-cineole generating significant quantities of 5α-
hydroxy-1,8-cineole 3c alongside the major product 6α-hydroxy-1,8-cineole 3b (Fig. 5). The
5β-hydroxy metabolite 3d made up the majority of the remainder of the oxidised 1,8-
cineole.
3.3 Comparison of different bacterial P450 systems for cineole oxidation
Bacterial P450 enzyme systems are currently preferably used for biocatalytic reactions due
to their ease of production and high activity levels. The use of whole-cell oxidation systems
is a convenient method to generate larger scale quantities of metabolites. A direct
comparison of the three systems studied here is complex due to differing requirements in
cofactor usage (NADPH vs. NADH), electron transfer partner systems (ferredoxin vs.
flavodoxin, fused single systems vs. three component system) and plasmid systems
(pCWori+ bicistronic and tricistronic vs. pET28 systems). However, it is useful to compare
the relative amount of product obtained for the most selective systems for 1,4- and 1,8-
cineole oxidation. All the samples were tested in BL21(DE3) cells.
All the systems generated similar levels of cell biomass (13-16 grams of cell wet
weight per litre of culture). The amount of P450 detected in the tricistronic CYP101A1
20

ACCEPTED MANUSCRIPT
systems (10 – 97 nM) was generally lower than for the CYP102A1 systems (86 – 107 nM,
Table S3). There was greater variation in the amount of P450 in each of the CYP101A1
systems, with, for example, the LF variant (97 nM) being formed in significantly greater
quantities than either the WFA or VFA mutants (10 - 14 nM, Table S3).
After 16 hours of reaction, the level of 6α-hydroxy-1,8-cineole 3b formed with the
GVQ and RLYFGVQ CYP102A1 variant was 2.25 ± 0.4 mM (4 mM substrate was added
in total to each turnover). The RLYFFAIP variant generated 3α-hydroxy-1,4-cineole 5f at
similarly high levels (2.0 ± 0.1 mM when tested under the same conditions). In line with the
lower levels of P450 production, the amount of hydroxylated metabolites were lower in the
CYP101A1 whole-cell oxidation reactions. The LF variant system generated 470 µM 8-
hydroxy-1,4-cineole 5a (plus additional minor products), while the WFA variant generated
up to 1 mM of 3α-hydroxy-1,4-cineole 5f from 1,4-cineole. The VFA CYP101A1 variant
produced 1 mM of 6α-hydroxy-1,8-cineole 3b metabolite but the levels of the 5α
hydroxylated 1,8-cineole 3c formed by the WFA variant were considerably lower ( 250
µM). For comparison, the amount of (1R)-6β-hydroxy-1,8-cineole 3a generated using the
CYP176A1 whole-cell oxidation system under the same conditions was 650 µM. This
could potentially be due to low levels of the endogenous E. coli flavodoxin reductase which
is used in this system to transfer electrons from NADPH to cindoxin.
Overall the whole-cell turnover of the cineoles with the CYP102A1 variants seems to
be better than with CYP101A1 variants. With 1,8-cineole the WT CYP101A1 system could
generate similar levels of the mixture of the three hydroxyl cineoles formed (> 2.5 mM)
suggesting that with further optimisation either system could be used for larger scale
selective metabolite generation (Fig. 6).
21

ACCEPTED MANUSCRIPT
4. Discussion
The ability to use engineered variants of the metalloproteins CYP101A1 and CYP102A1 to
alter the regioselectivity of 1,8- and 1,4-cineole hydroxylation could lead to new routes for
enantioselectively functionalised cineole analogues. The isolated CYP176A1 enzyme and a
Rhodococcus bacterium have been reported to be capable of the preparative enantioselective
oxidation of 1,8-cineole to (1R)-6β-hydroxy-1,8-cineole 3a [9, 42]. The non-selective
oxidation of 1,4-cineole by this enzyme and the fact that small levels of 1,8-cineole
impurities preferentially bind to enzyme over the 1,4-analogue, highlights that CYP176A1
has evolved for the selective binding and oxidation of 1,8-cineole. This is perhaps
unsurprising as this enzyme is responsible for the first step in the utilisation of 1,8-cineole as
a carbon source. (1R)-6α-Hydroxy-1,8-cineole 3b formation has also been reported from the
oxidation of 1,8-cineole by Bacillus cereus [106]. A number of regioselective
transformations of 1,8-cineole have also been reported but without any enantioselectivity
data. For example, 6α-hydroxy-1,8-cineole 3b is generated from biotransformations using
the fungi Cladosporium cladosporioides [9]. Mixtures of products are reported for other
biological oxidations of 1,8-cineole and 1,4-cineole [9].
