Improved oxidation of aromatic and aliphatic hydrocarbons using rate enhancing variants of P450Bm3 in combination with decoy molecules

Article abstract of DOI:10.1039/c5cc09247g

Enzyme performance can be improved using decoy molecules or engineered variants to accelerate the activity without affecting selectivity. Here we combine a rate accelerator variant of cytochrome P450Bm3 with decoy molecules to enhance the oxidation activity of a range of small organic molecules. This combined approach offers superior biocatalytic efficiency without modifying the product distribution.

Full text of DOI:10.1039/c5cc09247g

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Improved oxidation of aromatic and aliphatic
hydrocarbons using rate enhancing variants of
P450Bm3 in combination with decoy molecules†
Cite this: Chem. Commun., 2016,
52, 1036
Received 7th November 2015,
Accepted 17th November 2015
Samuel D. Munday,a Osami Shoji,b Yoshihito Watanabe,b Luet-Lok Wongc and
Stephen G. Bell*a
DOI: 10.1039/c5cc09247g
www.rsc.org/chemcomm
Enzyme performance can be improved using decoy molecules or use of these enzymes in synthesis.6,10 Wild-type P450Bm3 (WT)
engineered variants to accelerate the activity without affecting turns over most unnatural substrates at desultory rates but it is
selectivity. Here we combine a rate accelerator variant of cytochrome a versatile platform for designing biocatalysts via protein
P450Bm3 with decoy molecules to enhance the oxidation activity of a engineering.5,6,11–15
range of small organic molecules. This combined approach offers
It has been shown that chemically inert perfluorocarboxylic
superior biocatalytic efficiency without modifying the product acids (PFCs) can act as decoy molecules and greatly promote
distribution.
the oxidation of unnatural substrates such as benzenes and
short chain alkanes. The inert decoy fills part of the P450Bm3
Cytochrome P450s (CYPs) are heme-dependent enzymes that active site and constrains potential substrates to bind closer to
catalyse the oxidation of a broad array of organic substrates, the heme than they would without an added PFC, leading to
allowing them to perform physiological roles as varied as increased activity.15–17 The structure of a decoy molecule bound
steroid hormone and antibiotic synthesis and the breakdown WT (PDB 3WSP) showed that the PFC filled the substrate
of xenobiotics.1–3 The signature reaction of CYP enzymes is the channel but its shorter chain length left a significant amount
insertion of a single oxygen atom from molecular dioxygen into of space close to the heme to bind another molecule (see ESI†).18
a carbon–hydrogen bond to give the corresponding alcohol.4
The direct hydroxylation of organic compounds by chemical
methods is energy-intensive, often non-selective and leads to
the formation of side products and toxic wastes. CYP-mediated
oxidation reactions have the potential to provide efficient
biocatalytic routes to alcohols by C–H activation, which is not
feasible using classical organic synthetic methods.5–8
CYP102A1 (P450Bm3) from Bacillus megaterium is a soluble
119.5 kDa enzyme in which the P450 heme domain is fused to a
reductase domain.6 This catalytically self-sufficient enzyme
requires only NADPH and oxygen to function, and hydroxylates
medium-chain length (C11–C18) fatty acids at sub-terminal
positions.9 CYP102A1 and other CYP102 family enzymes exhibit
the highest turnover frequency (41000 minÀ1) of any CYP
enzyme and this overcomes one of the major hurdles to the
a Department of Chemistry, University of Adelaide, SA 5005, Australia.
E-mail: stephen.bell@adelaide.edu.au
b Department of Chemistry, Graduate School of Science, Nagoya University,
Fig. 1 GC-MS analysis of the turnovers of (a) benzene and (b) cyclohexane
using the WT (i) and KT2 (ii) variants of P450Bm3 with and without the decoy
molecule PFC9. In black is the turnover in the absence of the decoy
molecule and in red the turnover in the presence of the decoy molecule.
In (b) (ii) the turnover of cyclohexane with KT2 with PFC9 has been offset
slightly along the x-axis for clarity. The integrations of these peaks and the
internal standard are given in Table S1 (ESI†).
