Bio-based α,ω-Functionalized Hydrocarbons from Multi-step Reaction Sequences with Bio- and Metallo-catalysts Based on the Fatty Acid Decarboxylase OleTJE

Article abstract of DOI:10.1002/cctc.201701804

OleT from Jeotgalicoccus sp. ATCC 8456 catalyzes the decarboxylation of ω-functionalized fatty acids to the corresponding alkenols, which can themselves serve as starting material for the synthesis of polymers and fine chemicals. To show the versatility of possible reactions, a series of in vitro reaction cascades was developed where an alkenol produced by the decarboxylation of ω-hydroxy fatty acids can be further converted into alkenylamines and diols. By coupling OleT with an alcohol dehydrogenase or alcohol oxidase as well as an amino-transaminase, an oxidative decarboxylation followed by the oxidation of the terminal alcohol and a subsequent reductive transamination could be carried out. By using different cofactors or electron sources, the reactions could be performed sequentially or simultaneously. The combination of enzymatic decarboxylation with a ruthenium catalyst in a chemo-enzymatic cascade provides a novel way to synthesize long-chain diols.

Full text of DOI:10.1002/cctc.201701804

Accepted Article
Title: Bio-based α,ω functionalized hydrocarbons from multi-step
reaction sequences with bio- and metallo-catalysts based on the
fatty acid decarboxylase OleTJE
Authors: Samiro Bojarra, Dennis Reichert, Marius Grote, Álvaro
Gómez Baraibar, Alexander Denning, Bernd Nidetzky, Carolin
Mügge, and Robert Kourist
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201701804
Link to VoR: http://dx.doi.org/10.1002/cctc.201701804
A Journal of
www.chemcatchem.org

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
Bio-based α,ω-functionalized hydrocarbons from multi-step
reaction sequences with bio- and metallo-catalysts based on the
fatty acid decarboxylase OleTJE
Samiro Bojarra[a], Dennis Reichert[a,b], Marius Grote[a], Álvaro Gómez Baraibar[a], Alexander Dennig[c],
Bernd Nidetzky[c], Carolin Mügge[a], and Robert Kourist*[a,d]
Abstract: OleT from Jeotgalicoccus sp. ATCC 8456 catalyzes the
decarboxylation of ω-functionalized fatty acids to the corresponding
alkenols, which can themselves serve as starting material for the
synthesis of polymers and fine chemicals. To show the versatility of
possible reactions, we developed a series of in vitro reaction
cascades where an alkenol produced by the decarboxylation of
ω-hydroxy fatty acids can be further converted into alkenylamines
and diols. By coupling OleT with an alcohol dehydrogenase or
alcohol oxidase as well as an amino-transaminase, an oxidative
decarboxylation followed by the oxidation of the terminal alcohol and
a subsequent reductive transamination could be carried out. By
using different cofactors or electron sources, the reactions could be
performed sequentially or simultaneously. In addition, the
combination of enzymatic decarboxylation with a ruthenium catalyst
in a chemo-enzymatic cascade provides a novel way to synthesize
long-chain diols.
conversion of fatty acids to terminal alkenes.[5,6,7] The fatty acid
decarboxylase cytochrome P450 OleT (CYP152L1) from
Jeotgalicoccus sp. ATCC 8456 possesses the unique ability to
convert saturated and (E)-unsaturated long-chain fatty acids to
the corresponding 1-alkenes at neutral pH and room
temperature.[8–12] Besides being active as
a
terminal
decarboxylase, OleT also catalyzes the hydroxylation of fatty
acids in α-, β- and γ-position.[11,12,19] It might be that this
secondary reaction is related to the natural role of this enzyme
since it is known that β-hydroxy fatty acid esters have regulatory
functions in bacteria.[20] In contrast to other alkene producing
enzymes, OleT acts predominantly as H2O2-dependent
peroxygenase.[13,14] To minimize heme destruction by excess of
H2O2,[14] the amount of H2O2 present in the reaction mixture
needs to be carefully controlled.[15] For this purpose, we
developed a system for the in situ formation of H2O2 by
photocatalytic reduction of O2 (hν/FMN/H2O2 system).[12] Different
mutagenesis analyses revealed the specific residues in the
active site of the enzyme that are crucial for the activity of OleT
as peroxygenase in the decarboxylation of fatty acids.[13,16,17] By
using OleT as peroxygenase, we can circumvent the
complicated multi-step electron transport chain as well as the
recycling of expensive cofactors like NAD(P)H, which are usually
required for this type of reaction (Scheme 1).[8,10,18]
Introduction
As a result of the rapid depletion of fossil resources and
continuously increasing environmental concerns, the interest to
use renewable raw materials for the chemical industry has
increased in the last decade.[1,2] Nowadays, alkenes form the
basis for the production of plasticizers, lubricants and
surfactants.[3] Oils and fats are regarded as one of the most
important pools for the extraction of raw materials from
renewable resources.[4] The use of enzymes in these first-stage
valorization processes can facilitate the use of renewable
resources for the production of fine chemicals.
Four specific enzymes, OleT, UndA, UndB and recently
characterized CYP-Sm46, directly catalyze the enzymatic
Scheme 1. Oxidative decarboxylation of fatty acids to 1-alkene by OleT as
peroxygenase or monooxygenase. a) Photocatalytic in situ generation of H2O2
to promote OleT catalyzed oxidative decarboxylation of fatty acids.[12] b)
Enzymatic redox cascade process for the decarboxylation of fatty acids using
the CamAB system.[8]
[a]
[b]
S. Bojarra, D. Reichert, M. Grote, Dr. Á. Gomez Baraibar, Dr. C.
Mügge, and Prof. Dr. R. Kourist
Junior Research Group for Microbial Biotechnology, Ruhr-Universität
Bochum, Universitätsstraße 150, 44780 Bochum, Germany
D. Reichert
Institute for Biochemistry, Westfälische Wilhelms-Universität
Münster, Wilhelm-Klemm-Strasse 2, 48149 Münster, Germany –
current address
While OleT efficiently decarboxylates saturated and (E)-
unsaturated long-chain fatty acids, its activity drops with
decreasing chain length. Medium-chain (8C-14C) fatty acids are
not or only slowly decarboxylated.[21] Faber and co-workers
could still recently show that OleT is an efficient catalyst for the
decarboxylation of short-chain (4C-5C) organic acids with NADH
as electron donor, giving rise to gaseous 1-alkenes.[8]
Furthermore, medium-chain fatty acids could also be
decarboxylated with this system.
[c]
[d]
Dr. A. Dennig, Prof. Dr. Bernd Nidetzky
Institute of Biotechnology and Biochemical Engineering, Graz
University of Technology, Petersgasse 12, 8010 Graz, Austria
Prof. Dr. R. Kourist
Institute of Molecular Biotechnology, Graz University of Technology,
Petersgasse 14, 8010 Graz, Austria – permanent address
E-mail: kourist@tugraz.at
Supporting information for this article is given via a link at the end of
the document.
As the enzymatic decarboxylation proceeds under very mild
reaction conditions and can be expected to have a high
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
functional-group tolerance, we envisioned that OleT should be a
suitable catalyst for the synthesis of ω-alkenols and other olefins
from the corresponding ω-functionalized fatty acids.
