Deazaflavins as mediators in light-driven cytochrome P450 catalyzed hydroxylations

Article abstract of DOI:10.1039/b913863c

A light-driven deazaflavin-dependent direct enzyme regeneration system has been developed for a P450-BM3 catalyzed CH-activating hydroxylation, thereby avoiding the need for the expensive NADPH cofactor. The Royal Society of Chemistry 2009.

Full text of DOI:10.1039/b913863c

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Deazaflavins as mediators in light-driven cytochrome P450 catalyzed
hydroxylationsw
a
a
a
b
Felipe E. Zilly,z Andreas Taglieber,z Frank Schulz, Frank Hollmann and
a
Manfred T. Reetz*
Received (in Cambridge, UK) 10th July 2009, Accepted 21st September 2009
First published as an Advance Article on the web 14th October 2009
DOI: 10.1039/b913863c
2
0
A light-driven deazaflavin-dependent direct enzyme regeneration
system has been developed for P450-BM3 catalyzed
decomposition leads to formation of side products. The
presence of oxygen, which is obviously required for a mono-
oxygenase, was suspected to be a major limitation of this
a
CH-activating hydroxylation, thereby avoiding the need for
the expensive NADPH cofactor.
18
system, leading to the decoupling of the regeneration reaction
from the enzymatic reaction due to the rapid reaction of
2
reduced flavin species with molecular oxygen (Scheme 1).
1
1
,2
The P450 monooxygenase BM3 from Bacillus megaterium,
Such a decoupling results in a waste of reduction equivalents,
inefficient electron transfer and the generation of reactive
oxygen species such as hydrogen peroxide which can be
detrimental to enzyme activity.
like other members of the P450 family, catalyzes the mono-
oxygenation of non-activated substrates with concomitant
consumption of NADPH and molecular oxygen. Unlike other
protein family members, it represents a fusion protein
comprising a monooxygenase and a reductase domain. The
diflavin-reductase domain is reduced by NADPH. Subsequent
electron transfer to the heme domain of the monooxygenase
domain allows for initiation of a catalytic cycle. This domain
fusion ensures a highly concerted catalytic activity (up to
In order to avoid the decoupling reaction, we recently
1
8
proposed the use of deazaflavins, since reduced deazaflavins
2
2
were reported to react very slowly with molecular oxygen.
Furthermore, it has been demonstrated in mechanistic studies
2
and sensor applications that rat liver P450 reductase and
3
24,25
rabbit liver P450-2B4 monooxygenase
ꢀ
1
3
can undergo reduction
1
7 100 min
for hydroxylation of arachidonic acid) and
in the presence of soluble flavins, sacrificial electron donors
and light in an oxygen-free system. However, to the best of our
knowledge this type of electron transfer has never been used to
achieve catalysis, i.e., product formation with P450 mono-
oxygenases. By choosing deazaflavins as oxygen-stable
electron mediators, we hoped to achieve robust and efficient
light-driven direct regeneration of a P450 monooxygenase for
the first time.
makes the protein an interesting candidate for industrial
4,5
applications. The requirement of the unstable and prohibitively
expensive cofactor NADPH, however, hinders broad application.
6
–9
Hence, enzyme coupled cofactor recycling systems or whole
1
0–12
cell catalysis
are used to achieve regeneration of NADPH.
Both approaches have their respective drawbacks such as
undesired substrate/product oxidation and limited substrate
1
3,14
accessibility due to membrane impermeability.
approaches utilize artificial electron mediators and aim to
Other
As the model system, we chose the P450-BM3 catalyzed
oxidation of lauric acid. Reduction of the heme domain
represents the first step of the catalytic cycle in P450 mono-
oxygenase catalysis. Therefore, this step was studied in
exploratory experiments by shining light from a 100 Watt
lamp onto aerobic samples comprising purified BM3-WT,
EDTA and various types of soluble flavins. Selected samples
1
5–17
directly regenerate P450-BM3.
Herein, we report an
alternative approach based on the use of deazaflavins in a
light-driven process.
Recently, we described a light-driven direct cofactor
regeneration system for the flavin-dependent Baeyer–Villiger
monooxygenase PAMO and the enoate-reductase YqjM,
deriving reducing equivalents from the inexpensive sacrificial
1
8,19
electron donor EDTA.
