Exploring the substrate specificity of Cytochrome P450cin

Article abstract of DOI:10.1016/j.abb.2019.07.025

Cytochromes P450 are enzymes that catalyse the oxidation of a wide variety of compounds that range from small volatile compounds, such as monoterpenes to larger compounds like steroids. These enzymes can be modified to selectively oxidise substrates of interest, thereby making them attractive for applications in the biotechnology industry. In this study, we screened a small library of terpenes and terpenoid compounds against P450_cin and two P450_cin mutants, N242A and N242T, that have previously been shown to affect selectivity. Initial screening indicated that P450_cin could catalyse the oxidation of most of the monoterpenes tested; however, sesquiterpenes were not substrates for this enzyme or the N242A mutant. Additionally, both P450_cin mutants were found to be able to oxidise other bicyclic monoterpenes. For example, the oxidation of (R)- and (S)-camphor by N242T favoured the production of 5-endo-hydroxycamphor (65–77% of the total products, dependent on the enantiomer), which was similar to that previously observed for (R)-camphor with N242A (73%). Selectivity was also observed for both (R)- and (S)-limonene where N242A predominantly produced the cis-limonene 1,2-epoxide (80% of the products following (R)-limonene oxidation) as compared to P450_cin (23% of the total products with (R)-limonene). Of the three enzymes screened, only P450_cin was observed to catalyse the oxidation of the aromatic terpene p-cymene. All six possible hydroxylation products were generated from an in vivo expression system catalysing the oxidation of p-cymene and were assigned based on ¹H NMR and GC-MS fragmentation patterns. Overall, these results have provided the foundation for pursuing new P450_cin mutants that can selectively oxidise various monoterpenes for biocatalytic applications.

Full text of DOI:10.1016/j.abb.2019.07.025

ArchivesofBiochemistryandBiophysics672(2019)108060
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics
journal homepage: www.elsevier.com/locate/yabbi
Exploring the substrate specificity of Cytochrome P450cin
Jeanette E. Stok, Peter D. Giang, Siew Hoon Wong, James J. De Voss*
School of Chemistry and Molecular Biosciences, The University of Queensland, Australia
A R T I C L E I N F O
A B S T R A C T
Dedication: This paper is dedicated to Paul
Ortiz de Montellano: inspirational, insightful
and interested – what more could one ask?
Cytochromes P450 are enzymes that catalyse the oxidation of a wide variety of compounds that range from small
volatile compounds, such as monoterpenes to larger compounds like steroids. These enzymes can be modified to
selectively oxidise substrates of interest, thereby making them attractive for applications in the biotechnology
industry. In this study, we screened a small library of terpenes and terpenoid compounds against P450cin and two
P450cin mutants, N242A and N242T, that have previously been shown to affect selectivity. Initial screening
indicated that P450cin could catalyse the oxidation of most of the monoterpenes tested; however, sesquiterpenes
were not substrates for this enzyme or the N242A mutant. Additionally, both P450cin mutants were found to be
able to oxidise other bicyclic monoterpenes. For example, the oxidation of (R)- and (S)-camphor by N242T
favoured the production of 5-endo-hydroxycamphor (65–77% of the total products, dependent on the en-
antiomer), which was similar to that previously observed for (R)-camphor with N242A (73%). Selectivity was
also observed for both (R)- and (S)-limonene where N242A predominantly produced the cis-limonene 1,2-ep-
oxide (80% of the products following (R)-limonene oxidation) as compared to P450cin (23% of the total products
with (R)-limonene). Of the three enzymes screened, only P450cin was observed to catalyse the oxidation of the
aromatic terpene p-cymene. All six possible hydroxylation products were generated from an in vivo expression
system catalysing the oxidation of p-cymene and were assigned based on 1H NMR and GC-MS fragmentation
patterns. Overall, these results have provided the foundation for pursuing new P450cin mutants that can se-
lectively oxidise various monoterpenes for biocatalytic applications.
Keywords:
Cytochrome P450cin
Monoterpenes
Oxidation
Substrate specificity
Mutant
Binding
1. Introduction
materials because they are easily attainable, inexpensive and derived
from a renewable resource [3]. A large number of bacterial P450s have
Cytochromes P4501 (P450s) are family of versatile enzymes that
catalyse a wide range of oxidative transformations. Typically, P450s
employ a high-valent iron-oxo species to insert an oxygen atom into an
unactivated carbon hydrogen bond to create an alcohol. Significantly,
this is a one-step oxidation process in which the stereo- regio- and
enantiospecificity is often highly controlled and is difficult to replicate
with traditional organic synthetic methodologies [1]. Hence, P450s can
provide a suitable adjunct to chemical synthesis with a wide array of
potential biotechnological applications. One of these applications is the
production of terpene derivatives, which are essential components in
the manufacture of both flavour and fragrance compounds and also find
use in the pharmaceutical industry [2,3].
been found that naturally catalyse the oxidation of various mono-
terpenes [3]. One example is P450cam (CYP101A1), which catalyses the
oxidation of camphor 1 to 5-exo-hydroxycamphor 2a (Fig. 1; vide infra)
[5]. Such terpene oxidising P450s are potentially valuable tools that
can be employed biotechnologically to produce a variety of modified
terpene intermediates [1]. However, to increase the potential utility of
bacterial P450s for biocatalytic applications these proteins must be
engineered to increase their substrate range while maintaining catalytic
efficiency, and stereo- regio- and enantioselectivities. It has been es-
tablished that enzymes that can only catalyse a transformation at very
low levels can be a useful starting point for developing efficient cata-
lysts [6]. Therefore, protein engineering techniques such as site-di-
rected mutagenesis and directed evolution are often used to create P450
mutants that can expand the inherent substrate selectivity of these
enzymes and catalyse alternative oxidations [1,7].
Monoterpenes such as camphor, cineole, limonene and pinene, are
the main constituents of many plant oils [4] and are formed from the
linking of two isoprene units to create a C10 molecule. As monoterpenes
are found in nature they are particularly attractive chemical starting
Two bacterial P450s most commonly modified to enhance their
Abbreviations: P450, Cytochrome P450; P450cam, CYP101A1; P450BM3, CYP102A1; Pdx, putidaredoxin; PdR, putidaredoxin reductase; P450cin, CYP176A1; Cdx,
cindoxin; FdR, Escherichia coli flavodoxin reductase
* Corresponding author. The University of Queensland, Brisbane, 4072, Australia.
