In vitro double oxidation of n-heptane with direct cofactor regeneration

Article abstract of DOI:10.1002/adsc.201300143

A novel concept for the direct oxidation of cycloalkanes to the corresponding cyclic ketones in a one-pot synthesis in water with molecular oxygen as sole oxidizing agent was reported recently. Based on this concept we have developed a new strategy for the double oxidation of n-heptane to enable a biocatalytic resolution for the direct synthesis of heptanone and (R)-heptanols in a one-pot reaction. The bicatalytic cascade employs an NADH driven P450 BM3 monooxygenase variant (WT^NADH, 19A12^NADH or CM1 ^NADH) and an (S)-enantioselective alcohol dehydrogenase (RE-ADH). In the initial step n-heptane is hydroxylated under consumption of NADH to produce (R/S)-heptanol. In the second oxidation step the (S)-heptanol enantiomers are transformed to the corresponding ketones, reducing and thereby regenerating the cofactor. Characterization of initial hydroxylation step revealed high turnover frequencies (TOF) of up to 600 min^-1, as well as high coupling efficiencies using NADH as cofactor (up to 44%). In the cascade reaction a nearly 2-fold improved product formation was achieved, compared to the single hydroxylation reaction. The total product concentration reached 1.1 mM, corresponding to a total turnover number (TTN) of 2500. Implementation of an additional cofactor regeneration system (D-glucose/glucose dehydrogenase) enabled a further enhancement in product formation with a total product concentration of 1.8 mM and a TTN of 3500. Copyright

Full text of DOI:10.1002/adsc.201300143

FULL PAPERS
DOI: 10.1002/adsc.201300143
In Vitro Double Oxidation of n-Heptane with Direct Cofactor
Regeneration
Christina A. Mꢀller,a Beneeta Akkapurathu,a Till Winkler,b Svenja Staudt,c
Werner Hummel,b Harald Grçger,c,d and Ulrich Schwaneberga,*
a
Institute of Biotechnology, RWTH Aachen, Worringerweg 1, 52074 Aachen, Germany
Fax : (+49)-241-80-22387; phone: (+49)-241-80-24170; e-mail: u.schwaneberg@biotec.rwth-aachen.de
Institute of Molecular Enzyme Technology at the Heinrich-Heine-University of Dꢀsseldorf, Research Centre Jꢀlich,
b
Stetternicher Forst, 52426 Jꢀlich, Germany
Department of Chemistry and Pharmacy, University of Erlangen-Nꢀrnberg, Henkestrasse 42, 91054 Erlangen, Germany
Faculty of Chemistry, Bielefeld University, Universitꢁtsstrasse 25, 33615 Bielefeld, Germany
c
d
Received: February 20, 2013; Published online: && &&, 0000
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201300143.
Abstract: A novel concept for the direct oxidation of hydroxylation step revealed high turnover frequen-
cycloalkanes to the corresponding cyclic ketones in cies (TOF) of up to 600 minÀ1, as well as high cou-
a one-pot synthesis in water with molecular oxygen pling efficiencies using NADH as cofactor (up to
as sole oxidizing agent was reported recently. Based 44%). In the cascade reaction a nearly 2-fold im-
on this concept we have developed a new strategy proved product formation was achieved, compared
for the double oxidation of n-heptane to enable a bio- to the single hydroxylation reaction. The total prod-
catalytic resolution for the direct synthesis of heptan- uct concentration reached 1.1 mM, corresponding to
ACHTUNGTRENNUNGone and (R)-heptanols in a one-pot reaction. The bi- a total turnover number (TTN) of 2500. Implementa-
catalytic cascade employs an NADH driven P450 tion of an additional cofactor regeneration system
BM3 monooxygenase variant (WTNADH, 19A12NADH (d-glucose/glucose dehydrogenase) enabled a further
or CM1NADH) and an (S)-enantioselective alcohol de- enhancement in product formation with a total prod-
hydrogenase (RE-ADH). In the initial step n-hep- uct concentration of 1.8 mM and a TTN of 3500.
tane is hydroxylated under consumption of NADH
to produce (R/S)-heptanol. In the second oxidation
step the (S)-heptanol enantiomers are transformed Keywords: alkanes; cascade reactions; C H activa-
to the corresponding ketones, reducing and thereby tion; directed evolution; enzymatic resolution; oxida-
regenerating the cofactor. Characterization of initial tion; P450 BM3
À
Introduction
boron reagents[6] and chiral phosphorus ligands[7] or
by enzymatic ketone reduction.[8]
Petrochemical feedstocks are currently used as main
Chiral alcohols are used as building blocks for agro-
raw materials for the production of bulk and fine chemicals and pharmaceuticals.[9] The stoichiometric
chemicals. A selective oxyfunctionalization of hydro- need of transition metal reagents,[10] or large quanti-
carbons remains one of the most challenging tasks in ties of organic solvents, and the laborious separation
chemistry, especially with regard to simple alkanes and recycling steps for products, solvents and catalysts
and arenes,[1] mainly due to a low reactivity of C H renders these chemical synthesis routes unfavourable
À
bonds[2] and poor chemo- and stereoselectivity of in regard to environmental impact and sustainabil-
most chemocatalysts.[3] The Wacker oxidation[4] repre-
sents one of the most efficient chemical methods for
ity.[11]
ACHTUNGTRENNUNG
Biotechnological processes offer an economical at-
oxyfunctionalization of olefins and enables the syn- tractive and environmentally preferable access to
thesis of aldehydes and ketones.[5] Prochiral ketones chiral alcohols.[12] Nowadays most biocatalytic pro-
and olefins are further used for synthesis of chiral al- cesses rely on reduction of ketones making use of ke-
cohols, either by asymmetric reduction with chiral toreductases.[8] The direct biocatalytical activation of
alkanes is an interesting and more cost-effective alter-
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
ÞÞ
These are not the final page numbers!

Christina A. Mꢀller et al.
FULL PAPERS
native to the use of ketones in stereochemical synthe- duced to water. The electron transfer takes place
sis, making the enzymatic oxyfunctionalization a key from the nicotinamide cofactor NAD(P)H to the
parameter for a successful application.[13]
heme iron via an FAD-FMN electron transfer
The enzyme-catalyzed oxidation of non-activated chain.[20,21]
alkanes is in general achieved with higher selectivities
P450 BM3 from Bacillus megaterium is particularly
than with chemocatalysts.[14] The main enzyme classes interesting for in vitro applications and chemical syn-
that catalyze alkane oxygenation reactions are oxy- thesis since it is a soluble and catalytically self-suffi-
genases and peroxidases.[15] Peroxidases catalyze cient monooxygenase with the FAD-FMN reductase
oxygen transfer reactions, such as olefin epoxidations and hemoprotein domain on one polypeptide chain.[22]
and sulfoxidations. Recently, the peroxygenase from P450 BM3 has been engineered extensively to broad-
Agrocybe aegerita (AaP) was reported as the first per- en the substrate spectra, to enhance activities, to in-
oxidase to hydroxylate aliphatic chains.[16] Oxygenases crease thermal or process stability, and to accept alter-
comprise a more diverse catalytic family due to their native electron transfer systems.[23]
intrinsic physiological role in detoxification, synthesis
In particular, P450 BM3 mutants with changed sub-
of secondary metabolites or hydrocarbon degrada- strate specificity and selectivity for medium- and
tion.[15] Nevertheless, industrial applications of oxy- short-chain alkanes were successfully reengineered.[24]
genases are often hindered, mainly due to their low P450 BM3 wild-type hydroxylates C6–C10 alkanes with
reaction rates, stabilities and cofactor dependency.[17] low activities, but does not convert short-chain alka-
To overcome the “cofactor challenge” in cell-free bio- nes (<C6).[24] The engineered variant 19A12 prefers
catalysis two basic strategies have been investigated: as substrates short-chain alkanes (C3–C7). It hydroxy-
the use of enzymatic regeneration systems[18] and re- lates propane with a turnover frequency (TOF) of
generation by chemical, electrochemical or photo- 420 minÀ1 and a total turnover number (TTN) of
chemical set-ups.[19] The enzymatic regeneration strat- 10550.[25]
egy is commonly employed using dehydrogenases
Recently, we reported a novel process concept for
with cosubstrates like isopropyl alcohol or formate.[18] the double oxidation of cycloalkanes to produce cy-
P450 monoogygenases are one class of heme-con- cloalkanones in a one-pot synthesis.[26] This bicatalytic
taining oxygenases, capable of oxidizing non-activated reaction with a P450 monooxygenase and an alcohol
À
C H bonds at room temperature, in water with dehydrogenase takes place in water and employs mo-
oxygen as sole oxidant.[20] One atom of molecular di- lecular oxygen as sole oxidizing agent. Furthermore,
oxygen is often incorporated stereo- and regioselec- for regeneration of the nicotinamide cofactor no addi-
tively into the substrate, while the second atom is re- tional co-substrate is supplemented.
Based on this concept a novel strategy for double
oxidation of medium-chain alkanes using n-heptane
as substrate was developed (Scheme 1). In this report
we have implemented an enzyme cascade (Scheme 2)
consisting of two steps for the enantioselective syn-
Scheme 1. Reaction scheme for the bicatalytic direct oxida-
thesis of chiral alcohols starting from alkanes: the
first reaction comprises the P450 BM3-catalyzed hy-
droxylation of an alkyl chain (1), followed by
tion of n-heptane to 2-heptanone in aqueous medium, using
a monooxygenase–dehydrogenase cascade and oxygen as
sole oxidizing agent.
Scheme 2. Double oxidation of n-heptane using P450 BM3 monooxygenases and an alcohol dehydrogenase from Rhodococ-
cus erythropolis (RE-ADH) as cascade catalysts. In the first oxidation step n-heptane (1) is converted by a P450 BM3 mono-
oxygenase (WTNADH, 19A12NADH, and CM1NADH) to heptanol [enantiomeric mixture of (S)/(R)-2] under NADH consumption
yielding NAD+. In the second oxidation step enzymatic resolution[28] of the in situ produced (R)-heptanols [(R)-2] and syn-
thesis of the corresponding ketones (3) is achieved under reduction of NAD+ to NADH (cofactor regeneration). R=CH3,
C2H5 or C3H7; R’=C5H11, C4H9 or C3H7.
2
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