ω-Hydroxylauric acid (1a) can be obtained from the inexpensive,
renewable raw material palm kernel oil.[22] In addition, the
terminal oxidation of fatty acids by CYP450 monooxygenases to
terminally oxidized hydroxy fatty acids has recently been
established and demonstrated with 1a.[23] Accordingly, this fatty
acid is an industrially interesting substrate which can be
converted into further substances like ω-alkenols. ω-Alkenols
can serve as starting materials for the synthesis of functionalized
long-chain amines, acids or even diols. In this work, we present
a concept for the valorization of ω-alkenols like undec-10-en-1-ol
(1b) by enzymatic and chemo-enzymatic reaction sequences.
Scheme 2. Enzymatic decarboxylation of ω-functionalized fatty acids and potential follow-up reactions. ω-Hydroxylauric acid 1a is decarboxylated by OleT to give
ω-alkenol 1b, which can be further converted by additional enzymatic reactions to 1e, 1f or even 1g by using ruthenium-based metathesis catalysts. As electron
supply system for OleT, CamAB and the in situ light driven H2O2 generation system was used.
A series of in vitro reaction cascades combines the de-
functionalization by enzymatic decarboxylation to re-
Enzymatic decarboxylation of ω-hydroxy fatty acids.
a
OleT was expressed in a constructed E. coli JW5020 expression
strain with inactivated acyl-coenzyme A dehydrogenase (FadE)
to avoid the non-desired degradation of fatty acids by enzymes
present in the crude cellular extract (CE). OleT was either used
as cell-free extract or in purified form. An in situ light-driven H2O2
generation system (hν/FMN/H2O2) was used to fuel the oxidative
decarboxylation of hydroxy fatty acids to the corresponding
ω-alkenols. While we expected that OleT would be capable of
decarboxylating long-chain ω-hydroxy fatty acids like
functionalization of the ω-hydroxy group or the double bonds
(Scheme 2). The first and central step of all valorization chains is
the decarboxylation of 1a by OleT to form 1b. The further
oxidation of 1b to undec-10-en-1-al (1d) can be carried out by a
mutated NAD+ dependent alcohol dehydrogenase from
Escherichia coli (EcAdhZ3-LND)[24] or a FAD+ dependent alcohol
oxidase from Aspergillus fumigatus (LCAO_Af).[25] A further
oxidation of 1d by either of the used enzymes can lead to the
formation of undec-10-en-1-oic acid (1e). The reductive
amination of the aldehyde 1d by the amine-transaminase from
Chromobacterium violaceum (CvTA)[25] results in the formation
of undec-10-en-1-amine (1f). Furthermore, we can produce icos-
10-ene-1,20-diol (1g) from 1b in a chemo-catalytic olefin
metathesis reaction, using ruthenium-based metathesis
ω-hydroxypentadecanoic
acid
(2a,
15C)
and
ω-hydroxyhexadecanoic acid (3a, 16C), it was difficult to
anticipate whether the shorter hydroxylauric acid (1a, 12C)
would be converted. We were pleased observe decarboxylation
of fatty acids of different chain lengths (12C-OH (1a), 15C-OH
(2a) and 16C-OH (3a)) to give the corresponding terminal
alkenols with a chain length of one carbon atom less (1b-3b)
took place. Formation of ω-alkenols and hydroxy fatty acids was
determined by gas chromatography with MS and FID detection
catalysts.
A similar approach with hydroxystilbene-based
intermediates has recently led to the establishment of an
effective chemo-enzymatic cascade using a phenolic acid
decarboxylase with ruthenium catalysts.[26,27] The comparison of
two different electron donor systems will show whether the in
after
derivatization
with
N-methyl-N-(trimethylsilyl)
trifluoroacetamide (MSTFA). In addition, formation of the
terminal olefins was confirmed by the identification of the
characteristic 1H-NMR signals of the terminal double bonds in
the crude extract of small-scale biocatalytic reactions. We also
observed the formation of side-products. OleT is known to
convert stearic acid to 1-heptadecene and β-hydroxystearic acid
with two additional side-products that were tentatively assigned
to be the α- and γ-isomer, respectively. Similarly we assume that
it is likely that OleT converts 1a-3a to the corresponding
[12]
situ generation of H2O2 or the recyclization of NADH via the
CamAB[8] system from Pseudomonas putida is more proficient
with OleT. With this approach we can obtain 1e, 1f and 1g which
are interesting building blocks for the chemical industry.[2,28,29]
Results and Discussion
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
hydroxyacids (1c-3c).[11] For each substrate, three different
tentative hydroxylation products, the α-, β-, and γ-hydroxy fatty
acids, were identified. However, due to the low amounts formed,
confirmation by NMR was not possible. GC-MS-analysis
indicated the β-hydroxy isomer as the most prominent form
(Figures 1 and S 5).[11] While 1b was identified by comparison
with an authentic standard, the homologs 2b and 3b were
identified by a tentative assignment based on their mass spectra
(Figure S 2 – Figure S 4). MS analysis also suggested that the
hydroxylation is more pronounced with decreasing chain length.
Studies on the crystal structure of OleT indicate that the exact
accommodation of the carboxylic group towards the heme
function in the enzyme’s active site decides the outcome of the
reaction.[30] To determine the decarboxylation vs. hydroxylation
ratio, the signals of the three tentative hydroxylation products
were considered in their sum (Figure 1).
The selectivity of OleT between decarboxylation and
hydroxylation of 1a differed under the influence of various co-
side-products stem from a hydroxylation of the substrate (1c-3c). solvents and their concentrations as well as the buffer system
To optimize the reaction conditions to favor the
decarboxylation by OleT, we investigated the influence of
different buffer systems, the pH value and of co-solvents on the
ratio between decarboxylation and hydroxylation of 1a to give 1b
and 1c. The ratio between decarboxylation and hydroxylation of
fatty acids depends on the chain length.[12] While for long-chain
fatty acids the decarboxylation is the dominant reaction, the
used. The ratio of the signal intensities between ω-alkenols and
hydroxylation products was about 0.3 under initial reaction
conditions with 1a (0.5 mM), DMSO (5% (v/v)) and Tris buffer
(50 mM at pH 7.5). From analyzing the influence of different
buffers on the selectivity of OleT, the Britton-Robinson buffer
showed the best results for ω-alkenol production with a ratio of
Figure 1. Influence of different buffers (50 mM each, with pH 7) on the activity of OleT in the reaction with 1a. a) decarboxylation and hydroxylation reaction of 1a
by OleT.[11] b) Ratios of the normalized signal intensities of the decarboxylation vs. hydroxylation products in the reaction of 0.5 mM 1a, catalyzed by cell-free
extract of 60 mg (cdw) Containing OleTJE for 60 minutes in different buffer systems: bicine, bis-Tris Methane (Bis-Tris), tris(hydroxymethyl)aminomethane (Tris),
4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 3-(N-morpholino)propanesulfonic acid (MOPS), Phosphate, Britton–Robinson (Brit.-Rob.) and
borate c) Formation of three side-products was observed. As it is known that OleT hydroxylates stearic acids, the production of hydroxyfatty acids (α-, β-, or
γ-position) would be a possible explanation. GC-MS and GC-FID chromatograms showed a typical pattern for hydroxy-fatty acids (see also Figure S 3 and
Figure S 4).
