Immobilisation of P450 BM-3 and an NADP+ Cofactor Recycling System: Towards a Technical Application of Heme-Containing Monooxygenases in Fine Chemical Synthesis

Article abstract of DOI:10.1002/adsc.200303021

Cytochrome P450 monooxygenases are potentially a very useful class of hydroxylation catalysts; they are able to introduce oxygen at activated and non-activated carbon-hydrogen bonds and thus lead to regio- and/or stereochemically pure compounds. However, this potential is lowered by their intrinsic low activity and inherent instability. P450-catalysed biotransformations require a constant supply of NAD(P)H, making the process an expensive one. To render these catalysts more suitable for industrial biocatalysis, the immobilisation of P450 BM-3 (CYP 102A1) from Bacillus megaterium in a sol-gel matrix was combined with a cofactor recycling system based on NADP⁺-dependent formate dehydrogenase (EC 1.2.1.2) from Pseudomonas sp. 101 and tested for practical applicability. This approach was used for the conversion of β-ionone, octane and naphthalene to the respective hydroxy compounds with DMSO as cosolvent using sol-gel immobilised P450 BM-3 mutants.

Full text of DOI:10.1002/adsc.200303021

FULL PAPERS
‡
Immobilisation of P450 BM-3 and an NADP Cofactor
Recycling System:Towards a Technical Application of Heme-
Containing Monooxygenases in Fine Chemical Synthesis
Steffen C. Maurer, Holger Schulze, Rolf D. Schmid, Vlada Urlacher*
Institute of Technical Biochemistry, Stuttgart University, Allmandring 31, 70569 Stuttgart, Germany
Fax: ( ‡ 49)-711-685-3196, e-mail: itbvkha@po.uni-stuttgart.de
Received: January27, 2003; Accepted: April 5, 2003
Abstract: Cytochrome P450 monooxygenases are sol-gel matrix was combined with a cofactor recycling
‡
potentially a very useful class of hydroxylation system based on NADP -dependent formate dehy-
catalysts; they are able to introduce oxygen at drogenase (EC 1.2.1.2) from Pseudomonas sp. 101
activated and non-activated carbon-hydrogen bonds and tested for practical applicability. This approach
and thus lead to regio- and/or stereochemicallypure
compounds. However, this potential is lowered by naphthalene to the respective hydroxy compounds
their intrinsic low activityand inherent instabilit.y with DMSO as cosolvent using sol-gel immobilised
was used for the conversion of b-ionone, octane and
P450-catalysed biotransformations require a constant P450 BM-3 mutants.
supplyof NAD(P)H, making the process an expen-
sive one. To render these catalysts more suitable for Keywords: biotransformation; cofactor-recycling; cy-
industrial biocatalysis, the immobilisation of P450 tochrome P450 BM-3; hydroxylation; immobilisation;
BM-3 (CYP 102A1) from Bacillus megaterium in a sol-gel
Abbreviations: CYP: cytochrome P450; FDH: for-
mate dehydrogenase; GC: gas chromatography; KPi:
potassium phosphate; NAD(P)H: nicotinamide ad-
enine dinucleotide (phosphate); 10-pNCA: p-nitro-
phenoxydecanoic acid; TEOS: tetraethoxy orthosili-
cate
Introduction
composed of a heme-containing monooxygenase do-
main and an FAD- and FMN-containing reductase
Cytochromes P450 (CYPs) are heme-containing en- domain.[4,5] This multi-domain architecture renders
zymes catalysing a broad range of monooxygenation P450 BM-3 catalytically self-sufficient in terms of
reactions in catabolic and anabolic pathways.[ Several electron transfer from the cofactor NADPH to the
1]
groups have recentlystarted to investigate c yt ochrome
heme iron. The turnover frequencies determined for
À1
P450 monooxygenases for their potential use in bio- P450 BM-3 (>1000 min ) are, to date, among the
transformations.[ CYP P450 s catalyse the hydroxyla- highest reported for a P450 monooxygenase. The
tion of a wide range of hydrophobic compounds at natural substrates of P450 BM-3 are long-chain fatty
activated and non-activated carbons. Thereby, one acids (C to C ), which are exclusivelyhydroxylated at
2,3]
[6]
1
2
20
oxygen atom of O is transferred to the substrate. The the subterminal positions w-1, w-2 and w-3, partially
2
[
6,7]
other oxygen atom is reduced to H O with the simulta- with high enantioselectivity.
P450 BM-3 has fre-
neous oxidation of nicotinamide adenine dinucleotide quentlybeen investigated using rational protein design
phosphate) (NAD(P)H). These hydroxylations are in and/or directed evolution aimed at developing a more
2
(
most cases regio- and stereospecific. As regio- and suitable catalyst. Artificial mutants of P450 BM-3 also
stereospecific oxidation of aliphatic carbons is a reac- hydroxylate non-natural substrates like short-chain
[
8,9]
[10]
tion which is difficult to perform bymeans of classical
fattyacids (C to C ), indole, polycyclic aromatic
8 10
[
11]
[2,12]
[13]
organic synthesis, the biocatalysis by monooxygenases is hydrocarbons, alkanes
and styrenes.
a promising option, especiallywith regard to fine
chemistry.
