Bacterial CYP153A monooxygenases for the synthesis of omega-hydroxylated fatty acids

Article abstract of DOI:10.1039/c2cc18103g

CYP153A from Marinobacter aquaeolei has been identified as a fatty acid ω-hydroxylase with a broad substrate range. Two hotspots predicted to influence substrate specificity and selectivity were exchanged. Mutant G307A is 2- to 20-fold more active towards fatty acids than the wild-type. Residue L354 is determinant for the enzyme ω-regioselectivity.

Full text of DOI:10.1039/c2cc18103g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 5115–5117
www.rsc.org/chemcomm
COMMUNICATION
Bacterial CYP153A monooxygenases for the synthesis of
omega-hydroxylated fatty acidsw
Sumire Honda Malca, Daniel Scheps, Lisa Kuhnel, Elena Venegas-Venegas,
¨
Alexander Seifert, Bettina M. Nestl and Bernhard Hauer*
Received 27th December 2011, Accepted 23rd March 2012
DOI: 10.1039/c2cc18103g
CYP153A from Marinobacter aquaeolei has been identified as a
fatty acid x-hydroxylase with a broad substrate range. Two
hotspots predicted to influence substrate specificity and selectivity
were exchanged. Mutant G307A is 2- to 20-fold more active
towards fatty acids than the wild-type. Residue L354 is determinant
for the enzyme x-regioselectivity.
been achieved by strain development, particularly in Yarrowia
5
,13
and Candida hosts.
Recent research includes the synthesis of 174 g l o-OHFA
À1
À1
and 6 g l a,o-DCA from methyl tetradecanoate after 6 days
using an engineered Candida tropicalis strain. Lately, the non-heme
iron alkane monooxygenase (AlkB) from Pseudomonas putida
5
5
GPo1 was demonstrated to o-hydroxylate C –C12 fatty acid
8
methyl esters in vitro and in vivo using E. coli as host.
Bacterial CYP153A monooxygenases are soluble enzymes
that hydroxylate medium-chain (C –C ) alkanes with high
o-Hydroxy fatty acids (o-OHFAs) are valuable chemicals for
adhesives, lubricants, cosmetic intermediates and potential
1
anticancer agents. In addition, o-OHFAs serve as building
5
16
14,15
o-regioselectivity.
and Acinetobacter have been reported to convert primary alcohols
CYP153A enzymes from Mycobacterium
blocks for the synthesis of poly(o-hydroxy fatty acids), which
exhibit similar or superior physicochemical properties than
1
6
2
to a,o-diols, indicating that they are also active towards polar
compounds. With the aim of identifying fatty acid o-hydroxylases
among the CYP153A subfamily, we investigated the in vitro
activities and selectivities of monooxygenases from Polaromonas
sp. (CYP153A P. sp.), Mycobacterium marinum (CYP153A16)
and Marinobacter aquaeolei (CYP153A M. aq.) towards C8:0–C20:0
saturated and 9(Z)/(E)-C14:1–C18:1 monounsaturated fatty acids
polyethylene and other bioplastics. o-OHFAs can be oxidized
to a,o-dicarboxylic acids (a,o-DCAs), which are employed in
the preparation of polymers, coatings, fragrances, antiseptics
3
and polyketide antibiotics. Medium- to long-chain o-OHFAs
4
5,6
and a,o-DCAs are produced by chemical and biological
processes, with the latter being sought for their higher selectivity
and greener approach.
The selective terminal oxygenation of fatty acids occurs
(Materials and methods, ESIw). The results are presented in Fig. 1.
7
,8
Saturated fatty acids were oxidized by CYP153A P. sp. in
very low yields (o5% conversion). In contrast, CYP153A16
and CYP153A M. aq. were considerably more active, displaying
maximum activities towards C13:0 (92% conversion) and C14:0
naturally in mammals, plants and in certain fungi and bacteria.
