In vivo and in vitro hydroxylation of cineole and camphor by cytochromes P450CYP101A1, CYP101B1 and N242A CYP176A1

Article abstract of DOI:10.1016/j.molcatb.2016.03.004

Cytochromes P450 (P450s) are valuable enzymes that can generate a range of useful compounds via biocatalytic oxidations that complement traditional synthetic chemistry. In this study three bacterial P450s, P450_cam (CYP101A1), CYP101B1 and the m

Full text of DOI:10.1016/j.molcatb.2016.03.004

Accepted Manuscript
Title: In vivo and In vitro hydroxylation of Cineole and
Camphor by Cytochromes P450CYP101A1, CYP101B1 and
N242A CYP176A1
Author: Jeanette E. Stok Emma A. Hall Isobella S.J. Stone
Margaret C. Noble Siew Hoon Wong Stephen G. Bell James J.
De Voss
PII:
S1381-1177(16)30035-2
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcatb.2016.03.004
MOLCAB 3338
To appear in:
Journal of Molecular Catalysis B: Enzymatic
Received date:
Revised date:
Accepted date:
9-2-2016
4-3-2016
16-3-2016
Please cite this article as: Jeanette E.Stok, Emma A.Hall, Isobella S.J.Stone, Margaret
C.Noble, Siew Hoon Wong, Stephen G.Bell, James J.De Voss, In vivo and In
vitro hydroxylation of Cineole and Camphor by Cytochromes P450CYP101A1,
CYP101B1 and N242A CYP176A1, Journal of Molecular Catalysis B: Enzymatic
http://dx.doi.org/10.1016/j.molcatb.2016.03.004
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.

In vivo and In vitro hydroxylation of Cineole and Camphor by Cytochromes
P450 CYP101A1, CYP101B1 and N242A CYP176A1
a
b
a
a
Jeanette E. Stok , Emma A. Hall , Isobella S. J. Stone , Margaret C. Noble , Siew
b
a
Hoon Wong, Stephen G. Bell * and James J. De Voss *
aDepartment of Chemistry, School of Chemistry and Molecular Biosciences,
University of Queensland, St. Lucia, Brisbane QLD 4072, Australia
bDepartment of Chemistry, School of Physical Sciences, University of Adelaide,
Adelaide, 5005, Australia
*Corresponding authors
Dr James J. De Voss
Phone: +61 7 3365 3825
FAX:+ 61 7 3365 4273
E-mail: j.devoss@uq.edu.au
Dr Stephen G. Bell
Phone: +61 8 8313 4822
E-mail: stephen.bell@adelaide.edu.au
ꢀ
1

Graphical abstract
ꢀ
2

Highlights
1,8-cineole, (1R) and (1S)-camphor oxidation by CYP101A1 CYP101B1 and
N242A-CYP176A1.
Suite of seven hydroxycamphors and seven hydroxycineoles produced and
characterised.
Whole cell CYP101B1 system can generate between 300-600 mg/L of oxidation
products.
-
-
-
-Metabolites useful for mechanistic studies and structure elucidation.
ꢀ
3

Abstract
Cytochromes P450 (P450s) are valuable enzymes that can generate a range of useful
compounds via biocatalytic oxidations that complement traditional synthetic
chemistry. In this study three bacterial P450s, P450cam (CYP101A1), CYP101B1 and
the mutant N242A-P450cin (N242A-CYP176A1), were used to produce a range of
products from the oxidation of the monoterpenes (1R)- and (1S)-camphor and 1,8-
cineole. We demonstrate that both in vitro and in vivo catalytic turnover with these
P450s can produce a complement of up to seven hydroxycamphors and seven
hydroxycineoles, in addition to compounds produced from further oxidation. The
CYP101B1 whole cell catalytic system was found to produce 300-600 mg/L of
culture of oxidation products that could be easily separated chromatographically. The
CYP101B1 in vitro oxidation of 1,8-cineole primarily produced (1S)-5-
hydroxycineole, which was 78% of the total product formed. However, the amount of
(
1S)-5-hydroxycineole was reduced to 42% of the total products when isolated from
the CYP101B1 whole cell system. (1S)-6-Hydroxycineole (96% ee) could be
isolated from a whole cell catalytic turnover of 1,8-cineole by N242A-P450cin in a
yield of 46 mg/L (98% of the total product). However, the amount of product isolated
(
(1R)-5-endo-hydroxycamphor, 75% of the total products) from the whole cell
catalytic oxidation of (1R)-camphor with N242A-P450cin was much lower (6 mg/L)
due to the inefficient use of reducing equivalents (3.5 ± 0.5%) for substrate oxidation.
These compounds will assist in the identification of specific structures in mechanistic
investigations and structure elucidation, but further optimisation is required to
generate larger quantities for synthetic applications.
Keywords: cytochrome P450; cineole; hydroxycineole; camphor; hydroxycamphor
ꢀ
4

1
. Introduction
1
Cytochromes P450 (P450s) are a family of multipurpose enzymes that are able carry
out numerous oxidative transformations that include, but are not limited to, aliphatic
and aromatic hydroxylations, heteroatom oxidation and carbon-carbon bond cleavage
[1]. Typically, P450s catalyse the oxidation of their substrate by employing an
extremely potent, high-valent iron-oxo species that is able to directly insert an oxygen
atom into an unactivated carbon-hydrogen bond [2, 3]. The iron-oxo species is formed
within the P450 when molecular oxygen is activated following the delivery of two
electrons that have been originally sourced from a nicotinamide cofactor (NADH or
NADPH). In contrast to traditional methodologies of organic synthesis, P450
catalysed oxidation, and carbon hydroxylation in particular, is a process where the
regio- and stereoselectivity is often controlled in a single step [4, 5]. Hence, it is the
ability of P450s to efficiently catalyse oxidation reactions via this exceptionally
effective and selective oxidant, together with their catalytic versatility and substrate
diversity that make P450s particularly attractive for potential biocatalytic applications
[6, 7] This allows P450s to complement traditional chemical synthesis for the
production of a wide array of compounds.
P450cam (CYP101A1) is a bacterial P450 that catalyses the stereospecific
hydroxylation of the monoterpene (1R)-camphor 1 to 5-exo-hydroxycamphor 2a
(Figure 1). Originally described in 1968 [8], P450cam has become one of the most
well-known and extensively studied P450s. Much of the mechanistic understanding of
P450s in general has been established via studies employing P450cam as a model
ꢀ
ꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀꢀ
Abbreviations: P450, cytochrome P450; P450cam, CYP101A1; P450cin or WT-
1
P450cin, wild type CYP176A1; Cdx, cindoxin; Arx/ArR, [2Fe-2S] ferredoxin/flavin
dependant ferredoxin reductase from Novosphingobium aromaticivorans.
ꢀ
5