P450 BM-3, which was originallycloned from Bacillus operational and storage stability and simple product
megaterium, is a 119 kDa natural fusion protein separation. To date, this subject has received verylittle
The immobilisation of enzymes has proved very useful
in in vitro biotransformations as this leads to enhanced
[
14]
802
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI:10.1002/adsc.200303021
Adv. Synth. Catal. 2003, 345, 802 ± 810

‡
Immobilisation of P450 BM-3 and an NADP Cofactor Recycling System
FULL PAPERS
attention.[15±17] Therefore, the immobilisation of two of formate anions to carbon dioxide with concomitant
‡
P450 BM-3 mutants was examined on a broad varietyof
reduction of NAD to NADH and thus is of consid-
common immobilisation matrices. The best result was erable commercial interest as a catalyst for the regen-
achieved byencapsulation of the enz ym e in a sol-gel
eration of reduced coenzymes in fine chemical synthesis.
The great advantage of FDH is the low cost of the
matrix derived from tetraethoxyorthosilicate
18]
(
TEOS).[ This approach resulted in a biocatalyst that substrate formate and the simple removal of the
revealed long-term activitywhen immobilised on the
matrix.
reaction product CO . An FDH mutant also exhibits
high activitywith NADP and therefore can be used as
2
‡
A keyproblem preventing the use of P450 s in
biotechnologyis the supplyof electrons to the heme
iron. In vivo, the electrons are supplied byNAD(P)H,
which is however far too expensive to be used in
an NADPH-regenerating enzyme in combination with
[
29,30]
P450 BM-3 (Figure 1).
[
19]
technical applications. In addition, the use of whole Results and Discussion
cells in industrial transformation reactions is impeded
bythe complicated recoveryof the products, even
Immobilisation of P450 BM-3
though considerable progress has been made in the field
of in situ product removal techniques.[20] Therefore, an in
vitro reaction process would be highlydesirable if a cost
To our knowledge, to date no systematic investigation
on immobilisation of bacterial P450 monooxygenases
efficient wayof supplying electrons to the heme iron can
[
25]
be found. The use of electron mediators , direct has been published.
[
26,27]
electron supplyfrom electrodes
as well as enzymat-
The immobilisation procedure involved the highly
[15,17]
[28]
ic
suggested as sources of the required reduction equiv- hydroxylates a broad range of substrates.
alents. According to Taylor et al. a bacterial P450 can be
and organometallic
approaches have been purified P450 BM-3 mutant A74G, F87V, L188Q which
[
12]
[
17]
In our report we suggest an enzymatic cofactor immobilised on anion exchange materials. The elec-
‡
recycling system based on NAD -dependent FDH trostatic interactions built up byanion exchangers like
from Pseudomonas sp. 101. FDH catalyses the oxidation DEAE and SuperQ render them quite suitable for
immobilisation of the negativelycharged protein. In our
investigations this could also be achieved with P450 BM-
3
. However, as expected, the products of 10-pNCA
oxidation and other hydroxylation reactions (data not
shown) were also adsorbed to the matrix. This renders
DEAE and SuperQ unsuitable for our purposes. In
addition, washes with higher ionic strength of course
leached the enzyme from the carrier.
A varietyof further matrices was tested using the
respective standard immobilisation protocols (Table 1).
Immobilisation procedures based on hydrophobic in-
teractions proved unsuitable for efficient immobilisa-
tion of P450 BM-3, as judged from the throughout
negative results obtained with EP100 and MP1000
(
polypropylene derivatives), phenyl-, octyl- and butyl-
sepharose. Also, the enzyme was virtually unable to
adsorb on celite, a porous silicate matrix derived from
diatomaceous soil. The covalent attachment to Euper-
Figure 1. Cofactor recycling system for P450 BM-3 with
‡
NADP -dependent FDH.
Table 1. Immobilisation matrices under investigation; references include standard immobilisation protocols. Identical
procedures were used for immobilisation on phenyl-, octyl- and butylsepharose.
Immobi- EP100[22] MP1000[23] Phenylse- Octylse- Butylse- DEAE[17] SuperQ[17] Celite[21] Euper-
Sol-Gel[18]
pharose[ phaose
24]
pharose
git
[31]
lisation
matrix
Immobi- ±
lisation
result
±
±
±
±
‡
‡
±
±
‡
‡
: Immobilisation was observed; ±: no immobilisation could be observed
Adv. Synth. Catal. 2003, 345, 802 ± 810
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
803

FULL PAPERS
Steffen C. Maurer et al.
git, an oxirane-functionalised polymer, worked for
manyproteins, [ but not for active P450 BM-3.
Sol-gel entrapment was shown to be suitable for a
31]
[
18]
broad range of labile biological macromolecules, but
to date no P450 monooxygenase was immobilised in this
kind of matrix. Therefore, the encapsulation of the
monooxygenase in a sol-gel matrix derived from TEOS
was assayed. First tests were carried out using a slightly
modified version of the protocol proposed byShtelzer
and co-workers.[ Monooxygenase activity was ob-
served indicating entrapment of P450 BM-3 (up to
32]
7
.2 mg corresponding to 6.4 U P450 BM-3 in 1 g sol-gel).