Most of these reactions are catalyzed by cytochrome P450 mono-
oxygenases (P450s or CYPs). Some bacterial CYPs hypothesized as
9
fatty acid o-hydroxylases have not been characterized yet, while
(83% conversion), respectively. CYP153A16 was active towards
others display poor activity and/or regioselectivity. CYP124A1
from Mycobacterium tuberculosis, a methyl-branched fatty acid
o-hydroxylase, is not very active and o-specific towards linear fatty
C10:0–C16:0 fatty acids, while CYP153A M. aq. had a broader
scope (C9:0–C20:0). The substrate range of CYP153A P. sp. was
1
0
8:0
slightly shifted towards shorter compounds (C –C13:0). Mono-
acids. CYP102A1 (P450-BM3) from Bacillus megaterium and
other isoenzymes engineered for alkane terminal oxidation are not
highly o-regiospecific and were not tested towards fatty acids
unsaturated fatty acids were only oxidized by CYP153A16 and
CYP153A M. aq. In general, CYP153A M. aq. was more active
11,12
either.
towards these compounds, showing preference for 9(E)-C16:1
In contrast, microbial CYPs known to o-oxidize fatty
(93% conversion). Substrate conversions with CYP153A16 were
acids belong to the yeast CYP52 family. These enzymes are already
utilized for the yeast-based production of o-OHFAs and
a,o-DCAs. Significant improvements on product yields have
not higher than 35%. Interestingly, CYP153A16 and CYP153A
M. aq. were more active towards 9(Z)/(E)-C16:1–C18:1 than
towards C16:0–C18:0, but the reason is unknown to us in the
absence of enzyme structural data and substrate affinity studies.
In terms of regioselectivity, CYP153A P. sp. catalyzed
hydroxylations exclusively on the o-position of fatty acids.
CYP153A16 and CYP153A M. aq. preferred this position as
well, but their specificity varied depending on the CYP and
the chain length of the substrate. CYP153A M. aq. was highly
o-regioselective (491% o-OHFA), while CYP153A16 yielded
higher proportions of (o-1)-OHFAs (Table S1, ESIw).
Institute of Technical Biochemistry, University of Stuttgart,
Allmandring 31, 70569 Stuttgart, Germany.
E-mail: bernhard.hauer@itb.uni-stuttgart.de;
Fax: +49 711 68563196; Tel: +49 711 68563193
w Electronic supplementary information (ESI) available: Materials
and methods. Detailed substrate conversions and product distribu-
tions. Gas chromatograms and mass spectra. Michaelis–Menten plots.
See DOI: 10.1039/c2cc18103g
This journal is c The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 5115–5117 5115

View Article Online
Fig. 1 Substrate conversions (blue) and o-regioselectivities (green) of fatty acid oxidation reactions catalyzed by CYP153A monooxygenases
after 4 h. Reaction mixtures contained 3 mM CYP, 15 mM CamA, 30 mM CamB, 0.2 mM substrate, 2% DMSO, 0.2 mM NADH and cofactor
regeneration.
To our surprise, CYP153A16 could also a- and b-hydroxylate
12:0–C14:0 (o3% of the total product; Fig. S1 and S2, ESIw). We
Table 1 Oxidation profiles of CYP153A M. aq. wild-type and
mutants towards 1 mM nonanoic acid after a reaction time of 4 h
C
are unable to explain these events. However, considering that the
a- and b-carbon atoms are situated near the carboxyl group, we
inferred that iron-catalyzed oxygenation on such positions could
occur if the substrate entered the active site with its carboxyl
group coordinated towards the heme group. CYP153A16 and
CYP153A M. aq. converted C12:0–C14:0 and 9(Z)-C14:1 o-OHFAs
into a,o-DCAs, though in low yields (Table S1, ESIw).
Aldo-/ketoacids or epoxy products which might have arisen
from the 9(Z)-monoenoic fatty acids were not detected. We also
observed that using an o-hydroxylated fatty acid as starting
material (o-OH-C12:0) did not lead to higher a,o-DCA yields
Product distribution (%)
Variant
Conversion (%)
(o-1)-OH
o-OH
Wild-type
G307A
G307V
L354I
1.7
26.0
—
2.0
1.2
2.5
1.1
—
97.5
98.9
—
75.6
17.0
24.4
83.0
L354F
Reactions contained 3 mM CYP, 15 mM CamA, 30 mM CamB, 1 mM
substrate, 2% DMSO, 1 mM NADH and cofactor regeneration.