In this system flavin mediators
such as FAD, FMN or riboflavin are excited by irradiation
with visible light which allows for reaction with the sacrificial
electron donor to generate a reduced flavin mediator. Subsequent
electron transfer to the enzyme-bound flavin cofactor
initiates enzyme catalysis (Scheme 1). In this process EDTA
a
Max-Planck-Institut fu¨r Kohlenforschung, Kaiser-Wilhelm-Platz 1,
Mu¨lheim an der Ruhr, Germany. E-mail: reetz@mpi-muelheim.mpg;
Fax: +49-208-306-2985; Tel: +49-208-306-2000
Delft University of Technology, Julianalaan 136, Delft,
the Netherlands. E-mail: F.Hollmann@tudelft.nl;
Fax: +31-15-278-1415; Tel: +31-15-278-195
w Electronic supplementary information (ESI) available: Experimental
details and additional control experiments. See DOI: 10.1039/b913863c
z These authors contributed equally to this paper.
Scheme 1 Mechanism of the light-driven P450-BM3 catalyzed
hydroxylation (A) and structure of the deazaflavin mediator, 3,10-
dimethyl-5-deazaflavin (B). The decoupling of the regeneration
reaction from the enzymatic reaction can be largely prevented by
using a deazaflavin mediator and therefore the formation of reactive
b
21
2 2
oxygen species such as H O (red pathway) is essentially suppressed.
7
152 | Chem. Commun., 2009, 7152–7154
This journal is ꢁc The Royal Society of Chemistry 2009

View Article Online
+
hydroxylation, even if supplemented with NADP . In order
to test whether the rather unstable P450-BM3 reductase
domain is required for the light-driven reaction, we performed
corresponding experiments with the isolated P450-BM3 heme
domain. Under our set of conditions no product formation
was observed. In another system based on photoexcitation
28
reduction of the heme was achieved under anaerobic conditions.
Analysis of the product distribution of hydroxylated lauric
acid, achieved when using full-length P450-BM3, was found to
be essentially the same as in the normal reaction. This shows
that regioselectivity is not altered under the conditions of the
light-driven recycling system (see ESIw Table S1).
Fig. 1 Formation of the characteristic 450 nm absorption band in
the CO difference spectra obtained by light-driven reduction of the
P450-BM3 heme domain.
Next, the kinetics of the light-driven hydroxylation of lauric
acid were investigated (Fig. 3). The use of NADPH in a con-
ventional process leads to a quick and complete hydroxylation
of lauric acid within one hour. The light-driven reaction
proceeds at a significantly lower rate (only 12% conversion
after one hour). Enzyme stability was apparently not impaired
since product formation steadily continued for 17 h, reaching
74% conversion. Increasing the biocatalyst concentration
did not significantly alter the course of the reaction (data
not shown), indicating that the light-driven regeneration is
rate limiting.
+
were further spiked with NADP . The successful reduction
of the heme domain was monitored by the characteristic
II
Fe (heme)(CO) absorption band at 450 nm in the CO difference
spectra. A reduction of the heme can be monitored by the
appearance of the 450 nm absorption band after purging the
2
6
sample with CO. CO difference spectra show that under
aerobic conditions a pronounced reduction of the P450-BM3
heme domain is achieved only in the presence of deazaflavin
+
with additional NADP or under positive control conditions
ꢀ1
(
NADPH). In contrast to this, FAD as a mediator failed to
The turnover frequency (TOF) was estimated to be 117 h
yield notable reduction of the heme domain.
(Table 1). This rate is about 2000 times lower than that
observed for P450-BM3 using NADPH as a cofactor
This result corroborates our hypothesis of superior electron-
mediator properties of deazaflavins compared to flavins of the
isoalloxazine-type (FAD, FMN, riboflavin) under aerobic
conditions (Fig. 1).
ꢀ
1 3
(B5000 min ). However, it is in the same range as the
previously described catalytic rate of eukaryotic P450s.
3
5
Among the light-driven regeneration systems the deazaflavin
system shows the highest turnover number (TON) and is in the
range of the best reported direct regeneration systems for
cytochrome P450 enzymes (Table 1). The TOF is in the same
order of magnitude as the oxygen independent YqjM system
and about 10 fold higher than for using sodium dithionite as
reducing agent. Although the present deazaflavin-based system
is a very efficient direct regeneration system, so far it
cannot compete with well-studied enzymatic NADPH-recycling
systems. In contrast to the light-driven regeneration, in many
other cases of direct regeneration of monooxygenases, the
turnover frequency observed for the enzyme exceeds the total
turnover number. This suggests a low biocatalyst stability in
these systems due to the unnatural reaction conditions. The
light-driven regeneration approach also has the advantage
that it only uses inexpensive and environmentally benign
To study if the light-driven reduction of P450-BM3
can drive catalytic oxyfunctionalizations, we performed the
reaction in the presence of lauric acid, a typical BM3-WT
2
7
substrate, and analyzed the amount of product formed after
7 h (Fig. 2). As a control, the deazaflavin was omitted or
1
replaced by FAD under otherwise identical conditions. Some
+
samples were supplemented with NADP , since this has
previously been shown to improve PAMO catalyzed light-
1
8,19
driven reactions.