During immobilisation about half of the hydroxylation
activitywas lost (see Figure 2). Repeated washing steps
with 50 mM potassium phosphate (KPi) buffer lead to a
negligible loss of P450 BM-3 (about 2% of total P450
BM-3). Further washing did not lead to leaching of any
P450 BM-3.
Figure 2. Turnover of 0.25 mM 10-pNCA byP450 BM-3
monitored byabsorption measurement at 410 nm. Solid line:
0
.061 nmol/mL (0.0125 U/mL, specific activity: 1.7 U/mg) of
freshlypurified P450 BM-3 in solution (continuous in situ
measurement). Squares/dotted line: 0.061 nmol/mL
formation. Remarkably, we did not observe loss of the (0.0065 U/ml) of sol-gel immobilised P450 BM-3 (measured
A large amount of ethanol, which is expected to
[
33]
inactivate P450 BM-3, is released during the sol-gel
enzyme×s activity.
at particular time intervals after removal of the matrix).
Values displayed in this and the following figures originate
from measurements made at least three times. Standard
deviations were determined to be lower than 5% in all cases.
These standard deviations are given as error bars for the
individual data points. All curves shown in this work were
generated byleast-squares fitting of the experimental data
points.
Activity and Stability of Sol-Gel Immobilised P450
BM-3 A74G, F87V, L188Q
The oxidation of p-nitrophenoxydecanoic acid (10-
pNCA) was chosen as a model reaction, because it
allows the simple photometrical measurement of p-
[
34]
nitrophenolate which is produced during the reaction.
Upon its addition to the reaction mixtures, the (Figure 3). According to Figure 3, the half-life of sol-gel
immobilised enzyme forms a turbid suspension. There- immobilised P450 BM-3 is even longer than that of the
fore, direct kinetic measurements are not feasible for glycerol-stabilised one. Estimates of half-lives were
determining the activityof encapsulated P450 BM-3. To
circumvent this, the activityof sol-gel immobilised P450
BM-3 was assayed by incubating all the components of
the standard pNCA assaywith the respective immobi-
calculated assuming a first order exponential decayof
enzymatic activity.
The immobilised enzyme was remarkably stable also
at 258C. Half-lives of 29 days for the immobilised
lised enzyme preparations. Subsequent centrifugation enzyme, 2 days for the enzyme dissolved in 50 mM KPi
separated the yellow reaction product, p-nitropheno- and 5 days for the enzyme dissolved in 50 mM KPi with
late, from the immobilised catalyst. The absorption of 50% glycerol added were found (Figure 4). The stability
the supernatant was measured at 410 nm.
of the immobilised enzyme at 258C is crucial for its
Figure 2 illustrates the hydroxylation reaction of 10- application in a future P450 BM-3 bioreactor, because
pNCA using 0.061 nmol of P450 BM-3 in solution and the enzyme is most active in this temperature range (see
0
.061 nmol P450 BM-3 encapsulated in the sol-gel next section).
matrix. The specific activityof immobilised enz my e To determine the activityof the immobilised P450
0.89 U/mg P450 BM-3) was lower compared to that BM-3 as a function of temperature, the reaction was
observed for the free enzyme (1.7 U/mg P450 BM-3). studied at temperatures ranging from 48C to 508C
This maybe due to diffusion limitation in the nano- (Figure 5). Both the free and the immobilised enzyme
porous catalyst particles or to enzyme denaturation, but performed best at about 258C. Below 208C the immobi-
(
is most likelydue to the combination of both effects.
Investigations on long-term storage stabilityrevealed
lised enzyme was less active than P450 BM-3 in solution,
whichisprobablyduetorestricteddiffusion. Ontheother
advantages of the immobilised form of P450 BM-3. At hand, sol-gel immobilised P450 BM-3 seems to retain its
4
3
8C, no loss of activitywas observed during a period of
6 days. In contrast, the free enzyme has a half-life of
activitybetter at temperatures above 30 8C.
only26 da ys when stored in 50 mM KPi. The dissolved
enzyme could be stabilised by addition of 50% glycerol
804
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 802 ± 810

‡
Immobilisation of P450 BM-3 and an NADP Cofactor Recycling System
FULL PAPERS
Figure 3. P450 BM-3 activityat 4 8C. Squares/solid line: Sol-gel immobilised enzyme (measured at particular time intervals
after removal of the matrix). 100% activitycorrespond to 0.008 U/mL (0.075 nmol/mL of sol-gel immobilised P450 BM-3).
Circles/dotted line: Enzyme dissolved in 50% glycerol (measured at particular time intervals to allow comparison with sol-gel
immobilised enzyme). 100% activity correspond to 0.009 U/mL (0.044 nmol/mL of free P450 BM-3). Triangles/dashed line:
Enzyme dissolved in 50 mM KPi (measured at particular time intervals to allow comparison with sol-gel immobilised enzyme)
100% activitycorrespond to 0.009 U/mL (0.044 nmol/mL of free P450 BM-3).
Figure 4. P450 BM-3 activityat 25 8C. Squares/solid line: Sol-
gel immobilised enzyme (measured at particular time inter-
vals after removal of the matrix). 100% activitycorrespond to
Figure 5. Temperature profile of P450 BM-3 activity. Prior to
the experiments, the enzyme batches were incubated at the
respective temperature for 10 minutes. Squares/dotted line:
Sol-gel immobilised enzyme. 100% activity correspond to
0.008 U/mL (0.075 nmol/mL of sol-gel immobilised P450
BM-3). Circles/solid line: Enzyme dissolved in 50 mM KPi.