1.1 (using 5% DMSO, Figure 1b), which means that the
hydroxylation of 1a could be suppressed by a proper choice of
reaction buffer.
reaction. To verify this, different co-solvents like 1,4-dioxane,
tetrahydrofuran (THF), 1,1-dimethylformamide (DMF) and
dimethyl sulfoxide (DMSO) in concentrations of 5 and 30% (v/v)
were tested. In presence of 10% (v/v) DMSO, the highest overall
conversion could be observed (Figure S 7). The use of 30% (v/v)
DMSO led to an increase of the ratio from 0.43 with 5% DMSO
to about 0.7, indicating a growing preference for decarboxylation
in presence of this co-solvent (Figure S 8). This shift of the ratio
between decarboxylation and hydroxylation with increasing
DMSO concentration can be influenced by several factors. We
assume that a high concentration of DMSO leads to more
The
pH
activity
profile
regarding
the
decarboxylation/hydroxylation ratio was investigated within a
range of pH 3–9, where at pH 7 the best results were obtained in
favor of decarboxylation (Figure S 6). Therefore, for further
reactions, Britton-Robinson buffer at pH 7 and 10% (v/v) DMSO
were selected.
Due to its low solubility in water, fatty acids need to be
dissolved in a suitable co-solvent, which itself can influence the
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
soluble substrate in the reaction and therefore more product
formation by OleT. Crystallization studies on monooxygenase
P450 BM3 with DMSO demonstrated that DMSO can coordinate
to heme-bound iron as axial ligand,[31] thereby closing the
coordination sphere around the central metal - which can
significantly affect the enzymatic activity. When using OleT as a
monooxygenase, the presence of DMSO decreases the overall
activity of OleT.[32] Overall, to minimize a negative effect on other
enzymes in the cascades by using a high concentration of
DMSO (Scheme 2), 10% (v/v) was used for all following
experiments.
Preliminary experiments with 0.5 mM 1a and OleTCE in the
hν/FMN/H2O2 system allowed us to achieve full conversion
within 4 h (Table S 1, entry 6). When the concentration of 1a
was increased to 2 mM, the conversion did not exceed 60%
even after 24 h (i.e., 60% of 1a were consumed). In addition, by
using OleTCE, the formed 1b was over-oxidized to undec-
11-enoic acid (1e) and the remaining substrate 1a into
1,12-dodecandioic acid (1h), presumably by enzymatic
background reactions from the E. coli metabolism (Figure S 9b).
To avoid these undesirable background reactions, OleT was
purified. After the conversion of several ω-functionalized fatty
acids with OleT (Scheme 3a), different isomers of hydroxy fatty
acids were identified. GC-MS data indicated that the β-hydroxy
isomer (75% relative abundance of all detected hydroxy fatty
acids) was again the most prominent form. No over-oxidation
reactions of synthetized ω-alkenols or remaining substrate were
observed (Scheme 3a). Moreover, we recognized that with
reduced chain length, the overall conversion of the substrate
reduced and the ratio of decarboxylation and hydroxylation
shifted to favor the hydroxylation reaction (cf. above and Table 1,
entries 1-3). This means that the decarboxylating activity of OleT
in the hν/FMN/H2O2 system increases with growing chain length,
thus showing a similar trend to the one previously observed with
saturated fatty acids.[12] Long-chain fatty acids are assumed to
interact with hydrophobic amino acid residues in the substrate
channel of the active site, whereas smaller substrates are
accommodated entirely in the binding pocket, as shown for the
related enzyme CYPBS .
[8,33] This would imply that medium-chain
ß
fatty acids have less capacity to bind efficiently and form
O
OH
O
CO2
OleT
a)
g
a
R
R
R
R
R
b
+
+
OH
OH
OH
n
n
n
n
OH OH
hn
/FMN/H2O2 2 H2O
a
b
b
c
O
OH
O
CO2
OleT
g
R
b)
a
b
OH
n
n
O2 + 2 H+
2 H2O
2e-
OH OH
a
c
CamB
CamA
NAD+
HCOOH
NADH
CO2
FDH
Scheme 3. Oxidative decarboxylation and hydroxylation reactions of fatty acids to 1-alkenes with OleT by different electron supply systems. a) Decarboxylation of
saturated and functionalized fatty acids using the in situ light driven H2O2 generation system. b) Enzymatic redox cascade process for the decarboxylation of fatty
acids using the CamAB system.[8]
a productive binding mode.[8] Indeed, Faber and coworkers
could show the decarboxylation of short-chain carboxylic acids
with 99% alkene production in the CamAB system.[8] For ω-
amino lauric acid (4a) no conversion could be observed using
the light-driven system (Table 1, entry 7). We could furthermore
not detect any conversion with dodecandioic acid (5a),[8] even
though OleT activity towards this substrate has been
demonstrated before (Table 1, entry 8).[32]
Since our aim was to produce 1b from the bio-based
precursor 1a, further reactions were carried out with this hydroxy
fatty acid as a substrate. Results presented above describe the
optimal conditions for ω-alkenol synthesis in the hν/FMN/H2O2
system. To determine the influence of different electron donor
systems on the decarboxylation reaction with 1a as a substrate,
two methods were compared: hν/FMN/H2O2 (using H2O2 as
electron donor) and CamAB (using NADH as electron donor)
(Scheme 3). Interestingly, by using the CamAB electron supply
system, a full conversion of 2 mM of 1a could be achieved
(Table 1, entry 6). Furthermore, the formation of hydroxylated
by-products 1c was significantly reduced (Figure S 10).
Taken together, the ratio of decarboxylation to hydroxylation
shows a strong dependence on the chain length of the substrate
and the electron supply system used. Moreover, we could show
that OleT converts ω-hydroxy functionalized substrates (Table 1).
With the optimized hν/FMN/H2O2 system we could mainly
produce alkenols from the long-chain substrates 2a and 3a.
Similar results were observed earlier in the CamAB system with
hexadecanoic acid (16C) as substrate.[8] However, carboxylic
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
acids with short- to medium-chain lengths are rather poor
substrates for the decarboxylation with OleT by using the
Bis-Tris, Britton-Robinson and boric acid buffers) on 1e
formation, the speed of the reaction with EcAdhZ3-LND reduces
in Britton-Robinson buffer as compared to the results in Tris or
Bis-Tris buffer (Figure S 11a). In the case of LCAO_Af, the
reaction rate was identical in all buffers tested (Figure S 11b).
When comparing the influence of 10% vs. 30% (v/v) DMSO
used as co-solvent, it was found that 30% DMSO had a negative
effect on the activity of EcAdhZ3-LND, whereas in LCAO_Af
reactions, no difference was observed. According to reported
studies, the efficiency of the LCAO_Af reaction continuously
increases from pH 6 to a maximum pH 10,[25] whereas EcAdhZ3-
LND has an optimal pH between 5.5 and 6 for using NAD+ as
cofactor.[24] A pH of 7 appeared to offer the most favorable
conditions for both LCAO_Af and OleT.
hν/FMN/H2O2
system.
Nevertheless,
the
light-driven
decarboxylation is less complex in its application and therefore
offers more possibilities for application in multi-catalyst
approaches, especially due to the use of less regenerating
enzymes which might interfere with further enzymatic reactions.
We therefore used the light-driven electron supply system in all
following cascade approaches.