In Vitro Double Oxidation of n-Heptane with Direct Cofactor Regeneration
a second oxidation of the in situ-produced alcohol iso-
Accumulation of 18 mutations in this P450 BM3
mers (2) by means of an (S)-enantioselective ADH variant resulted in a significantly decreased expres-
from Rhodococcus erythropolis,[27] thus leading to the sion (E. coli BL21 Gold (DE3) lacIQ1) yielding 6% of
non-converted (R)-enantiomers as remaining prod- the wild-type production (see the Supporting Infor-
ucts.
mation, Figure S1). To investigate the effect of differ-
Remarkably, the second oxidation step does not ent amino acid substitutions on the hydroxylation of
only enable an enzymatic resolution[28] under accumu- n-heptane, the 18 previously reported mutated posi-
lation of the produced (R)-alcohols [(R)-2], but also tions were saturated individually in BM3 WT using
the synthesis of the corresponding ketones (3) with degenerated NNK oligonucleotides[31] (the targeted
direct regeneration of the oxidized nicotinamide co- codon is randomized to all 20 canonical amino acids;
factor. In order to achieve cofactor compatibility be- N=G/A/T/C; K=G/T). Screening of 180 clones per
tween both enzymes, the cofactor specificity of the saturation mutagenesis library with the NADPH de-
P450 BM3 monooxygenases was altered from pletion assay[32] showed that position 255 (32-fold im-
NADPH to NADH according to substitutions report- proved product formation) and to a lesser extent the
ed by Maurer et al.[29] As a proof of concept the positions R47, S226, H236 and L188 (up to 10-fold
double oxidation was performed with three different improvement) contributed individually to an opti-
P450 BM3 monooxygenases, WTNADH, 19A12NADH
,
mized n-heptane hydroxylation. Substitution of argi-
nine by proline at position 255 yielded the best result
in terms of NADPH consumption and product forma-
tion (see the Supporting Information, Figure S2). A
second substitution from proline to histidine at posi-
tion 329 was introduced in BM3 R255P via site-direct-
and CM1NADH
.
Results
The one-pot double oxidation of n-heptane by direct ed mutagenesis to finally generate the variant P450
coupling of two enzymes, the P450 BM3 monooxyge- BM3 CM1 (substitutions R255P/P329H). Characteri-
nase and the alcohol dehydrogenase from R. erythrop- zation of this double mutant for the hydroxylation of
olis (RE-ADH) was developed.
n-heptane using cell-free extracts yielded a turnover
As monooxygenase the wild-type (WT) enzyme frequency (TOF) of 567 minÀ1 compared to 13 minÀ1
and two reengineered P450 BM3 variants displaying in case of the wild-type, representing a 43-fold im-
higher activity for n-heptane (19A12,[25] CM1) were proved product formation. The highly active P450
tested. The switch in cofactor specificity from BM3 CM1 mutant was therefore chosen for evalua-
NADPH to NADH was achieved through introduc- tion of the monooxygenase-dehydrogenase cascade
tion of the reported substitutions W1046S and reaction in comparison to the wild-type, and to the
R966D.[29] For simplicity reasons the variants with al- mutant 19A12[25] as reference catalyst.
tered cofactor specificity are named WTNADH
19A12NADH, and CM1NADH
,
.
These three monooxygenase variants were charac- Characterization of NADH-Dependent P450 BM3
terized in detail (activity, coupling efficiency, selectivi- WTNADH, 19A12NADH, and CM1NADH Monooxygenases
ty and productivity) for the preliminary oxidation of in the Initial Step of the Cascade Reaction
n-heptane to 1-, 2-, 3-, or 4-heptanol. Finally, the
entire reaction concept was evaluated by determining P450 BM3 has a high specificity for NADPH over
product formation and total turnover numbers. The NADH.[33] Maurer et al. as well as Neeli et al. identi-
double oxidation reaction was performed in vitro, fied that mutations in the codon of W1046 in the re-
using purified enzymes (0.5 mM P450 BM3/10 U RE- ductase domain of P450 BM3 can cause a switch in
ADH) in a reaction scale of 1 mL.
cofactor specificity to NADH.[29,33] In this study, the
previously reported substitutions W1046S and
R966D[29] were introduced in the P450 BM3 wild-
type, 19A12 and CM1 by site-directed mutagenesis to
generate the three NADH-dependent (NADH) BM3
variants (WTNADH, 19A12NADH, and CM1NADH; see
Engineering of the P450 BM3 Mutant CM1 for
Improved n-Heptane Hydroxylation
The P450 BM3 monooxygenase 19A12 (containing Figure 1; light grey; corresponding NADPH consump-
substitutions R47C, L52I, V78F, A82S, K94I, P142S, tion values are shown in Figure 1 in dark grey).
T175I, A184V, L188P, F205C, S226R, H236Q, E252G,
WTNADH shows with both cofactors and n-heptane
R255S, A290V, A328F, L353V, I366V)[25] was previ- a low activity (132 minÀ1 for NADH; Figure 1A) and
ously reported as a reengineered variant for regiose- a moderate coupling efficiency (18% for NADPH;
lective
hydroxylation
of
alkanes
(propane, Figure 1C). The variants 19A12NADH and CM1NADH
butane)[25,30] and cycloalkanes (cyclohexane, cyclooc- prefer NADH over NADPH in contrast to WTNADH
tane and cyclodecane).[26]
(Figure 1A and B).
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3
ÞÞ
These are not the final page numbers!

