P450 BM3 Monooxygenase as an Efficient NAD(P)H-Oxidase for Regeneration of Nicotinamide Cofactors in ADH-Catalysed Preparative Scale Biotransformations

Article abstract of DOI:10.1002/adsc.201600241

Enzymatic oxidations of primary and secondary alcohols catalysed by nicotinamide dependent alcohol dehydrogenases on the preparative scale require cofactor regeneration systems. Of critical value from an economic and ecological perspective is the application of NAD(P)H-oxidases, which utilise molecular oxygen as a cost-effective, atom-efficient and environmentally benign oxidant to regenerate the cofactor NAD(P)⁺. Herein, the P450 BM3 monooxygenase from Bacillus megaterium is presented as an NAD(P)H-oxidase for the successful regeneration of both NADP⁺and NAD⁺on the preparative scale. This enzyme was exemplarily applied for ADH-catalysed oxidative kinetic resolutions of racemic secondary alcohols and the desymmetrisation of a meso-diol leading to enantiomerically enriched secondary alcohols in both cases. Furthermore, the ADH-catalysed oxidation of a primary alcohol targeting the corresponding aldehyde was performed. The obtained results significantly broaden the scope of feasible oxidative biotransformations, thereby increasing the number of synthetic reactions complying with key challenges of a modern and sustainable chemistry such as mild reaction conditions, environmentally benign solvents, and biodegradable non-toxic catalysts.

Full text of DOI:10.1002/adsc.201600241

FULL PAPERS
DOI: 10.1002/adsc.201600241
P450 BM3 Monooxygenase as an Efficient NAD(P)H-Oxidase
for Regeneration of Nicotinamide Cofactors in ADH-Catalysed
Preparative Scale Biotransformations
Claudia Holec,a Katharina Neufeld,a and Jçrg Pietruszkaa,b,*
a
Institut für Bioorganische Chemie, Heinrich-Heine-Universität Düsseldorf im Forschungszentrum Jülich, Stetternicher
Forst, Geb. 15.8, 52426 Jülich, Germany
E-mail: j.pietruszka@fz-juelich.de
Institut für Bio- und Geowissenschaften (IBG-1: Biotechnologie), Forschungszentrum Jülich, 52426 Jülich, Germany
b
Received: March 2, 2016; Revised: March 23, 2016; Published online: April 24, 2016
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201600241.
Abstract: Enzymatic oxidations of primary and sec- a meso-diol leading to enantiomerically enriched sec-
ondary alcohols catalysed by nicotinamide depen- ondary alcohols in both cases. Furthermore, the
dent alcohol dehydrogenases on the preparative ADH-catalysed oxidation of a primary alcohol tar-
scale require cofactor regeneration systems. Of criti- geting the corresponding aldehyde was performed.
cal value from an economic and ecological perspec- The obtained results significantly broaden the scope
tive is the application of NAD(P)H-oxidases, which of feasible oxidative biotransformations, thereby in-
utilise molecular oxygen as a cost-effective, atom-ef- creasing the number of synthetic reactions complying
ficient and environmentally benign oxidant to regen- with key challenges of a modern and sustainable
erate the cofactor NAD(P)+. Herein, the P450 BM3 chemistry such as mild reaction conditions, environ-
monooxygenase from Bacillus megaterium is present- mentally benign solvents, and biodegradable non-
ed as an NAD(P)H-oxidase for the successful regen- toxic catalysts.
eration of both NADP+ and NAD+ on the prepara-
tive scale. This enzyme was exemplarily applied for Keywords: alcohol dehydrogenases; biotransforma-
ADH-catalysed oxidative kinetic resolutions of race- tions; cofactor regeneration; enzyme catalysis;
mic secondary alcohols and the desymmetrisation of NAD(P)H-oxidase
Introduction
regio-, and chemoselectivity.[8] Interestingly, opposite
to this trend and even though the previously men-
The synthesis of optically pure secondary alcohols as tioned success of biocatalysts highlights their exten-
chiral synthons for the preparation of pharmaceuticals sive potential, the field of oxidative transformations is
and as building blocks for natural compounds is an dominated by classical chemistry.[9,10] Most examples
important goal in synthetic organic chemistry.[1] Over for oxidative kinetic resolutions (OKRs) of secondary
the last few decades significant progress has been alcohols rely on transition metal catalysts based on
made in developing a variety of methods to obtain Pd, Ru, Ir, V, or Fe, but also organocatalysed ap-
these compounds including kinetic resolutions of rac- proaches are known.[11–16]
emic mixtures, desymmetrisation of meso-compounds,
A main limitation hampering the application of en-
asymmetric reductions of ketones, and de novo func- zymes for oxidative biotransformations is the lack of
tionalisations.[2–5] Thereby, enzymatic routes based on efficient NAD(P)+ regeneration systems and the fact
alcohol dehydrogenases (ADHs, EC 1.1.1.x) represent that a vast majority of ADHs require stoichiometric
arguable the most efficient approach with regard to amounts of these expensive nicotinamide cofactors
reductive transformations.[6,7] Besides complying with for the preparative scale oxidation of primary and
key challenges of a modern and green chemistry by secondary alcohols.[8,17–20] Methods developed in
enabling mild reaction conditions, sustainable re- recent years to regenerate NAD(P)+ comprise sub-
agents, environmentally benign solvents, and biode- strate-coupled and enzyme-coupled systems as the
gradable catalysts, enzymes can be superior to con- most common approaches, but also chemical, electro-
ventional chemical catalysts due to higher enantio-, chemical and light-mediated solutions.[17,19–25] In the
Adv. Synth. Catal. 2016, 358, 1810 – 1819
1810
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPERS
asc.wiley-vch.de
substrate-coupled system, the applied ADH catalyses
Our study highlights the potential of the bacterial
simultaneously the desired oxidation [NAD(P)+! P450 BM3 monooxygenase as a novel and efficient
NAD(P)H] and the reduction of a ketone or aldehyde NAD(P)H-oxidase for the regeneration of NADP+
[NAD(P)H!NAD(P)+], which is added in excess as and NAD+ by demonstrating its successful application
a co-substrate to shift the equilibrium towards prod- in ADH-catalysed oxidations of structurally different
uct formation.[24,26] In enzyme-coupled systems, the secondary (OKR and oxidative desymmetrisation)
analogous reduction of a ketone or aldehyde as co- and primary alcohols.