Synthesis of undec-10 en-1-oic acid
Once the decarboxylation reaction of the ω-hydroxy fatty acid 1a
was completed, the oxidation of the terminal alcohol 1b to a
carboxylic group to produce 1e via the aldehyde 1d was
attempted. Two enzymes were tested for this task: an enzyme
variant of the NAD+-dependent alcohol dehydrogenase from
E. coli (EcAdhZ3-LND) where the SSN (serine, serine,
To analyze the specific activity of both hydroxy group
oxidizing enzymes, the alcohol 1b and the corresponding
aldehyde 1d were used as substrates. EcAdhZ3-LND showed a
asparagine) motif was substituted by a LND (leucine, asparagine, specific activity of 0.083 mU mg-1 towards 1b and 3.8 mU mg-1
aspartic acid) motif in the nucleotide-binding region,[24] and a
putative flavin-dependent long-chain alcohol oxidase from
Aspergillus fumigatus (LCAO_Af).[25]
To determine the influence of the conditions established
for the decarboxylation reaction on the other enzymes of the
cascade (Scheme 2), we investigated different buffers and co-
solvents. Analyzing the influence of different buffers (Tris-HCl,
for 1d. The higher activity towards 1d explains the strong
preference of EcAdhZ3-LND to convert aldehydes and therefore
the preferred synthesis of 1e, which was also observed with
endogenous alcohol dehydrogenases present in the crude
extract of E. coli. In contrast, LCAO_Af showed an overall lower
activity against 1b (0.023 mU mg-1) and 1d (1.58 mU mg-1).
Table 1. Substrate scope of the OleT in in situ light driven H2O2 generation system (hν/FMN/H2O2) and CamAB cascade process under individually optimized
conditions:
Ratio
Decarboxylation/
Hydroxylation
Conc.
[mM]
OleT
[mg mL-1]
Conversion
[%]
Entry
Substrate
Cofactor supply
1[a]
2[a]
3[a]
4[b]
5[b]
3a
2a
1a
1a
1a
ω-OH-C16
ω-OH-C15
ω-OH-C12
ω-OH-C12
ω-OH-C12
2
2
2
2
2
0.3
0.3
0.3
0.3
1.0
hν/FMN/H2O2
hν/FMN/H2O2
hν/FMN/H2O2
hν/FMN/H2O2
hν/FMN/H2O2
99
78
25
6
4.2
2.1
0.14
0.5
28
0.8
6[c]
7[d]
8[d]
1a
4a
5a
ω-OH-C12
2
0.3
0.1
0.1
CamAB
100
n.c.
n.c.
19
-
ω-NH2-C12
ω-COOH-C11
0.5
0.5
hν/FMN/H2O2
hν/FMN/H2O2
-
Reaction conditions: [a] substrate, OleT, DMSO (10% v/v), FMN (10 µM) and EDTA (50 mM), Britton-Robinson buffer (50 mM, pH 7), at 25°C for 20 h. [b]
substrate, OleT, DMSO (10% v/v), FMN (10 µM) and EDTA (50 mM), KPi buffer (100 mM, pH 7.5), at 25°C for 20 h. [c] substrate, OleT, EtOH (10% v/v), CamAB
(0.05 U mL-1), catalase (1200 U mL-1), FDH (2 U mL-1), ammonium formate (100 mM), and NAD+ (200 μM), KPi buffer (100 mM, pH 7.5) at 25°C for 20 h: [d]
substrate, OleT, DMSO (5% v/v) FMN (10 μM) and EDTA (50 mM), Tris-HCl (50 mM Tris, 200 mM NaCl, pH 7.5), at 25 °C for 2 h.[12] For all reactions purified
OleT was used. Conversion was determined by: sum (c(all products))·100% / c(substrate). Detection by GC-FID and GC-MS analysis. n.c. = no conversion.
This was most probably due to a low soluble expression of
LCAO_Af (2.16 mg L-1 expression yield in comparison to
EcAdhZ3-LND with 166 mg L-1) which was attributed to inclusion
body formation.
To increase the concentration of soluble LCAO_Af in the
crude extract, different expression media such as TB, LB,
TYGPN and autoinduction media were tested, with the best
result (11.75 mg L-1) obtained by using autoinduction medium
(Figure S 1). Additionally, commercially purchased AO from
Pichia sp. was also tested but did not convert 1b.
Figure 2 shows a comparison of both enzymes in the
oxidation of 1b to 1d, which is then further oxidized to form 1e.
While the LCAO_Af system shows very little activity in the
oxidation of the terminal alcohol, the EcAdhZ3-LND system
achieved 100% conversion of the substrate into the desired
carboxylic acid 1e, with the intermediate aldehyde not being
detectable at any time point.
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
To further extend the cascade and to directly obtain 1f from
1a,
a sequential enzyme cascade using OleTCE in the
hν/FMN/H2O2 system, LCAO_Af as well as CvTA was applied.
Only a low production of 1f (8%) in a one-pot reaction was
achieved (Figure S 15), indicating that the use of a two pot-two
step reaction with intermediary extraction produce better results
in this case. The decrease in amine formation in the one-pot
reaction was caused by substantial over-oxidation to 1e. In the
amination of alcohols via intermediary aldehydes overoxidation
is a significant risk.[37,38,39]
Figure 2. Comparison between EcAdhZ3-LND and LCAO_Af in the synthesis
of 1e. General conditions: 1b (2.5 mM), DMSO (10%), NAD+ (2 mM) or FAD+
(0.25 mM), Britton-Robinson buffer (50 mM, 300 mM NaCl, pH 7.0) and
EcAdhZ3-LND (full line) or LCAO_Af (dashed line) (CE of 90 mg cdw),
reaction volume (3 mL). To have the same conditions as with 1a as starting
material in the hν/FMN/H2O2 system, EDTA (50 mM) and FMN (10 µM) were
also added to the reactions. Samples were derivatized by MSTFA and
analyzed by GC-FID. For this reaction, we used commercially purchased 1b.
To analyze the capability of a cascade approach for the
direct synthesis of 1e from 1a, a sequential enzyme cascade of
conversions by OleTCE in the hν/FMN/H2O2 system and
subsequent oxidation by EcAdhZ3-LND was performed to avoid
undesired attack of EcAdhZ3-LND on 1a. The light-driven
biocatalysis (4 h reaction time) resulted in conversion of 1a to a
mixture of 1b (17%) and 1h as by-product (43%). We were
pleased to find a complete conversion of the intermediary 1b to
1e by EcAdhZ3-LND within 4 additional hours (Figure S 12).
Figure 3. Amine formation from alcohol 1b in a one-pot reaction with
LCAO_Af and CvTA. General conditions: Britton-Robinson buffer (50 mM,
300 mM NaCl, pH 7.0), 1b (0.5 mM), DMSO (10%), FAD+ (0.25 mM), catalase
(1700 U), L-alanine (20 mM), PLP (1 mM), LCAO_Af (CE of 60 mg cdw) and
CvTA (CE of 120 mg of cdw) at 25°C for 90 min. Samples were derivatized by
MSTFA and analyzed by GC-FID. For this reaction, we used commercially
purchased 1b. 1b = 10 undec-1-enol (dashed line); 1e = undec-11-enoic acid
(chain line); 1f = 10-undecen-1-amine (full line).
Synthesis of 10-undecen-1-amine
By combining the oxidation of the terminal alcohol to an
aldehyde with a subsequent enzymatic reductive amination, 1b
can be used as starting material for the synthesis of undec-
10-en-1-amine (1f). Amine formation from alcohol 1b was tested
by combining EcAdhZ3-LND and an amine transaminase
(CvTA) from Chromobacterium violaceum. This approach
resulted in significant oxidation to 1e (80.7%) with only 15%
formation of the desired 1f (Figure S 13). In contrast, a one-pot
reaction with LCAO_Af and CvTA achieved 85% of 1f (Figure 3).
The lower activity of LCAO_Af towards forming the intermediary
aldehyde increases its availability for the amination reaction.