—, not detected.
(
data not shown). From our previous and current results on
M. aq. is a leucine (L354), while in the CYP153A subfamily
additionally isoleucine and valine are observed. In P450-BM3
the substitution of the equivalent position with phenylalanine
alkane, primary alcohol and fatty acid oxidation (Table S2, ESIw),
we observed that CYP153A16 and CYP153A M. aq. behave as
predominantly fatty acid o-hydroxylases, while CYP153A P. sp. as
a predominantly alkane o-hydroxylase.
Next, we identified potential substrate-interacting residues
in CYP153A M. aq. and addressed them by site-directed
mutagenesis (Materials and methods, ESIw) in order to elucidate
their impact on specificity and selectivity. CYP153A M. aq. was
chosen due to its higher regioselectivity and in vitro stability
compared to CYP153A16. Previously, we reported how to identify
19
(A328F) influenced regioselectivity towards alkanes. We chose
isoleucine and phenylalanine as substituents for this position.
The four generated enzyme variants were purified for activity
assays towards C9:0 (Table 1). G307A was more active than
the wild-type enzyme by a factor of 15, while preserving its
o-regioselectivity. In contrast, G307V showed no activity. Variants
L354I/F displayed similar to lower activity than the wild-type, but
their regioselectivities changed significantly. The most dramatic
shift was observed with variant L354I, which produced more
(o-1)-OHFA (76% of the total product). This indicates that the
size and shape of the residue in position 354 strongly influences the
orientation of the substrate close to the heme active site and thus
17
key residues in CYPs from protein structure and sequence. There
we showed that three structural elements contain amino acids
pointing with their side chains towards the heme center and are
hence expected to be in contact with every substrate molecule
during the attack of the activated oxygen. In the present work, we
focused on two hotspot positions which can be easily identified
17
controls where the activated oxygen attacks.
Additional experiments confirmed that mutant G307A was
more active than the wild-type enzyme (Table 2; Fig. S3, ESIw).
In a former study on P450cam, adding mutation G248A to a
multiple-point variant caused an increased activity towards
18
from the protein sequence. These residues are also part of the
active site of a previously published homology model built for
15
CYP153A6. The first hotspot, a conserved alanine which is part
18
20
of the sequence motif AGxxT, is positioned in the I-helix.
small substrates. Such mutation was introduced in the light
Interestingly, a glycine is observed instead of alanine in all
members of the CYP153A subfamily (GGNDT motif). We
decided to replace the first glycine (G307 in CYP153A M. aq.)
with alanine. Valine was also selected as substituent given its
hydrophobic nature and short size. The second hotspot resides in
substrate recognition site 5 (SRS5), in position 5 after the highly
conserved ExxR motif. The corresponding residue in CYP153A
of structural data to force substrates to bind closer to the heme
iron. We noticed that residue G248 in P450cam is equivalent
to G307 in CYP153A M. aq. In our study we show that G307A
is catalytically more efficient than the wild-type owing to an
increased turnover number rather than a higher substrate affinity.
Although the reaction rates are low, it should be considered
that they correspond to enzymes reconstituted with non-natural
5
116 Chem. Commun., 2012, 48, 5115–5117
This journal is c The Royal Society of Chemistry 2012

View Article Online
Table 2 Substrate conversion, selectivity and kinetic constants of
CYP153A M. aq. wild-type and variant G307A towards fatty acids
Notes and references
1
2
3
4
A. Abe and K. Sugiyama, Anti-Cancer Drugs, 2005, 16, 543;
A. V. Rawlings, Int. J. Cosmet. Sci., 2003, 25, 63.
C. Liu, F. Liu, J. Cai, W. Xie, T. E. Long, S. R. Turner, A. Lyons
and R. A. Gross, Biomacromolecules, 2011, 12, 3291.