Furo-cho, Chikusa-ku, Nagoya, 464-8602, Japan
c Department of Chemistry, University of Oxford, Inorganic Chemistry Laboratory,
South Parks Road, Oxford, OX1 3QR, UK
† Electronic supplementary information (ESI) available: Experimenal, structure
details, NADPH consumption traces, MS data, GC-MS and HPLC analysis. See
DOI: 10.1039/c5cc09247g
1036 | Chem. Commun., 2016, 52, 1036--1039
This journal is ©The Royal Society of Chemistry 2016

View Article Online
Communication
ChemComm
Table 1 In vitro turnover activity data for the P450Bm3 variants and decoy molecule combinations with benzene and cyclohexane as substrates and
in the absence of substrate (decoy). N = NADPH turnover rate. C = coupling efficiency (%). PFR = product formation rate. Rates are in nmol (nmol-
P450)À1 minÀ1. All data are reported as mean Æ S.D. (n 4 3)
Decoy
Benzene
Cyclohexane
N
N
C
PFR
N
C
PFR
WT
—
29 Æ 0.3
71 Æ 1
0.6 Æ 0.05
3.0 Æ 0.2
13 Æ 0.6
8.6 Æ 0.6
0.2 Æ 0.02
2 Æ 0.2
23 Æ 1
30 Æ 2
212 Æ 8
416 Æ 7
626 Æ 14
2.2 Æ 0.4
31 Æ 2
28 Æ 2
15 Æ 1
0.7 Æ 0.07
65 Æ 6
WT/PFC8
WT/PFC9
WT/PFC10
35 Æ 1
56 Æ 0.4
327 Æ 11
168 Æ 7
522 Æ 6
119 Æ 8
91 Æ 7
45 Æ 4
KT2
—
77 Æ 6
220 Æ 4
374 Æ 7
847 Æ 10
3.0 Æ 0.2
4.6 Æ 0.4
10 Æ 0.3
8.8 Æ 0.8
2.3 Æ 0.3
10 Æ 1
38 Æ 1
74 Æ 1
353 Æ 9
900 Æ 19
1338 Æ 25
1797 Æ 45
19 Æ 1
35 Æ 4
36 Æ 1
27 Æ 1
70 Æ 5
KT2/PFC8
KT2/PFC9
KT2/PFC10
83 Æ 3
161 Æ 8
699 Æ 12
315 Æ 36
483 Æ 28
484 Æ 31
Enzyme–substrate interactions in the vicinity of the heme remain and cyclohexane, in line with what has been reported previously
important in the catalytic activity.
with other substrates (Table 1 and Fig. 1).10,21 Benzene oxidation
An earlier directed evolution study identified P450Bm3 variants
with enhanced activity for unnatural substrate oxidation but which
maintained the product selectivity of the WT.10,19–21 The crystal
structure of the substrate-free form of the KT2 variant (A191T/
N239H/I259V/A276T/L353I; PDB 3PSX) showed that it had the
same conformation of the I helix around the critical oxygen-
binding groove as fatty acid-bound WT.21 The heme axial water
interaction was weakened because the hydrogen bond between
this water and the carbonyl of Ala264 was broken. This should
promote substrate binding close to the heme. Kinetic studies also
showed that the rate of the first electron transfer step for substrate-
free KT2 was comparable to that of palmitic acid-bound WT.
Therefore, KT2 has a catalytically ready conformation, such that
substrate-induced changes to the structure play a less significant
role in promoting the electron transfer steps (see ESI†).21
Since the mechanisms by which decoy molecules and the
KT2 variant showed enhanced unnatural substrate activity are
different we explored whether these mechanisms would work in
concert to facilitate the oxidation of smaller organic molecules
such as benzene and cyclohexane by P450Bm3. We initially
investigated the activity of WT P450Bm3 and the KT2 variant with
and without added PFC8, PFC9 and PFC10.18,22 As reported
previously the addition of these PFCs improved the activity of
the turnovers of WT P450Bm3 with cyclohexane and benzene.22
These turnovers yielded phenol and cyclohexanol as the sole
products. Both were identified by co-elution experiments with
standards using GC-MS (Fig. 1 and Fig. S1, ESI†). PFC10 itself
induced the highest activity of NADPH oxidation in WT P450Bm3
followed by PFC9 and then PFC8 (Table 1 and ESI†). This trend in
NADPH oxidation activity was followed for the different PFC
molecules. Addition of PFC9 increased the coupling efficiency with
WT to a greater extent than PFC10 for both substrates (Table 1 and
Fig. 1).22 Overall the combined effects of increased activity and
coupling resulted in the largest improvements in the product for-
mation rate for PFC9 and PFC10 over PFC8 (Table 1). The maximal
augmentation of the activity upon addition of a decoy molecule
was 170-fold for cyclohexane and 225-fold for benzene.