100% activitycorrespond to 0.009 U/mL (0.044 nmol/mL of
free P450 BM-3).
0
.008 U/mL (0.075 nmol/mL of sol-gel immobilised P450
BM-3). Circles/dotted line: Enzyme dissolved in 50% glycerol
measured at particular time intervals to allow comparison
(
with sol-gel immobilised enzyme). 100% activity correspond
to 0.009 U/mL (0.044 nmol/mL of free P450 BM-3). Trian-
gles/dashed line: Enzyme dissolved in 50 mM KPi (measured
at particular time intervals to allow comparison with sol-gel
immobilised enzyme). 100% activity correspond to 0.009 U/
mL (0.044 nmol/mL of free P450 BM-3).
Adv. Synth. Catal. 2003, 345, 802 ± 810
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
805

FULL PAPERS
Steffen C. Maurer et al.
Influence of Acetone and DMSO
Solubilisation of the mostly hydrophobic monooxyge-
nase substrates requires at least small amounts of
organic solvents. Previous investigations showed that
acetone and DMSO are the most suitable solvents for
this purpose.[
9,11,12,35]
Hence the activityof immobilised
and free P450 BM-3 was compared using acetone and
DMSO. DMSO is the more favourable co-solvent
because it has a less deleterious effect on hydroxylation
activity(Figures 6 and 7), this seems to be the case in
particular for the immobilised enzyme (Figure 6).
Figure 6. P450 BM-3 activityin presence of DMSO. Squares/
solid line: Sol-gel immobilised enzyme. 100% activity corre-
Turnover of b-Ionone by Sol-gel Immobilised P450
BM-3 A74E, F87V, P386S; Hydroxylation of n-Octane spond to 0.008 U/mL (0.075 nmol/mL of sol-gel immobilised
and Naphthalene by P450 BM-3 A74G, F87V, L188Q
P450 BM-3). Circles/dotted line: Enzyme dissolved in 50 mM
KPi. 100% activitycorrespond to 0.009 U/mL (0.044 nmol/
mL of free P450 BM-3).
To test the general applicabilityof the immobilisation
approach, the hydroxylation of three further substrates
beside 10-pNCA was investigated. The conversion of b-
[
36]
ionone into 4-hydroxy-b-ionone
(Figure 8) bythe
P450 BM-3 A74E, F87V, P386S mutant was measured as
well as the hydroxylation of n-octane and naphthalene
byP450 BM-3 A74G, F87V, L188Q. For all these
reactions, educts and products were quantified byGC.
The transformation of b-ionone is of great interest to
[
37]
the flavour industry.
GC analysis of the reaction products of the heteroge-
neouslycatal sy ed reaction revealed that 83% of b-
ionone was converted; reaction product was exclusively
-hydroxy-b-ionone. Investigations on the enantiomeric
4
excess of 4-hydroxy-b-ionone formed in this reaction are
currentlyin progress.
In the case of naphthalene oxidation, GC analysis of
the reaction extract revealed 77% conversion, mainlyto
Figure 7. P450 BM-3 activityin presence of acetone. Squares/
solid line: Sol-gel immobilised enzyme. 100% activity corre-
spond to 0.008 U/mL (0.075 nmol/mL of sol-gel immobilised
P450 BM-3). Circles/dotted line: Enzyme dissolved in 50 mM
KPi. 100% activitycorrespond to 0.009 U/mL (0.044 nmol/
mL of free P450 BM-3).
1
-naphthol (85%). As minor component (15%) 2-
naphthol was detected. These results show similarity
to the observations reported byAppel et al. for this
reaction with the same P450 BM-3 mutant in solution: In
their investigation exclusively1-naphthol was detected
[
12]
as reaction product.
Finallywe studied the h yd rox yl ation of n-octane by
immobilised P450 BM-3. 79% of the educt was con-
verted. As reaction products the regioisomers 2-, 3- and
4
-octanol were identified in molar ratio 1:2.1:1.6. As for
naphthalene, these values correlate to those reported by
Appel et al. for the free enzyme.
[
12]
Figure 8. Conversion of b-ionone to 4-hydroxy-b-ionone
catalysed by the P450 BM-3 A74E, F87V, P386S mutant.
As substrates belonging to diverse substance classes
such as fattyacids (10-pNCA), terpenes ( b-ionone),
aromatics (naphthalene) and n-alkanes (n-octane) are
hydroxylated by sol-gel immobilised P450 BM-3 mu-
tants, we conclude that the method described here is Cofactor Recycling with FDH
applicable to a verybroad range of substrates. We
suppose that at least most substances which are sub- First we investigated whether dissolved P450 BM-3 and
strates of free P450 BM-3 can also be transformed bythe
sol-gel immobilised enzyme.