substrate is catalysed by an additional dehydrogen-
ase.[18,27,28] The preference of NAD(P)+-dependent
enzyme reactions towards reduction rather than oxi- Results and Discussion
dation challenges the establishment of these sys-
tems.[26] In addition, application of large amounts of P450 BM3 for NADP+ Regeneration
co-substrate is limited to ADHs with high stability to
organic compounds.[7]
Discovery of NADPH-Oxidase Activity of P450 BM3
NAD(P)H-oxidases (EC 1.6.3.1 NOX) represent Wild-Type
a promising solution to the previously mentioned bot-
tlenecks and have already been used for enzyme-cou- Enantiopure homoallylic alcohols are prominent mo-
pled cofactor regeneration systems converting tives abundantly present in complex natural and bio-
NAD(P)H to NAD(P)+.[26,29–32] In addition to the pre- logically active compounds.[44–48] Aiming at the synthe-
viously mentioned benefits of biocatalysts, application sis of one of these natural products in the course of
of these enzymes enables the use of molecular oxygen another project, we faced the need to access (S)-1-
abundantly present in air as an ideal and cost-effec- nonen-4-ol [(S)-1] as a respective building block. Ini-
tive oxidant acting as the final electron and hydrogen tially, a reductive enzymatic approach starting from
acceptor, thereby lacking toxic by-products.[11b] Fur- the corresponding ketone 2 was envisaged and screen-
thermore, the thermodynamic driving force shifts the ing of available ADHs revealed an enzyme from
reaction equilibrium towards product formation due Thermoanaerobacter brockii (TbADH) to be highly
to the high redox potential of oxygen; oxygen reduc- enantioselective; unfortunately, the opposite enantio-
tion is generally recognised as irreversible.[24a,26] mer (R)-1 was obtained. In addition, conventional
Thereby, hydrogen peroxide is formed in the case of chemical reducing approaches failed in common with
a two-electron-transfer and addition of catalase en- lipase-catalysed kinetic resolutions of the racemic al-
sures degradation of this product to prevent enzymes cohol rac-1 due to insufficient catalyst selectivity,
deactivation through this compound. In the case of making (S)-1-nonen-4-ol [(S)-1] a sophisticated target.
a four-electron transfer, water is formed as the final Considering the lack of alternatives, an OKR of the
product. Recently, Petschacher et al. engineered an racemic alcohol rac-1 using the previously identified
NADH-oxidase from Streptococcus mutants for effi- TbADH seemed to be the most promising approach
cient NAD+- and NADP+ regeneration and applied to produce the desired building block (S)-1. Thereby,
the enzyme to the ADH-catalysed oxidation of 2-hep- the undesired enantiomer (R)-1 is oxidised by the
tanol to 2-heptanone.[24a] Fraaije and co-workers con- enzyme to the corresponding ketone 2 along with
verted a Baeyer–Villiger monooxygenase [BVMO, NADP+-reduction to NADPH, so that alcohol (S)-1 is
flavin-containing enzyme using O2 and NAD(P)H for retained. Regeneration of the oxidised cofactor was
ketone, sulphide and heteroatom oxidation, EC planned using an enzyme-coupled approach via the
1.14.13.x.] into an NADPH-oxidase via rational pro- P450 BM3 wild-type as a co-enzyme and myristic acid
tein engineering.[33] The NADPH-oxidase activity (3) as co-substrate (Scheme 1). The frequently em-
occurs via uncoupling, which describes the process of phasised remarkably high coupling efficiency of this
oxidising NADPH without oxygenation of the sub- monooxygenase towards medium to long chain fatty
strate over a flavin-peroxide intermediate. Uncou- acids as its native substrates deemed this enzyme con-
pling reactions are also very prominent in other venient for our approach.[49,50] To identify suitable re-
classes of monooxygenases such as, for example, in action conditions for the projected OKR, we per-
the NADPH-dependent P450 monooxygenase from formed reactions on an analytical scale varying the
Bacillus megaterium (P450 BM3, EC 1.14.14.1). This monooxygenase and myristic acid (3) amount relative
self-sufficient and soluble bacterial enzyme is one of to the 1-nonen-4-ol (rac-1) content (Figure 1). In
the most frequently studied P450 enzymes with re- detail, 1–12 mM co-enzyme and 0.25–1.00 equivalents
markably high activities and levels of heterologous of co-substrate 3 were tested. The highest conversion
expression in E. coli.[34–36] This catalyst also obtained of approx. 30% was detected when 3 mM wild-type
prominence due to its chemo-, regio- and enantiose- monooxygenase were used. However, very unexpect-
lectivity in preparative scale applications for the syn- ed was the fact that the amount of added myristic
thesis of organic building blocks.[37–43]
acid (3) had no effect on the reaction outcome, which
Adv. Synth. Catal. 2016, 358, 1810 – 1819
1811
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPERS
asc.wiley-vch.de
best of our knowledge, this is the first time a P450
BM3 monooxygenase was applied as a cofactor-regen-
erating enzyme and the obtained insights urge a fur-
ther exploration of this approach. The previously un-
derestimated NADPH-oxidase activity of P450 BM3
increases the potential of this enzyme for NADP+-re-
generating applications by omitting co-substrate addi-
tion and hence, the need of co-product separation
compared to alternative enzyme-coupled cofactor re-
generation systems.
Enhanced NADPH-Oxidase Activity of P450 BM3
F87A
Encouraged by the previous results, we investigated
the efficiency of the P450 BM3 F87A mutant for
NADP+ regeneration in the course of the OKR of
racemic 1-nonen-4-ol (rac-1) in the absence of a co-
substrate. Selection of this variant was based on its
lower coupling efficiency compared to the wild-type,
which can be attributed to a higher accessibility of the
haem group of this biocatalyst due to a weaker
“shielding” of the active site by the alanine compared
to the phenylalanine at position 87.[51,52] In addition,
coordination of solvent molecules such as DMSO to
the haem iron – thereby displacing the haem water
ligand in place of a substrate molecule – presumably
decreases the coupling efficiency of this enzyme even
further.[53,54] Reactions were carried out in analogy to
the previous study on an analytical scale, but this time
varying the P450 BM3 F87A and DMSO amount rela-
tive to the 1-nonen-4-ol (rac-1) content (Figure 2). In
detail, 1–12 mM co-enzyme and 0–10 vol% of co-sol-
vent were tested. In agreement with the previous
study, the highest conversion of approx. 36% was de-
tected when 3 mM P450 BM3 F87A were used. Over-
all, higher conversions were obtained compared to
the wild-type study, thereby affirming the proposed
improved NADPH-oxidase activity of the F87A
mutant. Different from our expectations, the amount
of added DMSO had a negligible effect on the reac-
tion outcome, which indicated that addition of co-sol-
vent is not enhancing uncoupling of the monooxyge-
nase. In conclusion, we demonstrated that co-sub-
strate addition is not mandatory for P450 BM3-cata-
lysed NADPH-oxidation and P450 BM3 variants with
higher uncoupling rates such as the F87A mutant dis-
play higher suitability for cofactor regenerating appli-
cations.
Scheme 1. TbADH-catalysed OKR of racemic 1-nonen-4-ol
(rac-1) using a P450 BM3 wild-type-coupled NADP+ regen-
eration based on myristic acid (3) oxidation.
Figure 1. TbADH-catalysed OKR of racemic 1-nonen-4-ol
(rac-1) using P450 BM3 wild-type-coupled NADP+ regener-
ation based on myristic acid (3) oxidation; different co-
enzyme and co-substrate concentrations were investigated.
Reactions were performed on the analytical scale and sam-
ples were analysed by GC with respect to product forma-
tion.
indicated that addition of myristic acid (3) as a co-
substrate is non-essential. This result contradicts the
generally accepted high coupling efficiency of P450 Preparative Scale Synthesis of (S)-1-Nonen-4-ol
BM3 towards its native substrates and leads to the
conclusion that oxidation of NADPH occurs primarily To demonstrate the potential of P450 BM3 as an
through O2-reduction to water or reactive oxygen spe- NADP+-regenerating biocatalyst in preparative scale
cies via published uncoupling mechanisms.[51] To the biotransformations, we performed an OKR of racemic
Adv. Synth. Catal. 2016, 358, 1810 – 1819
1812
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPERS
asc.wiley-vch.de
a product-based calibration curve was not feasible
due to decomposition of the formed ketone 2.[55] The
acetone-based regeneration resulted in 48% conver-
sion of substrate rac-1 and the monooxygenase-based
alternative in 51%, respectively. These results con-
firmed applicability of both cofactor regeneration sys-
tems for the production of enantioenriched (S)-1-
nonen-4-ol [(S)-1]. The final conversions were used
for E value determination leading to values >100 in
both cases (E>20 is reported as a threshold value for
organic synthesis).[57] The feasibility of the pro-
nounced P450 BM3-based regeneration was further
verified by the obtained yields (Table 1, entry 1). Ap-
plication of the F87A mutant resulted in 41% yield
with 98% ee regarding the desired alcohol (S)-1 (82%
of the theoretically possible yield), whereas acetone
addition gave only 28% yield with 95% ee (56% of
the theoretically possible yield), respectively. The
lower yield of the acetone-based approach is probably
attributable to hampered extraction of compound (S)-
1 during work-up due to the presence of acetone.