Christina A. Mꢀller et al.
FULL PAPERS
Figure 1. Initial oxidation rates, turnover frequencies and coupling efficiencies of WTNADH, 19A12NADH, and CM1NADH mono-
oxygenases with the cofactors NADPH and NADH. All three BM3 variants contain the substitutions W1046S and R966D
for switching the cofactor preference from NADPH to NADH.[29] Panels A–C show: (A) initial cofactor oxidation (mmol co-
factor·mmol P450À1·minÀ1) determined via cofactor consumption, (B) turnover frequency (product formation in mmol pro-
duct·mmol P450À1·minÀ1) quantified by GC, and (C) coupling efficiency as ratio between product formation and cofactor oxi-
dation. Details of the biotransformation parameters are described in the Experimental Section.
The oxidation rate of NADH with variant fatty acids.[23b] WTNADH produces all 3 subterminal hy-
19A12NADH is 9% higher than with the variant droxylated heptanol regioisomers in amounts higher
CM1NADH (Figure 1A). Nevertheless, CM1NADH shows than 24%.
a 24% higher turnover frequency (TOF, Figure 1B)
In contrast, P450 BM3 monooxygenase 19A12NADH
than 19A12NADH, which can mainly be attributed to its and CM1NADH show a changed regiospecificity and
higher coupling efficiency (44% vs. 32%; Figure 1C).
a significant hydroxylation preference for two subter-
TOF values were calculated based on GC quantifi- minal carbon atoms. 19A12NADH hydroxylates mainly
cation of the overall amount of products (see the Sup- the C-2 atom (64% 2-heptanol) followed by C-3 (33%
porting Information, Figure S3). The TOF values 3-heptanol). CM1NADH produces preferentially 4-hep-
reached with NADH as cofactor are 485 minÀ1 for tanol (48%) followed by 3-heptanol (42%) and 2-hep-
19A12NADH and 604 minÀ1 for CM1NADH, which corre- tanol (9%) (for the original GC spectra see the Sup-
sponds to a specific activity of 4075 Ug P450À1 and porting Information, Figure S4).
5073 Ug P450À1, respectively.
Interestingly, the stereopreferences for chiral hepta-
Figure 2 shows the regioselectivity of P450 BM3 nols differ between WTNADH
,
19A12NADH
,
and
WTNADH, 19A12NADH and CM1NADH. WTNADH is a sub- CM1NADH (Table 1). WTNADH produces preferentially
terminal hydroxylase with a similar regioselectivity (R)-heptanols [ee of 49% for (R)-3-heptanol and 74%
(2-, 3- or 4-heptanol) to the hydroxylation pattern of for (R)-2-heptanol], whereas 19A12NADH produces
(S)-3-heptanol with a high preference (ee 86%) and
a nearly racemic mixture of 2-heptanol [ee 3.9% for
(S)-2-heptanol]. CM1NADH shows also a slight excess
of (S)-3-heptanol (ee 12%) and a clear preference for
(R)-2-heptanol (ee 46%). All details for the determi-
nation of enantioselectivity and the original GC spec-
tra can be found in the Supporting Information, Fig-
ure S5 and Figure S6.
Double Oxidation Reaction of n-Heptane:
Conversion and Total Turnover Numbers
Figure 3 shows the performance of WTNADH
,
19A12NADH, and CM1NADH in the initial hydroxylation
step (upper part) and in the double oxidation reaction
(lower part).
Figure 2. Regioselectivity of the n-heptane hydroxylation by
P450 BM3 monooxygenase WTNADH
,
19A12NADH and
The (S)-specific alcohol dehydrogenase from Rho-
dococcus erythropolis was selected for the oxidation
of 2-heptanol due to its high activity and selectivity.[34]
The volumetric activity of the semi-purified RE-ADH
CM1NADH. Product distribution was calculated for each var-
iant based on GC quantification of each regioisomer (1-, 2-,
3-, and 4-heptanols) and normalized to 100%.
4
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

In Vitro Double Oxidation of n-Heptane with Direct Cofactor Regeneration
Table 1. Enantioselectivity of the n-heptane hydroxylation catalyzed by P450 BM3 WTNADH, 19A12NADH and CM1NADH
.
P450 BM3 variant
2-Heptanol[a]
3-Heptanol[b]
ee [%]
er (S:R)
ee [%]
er (S:R)
WTNADH
74Æ0.1 (R)
4Æ3 (S)[c]
13:87
52:48
27:73
50Æ3 (R)
86Æ2 (S)
12Æ1 (S)
25:75
93:7
56:44
19A12NADH
CM1NADH
46Æ0.2 (R)
[a]
Enantiomeric excess (ee) and enantiomeric ratio (er) were determined by chiral GC, for 2-heptanol using a subsequent
lipase-catalyzed kinetic resolution [(R)-2-heptanol acylation].
[b]
[c]
For 3-heptanol, the absolute configuration was determined by polarimetry (see Experimental Section).
High deviation is caused by low differences in values of enantiomer “peaks”.
Figure 3. Conversion of n-heptane to heptanols (initial hydroxylation step; upper part) and conversion of n-heptane to hepta-
nones (double oxidation; lower part). Initial hydroxylations are shown for (A) WTNADH, (B) 19A12NADH or (C) CM1NADH
and the double oxidations in presence of the RE-ADH: (D) WTNADH, (E) 19A12NADH, and (F) CM1NADH. Total heptanol ( )
*
^
and total heptanone ( ) concentrations are shown as sum of the corresponding regioisomers (1-, 2-, 3-, and 4-heptanols)
which were quantified via GC after a reaction time of 0.17, 0.5, 1, 2, 4 and 24 h (see details in Experimental Section).
towards the heptanol isomers was determined under 0.23, 0.69, and 0.62 mM heptanols, respectively. In
the experimental conditions for the double oxidation contrast, during the double oxidation reaction the
cascade. (S)-2- and (S)-3-heptanol are the preferred product formation increases continuously over more
heptanol isomers of the RE-ADH, which had a volu- than 4 h. This can be attributed to the NADH cofac-
metric activity of 46.7 and 6.8 UmgÀ1 for each enan- tor recycling during the double oxidation. A detailed
tiomer, respectively. Activities for 1- and 4-heptanol comparison in terms of total product concentrations
as substrates reached only 1.3 and 0.8 UmgÀ1.
shows an increase from 0.23 to 0.56 mM for WTNADH
During the initial oxidation reaction n-heptane is (Figure 3A and D), from 0.69 to 1.3 mM for
hydroxylated by BM3 WTNADH or variants 19A12NADH (Figure 3B and E), and from 0.62 to
(19A12NADH, CM1NADH) to the heptanol isomers [see 1.1 mM CM1NADH (Figure 3C and F).
Scheme 2, product (S)/(R)-2]. In the subsequent step
RE-ADH determines the product ratio between
the (S)-specific RE-ADH catalyzes the further oxida- heptanones and heptanols in the coupled reaction due
tion of the preferred alcohols (S)-2 to the correspond- to its specificity for the (S)-isomers. This ratio differs
ing ketones (see Scheme 2; product 3), thus enabling significantly between 19A12NADH and CM1NADH
.
an in situ resolution of the (R)-enantiomers [see 19A12NADH/RE-ADH produces twice the amount of
Scheme 2; product (R)-2]. The second oxidation is ac- heptanones (0.85 mM vs. 0.43 mM heptanol), whereas
companied by the reduction of NAD+ to NADH. The CM1NADH/RE-ADH produces twice the amount of
regenerated cofactor NADH can be reused by the heptanols (0.78 mM vs. 0.35 mM heptanone). BM3
monooxygenase for hydroxylation of n-heptane.
WTNADH achieves a product concentration of approxi-
The n-heptane conversion in a single hydroxylation mately 0.56 mM when coupled to the alcohol dehy-
reaction with WTNADH, 19A12NADH, or CM1NADH stops drogenase (Figure 3D). Nevertheless, the concentra-
nearly after 1 hour yielding product concentrations of tion of heptanones achieved with BM3 WTNADH is
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
5
ÞÞ
These are not the final page numbers!