When the concentration of 1b was increased (1.5 mM), the
conversion slightly decreased and achieved 58% amine
(Figure S 14).
The challenge in asymmetric reductive amination, especially
when using L-alanine as amine donor, is to shift the equilibrium
to the product side. Therefore, to improve the amination step,
pyruvate must be removed during the reaction by using
additional enzymes like alanine dehydrogenase (AlaDH) or its
formation circumvented by using other amine donors.[34,35]
Furthermore, the incomplete transamination may not only be
due to the chemical equilibrium, but also to a poor solubility of
1b in the buffer medium.[36] Nevertheless, with undec-10-en-1-ol
as substrate, Pickl et al. showed only 16% conversion to the
respective amine; Significant amine formation was only
observed with 6C-8C alcohols.[25]
Synthesis of a long-chain terminal diol by olefin metathesis
The terminal alkenes from enzymatic decarboxylation of fatty
acids can also be used to produce long-chain terminal diols with
an internal double bond by ruthenium-catalyzed olefin
metathesis (Scheme 1). To achieve an efficient substrate
conversion, the choice of the proper catalyst is crucial. Apart of
its effectivity to convert the desired substrate, the reaction
conditions required are critical for the development of a chemo-
enzymatic cascade. Therefore, in a first step to establish the
olefin metathesis, different commercially available ruthenium-
based metathesis catalysts were tested for the conversion of 1b
in selected organic solvents. As a result, three catalysts C1-C3
(Figure 4) were identified that could convert 1b to 1g in the
presence of water and oxygen. GC-FID and GC-MS analyses
revealed several peaks corresponding to the desired product
and degradation products, possibly due to thermal
decomposition during GC analytics (Figures S 16 – S 18).
Therefore, the performance of the catalysts was assessed by
substrate consumption rather than by product formation
(Table S 3, Figure 5).
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
followed by extraction of the reaction products in ethyl acetate
(EtOAc). After evaporation of the solvent, the intermediates were
re-dissolved in isooctane and the metathesis performed in pure
organic solvent. Three different ω-hydroxy fatty acids (1a, 2a
and 3a) served as substrates and were successfully converted
to the respective long-chain alkenediols (icos-10-ene-1,20-diol
1g, Hexacos-13-en-1,26-diol (2g), Octacos-14-en-1,28-diol (3g))
within 3 hours of reaction, as judged by GC-MS measurements
(Figures S 16 and S 17). This sequential two-pot reaction can be
considered as the starting point, with both reactions being run
under their respective ideal conditions. Nevertheless, the
requirement of an intermediary extraction step makes this
reaction economically unfavorable.
Figure 4. Cross metathesis catalysts that were used for the olefin metathesis
reaction of 1b. C1
=
Dichloro [1,3-bis (2,6-isopropylphenyl)- 2-
imidazolidinylidene] -2-[[1-(methoxy(methyl) amino)-1-oxopropan- 2yl]oxy]
benzylideneruthenium (VI); C2 = Dichloro[1,3-bis (2,4,6-trimethylphenyl)-4- [(4-
ethyl-4-methyl piperazinyl)methyl]-2-imidazolidinylidene] -(2 isopropoxy
Secondly, the reaction was performed in a sequential one-
pot mode, with the enzymatic decarboxylation taking place in
purely aqueous buffer or already in presence of isooctane as the
second phase. Afterwards, with the addition of the catalyst C1
(and isooctane in the case of biocatalysis in pure buffer), the
metathesis was performed. In both cases, an enzymatic
conversion of ca. 90% 1a was estimated from GC-based
investigation of the aqueous phase. In the organic phase, only
the intermediate 1b and the desired metathesis product 1g were
found. These results indicate that OleT is active even in the
presence of isooctane as organic solvent; an important
prerequisite for the final one-pot reaction.
Finally, a simultaneous one-pot metathesis reaction was
performed. In this case, the reaction was conducted in the bi-
phasic buffer/isooctane system with both catalysts present from
beginning on. To protect the ruthenium compound from potential
light-induced degradation, the upper organic phase was shielded
from light during the course of the reaction. Even in this case,
the formation of the desired product 1g could be observed by
GC-MS measurements. However, conversions did not exceed
20%, which can be explained with the incompatibility of cell-free
extracts with the metathesis catalyst. Due to the high cost of
using purified enzyme, this option was not considered for the
cascade reaction. The most practical approach is therefore the
sequential cascade approach, with the in situ extraction of the
intermediate into the organic phase, followed by an efficient
metathesis reaction (Scheme 4).
benzylidene) ruthenium(VI) chloride and C3
trimethylphenyl)-4-[(trimethylammonio)methyl]-2-imidazolidinylidene]-
isopropoxybenzylidene)-ruthenium(VI)chloride.
= Dichloro [1,3-bis(2,4,6-
(2-
Based on these data, the reaction was then performed in a
biphasic system consisting of Tris buffer and isooctane. It was
found that only catalyst C1 catalyzes the conversion of 1b to 1g
in presence of an aqueous phase, with a conversion of 89% in
18 h reaction time while using a 5 mol-% catalyst loading
(Figure S 18).
In parallel, the metathesis was performed in isooctane with
C1 as catalyst in a semi-preparative scale. C1 converted
0.5 mmol 1b in 18 h and produced the desired 1g, which was
isolated after preparative TLC with 43% isolated yield.
Figure 5. Substrate conversion rates for the three selected catalysts in the
screened solvents. Conversion was determined by GC-FID as substrate
consumption.
Once the metathesis reaction of 1b was established, the
practicability of the combination of enzymatic and chemical
catalysis was tested. Different approaches, applying sequential
and simultaneous one- and two-pot conditions were tested.
Working in a biphasic system brings great advantages for the
planned chemo-enzymatic cascade reaction, since the
enzymatic decarboxylation can take place in aqueous buffer,
and the intermediately formed α-ω functionalized alkenol, being
less polar than the ω-hydroxy fatty acid, can immediately be
extracted into the organic phase where metathesis to form the
long-chain alkenediol is performed.
Scheme 1. Sequential chemo-enzymatic cascade to produce long-chain
alkendiol 1g from bio-based fatty acid 1a. In a first step, the light-driven
decarboxylation takes place in presence of an organic solvent (isooctane) to
aid the in situ extraction of the intermediate 1b into the organic phase. After
addition of the ruthenium catalyst C1, the metathesis takes place in presence
of water.
Firstly, the reaction was tested in a sequential two-pot mode,
with the decarboxylation by OleT run in purely aqueous buffer,
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
transfer were removed by ultracentrifugation through a centrifuge filter
unit with a 10 kDa cut-off membrane (3 mL, 4,000 × g at 4 °C, twice). The
remaining protein was resuspended in Britton-Robinson buffer (3 mL of
50 mM pH 7, 200 mM NaCl).
Conclusion
The conversion of ω-hydroxy fatty acids by the fatty acid
decarboxylase OleT gives rise to ω-alkenols, which can be
further converted to ω-unsaturated fatty acids, alkenylamines
and long-chain diols. We could show here the capacity of OleT
to produce terminal alkenes and the possibility to combine this
decarboxylase with various other enzymes and even with metal
catalysts in a number of different reaction sequences. On the
way to synthetic applications, however, several challenges
remain to be solved. In all examples, the extraction of the
intermediate product without purification turned out to be more
successful than cascade approaches. This underlines the
complexity of cascades,[40] and the requirement for a detailed
optimization of these multi-catalytic one pot reactions. The
formation of side-products by OleT and incomplete conversions
in cascade reactions must be overcome in order to facilitate the
downstream-processing. It should be also noted that current
applications of OleT rely on the use of purified enzyme.[8,9] The
total turnover numbers with cell-free extracts are considerably
lower.[11] Applications in whole cells in bacteria and yeast do not
exceed volumetric yield 100 mg L-1 or 10 mg L-1,[10] respectively.