FDH are active in the same buffer. To make sure that
onlyNADPH that is rec yc led from NADP is used for
‡
806
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 802 ± 810

‡
Immobilisation of P450 BM-3 and an NADP Cofactor Recycling System
FULL PAPERS
Thus, the combination of sol-gel immobilised P450
BM-3 and FDH in repeated batch reactions represents
as far as we know the first solid phase example of
combining a P450 monooxygenase with a cheap cofactor
recycling system. Sol-gel immobilisation of P450 BM-3
facilitates the separation of the biocatalyst from the
product and greatlyimproves the retention of enzymatic
stabilityand activit y. Therefore, the approach offers the
possibilityof designing a P450 bioreactor that can be
operated over a long period of time. The use of
immobilised FDH in such a system allows the exchange
of stoichiometric quantities of expensive NADPH with
‡
[38]
cheaper NADP . Combined with the broad substrate
spectrum of P450 BM-3, which can certainlybe ex-
tended further, a step towards the technical application
of heme-containing monooxygenases in fine chemical
Figure 9. Turnover of 0.25 mM 10-pNCA by0.02 U/ml P450
BM-3 monitored byabsorption measurement at 410 nm
(
continuous in situ measurement). No NADPH was supplied synthesis has been made. The transformation of b-
externallyand the reduction equivalents were generated by
.025 mM (solid line), 0.005 mM (dashed line), 0.0025 mM activated carbon atom. The possibility to hydroxylate
ionone presented here involves the hydroxylation at an
0
‡
(
dotted line) NADP and 0.02 U/ml FDH.
even non-activated carbons is demonstrated in the
reactions with n-octane and naphthalene described
above. The full potential of the approach will be
the oxidation of 10-pNCA no external NADPH was exploited in stereo- and regioselective hydroxylation
added to the reaction mixture. A 10-fold excess of 10- reactions at non-activated carbons. Therefore, deeper
‡
pNCA over NADP leads to a quantitative oxidation insights into the suggested system are required and
reaction within one hour reaction time (Figure 9). potential improvements are certainlypossible, which
‡
‡
Decreasing the NADP -concentration results in lower will also involve the further reduction of NADP .
reaction speed. In case of a 50- and 100-fold excess of 10- Experiments on long-term oxidation reactions using
‡
pNCA over NADP , the reaction rate decreases to 26% the methods described in this paper are currentlyunder
and 4.4% of the value obtained with a 10-fold excess of way.
1
0-pNCA, respectively.
Encouraged bythis result, sol-gel immobilisation of
This work presents a simple approach to practically
useful and stable hydroxylation catalysts based on P450
FDH was examined in two approaches. In the first set of BM-3.
experiments, co-immobilisation of both enzymes simul-
taneously(P450 BM-3 and FDH) was investigated. In
the second set, the two proteins were immobilised
Experimental Section
separatelyand then mixed in a ratio of 1:1 (m/m). Co-
immobilised enzymes from the first set and the mixture
from the second set were used in pNCA assays, which
again contained a 10-fold excess of 10-pNCA over
Chemicals
‡
All chemical reagents were purchased from Fluka, Aldrich,
Sigma or Riedel-de-Ha e» n. NADPH tetrasodium salt was
procured from J¸lich Fine Chemicals (Germany). All chem-
icals used were of analytical grade or higher. 10-pNCA was
synthesised as described elsewhere.[ b-Ionone was procured
from Merck, the 4-hydroxy derivative was synthesised accord-
oxidised NADP . After 3 hours, the co-immobilised
enzymes showed a conversion of up to 28% of the 10-
pNCA used, whereas the mixture of the separately
immobilised enzymes led to 75% conversion of the
educt.
This result could be reproduced thrice in repeated
batch experiments, indicating the theoretical possibility
to design a continuouslyoperated bioreactor based on
sol-gel encapsulated P450 BM-3.
34]
[39]
ing to Broom et al. The substrates b-ionone and 10-pNCA
were added to the reaction mixtures as 10 ± 25 mM stock
solutions in DMSO, which were prepared fresh everyda.y
‡
10 mM NADPH and NADP were dissolved fresh each dayin
5
0 mM KPi pH 8.1 and pH 6.5, respectively.
Conclusions
Enzymes
The exploitation of P450 s in technical applications has
attracted great attention. However, not much is known
‡
NADP -dependent formate dehydrogenase (FDH) was ob-
tained from Prof. V. I. Tishkov (Department of Chemical
about using P450 monooxygenases in organic syn- Enzymology, Lomonosov Moscow State University, 119899
thesis.[
3,15,17]
Moscow, Russia). FDH was stored in 0.1 M KPi, 1.2 M
Adv. Synth. Catal. 2003, 345, 802 ± 810
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
807

FULL PAPERS
Steffen C. Maurer et al.
[
40]
NH SO , 20% (v/v) glycerol, pH 7 at 48C. The obtained
CO difference spectroscopyas previouslydescribed,
using
4
4
À1
À1
enzyme batch displayed an activity of 47 U/mL.
DNA-modifying enzymes were purchased from MBI Fer-
mentas.
an extinction coefficient of 91 mM cm for the 450 minus
490 nm peak.