Importantly, from screening just a small library of mutants, regioselective (and in
some cases enantioselective) catalysts for the production of compounds arising from
oxidation at the C3 and the C8 positions of 1,4-cineole and the C5 and C6 positions of 1,8-
cineole were obtained in this study. This significantly expands upon the range of metabolites
which can be selectively generated from 1,4- and 1,8-cineole using biotransformations. For
both 1,4-and 1,8-cineole there was a preference for the CYP101A1 and CYP102A1 variants
for oxidation on the α (or endo) face of the substrate whereas CYP176A1 is unusual in that it
shows a preference for the exo (or β) face. The endo face would be considered to be less
sterically hindered in both substrates. In CYP176A1 there is a specific hydrogen bond
22

ACCEPTED MANUSCRIPT
between its asparagine 242 and the ether oxygen of 1,8-cineole which helps orientate the
substrate for hydroxylation at a specific C-H bond (Fig. 2); mutation of this residue results in
a significant increase in oxidation from the endo face [40].
For 1,8-cineole, CYP176A1 specifically generates the (1R)-6-hydroxycineole
enantiomeric series whilst the CYP101A1 and CYP102A1 mutants found here generate the
complementary (1S)-6-hydroxycineole series in good to excellent e.e. Two different
CYP101A1 mutants were able to selectively hydroxylate the C5 position of 1,8-cineole with
the WFA mutant yielding the (1S) compounds in high e.e. A lack of standards prevented
complete determination of the enantiomeric purity of the hydroxycineoles produced from
1,4-cineole oxidation except in the case of 5f formed via CYP101A1 mediated oxidation
which was shown crystallographically to be the (1S) isomer in good e.e.
A significant improvement in yield as well as specificity was achieved for both the
best CYP102A1 mutants. The small-scale CYP101A1 mutant screens produced catalysts
with enhanced selectivity for metabolites including the enantiospecific production of the
6α− and 5α− isomers. The CYP101A1 LF mutant, which produces enhanced levels of the
desaturated alkene product of 1,4-cineole 6, will be a useful starting point for mutations to
specific catalysts of this unusual P450 catalysed transformation.
Using CYP101A1 and CYP102A1 as biocatalysts has several advantages over
isolating new enzymes. Both have been structurally characterised and the electron transfer
proteins are well established. We have also shown that both systems are capable of oxidising
cineole in good yields in non-optimised whole-cell oxidation systems in shake flasks. In
addition, they have been extensively engineered and it has been shown that highly
enantioselective hydroxylations are possible with evolved variants of both. Therefore, the
variants of these enzymes used here will provide a good basis for further studies to enhance
23

ACCEPTED MANUSCRIPT
the enantioselectivity, regioselectivity and activity of P450 catalysed biooxidations of
cineole derivatives with these metalloenzymes.
24

ACCEPTED MANUSCRIPT
1,4-Cineole Nomenclature Footnote
Nomenclature footnote: The nomenclature of the hydroxy 1,4-cineoles is relatively consistent
within the literature but in IUPAC nomenclature the oxygen atom is labelled as position 7.
However, this oxygen is frequently not assigned an atom number in the literature. To be
consistent, we will employ the established literature numbering as shown. To discuss
stereochemistry we have employed the descriptors α and β. The α descriptor is used to define
the substituents that sit below the plane that passes through C2, C3, C5 and C6, while the β
descriptor defines substituents above this plane. 1,4-Cineole is an achiral, meso compound;
however, hydroxylation of either of the methylene bridges will produce enantiomeric compounds
and generate three new stereogenic centres at C1, C4 and the hydroxylated carbon. Hence, a
carbon atom that, following hydroxylation, generates the R-C1 isomer is defined as a pro-R
carbon and a carbon atom that produces the S-C1 isomer is a pro-S carbon.