As expected, in the absence of light
NADPH leads to almost complete conversion of lauric acid.
Deazaflavin shows in the corresponding experiment a strong
+
and light-dependent conversion if NADP is present as well.
No substantial hydroxylation is observed in the presence of
+
deazaflavin or NADP alone or in the absence of flavins.
Furthermore, FAD as an electron mediator does not lead to
Fig. 3 Time-course of light-driven hydroxylation of lauric acid by
Fig. 2 Light-driven hydroxylation of lauric acid by P450-BM3 under
aerobic conditions for 17 h. Light-driven reactions: n = 3, error bars
correspond to the standard deviation; dark reactions: n = 1.
P450-BM3. The experiment was performed as described in Fig. 2.
+
+
’: NADPH, K: Deazaflavin + NADP , m: FAD + NADP . All
n = 3. The error bars correspond to the standard deviation.
This journal is ꢁc The Royal Society of Chemistry 2009
Chem. Commun., 2009, 7152–7154 | 7153

View Article Online
Table 1 Comparison of direct flavo-enzyme regeneration systems
TON
ꢀ1
Enzyme, reference
Mediator
Source of reducing equivalents
TOF [h
]
Enzyme
Mediator
1
8
PAMO-P3
FMN, hn
EDTA
EDTA
EDTA
Cathode
Cathode
Cathode
Dithionite
NADPH
10
96
9.6
10
7.4
0.2
n/a
n/a
n/a
n/a
1
8,29
YqjM
P450-BM3 (this study)
FMN, hn
194
117
104
984
2268
17.4
4866
383
698
26
224
835
Deazaflavin, hn
FAD
Cobaltocene cation
Cobalt (III) sepulchrate
Dithionite
30
StyA
1
1
3
3
5
5
1
2
a
P450-BM3
P450-BM3
P450-BM3
P450-BM3
a
a
a
n/a
n/a
a
a
NADPH
a
33,34
n/a = not available. Further direct regeneration systems have been reported for other P450 monooxygenases.
components in a simple setup. It should be noted that the
present system was intended for a proof-of-principle and that
an optimization of the reaction conditions could lead to an
even more efficient setup.
7 R. Wichmann and D. Vasic-Racki, Adv. Biochem. Eng. Biotechnol.,
005, 92, 225–260.
H. Zhao and W. A. van der Donk, Curr. Opin. Biotechnol., 2003,
4, 583–589.
9 D. Holtmann and J. Schrader, in Modern Biooxidation, ed.
R. Schmid and V. B. Urlacher, Wiley-VCH, Weinheim, 2007.
10 H. Schewe, B. A. Kaup and J. Schrader, Appl. Microbiol. Biotechnol.,
2
8
1
Our experiments show that only the deazaflavin mediator in
+
the presence of catalytic amounts of NADP with respect to
2008, 78, 55–65.
the substrate is capable of mediating productively the light-
driven recycling pathway. The fact that the presence of
+
1
1
1 S. Schneider, M. G. Wubbolts, D. Sanglard and B. Witholt,
Appl. Environ. Microbiol., 1998, 64, 3784–3790.
2 U. Schwaneberg, C. Otey, P. C. Cirino, E. Farinas and
F. H. Arnold, J. Biomol. Screening, 2001, 6, 111–117.
3 R. R. Chen, Appl. Microbiol. Biotechnol., 2007, 74, 730–738.
4 Y. Ni and R. R. Chen, Biotechnol. Bioeng., 2004, 87,
804–811.
NADP is essential might be explained by NADP-induced
structural alterations necessary for catalysis, as previously
1
suspected for the Baeyer–Villiger monooxygenase PAMO.
1
1
8
A putative indirect pathway via the reduced nicotinamide
1
1
5 A. K. Udit, F. H. Arnold and H. B. Gray, J. Inorg. Biochem., 2004,
cofactor was previously excluded since the presence of
+
NADP was not necessary for the light-driven regeneration
9
6 F. Hollmann and A. Schmid, J. Inorg. Biochem., 2009, 103,
8, 1547–1550.
1
8
of the reductase YqjM. Furthermore, control experiments
using an alcohol dehydrogenase-based detection for NADPH
exclude this possibility as well (see ESIw Fig. S1).
313–315.
17 U. Schwaneberg, D. Appel, J. Schmitt and R. D. Schmid,
J. Biotechnol., 2000, 84, 249–257.
8 A. Taglieber, F. Schulz, F. Hollmann, M. Rusek and M. T. Reetz,
1
In summary, we report the first light-driven cytochrome
P450 catalyzed monoxygenation using a soluble deazaflavin
mediator in combination with an inexpensive sacrificial electron
donor. By using deazaflavins as electron mediators, instead of
normal flavins, a light-driven regeneration of flavin-dependent
enzymes can also proceed efficiently under aerobic conditions,
possibly due to suppression of the decoupling of the regeneration
reaction from the enzymatic reaction (Scheme 1). Furthermore,
the unnatural reaction conditions do not alter the regioselectivity
of the P450 catalyzed hydroxylation. The light-driven
regeneration system presented here requires only a simple
setup without additional regeneration enzyme and appears
to be generally applicable to flavin-dependent enzymes.
ChemBioChem, 2008, 9, 565–572.
19 F. Hollmann, A. Taglieber, F. Schulz and M. T. Reetz, Angew.
Chem., Int. Ed., 2007, 46, 2903–2906.
2
0 W. R. Frisell, C. W. Chung and C. G. Mackenzie, J. Biol. Chem.,
959, 234, 1297–1302.
21 V. Massey, J. Biol. Chem., 1994, 269, 22459–22462.
1
22 V. Massey and P. Hemmerich, Biochemistry, 1978, 17, 9–16.
2
3 J. L. Vermilion and M. J. Coon, J. Biol. Chem., 1978, 253,
812–8819.
8
2
4 J. A. Peterson, R. E. White, Y. Yasukochi, M. L. Coomes,
D. H. O’Keeffe, R. E. Ebel, B. S. Masters, D. P. Ballou and
M. J. Coon, J. Biol. Chem., 1977, 252, 4431–4434.
2
5 V. V. Shumyantseva, T. V. Bulko, R. D. Schmid and
A. I. Archakov, Biosens. Bioelectron., 2002, 17, 233–238.
6 T. Omura and R. Sato, J. Biol. Chem., 1964, 239, 2370–2378.
2
27 S. S. Boddupalli, R. W. Estabrook and J. A. Peterson, J. Biol.
Chem., 1990, 265, 4233–4239.
We thank Dina Klu
¨
tt for the synthesis of the deazaflavin
usig for help with GC
2
8 H. M. Girvan, D. J. Heyes, N. S. Scrutton and A. W. Munro,
J. Am. Chem. Soc., 2007, 129, 6647–6653.
¨
29 A. Taglieber, PhD thesis, Ruhr-Universitat Bochum, Bochum,
and Frank Kohler and Ulrich Ha
¨
analysis.
2
007.
0 F. Hollmann, K. Hofstetter, T. Habicher, B. Hauer and
A. Schmid, J. Am. Chem. Soc., 2005, 127, 6540–6541.
31 X. Fang and J. R. Halpert, Drug Metab. Dispos., 1996, 24,
1282–1285.
3
Notes and references
1
2
3
L. O. Narhi and A. J. Fulco, J. Biol. Chem., 1982, 257, 2147–2150.
L. P. Wen and A. J. Fulco, J. Biol. Chem., 1987, 262, 6676–6682.
M. A. Noble, C. S. Miles, S. K. Chapman, D. A. Lysek,
A. C. MacKay, G. A. Reid, R. P. Hanzlik and A. W. Munro,
Biochem. J., 1999, 339(Pt 2), 371–379.
32 L. A. Cowart, J. R. Falck and J. H. Capdevila, Arch. Biochem.
Biophys., 2001, 387, 117–124.
33 K. M. Faulkner, M. S. Shet, C. W. Fisher and R. W. Estabrook,
Proc. Natl. Acad. Sci. U. S. A., 1995, 92, 7705–7709.
34 M. S. Shet, K. M. Faulkner, P. L. Holmans, C. W. Fisher and
R. W. Estabrook, Arch. Biochem. Biophys., 1995, 318, 314–321.
35 C. Helvig and J. H. Capdevila, Biochemistry, 2000, 39,
5196–5205.
4
5
V. B. Urlacher and S. Eiben, Trends Biotechnol., 2006, 24, 324–330.
M. K. Julsing, S. Cornelissen, B. Buhler and A. Schmid,
Curr. Opin. Chem. Biol., 2008, 12, 177–186.
6
S. C. Maurer, H. Schulze, R. D. Schmid and V. B. Urlacher,
Adv. Synth. Catal., 2003, 345, 802–810.
7
154 | Chem. Commun., 2009, 7152–7154
This journal is ꢁc The Royal Society of Chemistry 2009