Christina A. Mꢀller et al.
FULL PAPERS
Table 2. Obtainable yields of the RE-ADH-catalyzed resolu-
tion of (R)/(S)-heptanols in the 2nd step of the double oxida-
tion cascade.
cade an additional NADH regeneration system based
on glucose/glucose dehydrogenase (GDH) was imple-
mented (see the Supporting Information, Scheme
S1).[12b,35] Additional cofactor regeneration over glu-
cose/GDH can improve the regeneration capacity and
compensate for cofactor losses as a result of the low
coupling efficiencies of the P450 monooxygenases,
where NADH is utilized for undesired (uncoupling)
reactions.
Yield [%][a]
(R)-2-Heptanol
(R)-3-Heptanol
WTNADH
89Æ1
51Æ3
72Æ3
82Æ2
20Æ1
58Æ3
19A12NADH
CM1NADH
[a]
Figure 4 shows the results of the double oxidation
See Table 1 for details on the theoretical composition
(enantiomeric ratio er) of the heptanol mixture which with WTNADH, 19A12NADH, and CM1NADH in the pres-
serves as substrate. Conversion of (S)-2-heptanol and
(S)-3-heptanol by RE-ADH was determined to be
>99%.
ence of the d-glucose/GDH regeneration system. Pro-
duced NAD+ during the initial oxidation step is uti-
lized simultaneously by the RE-ADH and the GDH
for the second coupled oxidation/regeneration step.
very low (69 mM vs. 0.5 mM heptanol) and differs sig- As both NAD+-consuming reactions run in parallel it
nificantly from 19A12NADH and CM1NADH
.
was interesting to characterize more in detail the co-
Notably, when starting from the alkane n-heptane factor demand of each regenerating enzyme. There-
the double oxidation concept also allows the produc- fore, the KM and Vmax values for NAD+ of both en-
tion of (R)-alcohols due to the selectivity of the RE- zymes were determined. The KM value of the RE-
ADH. Table 2 summarizes the yields obtained on (R)- ADH for NAD+ using 2-heptanol as substrate is 84Æ
2- and (R)-3-heptanol in the double oxidation cas- 9 mM and Vmax reaches 0.005 secÀ1. The GDH from
cade. The yields from the enzymatic resolution experi- Pseudomonas sp. displays a lower affinity for NAD+
ment are calculated based on the overall amount of with a KM of 113Æ15 mM when d-glucose is used as
the chiral [(R)/(S)-2- and (R)/(S)-3-heptanols] and substrate, and achieves a nearly 3-fold higher Vmax of
non-chiral products (2- and 3-heptanones), which 0.017 secÀ1 compared to RE-ADH (see the Support-
were set to 100% disregarding the conversion of the ing Information, Figure S7).
non-chiral 1- and 4-heptanols. The highest yields on
Based on the KM and Vmax values it was deduced
(R)-heptanols were obtained for BM3 WTNADH [89% that a key performance parameter was to adjust the
(R)-2-heptanol], followed by CM1NADH [72% (R)-2- glucose concentration to ensure sufficient NADH
and 58% (R)-3-heptanol]. Comparably low yields supply for the RE-ADH-catalyzed oxidation in order
were achieved with 19A12NADH [51% (R)-2- and 20% to maximize product formation (see the Supporting
(R)-3-heptanol].
Information, Figure S8). The best result in terms of
The calculated yields for the resolution of (R)-hep- product concentration and ratio between heptanone
tanols are higher than the expected values from the and heptanol was obtained with a d-glucose concen-
single, non-coupled hydroxylation reaction (Table 2 tration between 2 and 5 mM. A concentration of
vs. Table 1). The differences are caused by a possible 2 mM was finally employed for the enzyme cascade as
shift in the product distribution during the double oxi- shown in Figure 4.
dation due to the regeneration and therefore higher
availability of the cofactor.
Total product concentrations in the double oxida-
tion cascades with supporting cofactor regeneration
In order to explore whether the cofactor regenera- (d-glucose/GDH) are significantly higher. In the case
tion is a limiting factor in the double oxidation cas- of BM3 WTNADH a maximal product concentration of
Figure 4. Double oxidation of n-heptane with a supporting regeneration system (d-glucose/GDH) for the cofactor NADH.
Panels A–C show: (A) BM3 WTNADH, (B) 19A12NADH and (C) CM1NADH. Total heptanol ( ) and total heptanone ( ) concen-
trations are displayed as sum of the corresponding regioisomers which were quantified via GC after a reaction time of 0.17,
0.5, 1, 2, 4 and 24 h (see details in the Experimental Section).
*
^
6
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

In Vitro Double Oxidation of n-Heptane with Direct Cofactor Regeneration
Table 3. Total turnover numbers of n-heptane oxidation with BM3 WTNADH, 19A12NADH and CM1NADH in the absence of RE-
ADH (entry 1), in presence of RE-ADH (entry 2) and in presence of RE-ADH with additional NADH regeneration (d-glu-
cose/GDH; entry 3).
Entry
TTN[a]
WTNADH
19A12NADH
CM1NADH
1[b]
2[c]
3[d]
only P450
P450+RE-ADH
P450+RE-ADH+GDH
462Æ56
1128Æ27
1457Æ24
1374Æ2
1247Æ30
2196Æ167
3591Æ186
2514Æ92
3146Æ213
[a]
TTN: total turnover number (mmol total product·mmol catalystÀ1); total product is the sum of all quantified heptanol and
heptanone isomers (24 h).
The reaction mixture contained n-heptane (50 mM) in KPi buffer (1 mL, 100 mM; pH 8.0), ethanol (1.3% v/v), NADH
(2 mM), catalase (600 U), P450 BM3 (0.5 mM).
Additionally to entry 1: RE-ADH (10 U).
Additionally to entry 1: RE-ADH (10 U), d-glucose (2 mM) and GDH (1 U).
[b]
[c]
[d]
0.73 mM is obtained (215% higher than without
Enzymatic cascade reactions are interesting as an
GDH; Figure 4A). The variants 19A12NADH and alternative to chemical synthesis routes in that chal-
CM1NADH produced 1.57 and 1.8 mM total product, lenges in cofactor supply, insufficient coupling rates,
which correspond to a 129% and 188% increased and stability issues can be addressed successfully. In
product formation (Figure 4B and C). Product distri- this regard, significant progress has been achieved to
bution patterns in the double oxidation cascades as improve stabilities and coupling rates of monooxyge-
well as conversion and yields in the second oxidation nases by rational design and/or directed evolu-
step remained within standard deviations in the pres- tion.[23c,25,36] The concept of the double oxidation cas-
ence and absence of the d-glucose/GDH regeneration cade using molecular oxygen as sole oxidant and
system.
direct regeneration of the cofactor NADH is syntheti-
Based on the quantified product concentrations cally attractive since use of additional cosubstrates
after a reaction time of 24 h, the total turnover num- can be omitted or at least be reduced.
bers (TTN) were determined for the single and the
double oxidation reactions as well as for the double was selected as substrate for the coupled direct oxida-
oxidation reactions with supporting cofactor regenera- tion with P450 BM3 monooxygenases (WTNADH
tion (d-glucose/GDH) (see Table 3). The employed 19A12NADH, CM1NADH) and the (S)-enantioselective
monooxygenases (WTNADH, 19A12NADH, CM1NADH
alcohol dehydrogenase from R. erythropolis (RE-
To validate the one-pot double oxidation n-heptane
,
)
displayed a >1.8-fold improved TTN in the double ADH). In all three P450s the cofactor specificity was
oxidation cascades compared to the single hydroxyl- switched from NADPH to NADH. With the variants
ation reaction in absence of RE-ADH (Table 3, 19A12NADH and CM1NADH not only the cofactor pref-
entry 1 vs. 2).
erence changes but also higher product yields using
Supplementation of an NADH regeneration system NADH were achieved (43% more for 19A12NADH and
with d-glucose as a co-substrate and GDH yielded 80% more for CM1NADH).
higher product concentrations for all three BM3 var-
P450 BM3 WTNADH, 19A12NADH, and CM1NADH
iants and consequently TTNs increased up to 3-fold were characterized in detail with regard to activity,
(Table 3, entry 1 vs. 3). 19A12NADH and CM1NADH ach- coupling efficiency, selectivity and product formation
ieved comparably high TTN values (up to 3591) in the oxidation of n-heptane to 1-, 2-, 3-, or 4-hepta-
which are under all tested reaction conditions more nol and in the double oxidation cascades without and
than 2-fold higher than the corresponding TTN-values with a supporting cofactor regeneration system.
of WTNADH (Table 3).
Remarkably, the regio- and stereoselectivity of
WTNADH, 19A12NADH, and CM1NADH differs signifi-
cantly (Figure 2, Table 1). For instance, WTNADH is
(R)-selective for the production of 2- and 3-heptanols,
whereas 19A12NADH and CM1NADH are (S)-selective
Discussion
Monooxygenases are rarely employed on an industrial with respect to 3-heptanol. In this case only two sub-
scale, despite their synthetic potential for oxyfunc- stitutions (R255P, P329H) are required in CM1NADH
À
tionalization of alkanes. C H bond activation with O2 to partially invert the stereoselectivity of the enzymes.
as clean oxidant has been described as a dream reac- Additionally, 19A12NADH is poorly stereoselective for
tion in organic chemistry.[11]
the 2-heptanol production in contrast to the (R)-enan-
tioselective WTNADH and CM1NADH
.
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
7
ÞÞ
These are not the final page numbers!