Scheme 1 The products formed from the turnovers of P450Bm3 enzyme/
decoy molecule combinations with benzene substrates. Abbreviations used;
2-methoxyphenol (2-MP), 4-methoxyphenol (4-MP), 2,5-dimethylphenol
(2,5-DMP), 2,4-dimethylphenol (2,4-DMP), 4-methylbenzyl alcohol (4-MBA),
2,6-dimethylphenol (2,6-DMP), 3-methylbenzylalcohol (3-MBA), 2-methyl-
benzylalcohol (2-MBA), 2,3-dimethylphenol (2,3-DMP), 3,4-dimethylphenol
(3,4-DMP) and 6,6-dimethylcyclohexa-2,4-dienone (6,6-DMCHD).
In the absence of added PFCs the KT2 variant showed
increased NADPH oxidation activity, coupling efficiency and
hence the product formation rate over the WT with both benzene
This journal is ©The Royal Society of Chemistry 2016
Chem. Commun., 2016, 52, 1036--1039 | 1037

View Article Online
ChemComm
Communication
Table 2 In vitro turnover activity data for the P450Bm3 variants and
decoy molecule combinations with toluene, anisole, p-, m- and o-xylene.
See Table 1 for details
Encouraged by these results we assessed if the decoy molecule/
rate enhancement mutant combinations would work for substi-
tuted benzenes. The oxidation of toluene and anisole by P450Bm3
have been reported to be highly selective (490%) for the ortho-
hydroxylated product.21–23 The turnovers of toluene generated
p-cresol and benzyl alcohol as minor products (o9% total product)
while 4-methoxyphenol (o10%) is the sole minor product identi-
fied for anisole oxidation.22,23 The biocatalytic oxidation of xylenes
by P450Bm3 has also been reported. The turnover of p-xylene is
documented to generate only 2,5-dimethylphenol24 and m-xylene
oxidation produced 2,4-dimethylphenol (87%), 2,6-dimethylphenol
(11%) and 3-methylbenzylalcohol (2%).25 The P450Bm3 catalysed
oxidation of o-xylene was more complex, yielding five products.
The major product is 2-methylbenzyl alcohol (47%) with 2,3- and
3,4-dimethylphenol being generated as two of the minor products,
27% and 10%, respectively. The other two minor metabolites arise
from a methyl shift resulting in formation of 2,6-dimethylphenol
(8%) and the dearomatisation product, 6,6-dimethylcyclohexa-2,4-
dienone (8%, Scheme 1).25 The analysis of the turnovers of
these larger substrates with the PFCs and KT2 would be a more
stringent test to assess if these mutant and decoy molecule
combinations would alter the product distribution as well as
heighten the activity.