E-mail address: j.devoss@uq.edu.au (J.J. De Voss).
https://doi.org/10.1016/j.abb.2019.07.025
Received 13 November 2018; Received in revised form 25 July 2019; Accepted 25 July 2019
Availableonline27July2019
0003-9861/©2019ElsevierInc.Allrightsreserved.

J.E. Stok, et al.
ArchivesofBiochemistryandBiophysics672(2019)108060
catalyse the selective oxidation of cineole 3 to (1S)-6α-hydroxycineole
4b (90% of total products formed) rather than (1R)-6β-hydroxycineole
4a, the isomer that had originally been observed with the wildtype
enzyme (Fig. 2) [23]. When the asparagine was replaced with a
threonine (N242T) the product profile was altered further resulting in
three hydroxycineole isomers: 4a (47%), 4b (22%) and 4c (31%) [24].
The aim of this study was to examine a range of compounds, pri-
marily terpenes, to determine their potential as substrates for wildtype
P450cin, in addition to the N242A and N242T mutants. Since previous
reports had indicated that N242A could alter the product profile as
compared to the wildtype P450cin, we were interested in determining
whether this extended to other non-natural P450cin substrates, espe-
cially other terpenes. P450cin and N242A have previously been shown
to oxidise camphor 1, which is a bicyclic monoterpene similar to ci-
neole 3 [25,26]. The N242A mutant catalysed oxidation of camphor
resulted in comparatively different product profiles for each isomer of
camphor [26], whereas P450cin gave a similar profile for both (1R) and
(1S)-camphor 1 [25]. Therefore, we initially screened a library of ter-
pene and terpenoid compounds (1, 5–26; Fig. 3) to determine whether
they could be catalytically oxidised by P450cin, N242A and/or N242T.
Once we had determined if these compounds were substrates for these
enzymes, a number of catalytic parameters were determined. It is an-
ticipated that this study may highlight pathways for further diversifi-
cation of the substrate range of P450cin and its mutants in the oxidation
of terpenes.
Fig. 1. The catalytic oxidation of camphor 1 to 5-exo-hydroxycamphor 2a by
P450cam (CYP101A1) [5].
biocatalytic properties and catalyse the oxidation of a broad range of
substrates, including terpenes, are P450cam and P450BM3 (CYP102A1)
[3,7]. P450cam from Pseudomonas putida is responsible for the oxidation
of the monoterpene (1R)-camphor 1 and allows P. putida to utilize
camphor as its sole carbon and energy source (Fig. 1) [5]. A number of
studies have employed site-directed mutagenesis to modify P450cam
permitting it to catalyse the oxidation of monoterpenes other than
camphor [8–12]. Additionally, P450cam requires two auxiliary proteins,
putidaredoxin (Pdx) and putidaredoxin reductase (PdR), and an elec-
tron source (NADH) in order to facilitate oxidation. To improve the
biocatalytic potential of P450cam whole-cell in vivo systems, which in-
clude its redox partners Pdx and PdR, can be used to simplify the
process and make it more appealing for biotechnological applications
[10,13].
In contrast to P450cam, P450BM3 does not naturally catalyse the
oxidation of terpenes but rather catalyses the oxidation of long-chain
fatty acids at the ω-1, ω-2 and ω-3 positions [14]. However, unlike
P450cam, P450BM3 does not require auxillary proteins to deliver elec-
trons as it is a natural fusion of the P450 and its redox partner(s) in one
polypeptide. Thus, the fact that P450BM3 can perform oxidations
without any additional proteins, together with its substrate pro-
miscuity, makes it is an attractive P450 to engineer to accept a range of
alternative substrates [15]. A number of mutagenesis studies have
specifically explored terpene oxidation catalysed by P450BM3 and have
included monoterpene substrates such as limonene and pinene [16–20].
P450cin (CYP176A1) is another bacterial P450 with potential bio-
catalytic capabilities for the oxidation of terpenes. P450cin (CYP176A1)
was originally isolated from an organism that can live on cineole as its
sole carbon source and catalyses the enantiospecific hydroxylation of a
monoterpene 1,8-cineole 3 to produce (1R)-6β-hydroxycineole 4a
(Fig. 2). This is the first step in the biodegradation of cineole [21].
P450cin was found to have an asparagine residue (N242) in the position
where most other P450s have a mechanistically important active site
threonine [21]. Typically, P450s have been shown to use this conserved
threonine to ensure the protonation of the distal oxygen of a hydro-
peroxy intermediate as protonation of the proximal oxygen leads to
hydrogen peroxide formation (uncoupling the electron consumption
without product formation) [22]. However, studies indicated that the
asparagine in P450cin was not a functional replacement for threonine,
but instead was observed to control the stereo- and regioselectivity of
the oxidation [23,24]. Additionally, the N242A mutant was found to
2. Materials and methods
2.1. Chemicals
Valencene 16, guaiazulene 17, α-bisabolol 19, (1R)-nopol 21, cis-
jasmone 22, isophorone 23, β-ionone 25 and α-ionone 26 were gifts
from Dr. Steven G. Bell. Camphane 24 has been synthesised and re-
ported previously [27].
A mixture of cis and trans (R)-limonene 1,2-epoxide 31 was pre-
pared according to previously published methods [28]. Briefly, (R)-li-
monene 7 was stirred with meta-chloroperbenzoic acid with sodium
bicarbonate in dichloromethane. GC-MS data obtained for the (R)-li-
monene 1,2-epoxide isomers was consistent with literature values [29].
All other chemicals used in this study were analytical grade re-
agents.
2.2. Purification of enzymes: P450cin and mutants
P450cin, mutants N242A and N242T, and the redox partners were all
expressed and purified using previously published methods [21,30,31].