Christina A. Mꢀller et al.
FULL PAPERS
19A12NADH was selected as reference based on its ciency of P450 BM3 was significantly improved by
catalytic performance for the conversion of n-heptane protein engineering for NADH from 9 to 44% in this
(18 substitutions) showing a TTN of 5360Æ1020, study, and up to 95% (NADPH) in others.[25] The re-
which is to the best of our knowledge the highest re- gioselectivity of P450 BM3 could also be improved
ported value for biocatalytic hydroxylation of this significantly for non-native substrates like limon-
compound.[24] Our total turnover numbers for
ene,[41] steroids,[42] and 2-arylacetic acid esters.[43] In
ACHTUNGTRENNUNG
19A12NADH reach up to 3150. Differences to the previ- this regard, tuning of P450 enzymes for non-native
ously reported TTN might be attributed to the altered transformations was recently reviewed in a compre-
cofactor specificity (NADH instead of NADPH) and hensive way by Fasan.[44]
the employed regeneration system (isocitrate dehy-
drogenase vs. GDH) as well as the reaction condi- opens up the possibility of a non-racemic resolution
tions. of enantiomers starting from an n-alkane substrate
The stereopreferences of the employed ADH
The catalytic characterization of BM3 CM1NADH and yielding (R)- or (S)-alcohols as value added prod-
showed a comparably high activity with n-heptane ucts in double oxidation cascade reactions. The specif-
(TOF value of 604 minÀ1 or 10 sÀ1 with NADH; Fig- icity of the RE-ADH [(S)-selective], which accepts
ure 1B) as found for other engineered P450 BM3 var- only (S)-alcohols for NADH production reduces the
iants for the oxidation of non-natural substrates like amount of regenerated cofactor. In an ideal case two
octane (variant 53-5H, TOF of 660 minÀ1 using enzymes with matching KM values and specificities,
NADPH).[37] A fine compilation of kcat values for and a highly coupled cofactor oxidation would be suf-
P450 BM3 is reported in the review by Whitehouse ficient to overcome constraints in the cofactor availa-
et al.[23b]
bility.
The oxyfunctionalization of alkanes under mild
The developed double oxidation concept is a tuna-
conditions was summarized in a recent review.[38] Oxi- ble and modular system in which the employed en-
dation of n-heptane to the corresponding alcohols zymes (monooxygenase and alcohol dehydrogenase)
and ketones was reported for a MnACTHNUTRGNE(NUG TDCPP)Cl com- can be exchanged depending on the targeted specifici-
plex together with H2O2 as oxidant with a TTN of ty and selectivity. General options are to focus on the
202.[38,39] Total turnover numbers for 19A12NADH and resolution of enantiomeric mixtures for the produc-
CM1NADH are 15-fold higher (see Table 2) and the tion of chiral alcohols or to maximize the ketone pro-
modular concept with an RE-ADH allows us in addi- duction.
tion to produce chiral alcohols.
Some general trends for the described double oxi-
dation cascade transformations of n-heptane were Conclusions
identified. Coupling of two oxidation reactions with
direct cofactor regeneration boosts the product for- A double oxidation cascade concept using the trans-
mation by more than 1.8-fold (up to 1.3 mM/ formation of n-heptane into chiral heptanols and hep-
150 mgLÀ1 total product; Figure 3). The cofactor recy- tanone as a model reaction has been developed by
cling during the second oxidation step enables an im- coupling a monooxygenase- and an alcohol dehydro-
proved product formation, which is reflected in higher genase-catalyzed reaction with and without a support-
total turnover numbers (up to 2500) compared to the ing cofactor regeneration system. The cascade oxida-
single hydroxylation reaction. The product formation tion reactions were performed in water at room tem-
can however be further increased (3-fold) to 1.8 mM/ perature using molecular oxygen as an oxidant and
207 mgLÀ1 by means of a supporting cofactor regener- the obtained TTNs (>3000) exceed the ones from
ation system; in the case of d-glucose/GDH TTNs up chemical catalysts by 15-fold.
to 3500 without changes in the stereoselectivities were
obtained (Figure 4A–C).
Interestingly, the stereopreferences for n-heptane
hydroxylation were partially inverted by substitutions
These results demonstrate that the direct regenera- R255P and P329H. The reported bicatalytic cascade is
tion of NADH through enzyme-coupling compensates especially attractive for cell-free chemical syntheses,
the losses of the cofactor due to the uncoupling of the can be handled in chemical synthesis labs (without bi-
monooxygenase-catalyzed oxidation. The coupling ef- osafety specifications) and represents the first concept
ficiencies of P450 BM3 achieved in this work are in which a chiral resolution of alcohols is implement-
comparable to those of previous evolutionary studies ed after functionalization of an alkane.
for non-natural substrates which are often below
The double oxidation concept employs alcohol de-
80%.[23c,40] Nevertheless, enhancing coupling efficiency hydrogenases and monooxygenases in a modular
of monooxygenases and thereby reducing the amount manner. As a toolbox it will enable, for instance, the
of generated hydrogen peroxide remains an important synthesis of enantiomerically pure alcohols or the cor-
challenge to improve process stability and total turn- responding ketones by employing alternative and
over numbers of P450s in general. The coupling effi- highly selective enzymes like the ADH from Lactoba-
8
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

In Vitro Double Oxidation of n-Heptane with Direct Cofactor Regeneration
cillus brevis (Lb-ADH).[34] The concept of double oxi-
dation cascades can further be expanded to other ste-
reoselective oxidoreductases or further enzyme
classes such as aminotransferases or lyases.
GCTTATGGCCAACTGCTCATGCGTTTTCCCTATATGC-3’,
P329H-reverse: 5’-GCATATAGGGAAAACGCATGAGCAGTTGGC
CATAAGC-3’.
Construction of NADH-Dependent Mutants
To switch cofactor specificity of the P450 BM3 variants the
mutations W1046S and R966D were introduced in the P450
BM3 genes (wild type, 19A12 and CM1) by site-directed
mutagenesis, using following primers: W1046S-forward: 5’-
GATACGCAAAAGACGTGAGCGCTGGGTAATAAGAATTCG-3’,
Experimental Section
Chemicals and Reagents
All chemicals used were of analytical grade or higher and
were purchased, if not stated otherwise, from Sigma Aldrich
(Steinheim, Germany), AppliChem (Darmstadt, Germany),
TCI Europe (Eschborn, Germany) and Carl Roth (Karls-
ruhe, Germany). Enzymes and dNTPs were obtained from
Fermentas (St. Leon-Rot, Germany), New England Biolabs
(Frankfurt, Germany) and Sigma Aldrich (Steinheim, Ger-
many). The HPLC-purified oligonucleotides employed for
mutagenesis were procured from Eurofins MWG Operon
(Ebersberg, Germany).
W1046S-reverse:
CACGTCTTTTGCGTATC-3’, R966D-forward: 5’-GCTTCA
TACCGCTTTTTCTGACATGCCAAATCAGCC-3’, R966D-re-
5’-CGAATTCTTATTACCCAGCGCT
verse: 5’-GGCTGATTTGGCATGTCAGAAAAAGCGGTATGAAGC-
3’.
The mutagenic PCR using a two-stage modified Quik-
Change protocol and the transformation procedure were
performed as described above. Correct insertion of both mu-
tations was corroborated by sequencing.
Expression and Purification of P450 BM3 WTNADH
19A12NADH and CM1NADH Mutants
,
Strains and Expression Vectors
Cloning of the three P450 BM3 monooxygenase variants
used in this work (WT, 19A12 and CM1) was performed
using the pET28a(+)-derived pALXtreme-1a vector and
transformation into chemically competent E. coli BL21
Gold (DE3) lacIQ1 cells.[45] The construction of the P450
BM3 monooxygenase variant harbouring the 19A12 gene[25]
was already described elsewhere.[30] The expression vector
pKA1 harbouring the (S)-alcohol dehydrogenase from Rho-
dococcus erythropolis (RE-ADH)[27b] was transformed into
chemically competent E. coli BL21 Gold (DE3) cells. Chem-
ically competent cells were prepared using a standard proto-
col[46] yielding a transformation efficiency of 1ꢃ107 and 5ꢃ
106 cfumgÀ1 pUC19, respectively.
Expression of the NADH-dependent monooxygenase mu-
tants and preparation of crude cell extracts for further pu-
rification were performed as previously described.[30] The
fresh cell lysates were subsequently purified by means of
anion exchange chromatography using a Toyopearl DEAE
650S matrix (Tosoh Bioscience, Stuttgart, Germany) and an
established protocol.[48] An ꢄktaprime Plus system with UV
detector (GE Healthcare, Mꢀnchen, Germany) was used for
purification and collection of the eluate. The eluted protein
fractions were pooled together and concentrated using an
Amicon Ultra-4 centrifugation tube (30 kDa cut-off; Milli-
pore, Schwalbach, Germany), followed by desalting and
buffer exchange in KPi buffer (100 mM; pH 8.0) using
a PD-10 gel-filtration column (GE Healthcare, Mꢀnchen,
Germany). A purity of 90% was estimated after electropho-
resis in a 10% sodium dodecyl polyacrylamide (SDS-
PAGE)[49] gel. The purified monooxygenase mutants were fi-
nally shock-frozen in liquid nitrogen, lyophilized for 48 h at
À548C in an Alpha 1–2 LD plus Freeze dryer (Christ, Oster-
ode am Harz, Germany) and stored at À208C until further
use.
Generation of the Mutant P450 BM3 CM1
Altogether 18 previously reported positions of the P450
BM3 monooxygenase variant 19A12[25] were saturated in the
BM3 WT using NNK-degenerated oligonucleotides.[31]
A
two-stage PCR was performed according to the modified
QuikChange Site-Directed mutagenesis protocol.[47] Expres-
sion of the mutant libraries and screening for activity to-
wards n-heptane was achieved following a published proce-
dure.[23c] The reaction mixture for measuring NADPH deple-
tion[32] contained n-heptane (1 mM), DMSO as cosolvent
(4% v/v), crude cell lysate (50 mL) and NADPH (0.2 mM) in
a final volume of 250 mL KPi buffer (100 mM; pH 8.0).
These studies led to identification of the key position P255
for enhanced hydroxylation activity towards n-heptane.
Based on the NADPH depletion (relative activity) and the
sequencing results (MWG Eurofins DNA, Ebersberg, Ger-
many), the improved variant P450 BM3 containing the mu-
tation R255P was identified. A second, previously identified
beneficial substitution P329H was combined via site-directed
mutagenesis to generate the variant P450 BM3 CM1. Fol-
lowing oligonucleotide primers were used: R255-forward:
5’-GCTTGATGACGAGAACATTNNKTATCAAATTATTACATTC-
Expression and Purification of RE-ADH
The RE-ADH was expressed and partially purified accord-
ing to a published protocol[27b] with minor modifications. LB
medium containing chloramphenicol (50 mL, 1.7 mg) was
used and after reaching an OD600 of 0.5 the expression was
induced by addition of isopropyl b-d-thiogalactopyranoside
(IPTG; 25 mM) and ZnCl2 (1 mM) for 20 h at 308C. Cells
were harvested by centrifugation (10 min, 2900 g at 48C)
and the cell pellet was frozen. For preparation of the cell ex-
tract the frozen pellet (1.3 g wet cells) was then resuspended
in KPi buffer (20 mL, 100 mM; pH 6.0) prior to disruption
with an Avestin EmulsiFlex-C3 high pressure homogenizer
(Ottawa, ON, Canada) by applying three cycles of 1500 bar.
The cell debris was removed by centrifugation (15 min,
16000ꢃg at 48C) in a Sorvall RC-6 Plus centrifuge (Thermo
Scientific, Rockford, IL, USA) and the supernatant was par-
3’,
R255-reverse:
5’-GAATGTAATAATTTGATAMN
P329H-forward: 5’-
NAATGTTCTCGTCATCAAGC-3’,
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
9
ÞÞ
These are not the final page numbers!