Sol-Gel Immobilisation
Construction of a P450 BM-3 Expression System
The fluoride-catalysed formation of the silica sol was started by
mixing 5 volumes of TEOS with 1 part of 20% (w/v) poly-
ethylene glycol 6000 in 50 mM KPi and 0.25 parts of 0.2 M NaF
The genes encoding P450 BM-3 mutants A74G, F87V, L188Q
and A74D, F87V, P386S were amplified from pT-USC1BM-
[
29]
3
and pET22-BM-3mt, respectively, by PCR using primers on ice. After 3 minutes, 2.5 volumes of a 2 ± 12 mM solution of
designed to facilitate cloning into pET28a(‡) vector. The
P450 BM-3 in 50 mM KPi buffer was added. The mixture was
stirred on ice for another 15 minutes or until a homogeneous
emulsion had formed. This hydrogel was subsequently stored
at 48C for at least 3 days to allow formation of the xerogel. The
resulting hard and brittle solid was washed 5 times in a
centrifugation tube with 50 mM KPi to remove all traces of
non-immobilised P450 BM-3. The samples were subsequently
lyophilised. The obtained immobilisate was 60% of the
theoreticallypossible yi eld calculated from the amount of
oligonucleotide primer for the 5'-end of the gene: 5'-
GCGGATCCATGACAATTAAAGAAATGCCTCAGC was
designed to introduce a BamHI restriction site upstream of
the start codon. The primer for the 3'-end of the gene: 5'-
GCGAATTCTTACCCAGCCCACACGTCTTTTGCG in-
troduced an EcoRI restriction site downstream after the stop
codon. The gene was amplified in 25 cycles of 2 min 958C,
followed by2 min annealing at 62 8C and extension 4 min at
À1
728C in an Eppendorf Thermal Cycler. The amplified gene was
TEOS used. The P450 BM-3 content [nmol g ] of the sol-gels
subsequentlycloned into the expression vector by BamHI/
EcoRI restriction sites. Initial cloning has been done in strain
DH5a (Clontech, Heidelberg, Germany), which gives high
transformation efficiencyand good plasmid yi eld followed by
heterologous expression in BL21(DE3) Escherichia coli cells
was determined bydividing the total amount of P450 BM-3
used minus the P450 amount in the washing solutions bythe
total mass of the sol-gel obtained. Depending on the P450 BM-
3 concentration of the particular enzyme batch used for
immobilisation, loadings of 13 ± 60 nmol catalyst per g sol-gel
were obtained. The residual P450 BM-3 activityafter immobi-
lisation was 50% of the original activity. Immobilised enzyme
preparations were stored at À 208C until required.
(
Novagen, Madison, USA) with N-terminal His6 tag to
facilitate purification of the protein on a nickel affinitycolumn.
Attachment of the His polypeptide had no effect on the
6
activityof the gene product.
FDH was immobilised in the same wayas P450 BM-3.
Prior to use, sol-gel immobilised enzymes were ground to a
fine powder in order to reduce mass transfer limitation effects
in the nanoporous catalyst particles.
Expression and Purification of P450 BM-3
The first culture (2 mL) was inoculated from a single colony
into 2 mL Luria-Bertani (LB)-medium supplied with kanamy-
cin (30 mg/mL) and grown at 378C with shaking at 150 rpm
until OD578 reaches 0.6 ± 1.0. This culture was used to inoculate
Activity Assays with 10-pNCA
All activitymeasurements were executed at least three times to
minimise standard errors.
200 mL LB-medium, supplied with kanamycin (30 mg/mL).
The cells were grown at 378C with shaking at 160 rpm to an
OD578 ~0.8 and expression of the gene was induced byaddition
of isopropyl-b-D-thiogalactopyranoside (IPTG) to 0.35 mM.
After a further 6 h of growth at 308C the cells were harvested
bycentrifugation (20 min, 6000 rpm, 4 8C). The cell pellet was
then resuspended in loading buffer (50 mM KPi, 800 mM
NaCl, 0.1 mM PMSF, pH 7.5) and lysed by sonication on ice
P450 BM-3 activity assays with the enzyme in solution were
performed as previouslydescribed byfollowing the kinetics of
p-nitrophenolate production at 410 nm.[
29]
For immobilised enzymes this procedure was modified: 5 mg
of sol-gel immobilised P450 BM-3 were incubated with 965 mL
50 mM KPi, pH 8.1, 10 mL 25 mM 10-pNCA in DMSO and
25 mL 10 mM NADPH stock solution in a shaker at 1200 rpm.
After 1 hour the solid was removed bycentrifugation. p-
Nitrophenolate production was measured bydetermining the
absorption difference at 410 nm of the supernatant against a
blank, which was treated exactlyas the actual sample except
that no P450 BM-3 was encapsulated in the used sol-gel.
When activityof immobilised P450 BM-3 was compared to
activity of the enzyme in solution, activity of the free enzyme
was assayed as described for the immobilised enzyme, using the
same amount of P450 BM-3 in both cases.
(
5  2 min, output control ˆ 4, dutyc yc le 40%; BransonSoni-
fier W250, Dietzenbach, Germany).