2.3. Catalytic turnover
The catalytic turnover of a number of substrates with P450cin and
the N242A and N242T mutants were performed employing the fol-
lowing method. A solution of Buffer A (50 mM Tris, pH 7.4, 100 mM
KCl; 1 mL) containing the appropriate enzyme (0.5 μM), cindoxin (Cdx)
(4 μM), Escherichia coli flavodoxin reductase (FdR) (1 μM), catalase
(1 μM), NADPH (1–2 mM) and relevant substrate (5 mM) were in-
cubated at room temperature (approximately 25 °C). The reaction time
was altered depending on the enzyme used (P450cin, 30 min; N242A or
N242T, 1 h). The reaction mixture was extracted using ethyl acetate
and dried over MgSO4. The products were characterised by GC-MS
(Econo-cap capillary column EC-1) as previously described [21,23]
using the following conditions: 50 °C for 2 min, 16 °C/min until 250 °C,
hold for 20 min. Products observed during the turnovers were identified
by comparison to the MS library or authentic standards. Product ratios
of structurally similar compounds were determined via integration of
the GC-MS Total ion current trace.
Fig. 2. The catalytic oxidation of cineole 3 by P450cin (CYP176A1) and its
mutants N242A and N242T. Numbers indicate the percentage of the hydro-
xycineole 4 formed from the reaction [23,24].
2

J.E. Stok, et al.
ArchivesofBiochemistryandBiophysics672(2019)108060
Fig. 3. Structures of all compounds used in this study. Letters (A-E) indicate each group type: A. Bicyclic monoterpenes (1,3,5,6); B. Monocyclic monoterpenes (7-
12); C. Acyclic monoterpenes (13–15); D. Sesquiterpenes (16–19); and E. Compounds C9–C13 (20–26).
2.4. Spin state change
equation was employed if the affinity of the substrate for the enzyme
satisfied the assumptions the equation. If the compound displayed high
affinity for the enzyme the dissociation constant was calculated using a
tight binding quadratic equation [21].
The percentage spin state change was calculated for P450cin and the
N242A and N242T mutants with a number of compounds (1, 5–8,
10–14, 19, 21–25) [25]. Briefly, the compound (5 mM, final con-
centration) was added to a solution of the P450 (2 μM in Buffer A) and
the difference in absorbance (A392 –A417) measured. The percentage
spin state change was calculated from the difference observed and
standardised against P450cin with cineole 3.
2.6. Rate of NADPH consumption
The rate of NADPH consumption was measured using the following
procedure [23]. Enzyme (0.5 μM), Cdx (4 μM), E. coli flavodoxin re-
ductase (1 μM), catalase (1 μM) and substrate (5 mM) were all com-
bined in Buffer A. The reaction was initiated with NADPH (200 μM) and
the rate of NADPH consumption recorded over time via the decrease in
absorbance at 340 nm. The background was determined by measuring
the rate of the reaction in the absence of the P450. NADPH consump-
tion was standardised against P450cin with cineole and reported as a
percentage.
2.5. Dissociation constant
The dissociation constants (Kd) for P450cin, N242A and N242T with
a variety of compounds (5–7 and 12) were determined using a pre-
viously reported method [21]. Briefly, the appropriate compound (final
concentration dependent on the compound and enzyme) was added to a
solution of P450 (2 μM) in Buffer A and the absorbance difference
measured (A392-A417). The A392-A417 difference was then plotted
against the concentration of the compound. A typical hyperbolic
3

J.E. Stok, et al.
ArchivesofBiochemistryandBiophysics672(2019)108060
Table 1
Catalytic oxidation and percentage high spin (HS) data (P450cin low spin Fe(III)·H2O to high spin Fe(III)) of a small library of terpene/terpenoid compounds with
P450cin, N242A and N242T. Letters (A-E) indicate each group type: A. Bicyclic monoterpenes; B. Monocyclic monoterpenes; C. Acyclic monoterpenes; D.
Sesquiterpenes; and E. Compounds C9–C13.P = Products detected, ND = Not Detected.
P450cin
N242A
N242T
Products
HS
Products
HS
Products
HS
A
B
3
P [25]
P [25]
P [25]
P
P
P
P
P [38]
ND
P
P
P
100 [21]
P [26]
P [26]
P [26]
P
P
P
30 [23]
P
P
P
ND
P
P
P
ND
11 [24]
(1R)-1
(1S)-1
5
27
32
70
56
32
60
11
2 [25]
1 [25]
6
1
3
3
84
89
79
71
64
52
30
1 [26]
1 [26]
8
5
1
3
1
1
1
4
1
36
16
17
18
13
5
3
1
2
1
6
(R)-7
(S)-7
8
P
P
1 [38]
9
ND
P
ND
ND
P
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
5
1
33
8
4
1
P
8
6
1
1
11
21
6
1
1
1
1
23
10
16
1
1
1
ND
2
8
C
P
P
1
1
6
P
ND
ND
ND
ND
ND
ND
ND
P
ND
ND
ND
ND
ND
ND
ND
P
P
P
ND
ND
D
4
2
9
1
1
ND
ND
E
11
5
7
1
1
1
2
5
1
P
P
34
32
110
29
2
1
3
2
10
10
ND
36
1
1
P [24]
ND
P
59 [24]
25
1
1
2.7. Coupling (NADPH consumed: product formed)
derivatisation was used to differentiate between the multiple products
by GC-MS (See Section 2.3 for details).
The percentage of coupling (NADPH consumed: product formed)
was determined employing a previously published method [23]. A
catalytic turnover reaction (see Section 2.3) was performed with the
following modifications. An internal standard, α-pulegone (1 mM) was
added once the reaction was completed before extraction and GC-MS
analysis. Catalytic oxidation products of 5, 6, and (R) and (S)-7, were
compared to a standard curve (mixture of cis- and trans-limonene oxide
31) and the amount of product was determined. Coupling was then
calculated as a percentage of product formed vs NADPH consumed.
2.8.1. Key data for metabolite identification
1H NMR: p-cymene oxidation products.
4-Hydroxycymene 36: 1H NMR (500 MHz, CDCl3) δ 7.03 (d, 1H),
6.72 (dd, 1H), 6.65 (d, 1H), 2.81 (m, 1H), 1.21 (d, 6H). This is con-
sistent with the literature [32].
1-Hydroxycymene 37: 1H NMR (500 MHz, CDCl3) δ 7.29 (d, 2H),
4.66 (s, 2H), 2.90 (m, 1H), 1.25 (d, 6H). This is consistent with the
literature [33].
8-Hydroxcymene 40: 1H NMR (500 MHz, CDCl3) δ 7.38, 7.16 (2 x
m, 2 × 2H), 2.30 (s, 3H), 1.60 (s, 6H). This is consistent with the lit-
erature [34].