Thus, this result reveals a further advantage of the
P450 BM3-based regeneration system compared to
known alternatives.
Figure 2. TbADH-catalysed OKR of racemic 1-nonen-4-ol
(rac-1) using P450 BM3 F87A-coupled NADP+ regeneration
based on ꢁuncouplingꢂ; different co-enzyme and DMSO con-
centrations were investigated. Reactions were performed on
the analytical scale and samples were analysed by GC with
respect to product formation.
Oxidative Kinetic Resolutions of 1-Phenylethanol
1-nonen-4-ol (rac-1) catalysed by TbADH using the To broaden the scope of possible applications, our
P450 BM3 F87A mutant for NADP+ regeneration; next attempt was to use the P450 BM3 F87A-based
controls verified that the oxidation of compound rac- NADP+ regeneration system for supporting the
1 is not catalysed by molecular oxygen or P450 BM3 ADH-catalysed OKR of a structurally different sub-
F87A itself. An analogous biotransformation was fur- strate. Racemic 1-phenylethanol (rac-5) was selected
ther conducted using a substrate-coupled approach as the substrate of choice due to its structurally differ-
for cofactor regeneration based on the reduction of ent aromatic character compared to the previously ex-
acetone to isopropyl alcohol by TbADH. Hereby, amined linear aliphatic 1-nonen-4-ol (rac-1)
a comparison between the unprecedented P450 BM3- (Scheme 2). In addition, ADHs for the production of
based and the well-known co-substrate-based regener- both enantiomers of compound 5 starting from the re-
ation was intended to allow a valuation of the applica- spective ketone 6 were known and their preparative
bility of the presented approach. Reaction progress scale applicability has been recently proven.[4a,58–60]
was followed by GC analysis based on the decrease of TbADH provides selectively alcohol (R)-5 and an
the starting material relative to an internal standard; enzyme from Lactobacillus brevis (LbADH) has been
Table 1. Preparative scale oxidative kinetic resolutions of racemic secondary alcohols catalysed by alcohol dehydrogenases
using P450 BM3 F87A for NADP+ regeneration.[a]
Entry
Alcohol dehydrogenase from
Substrate
Conversion [%][b]
Yield [%]
ee value [%][c]
E value[d]
1
2
3
Thermoanaerobacter brockii
Thermoanaerobacter brockii
Lactobacillus brevis
rac-1
rac-5
rac-5
51 (after 8 h)
50 (after 2 h)
48 (after 4 h)
41
49
48
98 (S)
95 (R)
99 (S)
>100
>100
>100
[a]
Reaction conditions (S=150 mL): 63 U ADH,[56] 3 mM P450 BM3 F87A (crude cell lysate), 1.5 mmol substrate,
600 UmLÀ1 catalase and 300 mM NADP+ in KPi buffer (100 mM, pH 7.0, containing 1 mM MgCl2) initially saturated with
oxygen.
Determined by GC analysis according to calibration curves based on internal standard.
Determined by GC analysis; absolute configurations were assigned by comparison with authentic samples.
ln[(1Àc)(1Àee(S))]/ln[(1Àc)(1Àee(S))][2a] with ln=natural logarithm, c=conversion and ee(S)=ee of the remaining sub-
strate.
[b]
[c]
[d]
Adv. Synth. Catal. 2016, 358, 1810 – 1819
1813
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPERS
asc.wiley-vch.de
Scheme 2. TbADH- and LbADH-catalysed preparative
scale OKRs of racemic 1-phenylethanol (rac-5) using P450
BM3 F87A-based cofactor regeneration. Reaction conditions
(S=150 mL): 63 U ADH,[56] 1.5 mmol substrate, 300 mM
NADP+, 3 mM P450 BM3 F87A and 600 UmLÀ1 catalase in
KPi buffer (100 mM, pH 7.0, containing 1 mM MgCl2) ini-
tially saturated with oxygen.
Scheme 3. RasADH-catalysed preparative scale desymmetri-
sation of meso-diol 7 to ketone (R)-8 using different
NADP+ regeneration systems. Reaction conditions (S=
150 mL): 63 U RasADH,[63] 1.5 mmol substrate and 300 mM
NADP+ with a cofactor regeneration system in KPi buffer
(100 mM, pH 7.0, containing 0.8 mM CaCl2), 24 h.[64]
reported to generate the opposite enantiomer (S)-5.[61]
The application of these enzymes was envisaged to
perform OKRs aiming at the different products (R)-5
and (S)-5, respectively. The results in Table 1 illustrate organic synthesis.[62] Recently, we reported an oxida-
that the P450 BM3-based cofactor regeneration tive desymmetrisation of cis-4-cyclopentene-1,3-diol
system effectively supports the OKRs of 1-phenyle- (7) catalysed by an alcohol dehydrogenase from Ral-
thanol (rac-5) by providing sufficient NADP+ stonia sp. (RasADH) affording (R)-4-hydroxy ketone
amounts. Utilisation of TbADH resulted in 50% and (R)-8 in high yield with an excellent ee; NADP+ co-
LbADH in 48% conversion of compound 5, respec- factor regeneration was carried out using cyclohexa-
tively. Application of TbADH gave alcohol (R)-5 in none (9) as co-substrate.[4a] Here, the efficiencies of
49% yield with 95% ee and LB-ADH lead to product the substrate-coupled cyclohexanone-based and the
(S)-5 in 48% yield with 99% ee; in both cases an E enzyme-coupled P450 BM3 F87A-based cofactor re-
value >100 was determined (Table 1, entries 2 and 3). generation system were compared with regard to this
Controls verified that the oxidation of substrate rac-5 biotransformation on a preparative scale (Scheme 3).
is not catalysed by molecular oxygen or P450 BM3 The oxidation of meso-diol 7 was followed by GC
F87A itself. When acetone-based regeneration was analysis. Controls verified that neither P450 BM3
applied, application of TbADH led to 51% conver- F87A nor oxygen effect the meso-diol 7 oxidation. In
sion and 99% ee of compound (R)-5, whereas addition, no formation of the diketone was observed.
LbADH resulted in 45% conversion and 94% ee of Interestingly, both reaction systems gave similar re-
alcohol (S)-5; the lower yield in the latter case could sults. Cyclohexanone-based regeneration provided
be a consequence of a poor acetone tolerance of ketone (R)-8 with 78% yield (94% conversion) and
LbADH. In summary, our findings prove that the 94% ee and the monooxygenase-based alternative re-
P450 BM3 F87A-based NADP+ regeneration system sulted in 77% yield (97% conversion) and 95% ee, re-
effectively supports ADH-catalysed oxidations, there- spectively. These findings confirm that the P450 BM3
by being especially beneficial with regard to ADHs F87A-based NADP+ regeneration system is a feasible
that exhibit low tolerance toward organic solvents/ alternative to previously established methods, thereby
compounds. In addition, the general applicability of further expanding its scope towards desymmetrisation
the P450 BM3 F87A mutant for NADP+ regeneration reactions.
was illustrated by broadening the set of examples.