The low solubility of fatty acids in water is another serious
problem. While the capacity of OleT to cleave non-activated C-
C-bonds under very mild reactions is of high synthetic interest,
the current state of OleT is far from concrete applications.
Despite these current limitations, the unique reaction catalyzed
by OleT is an important addition of the biocatalytic toolbox for
the valorization of bio-based resources.
For biocatalyses with purified OleT, protein concentration was
determined by Bradford assay. The assay was carried out in 96-well
plates. 50 µL of diluted protein samples (1:10, 1:100, 1:200) and BSA
standards (0, 20, 30, 40, 50, 60, 80 and 100 µg mL-1) were mixed with
200 µL Bradford reagent (Coomassie Brilliant Blue G-250). After 15 min
incubation time, the absorbance was measured at 595 nm using a
FLUOstar Omega UV/Vis spectrophotometer (BMG Labtech). Samples
for standard curves as well as protein samples were always prepared in
triplicates and used for the calculation of protein concentrations.
Cloning, expression and purification of EcAdhZ3-LND, LCAO_Af and
CvTA
A
pET28a vector with NAD+-dependent alcohol dehydrogenase
(EcAdhZ3-LND) from Escherichia coli[24] was kindly provided by the
group of Volker Sieber (Wissenschaftszentrum Straubing, Germany).
Transaminase from Chromobacterium violaceum (CvTA) was kindly
supplied by Karim Cassimjee (KTH Stockholm, Sweden). The alcohol
oxidase from Aspergillus fumingatus (LCAO_Af) (XP_753079.1) was
cloned into the pET28a vector with an N-terminal His-tag.[25] For that, the
codon-optimized LCAO_Af was synthetized at GeneArt life technology
(Regensburg, Germany) with corresponding restriction sites NdeI (N-
terminus) and XhoI (C-terminus). After cloning into the pET28a vector,
E. coli XL blue was transformed for the amplification of a constructed
plasmid.
Chemically competent E. coli BL21 (DE3) was used as expression strain
for EcAdhZ3-LND, LCAO_Af and CvTA. For efficient protein expression
enzymes were overexpressed in Terrific Broth (TB)[41] medium (250 mL)
with kanamycin (100 μg mL-1) by addition of Isopropyl β-D-1-
thiogalactopyranoside (IPTG 0.5 mM) and incubated overnight at 20 °C.
The cells were harvested by centrifugation (12,000 × g, 20 min, 4 °C) and
washed with phosphate-puffer (20 mM pH 7, 200 mM NaCl). After
sonication (3 min, 50% pulse cycle, 60% amplitude, 15 sec sonication /
15 sec pause) on ice, the cell debris was removed by centrifugation
(Beckman Coulter, Avanti J-26S XP – 10,000 × g, 20 min, 4°C). To
improve the amount of soluble LCAO_Af, different media such TB,
Lysogeny Broth medium (LB),[42] TYGPN (medium broth from
AMRESCO) and autoinduction medium (media components: 12 g L-1
Tryptone, 24 g L-1 yeast extract in 795 mL; buffer component: 1M KPi
buffer pH 7 (90 mL); induction components: 50 g L-1 glucose (10 mL),
20 g L-1 lactose (100 mL), 5 mL glycerin) in 50 mL volume were tested,
where the best result with autoinduction medium were obtained
(Figure S 1).
Experimental Section
General
Unless otherwise stated, all chemicals and compounds used during this
work were purchased from Sigma Aldrich (Steinheim, Germany), TCI
chemicals
(Eschborn,
Germany),
VWR
international
(Langenfeld/Rheinland, Germany), Alfa Aesar (Karlsruhe, Germany), or
Carl Roth (Karlsruhe, Germany) in the highest purity degree available.
Ruthenium catalysts were purchased from Apeiron Catalysts (Wrocław,
Poland).
Syntheses were carried out under standard air-exposed atmosphere
unless otherwise stated. Isolation of the reaction products was conducted
via column chromatography on silica gel type 60 (Carl Roth, Karlsruhe,
Germany), or preparative silica gel GF TLC 20x20 cm plates (Analtech,
Newark, USA). NMR spectra were recorded on a Bruker DPX200 NMR
spectrometer. Chemical shifts δ (in ppm) are given relative to TMS and
referenced to the undeuterated residues of deuterated chloroform as
internal standard.
Light-driven biotransformations
Assays for the light-driven biocatalysis were carried out in clear glass
tubes placed in a water bath with LED as a light source.[12] The influence
of different reaction conditions on the decarboxylation as well as
hydroxylation was analyzed in 3 mL reaction volume, mixed with a
magnetic stirrer at 100 rpm and incubated at 25 °C. The standard
biocatalytic reaction with OleT was carried out with different substrate
concentrations (0.5 to 2 mM), ethylenediaminetetraacetate disodium salt
(EDTA, 50 mM), flavinmononucleotide (FMN, 10 µM), dimethylsulfoxide
(DMSO, 5-10%) and OleTCE (400 μL) or purified OleT (0.3 or 1.0 mg mL-
1) in Britton-Robinson buffer (50 mM pH 7, 200 mM NaCl). The reaction
was started by addition of FMN and crude extract (60 mg of cell dry
Preparation of OleT for biocatalysis
OleT expression and purification was performed depending on the used
electron supply system.[8,12] Biocatalyses and analytics with the CamAB
system were performed according to an earlier established protocol.[8]
Before using the crude extract of OleT for biocatalyses, small molecules
which might interfere with hydrogen peroxide formation or electron
8
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
weight, cdw) or purified enzyme. Biocatalysis reactions were run for 4 h
or overnight. For analysis, samples (400 μL) were withdrawn at the
desired times and the reaction was stopped with hydrochloric acid (40 μL,
37% w/v). Myristic acid (0.3 mM) was added as internal standard and the
sample further processed for GC-FID or GC-MS analysis (cf. below).
28H); 13C NMR (200 MHz, chloroform-d) δ 25.75, 27.19, 29.05, 29.10,
29.30, 29.44, 29.59, 29.73, 32.56, 32.77, 62.97, 129.88.
General procedure: Solvent and catalyst screening for metathesis
conditions
To investigate the influence of the co-solvent on OleT activity and
stability, N,N-dimethylformamide (DMF), tetrahydrofuran (THF) and 1,4-
dioxane were additionally used as alternative additives to DMSO. The
investigation of optimum pH, Britton-Robinson buffer (H3BO3, H3PO4,
CH3COOH, each 50 mM) with high buffering capacity in the pH range of
2 to 12 was selected. Other buffers (50 mM of Bicine/ Bis-Tris/ Borate/ 4-
Olefin metathesis reaction of 1b was performed using 1.5 mL closed
glass vials under air-exposed conditions in 1 mL solvent. For the reaction,
20 µmol substrate was used in combination with 5 mol-% catalyst C1, C2
(Dichloro[1,3-bis (2,4,6-trimethylphenyl)-4- [(4-ethyl-4-methylpiperazinyl)
methyl]- 2-imidazolidinylidene]- (2-isopropoxybenzylidene) ruthenium(VI)
chloride), AquaMet) and C3 (Dichloro [1,3-bis(2,4,6-trimethylphenyl) -4-
(2-hydroxyethyl)-1-piperazineethanesulfonic
acid
(HEPES)/
3-(N-
[(trimethylammonio)methyl]
-2-imidazolidinylidene]-
(2-
morpholino) propanesulfonic acid (MOPS)/ Phosphate/ Tris-HCl, or
Imidazole) were used at a pH of 7. All reactions were performed as
duplicates.
isopropoxybenzylidene)-ruthenium(VI)chloride, StickyCat). The synthesis
was conducted in tetrahydrofuran (THF), 2-methyl-tetrahydrofuran (2-
MeTHF), and 2,2,4-trimethylpentane (isooctane) at 50 °C. After the
reaction, the mixtures were cooled down to room temperature. 100 µL
samples were taken and centrifuged (2 min, 17 000 × g). Subsequently,
the supernatant was analyzed by GC-FID and GC-MS measurements.