Christina A. Mꢀller et al.
FULL PAPERS
tially purified using heat-treatment incubation for 15 min at
658C. A second centrifugation for removal of precipitated
proteins was performed. The supernatant was further clari-
fied by filtration through a 0.45 mm filter (Roth, Karlsruhe,
Germany) and stored at 48C. A purity of nearly 90% was
estimated according to SDS-PAGE visualization.[49]
Regioselectivity: Product distribution (normalized to
100%) was calculated based on GC quantification of each
regioisomer (1-, 2-, 3-, 4-heptanol) as mentioned above.
Enantioselectivity: Pure enantiomeric standards of 3-hep-
tanol were not commercially available, but the racemic 3-
heptanol could be separated on a chiral GC column (see
below). Therefore, the absolute configuration of each GC
peak was assigned based on polarimetry measurements and
correlation to the optical rotation (sign of the rotation) ac-
cording to literature values [À4.98 for (R)-(À)-3-heptanol[51]
and +5.88 for (S)-(+)-3-heptanol[52]]. Samples of each enan-
tiomer in excess were obtained after conversion of (rac)-3-
heptanol, either with the (R)-specific alcohol dehydrogenase
from Lactobacillus brevis (LbADH)[34] or the (S)-specific
ADH from Rhodococcus erythropolis (RE-ADH).[27a] The
LbADH reaction was performed using recombinant E. coli
whole cells (optical density of 22) in KPi-buffer (25 mL,
0.1M, pH 6.0), containing acetone (2 mL, 8% v/v), MgSO4
(4.25 mg), NADP+ (2 mM) and 3-heptanol (0.4M), for 24 h
at room temperature. The products were extracted twice by
addition of chloroform (10 mL). After centrifugation the or-
ganic layers were combined, dried over NaSO4 and concen-
trated to 1 mL using an evaporator IKA RV 10 basic (IKA-
Werke, Staufen, Germany). Subsequently, the extracted re-
action mixture was used for polarimetric measurement with
a Perkin–Elmer 241 MC Polarimeter (Waltham, USA) and
for chiral GC analysis.
The 2-heptanol enantiomers could not be separated on
a chiral GC column, therefore the calculation of the enan-
tiomeric ratio (er) and enantiomeric excess (ee) was per-
formed based on the lipase-catalyzed resolution of the (S)-
2-heptanol enantiomer as described by Patel et al.[53] with
some modifications. The products of the P450 BM3-cata-
lyzed hydroxylation of n-heptane were extracted in
1 volume of n-heptane (1 mL) and split into two equal vol-
umes of 0.4 mL. One volume was treated with lipase B for
acylation of the (R)-2-heptanol (20 U, 4 mg of lipase from
Candida Antarctica/acrylic resin, Sigma Aldrich, Steinheim,
Germany) and succinic anhydride (0.2 mg), one volume
without the addition of lipase as reference reaction. Both
samples were incubated in parallel (358C, 24 h, 600 rpm).
For chiral separation and quantification of (R)- and (S)-3-
heptanols, as well as analysis of the remaining (S)-2-hepta-
nol after acylation of the (R)-enantiomer, a GC-2010 Plus
with an FID detector (Shimadzu, Japan) was fitted with
a column FS-Hydrodex b-TBDAc (Macherey–Nagel GmbH,
Dueren, Germany). The following oven temperature pro-
gram was used: 508C, 5 8C minÀ1 to 958C, 9.5 min. Retention
times for (R)- and (S)-3-heptanol, respectively: 9.88,
10.01 min; for (S/R)-2-heptanol: 10.65 min (for the original
GC spectra see the Supporting Information, Figure S5 and
Figure S6).
Kinetic Characterization of the Purified P450 BM3
Monooxygenases in Terms of Product Formation and
Coupling Efficiency
The concentration of active, purified and lyophilized P450
BM3 monooxygenase was determined from the CO-differ-
ence spectra of the reduced heme iron using a standard pro-
cedure.[50] CO-difference spectra and NADH oxidation were
recorded with a Varian Cary 50 spectrophotometer (Agilent
Technologies, Darmstadt, Germany). To assay activity of the
P450 BM3 mutants NADH or NADPH depletion was moni-
tored in a cuvette at 340 nm in triplicate. Quantification of
the exact amount of consumed NAD(P)H was accomplished
using the extinction coefficient eNAD(P)H =6.22 mMÀ1 cmÀ1.
The reaction mixture consisted of n-heptane (10 mM), etha-
nol as cosolvent (2% v/v), P450 BM3 (0.2 mM) in KPi buffer
(950 mL, 100 mM; pH 8.0). After 5 min of pre-incubation the
reaction was started by addition of NADH or NADPH from
a stock solution (50 mL, 8 mM). When the cofactor was
almost consumed the reaction was quenched by addition of
concentrated HCl (100 mL, 37% v/v) and directly extracted
with methyl tert-butyl ether (MTBE; 400 mL) containing
decane (1 mM) as internal standard. The organic layer was
dried over NaSO4 and analyzed by gas chromatography
(GC) for product quantification. Maximal initial product
formation rates were calculated for the first 20 sec of the re-
action. The molecular weight value of P450 BM3 and var-
iants used in the calculations was 119 kDa.
Quantification of Heptanol and Heptanone
Production
Quantification of the reaction products was performed using
a Shimadzu GC-2010 gas chromatograph coupled to a flame
ionization detector (FID) (Shimadzu, Japan). For quantifica-
tion the GC was fitted with a column FS-Supreme-5 ms
(30 mꢃ0.25 mmꢃ0.25 mm film thickness; CS-Chromatogra-
phie, Germany) and following column oven temperature
program was used for separation: 408C for 19.5 min, 208C
minÀ1 to 1008C, 3 min. For each corresponding alcohol and
ketone a 6-point calibration curve was generated using au-
thentic standards and a defined amount of decane (1 mM)
as internal standard. Retention times for 4-, 3-, 2-hepta-
nones, 4-, 3-, 2-, 1-heptanols and decane, respectively: 13.6,
15.2, 15.8, 16.4, 17.1, 17.9, 22.7 and 23.3 min (see original
GC spectra in the Supporting Information, Figure S3).
Determination of the Selectivity of the P450 BM3
Variants
RE-ADH and GDH Activity Determination
The activity of the heat-treated RE-ADH and of the com-
mercially available glucose dehydrogenase (GDH) from
Pseudomonas sp. (Sigma Aldrich, Steinheim, Germany) was
assayed spectrometrically with a Varian Cary 50 (Agilent
Technologies, Darmstadt, Germany) by measuring the in-
crease in absorption of NADH at 340 nm.
The distribution of the regioisomers and the enantiomeric
excess (ee values) for the hydroxylation of n-heptane were
analyzed via gas chromatography. The same set up of the
hydroxylation reaction for the determination of the total
turnover number, containing solely P450 BM3, was used.
10
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