P450 BM3 for NAD+ Regeneration
Oxidative Desymmetrisation of
cis-4-Cyclopentene-1,3-diol
Generation of a NAD+ Regenerating P450 BM3
Variant
Enantiomerically pure 4-hydroxy-2-cyclopentenones
are common building blocks for a variety of natural Intrigued by our previous results, we embarked on
compounds and therefore, highly desirable motives in a series of experiments to investigate whether P450
Adv. Synth. Catal. 2016, 358, 1810 – 1819
1814
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPERS
asc.wiley-vch.de
BM3 F87A is also applicable for NAD+ regeneration. substrate for NAD+ regeneration and GC analysis
In 2007, Maurer et al. described a NADH-dependent showed formation of the corresponding aldehyde 11
P450 BM3 monooxygenase variant; shifting of the co- with 58% conversion. Subsequent studies were ac-
factor dependency from NADPH to NADH was ach- complished to compare the efficiency of different co-
ieved by two amino acid substitutions in the reductase factor regeneration systems providing NAD+ for
domain (R966D and W1046S).[65] Determination of HlADH-catalysed vanillin (11) formation. Preparative
kinetic constants indicated a higher affinity [KM scale reactions were performed using both substrate-
(NADH)=12 mM] of this mutant for NADH along coupled acetone- and benzaldehyde-based as well as
with an improved turnover rate [kcat (NADH)= enzyme-coupled P450 BM3-based cofactor regenera-
14601 minÀ1] compared to the wild-type enzyme [KM tion systems (Scheme 4).[70] In case of the latter, be-
(NADH)=1430 mM
and
kcat
(NADH)= sides the F87A R966D W1046S triple mutant we also
2810 minÀ1].[65] Based on these data and our previous tested the previously established F87A variant as a ref-
findings, we envisaged the triple mutant F87A R966D erence. Oxidation of alcohol 12 by the monooxyge-
W1046S as a potential NAD+ regenerating P450 BM3 nase or via auto-oxidation in the presence of oxygen
F87A variant. The respective gene was created using was excluded by performing the respective controls.
a sequential QuikChangeꢃ mutagenesis approach In addition, the over-oxidation of aldehyde 11 to car-
starting from the F87A mutant and the obtained bio- boxylic acid was not observed.[71] The oxidation prog-
catalyst was used for further investigations.
ress was monitored via GC analysis indicating satura-
tion kinetics for all reactions; formation of side prod-
ucts was less than 7%. Sample work-up and crude
product separation by column chromatography ena-
bled yield determination for vanillin (11) and the
Targeting Biocatalytic Vanillin Production
The ability of the created F87A R966D W1046S values were in agreement with the previously deter-
mutant to regenerate NAD+ was confirmed via the mined conversions (Table 2). Higher productivities
successful oxidation of a primary alcohol by an were observed with the enzyme-coupled P450 BM3-
NAD+/NADH-dependent dehydrogenase. This appli- based cofactor regeneration systems compared to the
cation broadened the scope of the P450 BM3-based substrate-coupled alternatives. This outcome was sur-
cofactor regeneration system towards both an alterna- prising as the HlADH is known to exhibit high toler-
tive cofactor as well as a different biotransformation. ance to organic solvents and thus, higher yields than
The horse liver alcohol dehydrogenase (HlADH) was 16% and 7% (16% and 8% conversion; Table 2, en-
selected as one of the most widely used oxidoreduc- tries 3 and 4) were expected using benzaldehyde and
tases due to its broad substrate spectrum comprising acetone as co-substrates, respectively.[7,72] Similarly
primary and secondary alcohols or aldehydes and ke- surprising was the fact that the F87A and the F87A
tones, respectively.[7] The production of vanillin (4-hy-
droxy-3-methoxybenzaldehyde, 11) starting from va-
nillyl alcohol (3-methoxy-4-hydroxybenzylalcohol, 12)
using biocatalysis seemed to be a valuable and sus-
tainable approach with respect to the importance of
this compound as an industrial food additive and its
manifold use in flavouring applications. Extensive ef-
forts to establish more ecological routes towards va-
nillin (11) – aiming at omitting the chemical synthesis
starting from toxic phenol – support the relevance of
our example. In 1998, Li and Frost published an enzy-
matic access to vanillin (11) starting from glucose
over vanillic acid as an intermediate, the latter being
reduced to the desired product using an aryl aldehyde
dehydrogenase.[67,68] Recently, Furuya and co-workers
developed a two-pot bioprocess towards vanillin (11)
starting from ferulic acid by a co-enzyme-independent
decarboxylase/oxygenase sequence.[69] These examples
stress the consistent interest for a biocatalytic vanillin
(11) production in the past and the present.
Initially, we confirmed that vanillyl alcohol (12) is
a substrate of the HlADH by performing the oxida-
tion on an analytical scale using a substrate-coupled
Scheme 4. HlADH-catalysed preparative scale oxidations of
vanillyl alcohol (12) to vanillin (11) using different NAD+
regeneration systems. Reaction conditions (S=150 mL):
63 U HlADH,[66] 1.5 mmol substrate and 300 mM NAD+
with a cofactor regeneration system in KPi buffer (100 mM,
cofactor regeneration system. Acetone served as co- pH 7.0, containing 1 mM MgCl2), 24 h.
Adv. Synth. Catal. 2016, 358, 1810 – 1819
1815
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPERS
asc.wiley-vch.de
Table 2. HlADH-catalysed preparative scale oxidations of
vanillyl alcohol (12) to vanillin (11) using different NAD+
regeneration systems.
friendly conditions, thereby greatly avoiding the use
of toxic reagents or active metal catalysts. We have
shown that the established P450 BM3 F87A-based
NAD(P)+ regeneration system is not only broadly ap-
plicable to diverse biotransformations, but also com-
petitive to the known co-substrate-based alternatives.
Examples are the TbADH-catalysed OKR of racemic
1-nonen-4-ol (rac-1) and 1-phenylethanol (rac-5) as
well as the RasADH-catalysed oxidative desymmetri-
sation of meso-diol 7 leading to enantiomerically pure
Entry Co-enzyme/Co-sub-
strate
Conversion
[%][a]
Yield
[%]
1
2
3
4
F87A R966D W1046S 40
33
35
16
7
F87A
48
16
8
benzaldehyde
acetone
[a]
Determined by GC analysis according to a calibration products in all cases. In case of the TbADH-catalysed
curve based on substrate/product mixtures.