S. Huf, S. Krugener, T. Hirth, S. Rupp and S. Zibek, Eur. J. Lipid
Sci. Technol., 2011, 113, 548.
L. Cotarca, P. Maggioni and A. Nardelli, U.S. Pat., 5,872,267,
Industrie Chimiche Caffaro, S.p.A., 1999; M. M. S. Stephan and
B. Mohar, Org. Process Res. Dev., 2006, 10, 481; K. Suzuki, T. Eto,
T. Osuka, S. Abe and S. Yoshikawa, U.S. Pat., 4,246,182, Soda
Koryo Kabushiki Kaisha, 1981; T. Yokota and A. Watanabe,
U.S. Pat., 5,191,096, Nippon Mining Co., Ltd., 1992.
W. Lu, J. E. Ness, W. Xie, X. Zhang, J. Minshull and R. A. Gross,
J. Am. Chem. Soc., 2010, 132, 15451.
Fatty acid substrate
Enzyme variant/
a,b
c
parameter
C
8:0
C
9:0
C
14:0
Wild-type
Conversion (%)
o-Selectivity (%)
—
—
1.7
97.5
48.4
97.1
0.036 Æ 0.004
4.3 Æ 0.2
119
K
M
/mM
5.15 Æ 0.04
0.13 Æ 0.01
0.025
0.217 Æ 0.009
0.24 Æ 0.02
1.1
À1
k
k
cat/min
cat/K /min mM
À1
M
G307A
Conversion (%)
o-Selectivity (%)
20.3
98.4
26.0
98.9
68.6
96.8
0.035 Æ 0.004
7.5 Æ 0.5
214
5
K
M
/mM
4.84 Æ 0.36
2.55 Æ 0.26
0.527
0.245 Æ 0.008
4.0 Æ 0.3
16.3
À1
6 J. Oriedo, B. Vincent and M. Subramanian, WO Pat., 2003/020017,
The Dow Chemical Company, 2003; B. Qin, R. Zhang and Y. Qiu,
CN Pat., 101328165 A20081224, Cathay Industrial Biotech, 2008.
7 F. Pinot, FEBS J., 2011, 278, 181; I. N. Van Bogaert, S. Groeneboer,
K. Saerens and W. Soetaert, FEBS J., 2011, 278, 206.
8 M. Schrewe, A. O. Magnusson, C. Willrodt, B. Bu
A. Schmid, Adv. Synth. Catal., 2011, 353, 3485.
9
k
k
cat/min
cat/K /min mM
À1
M
a
Conversions and selectivities were calculated after 4 h with 3 mM
CYP, 15 mM CamA, 30 mM CamB, 1 mM substrate, 2% DMSO,
b
mM NADH and cofactor regeneration. Steady-state kinetic para-
1
meters were determined by GC analysis of substrate conversion. Reaction
¨
hler and
N. M. Broadway, F. M. Dickinson and C. Ratledge, J. Gen.
Microbiol., 1993, 139, 1337; S. Torres, C. R. Fjetland and
P. J. Lammers, BMC Microbiol., 2005, 5.
mixtures contained 0.75–1 mM CYP + CamA + CamB (1 : 5 : 10 ratio),
c
% DMSO, 1 mM NADH and cofactor regeneration. Conversion of
2
C8:0 to the corresponding o-OHFA by the wild-type enzyme was detected
1
0 J. B. Johnston, P. M. Kells, L. M. Podust and P. R. Ortiz de
Montellano, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 20687.
1 E. T. Farinas, U. Schwaneberg, A. Glieder and F. H. Arnold, Adv.
Synth. Catal., 2001, 343, 601; O. Lentz, A. Feenstra, T. Habicher,
B. Hauer, R. D. Schmid and V. B. Urlacher, ChemBioChem, 2006,
7, 345.
with 41 mM substrate. Kinetic constants for C8:0 were thus measured
after adjusting the GC method. —, Not detected.
1
redox partners. Nevertheless, in comparison with other bacterial
CYPs, wild-type CYP153A M. aq. is already more active
1
2 P. Meinhold, M. W. Peters, A. Hartwick, A. R. Hernandez and
F. H. Arnold, Adv. Synth. Catal., 2006, 348, 763.
À1
towards linear fatty acids (3.5 min for C14:0) than CYP124A
À1
1
3 P. Fickers, P. H. Benetti, Y. Wache, A. Marty, S. Mauersberger,
M. S. Smit and J. M. Nicaud, FEMS Yeast Res., 2005, 5, 527;
M. S. Smit, M. M. Mokgoro, E. Setati and J. M. Nicaud,
Biotechnol. Lett., 2005, 27, 859.
10
(
0.07 min for C16:0
)
at equal substrate saturating concentrations
À1
(
5 Â K ). The ability of a P450-BM3 variant (160 min and 52%
M
12
o-selectivity towards octane) to o-hydroxylate fatty acids remains
unknown.
1
4 M. Bordeaux, A. Galarneau, F. Fajula and J. Drone, Angew. Chem.,
Int. Ed., 2011, 50, 2075; M. Kubota, M. Nodate, M. Yasumoto-
Hirose, T. Uchiyama, O. Kagami, Y. Shizuri and N. Misawa, Biosci.,
Biotechnol., Biochem., 2005, 69, 2421; T. Maier, H. H. Forster,
O. Asperger and U. Hahn, Biochem. Biophys. Res. Commun., 2001,
In conclusion, we report the selective terminal oxygenation
of a set of saturated and monounsaturated fatty acids using
three monooxygenases from the bacterial CYP153A subfamily.
CYP153A from Marinobacter aquaeolei appears to be broadly
applicable to the hydroxylation of medium-chain saturated and
long-chain monounsaturated fatty acids. Furthermore, our
mutagenesis study performed by a systematic comparison of
protein sequences allowed us to identify key residues influencing
activity and regioselectivity in the hydroxylation of fatty acids.
We thank P. Scheller for his collaboration in protein
purification. We gratefully acknowledge financial support
from the German Federal Ministry of Education and Research
286, 652; J. B. van Beilen, E. G. Funhoff, A. van Loon, A. Just,
L. Kaysser, M. Bouza, R. Holtackers, M. Rothlisberger, Z. Li and
B. Witholt, Appl. Environ. Microbiol., 2006, 72, 59.
1
5 E. G. Funhoff, U. Bauer, I. Garcia-Rubio, B. Witholt and
J. B. van Beilen, J. Bacteriol., 2006, 188, 5220.
1
6 T. Fujii, T. Narikawa, F. Sumisa, A. Arisawa, K. Takeda and J. Kato,
Biosci., Biotechnol., Biochem., 2006, 70, 1379; N. Fujita, F. Sumisa,
K. Shindo, H. Kabumoto, A. Arisawa, H. Ikenaga and N. Misawa,
Biosci., Biotechnol., Biochem., 2009, 73, 1825; D. Scheps, S. Honda
Malca, H. Hoffmann, B. M. Nestl and B. Hauer, Org. Biomol. Chem.,
2011, 9, 6727.
17 A. Seifert and J. Pleiss, Curr. Drug Metab., 2011, 13, 197.
1
8 J. Mestres, Proteins, 2005, 58, 596.
(
BMBF) in the frame of the ‘‘Systems Biology in Pseudomonas
1
9 E. Weber, A. Seifert, M. Antonovici, C. Geinitz, J. Pleiss and
V. B. Urlacher, Chem. Commun., 2011, 47, 944.
0 F. Xu, S. G. Bell, J. Lednik, A. Insley, Z. H. Rao and L. L. Wong,
Angew. Chem., Int. Ed., 2005, 44, 4029.
for Industrial Biocatalysis’’ project as well as the European
Union’s 7th Framework Programme FP7/2007-2013 under
grant agreement no. 266025.
2
This journal is c The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 5115–5117 5117