In Vitro Double Oxidation of n-Heptane with Direct Cofactor Regeneration
For RE-ADH: The reaction mixture (999 mL; cuvette)
Acknowledgements
contained the corresponding alcohol (1-, 2-, 3- or 4-hepta-
nol; 10 m M) and NAD+ (1 mM) in KPi buffer (100 mM;
pH 8.0). The reaction was started by addition of RE-ADH
(1 mL; partially purified). One unit was defined as the
amount of enzyme that reduces 1 mmol of NAD+ per
minute. Total protein concentration was determined by the
method of Bradford using bovine serum albumin as stan-
dard.[54] A control reaction with ethanol (4% v/v), which is
used in the cascade reaction as cosolvent, was also recorded
and revealed no activity of RE-ADH towards ethanol. The
KM value of the RE-ADH for the cofactor NAD+ using 2-
heptanol as substrate was determined in an analogous
manner, varying the NAD+ concentration (1 to 4000 mM).
For GDH: the reaction mixture contained KPi buffer
(1 mL, 100 mM; pH 8.0), d-glucose (10 mM) and NAD+
(1 mM). The reaction was started by addition of GDH
(1 mL). The KM value of the GDH for the cofactor NAD+
using d-glucose as substrate was determined in an analogous
manner, varying the NAD+ concentration (3 to 2000 mm).
This work was supported by Deutsche Bundesstiftung
Umwelt (ChemBioTec project AZ 13234; Nachhaltige Bioka-
talytische Oxidationsprozesse). We also would like to thank
Dr. Jan Marienhangen (FZ Jꢀlich), Dr. Anna Joelle Ruff and
Alexander Dennig (RWTH Aachen) for fruitful discussions.
References
[1] E. Roduner, W. Kaim, B. Sarkar, V. B. Urlacher, J.
Pleiss, R. Glꢁser, W.-D. Einicke, G. A. Sprenger, U.
Beifuß, E. Klemm, C. Liebner, H. Hieronymus, S.-F.
Hsu, B. Plietker, S. Laschat, ChemCatChem 2013, 5,
82–112.
[2] A. E. Shilov, G. B. Shul’pin, Chem. Rev. 1997, 97, 2879–
2932.
[3] R. H. Crabtree, J. Chem. Soc. Dalton Trans. 2001,
2437–2450.
[4] J. Smidt, W. Hafner, R. Jira, J. Sedlmeier, R. Sieber, R.
Rꢀttinger, H. Kojer, Angew. Chem. 1959, 71, 176–182.
[5] G. O. Spessard, G. L. Miessler, Organometallic Chemis-
try, 1st edn., Prentice-Hall, New Jersey, 1996.
One-Pot Double Oxidation of n-Heptane by P450
BM3 and RE-ADH, with and without
Supplementation of GDH
[6] a) A. Hirao, S. Itsuno, S. Nakahama, N. Yamazaki, J.
Chem. Soc. Chem. Commun. 1981, 315–317; b) E. J.
Corey, R. K. Bakshi, S. Shibata, C. P. Chen, V. K.
Singh, J. Am. Chem. Soc. 1987, 109, 7925–7926.
[7] W. Tang, X. Zhang, ChemInform 2003, 34, 3029–3069.
[8] G. W. Huisman, J. Liang, A. Krebber, Curr. Opin.
Chem. Biol. 2010, 14, 122–129.
Conversion experiments and determination of total turnover
number (TTN): To determine product formation the in vitro
double oxidation of n-heptane was monitored over the first
24 h (1 mL, stirred flask, room temperature). The reaction
mixture contained n-heptane (7 mL, 50 mM) and ethanol
(13 mL, 1.3% v/v) as cosolvent for enhanced solubility of the
organic substrate, P450 BM3 (0.5 mM), RE-ADH (10 U, re-
ferring to the activity determined with 2-heptanol) and cata-
lase (600 U, to remove traces of hydrogen peroxide) in KPi
buffer (1 mL, 100 mM; pH 8.0). After an incubation period
of 5 min the reaction was started by addition of NADH
(2 mM). After reaction times of 10, 30, 60, 120, 240 min and
24 h, the reaction mixture (1 mL) was transferred to a reac-
tion vessel containing HCl (100 mL) for stopping the reac-
tion and the aqueous phase was directly extracted with
MTBE (400 mL, 1 mM decane as internal standard) for GC
quantification as described above. In parallel, a second oxi-
dation reaction containing an additional NADH-regenera-
tion system was performed. In this case glucose dehydrogen-
ase (1 U, referring to the activity determined with d-glucose)
and d-glucose (2 mM) were added to the reaction mixture.
As a reference reaction the sole hydroxylation of n-heptane,
using only P450 BM3 as catalyst was conducted in the same
way.
[9] T. Daußmann, H. G. Hennemann, T. C. Rosen, P. Dꢀn-
kelmann, Chem. Ing. Tech. 2006, 78, 249–255.
[10] J. A. Labinger, J. E. Bercaw, Nature 2002, 417, 507–514.
[11] R. A. Sheldon, I. Arends, U. Hanefeld, Green chemistry
and catalysis, Wiley-VCH, Weinheim, 2007.
[12] a) A. Schmid, J. S. Dordick, B. Hauer, A. Kiener, M.
Wubbolts, B. Witholt, Nature 2001, 409, 258–268; b) A.
Liese, K. Seelbach, C. Wandrey, Industrial Biotransfor-
mations, 2nd edn., Wiley-VCH, Weinheim, 2006.
[13] H. Arakawa, M. Aresta, J. N. Armor, M. A. Barteau,
E. J. Beckman, A. T. Bell, J. E. Bercaw, C. Creutz, E.
Dinjus, D. A. Dixon, K. Domen, D. L. DuBois, J.
Eckert, E. Fujita, D. H. Gibson, W. A. Goddard, D. W.
Goodman, J. Keller, G. J. Kubas, H. H. Kung, J. E.
Lyons, L. E. Manzer, T. J. Marks, K. Morokuma, K. M.
Nicholas, R. Periana, L. Que, J. Rostrup-Nielson, W. M.
Sachtler, L. D. Schmidt, A. Sen, G. A. Somorjai, P. C.
Stair, B. R. Stults, W. Tumas, Chem. Rev. 2001, 101,
953–996.
[14] F. Hollmann, I. W. C. E. Arends, K. Buehler, A. Schall-
mey, B. Bꢀhler, Green Chem. 2011, 13, 226–265.
[15] B. Bꢀhler, A. Schmid, J. Biotechnol. 2004, 113, 183–210.
[16] S. Peter, M. Kinne, X. Wang, R. Ullrich, G. Kayser,
J. T. Groves, M. Hofrichter, FEBS J. 2011, 278, 3667–
3675.
Yields for resolution of (R)-heptanols: Yields were calcu-
lated for each (R)-heptanol as product of the second RE-
ADH-catalyzed oxidation reaction, starting from the enan-
tiomeric mixture [(R)/(S)-2-heptanol or (R)/(S)-3-heptanol]
and not heptane as substrate. For the calculation of yields
conversion of the non-chiral 1- and 4-heptanol was not con-
sidered. The products were quantified using the area of each
enantiomeric peak by chiral GC analysis as mentioned
above.
[17] J. B. van Beilen, W. A. Duetz, A. Schmid, B. Witholt,
Trends Biotechnol. 2003, 21, 170–177.
[18] A. Weckbecker, H. Grçger, W. Hummel, Adv. Bio-
chem. Eng./Biotechnol. 2010, 120, 195–242.
[19] a) F. Hollmann, K. Hofstetter, A. Schmid, Trends Bio-
technol. 2006, 24, 163–171; b) S. Z. Cekic, D. Holtmann,
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
11
ÞÞ
These are not the final page numbers!