OKR of racemic 1-nonen-4-ol (rac-1) and the biocata-
lytic production of vanillin (11) starting from vanillyl
alcohol (12) even superior results compared to widely
R966D W1046S variant gave comparable results used co-substrate-based approach could be achieved
(Table 2, entries 1 and 2); the reputedly NADPH-de- using the established P450-based cofactor regenera-
pendent single mutant leading to an even slightly tion system. By offering a solution to a major bottle-
higher yield of 35% (conversion 48%) than the de- neck regarding enzymatic oxidative applications, we
signed NADH-dependent triple mutant with 33% believe that the presented regeneration system is an
(conversion 40%), respectively. These findings re- important contribution to the field of sustainable
vealed that the P450 BM3 F87A mutant is likewise chemistry and will support the establishment of novel
capable of regenerating NADP+ as well as NAD+ and biotransformations in the future.
is therefore a universally applicable co-enzyme for co-
factor regeneration. Moreover, superiority of the
P450 BM3-based cofactor regeneration compared to
co-substrate-based alternatives was confirmed; the
Experimental Section
latter suffering from unfavourable reaction equilibria
and ADH deactivation in the presence of high co-sub-
strate concentrations.
General Procedure for Preparative Scale
Biotransformations
A 500-mL flask was charged with ADH (63 U), P450 BM3
F87A crude cell lysate (3 mM), NADP+ (300 mM) and cata-
lase from Micrococcus lysodeikticus (600 UmLÀ1)[73] in KPi
Conclusions
buffer (100 mM, pH 7.0, containing 1 mM MgCl2) using
a final volume of 150 mL. Initially, the solution was saturat-
ed with O2 by introducing the gas for 5 min while stirring.
Subsequently, the substrate (1.5 mmol) was added, the flask
was closed with a septum and the reaction was incubated at
Preparative scale oxidative biotransformations with
alcohol dehydrogenases bear a great potential for syn-
thetic organic chemistry in respect to catalytic effi-
ciency and environmental compatibility. Thereby, 308C and 120 rpm. For monitoring the reaction progress
200–500 mL samples were taken, extracted and analysed by
GC. The conversions were calculated based on previously
monitored calibration curves (Supporting Information S4).
The reaction was stopped at 50% conversion for OKRs (2–
8 h) or until no increase in conversion was noticed for the
oxidative desymmetrisation and oxidation of the primary al-
cohol. The solution was saturated with (NH4)2SO4 to allow
protein precipitation overnight at 48C. The reaction was fil-
tered over a pad of celite and extracted with MTBE or ethyl
acetate (5”50 mL). The combined organic layers were dried
highly efficient systems for the regeneration of oxi-
dised nicotinamide cofactors represent an essential
presupposition for the success of these applications
and the lack of NAD(P)+ regeneration systems is
a major bottleneck. We successfully established the
P450 BM3 monooxygenase variant F87A as an effi-
cient NAD(P)H-oxidase for the regeneration of oxi-
dised nicotinamide cofactors in ADH-catalysed oxida-
tions of primary and secondary alcohols on the prepa-
rative scale. The astonishingly high oxidase activity of over MgSO4 and the solvent was removed under reduced
pressure.
OKR of 1-nonen-4-ol (1): The biotransformation was per-
this enzyme extends potential applications of P450
BM3 and makes it a valuable alternative to estab-
lished NAD(P)+ regeneration systems; the latter
being primarily limited by low tolerance of oxidising
enzymes to organic co-substrates and unfavourable
reaction equilibria. In contrast to many chemical
methods for oxidative transformations, the application
of P450 BM3 biocatalysts as NAD(P)H-oxidases for
cofactor regeneration enables the performance of oxi-
formed as described above with partially purified TbADH
(63 U)[56] starting from the commercially available alcohol
rac-1 (255 mL, 1.5 mmol). For monitoring the reaction prog-
ress 200 mL samples were taken, extracted with 500 mL
MTBE containing 2.5 mM 1-hexanol as an internal standard
and analysed by GC. The reaction was stopped after 8 h.
Flash column chromatography on silica gel with n-pentane/
diethyl ether (95:5) afforded compound (S)-1 as a colourless
dative enzymatic chemistry under environmentally oil; yield: 87 mg (0.62 mmol, 41%, 98% ee). Analytical data
Adv. Synth. Catal. 2016, 358, 1810 – 1819
1816
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPERS
asc.wiley-vch.de
are consistent with those reported in the literature.[74,75]
1H NMR (600 MHz, CDCl3): d=5.83 (m, 1H, 2-H), 5.15 (m,
1H, 1-Hb), 5.13 (m, 1H, 1-Ha), 3.65 (m, 1H, 4-H), 2.31 (m,
1H, 3-Ha), 2.14 (m, 1H, 3-Hb), 1.60–1.54 (m, 1H, OH),
1.51–1.40 (m, 3H, 5-Ha, 6-H), 1.39–1.24 (m, 5H, 5-Hb, 7-H,
8-H), 0.89 (t, J=6.9 Hz, 3H, 9-H); 13C NMR (151 MHz,
CDCl3): d=14.2 (C-9), 22.8 (C-8), 25.5 (C-5), 32.0 (C-7),
36.9 (C-6), 42.1 (C-3), 70.9 (C-4), 118.2 (C-1), 135.1 (C-2);
[a]2D0: À7.6 (c=1.1 in CHCl3, ee 98%); GC [CP-Chirasil-Dex
CB, 608C (5 min iso), 58C minÀ1 to 1008C (10 min iso),
208C minÀ1 to 1508C (5 min iso)]; tR (S)-1: 21.1 min, tR (R)-
1: 21.5 min. For acetone-based regeneration 5 vol% acetone
instead of the monooxygenase and catalase were used; in
addition, the reaction was not saturated with oxygen. The
reaction was stopped after 3 h; product (S)-1: yield: 60 mg
(0.42 mmol, 28%, 95% ee):
OKR of 1-phenylethanol (5): The biotransformation was
performed as described above with partially purified
TbADH (63 U)[56] or LbADH crude cell lysate starting from
commercially available alcohol rac-5 (181 mL, 1.5 mmol).
For monitoring the reaction progress 200 mL samples were
taken, extracted with 500 mL MTBE containing 2.5 mM 2-
phenylethanol as an internal standard and analysed by GC.
The reaction was stopped after 2 h for TbADH and 4 h for
LbADH. Flash column chromatography on silica gel with n-
pentane/diethyl ether (9:1) afforded alcohol (R)-5 (yield:
91 mg, 0.74 mmol, 49%, ee 95% for TbADH) and alcohol
(S)-5 (yield: 86 mg, 0.73 mmol, 48%, ee 99% for LbADH)
as colourless oils. Analytical data are consistent with those
reported in the literature.[76–78] 1H NMR (600 MHz, CDCl3):
d=7.40–7.33 (m, 4H, 3-H, 4-H), 7.30–7.24 (m, 1H, 5-H),
4.90 (dq, J=6.5, 3.7 Hz, 1H, 2-H), 1.84 (d, J=3.7 Hz, 1H,
OH), 1.50 (d, J=6.6 Hz, 3H, 1’-H); 13C NMR (151 MHz,
CDCl3): d=25.3 (C-1’), 70.6 (C-1), 125.5 (C-3), 127.6.0 (C-
5), 128.7 (C-4), 146.0 (C-2). (S)-5: [a]2D0: À42.6 (c=1.0 in
CHCl3, ee 99%), (R)-5: [a]2D0: +45.4 (c=1.0 in CHCl3, ee
95%); GC [FS-Lipodex G, 908C (15 min iso), 108C minÀ1 to
1508C (5 min iso)]; tR (S)-5: 11.4 min, tR (R)-5: 12.3 min.