Purification took advantage of the His6 tag, utilising
Quiagen Ni-NTA superflow resin. After centrifugation cell
lysates were 0.22 mm-filtered and loaded onto the pre-equili-
brated affinitycolumn. The column was subsequentlywashed
with loading buffer, washing buffer A (50 mM KPi, 800 mM
NaCl, pH 6.2) and washing buffer B (50 mM KPi, 800 mM
NaCl, 250 mM glycine, pH 7.5). As in our experiments the
standard Ni-NTA eluting agent imidazole led to partial
aggregation and inactivation of BM-3, enzyme elution was
performed using L-histidine (50 mM KPi, 80 mM L-histidine,
pH 7.5). The purified protein was supplemented with 50%
glycerol and stored at À 208C. Prior to use the buffer was
exchanged to 50 mM KPi byultrafiltration utilising a 50 kDa
Amicon membrane. P450 concentrations were measured by
The pNCA assaywith FDH and P450 BM-3 in solution was
‡
executed byadding 2.5 mL 10 mM NADP stock solution to
1 mL of FDH/BM-3 assaysolution (containing 0.25 mM 10-
pNCA, 0.05 mM P450 BM-3, 0.02 U/mL FDH, 100 mM KPi,
300 mM formate, pH 8.1) resulting in a 10-fold excess of 10-
‡
‡
pNCA over NADP . Immediatelyafter NADP
addition,
absorption measurement at 410 nm was started.
808
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 802 ± 810

‡
Immobilisation of P450 BM-3 and an NADP Cofactor Recycling System
FULL PAPERS
For activityassa yi ng of immobilised FDH in combination
with immobilised P450 BM-3 5 mg of each immobilised
enzyme preparation (or 10 mg of the co-immobilised enzymes)
was added to 1 mL of FDH/BM-3 assaysolution. The reaction
[3] V. Urlacher, R. D. Schmid, Curr. Opin. Biotechnol. 2002,
13, 557 ± 564.
[
[
[
4] L. P. Wen, A. J. Fulco, J. Biol. Chem. 1987, 262, 6676 ±
682.
5] A. J. Fulco, R. T. Ruettinger, Life Sci. 1987, 40, 1769 ±
775.
6] S. S. Boddupalli, R. W. Estabrook, J. A. Peterson, J. Biol.
Chem. 1990, 265, 4233 ± 4239.
6
‡
was started byaddition of 2.5 mL 10 mM NADP , again leading
‡
to a 10-fold excess of 10-pNCA over NADP . Further steps
1
were carried out as described for immobilised BM-3 without
FDH, except that the reaction time was prolonged to 3 hours.
To quantifypNCA turnover, an extinction coefficient of p-
À1
À1
[7] G. Truan, M. R. Komandla, J. R. Falck, J. A. Peterson,
Arch. Biochem. Biophys. 1999, 366, 192 ± 198.
[8] T. W. Ost, C. S. Miles, J. Murdoch, Y. Cheung, G. A.
Reid, S. K. Chapman, A. W. Munro, FEBS Lett. 2000,
nitrophenolate at 410 nm of 13.2 mM cm was used. For
determining 10-pNCA turnover the respective reaction and
reference solutions were diluted in the ratio 1 to 10 with 50 mM
KPi, pH 8.1 to give absorption values in the linear range of the
photometer detector.
4
86, 173 ± 177.
[
9] Q. S. Li, U. Schwaneberg, M. Fischer, J. Schmitt, J. Pleiss,
S. Lutz-Wahl, R. D. Schmid, Biochim. Biophys. Acta
Turnover of b-Ionone, Naphthalene and n-Octane with
Immobilised P450 BM-3 Mutants
2
001, 1545, 114 ± 121.
[
10] Q. S. Li, U. Schwaneberg, P. Fischer, R. D. Schmid,
Chemistry 2000, 6, 1531 ± 1536.
[11] Q. S. Li, J. Ogawa, R. D. Schmid, S. Shimizu, Appl.
Environ. Microbiol. 2001, 67, 5735 ± 5739.
[12] D. Appel, S. Lutz-Wahl, P. Fischer, U. Schwaneberg,
R. D. Schmid, J. Biotechnol. 2001, 88, 167 ± 171.
1
(
00 mg of the respective immobilised P450 BM-3 mutant
35 nmol/g) was suspended in 9.3 mL 50 mM KPi pH 7.5,
supplemented with 100 mL 20 mM substrate in DMSO and
00 mL 10 mM NADPH. The reaction mixture was subse-
6
quentlyshaken for 2 h. After centrifugation the supernatant
was extracted twice with 4 mL diethyl ether. The combined
organic layers were dried with sodium sulphate and concen-
trated to a volume of 100 mL.
Reaction products and unreacted educts were identified by
GC analysis using a Fisons GC 8000 gas chromatograph
equipped with a flame ionisation detector (FID) and a 60 m
Zebron ZB 1 (polydimethylsiloxane) column. The temper-
ature gradients were as follows: b-Ionone : 1) 908C for 5 min, 2)
[
[
[
13] Q. S. Li, J. Ogawa, R. D. Schmid, S. Shimizu, FEBS Lett.
001, 508, 249 ± 252.
14] W. Tischer, F. Wedekind, Top. Curr. Chem. 1999, 200,
5 ± 126.
2
9
15] P. Fernandez-Salguero, C. Gutierrez-Merino, A. W.
Bunch, Enzyme Microb. Technol. 1993, 15, 100 ± 104.
[16] S. B. Lamb, D. C. Lamb, S. L. Kelly, D. C. Stuckey,FEBS
Lett. 1998, 431, 343 ± 346.