H
O
9
4
8
10
5
β-substituent
3
1
Pro-R
Pro-S
6
2
7
α-substituent
25

ACCEPTED MANUSCRIPT
Acknowledgements
This work was in part 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 an Australian Government Research Training Program Scholarship (MPhil to
JHZL). We than the Institute of Photonics and Advanced Sensing (IPAS, Adelaide) for
supporting a Student Summer Scholarship (SAB). The authors thank Prof. Luet-Lok Wong
(University of Oxford, UK) for the gene constructs of the CYP101A1 and CYP102A1 variants.
Abbreviations
P450, cytochrome P450; Cpd I, Compound I; CYP101A1, P450cam; CYP102A1, P450BM3
;
CYP176A1, P450cin; Cdx, cindoxin; Pdx/PdR, [2Fe-2S] ferredoxin/flavin dependant ferredoxin
reductase from Pseudomonas putida; VFA, CYP101A1 F87V/Y96F/L244A mutant; LF,
CYP101A1 F87L/Y96F; VFA, CYP101A1 F87V/Y96F/L244A; AFA, CYP101A1
F87A/Y96F/V247A; GVQ, CYP102A1 A74G/F87V/L188Q; RLYFGVQ, CYP102A1
R47L/Y51F/A74G/F87V/L188Q; RLYFAIP, CYP102A1 R47L/Y51F/F87A/I401P; PFR,
Product Formation Rate; FAD, flavin adenine dinucleotide; FMN, flavin mononucleotide; EMM,
E. coli minimal media; PFC, perfluorocarboxylic acid; IPTG, Isopropyl β-D-1-
thiogalactopyranoside; WT, wild-type.
26

ACCEPTED MANUSCRIPT
References
[1] M. Ashour, M. Wink, J. Gershenzon, Annual Plant Reviews Volume 40: Biochemistry of Plant
Secondary Metabolism, Wiley-Blackwell, 2010, pp. 258-303.
[2] J.S. Dickschat, Nat Prod Rep 33 (2016) 87-110.
[3] J. Schrader, in: R.G. Berger (Ed.), Flavours and Fragrances: Chemistry, Bioprocessing and
Sustainability, Springer Berlin Heidelberg, Berlin, Heidelberg, 2007, pp. 507-574.
[4] C.A. Black, M. Parker, T.E. Siebert, D.L. Capone, I.L. Francis, Aust J Grape Wine Res 21 (2015)
582-600.
[5] F. Bakkali, S. Averbeck, D. Averbeck, M. Idaomar, Food Chem Toxicol 46 (2008) 446-475.
[6] M.C. Chang, J.D. Keasling, Nature chemical biology 2 (2006) 674-681.
[7] J. Mi, H. Schewe, M. Buchhaupt, D. Holtmann, J. Schrader, World J Microbiol Biotechnol 32
(2016) 112.
[8] L.C.A. Barbosa, C.A. Filomeno, R.R. Teixeira, Molecules 21 (2016).
[9] R. Azerad, ChemPlusChem 79 (2014) 634-655.
[10] E.S. Kovats, Helv Chim Acta 46 (1963) 2705-&.
[11] R.M. Carman, I.C. Macrae, M.V. Perkins, Aust J Chem 39 (1986) 1739-1746.
[12] Y. Asakawa, M. Toyota, T. Ishida, Xenobiotica; the fate of foreign compounds in biological
systems 18 (1988) 1129-1134.
[13] A.F. Barton, B. Dell, A.R. Knight, J Agric Food Chem 58 (2010) 10147-10155.
[14] J.-A.M. Rasmussen, K.A. Henderson, M.J. Straffon, G.J. Dumsday, J. Coulton, M. Zachariou,
Aus J Chem 58 (2005) 912-916.
[15] I.C. Macrae, V. Alberts, R.M. Carman, I.M. Shaw, Aust J Chem 32 (1979) 917-922.
[16] R.M. Carman, W.T. Robinson, C.J. Wallis, Aust J Chem 58 (2005) 785-791.
[17] W. Liu, J.P. Rosazza, Appl Environ Microbiol 59 (1993) 3889-3893.