Oxidative desymmetrisation: The biotransformation was
performed as described above with RasADH crude cell
lysate (63 U)[63] starting from commercially available meso-
diol 7 (150 mg, 1.5 mmol) in KPi buffer (100 mM, pH 7.0,
containing 0.8 mM CaCl2).[64] For monitoring the reaction
progress 500 mL samples were taken, extracted with 500 mL
of MTBE and analysed by GC. The crude product was puri-
fied on a pad of silica by washing the product with 200 mL
petroleum ether/ethyl acetate (9:1) and elution with 100%
ethyl acetate afforded compound (R)-4-hydroxy-2-cyclopen-
the reaction was not saturated with oxygen; product (R)-8:
yield: 115 mg (1.17 mmol, 78%, 95% ee).
Biocatalytic synthesis of vanillin (11): The biotransforma-
tion was performed as described above with HlADH crude
cell lysate (63 U),[66] P450 BM3 F87A or F87A R966D
W1046S (3 mM) and NAD+ (300 mM) starting from alcohol
12 (231 mg, 1.5 mmol). For monitoring the reaction progress
200 mL samples were taken, extracted with 500 mL of MTBE
and analysed by GC. Flash column chromatography on
silica gel with petroleum ether/ethyl acetate (9:1) afforded
vanillin (11) as a light yellow solid; yield with F87A: 80 mg
(0.53 mmol, 35%); yield with F87A R966D W1046S: 75 mg
(0.49 mmol, 33%) . Analytical data are consistent with those
reported in the literature.[81] 1H NMR (600 MHz, CDCl3):
d=9.83 (s, 1H, 1-H), 7.45–7.41 (m, 3H, 6-H, 7-H), 7.05 (d,
J=8.5 Hz, 1H, 3-H), 6.20 (s, 1H, OH), 3.97 (s, 3H, OMe);
13C NMR (151 MHz, CDCl3): d=56.3 (OMe), 109.0 (C-6),
114.5 (C-3), 127.7 (C-7), 130.0 (C-2), 147.3 (C-4), 151.8 (C-
5), 191.0 (C-1). For the substrate-coupled approach benzal-
dehyde (306 mL, 3 mmol) or acetone (7.5 mL, 5 vol%) in-
stead of the monooxygenase and catalase were used; in ad-
dition, the reaction was not saturated with oxygen. The ben-
zaldeyde-based approach (yield: 40 mg, 0.26 mmol, 17%)
and the acetone-based one (yield: 16 mg, 0.10 mmol, 7%)
yielded compound 11 as a light yellow solid.
Acknowledgements
The authors thank the Fonds of the Chemical Industry
(scholarship to C.H.), the Ministry of Innovation, Science
and Research of the German federal state of North Rhine-
Westphalia and the Heinrich Heine University Düsseldorf
(scholarship within the CLIB-Graduate Cluster Industrial
Biotechnology for K. N.). We further thank Prof. Wolfgang
Kroutil for providing the RasADH gene, and gratefully ac-
knowledge Birgit Henßen and Dr. Patrick Bongen for their
ongoing support with HPLC and GC measurements.
References
[1] M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Keß-
eler, R. Stürmer, T. Zelinski, Angew. Chem. 2004, 116,
806–843; Angew. Chem. Int. Ed. 2004, 43, 788–824.
[2] Kinetic resolutions: a) J. M. Keith, J. F. Larrow, E. N.
Jacobsen, Adv. Synth. Catal. 2001, 343, 5–26; b) H. Pel-
lissier, Adv. Synth. Catal. 2011, 353, 1613–1666;
c) D. E. J. E. Robinson, S. D. Bull, Tetrahedron: Asym-
metry 2003, 14, 1407–1446; d) P. Somfai, Angew. Chem.
1997, 109, 2849–2851; Angew. Chem. Int. Ed. 2004, 43,
788–824; e) A. C. Spivey, A. Maddaford, A. J. Red-
grave, Org. Prep. Proced. Int. 2000, 32, 331–365; f) Y.
Yang, J. Liu, Z. Li, Angew. Chem. 2014, 126, 3184–
3188; Angew. Chem. Int. Ed. 2004, 43, 788–824.
tenone [(R)-8] as
a light yellow oil; yield: 112 mg
(1.15 mmol, 77%, 94% ee). Analytical data are consistent
with those reported in the literature.[79,80] 1H NMR
(600 MHz, CDCl3): d=7.58 (dd, J=5.5, 2.5 Hz, 1H, 3-H),
6.23 (dd, J=5.5, 1.5 Hz, 1H, 2-H), 5.09–5.06 (m, 1H, 4-H),
2.84 (dd, J=18.2, 6.1 Hz, 1H, 5-Ha), 2.25 (dd, J=18.2,
2.1 Hz, 1H, 5-Hb); 13C NMR (151 MHz, CDCl3): d=207.1
(C-1), 163.7 (C-3), 135.2 (C-2), 70.5 (C-4), 44.4 (C-5); [a]2D0:
+70.5 (c=1.0 in CHCl3, ee 94%); GC [FS-Lipodex E, 608C
(5 min iso), 108C minÀ1 to 1508C (5 min iso)]; tR (R)-8:
22.2 min, tR (S)-8: 22.7 min. In case of the substrate-coupled
approach cyclohexanone (310 mL, 3 mmol) without addition
of the monooxygenase and catalase was used; in addition,
[3] For de novo functionalisations, see: a) H. M. L. Davies,
J. Du Bois, J.-Q. Yu, Chem. Soc. Rev. 2011, 40, 1855–
1856; b) K. Godula, D. Sames, Science 2006, 312, 67–
72; c) F. Kakiuchi, N. Chatani, Adv. Synth. Catal. 2003,
345, 1077–1101.
Adv. Synth. Catal. 2016, 358, 1810 – 1819
1817
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPERS
asc.wiley-vch.de
[4] Desymmetrisation of meso-compounds: a) C. Holec, D.
Sandkuhl, D. Rother, W. Kroutil, J. Pietruszka, Chem-
CatChem 2015, 7, 3125–3130; b) J. C. Anderson, S. V.
Ley, S. P. Marsden, Tetrahedron Lett. 1994, 35, 2087–
2090; c) A. K. Ghosh, X. Xu, Org. Lett. 2004, 6, 2055–
2058; d) C. R. Johnson, S. J. Bis, Tetrahedron Lett. 1992,
33, 7287–7290; e) P. M. Khan, R. Wu, K. S. Bisht, Tetra-
hedron 2007, 63, 1116–1126; f) K. Laumen, M. P.
Schneider, J. Chem. Soc. Chem. Commun. 1986, 1298–
1299.
[5] Asymmetric reductions: a) K. Nakamura, R. Yamana-
ka, T. Matsuda, T. Harada, Tetrahedron: Asymmetry
2003, 14, 2659–2681; b) T. Ohkuma, Proc. Jpn. Acad.
Ser. B Phys. Biol. Sci. 2010, 86, 202–219; c) R. Noyori,
in: Asymmetric Catalysis in Organic Synthesis, Wiley,
New York, 1994.
[22] S. Kara, J. Schrittwieser, F. Hollmann, M. Ansorge-
Schumacher, Appl. Microbiol. Biotechnol. 2014, 98,
1517–1529.
[23] E. E. Ferrandi, D. Monti, I. Patel, R. Kittl, D. Haltrich,
S. Riva, R. Ludwig, Adv. Synth. Catal. 2012, 354, 2821–
2828.