2.8. P450cin: catalysed oxidation of p-cymene 12 in vivo and product
identification
MS fragmentation patterns of TFA-derivatised p-cymene oxidation
products
A previously constructed bicistronic plasmid (pCW.P450cin.Cdx)
[25] was transformed into E. coli DH5αF’IQ and used to inoculate
Terrific Broth (500 mL in a 1.8 L Fernbach flask) containing ampicillin
(50 μg/mL). This culture was incubated at 37 °C (approx. 180 rpm; In-
nova 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 for 17 h. The culture was then pelleted via centrifugation
(5000 g; 20 min) to remove the supernatant and the cells were re-
suspended in 4 × 250 mL minimal medium (per L: 6.4 g
Na2HPO4·7H2O, 1.5 g KH2PO4, 0.5 g NaCl, 1 g NH4Cl, 0.24 g MgSO4,
0.01 g CaCl2 and 0.4% glucose). p-Cymene 12 (20 μL) was added to
each flask at the following intervals: 30 min; 60 min; 2 h; and 3 h (Total
amount of 12: 275 mg). After the second hour, glucose (final con-
centration 0.2%) was added and the cultures incubated for a further
17 h. The cells were pelleted at 5000 g for 20 min and the supernatant
collected. The supernatant was extracted with ethyl acetate, washed
with saturated NaCl and dried over MgSO4. The solvent was removed in
vacuo giving a yield of 90 mg of multiple oxidation products (33%),
which were characterised by NMR. Trifluoroacetic anhydride
10-Hydroxycymene 35 m/z (EI, 70eV) 91 (42), 115 (52), 117 (46),
133 (18), 149 (24), 185 (10), 213 (51), 231 (100), 246 (M+; 52).
4-Hydroxycymene 36: m/z (EI, 70eV) 91 (32), 115 (37), 117 (41),
133 (15), 105 (10), 185 (4), 213 (11), 231 (100), 246 (M+; 39).
1-Hydroxycymene 37: m/z (EI, 70eV) 91 (22), 115 (29), 117 (36),
133 (12), 185 (3), 213 (7), 231 (100), 246 (M+; 33).
2-Methyl-4-isopropylphenol 38: m/z (EI, 70eV) 91 (26), 115 (24),
117 (95), 132 (22), 133 (21), 213 (13), 231 (100), 246 (M+; 41).
6-Hydroxycymene 39: m/z (EI, 70eV) 65 (22), 91 (81), 115 (61),
131 (74), 159 (71), 213 (26), 228 (100), 246 (M+; 12).
8-Hydroxcymene 40: m/z (EI, 70eV) 69 (11), 91 (25), 117 (83), 132
(21), 133 (22), 203 (7), 231 (100), 246 (M+; 45).
3. Results
3.1. Catalytic turnover/spin state change
Previous experiments with P450cin catalysed oxidation of camphor
1 indicated that P450cin could catalyse the oxidation of substrates other
4

J.E. Stok, et al.
ArchivesofBiochemistryandBiophysics672(2019)108060
than cineole 3, although with reduced specificity and efficiency (cou-
pling of substrate oxidation to NADPH consumption) [24,25]. Thus, a
number of additional terpenes/terpenoid compounds (Fig. 3: 5–26)
were screened to assess the ability of P450cin and/or its mutants N242A
and N242T to catalytically oxidise alternative substrates (Table 1). In
general, both P450cin and N242A were found to oxidise the same
compounds (5–8, 10, 13, 14, 23, 24) with four compounds (11, 12, 21,
and 26) only oxidised by P450cin and not the mutants. Due to its poor
catalytic turnover rates in our initial screen (with 1, 3, 5–8 and 10)
N242T was not further evaluated for product formation with other
substrates and its kinetic parameters were not determined. (The cou-
pling of N242T during cineole 3 oxidation was previously determined
[24] to be 17% that of the wildtype enzyme.) The product yield from
P450cin and N242A catalysed oxidation of 8, 10, 13, 14, 22–24 was
poor and this was also reflected in their limited ability to convert
P450cin low spin Fe(III)·H2O to high spin Fe(III), a well-known indicator
of substrate binding at the active site of a P450 (Table 1). In general, the
percentage high spin conversion was determined with all isoforms for
compounds that had generated products following catalytic turnover
with wildtype P450cin, but was also determined for a small number of
compounds that were not detectably oxidised. Interestingly, a number
of terpenes were found to give a significantly higher percentage high
spin value for N242A compared to the wildtype enzyme; these included
(R)-limonene 7, pulegone 8, (−)-menthol 10, 22 and 23. P450cin failed
to catalyse the oxidation of a number of other substrates including all
four sesquiterpenes (C15) 16–19. This observation suggested that the
active site of P450cin may be too small to accommodate the extra bulk
of the larger sesquiterpene substrates.
Fig. 4. GC-MS trace comparing the oxidation products of the catalytic turnover
of (R)- and (S)-limonene 7 with P450cin WT and N242A, and the cis and trans
(R)-limonene 1,2-epoxide 31 standards. A. cis and trans (R)-limonene 1,2-ep-
oxide; B. WT with (R)-limonene; C. WT with (S)-limonene; D. N242A with (R)-
limonene; and E. N242A with (S)-limonene. Other products (RT = 7–8 min)
include: carveol 33, carvone 34 and perillyl alcohol 32.
3.2. P450cin: Unsaturated monoterpene hydrocarbons: α-pinene 5, β-pinene
6 and limonene 7
Pinenes 5 and 6, and limonene 7 all gave sufficient quantities of
products to justify further detailed analysis of the interaction of these
terpenes with P450cin and N242A. For P450cin, the ability for 5, 6 and
(S)-7 to shift the resting heme iron spin state from low spin Fe(III)·H2O
(417 nm) to high spin Fe(III) (392 nm) upon binding was relatively high
(Table 1: 56–70%). The change in the spin state observed for (R)-7
equivalents (NADPH) to the generation of terpene oxidation products
(Table 2: 18–44%) was much lower than with the natural substrate
cineole 3 but very similar to that which had been observed previously
for P450cin with camphor 1 (25–35%) [25].
P450cin oxidation of 5, 6, (R)- and (S)-7 were all found to produce a
mixture of oxygenated products (Fig. 4: B and C; Data for 5 and 6 not
shown). Comparison to a commercial MS fragmentation library and
authentic standards revealed that the products generated from oxida-
tion of 5, 6, (R)- and (S)-7 were mainly alcohols, carbonyl compounds
and epoxides (Fig. 5). For both pinenes 5 and 6 these products did not
include myrentol 27, but mainly pinocarveol 28 and the epoxides 29
and 30. In addition to cis and trans-limonene 1,2-epoxide 31, P450cin
oxidation of (R)- and (S)-7 also generated: perillyl alcohol 32; carveol
33; and carvone 34. It was observed that the stereochemistry of 7
dictated whether P450cin produces the cis or trans-limonene 1,2-epoxide
(32
camphor 1 (27–32%). P450cin was observed to have dissociation con-
stants (Kd) with pinenes and (Table 2) of 1.2 0.2 and
2.6 0.3 μM, respectively, similar to those reported for cineole 3 with
P450cin (0.7 μM) [21]. However, the dissociation constants found for
P450cin with (R)- and (S)-limonene 7 (Table 2: 27 3 and 17 2 μM,
3%) was similar to that previously reported for (1R) and (1S)-
5
6
respectively) were approximately an order of magnitude higher than
P450cin with the pinenes 5 and 6. The rate of NADPH consumption was
determined for P450cin with 5, 6 and (R)- and (S)-7 and found to be
approximately the same, if not faster than the rate reported for P450cin
with cineole 3 (Table 2: 83–140%). However, the coupling of reducing
Table 2
Kinetic constants determined for P450cin and N242A. Kd (μM): Dissociation constants (Kd) were calculated by measuring the absorbance difference from A392-A417
upon substrate binding. Rate (%): The rate of NADPH consumption is given as a percentage of the rate observed for P450cin in the presence of cineole, and was linear
over the time of the experiment. The rate of NADPH consumption by P450cin with cineole is typically 218 μM min−1 per μM P450 when the ratio of enzymes are 1:2:8
(P450:FdR; Cdx) [30]. Coupling (%): The coupling of NADPH to the generation of products was determined using a mixture of cis/trans limonene oxide 31 as the
product standard. See Materials and Methods for experimental detail.
Substrate
P450cin
Kd (μM)
0.7
N242A
Rate (%)
25
Rate (%)
100
Coupling (%)
Kd (μM)
Coupling (%)
3 [21,23,30]
(1R)-1 [25,26]
(1S)-1 [25,26]
5
6
(R)-7
(S)-7
80
35
25
18
27
44
24
2
2
3
2
2
6
1
0.3
49
3
101
81
4
33
29
1
1
1.2
1.5
0.09
0.08
4.6
0.1
0.2
0.05
0.04
0.5
0.7
55
55
19
18
21
18
5
6
2
1
2
2
3.5
3.4
38
0.5
0.6
5
5
1.2
2.6
27
0.2
0.3
3
138
118
83
14
4
20
30
3
1
4
17
2
140
10
6.6
5.8
0.2
5

J.E. Stok, et al.
ArchivesofBiochemistryandBiophysics672(2019)108060
catalytic turnover of (R)- and (S)-7 gave a much more complex mixture
of different oxidation products (Fig. 4: B and C). For example, cis-(R)-
limonene 1,2-epoxide 31 was only 23% of the total products following
the catalytic turnover of (R)-7 with P450cin compared to 80% with
N242A.
3.4. P450cin: Aryl monoterpene: p-cymene 12
P450cin oxidation of p-cymene was also analysed using an in vivo
system [25] that generated sufficient quantities of the products for
characterisation. 1H NMR of the crude product mixture allowed the
identification of the three major oxidation products by comparison to
reported literature values: 4-hydroxycymene 36; 1-hydroxycymene 37;
and 8-hydroxcymene 40 (Fig. 7). Derivatisation with trifluoroacetic
anhydride and analysis by GC-MS confirmed these assignments. Three
other minor products were observed in the GC-MS (Fig. 7B) and were
tentatively assigned as, 10-hydroxycymene 35, 2-methyl-4-iso-
propylphenol 38 and 6-hydroxycymene 39, based on their m/z 246
(TFA protected monohydroxylated p-cymene) and by comparison with
a commercial MS fragmentation library. Oxidation products 35 and 39
were expected, as they are direct hydroxylation products of p-cymene.
However, 38 was unexpected, as only five possible hydroxylation pro-
ducts can be generated. It is thought that 38 may be produced from a
rearrangement resulting in a methyl migration analogous to the NIH
shift [35,36]. Unusual binding spectra were observed by UV/Visible
spectroscopy, preventing simple Kd determination: low to high spin
conversion was seen at low concentrations of p-cymene (< 150 μM) but
the reverse was observed at higher concentrations (0.2–1 mM) with
wildtype P450cin. The percentage high spin conversion for p-cymene
Fig. 5. Possible oxidation products from the catalytic turnover of α-pinene 5, β-
pinene 6 (R)- and (S)-limonene 7.
(determined at 100 μM final concentration) with P450cin was 21
A moderate rate of NADPH consumption of P450cin with p-cymene 12
(41 4% when compared to P450cin) with 3 was also observed.
1%.
Fig. 6. P450cin catalysed oxidation of limonene 7 to cis or trans-limonene 1,2-
epoxide 31; cis-(4R)-limonene 1,2-epoxide from (R)-limonene, trans-(4S)-limo-
nene 1,2-epoxide from (S)-limonene.
3.5. N242T: camphor 1
The catalytic turnover of (1R)- and (1S)-camphor 1 with wildtype
P450cin and N242A has been previously analysed [25,26]. To complete
this series, the catalytic turnover of both (1R)- and (1S)-1 with N242T
was performed. During catalytic turnover of (1R)-1 with the N242T
mutant it was observed to produce predominantly 5-endo-hydro-
xycamphor 2b (77% of the products formed; Fig. 8) Additionally, the
oxidation of (1S)-1 by N242T also yielded mainly 5-endo-hydro-
xycamphor 2b (65% of the products formed) with a significant amount
of the 3-exo and 3-endo-hydroxycamphors 2c-d also present (28%).
31; (R)-limonene produced the cis-(4R)-limonene 1,2-epoxide, whereas
the (S)-limonene gave predominantly the trans-(4S)-limonene 1,2-ep-
oxide (Fig. 6).
3.3. N242A: Unsaturated monoterpene hydrocarbons: α-pinene 5, β-pinene
6 and limonene 7
Binding of pinenes 5, 6 or (S)-limonene 7 to N242A produced a spin
state change of a similar magnitude to that of these compounds with
P450cin (Table 1). However, there was a two-fold increase in the spin
state change produced by (R)-7 with N242A when compared to wild-
4. Discussion
P450cin and the available mutants N242A and N242T were screened
to determine their ability to catalyse the oxidation of a range of ter-
penes and terpenoid compounds. It was demonstrated that these en-
zymes were limited to the oxidation of terpene and terpenoids that were
similar in size to P450cin's natural C10 substrate, cineole 3, rather than
the larger sesquiterpenes (C15). Interestingly, nonpolar monoterpene
hydrocarbons, such as p-cymene 12, limonene 7, α-pinene 5 and β-
pinene 6 that all lack an oxygen in their structure, were found to be
relatively good substrates for wildtype P450cin (18–44% coupling for
5–7; Table 2). These results were similar to those previously observed
for P450cin catalysed oxidation of camphor ((1S)-camphor 25%; (1R)-
camphor 35%) [25]. However, both the rate of reaction and binding
constants for pinenes 5–6 were found to be more analogous to the in-
teraction between P450cin and cineole than to P450cin and camphor.
Therefore, this provides some preliminary evidence indicating that
P450cin may be a useful enzyme for modification of other mono-
terpenes, especially bicyclic monoterpenes. The in vitro binding beha-
viour of P450cin with p-cymene was found to be unusual but a number
of products were observed to be produced by catalytic oxidation with
type (Table 1: N242A 64
1% vs P450cin 32
3%). The dissociation
constants obtained for N242A with 5, 6 and (R)- and (S)-7 were ap-
proximately an order of magnitude smaller than those obtained with
the same substrates with P450cin (Table 2) indicating a significant in-
crease in binding affinity with the mutant. The rate of NADPH con-
sumption of 5, 6 and (R)- and (S)-7 with N242A dropped considerably
when compared with P450cin and the same substrates. However, cou-
pling of reducing equivalents to the generation of products remained
similar to that observed with the wildtype enzyme (Table 2) with the
exception of (S)-limonene 7, which was considerably lower than that
observed for P450cin
.
The catalytic turnover of (R)- and (S)-7 with N242A produced
predominantly the cis-limonene 1,2-epoxide 31 (Fig. 4: D and E). The
cis-limonene 1,2-epoxide 31 comprised 80% of the products following
(R)-limonene 7 oxidation and 57% of the products during (S)-limonene
5 oxidation. This suggests that the N242A mutant is more selective for
the generation of a single product that the wildtype enzyme where the
6

J.E. Stok, et al.
ArchivesofBiochemistryandBiophysics672(2019)108060
Fig. 7. A. Possible oxidation products (35–40) formed from a P450cin in vivo oxidation of p-cymene 12 in minimal medium. B. GC-MS m/z 246 ion trace of the TFA-
derivatised oxidation products from an in vivo turnover of p-cymene 12.
P450cin. Thus, the in vivo metabolism of this substrate by a bicistronic
P450cin system was characterised to elucidate the nature of the products
formed. The three major products formed from P450cin catalysed oxi-
dation of p-cymene were 36, 37 and 40 in a ratio of 2:1:6. These cor-
respond to hydroxylation at the easily oxidised benzylic methine, the
slightly less easily oxidised benzylic methyl and the least sterically
encumbered aromatic position. Careful analysis of the GC-MS trace of
the derivatised products showed the remaining two expected products
35 and 39, corresponding to oxidation at the less reactive methyl and
hindered aromatic position, were present at low levels (Fig. 7). Inter-
estingly, a sixth product was identified by comparison of its MS frag-
mentation pattern with a commercial library as 38. This would arise
from ipso attack on the methyl bearing aromatic carbon followed by an
NIH shift of the methyl group and subsequent rearomatisation. Such a
shift has previously been reported as a minor product in the P450BM-3
catalysed oxidation of xylene [36] but, interestingly, was found to occur
more in o-xylene and not at all in p-xylene.
In general, it was observed that the percentage high spin was not a
satisfactory indication of the potential for a compound to be a substrate
for the enzyme. The percentage high spin shows whether the compound
has displaced the resting water in the active site by shifting the heme
iron from low (417 nm) to high spin (392 nm). Frequently, the per-
centage high spin enzyme was observed to increase with the N242A
mutant as compared to wildtype P450cin (8, 10, 22, 23; Table 1), but
this increase did not coincide with an increased amount of product
observed in the N242A catalysed oxidation of these compounds when
compared to P450cin (Table 1). In the case of β-ionone 25, N242T was
observed to have a 35% spin state change but no products were
detected from the catalytic turnover. It therefore appears that the
coupling of reducing equivalents to the production of oxidation pro-
ducts may be the most important indicator when optimising these
oxidations. Consequently, screening compounds by catalytic activity
together with the coupling information appears to be the best strategy
to determine which enzyme/substrate combinations could be the most
viable for further modification.
Both the N242A and N242T mutants of P450cin were explored as
potential catalysts for monoterpene oxidation. It has previously been
observed that amino acid substitutions at this position (N242) have
resulted in significantly altered product profiles for the oxidation of
both cineole and camphor [23,24,26]. In a similar fashion to N242A
[26], N242T catalysed oxidation of (1R)-camphor was observed to in-
crease the production of the 5-endo-hydroxycamphor 2b and decrease
the oxidation at the 3-position to 2c/d. However, unlike the N242A
catalysed oxidation of (1S)-camphor 1, this preference for the produc-
tion of 2b was maintained during the oxidation of the (1S)-camphor by
N242T. This further suggests that amino acid substitutions at the N242
position are important in directing the regioselectivity of the products
generated and that the type of amino acid can also change the expected
stereoselectivity of the oxidation. Thus, subtle changes in the active site
induced by the N242A mutation may be an important starting point by
which to produce new oxygen containing monoterpenes. Interestingly,
wildtype P450cin was also found to differentiate between different
isomers of monoterpenes such as limonene 7. P450cin predominantly
converted (R)-7 to cis-limonene 1,2-epoxide 31, whereas trans-31 was
produced from the oxidation of (S)-7. This again demonstrates the
importance of the stereochemistry of the substrate itself in dictating the
7

J.E. Stok, et al.
ArchivesofBiochemistryandBiophysics672(2019)108060
Fig. 8. Oxidation products of the catalytic turnover of (1R)- and (1S)-1 with N242T, and for comparison, previously published results for P450cin [25] and N242A
[26]. Only the R stereochemistry is shown. Numbers indicate the amount (%) of each isomer formed from both (1R)- and (1S)-1 as determined by GC-MS.
outcome of the oxidation.
Acknowledgments
Furthermore, in contrast to wildtype P450cin and with the exception
of cineole, oxidation of monoterpenes by N242A often generates mainly
one product. For example, N242A catalysed oxidation of (R)-limonene
7 produced cis-(R)-limonene 1,2-epoxide 31, which comprised 80% of
the products of the oxidation. This is similar to the selectivity pre-
viously observed with (1R)-camphor 1 where N242A converted (1R)-1
to predominantly 5-endo-hydroxycamphor 2b (73% of the total pro-
ducts formed) [26]. It therefore appears that the substitution at the
N242 position not only significantly alters the product profile, but can
potentially direct the oxidation towards a single product. However,
critically, there are a number of factors that must be improved for the
potential use of the N242A mutant in biocatalytic applications. The rate
of the reaction and coupling efficiency were found to be compromised
by the introduction of this mutation. Therefore, other modifications
would be required to improve the viability of N242A, such as the in-
troduction of “accelerator” mutations that could increase the rate of the
reaction without disturbing the specificity. These types of accelerator
mutations have been employed in other P450s such as P450BM3 [37].
In sum, this study has shown that P450cin is a good scaffold to ex-
plore for the oxidation of a variety of C10 bicyclic monoterpenes. The
N242 residue has also been shown to be a useful site to alter when
attempting to control the specificity of the oxidation. Future work will
examine both other amino acid substitutions at this site, in addition to
other mutations that can refine the specificity and maintain the rate and
coupling efficiency of the original P450cin oxidation.
This work was in part supported by ARC grants DP140103229 and
DP170102220 (JJDV).
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.abb.2019.07.025.
References
[1] V.B. Urlacher, M. Girhard, Cytochrome P450 monooxygenases: perspectives for
synthetic application, Trends Biotechnol. 30 (1) (2012) 26–36.
[2] P.K. Ajikumar, K. Tyo, S. Carlsen, O. Mucha, T.H. Phon, G. Stephanopoulos,
Terpenoids: opportunities for biosynthesis of natural product drugs using en-
gineered microorganisms, Mol. Pharm. 5 (2) (2008) 167–190.
[3] H. Schewe, M. Mirata, D. Holtmann, J. Schrader, Biooxidation of monoterpenes
with bacterial monooxygenases, Process Biochem. 46 (10) (2011) 1885–1899.
[4] P.W. Trudgill, Microbial metabolism and transformation of selected monoterpenes,
in: C. Ratledge (Ed.), Biochemistry of Microbial Degradation, Kluwer Academic
Publishers, London, 1994, pp. 33–61.
[5] T.L. Poulos, B.C. Finzel, I.C. Gunsalus, G.C. Wagner, J. Kraut, The 2.6-A crystal-
structure of Pseudomonas putida Cytochrome P450, J. Biol. Chem. 260 (30) (1985)
6122–6130.
[6] J.A. McIntosh, C.C. Farwell, F.H. Arnold, Expanding P450 catalytic reaction space
through evolution and engineering, Curr. Opin. Chem. Biol. 19 (2014) 126–134.
[7] E.M.J. Gillam, Engineering cytochrome P450 enzymes, Chem. Res. Toxicol. 21 (1)
(2008) 220–231.
[8] S.G. Bell, X. Chen, F. Xu, Z. Rao, L.-L. Wong, Engineering substrate recognition in
catalysis by cytochrome P450cam, Biochem. Soc. Trans. 31 (Pt 3) (2003) 558–562.
[9] S.G. Bell, C.F. Harford-Cross, L.-L. Wong, Engineering the CYP101 system for in vivo
oxidation of unnatural substrates, Protein Eng. 14 (10) (2001) 797–802.
[10] D. Kim, P.R. Ortiz de Montellano, Tricistronic overexpression of cytochrome
8

J.E. Stok, et al.
ArchivesofBiochemistryandBiophysics672(2019)108060
hydroxylation in p450cin, J. Biol. Chem. 283 (16) (2008) 10804–10812.
[24] K.E. Slessor, A.J. Farlow, S.M. Cavaignac, J.E. Stok, J.J. De Voss, Oxygen activation
by P450(cin): protein and substrate mutagenesis, Arch. Biochem. Biophys. 507 (1)
(2011) 154–162.
[25] K.E. Slessor, D.B. Hawkes, A. Farlow, A.G. Pearson, J.E. Stok, J.J. De Voss, An in
vivo cytochrome P450cin (CYP176A1) catalytic system for metabolite production,
J. Mol. Catal. B Enzym. 79 (2012) 15–20.
[26] J.E. Stok, E.A. Hall, I.S.J. Stone, M.C. Noble, S.H. Wong, S.G. Bell, J.J. De Voss, In
vivo and in vitro hydroxylation of cineole and camphor by cytochromes
P450CYP101A1, CYP101B1 and N242A CYP176A1, J. Mol. Catal. B Enzym. 128
(2016) 52–64.
P450cam , putidaredoxin, and putidaredoxin reductase provides a useful cell-based
catalytic system, Biotechnol. Lett. 31 (9) (2009) 1427–1431.
[11] R.J. Sowden, S. Yasmin, N.H. Rees, S.G. Bell, L.-L. Wong, Biotransformation of the
sesquiterpene (+)-valencene by cytochrome P450cam and P450BM-3, Org, Biomol.
Chem. 3 (1) (2005) 57–64.
[12] J.H.Z. Lee, S.H. Wong, J.E. Stok, S.A. Bagster, J.R. Beckett, J.K. Clegg, A. Brock,
J.J. De Voss, S.G. Bell, Selective hydroxylation of 1,8- and 1,4-cineole using bac-
terial P450 variants, Arch. Biochem. Biophys. 663 (2019) 54–63.
[13] S.G. Bell, R.J. Sowden, L.-L. Wong, Engineering the haem monooxygenase cyto-
chrome P450cam for monoterpene oxidation, Chem. Commun. (7) (2001) 635–636.
[14] S. Schneider, M. Wubbolts, D. Sanglard, B. Witholt, Production of chiral hydroxy
long chain fatty acids by whole cell biocatalysis of pentadecanoic acid with an E.
coli recombinant containing cytochrome P450BM-3 monooxygenase, Tetrahedron:
Asymmetry 9 (16) (1998) 2833–2844.
[27] H. Toivonen, S.A. Laurema, P.J. Ilvonen, Reduction of 2- and 3-nitrobornanes to
bornane, Tetrahedron Lett. 34 (1971) 3203–3204.
[28] D.E. Cane, G. Yang, R.M. Coates, H.-J. Pyun, T.M. Hohn, Trichodiene synthase.
Synergistic inhibition by inorganic pyrophosphate and aza analogs of the bisabolyl
cation, J. Org. Chem. 57 (1992) 3454–3462.
[15] C.J.C. Whitehouse, S.G. Bell, L.-L. Wong, P450 BM3(CYP102A1): connecting the
dots, Chem. Soc. Rev. 41 (3) (2012) 1218–1260.
[16] R.J.F. Branco, A. Seifert, M. Budde, V.B. Urlacher, M.J. Ramos, J. Pleiss, Anchoring
effects in a wide binding pocket: the molecular basis of regioselectivity in en-
gineered cytochrome P450 monooxygenase from B. megaterium, Proteins 73 (3)
(2008) 597–607.
[17] H. Schewe, D. Holtmann, J. Schrader, P450BM-3-catalyzed whole-cell bio-
transformation of α-pinene with recombinant Escherichia coli in an aqueous–or-
ganic two-phase system, Appl. Microbiol. Biotechnol. 83 (5) (2009) 849–857.
[18] H. Schewe, B.-A. Kaup, J. Schrader, Improvement of P450(BM-3) whole-cell bio-
catalysis by integrating heterologous cofactor regeneration combining glucose fa-
cilitator and dehydrogenase in E. coli, Appl. Microbiol. Biotechnol. 78 (1) (2008)
55–65.
[29] S. Marine, J. Clemons, Determination of limonene oxidation products using SPME
and GC-MS, J. Chromatogr. Sci. 41 (1) (2003) 31–35.
[30] D.B. Hawkes, K.E. Slessor, P.V. Bernhardt, J.J. De Voss, Cloning, expression and
purification of cindoxin, an unusual fmn-containing cytochrome P450 redox
partner, Chembiochem 11 (8) (2010) 1107–1114.
[31] C.M. Jenkins, M.R. Waterman, NADPH-flavodoxin reductase and flavodoxin from
Escherichia coli: characteristics as a soluble microsomal P450 reductase,
Biochemistry 37 (17) (1998) 6106–6113.
[32] M. Razzaghi-Abyaneh, M. Shams-Ghahfarokhi, T. Yoshinari, M.-B. Rezaee,
K. Jaimand, H. Nagasawa, S. Sakuda, Inhibitory effects of Satureja hortensis L.
essential oil on growth and aflatoxin production by Aspergillus parasiticus, Int. J.
Food Microbiol. 123 (3) (2008) 228–233.
[19] A. Seifert, S. Vomund, K. Grohmann, S. Kriening, V.B. Urlacher, S. Laschat, J. Pleiss,
Rational design of a minimal and highly enriched CYP102A1 mutant library with
improved regio-, stereo- and chemoselectivity, Chembiochem 10 (5) (2009)
853–861.
[33] B. Basu, B. Mandal, S. Das, P. Das, A.K. Nanda, Chemoselective reduction of alde-
hydes by ruthenium trichloride and resin-bound formates, Beilstein J. Org. Chem. 4
(1) (2008) 53.
[20] H. Venkataraman, S. Beer, D. Geerke, N. Vermeulen, J. Commandeur, Regio‐and
stereoselective hydroxylation of optically active α‐ionone enantiomers by en-
gineered cytochrome P450 BM3 mutants, Adv. Synth. Catal. 354 (11‐12) (2012)
2172–2184.
[34] C. Russo-Caia, S. Steenken, Photo- and radiation-chemical production of radical
cations of methylbenzenes and benzyl alcohols and their reactivity in aqueous so-
lution, Phys. Chem. Chem. Phys. 4 (8) (2002) 1478–1485.
[35] D.M. Jerina, J.W. Daly, Arene oxides: a new aspect of drug metabolism, Science 185
[21] D.B. Hawkes, G.W. Adams, A.L. Burlingame, P.R. Ortiz de Montellano, J.J. De Voss,
Cytochrome P450cin (CYP176A), isolation, expression, and characterization, J.
Biol. Chem. 277 (31) (2002) 27725–27732.
[22] P.R. Ortiz de Montellano, J.J. De Voss, Substrate oxidation by cytochrome P450
enzymes, in: P.R. Ortiz de Montellano (Ed.), Cytochrome P450: Structure,
Mechanism, and Biochemistry, Kluwer Academic/Plenum Publishers, New York,
2005, pp. 183–245.
(4151) (1974) 573–582.
[36] C.J.C. Whitehouse, N.H. Rees, S.G. Bell, L.-L. Wong, Dearomatisation of o-Xylene by
P450BM3 (CYP102A1), Chem. Eur J. 17 (24) (2011) 6862–6868.
[37] S.D. Munday, S. Dezvarei, S.G. Bell, Increasing the activity and efficiency of ste-
reoselective oxidations by using decoy molecules in combination with rate-enhan-
cing variants of P450BM3, ChemCatChem 8 (2016) 2789–2796.
[38] J.E. Stok, S. Yamada, A.J. Farlow, K.E. Slessor, J.J. De Voss, Cytochrome P450(cin)
(CYP176A1) D241N: investigating the role of the conserved acid in the active site of
cytochrome P450s, Biochim. Biophys. Acta 1834 (3) (2013) 688–696.
[23] Y.T. Meharenna, K.E. Slessor, S.M. Cavaignac, T.L. Poulos, J.J. De Voss, The critical
role of substrate-protein hydrogen bonding in the control of regioselective
9