Christina A. Mꢀller et al.
FULL PAPERS
G. Gꢀven, K. M. Mangold, U. Schwaneberg, J. Schrad-
er, Electrochem. Commun. 2010, 12, 1547–1550.
[20] I. G. Denisov, T. M. Makris, S. G. Sligar, I. Schlichting,
Chem. Rev. 2005, 105, 2253–2277.
[35] H. Grçger, S. Borchert, M. Kraußer, W. Hummel, in:
Encyclopedia of Industrial Biotechnology, (Ed.: M.
Flickinger), John Wiley & Sons, New York 2010,
pp 2094–2110.
[21] D. W. Nebert, F. J. Gonzalez, Annu. Rev. Biochem.
[36] T. S. Wong, F. H. Arnold, U. Schwaneberg, Biotechnol.
Bioeng. 2004, 85, 351–358.
[37] P. Meinhold, M. W. Peters, M. M. Chen, K. Takahashi,
F. H. Arnold, ChemBioChem 2005, 6, 1765–1768.
[38] M. Bordeaux, A. Galarneau, J. Drone, Angew. Chem.
2012, 124, 10870–10881; Angew. Chem. Int. Ed. 2012,
51, 10712–10723.
1987, 56, 945–993.
[22] L. O. Narhi, A. J. Fulco, J. Biol. Chem. 1987, 262, 6683–
6690.
[23] a) K. L. Tee, U. Schwaneberg, Comb. Chem. High
Throughput Screening 2007, 10, 197–217; b) C. J. White-
house, S. G. Bell, L. L. Wong, Chem. Soc. Rev. 2012, 41,
1218–1260; c) A. Dennig, J. Marienhagen, A. J. Ruff, L.
Guddat, U. Schwaneberg, ChemCatChem 2012, 4, 771–
773; d) O. Salazar, P. C. Cirino, F. H. Arnold, ChemBio-
Chem 2003, 4, 891–893.
[24] R. Fasan, Y. T. Meharenna, C. D. Snow, T. L. Poulos,
F. H. Arnold, J. Mol. Biol. 2008, 383, 1069–1080.
[25] R. Fasan, M. M. Chen, N. C. Crook, F. H. Arnold,
Angew. Chem. 2007, 119, 8566–8570; Angew. Chem.
Int. Ed. 2007, 46, 8414–8418.
[26] S. Staudt, E. Burda, C. Giese, C. A. Mꢀller, J. Marien-
hagen, U. Schwaneberg, W. Hummel, K. Drauz, H.
Grçger, Angew. Chem. 2013, 125, 2415–2419; Angew.
Chem. Int. Ed. 2013, 52, 2359–2363.
[27] a) K. Abokitse, W. Hummel, Appl. Microbiol. Biotech-
nol. 2003, 62, 380–386; b) W. Hummel, K. Abokitse, K.
Drauz, C. Rollmann, H. Grçger, Adv. Synth. Catal.
2003, 345, 153–159.
[28] In this concept the term ꢅenzymatic resolutionꢆ refers
to the separation of enantiomers in a non-racemic mix-
ture of (S)- and (R)-heptanols (the composition will
depend on the enantioselectivity of the employed P450
BM3 monooxygenase variant).
[29] S. C. Maurer, K. Kuhnel, L. A. Kaysser, S. Eiben, R. D.
Schmid, V. B. Urlacher, Adv. Synth. Catal. 2005, 347,
1090–1098.
[30] S. Staudt, C. A. Mꢀller, J. Marienhagen, C. Boing, S.
Buchholz, U. Schwaneberg, H. Grçger, Beilstein J. Org.
Chem. 2012, 8, 186–191.
[31] A. E. Firth, W. M. Patrick, Nucleic Acids Res. 2008, 36,
W281–W285.
[39] A. Thellend, P. Battioni, D. Mansuy, J. Chem. Soc.
Chem. Commun. 1994, 1035–1036.
[40] a) C. J. Whitehouse, S. G. Bell, H. G. Tufton, R. J.
Kenny, L. C. Ogilvie, L. L. Wong, Chem. Commun.
2008, 966–968; b) J. A. Dietrich, Y. Yoshikuni, K. J.
Fisher, F. X. Woolard, D. Ockey, D. J. McPhee, N. S.
Renninger, M. C. Chang, D. Baker, J. D. Keasling, ACS
Chem. Biol. 2009, 4, 261–267; c) V. B. Urlacher, A.
Makhsumkhanov, R. D. Schmid, Appl. Microbiol. Bio-
technol. 2006, 70, 53–59.
[41] A. Seifert, M. Antonovici, B. Hauer, J. Pleiss, Chem-
BioChem 2011, 12, 1346–1351.
[42] J. C. Lewis, S. M. Mantovani, Y. Fu, C. D. Snow, R. S.
Komor, C. H. Wong, F. H. Arnold, ChemBioChem
2010, 11, 2502–2505.
[43] M. Landwehr, L. Hochrein, C. R. Otey, A. Kasrayan,
J. E. Bꢁckvall, F. H. Arnold, J. Am. Chem. Soc. 2006,
128, 6058–6059.
[44] R. Fasan, ACS Catal. 2012, 2, 647–666.
[45] M. Blanusa, A. Schenk, H. Sadeghi, J. Marienhagen, U.
Schwaneberg, Anal. Biochem. 2010, 406, 141–146.
[46] H. Inoue, H. Nojima, H. Okayama, Gene 1990, 96, 23–
28.
[47] W. Wang, B. A. Malcolm, BioTechniques 1999, 26, 680.
[48] U. Schwaneberg, A. Sprauer, C. Schmidt-Dannert,
R. D. Schmid, J. Chromatogr. A 1999, 848, 149–159.
[49] U. K. Laemmli, S. F. Quittner, Virology 1974, 62, 483–
499.
[50] T. Omura, R. Sato, J. Biol. Chem. 1964, 239, 2379–2385.
[51] J. Zhong, H. Guo, M. Wang, M. Yin, M. Wang, Tetrahe-
dron: Asymmetry 2007, 18, 734–741.
[32] A. Glieder, P. Meinhold, in: Directed Enzyme Evolu-
tion – Screening and Selection Methods, (Eds.: F. H.
Arnold, G. Georgiou), Humana Press Inc., Totowa,
New Jersey 2003, pp 157–170.
[52] E. Keinan, E. K. Hafeli, K. K. Seth, R. Lamed, J. Am.
Chem. Soc. 1986, 108, 162–169.
[53] R. N. Patel, A. Banerjee, V. Nanduri, A. Goswami,
F. T. Comezoglu, J. Am. Oil Chem. Soc. 2000, 77, 1015–
1019.
[33] R. Neeli, O. Roitel, N. S. Scrutton, A. W. Munro, J.
Biol. Chem. 2005, 280, 17634–17644.
[34] W. Hummel, Adv. Biochem. Eng./Biotechnol. 1997, 58,
145–184.
[54] M. M. Bradford, Anal. Biochem. 1976, 72, 248–254.
12
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 0000, 000, 0 – 0
ÝÝ
These are not the final page numbers!

FULL PAPERS
In Vitro Double Oxidation of n-Heptane with Direct
13
Cofactor Regeneration
Adv. Synth. Catal. 2013, 355, 1 – 13
Christina A. Mꢀller, Beneeta Akkapurathu, Till Winkler,
Svenja Staudt, Werner Hummel, Harald Grçger,
Ulrich Schwaneberg*
Adv. Synth. Catal. 0000, 000, 0 – 0
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
13
ÞÞ
These are not the final page numbers!