Alkenylamine synthesis
For the synthesis of a terminal alkenylamine from ω-hydroxy fatty acids
in a sequential one-pot enzyme cascade, OleT, EcAdhZ3-LND or
LCAO_Af and a CvTA were used. All enzymes were used as crude
extract. First, light-driven decarboxylation by OleTCE was performed for 4
or 20 hours. Subsequently, the illumination was interrupted and the other
enzymes, NAD+ and L-alanine as amino donor were added. The
biocatalysis reactions were carried out with substrate (1 mM), EDTA
(50 mM), FMN (10 µM), DMSO (10%), L-alanine (10-50 mM), FAD
(0.25 mM), PLP (1 mM), enzymes OleTCE and CvTA (400 μL)
respectively, as well as LCAO_Af (200 µL) in Britton-Robinson buffer (a
long range buffer consisting in a mixture of phosphoric, boric and acetic
acids) (50 mM pH 7, 200 mM NaCl) with a total volume of 3 mL. The
reactions were incubated at 25 °C for 20 h with 100 rpm. For analysis,
samples were processed as described above. All reactions were
performed as triplicates.
Analytics
For the analysis of the reactions by GC-MS or GC-FID samples were
derivatized by N-methyl-N-(trimethylsilyl)trifluoroacetamide (MSTFA) in
order to lower their boiling points and make them easier to separate. For
this, samples from the bioconversion (400 µL) were extracted with ethyl
acetate (700 μL). After centrifugation (12,000× g for 1 min) the organic
phase was removed. The extraction was repeated once and the
combined supernatants evaporated. A derivatization was then carried out
with MSTFA (25 μL) for 20 minutes at 65 °C. Subsequently, ethyl acetate
(150 µL) was added and the samples were transferred into GC vials.
Substrate and product identification and quantification was performed by
GC-MS using a DSQII TraceGC Ultradevice (Thermo Fisher Scientific)
using a Zebron ZB-1MS (30 m × 0.25 mm × 0.25 μm, Phenomenex)
column. 7 μL sample volumes were injected at a split ratio of 1:10 and a
temperature of 280 °C using helium as carrier gas with a flow rate of
1.2 mL min-1. The analytes were separated according to the temperature
programs shown in Figure S 22. The ion source of the mass
spectrometer was heated to 280 °C and ions were detected in the
positive mode in the range of m/z = 50-650. Signals were recorded from
4 minutes after injection to avoid detector overloading. Ions were
detected with a detector gain of 3×105 (multiplier voltage 1390 V) and
8,000 scans sec-1.
Production of Icos-10-en-1,20-diol
The olefin metathesis was performed using standard Schlenk technique
under nitrogen atmosphere. Commercially available 1b (1.2 mmol,
200 mg) and catalyst C1 (Dichloro[1,3-bis(2,6-isopropylphenyl)-2-
imidazolidinylidene]-2-[[1-(methoxy(methyl)amino)-1-oxopropan-
2yl]o’xy] benzylideneruthenium (VI)) (GreenCat) (4 mol-%) were mixed in
anhydrous dichloromethane (DCM) for 18 h under nitrogen atmosphere
at reflux conditions (40 °C). After cooling down to room temperature, vinyl
ether (30 mg) was added in order facilitate the separation of the product
from the catalyst. For purification, preparative silica gel thin layer
chromatography was performed in five cycles with light petroleum:EtOAc
(9:1(v/v) as mobile phase. Isolated 1g (99 mg, 51% yield) was obtained
as brown wax.
Alternatively, samples were analyzed by GC-FID using a Shimadzu GC-
2010 plus (Shimadzu, Duisburg) using a CP-Sil 5 CB (30 m × 0.25 mm ×
0.25 μm, Agilent) column for separation. In contrast to GC-MS
measurements, this method has a much broader linear measuring range
and generally smaller errors in the quantification. 7 μL sample volumes
were injected at a split ratio of 1:10 and a temperature of 300 °C. A flow
rate of 40 mL min-1 was used for the carrier gas (synthetic air). The
analytes were separated using the temperature program shown in
Figure S 23. The detector was heated to 350 °C.
1
1b: H NMR (200 MHz, chloroform-d) δ 5.98 – 5.64 (m, 1H), 5.09 – 4.77
(m, 2H), 3.74 – 3.49 (m, 2H), 2.24 – 1.88 (m, 2H), 1.71 – 1.47 (m, 2H),
1.47 – 1.08 (m, 12H); 13C NMR (200 MHz, chloroform-d) δ 25.74, 28.93,
29.12, 29.42, 29.55, 32.81, 33.81, 63.08, 114.11, 139.22.
2b: 1H NMR (200 MHz, chloroform-d) δ 5.87-5.61 (m, 1H), 5.00-4.78 (m,
2H), 3.65-3.47 (t, J = 6.0, 2H), 2.06-1.87 (m, 2H), 1.36-1.06 (m).
3b: 1H NMR (200 MHz, chloroform-d) δ5.81-5.62 (m, 1H), 5.00-4.77 (m,
2H), 3.69-3.44 (t, J = 6.0, 2H), 2.05-1.87 (m, 2H), 1.34-1.13 (m).
1
1g: H NMR (200 MHz, chloroform-d) δ 5.54 – 5.31 (m, 2H), 3.63 (t, J =
6.5 Hz, 4H), 2.07 – 1.88 (m, 4H), 1.77 – 1.49 (m, 4H), 1.44 – 1.19 (m,
9
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
[35] D. Koszelewski, K. Tauber, K. Faber, W. Kroutil, Trends Biotechnol.
2010, 28, 324–332.
Acknowledgements
[36] U. Kaulmann, K. Smithies, M. E. Smith, H. C. Hailes, J. M. Ward,
Enzyme Microb. Technol. 2007, 41, 628–637.
[37] M. Schrewe, N. Ladkau, B. Bühler, A. Schmid, Adv. Synth. Cat. 2013,
355, 1693–1697.
[38] A. Karau, V. Sieber, T. Haas, H. Haeger, K. Grammann, B. Buehler, L.
Blank, A. Schmid, G. Jach, B. Lalla, A. Mueller, K. Schullehner, P.
Welters, T. Eggert, A. Weckbecker (Evonik Degussa Gmbh),
US20100324257 A1, 2008.
[39] N. Ladkau, M. Assmann, M. Schrewe, M. K. Julsing, A. Schmid, B.
Bühler, Metab. Eng. 2016, 36, 1–9.