WT +
PFC9
WT +
PFC10
KT2 +
PFC9
KT2 +
PFC10
WT
KT2
Toluene
N
C
22 Æ 0.9 355 Æ 3 622 Æ 8
68 Æ 2
549 Æ 7 822 Æ 12
41 Æ 1 33 Æ 1
225 Æ 8 271 Æ 9
1.6 Æ 0.5 40 Æ 1
25 Æ 2
10 Æ 3
7 Æ 2
PFR 0.3 Æ 0.1 144 Æ 3 158 Æ 10
Anisole
N
C
122 Æ 2
262 Æ 2 581 Æ 3
189 Æ 12 399 Æ 9 729 Æ 16
1.4 Æ 0.1 18 Æ 3
11 Æ 0.5 7.5 Æ 1
64 Æ 3 14 Æ 2
19 Æ 2
76 Æ 6
12 Æ 0.1
87 Æ 2
PFR 2.0 Æ 0.1 48 Æ 9
p-Xylene
N
C
63 Æ 5
19 Æ 2
687 Æ 7 943 Æ 7
56 Æ 4 41 Æ 4
386 Æ 26 394 Æ 35
267 Æ 10 1010 Æ 33 1080 Æ 8
34 Æ 0.9 54 Æ 2
42 Æ 2
PFR 12 Æ 2
90 Æ 1
549 Æ 22 460 Æ 26
m-Xylene
N
C
33 Æ 0.6 546 Æ 3 698 Æ 8
178 Æ 5
830 Æ 7 1050 Æ 12
3.5 Æ 0.4 55 Æ 1
40 Æ 0.4 26 Æ 2
57 Æ 2
43 Æ 2
PFR 1.1 Æ 0.1 302 Æ 5 278 Æ 4
46 Æ 4
476 Æ 15 455 Æ 12
o-Xylene
N
C
29 Æ 1
284 Æ 5 468 Æ 5
28 Æ 2
106 Æ 2
30 Æ 3
32 Æ 2
738 Æ 18 1020 Æ 16
1.9 Æ 0.1 53 Æ 2
45 Æ 1
34 Æ 1
PFR 0.6 Æ 0.06 149 Æ 8 131 Æ 6
334 Æ 17 346 Æ 10
All five substrates were tested with WT and KT2 P450Bm3 with
PFC8 through to PFC10. Despite the larger size of these substituted
was 12-fold faster and a 100-fold increase was observed for benzenes, a similar pattern of activity enhancement was observed
cyclohexane. However, the activities were lower than the best with the decoy molecules. The turnovers containing PFC10 showed
WT/PFC combinations (Table 1). Addition of the PFC decoy the highest activity of NADPH oxidation and those with PFC9 were
molecules to KT2 engendered further enhancement of the the more efficiently coupled. In certain instances, e.g. with toluene,
oxidation of benzene and cyclohexane. The activity was higher the faster rates of NADPH oxidation with PFC10 resulted in higher
than with the WT in all cases. The trends in the catalytic para- product formation rates while in others the higher coupling
meters of the different PFCs with KT2 mirrored those obtained obtained with PFC9 resulted in the maximum activity, e.g.
with the WT (Table 1). As a consequence the product formation rate p-xylene with KT2/PFC9 (Table 2). In all cases the greatest product
of KT2 with PFC9 and PFC10 are similar and both are greater than formation rates were obtained with KT2/PFC9 or KT2/PFC10.
when PFC8 is used. KT2 and PFC10 was the most active combi-
The oxidation of both toluene and anisole was faster than
nation for benzene oxidation, with a product formation rate of that of benzene for all the enzyme/decoy molecule combina-
74 minÀ1. The gain in activity with cyclohexane was even greater, tions tested. The maximum enhancements in the rate of
with the product formation rate being 484 minÀ1 and 483 minÀ1 product formation, over WT P450Bm3, were approximately
with the KT2/PFC10 and KT2/PFC9 combinations. This was four- 900-fold for toluene and 40-fold for anisole (Table 2 and
fold higher than the best WT pairing (PFC9, Table 1 and Fig. 1).
Table S2, ESI†). The larger gain with toluene predominantly arose
Fig. 2 GC-MS analysis of the turnovers of (a) o-Xylene using (i) WT P450Bm3 and (ii) KT2 and PFC10, the products are labelled; 2-methylbenzylalcohol
(2-MBA), 2,3-dimethylphenol (2,3), 3,4-dimethylphenol (3,4), 2,6-dimethylphenol (2,6), and 6,6-dimethylcyclohexa-2,4-dienone (6,6). (b) (i) p-Xylene
using the KT2 and PFC8, the products are labelled 2,5-dimethylphenol (2,5), 2,4-dimethylphenol (2,4) and 4-methylbenzyl alcohol (4-MBA) Impurities are
labelled *.