system [9]. However, these mechanistic investigations have not only allowed a more
detailed understanding of P450 catalysed oxidation but they have also revealed the
biocatalytic potential of P450cam. Structural investigations including both the
crystallographic [10-16] and in silico studies [17-19] have established that a variety of
different molecules can either bind or are predicted to bind within the active site of
P450cam. Further investigations have demonstrated that P450cam is also able to catalyse
the oxidation of some of these molecules to chemically useful compounds, which has
consequently led to development of mutants that are optimised to catalyse the
oxidation of alternative substrates [20-33]. For example, a number of active-site
mutations introduced into P450cam were found to improve its selectivity toward two
monoterpenes: S-limonene and (+)--pinene [22, 24, 34]. This direct manipulation of
P450cam can therefore provide access to a range of compounds for use as standards in
structure elucidation and mechanistic studies and also for the potential production of
compounds for the fragrance, flavour and pharmaceutical industries [6, 35].
Mechanistic investigations of another bacterial P450, P450cin (CYP176A1) have also
highlighted the biocatalytic potential of P450s. P450cin was originally isolated from
Citrobacter braakii and catalyses the enantiospecific hydroxylation of a monoterpene
1
,8-cineole 3 to produce (1R)-6β-hydroxycineole 4a (Figure 2), which is the first step
in the biodegradation of cineole in C. braakii [36-38]. During the initial analysis of
P450cin it was observed that it did not possess the almost universally conserved
threonine found in other P450sꢀ [39]. Instead, in WT-P450cin (Wild type P450cin) this
conserved threonine residue was replaced with an asparagine. To probe this anomaly,
an N242A-P450cin mutant was prepared to examine the role of this asparagine in WT-
P450cin [40, 41]. It was determined that the Asn242 was not a functional replacement
ꢀ
6

for the mechanistically important threonine found in other P450s, but rather it was
important for determining the regio- and enantioselectivity of the oxidation of cineole
[40, 41]. Interestingly, this single amino acid change (N242A) was also shown to
result in formation of only 5% of the product observed with WT-P450cin ((1R)-6β-
hydroxycineole 4a) and instead resulted in conversion of cineole 3 to predominately
(
1S)-6-hydroxycineole 4c (90%; Figure 2) [40]. Further mutagenesis investigations
have demonstrated that a range of hydroxycineole isomer mixtures can also be
obtained from a number of single amino acid mutations [41, 42], again underlining
the potential of P450cin mutants in the generation of chemically useful compounds.
In addition, P450cin has also been observed to catalyse the oxidation of other
monoterpenes, such as (1R)- and (1S)-camphor [43, 44]. Camphor 1 is structurally
similar to P450cin’s natural substrate cineole 3 and was initially investigated to
analyse the interaction between the active site and the oxygen of camphor [43]. An in
vivo system was constructed where Escherichia coli had been transformed with a
bicistronic plasmid that expressed P450cin and cindoxin (Cdx) with endogenous E. coli
flavodoxin thought to be facilitating the terminal electron transfer. Both (1R)- and
(
1S)-camphor 1 oxidation by P450cin was observed. However, surprisingly no
selectivity was observed between (1R)- and (1S)-camphor 1, as both were found to be
converted to a similar mixture of five hydroxycamphor isomers (Figure 1) [43].
A group of P450s from Novosphingobium aromaticivorans were isolated to explore
their potential as biocatalysts and to investigate the mechanism of electron transfer
[45]. It was observed that several of these P450s, which are also from the CYP101
family, were able to catalyse the oxidation of a number of mono- and norisoprenoids.
ꢀ
7

Subsequent research led to the development of in vivo systems for the expression of
each of the P450s in E. coli that included both redox partners required to generate a
catalytically active system, the [2Fe-2S] ferredoxin Arx and the flavin dependant
ferredoxin reductase ArR [46]. One of these P450s, CYP101B1, was able to catalyse
the turnover of (1R)-camphor 1 [45] to four different hydroxycamphors including 5-
exo-hydroxycamphor (2a; 46%), the natural product of P450cam catalysed oxidation of
camphor. The three other metabolites were assigned by coelution experiments with
the Y96F mutant of P450cam which in turn had been reported based on the GC
retention times and relative polarity of the metabolites [47].
The aim of this present study was to continue to explore monoterpene oxidation by
P450cam, WT-P450cin, CYP101B1 and N242A-P450cin. In particular, we wanted to
confirm the identity of the hydroxycamphors produced by the CYP101B1 and
compare them to those previously reported from P450cam and P450cin. Additionally, as
N242A-P450cin produced a different hydroxycineole isomer to that produced by WT-
P450cin during cineole 3 oxidation, it was of interest to determine the selectivity of
this mutant with another monoterpene, camphor 1. Due to the structural similarity
between cineole 3 and camphor 1, a detailed investigation into the products obtained
from cineole 3 oxidation by both P450cam and CYP101B1 was also undertaken to
compare with those seen from N242A-P450cin and WT-P450cin. In vivo systems for
each P450 were also utilised in order to obtain structural information for each isomer
and assess the efficiency of the system. Collectively, it was anticipated that by
assessing the activity of these four P450s with cineole 3 and camphor 1, a collection
of hydroxycamphor 2 and hydrocineole 4 isomers could be produced that may be
useful for structure elucidation purposes and potential biocatalytic applications.
ꢀ
8

2
2
. Experimental
.1 Chemicals
The two entantiomers of both 5-ketocineole (1S)- and (1R)-5 and (1S)-5--
hydroxycineole 4d and its (1R) enantiomer 4e were prepared as racemic mixtures
according to a previously published synthesis [48]. The two enantiomers of 6-
ketocineole (1S) and (1R)-8 and 6--hydroxycineole 4b and 4c were also synthesised
according to previously published methods [49]. Both enantiomers of 5--
hydroxycineole 4f and 4g were produced from the reduction of racemic 5 with lithium
aluminium hydride following the method of Carman et al [48]. A number of
hydroxycamphor isomers had previously been generated from an in vivo WT-P450cin
catalytic system: 3-exo-hydroxycamphor 2d, 3-endo-hydroxycamphor 2c, 6-endo-
hydroxycamphor 2e, 5-exo-hydroxycamphor 2a and 5-endo-hydroxycamphor 2bꢀ[43].
All other chemicals were analytical grade reagents.
2.2 Protein Purification
All proteins were expressed and purified via previously reported methods [39, 46, 50-
–
1
–1
5
3]. CYP101B1 was quantitated using 417 = 113 mM cm [46]. The concentration
-
1
-1
of all other P450s were determined by CO binding (Ɛ448 = 91 000 M cm ) [54].
2.3 Coupling (NAD(P)H consumed: product formed)
For CYP101B1 the product concentration in incubation mixtures was calculated by
calibrating the total ion count response of the GC detector to products when available
(or the substrate if the product was not available). Where more than one product was
produced the detector response was assumed to be equal, e.g. the camphor metabolites
were assumed to have equal detector responses as were those of 1,8-cineole. The
ꢀ
9

coupling efficiency was the percentage of NADH consumed that led to product
formation.
The coupling for P450cam was determined using a previously reported method [40].
An appropriate standard curve was constructed using known concentrations of
internal standard, α-terpineol, and (1R)-6β-hydroxycineole 4a. The catalytic turnover
was performed in 50 mM Tris, 100 mM KCl, pH 7.4: P450cam (1 µM), putidaredoxin
(8 µM), putidaredoxin reductase (2 µM), cineole (5 mM), catalase (1 µM) and NADH
(2.3 mM). The reaction was incubated at room temperature for 30 min and α-terpineol
(6.24 mM, final concentration) was added once the reaction was completed and the
products were extracted into ethyl acetate. The products were then characterised by
GCMS (EC-1 column, Alltech) as previously reported [39, 40] using the following
method: 50 °C for 2 min, 16 °C/min until 250 °C, hold for 20 min. The quantity of
product formed was calculated by comparison to the standard curve. Coupling was
calculated as a percentage of product formed verses NADH consumed. Coupling for
(
1R)- and (1S)-camphor 1 with N242A-P450cin was performed in a similar fashion as
that outlined above for cineole and P450cam using the following components: N242A-
P450cin (1 µM), Cdx (8 µM), E. coli flavodoxin reductase (2 µM), (1R)- or (1S)-
camphor (5 mM), catalase (1 µM) and NADPH (2.3 mM).
2
.4 Dissociation Constant (K )
d
To determine the binding constant the CYP101B1 the enzyme was diluted to 2.0 µM
using 50 mM Tris, pH 7.4, total volume 0.6 mL and used to baseline the
spectrophotometer. Aliquots of substrate (0.5–2 µL) were added using a Hamilton
syringe from a 1, 10 or 100 mM stock solution in ethanol. The solution was mixed
ꢀ
10

and the peak-to-trough difference in absorbance recorded between 700 nm and 250
nm. Further aliquots of substrate were added until the peak-to-trough difference did
not shift further. The apparent binding constants, K
peak-to-trough difference against substrate concentration to a hyperbolic function
Eqn. 1):
d
, were obtained by fitting the
(

Amax  [S]
A 
Eqn. 1
K [S]
d
Where ΔA is the peak-to-trough absorbance difference, ΔAmax is the maximum
absorbance difference and [S] is the substrate concentration.
The dissociation constant (K ) for cineole with P450cam in 50 mM Tris, 100 mM KCl,
d
pH 7.4 was determined through the method previously described for WT-P450cin [39].
Eqn. 1 was also applied to determine the dissociation constant. As cineole is not the
natural substrate for P450cam, the affinity of the substrate was expected to be
sufficiently low so as to satisfy the assumptions of this equation. Determination of the
binding constant for camphor with N242A-P450cin was performed in a similar fashion
to that described above for P450cam using a modified Michaelis-Menten equation [39]
to compensate for the high affinity of camphor for N242A-P450cin
.
2.5 Spin State Change
For substrate binding CYP101B1 was diluted to 2.0 µM using 50 mM Tris, pH 7.4,
total volume 2.5 mL and P450cam and N242A-P450cin were diluted into 50 mM Tris,
1
00 mM KCl, pH 7.4. Either camphor or cineole (2 mM, final concentration) was
added to the P450. The high spin heme content was estimated, to approximately ±5
, by comparison with a set of spectra generated from the sum of the appropriate
percentages of the spectra of the substrate-free form (>95 % low spin, Soret
%
ꢀ
11

maximum at 418 nm) and camphor-bound form (>95 % high spin, Soret maximum at
92 nm) of wild-type P450cam. The spin state change for N242A-P450cin was
3
standardised against WT-P450cin and its substrate cineole for which complete high
spin configuration was assumed.
2.6 NAD(P)H Consumption
For CYP101B1 NADH turnover assays were performed with mixtures (1.2 mL)
containing 50 mM Tris, pH 7.4, CYP101B1 (0.5 μM), Arx (5 μM), ArR (0.5 μM) and
–
1
bovine liver catalase (100 μg mL ). The mixtures were oxygenated and then
equilibrated at 30 °C for 2 min. Substrates were added as 100 mM stock solutions in
ethanol to a final concentration of 1 mM (0.5 mM for -ionone). NADH was added to
~
320 μM, final A340 = 2.00, and the absorbance at 340 nm was monitored. The rate of
–
1
–1
NADH turnover was calculated using ε340 = 6.22 mM cm (Table 1).
For P450cam and N242A-P450cin NAD(P)H consumption was determined as
previously reported [40]. The reaction mixture was prepared in 50 mM Tris, 100 mM
KCl, pH 7.4: P450cam (1 µM), putidaredoxin (8 µM), putidaredoxin reductase (2 µM),
catalase (1 µM) and cineole as the substrate (5 mM). The reaction was initiated by the
addition of NADH (1 mM). The rate of NADH consumption was monitored by the
decrease in absorbance at 340 nm over time. NADH consumption was standardised
against P450cam with camphor as a substrate to calculate NADH consumption as a
percentage. N242A-P450cin NADPH consumption was performed in a similar fashion
to P450cam, using the appropriate redox partners, substrate and electron source (See
Section 2.3). NADPH consumption was standardised again WT-P450cin and cineole
as a percentage.
ꢀ
12

2.7 CYP101B1: Catalytic turnover and product identification
Where authentic product standards were available they were utilised for identification
purposes via GC co-elution experiments. Extraction of the products from activity
assays and product analysis by GC were performed as before [46]. Gas
chromatography (GC) analyses were carried out on a Shimadzu GC-17A instrument
coupled to a QP5050A MS detector using a DB-5 MS fused silica column (30 m x
0.25 mm, 0.25 µm) and helium as the carrier gas. The product concentration in
incubation mixtures was calculated by calibrating the total ion count response of the
GC detector to products when available (or the substrate if the product was not
available). Where more than one product was produced the detector response was
assumed to be equal, e.g. the (1S)-camphor and cineole metabolites were assumed to
have equal detector responses. The coupling efficiency was the percentage of NADH
consumed that led to product formation.
GCMS analyses were carried out with the injector held at 250 C and the interface at
280 C. The oven temperature was varied as follows: 80 C for 3 minutes followed by
1
a gradient of 10 C min for 14 min and held at 220 C for a further 3 minutes.
2.8 P450cam and N242A-P450cin: Catalytic turnover and product identification
The catalytic turnover with P450cam and N242A-P450cin was performed as outlined
above (See Section 2.3) varying the concentration of NADH or NADPH to evaluate
over-oxidation (2-7.5 mM). Products observed as a result of the turnover were
characterised using GCMS (See Section 2.3) and identified by comparison to
synthetic or known standards.
ꢀ
13

2.8.1 Key data for metabolite identification
GCMS data for hydroxycineoles 4 and ketocineoles 5 and 8 were identified by
comparison to literature [40, 55, 56]. See Section 2.10.1 for NMR data of compound
4d.
Compound 4a m/z (EI, 70 eV) 43 (100), 69 (34), 71 (48), 83 (21), 93 (32), 108 (54),
+.
1
11 (24), 126 (36), 170 (M ,6)
Compound 4b/c m/z (EI, 70 eV) 43 (100), 55 (15), 69 (14), 71 (20), 93 (14), 108 (30),
+.
1
11 (13), 126 (20), 170 (M , 5)
Compound 4d/e m/z (EI, 70 eV) 43 (100), 59 (21), 69 (27), 87 (21), 93 (23), 108 (19),
+
.
1
27 (9), 155 (8), 170 (M , 11)
Compound 4f/g m/z (EI, 70 eV) 43 (100), 85 (9), 93 (15), 108 (6), 111 (4), 127 (4),
+.
1
37 (4), 155 (13), 170 (M , 1)
Compound 5 m/z (EI, 70 eV) 43 (100), 55 (30), 82 (22), 83 (20), 111 (8), 125 (9), 153
+.
(
17), 168 (M , 3)
Compound 8 m/z (EI, 70 eV) 43 (100), 55 (16), 67 (17), 69 (9), 82 (38), 111 (4), 140
+.
(
4), 168 (M , 2).
2.9 Characterisation of catalytic turnover products by enantioselective GC
2.9.1 Derivatisation of hydroxycineoles 4
A hydroxycineole 4 (approximately 1 mg) was dissolved in dichloromethane and
trifluoroacetic anhydride added dropwise to the solution and incubated at room
temperature for 30 min. The solvent was removed in vacuo resuspended in ethyl
acetate and characterised by enantioselective GC. Enantioselective GC analysis of the
ꢀ
14

reaction products was performed on Cyclodextrin-B column (25 m x 0.25 mm; 90 °C
for 5 min, 1 °C/min until 200 °C, hold for 10 min, 1.1 mL/min).
2.9.2 Oxidation of hydroxycineoles 4 from P450cam catalysed oxidation of cineole 3
for enantioselective GC analysis
Ethyl acetate was removed from the catalytic product in vacuo, which was then
resuspended in acetone and stirred on ice. Jones’ reagent was added dropwise until an
orange colour persisted. Isopropanol was added until the green chromium salts
precipitated out, leaving a purple solution. The solution was then filtered through
celite and washed with pentane. The solution was dried in vacuo to obtain the
product, which was dissolved in ethyl acetate and characterised by GCMS (Section
2.3) and enantioselective GC (Section 2.9.1).
2.10 CYP101B1: In vivo production and product identification
To isolate and identify products where standards were not available a whole-cell
oxidation system utilizing the plasmids pETDuetArx/ArR and
pRSFDuetArx/CYP101B1 was used to oxidise (1R)-1, (1S)-1 and 3 as described
previously with a few alterations [46]. The plasmids were transformed into competent
−
1
BL21(DE3) cells and grown on LB plates containing 30 μg mL kanamycin and 100
−1
μg mL ampicillin. A single colony was inoculated into 500 mL 2YT broth in a 2 L
flask and grown at 37°C overnight. Protein expression was induced by the addition of
6
0 μM IPTG (from a 0.4 M stock in H
and the shaker speed to 110 rpm. The growths were allowed to continue for another
4 h before the culture was harvested by centrifugation and washed in E. coli minimal
media (EMM) [21]. The cell pellet was resuspended in double the volume of
2
O), and the temperature was reduced to 20°C
2
ꢀ
15

antibiotic-containing media (EMM kan/amp). The substrate (2 mM from a 1 M stock
in ethanol) was added to the resuspended cells (200 mL in a 2 L baffled flask) the
reactions were shaken at 160 rpm and 30°C. Three further aliquots of substrate were
added after 3, 6 and 20 hours. The reaction was allowed to proceed to 24 hours. This
system expressed CYP101B1 at approximately 650 nM.
2
.10.1 Key data for metabolite identification
CYP101B1 (cineole)
The supernatant (200 mL) from a CYP101B1 in vivo turnover of cineole 3 was
extracted in ethylacetate (3 x 100 mL), washed with brine (100 mL) and dried with
magnesium sulphate and the organic extracts were pooled and the solvent was
removed by vacuum distillation and then under a stream of nitrogen. The products
were purified using silica gel chromatography using a hexane/ethyl acetate stepwise
gradient ranging from 80:20 to 50:50 hexane to ethyl acetate using 2.5 % increases
every 100 mL. The composition of the fractions was assessed by TLC and GCMS and
those containing single products (95 %) were combined for characterisation. The
solvent was removed under reduced pressure and yielded 4d (235 mg/L), (1S)-5 (145
mg/L), 6 (120 mg/L) and 7 (65 mg/L) (47%). The purified product was dissolved in
CDCl
3
and the organics characterised by NMR spectroscopy and GCMS. NMR
spectra were acquired on a Varian Unity-plus and spectrometer operating at 600 MHz
1
13
1
13
for H and 150 MHz for C. A combination of H, C, COSY, HSQC and HMBC
experiments was used to determine the structures of the products.
1
(
1S)-5-Ketocineole 5 (92% ee): H NMR (600 MHz, CDCl
3
) δ 2.38 (dd, J = 18.8, 3.1
Hz, 1H, exo H6), 2.29 – 2.20 (m, 2H, endo H6 & exo H7), 2.17 (dd, J = 3.9, 2.0 Hz,
ꢀ
16

1H, H4), 1.91 – 1.83 (m, 1H, exo H8), 1.75 (dddd, J = 14.1, 11.7, 6.3, 2.0 Hz, 1H,
endo H7), 1.66 – 1.59 (m, 1H, endo H8), 1.32 (s, 3H, H10), 1.24 (s, 3H, H11), 1.15 (s,
13
3
H, H9); C NMR (151 MHz, CDCl
3
) δ 213.22 (C5), 73.60 (C1), 73.58 (C3), 51.89
(C4), 49.10 (C6), 30.64 (C10), 30.38 (C8), 26.94 (C9), 26.28 (C11), 18.30 (C7). This
is consistent with the literature [48]. [] -22.8 ± 1.1 (0.22 w/v%; ethanol) See
D
Section 2.8.1 for GCMS data.
1
(
1S)-5-Hydroxycineole 4d: H NMR (600 MHz, CDCl
3
) δ 4.46 (dt, J = 9.3, 3.9 Hz,
1H, exo H5), 2.12 (ddd, J = 14.1, 9.5, 3.3 Hz, 1H, exo H6), 2.07 – 2.03 (m, 1H, exo
H7), 1.82 – 1.74 (m, 1H, endo H7), 1.68 – 1.59 (m, 1H, exo H8), 1.54 (dd, J = 6.1, 3.5
Hz, 1H, H4), 1.51 (ddd, J = 13.3, 12.0, 3.4 Hz, 1H, endo H8), 1.32 (dd, J = 14.2, 2.8
1
3
Hz, 1H, endo H6), 1.30 (s, 3H, H10), 1.22 (s, 3H, H11), 1.07 (s, 3H, H9); C NMR
151 MHz, CDCl ) δ 73.36 (C1), 71.00 (C3), 65.08 (C5), 42.84 (C6), 40.33 (C4),
1.07 (C8), 28.98 (C10), 28.34 (C11), 27.12 (C9), 13.91 (C7). [] -3.9 ± 1.0 (0.18
(
3
3
D
w/v%; ethanol; lit.-18.5[48]) Section 2.8.1 for GCMS data.
1
8
-Hydroxy-5-ketocineole 6: H NMR (600 MHz, CDCl
3
) δ 4.61 – 4.55 (m, 1H, exo
H8), 2.49 (d, J = 4.7 Hz, 1H, H4), 2.38 – 2.28 (m, 3H, exo H7 & exo & endo H6),
1.63 (d, J = 14.4 Hz, 1H, endo H7), 1.28 (s, 3H, H10 or H11), 1.27 (s, 1H, H10 or
H11), 1.15 (s, 3H, H9); 13C NMR (151 MHz, CDCl
) δ 210.48 (C5), 72.58 (C3),
3
72.11 (C1), 63.46 (C8), 58.99 (C4), 47.72 (C6), 40.68 (C7), 29.56 (C9), 26.58 (C10 or
C11), 24.73 (C10 or C11). m/z (EI, 70 eV) 43 (100), 71 (14), 83 (42), 109 (22), 124
+.
(
7), 166 (8), 169 (12), 184 (M , 1).
1
5
,8-Dihydroxycineole 7: H NMR (600 MHz, CDCl
3
) δ 4.55 (dd, J = 9.4, 2.4 Hz,
ꢀ
17

2H, exo H5 & exo H8), 4.00 (Br s, 2H, OH ), 2.27 (ddd, J = 12.9, 9.7, 2.7 Hz, 2H, exo
H6 & exo H7), 1.98 (t, J = 3.4 Hz, 1H, H4), 1.59 (dd, J = 13.7, 2.2 Hz, 2H, endo H6
endo H7), 1.23 (s, 6H, H10 & H11), 1.14 (s, 3H, H9); 13C NMR (151 MHz,
&
CDCl3) δ 73.13 (C3), 71.21 (C1), 67.43 (C5 & 8), 43.84 (C6 & 7), 43.73 (C4), 28.66
(C10 & 11), 26.45 (C9). m/z (EI, 70 eV) 43 (100), 58 (20), 85 (27), 109 (14), 125 (6),
+.
1
27 (11), 153 (4), 168 (2), 186 (M , 0.3).
CYP101B1 (camphor 1)
The catalytic turnover of (1R)-1 with CYP101B1 resulted in a mixture of 2a, 2b, 2e,
2f and 2g in a yield of 325 mg/L (27%; Figure 6). A similar set of products were
observed for the catalytic turnover of (1S)-1: ent-2a, ent-2b, ent-2e, ent-2f and ent-2g.
These products were compared to GCMS standards[43] to identify which
hydroxycamphors were produced. It has previously been established that 6-endo-
hydroxycamphor 2e and 6-exo-hydroxycamphor coelute on the GCMS [57]. To
1
determine which hydroxycamphor was produced H NMR was performed that clearly
showed a peak at 4.15 ppm, consistent with the 6-endo-hydroxycamphor 2e [43] and
lacked the peak at 3.74 ppm [57] that would have been expected for the 6-exo-
hydroxycamphor. 5-exo-Hydroxycamphor 2a and 5-endo-hydroxycamphor 2b were
observed on GCMS when compared to previously isolated standards [43]. 8-
1
Hydroxycamphor 2g was also tentatively identified by H NMR of the crude products
(
See Section 3.1.3) but a corresponding separate peak was not identified by GCMS
58].
[
1
9
3
-Hydroxycamphor 2f: H NMR (600 MHz, CDCl
3
) δ 3.74 (d, J = 10.9 Hz, 1H, H9),
.52 (d, J = 10.9 Hz, 1H, H9), 2.26 – 2.41 (m, 2H, H4 & exo H3), 1.98 (ddd, J = 16.2,
ꢀ
18

8.1, 4.1 Hz, 1H, exo H5), 1.90 (d, J = 17.8 Hz, 1H, endo H3), 1.77 – 1.70 (m, 1H, exo
1
3
H6), 1.51 – 1.39 (m, 2H, endo H6 & endo H5), 0.97 (s, 6H, H8 & H10); C NMR
151 MHz, CDCl ) δ 218.26 (C2), 64.80 (C9), 57.38 (C1), 51.32 (C7), 42.95 (C3),
9.37 (C4), 29.77 (C5), 26.61 (C6), 14.93 (C8), 10.09 (C10). m/z (EI, 70 eV) 55 (14),
(
3
3
6
+.
7 (25), 79 (13), 95 (100), 108 (40), 109 (18), 137 (12), 168 (M , 9) [59, 60].
2.11 Construction of pCW-N242A-P450cin/Cdx
pCW-P450cin/Cdx [43] was digested with PpuMI and NcoI and the 790 bp fragment
obtained was discarded. Concurrently, pCW-N242A-P450cin [40] was also digested
with PpuMI and NcoI and the excised 790 bp fragment was ligated into the similarly
digested linearised pCW-P450cin/Cdx [43] plasmid. The pCW-N242A-P450cin/Cdx
construct was then verified by sequencing (AGRF, Brisbane, Qld).
2
.12 N242A-P450cin: In vivo production and product identification
The plasmid pCW-N242A-P450cin was transformed into E. coli DH5αF’IQ and used
to inoculate Terrific Broth (500 mL in 1.8 L Fernbach flask) containing ampicillin (50
mg/mL). This culture was incubated at 37 C (approx. 180 rpm; Innova 4000, New
Brunswick Scientific) until an OD600 of approximately 0.6 was attained. Protein
expression was then induced with isopropyl β-D-1-thiogalactopyranoside (1 mM) and
the incubation continued at 27.5 C. Cineole 3 (0.25 mL) was added to the culture
(500 mL) approximately every 12 h (approx. 180 rpm; Innova 40, New Brunswick
Scientific). After three days of expression, a total of 1.25 mL of cineole had been
added. The cells were pelleted at 4000 g for 30 min and the supernatant collected.
NaCl was added to the supernatant, extracted with ethyl acetate, filtered through celite
and dried over MgSO
4
. The solvent was removed in vacuo and the crude material
ꢀ
19

purified by flash chromatography (30% ethyl acetate in hexane) to yield (1S)-6-
hydroxycineole 4c (23 mg; 2%) GCMS indicated that it was 98% 4c with the 4a
isomer comprising of the remaining 2%. Spectral data matched the literature [49]. The
concentration of N242A-P450cin in the in vivo system was calculated to be
approximately 100 nM.
A similar procedure was performed to produce isolatable quantities of
hydroxycamphors 2 from in vivo oxidation of (1R)-1 using this N242A-P450cin
system. A total of 1.5 g of camphor (1R)-1 was added (in ethanol) to 500 mL of
culture over the 3 days. This resulted in the production of a mixture of
hydroxcamphors (3 mg; 0.2%) comprising of 75% 5-endo-hydroxycamphor 2b, 9%
5-exo-hydroxycamphor 2a and 16% 6-endo-hydroxycamphor 2e as determined by
GCMS.
3
3
3
. Results
.1 CYP101B1
.1.1 CYP101B1: In vitro characterisation of interaction of CYP101B1 with (1R)- and
(
1S)- 1 and 3
Substrate binding, rate of nicotinamide cofactor oxidation and coupling data were
obtained for CYP101B1 with (1R)- and (1S)-camphor 1 and cineole 3 (Table 1).
CYP101B1 in the presence of 3 was found to undergo a significant spin state change
(75% based upon that observed for P450cam with (1R)-1), which is the observed shift
from lowspin Fe(III):H O (max 417 nm) to highspin Fe(III) (max 392nm) in the
2
presence of substrate and often interpreted as an indicator of productive binding. In
the presence of both (1R)- and (1S)-1 CYP101B1 gave moderate spin state changes:
ꢀ
20

40 and 55%, respectively. Interestingly, when compared to CYP101B1 with either
enantiomer of 1, the catalytic oxidation of 3 with CYP101B1 consumed cofactor at
approximately the same rate but had a significantly faster rate of product formation
-
1
(
355 ± 10 min ) due to more efficient use of the reducing equivalents (Coupling
8%). However, the dissociation constant (K ) for 3 with CYP101B1 (580 ± 15 M)
was approximately 2-fold higher than that of either camphor isomer ((1S)-1 260 ± 8
M; (1R)-1 334 ± 11 M).
3
d

3.1.2 CYP101B1: Identification of products from CYP101B1 oxidation of 3
The catalytic oxidation of 3 by CYP101B1, in both a whole cell and an in vitro
system was performed (Figure 3). Five products were observed by GCMS in the in
vitro oxidation (Figure 3A: A-E) with two additional products isolated from the
whole cell system (Figure 3B: F and Figure 4: 6 and 7). The major product was D (8.1
min), constituting approximately 78% of the total products observed during in vitro
turnover was identified as (1S)-5-hydroxycineole 4d (Figure 4). A product resulting
from multiple oxidation of 3, (1S)-5-ketocineole (1S)-5 (Figure 3: A 7.1 min, 13% of
the total products in vitro) was also identified (Figure 4). Three minor products, B, C
and E (Figure 3), which were found by GCMS analysis to constitute 0.9%, 1.4% and
6.8%, respectively of the total products during in vitro turnover. They were all
+
observed to have a mass (M 170) corresponding to the monohydroxylation of cineole
and were tentatively identified on the basis of the retention time and mass spectral
fragmentation patterns of authentic standards as: 6--hydroxycineole 4b/4c (C), 6--
hydroxycineole 4a/ent-4a (B) and 5--hydroxycineole 4f/4g (E). A whole cell
turnover was carried out to confirm the structures of the metabolites and both 4d
(42% of the isolated products) and 5 (26% of the isolated products) were isolated and
ꢀ
21

characterised by NMR (Figure 4). Two other products resulting from multiple
oxidation of cineole 3 were also isolated and characterised by NMR and GCMS
following silica chromatography (Figure 4). One was 8-hydroxy-5-ketocineole 6
(
Figure 3B: F 10.3 min; 21% of the isolated products) and the other 5,8-
dihydroxycineole 7 (11% of the isolated products), which was not present in the
original GCMS analysis.
3
.1.3 CYP101B1: Identification of products from CYP101B1 oxidation of (1R)- and
1S)-1
The in vitro catalytic turnover of (1R)-1 by CYP101B1 has been previously reported
45], however this system was reconstituted with putidaredoxin reductase (PdR) from
(
[
P450cam and the ferridoxin Arx. The whole cell oxidation employs the ArR/Arx
redox-system isolated from N. aromaticivorans and was used to generate products
from the oxidation of 1 catalysed by CYP101B1 (Figure 5). GCMS analysis indicated
that the in vitro oxidation of (1R)- and (1S)-1 by CYP101B1 both gave the same set of
five different hydroxycamphors 2, albeit with a somewhat different distribution of
1
isomers (Table 2). Comparison to known standards [43] and H NMR were used to
determine that four of the hydroxycamphors from (1R)-1 were 6-endo-
hydroxycamphor 2e (ret. time 9.7 min), 5-exo-hydroxycamphor 2a (10.1 min), 5-
endo-hydroxycamphor 2b (10.5 min) and 9-hydroxycamphor 2f (10.8 min). Another,
very minor product was tentatively identified as 3-hydroxycamphor based upon its
mass spectral fragmentation pattern and retention time in comparison to authentic
standards (2c/2d 8.7 min; Figure 5) [43]. (As the two epimers 2c/2d are inseparable
1
by standard GC [43] we were unable to distinguish between them.) Finally, in the H
NMR of the crude product a small amount (~5%) of 8-hydroxycamphor 2g was
ꢀ
22

tentatively identified by its characteristic resonance at  3.33 (brd s) for the
hydroxymethylene at C8 [58]. A peak was not identified in the GCMS corresponding
to this compound, presumably because of co-elution with another product. Overall
there was a reduction in the formation of 2b during the catalytic turnover of (1S)-1
compared to (1R)-1, which was compensated for by smaller increases in 2c/d, 2e and
2f.
3.2 P450cam
3.2.1 P450cam: Interaction with 3
The binding constant (54.6 ± 5.3 M) of P450cam with 3 indicates that they interact
relatively well (Table 3). The percentage spin state change, which measures the
ability of the substrate to displace the resting state water reveals a significant change
(
68 ± 5%) upon binding cineole 3 but less than P450cam with its natural substrate (1R)-
camphor (1R)-1 (Assumed 100%). The spin state change of P450cam with cineole is
appears to be similar to (1S)-camphor (1S)-1 with P450cam, which has been reported to
be a significant but incomplete shift to high spin [61]. Oxidation of cineole 3 by
P450cam is also relatively well coupled (60 ± 6%) but the rate of the reaction, as
measured by NADH consumption is significantly less (12%) than that of P450cam
catalysed (1R)-camphor 1 oxidation.
3.2.2 P450cam: Identification of products from P450cam catalysed oxidation of 3
GCMS analysis revealed three new peaks following P450cam mediated oxidation of
cineole (Figure 7 and 8). Comparison of retention times and mass spectral
fragmentation with known compounds [40, 55, 56] identified the peaks as
hydroxycineoles 4c (ret. time 7.27 min), 4d/e (7.43 min) and 4f/g (7.45 min; Ratio:
ꢀ
23

4c:4d/e:4f/g, 18:68:14). In the presence of excess NADH a fourth peak was produced
and shown to be the 5-ketocineole 5 (Figure 8; 6.77 min). Enantioselective GC (-
cyclodextrin column; Figure 9) revealed that both enantiomers of 5-ketocineole ((1R)-
5
: 24.5 min; (1S)-5: 25 min) were present, with the major enantiomer (49% ee). (1S)-5
has been reported as being the second eluting isomer on a -cyclodextrin column
48].
[
The hydroxycineoles 4 were derivatised to their corresponding trifluroacetate esters to
determine the enantioselectivity of the reaction. Comparison to similarly derivatised
hydroxycineole enantiomers 4b (ret. time 15.3 min) and 4c (ret. time 16.1 min),
revealed that only 4c is a product of the enzymatic turnover of P450cam with cineole
(
Figure 10). Unfortunately, derivatisation of compounds 4d/4e and 4f/4g failed to
induce separation of stereoisomers via enantioselective GC analysis.
In an effort to determine the enantioselectivity of P450cam mediated production of 4c-
g, the crude hydroxycineole mixture was chemically oxidised to the corresponding
ketones (5 and 8). As expected (1S)-6-hydroxycineole 4c was converted specifically
to (1S)-6-ketocineole (1S)-8 (ret. time 27.2 min) as determined by entantioselective
GC (Figure 11; (1R)-6-ketocineole (1R)-8, ret time 27.9 min). Both enantiomers of 5-
ketocineole 5 were produced from the oxidation of hydroxycineoles 4e/4f and 4d/4g
(Figure 11) with a small enantiomeric excess of ((1S)-5 (29% ee). This level of
entantiomeric excess of (1S)-5 differs to that observed with over-oxidation by P450cam
itself (vida infra; 49% ee). The simplest explanation for these observations is that the
production of 4d-g by P450cam is not a particularly stereospecific process but there is
ꢀ
24

a somewhat higher level of stereoselectivity in the second oxidation which produces
5.
3.3 N242A-P450cin
3.3.1 N242A-P450cin: Interaction with (1R)- and (1S)-1
The interaction of N242A-P450cin with (1R)- and (1S)-1 was investigated (Table 4). It
was found that that the dissociation constant (K ) for both camphor isomers ((1R)-1:
.2 ± 0.1 M and (1S)-1 1.5 ± 0.2 M) was not to dissimilar to that seen for WT-
d
1
P450cin with cineole 3 (0.7 M) [39] and N242A-P450cin with 3 (0.3 M; Table 4).
However, comparison of the dissociation constants of both the camphor isomers with
WT-P450cin and N242A-P450cin indicate a 50-90 fold increase in affinity of camphor
for N242A-P450cin. Interestingly, both the spin state change and rate of NADPH
consumption was much higher in the presence of either (1R)- or (1S)-camphor ((1R)-
or (1S)-1) when compared to cineole 3. However, there was a dramatic decrease in
N242A-P450cin-mediated product formation from camphor (coupling: (1R)-1 3.5 ±
0
.5 and (1S)-1 3.6 ± 0.6%), as compared to cineole (49 ± 3%) as the bulk of the
NADPH-reducing equivalents are directed to the formation of either water or
hydrogen peroxide.
3.3.2 N242A-P450cin: Identification of products from N242A-P450cin oxidation of
(
1R) and (1S)-1
Due to the differential selectivity that had previously been reported for N242A-
P450cin catalytic oxidation of 3 (Figure 2), N242A-P450cin catalysed oxidation of (1R)
and (1S)-1 was examined to see if this change in selectivity was observed for other
terpenes. The same five products that had been generated during the catalytic turnover
ꢀ
25

of (1R)- or (1S)-1 with WT-P450cin were also produced following N242A-P450cin
catalysed oxidation. However, unlike WT-P450cin, N242A-P450cin was found to
produce a completely different distribution of products for each enantiomer (Figure
12 and Table 5). N242A-P450cin catalysed oxidation of (1R)-1 gave predominantly 2b
(73%; Table 5), whereas the major product from (1S)-1 turnover was the
hydroxycamphor ent-2e (50%).
An in vivo system was also used to determine if N242A-P450cin catalytic oxidation of
(
1R)-1 could produce useful quantities of 5-endo-hydroxycamphor 2b. Following
column chromatography, this system was found to produce a mixture of three
hydroxcamphors (2a, 2b and 2e) at approximately 6 mg/L, in which 2b was the major
component (75%).
3.3.3 N242A-P450cin: Production of 4c from in vivo N242A-P450cin catalysed
oxidation of 3
A large scale in vivo oxidation was performed to determine if N242A-P450cin
oxidation of 3 could generate a useful quantity of 4c. Isolation of the products,
followed by column chromatography and NMR indicated that the hydroxycineole 4c
(
96% ee) was produced at approximately 46 mg/L of original culture. (A small
amount of 4a was still present following column chromatography.)
4
. Discussion
Although CYP101B1 shares a 44% sequence identity with P450cam [53] it was
previously observed to generate a collection of four hydroxycamphors 2 following
camphor (1R)-1 oxidation rather than just the 5-exo-hydroxycamphor 2a that is seen
ꢀ
26

for P450cam [45]. Using a new whole cell oxidation system that contained the
complementary redox partners, Arx and ArR, a careful analysis of this reaction was
undertaken to generate sufficient quantities of the products in order to confirm the
identity of these four products. Following isolation they were found to be 6-endo-
hydroxycamphor 2e, 5-exo-hydroxycamphor 2a, 5-endo-hydroxycamphor 2b and 9-
hydroxycamphor 2f. A fifth product in very small quantities was also observed and is
thought to be 3-hydroxycamphor 2c/d based on GC retention time and its MS
1
fragmentation pattern. H NMR indicated that what was originally reported as the 6-
exo-hydroxycamphor based on its GCMS retention time [45, 47] was in fact 6-endo-
hydroxycamphor 2e. These two isomers are known to coelute during GC analysis [57]
but were easily distinguished by the characteristic shift of the CHOH proton (4.15
ppm in 2e and 3.74 ppm in 6-exo-hydroxycamphor) [57].ꢀ Interestingly, the major
product found from CYP101B1-catalysed oxidation of (1R) and (1S)-1 by is 2a or
ent-2a (47%; Table 2). This is also the only product from P450cam catalysed oxidation
of (1R)-1 [8] and the major product (≥ 98%) from the equivalent turnovers by
CYP101D1 and CYP101D2 [45]. In P450cam it was similarly observed that
substituting (1R)-1 with its enantiomer (1S)-1 results in formation of the enantiomeric
5-exo-hydroxycamphor ent-2a [16, 61]. Overall, CYP101B1 oxidation of both
camphor enantiomers appears to be mainly directed towards the C5 and C6 positions
of 1, but somewhat unusually a significant amount of oxidation at the C9 methyl and
little at the C3 methylene is observed (Figure 6). P450cam, CYP101D1 and
CYP101D2 contain a residue equivalent to Tyr96 in their active site. This tyrosine is
known to hydrogen bond to the camphor carbonyl group and helps orientate the
substrate in the active site for stereoselective oxidation. This residue is replaced with
a histidine in CYP101B1. In CYP101C1 from N. aromaticivorans, which only
ꢀ
27

oxidises camphor at low activity and selectivity this residue is Met82. The loss of this
hydrophilic substrate-enzyme interaction reduces the selectivity of the C-H bond
oxidation [62-65]. The active sites of P450cam, CYP101B1 and CYP101D1 must
position the molecule in order to facilitate oxidation at C5 and 6. However, this
preference for C5 and C6 may also be due to both the inherent electronic and steric
reactivity of camphor at these positions. It has been suggested on the basis of
theoretical calculations of the heats of radical formation, that C3, C5, C6 and C9 are
thermodynamically similar [66]. However, it has been observed experimentally that in
general methyl groups are much less likely to be oxidised, presumably due to their
higher C-H bond strength compared to methylenes and methines [67]. We observed
that WT-P450cin-mediated oxidation of camphor appeared to proceed with little
enzyme determined regio- or stereoselectivity and gave predominantly hydroxylation
at C3 (~50%), C5 and C6 (Figure 1) [43]. This suggests that the CYP101 P450s may
have evolved to enhance the natural preference for oxidation at the C5 and C6
positions. In addition, it was observed that CYP101B1, WT-P450cin and N242A-
P450cin all produced less ent-2b from (1S)-1 when compared to 2b produced from
(
1R)-1 from the same enzymes (vide infra; [43]).ꢀ
In comparison to WT-P450cin, N242A-P450cin possesses dramatically increased the
affinity for both isomers of 1 (50-90 fold; Table 4) but significantly reduced the
coupling of the reaction with only 3.5% of the reducing equivalents being used to
produce 2 compared to 25-35% for WT-P450cin [43]. Interestingly, N242A-P450cin
was found to reduce the preference for oxidation at C3 of 1 in both the enantiomers in
comparison with WT-P450cin and instead increased oxidation at C5 for (1R)-1 and C6
for (1S)-1 (Table 5). Whilst this change in selectivity is synthetically interesting
ꢀ
28

(
(1R)-1 oxidation yields 73% 2b), it reflects only a small change in the relative
energies of oxidation at these positions and its origins are therefore difficult to
determine. Interestingly, when compared to both WT-P450cin and the CYP101 P450s,
N242A-P450cin showed the biggest differentiation between the two enantiomers of
camphor. Where the other P450s had produced either the same product or a set of the
same products with a similar pattern of distribution, (1R)- and (1S)-1 turnover by
N242A-P450cin gave the same set of products in very different ratios. This shift in
selectivity in N242A-P450cin has previously been observed following oxidation of 3
where WT-P450cin produces 4a whereas N242A-P450cin gives predominantly 4c
(90%; Figure 2) [40, 41]. Taken together, these results indicate that while mutagenesis
of the active site N242 in P450cin does not effect the mechanism of the enzyme it will
have a significant effect on the regioselectivity of this P450 and the selectivity of
further mutants at this position would be of interest.
Cineole 3 oxidation was also investigated with CYP101B1 and P450cam. CYP101B1-
catalysed in vitro oxidation of 3 produces mainly (1S)-5--hydroxycineole 4d (78%,
Figure 3) as well a number of other products that were the result of multiple oxidation
events at C5 (total selectivity for C5 > 90%) and very small amounts of product from
oxidation at C6. The hydroxycineoles 4 were formed by the P450cam-catalysed
oxidation of 3 and also indicates a preference for reaction at C5 (Figure 7) with 5-
hydroxycineoles accounting for 82% of total the product. The only product that is the
result of P450cam mediated oxidation at the C6 of 3 is 4c, and interestingly, this was
not further oxidised to its corresponding ketone (1S)-8 during the turnover with
excess NADH when (1S)-5 was formed in significant amounts. The predominance of
C5 oxidation by these P450s, and in particular the -orientation (endo), may reflect
ꢀ
29