We thank Ioannis Zachos (TU Munich) for fruitful discussions
about the enzymatic decarboxylation of fatty acids. The Federal
Ministry for Innovation, Science and Research of North Rhine-
Westphalia (grant number PtJ-TRI/1411ng006) is gratefully
acknowledged for financial support.
[40] R. Kourist, S. Schmidt, K. Castiglione, Chemistry 2017, DOI:
Keywords: Biocatalysis; Renewable resources; Decarboxylase;
10.1002/chem.201703353.
[41] K.D. Tartoff, and C.A. Hobbs, Bethesda Res. Lab. Focus 1987, 12.
[42] G. Bertani, J. Bacteriol. 1951, 62, 293–300.
Olefin Metathesis; Cascade reaction.
[1]
[2]
[3]
[4]
A. Behr, J. P. Gomes, Eur. J. Lipid Sci. Technol. 2010, 112, 31–50.
K. Hu, D. Zhao, G. Wu, J. Ma, RSC Adv. 2015, 5, 85996–86005.
M. Jelling, J. Polym. Sci. B Polym. Lett. Ed. 1990, 28, 181.
J. O. Metzger, U. Bornscheuer, Appl. Microbiol. Biotechnol. 2006, 71,
13–22.
[5]
[6]
M.-K. Kang, J. Nielsen, J. Ind. Microbiol. Biotechnol. 2017, 44, 613–622.
C. E. Wise, J. L. Grant, J. A. Amaya, S. C. Ratigan, C. H. Hsieh, O. M.
Manley, T. M. Makris, J. Biol. Inorg. Chem. 2017, 22, 221–235.
H. Xu, L. Ning, W. Yang, B. Fang, C. Wang, Y. Wang, J. Xu, S. Collin,
F. Laeuffer, L. Fourage, S. Li, Biotechnol. Biofuels 2017, DOI
10.1186/s13068-017-0894-x.
A. Dennig, M. Kuhn, S. Tassoti, A. Thiessenhusen, S. Gilch, T. Bulter,
T. Haas, M. Hall, K. Faber, Angew. Chem. Int. Ed. 2015, 54, 8819–
8822.
[7]
[8]
[9]
A. Dennig, M. Kuhn, S. Tassoti, A. Thiessenhusen, S. Gilch, T. Bülter,
T. Haas, M. Hall, K. Faber, Angew. Chem. 2015, 127, 8943–8946.
[10] Y. Liu, C. Wang, J. Yan, W. Zhang, W. Guan, X. Lu, S. Li, Biotechnol.
Biofuels 2014, 7, 28.
[11] M. A. Rude, T. S. Baron, S. Brubaker, M. Alibhai, S. B. Del Cardayre, A.
Schirmer, Appl. Environ. Microbiol. 2011, 77, 1718–1727.
[12] I. Zachos, S. K. Gassmeyer, D. Bauer, V. Sieber, F. Hollmann, R.
Kourist, Chem. Commun. 2015, 51, 1918–1921.
[13] S. Matthews, J. D. Belcher, K. L. Tee, H. M. Girvan, K. J. McLean,
S.E.J. Rigby, J. Biol. Chem. 2017, 292, 5128–5143.
[14] S. Matthews, K. L. Tee, N. J. Rattray, K. J. McLean, D. Leys, D. A.
Parker, R. T. Blankley, A. W. Munro, FEBS Lett. 2017, 591, 737–750.
[15] H. Shalan, M. Kato, L. Cheruzel, Biochim. Biophys. Acta 2017, DOI:
10.1016/j.bbapap.2017.06.002.
[16] C. H. Hsieh, T. M. Makris, Biochem. Biophys. Res. Commun. 2016, 476,
462–466.
[17] B. Fang, H. Xu, Y. Liu, F. Qi, W. Zhang, H. Chen, C. Wang, Y. Wang,
W. Yang, S. Li, Sci. Rep. 2017, 7, 44258.
[18] L. McIver, C. Leadbeater, D. J. Campopiano, R. L. Baxter, S. N. Daff, S.
K. Chapman, A. W. Munro, Eur. J. Biochem. 1998, 257, 577–585.
[19] J.-B. Wang, R. Lonsdale, M. T. Reetz, Chem. Commun. 2016, 52,
8131–8133.
[20] K. Kai, H. Ohnishi, M. Shimatani, S. Ishikawa, Y. Mori, A. Kiba, K.
Ohnishi, M. Tabuchi, Y. Hikichi, ChemBioChem 2015, 2309–2318.
[21] C. H. Hsieh, X. Huang, J. A. Amaya, C. D. Rutland, C. L. Keys, J. T.
Groves, R. N. Austin, T. M. Makris, Biochemistry 2017, 56, 3347–3357.
[22] J. O. Metzger, Eur. J. Lipid Sci. Technol. 2009, 111, 865–876.
[23] D. Scheps, S. H. Malca, S. M. Richter, K. Marisch, B. M. Nestl, B.
Hauer, Microb. Biotechnol. 2013, 694–707.
[24] A. Pick, W. Ott, T. Howe, J. Schmid, V. Sieber, J. Biotechnol. 2014, 189,
157–165.
[25] M. Pickl, M. Fuchs, S. M. Glueck, K. Faber, ChemCatChem 2015, 7,
3121–3124.
[26] Á. Gómez Baraibar, D. Reichert, C. Mügge, S. Seger, H. Gröger, R.
Kourist, Angew. Chem. Int. Ed. 2016, 55, 14823–14827.
[27] Á. Gómez Baraibar, D. Reichert, C. Mügge, S. Seger, H. Gröger, R.
Kourist, Angew. Chem. 2016, 128, 15043–15047.
[28] A. Wetzel, A. Limbach, M. Brinks, M. Schelwies (Basf Se, Basf Schweiz
Ag), WO 2014023549 A1, 2014.
[29] E. Ricca, B. Brucher, J. H. Schrittwieser, Adv. Synth. Cat. 2011, 353,
2239–2262.
[30] J. Belcher et al., J. Biol. Chem. 2014, 289, 6535–6550.
[31] J. Kuper, T. S. Wong, D. Roccatano, M. Wilmanns, U. Schwaneberg, J.
Am. Chem. Soc. 2007, 129, 5786–5787.
[32] A. Dennig, S. Kurakin, M. Kuhn, A. Dordic, M. Hall, K. Faber, Eur. J.
Org. Chem. 2016, 2016, 3473–3477.
[33] O. Shoji, Y. Watanabe, J. Biol. Inorg. Chem. 2014, 19, 529–539.
[34] R. C. Simon, N. Richter, E. Busto, W. Kroutil, ACS Catal. 2014, 4, 129–
143.
10
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Samiro Bojarra, Dennis Reicher, Marius
Grote, Álvaro Gomez Baraibar, Alexander
Dennig, Bernd Nidetzky, Carolin Mügge,
and Robert Kourist]*
Page No. – Page No.
Bio-based α,ω-functionalized
hydrocarbons from multi-step reaction
sequences with bio- and metallo-
catalysts based on the fatty acid
decarboxylase OleTJE.
OleT from Jeotgalicoccus sp. ATCC 8456 holds great industrial potential for using
natural fats as renewable feedstock. Different cascades with OleT to produce high
value products by enzymatic and chemo-enzymatic reaction sequences are
presented.
11
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701804
ChemCatChem
FULL PAPER
[1] R. Kourist, S. Schmidt, K. Castiglione, Chemistry 2017, DOI: 10.1002/chem.201703353.
12
This article is protected by copyright. All rights reserved.