[24] a) B. Petschacher, N. Staunig, M. Müller, M. Schür-
mann, D. Mink, S. De Wildeman, K. Gruber, A. Glied-
er, Comput. Struct. Biotechnol. J. 2014, 9, e201402005;
b) for an example utilizing stoichiometric amounts of
co-substrate, see: I. Lavandera, A. Kern, V. Resch, B.
Ferreira-Silva, A. Glieder, W. M. F. Fabian, S. de Wilde-
man, W. Kroutil, Org. Lett. 2008, 2155–2158.
[25] S. Kochius, A. Magnusson, F. Hollmann, J. Schrader, D.
Holtmann, Appl. Microbiol. Biotechnol. 2012, 93, 2251–
2264.
[6] M. Müller, M. Wolberg, T. Schubert, W. Hummel, Adv.
Biochem. Eng./Biotechnol. 2005, 92, 261–287.
[7] W. Kroutil, H. Mang, K. Edegger, K. Faber, Curr.
Opin. Chem. Biol. 2004, 8, 120–126.
[26] M. Pickl, M. Fuchs, S. M. Glueck, K. Faber, Appl. Mi-
crobiol. Biotechnol. 2015, 99, 6617–6642.
[27] F. R. Bisogno, I. Lavandera, W. Kroutil, V. Gotor, J.
Org. Chem. 2009, 74, 1730–1732.
[8] T. Fischer, J. Pietruszka, Top. Curr. Chem. 2010, 297, 1–
34.
[28] C. Rodriguez, I. Lavandera, V. Gotor, Curr. Org.
Chem. 2012, 16, 2525–2541.
[9] a) S. K. Alamsetti, G. Sekar, Chem. Commun. 2010, 46,
7235–7237; b) C.-T. Chen, S. Bettigeri, S.-S. Weng, V. D.
Pawar, Y.-H. Lin, C.-Y. Liu, W.-Z. Lee, J. Org. Chem.
2007, 72, 8175–8185; c) D. C. Ebner, J. T. Bagdanoff,
E. M. Ferreira, R. M. McFadden, D. D. Caspi, R. M.
Trend, B. M. Stoltz, Chem. Eur. J. 2009, 15, 12978–
12992; d) R. A. Shiels, K. Venkatasubbaiah, C. W.
Jones, Adv. Synth. Catal. 2008, 350, 2823–2834.
[10] M. Wills, Angew. Chem. 2008, 120, 4334–4337; Angew.
Chem. Int. Ed. 2008, 47, 4264–4267.
[11] a) E. M. Ferreira, B. M. Stoltz, J. Am. Chem. Soc. 2001,
123, 7725–7726; b) H. Mizoguchi, T. Uchida, T. Katsu-
ki, Angew. Chem. 2014, 126, 3242–33246; Angew.
Chem. Int. Ed. 2014, 53, 3178–3182; c) Y. Nishibayashi,
A. Yamauchi, G. Onodera, S. Uemura, J. Org. Chem.
2003, 68, 5875–5880.
[29] a) B. R. Riebel, P. R. Gibbs, W. B. Wellborn, A. S. Bom-
marius, Adv. Synth. Catal. 2002, 344, 1156–1168;
b) B. R. Riebel, P. R. Gibbs, W. B. Wellborn, A. S.
Bommarius, Adv. Synth. Catal. 2003, 345, 707–712.
[30] L. Engels, M. Henze, W. Hummel, L. Elling, Adv.
Synth. Catal. 2015, 357, 1751–1762.
[31] J. Rocha-Martín, D. Vega, J. M. Bolivar, C. A. Godoy,
A. Hidalgo, J. Berenguer, J. M. Guisµn, F. López-Galle-
go, BMC Biotechnol. 2011, 11, 101.
[32] J. Zhang, S. Wu, J. Wu, Z. Li, ACS Catal. 2015, 5, 51–
58.
[33] P. B. Brondani, H. M. Dudek, C. Martinoli, A. Mattevi,
M. W. Fraaije, J. Am. Chem. Soc. 2014, 136, 16966–
16969.
[34] R. Fasan, ACS Catal. 2012, 2, 647–666.
[35] G. Grogan, Curr. Opin. Chem. Biol. 2011, 15, 241–248.
[36] S. Schulz, M. Girhard, V. B. Urlacher, ChemCatChem
2012, 4, 1889–1895.
[37] J. R. Falck, Y. K. Reddy, D. C. Haines, K. M. Reddy,
U. M. Krishna, S. Graham, B. Murry, J. A. Peterson,
Tetrahedron Lett. 2001, 42, 4131–4133.
[12] S. Arita, T. Koike, Y. Kayaki, T. Ikariya, Angew. Chem.
2008, 120, 2481–2483; Angew. Chem. Int. Ed. 2008, 47,
2447–2449.
[13] J. Zhang, X. Yang, H. Zhou, Y. Li, Z. Dong, J. Gao,
Green Chem. 2012, 14, 1289–1292.
[14] A. T. Radosevich, C. Musich, F. D. Toste, J. Am. Chem.
Soc. 2005, 127, 1090–1091.
[38] S. Kille, F. E. Zilly, J. P. Acevedo, M. T. Reetz, Nat.
Chem. 2011, 3, 738–743.
[15] T. Kunisu, T. Oguma, T. Katsuki, J. Am. Chem. Soc.
[39] K. Kühnel, S. C. Maurer, Y. Galeyeva, W. Frey, S. La-
schat, V. B. Urlacher, Adv. Synth. Catal. 2007, 349,
1451–1461.
2011, 133, 12937–12939.
[16] M. Tomizawa, M. Shibuya, Y. Iwabuchi, Org. Lett.
2009, 11, 1829–1831.
[40] P. Le-Huu, T. Heidt, B. Claasen, S. Laschat, V. B. Ur-
[17] W. Hummel, H. Grçger, J. Biotechnol. 2014, 191, 22–
31.
lacher, ACS Catal. 2015, 5, 1772–1780.
[41] 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.
[42] K. Neufeld, B. Henßen, J. Pietruszka, Angew. Chem.
2014, 126, 13469–13473; Angew. Chem. Int. Ed. 2014,
53, 13253–13257.
[18] A. Weckbecker, H. Grçger, W. Hummel, in: Biosystems
Engineering I, Vol. 120, (Eds.: C. Wittmann, R. Krull),
Springer, Berlin, Heidelberg, 2010, pp 195–242.
[19] V. Uppada, S. Bhaduri, S. B. Noronha, Curr. Sci. 2014,
106, 946–957.
[20] W. Kroutil, H. Mang, K. Edegger, K. Faber, Adv.
[43] K. Neufeld, J. Marienhagen, U. Schwaneberg, J. Pie-
Synth. Catal. 2004, 346, 125–142.
truszka, Green Chem. 2013, 15, 2408–2421.
[21] F. Hollmann, K. Hofstetter, A. Schmid, Trends Biotech-
nol. 2006, 24, 163–171.
[44] K. R. Hornberger, C. L. Hamblett, J. L. Leighton, J.
Am. Chem. Soc. 2000, 122, 12894–12895.
Adv. Synth. Catal. 2016, 358, 1810 – 1819
1818
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

FULL PAPERS
asc.wiley-vch.de
[45] N. Makita, Y. Hoshino, H. Yamamoto, Angew. Chem.
2003, 115, 971–973; Angew. Chem. Int. Ed. 2003, 42,
941–943.
[46] K. C. Nicolaou, D. W. Kim, R. Baati, Angew. Chem.