1038 | Chem. Commun., 2016, 52, 1036--1039
This journal is ©The Royal Society of Chemistry 2016

View Article Online
Communication
ChemComm
from a more significant increase in the coupling efficiency when P450Bm326 and could be used to drive substrate oxidation by CYP
compared to the anisole turnovers. enzymes at high activities and improved product yields. This would
The best KT2/PFC combinations resulted in product enable the generation of the levels of product required in synthetic
formation rates which were almost double that of the WT/ chemistry applications for reactions which show promising regio-
PFC pairings. The oxidation of toluene and anisole resulted in and stereoselectivity with WT or mutant forms of the enzyme.
the generation of o-cresol (94–95%) and o-methoxyphenol
Note added after first publication: This article replaces the
(88–92%), respectively (Fig. S2, S3 and Table S3, ESI†). Benzyl version published on 23rd November 2015, which contained
alcohol and p-cresol from toluene and 4-methoxyphenol from errors in Scheme 1.
anisole were minor metabolites in these turnovers. The product
distributions were virtually unchanged when compared to the of a MPhil Scholarship (to SDM) and part-funding the project
WT enzyme (Scheme 1; Fig. S2 and Table S3, ESI†). through the award of a Priority Partner Project Grant (with
The authors thank the University of Adelaide for the award
Xylene oxidation was enhanced using the PFC decoys with Nagoya University).
KT2. The maximum activity enhancements over the WT were;
p-xylene, 45-fold; m-xylene, 430-fold; o-xylene, 575-fold (Table 2
Notes and references
and Fig. S4, ESI†). The product formation rates were higher than
those achieved with toluene. The product distributions across all the
turnovers were comparable to those reported for the WT and the
KT2 variant (Scheme 1, Fig. S2 and S3, Table S3, ESI†). For example,
the ratios of the five products formed by the WT enzyme and
o-xylene, including those arising from a shift in a methyl group,
are similar to those obtained with the KT2/PFC10 combination,
which has the highest activity (Fig. 2 and Table S3, ESI†).
1 P. R. Ortiz de Montellano, Cytochrome P450: Structure, Mechanism
and Biochemistry, Plenum Publishers, New York, 2005.
2 A. Sigel, H. Sigel and R. Sigel, The Ubiquitous Roles of Cytochrome
P450 Proteins, John Wiley & Sons, Weinheim, 2007.
3 F. P. Guengerich, Chem. Res. Toxicol., 2001, 14, 611–650.
4 P. R. Ortiz de Montellano, Chem. Rev., 2010, 110, 932–948.
5 S. G. Bell, N. Hoskins, C. J. C. Whitehouse and L. L. Wong, Met. Ions
Life Sci., 2007, 437–476.
6 C. J. Whitehouse, S. G. Bell and L. L. Wong, Chem. Soc. Rev., 2012,
41, 1218–1260.
One point of difference to the previously reported data for
p-xylene was that three products were observed in the GC-MS
analysis of all the turnovers including those of the WT enzyme
7 G. D. Roiban and M. T. Reetz, Chem. Commun., 2015, 51, 2208–2224.
8 R. Fasan, ACS Catal., 2012, 2, 647–666.
9 S. S. Boddupalli, R. W. Estabrook and J. A. Peterson, J. Biol. Chem.,
1990, 265, 4233–4239.
(Fig. 2, Fig. S5 and Table S3, ESI†).24 These corresponded to 10 C. J. Whitehouse, S. G. Bell, H. G. Tufton, R. J. Kenny, L. C. Ogilvie
and L. L. Wong, Chem. Commun., 2008, 966–968.
11 A. Seifert, S. Vomund, K. Grohmann, S. Kriening, V. B. Urlacher,
4-methylbenzyl alcohol (0–6%), 2,4- and 2,5-dimethylphenol
(15–17% and 77–85%, respectively). The 2,4-dimethylphenol
S. Laschat and J. Pleiss, ChemBioChem, 2009, 10, 853–861.
product must arise via the NIH shift of a methyl group an 12 J. A. McIntosh, P. S. Coelho, C. C. Farwell, Z. J. Wang, J. C. Lewis,
activity that is also observed with o-xylene (Scheme S1, ESI†).25
T. R. Brown and F. H. Arnold, Angew. Chem., Int. Ed. Engl., 2013, 52,
9309–9312.