[17] M. Taylor, D. C. Lamb, R. J. Cannell, M. J. Dawson, S. L.
Kelly, Biochem Biophys. Res. Commun. 2000, 279, 708 ±
9
1
0 to 3008C at 108C/min, 3) 3008C for 10 min. Naphthalene: 1)
008C for 10 min, 2) 100 to 1408C at 28C/min, 3) 140 to 2508C
at 108C/min, 4) 2508C for 5 min. Octane: 1) 458C for 8 min, 2)
5 to 1008C at 28C/min, 3) 100 to 2408C at 108C/min. Pure
4
7
11.
samples of substrates and potential reaction products were
available. Equal amounts of these substances dissolved in
diethyl ether were applied to the column. From the resulting
GC trace the ratio of the peak areas corresponding to the
substrates and products were calculated. These ratios were
used to determine the molar ratios of substrates and products
emerging from the biotransformations. Therefore equal dieth-
yl ether-water partition coefficients for educts and products
were assumed.
[
[
18] I. Gill, Chem. Mater. 2001, 13, 3404 ± 3421.
19] Catalogue price of NADPH and NADH about 800 E/g
and 90 E/g, respectively.
20] G. J. Lye, J. M. Woodley, TIBTECH 1999, 17, 395 ± 402.
21] C. H. Lee, K. L. Parkin, Biotechnol. Bioeng. 2001, 75,
219 ± 227.
[
[
[22] A. Sanchez, J. L. del Rio, F. Valero, J. Lafuente, I. Faus,
C. Sola, J. Biotechnol. 2001, 84, 1 ± 12.
[
[
23] J. C. Chen, S. W. Tsai, Biotechnol. Prog. 2000, 16, 986 ±
92.
24] N. S. Dosanjh, J. Kaur, Biotechnol. Appl. Biochem. 2002,
6, 7 ± 12.
9
Acknowledgements
3
We thank BASF AG (Ludwigshafen, Germany) for financial
support and Professor Vladimir I. Tishkov for providing us with
NADP -dependent formate dehydrogenase.
[25] U. Schwaneberg, D. Appel, J. Schmitt, R. D. Schmid, J.
Biotechnol. 2000, 84, 249 ± 257.
[26] G. Gilardi, Y. T. Meharenna, G. E. Tsotsou, S. J. Sadeghi,
M. Fairhead, S. Giannini, Biosens. Bioelectron. 2002, 17,
‡
1
33 ± 145.
[
[
27] J. Kazlauskaite, A. C. G. Westlake, L.-L. Wong, H. A. O.
Hill, Chem. Commun. 1996, 18, 2189 ± 2190.
28] F. Hollmann, B. Witholt, A. Schmid, J. Mol. Cat. B 2002,
References and Notes
[
1] J. T. Groves, Y.-H. Han, in Cytochrome P450: Structure,
Mechanism and Biochemistry, (Ed.: P. R. E. Ortiz de
Montellano), Plenum Press, New York, 1995, pp 3 ± 48.
2] E. T. Farinas, U. Schwaneberg, A. Glieder, F. H. Arnold,
Adv. Synth. Catal. 2001, 343, 601 ± 606.
7
91, 1 ± 10.
[29] V. I. Tishkov, A. G. Galkin, V. V. Fedorchuk, P. A.
Savitsky, A. M. Rojkova, H. Gieren, M. R. Kula, Bio-
technol. Bioeng. 1999, 64, 187 ± 193.
[
Adv. Synth. Catal. 2003, 345, 802 ± 810
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
809

FULL PAPERS
Steffen C. Maurer et al.
[
30] K. Seelbach, B. Riebel, W. Hummel, M.-R. Kula, V. I.
Tishkov, A. M. Egorov, C. Wandrey, U. Kragl, Tetrahe-
dron Lett. 1996, 37, 1377 ± 1380.
[35] O. Lentz, Q. S. Li, U. Schwaneberg, S. Lutz-Wahl, P.
Fischer, R. D. Schmid, J. Mol. Cat. B 2001, 15, 123 ± 133.
[36] V. Urlacher, manuscript in preparation.
[37] V. Krasnobajew, D. Helmlinger, Helv. Chim. Acta. 1982,
65, 1590 ± 1601.
[
[
[
[
31] E. Katchalski-Katzir, D. M. Kraemer, J. Mol. Cat. B
2
000, 10, 157 ± 176.
‡
32] S. Shtelzer, S. Rappoport, D. Avnir, M. Ottolenghi, S.
Braun, Biotechnol. Appl. Biochem. 1992, 15, 227 ± 235.
33] U. Schwaneberg, PhD thesis, Universit‰t Stuttgart,
Germany, 1999.
34] U. Schwaneberg, C. Schmidt-Dannert, J. Schmitt, R. D.
Schmid, Anal. Biochem. 1999, 269, 359 ± 366.
[38] Catalogue price of NADPH and NADP about 800 E/g
and 260 E/g, respectively.
[39] S. J. Broom, R. M. Ede, A. L. Wilkins, Tetrahedron Lett.
1992, 33, 3197 ± 3200.
[40] T. Omura, R. J. Sato, J. Biol. Chem. 1964, 239, 2370 ±
2378.
810
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 802 ± 810