[18] J.P. Rosazza, J.J. Steffens, F.S. Sariaslani, A. Goswami, J.M. Beale, S. Reeg, R. Chapman, Appl
Environ Microbiol 53 (1987) 2482-2486.
[19] M. Miyazawa, Y. Hashimoto, J Agric Food Chem 50 (2002) 3522-3526.
[20] M. Miyazawa, M. Shindo, T. Shimada, Drug Metab Dispos 29 (2001) 200-205.
[21] R. Marmulla, J. Harder, Frontiers in microbiology 5 (2014) 346.
[22] J.L. Bicas, A.P. Dionísio, G.M. Pastore, Chem Rev 109 (2009) 4518-4531.
[23] H. Schewe, M.A. Mirata, J. Schrader, Adv Biochem Eng Biotechnol 148 (2015) 251-286.
[24] K.E. Slessor, D.B. Hawkes, A. Farlow, A.G. Pearson, J.E. Stok, J.J. De Voss, J Mol Catal B:
Enzym 79 (2012) 15-20.
[25] M.J. Cryle, J.E. Stok, J.J. De Voss, Aust J Chem 56 (2003) 749-762.
[26] F.P. Guengerich, Chem Res Toxicol 14 (2001) 611-650.
[27] P.R. Ortiz de Montellano (Ed.), Cytochrome P450: Structure, Mechanism, and Biochemistry
Springer International Publishing, Switzerland, 2015.
[28] S.G. Bell, L.L. Wong, in: M. Sono, J.H. Dawson (Eds.), Encyclopedia of Inorganic Chemistry
2005.
[29] F. Hannemann, A. Bichet, K.M. Ewen, R. Bernhardt, Biochimica et biophysica acta 1770
(2007) 330-344.
[30] A.W. Munro, H.M. Girvan, K.J. McLean, Biochimica et biophysica acta 1770 (2007) 345-359.
[31] T.L. Poulos, Chem Rev 114 (2014) 3919-3962.
[32] J. Rittle, M.T. Green, Science 330 (2010) 933-937.
[33] J.T. Groves, G.A. McClusky, J Am Chem Soc 98 (1976) 859-861.
27

ACCEPTED MANUSCRIPT
[34] V.B. Urlacher, S.G. Bell, L.L. Wong, in: R.D. Schmid, V.B. Urlacher (Eds.), Modern
Biooxidation, Wiley, New York, 2007, pp. 99-122.
[35] G.D. Roiban, M.T. Reetz, Chem Commun 51 (2015) 2208-2224.
[36] V.B. Urlacher, M. Girhard, Trends Biotechnol 30 (2012) 26-36.
[37] S.G. Bell, X. Chen, R.J. Sowden, F. Xu, J.N. Williams, L.L. Wong, Z. Rao, J Am Chem Soc 125
(2003) 705-714.
[38] J. Dietrich, Y. Yoshikuni, K. Fisher, F. Woolard, D. Ockey, D. McPhee, N. Renninger, M.
Chang, D. Baker, J.D. Keasling, ACS Chem Biol 4 (2009) 261-267.
[39] R.J. Sowden, S. Yasmin, N.H. Rees, S.G. Bell, L.L. Wong, Org Biomol Chem 3 (2005) 57-64.
[40] J.E. Stok, E.A. Hall, I.S.J. Stone, M.C. Noble, S.H. Wong, S.G. Bell, J.J. De Voss, Journal of
Molecular Catalysis B-Enzymatic 128 (2016) 52-64.
[41] R.J. Branco, A. Seifert, M. Budde, V.B. Urlacher, M.J. Ramos, J. Pleiss, Proteins 73 (2008)
597-607.
[42] D.B. Hawkes, G.W. Adams, A.L. Burlingame, P.R. Ortiz de Montellano, J.J. De Voss, J Biol
Chem 277 (2002) 27725-27732.
[43] J.E. Stok, K.E. Slessor, A.J. Farlow, D.B. Hawkes, J.J. De Voss, Adv Exp Med Biol 851 (2015)
319-339.
[44] D.B. Hawkes, K.E. Slessor, P.V. Bernhardt, J.J. De Voss, Chembiochem 11 (2010) 1107-1114.
[45] K.E. Slessor, A.J. Farlow, S.M. Cavaignac, J.E. Stok, J.J. De Voss, Arch Biochem Biophys 507
(2011) 154-162.
[46] J.E. Stok, S. Yamada, A.J. Farlow, K.E. Slessor, J.J. De Voss, Biochim Biophys Acta Prot
Proteom 1834 (2013) 688-696.
[47] Y.T. Meharenna, H. Li, D.B. Hawkes, A.G. Pearson, J. De Voss, T.L. Poulos, Biochemistry 43
(2004) 9487-9494.
[48] T.L. Poulos, B.C. Finzel, A.J. Howard, J Mol Biol 195 (1987) 687-700.
[49] J.A. Peterson, M.C. Lorence, B. Amarneh, J Biol Chem 265 (1990) 6066-6073.
[50] M.J. Honeychurch, A.O. Hill, L.L. Wong, FEBS letters 451 (1999) 351-353.
[51] S.G. Bell, R.J. Sowden, L.-L. Wong, Chem Commun (2001) 635-636.
[52] D.P. Nickerson, C.F. Harford-Cross, S.R. Fulcher, L.L. Wong, FEBS letters 405 (1997) 153-
156.
[53] S.G. Bell, C.F. Harford-Cross, L.-L. Wong, Protein Eng 14 (2001) 797-802.
[54] S.G. Bell, N. Hoskins, C.J.C. Whitehouse, L.L. Wong, Met Ions Life Sci (2007) 437-476.
[55] F. Xu, S.G. Bell, Z. Rao, L.L. Wong, Protein Eng Des Sel 20 (2007) 473-480.
[56] K. Sakurai, H. Shimada, T. Hayashi, T. Tsukihara, Acta Crystallogr Sect F Struct Biol Cryst
Commun 65 (2009) 80-83.
[57] C.F. Harford-Cross, A.B. Carmichael, F.K. Allan, P.A. England, D.A. Rouch, L.L. Wong, Protein
Eng 13 (2000) 121-128.
[58] J.P. Jones, E.J. O'Hare, L.L. Wong, Eur J Biochem 268 (2001) 1460-1467.
[59] F. Xu, S.G. Bell, J. Lednik, A. Insley, Z. Rao, L.-L. Wong, Angew Chem Int Ed Engl 44 (2005)
4029-4032.
[60] A. Eichler, L. Gricman, S. Herter, P.P. Kelly, N.J. Turner, J. Pleiss, S.L. Flitsch, Chembiochem
17 (2016) 426-432.
[61] O. Sibbesen, Z. Zhang, P.R. Ortiz de Montellano, Arch Biochem Biophys 353 (1998) 285-296.
[62] P.P. Ho, A.J. Fulco, Biochimica et biophysica acta 431 (1976) 249-256.
[63] C.J. Whitehouse, S.G. Bell, L.L. Wong, Chem Soc Rev 41 (2012) 1218-1260.
[64] H. Li, T.L. Poulos, Nat Struct Biol 4 (1997) 140-146.
28

ACCEPTED MANUSCRIPT
[65] K.Y. Choi, E. Jung, D.H. Jung, B.P. Pandey, H. Yun, H.Y. Park, R.J. Kazlauskas, B.G. Kim, FEBS J
279 (2012) 1650-1662.
[66] S.D. Munday, S. Dezvarei, S.G. Bell, ChemCatChem 8 (2016) 2789-2796.
[67] M. Dietrich, S. Eiben, C. Asta, T.A. Do, J. Pleiss, V.B. Urlacher, Appl Microbiol Biotechnol 79
(2008) 931-940.
[68] M. Budde, S.C. Maurer, R.D. Schmid, V.B. Urlacher, Appl Microbiol Biotechnol 66 (2004)
180-186.
[69] M.C. Gustafsson, O. Roitel, K.R. Marshall, M.A. Noble, S.K. Chapman, A. Pessegueiro, A.J.
Fulco, M.R. Cheesman, C. von Wachenfeldt, A.W. Munro, Biochemistry 43 (2004) 5474-5487.
[70] S.D. Munday, N.K. Maddigan, R.J. Young, S.G. Bell, Biochimica et biophysica acta 1860
(2016) 1149-1162.
[71] M.G. Joyce, I.S. Ekanem, O. Roitel, A.J. Dunford, R. Neeli, H.M. Girvan, G.J. Baker, R.A.
Curtis, A.W. Munro, D. Leys, FEBS J 279 (2012) 1694-1706.
[72] T. Kitazume, D.C. Haines, R.W. Estabrook, B. Chen, J.A. Peterson, Biochemistry 46 (2007)
11892-11901.
[73] R. Neeli, H.M. Girvan, A. Lawrence, M.J. Warren, D. Leys, N.S. Scrutton, A.W. Munro, FEBS
Lett. 579 (2005) 5582-5588.
[74] P. Meinhold, M.W. Peters, A. Hartwick, A.R. Hernandez, F.H. Arnold, Adv Synth Catal 348
(2006) 763-772.
[75] W.C. Huang, A.C. Westlake, J.D. Marechal, M.G. Joyce, P.C. Moody, G.C. Roberts, J Mol Biol
373 (2007) 633-651.
[76] S. Kille, F.E. Zilly, J.P. Acevedo, M.T. Reetz, Nat Chem 3 (2011) 738-743.
[77] G. Di Nardo, G. Gilardi, Int J Mol Sci 13 (2012) 15901-15924.
[78] R. Singh, M. Bordeaux, R. Fasan, ACS Catal 4 (2014) 546-552.
[79] R. Singh, J.N. Kolev, P.A. Sutera, R. Fasan, ACS Catal 5 (2015) 1685-1691.
[80] P.S. Coelho, E.M. Brustad, A. Kannan, F.H. Arnold, Science 339 (2013) 307-310.
[81] T.K. Hyster, C.C. Farwell, A.R. Buller, J.A. McIntosh, F.H. Arnold, J Am Chem Soc 136 (2014)
15505-15508.
[82] C.C. Farwell, R.K. Zhang, J.A. McIntosh, T.K. Hyster, F.H. Arnold, ACS Cent Sci 1 (2015) 89-
93.
[83] C.J. Whitehouse, S.G. Bell, H.G. Tufton, R.J. Kenny, L.C. Ogilvie, L.L. Wong, Chem Commun
(2008) 966-968.
[84] C.J. Whitehouse, S.G. Bell, W. Yang, J.A. Yorke, C.F. Blanford, A.J. Strong, E.J. Morse, M.
Bartlam, Z. Rao, L.L. Wong, ChemBioChem 10 (2009) 1654-1656.
[85] C.J. Whitehouse, W. Yang, J.A. Yorke, B.C. Rowlatt, A.J. Strong, C.F. Blanford, S.G. Bell, M.
Bartlam, L.L. Wong, Z. Rao, ChemBioChem 11 (2010) 2549-2556.
[86] C.J. Whitehouse, W. Yang, J.A. Yorke, H.G. Tufton, L.C. Ogilvie, S.G. Bell, W. Zhou, M.
Bartlam, Z. Rao, L.L. Wong, Dalton Trans 40 (2011) 10383-10396.
[87] Q.S. Li, J. Ogawa, R.D. Schmid, S. Shimizu, FEBS letters 508 (2001) 249-252.
[88] Q.S. Li, J. Ogawa, R.D. Schmid, S. Shimizu, Appl Environ Microbiol 67 (2001) 5735-5739.
[89] G.D. Roiban, R. Agudo, A. Ilie, R. Lonsdale, M.T. Reetz, Chem Commun 50 (2014) 14310-
14313.
[90] A. Seifert, S. Vomund, K. Grohmann, S. Kriening, V.B. Urlacher, S. Laschat, J. Pleiss,
ChemBioChem 10 (2009) 853-861.
[91] Q.S. Li, U. Schwaneberg, P. Fischer, R.D. Schmid, Chem Eur J 6 (2000) 1531-1536.
[92] B. Unterweger, D.M. Bulach, J. Scoble, D.J. Midgley, P. Greenfield, D. Lyras, P. Johanesen,
G.J. Dumsday, Appl Environ Microbiol 82 (2016) 6507-6517.
29