The rate enhancement for KT2/PFC combinations over the
13 S. Kille, F. E. Zilly, J. P. Acevedo and M. T. Reetz, Nat. Chem., 2011, 3,
WT/PFC counterparts arose mainly from increased NADPH
oxidation activity rather than enhanced coupling (Tables 1 and 2).
In addition, the regioselectivity was unchanged when PFCs
738–743.
14 G. Di Nardo and G. Gilardi, Int. J. Mol. Sci., 2012, 13, 15901–15924.
15 O. Shoji, T. Fujishiro, H. Nakajima, M. Kim, S. Nagano, Y. Shiro and
Y. Watanabe, Angew. Chem., Int. Ed. Engl., 2007, 46, 3656–3659.
were added to the WT or KT2. These observations are consis- 16 N. Kawakami, O. Shoji and Y. Watanabe, Angew. Chem., Int. Ed.
Engl., 2011, 50, 5315–5318.
17 F. E. Zilly, J. P. Acevedo, W. Augustyniak, A. Deege, U. W. Hausig and
tent with the catalytically ready structure of the KT2 variant,
pre-disposing it to electron transfer, with little effect on sub-
M. T. Reetz, Angew. Chem., Int. Ed. Engl., 2011, 50, 2720–2724.
strate binding orientation and product selectivity. The PFC 18 Z. Cong, O. Shoji, C. Kasai, N. Kawakami, H. Sugimoto, Y. Shiro and
Y. Watanabe, ACS Catal., 2015, 5, 150–156.
19 C. J. Whitehouse, S. G. Bell, W. Yang, J. A. Yorke, C. F. Blanford,
decoys, on the other hand, fill part of the active site and
constrain the substrate to bind closer to the heme. This is a
A. J. Strong, E. J. Morse, M. Bartlam, Z. Rao and L. L. Wong,
determining factor for enhanced coupling but the PFC does not
interfere with substrate binding close to the heme which deter-
mines the regioselectivity. For substituted benzenes the phenol
ChemBioChem, 2009, 10, 1654–1656.
20 C. J. Whitehouse, W. Yang, J. A. Yorke, B. C. Rowlatt, A. J. Strong,
C. F. Blanford, S. G. Bell, M. Bartlam, L. L. Wong and Z. Rao,
ChemBioChem, 2010, 11, 2549–2556.
product formed is dependent on the carbocation stability during 21 C. J. Whitehouse, W. Yang, J. A. Yorke, H. G. Tufton, L. C. Ogilvie,
S. G. Bell, W. Zhou, M. Bartlam, Z. Rao and L. L. Wong, Dalton
Trans., 2011, 40, 10383–10396.
22 O. Shoji, T. Kunimatsu, N. Kawakami and Y. Watanabe, Angew.
NIH shift rearrangement of the arene oxide intermediates rather
than the proximity of arene C–H bonds to the ferryl oxygen.
In conclusion we have shown that the combination of a rate
accelerating mutant and a decoy molecule is a simple and efficient
method to improve the activity of P450Bm3 without altering the
product distribution. The improvements were greatest for cyclo-
hexane and the KT2/PFC decoy molecule combinations seem to be
well suited for the oxidation of smaller alkanes. This approach
provides additional evidence for multiple binding sites in
Chem., Int. Ed. Engl., 2013, 52, 6606–6610.
23 A. Dennig, N. Lulsdorf, H. Liu and U. Schwaneberg, Angew. Chem.,
Int. Ed. Engl., 2013, 52, 8459–8462.
24 A. Dennig, J. Marienhagen, A. J. Ruff, L. Guddat and
U. Schwaneberg, ChemCatChem, 2012, 4, 771–773.
25 C. J. Whitehouse, N. H. Rees, S. G. Bell and L. L. Wong, Chem. – Eur. J.,
2011, 17, 6862–6868.
26 B. Rowlatt, J. A. Yorke, A. J. Strong, C. J. Whitehouse, S. G. Bell and
L. L. Wong, Protein Cell, 2011, 2, 656–671.
This journal is ©The Royal Society of Chemistry 2016
Chem. Commun., 2016, 52, 1036--1039 | 1039