2002, 114, 3853–3856; Angew. Chem. Int. Ed. 2002, 41,
3701–3704.
[47] T. E. Smith, S. J. Fink, Z. G. Levine, K. A. McClelland,
A. A. Zackheim, M. E. Daub, Org. Lett. 2012, 14,
1452–1455.
[48] L. Wang, P. E. Floreancig, Org. Lett. 2004, 6, 569–572.
[49] M. A. Noble, C. S. Miles, S. K. Chapman, D. A. Lysek,
A. C. Mackay, G. A. Reid, R. P. Hanzlik, A. W. Munro,
Biochem. J. 1999, 339, 371–379.
[50] A. W. Munro, S. Daff, J. R. Coggins, J. G. Lindsay, S. K.
Chapman, Eur. J. Biochem. 1996, 239, 403–409.
[51] For review, see: C. J. C. Whitehouse, S. G. Bell, L.-L.
Wong, Chem. Soc. Rev. 2012, 41, 1218–1260.
[52] J. Kuper, K. L. Tee, M. Wilmanns, D. Roccatano, U.
Schwaneberg, T. S. Wong, Acta Crystallogr. Sect. F
Struct. Biol. Cryst. Commun. 2012, 68, 1013–1017.
[53] T. S. Wong, F. H. Arnold, U. Schwaneberg, Biotechnol.
Bioeng. 2004, 85, 351–358.
[63] One unit is defined as the amount of enzyme needed
for the reduction of 1 mmolminÀ1 benzaldehyde.
[64] For RasADH applications KPi buffer (100 mM,
pH 7.0) supplemented with 0.8 mM CaCl2 was used due
its stabilising effect. For literature, see: J. Kulig, A.
Frese, W. Kroutil, M. Pohl, D. Rother, Biotechnol.
Bioeng. 2013, 110, 1838–1848.
[65] S. C. Maurer, K. Kühnel, L. A. Kaysser, S. Eiben, R. D.
Schmid, V. B. Urlacher, Adv. Synth. Catal. 2005, 347,
1090–1098.
[66] One unit is defined as the amount of enzyme needed
for the oxidation of 1 mmolminÀ1 ethanol.
[67] K. Li, J. Frost, J. Am. Chem. Soc. 1998, 120, 10545–
10546.
[68] A. Liese, M. Villela Filho, Curr. Opin. Biotechnol. 1999,
10, 595–603.
[69] T. Furuya, M. Miura, M. Kuroiwa, K. Kino, Nat. Bio-
technol. 2015, 32, 335–339.
[70] Acetaldehyde was also tested as a co-substrate for re-
generation of NADH and resulted in a conversion of
67% on the analytical scale representing it as a suitable
co-substrate; acetaldehyde-based regeneration on the
preparative scale was rendered impractical due to the
high volatility of acetaldehyde.
[71] P. Kçnst, H. Merkens, S. Kara, S. Kochius, A. Vogel, R.
Zuhse, D. Holtmann, I. W. C. E. Arends, F. Hollmann,
Angew. Chem. 2012, 124, 10052–10055; Angew. Chem.
Int. Ed. 2012, 51, 9914–9917.
[72] The highest conversion was detected with 3 equiv. of
benzaldehyde; increasing the benzaldehyde amount up
to 40 equiv. did not result in higher conversions.
[73] The omission of catalase resulted in a slightly lower
conversion; however, catalase was preventively added
to exclude enzyme deactivation by hydrogen peroxide.
[74] S. M. Wickel, C. A. Citron, J. S. Dickschat, Eur. J. Org.
Chem. 2013, 2013, 2906–2913.
[54] D. Roccatano, T. S. Wong, U. Schwaneberg, M. Zacha-
rias, Biopolymers 2006, 83, 467–476.
[55] Isolation/identification of the decomposition product(s)
of ketone 2 was not possible due to a further decompo-
sition during work-up and isolation.
[56] One unit is defined as the amount of enzyme needed
for the reduction of 1 mmol minÀ1 acetophenone.
[57] U. T. Bornscheuer, R. K. Kazlauskas, in: Hydrolases in
Organic Synthesis, Wiley, Weinheim, 2006.
[58] T. Classen, M. Korpak, M. Schçlzel, J. Pietruszka, ACS
Catal. 2014, 4, 1321–1331.
[59] T. Fischer, J. Pietruszka, Adv. Synth. Catal. 2007, 349,
1533–1536.
[60] M. Bischop, V. Doum, A. Nordschild (nØe Rieche), J.
Pietruszka, D. Sandkuhl, Synthesis 2010, 527–537.
ˇ
[75] A. D. Wadsworth, D. P. Furkert, M. A. Brimble, J. Org.
[61] a) Z. Findrik, “. Vasic’-Racki, S. Lütz, T. Daußmann,
Chem. 2014, 79, 11179–11193.
C. Wandrey, Biotechnol. Lett. 2005, 27, 1087–1095; b) S.
Leuchs, L. Greiner, Chem. Biochem. Eng. Q. 2011, 25,
267–281.
[76] . García-MuÇoz, A. I. Ortega-Arizmendi, M. A.
García-Carrillo, E. Díaz, N. Gonzalez-Rivas, E.
Cuevas-YaÇez, Synthesis 2012, 44, 2237–2242.
[77] J. M. Zimbron, M. Dauphinais, A. B. Charette, Green
Chem. 2015, 17, 3255–3259.
[78] D. J. Blair, C. J. Fletcher, K. M. P. Wheelhouse, V. K.
Aggarwal, Angew. Chem. 2014, 126, 5658–5661; Angew.
Chem. Int. Ed. 2014, 53, 5552–5555.
[79] G. Dickmeiss, V. De Sio, J. Udmark, T. B. Poulsen, V.
Marcos, K. A. Jørgensen, Angew. Chem. 2009, 121,
6778–6781; Angew. Chem. Int. Ed. 2009, 48, 6650–6653.
[80] T. T. Curran, D. A. Hay, C. P. Koegel, J. C. Evans, Tetra-
hedron 1997, 53, 1983–2004.
[62] For selected examples, see: a) J. Iriondo-Alberdi, J. E.
Perea-Buceta, M. F. Greaney, Org. Lett. 2005, 7, 3969–
3971; b) S. D. Knight, L. E. Overman, G. Pairaudeau, J.
Am. Chem. Soc. 1995, 117, 5776–5788; c) L. Q. Lu, Y.
Li, K. Junge, M. Beller, Angew. Chem. 2013, 125, 8540–
8544; Angew. Chem. Int. Ed. 2013, 52, 8382–8386;
d) K. C. Nicolaou, P. Heretsch, A. ElMarrouni, C. R. H.
Hale, K. K. Pulukuri, A. K. Kudva, V. Narayan, K. S.
Prabhu, Angew. Chem. 2014, 126, 10611–10615; Angew.
Chem. Int. Ed. 2014, 53, 10443–10447; e) S. A. Ruider,
T. Sandmeier, E. M. Carreira, Angew. Chem. 2015, 127,
2408–2412; Angew. Chem. Int. Ed. 2015, 54, 2378–2412;
f) G. Singh, A. Meyer, J. AubØ, J. Org. Chem. 2014, 79,
452–458.
[81] Y. I. Fengping, J. Xiaoyan, N. I. U. Jihua, Z. Lirong, W.
Zhen, Asian J. Chem. 2015, 27, 885–888.
Adv. Synth. Catal. 2016, 358, 1810 – 1819
1819
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim