Artificial Light-Harvesting Complexes Enable Rieske Oxygenase Catalyzed Hydroxylations in Non-Photosynthetic cells

Article abstract of DOI:10.1002/anie.201914519

In this study, we coupled a well-established whole-cell system based on E. coli via light-harvesting complexes to Rieske oxygenase (RO)-catalyzed hydroxylations in vivo. Although these enzymes represent very promising biocatalysts, their practical applicability is hampered by their dependency on NAD(P)H as well as their multicomponent nature and intrinsic instability in cell-free systems. In order to explore the boundaries of E. coli as chassis for artificial photosynthesis, and due to the reported instability of ROs, we used these challenging enzymes as a model system. The light-driven approach relies on light-harvesting complexes such as eosin Y, 5(6)-carboxyeosin, and rose bengal and sacrificial electron donors (EDTA, MOPS, and MES) that were easily taken up by the cells. The obtained product formations of up to 1.3 g L^?1 and rates of up to 1.6 mm h^?1 demonstrate that this is a comparable approach to typical whole-cell transformations in E. coli. The applicability of this photocatalytic synthesis has been demonstrated and represents the first example of a photoinduced RO system.

Full text of DOI:10.1002/anie.201914519

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Artificial light-harvesting complexes enable Rieske oxygenase-
catalyzed hydroxylations in non-photosynthetic cells
Authors: Fatma Feyza Özgen, Michael Ernst Runda, Bastien O. Burek,
Peter Wied, Jonathan Z. Bloh, Robert Kourist, and Sandy
Schmidt
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: Angew. Chem. Int. Ed. 10.1002/anie.201914519
Angew. Chem. 10.1002/ange.201914519
Link to VoR: http://dx.doi.org/10.1002/anie.201914519
http://dx.doi.org/10.1002/ange.201914519

10.1002/anie.201914519
Angewandte Chemie International Edition
COMMUNICATION
Artificial light-harvesting complexes enable Rieske oxygenase-
catalyzed hydroxylations in non-photosynthetic cells
F. Feyza Özgen,[a] Michael E. Runda,[a] Bastien O. Burek,[b] Peter Wied,[a] Jonathan Z. Bloh,[b] Robert
Kourist,[a] and Sandy Schmidt*[a]
Abstract: In this study, we coupled a well-established whole-cell
system based on E. coli via light-harvesting complexes to Rieske
oxygenase (RO)-catalyzed hydroxylations in vivo. Although these
enzymes represent very promising biocatalysts, their practical
applicability is hampered by their dependency on NAD(P)H as well as
their multi-component nature and intrinsic instability in cell-free
systems. In order to explore the boundaries of E. coli as chassis for
artificial photosynthesis, and due to the reported instability of ROs, we
used these challenging enzymes as model system. The light-driven
approach relies on light-harvesting complexes such as eosin Y, 5(6)-
carboxyeosin or rose bengal and sacrificial electron donors (EDTA,
MOPS and MES) that were easily up-taken by the cells. The obtained
product formations of up to 1.3 g/L and rates of up to 1.6 mM/h
demonstrate that this is a comparable approach to typical whole-cell
transformations in E. coli. The applicability of this photocatalytic
synthesis has been demonstrated and represents the first example of
a photo-induced RO system.
NADPH, the often-used glucose dehydrogenase utilizes only a
part of the electron pairs from each glucose molecule. In order to
solve this challenge, many alternative solutions are currently
under consideration.[16–18] Next to linking photochemistry to
enzymes in vitro for cofactor regeneration,[18–24] autotrophic and
chemolithoautotrophic organisms have recently received
attention as they are capable of utilizing inorganic compounds as
electron donors.[25–29] Light-driven whole-cell reactions in
cyanobacteria show the same reaction rates as E. coli.[26,27,29] Yet,
the strong absorption of the photosynthetic apparatus lead to self-
shading of the cells at high densities, thus resulting in a low light
utilization and a reduced photosynthetic activity.[30] On the other
hand, introducing artificial photosynthesis in heterotrophic
bacteria such as E. coli offers the advantage of utilizing a
genetically easy-to-manipulate organism along with the capability
of producing high amounts of soluble protein within the cells.
Additionally, these systems are less prone to the inhibiting effects
of self-shading at high cell densities. Currently reported artificial
light-driven approaches in heterotrophs comprise the use of
inorganic-bio hybrid systems and the coupling of organic
photosensitizers to biotransformations in vivo.[31,32] One of the
earliest examples of a whole-cell reaction was reported using
recombinant E. coli coupled to photocatalytic H2 production via an
extracellular photosensitizer (TiO2) and methyl viologen as
electron mediator.[31] Similarly, the light-driven H2 evolution and
C=C or C=O bond hydrogenation by Shewanella oneidensis using
methyl viologen was shown.[33] These are interesting systems,
however, the toxicity of methyl viologen is well known, thus
hampering large-scale applications. A direct and perhaps the
most applicable approach has been reported by Park and
coworkers.[32] This light-driven catalysis is based on in vivo
photoreduction of a P450 by using different fluorescent dyes and
sacrificial electron donors.[32] Although operating at low product
concentrations, it represents a highly promising system for the
challenging multi-component ROs since these enzymes usually
exhibit high catalytic activities despite low expression levels, and
thus a high potential for artificial photosynthesis approaches in
E. coli (Figure 1).
Nature's creativity in developing solutions for C-H-bond
functionalization reactions like hydroxylations at activated or non-
activated C-H-bonds is remarkably shown by an expansive list of
metal-dependent enzymes.[1,2] These enzymes, like the Rieske
non-heme iron oxygenases (ROs) are able to activate molecular
oxygen in order to generate reactive oxygen species capable of
hydroxylating alkyl substrates, but also to promote further
oxidative transformations.[3–11] For many of these reactions no
'classical' chemical counterpart is known. Due to their
dependency on complex electron transport chains[12] as well as to
provide an efficient in situ cofactor regeneration, the majority of
synthetic applications of ROs relies on recombinant whole-cell
catalysts. Generally, for such reactions various concepts have
been developed that rely on electron supply via the metabolism of
living heterotrophic cells.[13–15] In synthetic applications, the
nicotinamide cofactors are recycled by using energy-rich organic
molecules as electron donors. In most cases, only a small fraction
of the electrons provided by these sacrificial co-substrates is
utilized, resulting in a poor atom efficiency.[14] Moreover, when
glucose is supplied as sacrificial substrate for the recycling of
[a]
Fatma Feyza Özgen, MSc; Michael Ernst Runda, BSc; Dipl. Ing.
Peter Wied; Prof. Dr. Robert Kourist; Dr. Sandy Schmidt
Institute of Molecular Biotechnology
Graz University of Technology
Petersgasse 14/1, 8010 Graz, Austria
E-mail: s.schmidt@tugraz.at
[b]
[+]
Bastien O. Burek, MSc; Dr. Jonathan Z. Bloh
DECHEMA-Forschungsinstitut
Theodor-Heuss-Allee 25, 60486 Frankfurt am Main, Germany
Figure 1. In vivo photoactivation of a Rieske non-heme iron oxygenase by an
artificial light-harvesting complex. The catalytic turnover of the oxygenase-
component is mediated by the excited photosensitizer that transfers electrons
from the sacrificial electron donor to the oxygenase within the cytoplasm of E.
coli.
These authors contributed equally to this work.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201914519
Angewandte Chemie International Edition
COMMUNICATION
Herein, we demonstrate the general feasibility of this light-induced
concentrations of 1a and 2a were lower than previously
reported,[34] which we mostly contribute to a different expression
protocol (19 hours instead of 2 hours) and a lower cell density
(100 gWCW/L instead of 200 gWCW /L) than previously reported.[34]
However, we first investigated the light-driven system by using
CDO-containing whole-cells at 100 gWCW/L in order to avoid self-
shading at too high cell densities and used cells expressed under
the conditions mentioned above (19 hours).
approach together with
a characterization of the crucial
parameters determining the catalytic efficiencies of the light-
driven in vivo enzymatic reaction.
We explored eosin Y (2,4,5,7-tetrabromofluorescein, EY) and its
xanthene derivatives as well as safranin O (SO) as efficient
photosensitizers to drive RO-catalyzed hydroxylation reactions
under illumination with LED or fluorescent lamps. As electron
donors, either 3-(N-morpholino)propane-sulfonic acid (MOPS), 2-
(N-morpholino)ethanesulfonic acid (MES) or ethylendiamin-
tetraacetic acid (EDTA) have been investigated.[23] In this way, the
successive transfer of electrons reduces the catalytic iron and
drives the conversion of (R)-limonene 1 into (1R,5S)-carveol 1a,
toluene 2 into benzyl alcohol 2a or indene 3 to 1-indenol 3a and
cis-or trans-1,2-indandiol 3b by the ROs under visible light
irradiation and without the need of NAD(P)H as redox partner
(Scheme 1).
We became interested in different photosensitizer / electron donor
combinations to drive the light-driven whole-cell hydroxylation
catalyzed by the ROs (Figures S6-S12, Table 1, Table S9 and 10).
Table 1. Photobiocatalytic hydroxylation of (R)-limonene 1, toluene 2 and
indene 3 catalyzed by CDO M232A and NDO H295A, respectively, under dark
and light conditions.
Whole-cell
Product
activity[c]
Reaction
conditions
Subs-
trate
Enzyme
de or
dr[b]
[%]
Conc.
mU/gWCW
8.0
[mM][a]
whole-cells
RO
hυ
dark / glucose
1.1 ± 0.1
HO
photosensitizer, e- donor,
30°C, buffer pH 7.2,
O2
CDO
M232A
> 99
n.d.
dark / CE/ MES
light / CE / MES
light / CE / MES
dark / glucose
n.d.
2.5
0.1 ± 0.04
0.4 ± 0.05
0
1
2
3
1a
2a
3a
1
2
whole-cells
RO
hυ
Empty
vector
n.d.
8.8
OH
photosensitizer, e- donor,
30°C, buffer pH 7.2,
O2
0.6 ± 0.02
0
NDO
H295A
dark / CE/ MES
light / CE / MES
n.d.
2.8
OH
n.a.
whole-cells
RO
hυ
0.2 ± 0.01
OH
Empty
vector
light / CE / MES
0
n.d.
photosensitizer, e- donor,
30°C, buffer pH 7.2,
O2
OH
3b
dark / glucose
dark / CE/ MES
light / CE / MES
dark / glucose
dark / CE/ MES
light / CE / MES
100 : 0
100 : 0
90 : 10
100 : 0
n.a.
n.d.
n.d.
124
n.d.
n.d.
107
4.8 ± 0.8
1.5 ± 0.2
8.3 ± 0.08
2.3 ± 0.17
0.5 ± 0.03
8.5 ± 0.4
CDO
M232A
Scheme 1. Light-driven whole-cell oxyfunctionalization reactions catalyzed by
CDO or NDO, respectively.
NDO
H295A
3
We have chosen cumene dioxygenase (CDO, from
Pseudomonas fluorescens IP01) and naphtalene dioxygenase
(NDO, from Pseudomonas sp. NCIB 9816-4) as model enzymes
(Figure S1, Tables S1-S7) since both enzymes have been
extensively studied for a long time and many redesigned variants
have been investigated. We became particularly interested in two
variants of CDO and NDO, who have been engineered toward the
asymmetric dihydroxylation of olefins. CDO variant M232A
converts 1 almost exclusively to 1a (ee > 98%),[34] whereas NDO
86 : 14
Empty
vector
light / CE / MES
18 : 82
n.d.
0.7 ± 0.2
Reaction conditions dark: [substrate] = 10 mM, [glucose] = 20 mM, [whole cells]
100 gWCW/L (E. coli JM109 (DE3)_pDTG141_NDO H295A or E. coli
=
JM109_pCDOv1_CDO M232A), sodium phosphate buffer (pH 7.2, 50 mM), 24
hours. Reaction conditions light: [substrate] = 10 mM, [whole cells] = 100 gWCW/L
(E. coli JM109 (DE3)_pDTG141_NDO H295A or E. coli JM109_pCDOv1_CDO
M232A), MES buffer (50 mM), white light illumination (max. 112 µE L-1 s-1) 24
hours; n.a. not applicable; n.d. not determined. [a] For 3, product concentrations
refer to the sum of 3a and 3b; [b] Diastereomeric ratio cis:trans-3b has been
determined after 4-6 hours of reaction; [c] Determined from the linear range of
product formation determined from the kinetic profiles for each reaction (Figures
S20-S25).
variant H295A shows
a
different ratio between allylic
monohydroxylation and cis-dihydroxylation for several substituted
arenes.[35] First, we investigated the expression of the whole RO
system under different culture conditions (SDS-PAGE, Figure S2)
and confirmed the RO activity by an agar plate assay based on
indigo formation (Figure S3, S4) and could identify significant
activity toward indole.[36] To identify the cell density leading to the
highest product formation catalyzed by CDO expressed under
different culture conditions, biotransformations supplemented
with glucose (20 mM) and 1 (10 mM) as substrate have been
performed under dark conditions (Table S8). As expected, CDO-
containing whole cells (100 gWCW/L) expressed in TB medium at
30°C gave the highest activity for 1 and the product 1a was
obtained with an ee of > 99% (Figure S5). The obtained product
We first chose MES since it has been successfully used as
efficient electron donor previously,[37] is non-toxic and can be up
taken by E. coli cells.[38,39] The E. coli strain herein used is lacking
a natural uptake system for flavins,[41] thus we decided to choose
a PS that can easily enter the cells[32] while showing similar redox
properties such as flavins. 5(6)-carboxyeosin (CE) has been
chosen first since it possesses excellent photosensitizer
properties (Figure S8) with an ERedox of -1.06 V, which is similar to
the ERedox of proflavine.[40] Performing the photoenzymatic
This article is protected by copyright. All rights reserved.

10.1002/anie.201914519
Angewandte Chemie International Edition
COMMUNICATION
hydroxylation of 1 and 2 with 50 mM MES and 100 µM 5(6)-
carboxyeosin (CE) at a cell density of 100 gWCW/L resulted in a
smooth formation of the desired products 1a (up to » 360 µM in
24 hours) and 2a (up to » 200 µM in 24 hours) under illumination
with white light. Nonetheless, the obtained concentrations of 1a
and 2a always remained low (£ 360 µM, Table 1). Due to the
known toxicity of 1 as well as 1a on whole-cells, we turned our
attention toward indene 3 as typical substrate for ROs. In a next
To investigate whether the observed product formation was
strictly light-dependent and only proceeded through the electron
transfer mediated by the photosensitizer, we conducted control
reactions with an empty vector control, but performed in light and
dark with and without electron donor (Figure 2A, Figure S19).
A)
set
of
experiments,
we
investigated
the
same
photosensitizer/electron donor combination for both model
enzymes (Table 1) and were pleased to find that conversions of
up to 85% were achieved with 3. Since 3 was the most promising
substrate to in-depth characterize our light-driven system, we
turned our attention to the investigation of further
photosensitizer/electron donor combinations. We investigated EY,
rose bengal (RB) and CE with EDTA, but also with MOPS and
MES as sacrificial electron donors that can be up taken by cells
(Table 2).[39,40,43] The electron donor can constitute an obstacle in
this photobiocatalytic setup,[23] as EDTA may suffer e.g. from
incompatibility with the RO due to its ability to sequester the Fe3+
ion which is located in the active site of the oxygenase. However,
we did not see any activity loss of NDO H295A and CDO M232A
in dark reactions supplemented with EDTA (Figure S13) nor any
toxicity effects of MES/MOPS on the cells (Figure S14). Moreover,
when performing the light-driven hydroxylations with lysed cells,
product formations of only 5.6 mM compared to 8.5 mM with
whole cells were achieved (Table S11), indicating that the cells
do not suffer from the electron donors or decomposition products
thereof. The obtained product formations with 3 as substrate are
summarized in Table 2. Reactions supplemented with EDTA led
in general to a lower product formation compared to reactions with
MOPS and MES. However, also with EDTA product
concentrations of up to 7.5 mM could be achieved, leading to the
assumption that EDTA is not sequestering the catalytic iron ion
from the enzyme’s active site. Moreover, we were pleased to find
that the utilization of 3 as substrate boosted the product formation
in the mM range, leading to a conversion of up to >85% within 24
hours when using CE in combination with MES (Table 1 and 2).
The determination of the incident photon flux density (Figure
S11B) revealed that the light intensity at each position of the light
reactor varies in a range between 32-112 µE L-1 s-1, which causes
a light-intensity-dependent photochemical background reaction.
This photochemical background reaction is only observed when 3
is used as substrate and leads to the accumulation of trans-3b
within 6-20 hours of reaction (Figure S15-S17). Moreover, 1-
indanone formation has been observed, which we attribute to an
isomerization reaction of 3a (Table S12).
B)
Figure 2. A) Performed control reactions using NDO H295A under light ( ) and
dark (•) conditions with (+) or without (-) 100 µM EY, RB or CE in the presence
of NDO H295A (red bars) or with an empty vector control (grey bars) in 50 mM
MOPS. Values for the empty vector control were the highest that have been
achieved when the max. light intensity of 112 µE L-1 s-1 were applied. B) Effect
of photosensitizer concentration on product yield. Different concentrations of CE
used in combination with MES as electron donor in the light-driven whole-cell
hydroxylation reaction employing NDO H295A. Reaction conditions: 0-320 µM
photosensitizer, 10 mM 3, 50 mM MES, 100 gWCW/L whole cells (E. coli JM109
(DE3)_pDTG141_NDO H295A, 19h expression), 50 mM MES, white light (max.
112 µE L-1 s-1), 30°C, 140 rpm.
Indeed, the obtained product formations with the empty vector
controls were much lower under dark and light conditions. We
attribute the turnover in the dark to the production of
carbohydrates, which under “famine conditions” were consumed
to regenerate NAD(P)H.[26,27,44,45] However, under light conditions
a photochemical background reaction has been observed, which
varies depending on the applied light intensity between 0.15 to
3.5 mM and depending on the applied photosensitizer/electron
donor combination (Figure 2A, Figure S19). The background
reaction contributed up to 12% of total product formation when
using CE/MOPS and 35% when using RB/MOPS (Figure 2A),
indicating that the photochemical background reaction depends
on the photosensitizer.
To determine the incident photon flux, chemical actinometry was
performed using the well-described ferrioxalate actinometer
(Table 2 and Table S10).[44] Although the cell suspension showed
strong scattering and optical absorption by other cell or solution
components, we were able to estimate quantum yields (QY, Table
S10). Additionally, apparent quantum yields (AQY) were
calculated as the ratio of two times the observed product
formation rate to the incident photon flux, as two photons are
required per turnover (Table 2). Admittedly, the given AQYs are
lower than the typical values achieved in photochemical reactions,
however, these values lay a promising foundation for further
optimization of this artificial photosynthetic systems.
This article is protected by copyright. All rights reserved.

10.1002/anie.201914519
Angewandte Chemie International Edition
COMMUNICATION
Table 2. The combination of photosensitizer and electron donor is a crucial factor for the efficiency of the light-driven reaction.
OH
whole-cells containing RO
OH
hυ
photosensitizer, electron donor,
30°C, buffer pH 7.2, O2
OH
3
3a
3b
Apparent
quantum yield
Enzyme
Photosensitizer,
electron donor
Products
Whole-cell activity
Maximum
Diastereomeric ratio
Distribution
3a : 3b
[%][c]
Specific activity[d]
Productivity
concentration
cis : trans-3b
[%]
[mU/gWCW
]
[mM/h]
[mM][a]
[%][b]
CDO M232A
0.18
0.13
0.25
1.59
0.94
1.65
0.26
0.61
0.75
0.23
0.37
0.52
0.32
0.89
0.45
0.30
0.86
0.64
0.09
0.06
0.12
0.78
0.46
0.81
0.13
0.30
0.37
0.11
0.18
0.26
0.16
0.44
0.22
0.15
0.42
0.32
EY/EDTA
EY / MOPS
EY / MES
100 : 0
100 : 0
100 : 0
96 : 4
95 : 5
95 : 5
94 : 6
88 : 12
90 : 10
96 : 4
83 : 17
86 : 14
97 : 3
95 : 5
97 : 3
94 : 6
95 : 5
86 : 14
3 : 97
0 : 100
4 : 96
4 : 96
4 : 96
3 : 97
2 : 98
0 : 100
0 : 100
4 : 96
2 : 98
4 : 96
0 : 100
2 : 98
3 : 97
6 : 94
18 : 82
0 : 100
29
21
3.7 ± 0.4
6.8 ± 0.03
4.7 ± 0.2
6.8 ± 0.3
7.3 ± 0.6
5.8 ± 0.4
4.0 ± 0.01
8.6 ± 0.6
8.3 ± 0.08
4.7 ± 0.2
7.7 ± 0.4
7.4 ± 0.3
7.5 ± 1.0
7.5 ± 1.2
7.9 ± 0.6
5.5 ± 0.5
7.3 ± 0.4
8.5 ± 0.4
41
RB / EDTA
RB / MOPS
RB / MES
CE / EDTA
CE / MOPS
CE / MES
EY /EDTA
EY / MOPS
EY / MES
265
156
275
43
102
124
47
NDO H295A
61
87
RB / EDTA
RB / MOPS
RB / MES
CE / EDTA
CE / MOPS
CE / MES
53
148
79
50
143
107
Reaction conditions: [3] = 10 mM; [whole cells] = 100 gWCW/L (E. coli JM109 (DE3)_pDTG141_NDO H295A or E. coli JM109_pCDOv1_CDO M232A); sodium
phosphate buffer (pH 7.2, 50 mM) when using 25 mM EDTA, otherwise 50 mM MES/MOPS buffer; white light illumination (max. 112 µE L-1 s-1). [a] Sum of 3a and
3b; time points for determination were chosen at max. product concentration during the time course of the reaction; [b] The diastereomeric ratio has been determined
after 4-6 hours of reaction. [c] Determined after 24 hours. [d] Determined from the liner range of product formation determined from the kinetic profiles for each reaction
(Figures S19-S24).
This capability seems to be limited by the applied light intensity,
since the overall background reaction remained low in all cases
(< 0.5 mM) when lower light intensities were used, thus confirming
that the reaction is truly light-driven.
Additionally, we investigated the influence of the photosensitizer
(Figure 2B) and electron donor concentration, as well as the cell
densities (Figure S18) on the efficiency of the photocatalytic
activation of NDO H295A. Under light illumination, the RO
donor concentration (CE constant), it can be seen that the product
formation is influenced by the concentration of the electron donor
when the cell density is higher, because more photo-induced
electrons are transferred to the enzyme. However, the system
seems to be limited by the applied light intensities (max. 112 µE
L-1 s-1), i.e. that from a certain cell density on, the product
concentration is not “controlled” anymore by the concentration of
the electron donor, and at that point the light intensity in the
system becomes the limiting factor.
catalyzed hydroxylation of
3 is most efficient when CE
concentrations of 80 to 100 µM were used. Between 10 µM and
80 µM of CE, an increase in product formation is observed (Figure
2B). However, when using >100 µM of CE, no further increase
can be seen, i.e. that above 100 µM CE either the concentration
is not limited or the transport of the photosensitizer inside the cells
is hampered (Figure 2B). However, we observed photobleaching
over time. When additional 100 µM of CE were added to the light-
driven hydroxylation, the product formation accelerates again,
and after 6 hours after adding additional CE already 7.2 mM of
product has been formed in contrast to only 3 mM without adding
additional CE (Figure 3A).
Light intensity plays a crucial role on the efficiency of the light-
driven catalysis, and influences the extend of the photochemical
background reaction. When reducing the light intensity by 75%
(Figure S10B), the obtained product concentrations decreased by
only 20%. Finally, we followed the light-driven reaction in a time-
course experiment over 24 hours under optimized conditions
(Figure 3B). The product formation proceeds smoothly within 24
hours of reaction, however, in case of NDO H295A/MES/CE and
CDO M232A/MOPS/CE no significant product increase is
observed after 20 hours. Noteworthy, when using SO in
combination with MOPS or EDTA, obtained product
concentrations remained always lower compared to other
photosensitizer/electron donor combinations.
We further investigated the effect of increasing cell densities on
the efficiency of the light-driven hydroxylation reaction (Figure
S18). When increasing the cell density as well as the electron
This article is protected by copyright. All rights reserved.

10.1002/anie.201914519
Angewandte Chemie International Edition
COMMUNICATION
Experimental Section
All experimental details can be found in the Supporting Information.
A)
Acknowledgements
This project has received funding from the European Union’s Horizon 2020
research and innovation program under the Marie Skłodowska-Curie grant
agreement No. 764920. We thank Prof. Dr. Rebecca Parales for providing
us with the vector pDTG141 harboring the NDO genes and Prof. Dr.
Hideaki Nojiri for plasmid pIP107D harboring the CDO genes.
Keywords: biocatalysis • oxyfunctionalization • photocatalysis •
photoinduced electron transfer • Rieske dioxygenases
[1]
[2]
J. C. Lewis, P. S. Coelho, F. H. Arnold, Chem. Soc. Rev. 2011, 40,
2003–2021.
J. Dong, E. Fernández-Fueyo, F. Hollmann, C. E. Paul, M. Pesic, S.
Schmidt, Y. Wang, S. Younes, W. Zhang, Angew. Chem. Int. Ed.
2018, 57, 9238–9261; Angew. Chem. 2018, 130, 9380-9404.
D. J. Ferraro, L. Gakhar, S. Ramaswamy, Biochem. Biophys. Res.
Commun. 2005, 338, 175–190.
B)
[3]
[4]
[5]
[6]
S. M. Barry, G. L. Challis, ACS Catal. 2013, 3, 2362–2370.
W. A. Tan, R. E. Parales, Green Biocatal. 2016, 457–471.
W. K. Dawson, R. Jono, T. Terada, K. Shimizu, PLoS One 2016, 11,
e0162031.
[7]
R. E. Parales, S. M. Resnick, in Biodegrad. Bioremediation (Eds.: A.
Singh, O.P. Ward), Springer Berlin Heidelberg, Berlin, Heidelberg,
2004, 175–195.
[8]
R. E. Parales, S. M. Resnick, C. L. Yu, D. R. Boyd, N. D. Sharma, D.
T. Gibson, J. Bacteriol. 2000, 182, 5495–5504.
[9]
O. Kweon, S. J. Kim, S. Baek, J. C. Chae, M. D. Adjei, D. H. Baek, Y.
C. Kim, C. E. Cerniglia, BMC Biochem 2008, 9, 4268-4273.
F. F. Özgen, S. Schmidt, in Biocatalysis (Eds.: Q. Husain, M.F. Ullah),
Springer International Publishing, Cham, 2019, 57–82.
D. J. Ferraro, A. Okerlund, E. Brown, S. Ramaswamy, IUCrJ 2017, 4,
648–656.
[10]
[11]
[12]
[13]
[14]
Y. Ashikawa, Z. Fujimoto, H. Noguchi, H. Habe, T. Omori, H. Yamane,
H. Nojiri, Structure 2006, 14, 1779–1789.
Figure 3. A) Effect of photobleaching of CE on the time course of the light-
driven hydroxylation catalysed by NDO H295A. The red curve visualizes the
addition of again 100 µM CE after 4 h of biotransformation, whereas the black
curve shows the light-driven biotransformation without adding additional CE. B)
Kinetic profile obtained for the light-driven whole-cell hydroxylation reaction
employing CDO M232A and NDO H295A with SO, CE and EY in combination
with either EDTA, MOPS or MES as electron donors. Reaction conditions:
100 µM photosensitizer, 10 mM 3, 25 mM EDTA or 50 mM MOPS/MES, in A)
25-300 gWCW/L and in B) 100 gWCW/L whole cells (E. coli JM109
(DE3)_pDTG141_NDO H295A or E. coli JM109_pCDOv1_CDO M232A, 19h
expression), white light (max. 112 µE L-1 s-1), 30°C, 140 rpm, 24 hours.
C. Schmidt-Dannert, F. Lopez-Gallego, Microb. Biotechnol. 2016, 9,
601–609.
L. M. Blank, B. E. Ebert, K. Buehler, B. Bühler, Antioxid. Redox
Signaling 2010, 13, 349–394.
[15]
[16]
[17]
V. Uppada, S. Bhaduri, S. B. Noronha, Curr. Sci. 2014, 106, 946.
J. Kim, C. B. Park, Curr. Opin. Chem. Biol. 2019, 49,122–129
B. O. Burek, S. Bormann, F. Hollmann, J. Z. Bloh, D. Holtmann,
Green Chem. 2019, 21, 3232-3249.
[18]
[19]
W. Zhang, F. Hollmann, Chem. Commun. 2018, 54, 7281–7289.
S. H. Lee, D. S. Choi, S. K. Kuk, C. B. Park, Angew. Chem. Int. Ed.
2018, 57, 7958–7985.
To conclude, we have shown the photo-activation of two different
ROs in an E. coli-based whole-cell system by coupling light-
harvesting complexes to hydroxylation reactions in vivo. This was
successfully conducted by using several photosensitizers for the
bioconversion of three different substrates, hence representing
the first example of photo-induced RO systems. Particularly for
challenging multi-component oxygenases, this system offers the
advantage of relying on the well-studied genetic toolbox of E. coli
as host, thereby facilitating a broad applicability of light-driven
artificial photosynthesis. The obtained product formations of up to
1.3 g/L and rates of up to 1.6 mM/h demonstrate that competitive
productivities compared to cyanobacteria were achieved.[28]
The coupling of artificial light-harvesting complexes to enzymes
inside cells provides a versatile route to accessing diverse and
selective visible-light-driven chemical syntheses especially when
unstable or multi-component enzymes are used.
[20]
[21]
T. Gulder, C. J. Seel, ChemBioChem 2019, 20, 1871.
L. Schmermund, V. Jurkaš, F. F. Özgen, G. D. Barone, H. C.
Büchsenschütz, C. K. Winkler, S. Schmidt, R. Kourist, W. Kroutil,
ACS Catal. 2019, 4115–4144.
[22]
[23]
D. Adam, L. Bösche, L. Castañeda-Losada, M. Winkler, U. P. Apfel,
T. Happe, ChemSusChem 2017, 10, 894–902.
L. C. P. Gonçalves, H. R. Mansouri Khosravi, S. PourMehdi, M.
Abdellah, B. S. Fadiga, E. Bastos, J. Sa, M. Mihovilovic, F. Rudroff,
Catal. Sci. Technol. 2019, 9, 2682-2688.
[24]
[25]
[26]
[27]
A. Taglieber, F. Schulz, F. Hollman, M. Rusek, M. T. Reetz,
ChemBioChem 2008, 9, 565–572.
L. Assil-Companioni, S. Schmidt, P. Heidinger, H. Schwab, R. Kourist,
ChemSusChem 2019, 12, 2361-2365.
S. Böhmer, K. Köninger, Á. Gómez-Baraibar, S. Bojarra, C. Mügge,
S. Schmidt, M. M. M. Nowaczyk, R. Kourist, Catalysts 2017, 7, 240.
K. Köninger, Á. Gómez Baraibar, C. Mügge, C. E. Paul, F. Hollmann,
M. M. Nowaczyk, R. Kourist, Angew. Chem. Int. Ed. 2016, 55, 5582–
5585; Angew. Chem. 2016, 128, 5672.
[28]
A. Hoschek, B. Bühler, A. Schmid, Angew. Chem. Int. Ed. 2017, 56,
This article is protected by copyright. All rights reserved.

10.1002/anie.201914519
Angewandte Chemie International Edition
COMMUNICATION
15146–15149; Angew. Chem. 2017, 129, 15343-15346.
[29]
A. Hoschek, A. Schmid, B. Bühler, ChemCatChem 2018, 10, 5366–
5371.
[30]
[31]
H. Qiang, A. Richmond, Y. Zarmi, Eur. J. Phycol. 1998, 33, 165–171.
Y. Honda, H. Hagiwara, S. Ida, T. Ishihara, Angew. Chem. Int. Ed.
2016, 55, 8045–8048; Angew. Chem. 2016, 128, 8177-8180.
J. H. Park, S. H. Lee, G. S. Cha, D. S. Choi, D. H. Nam, J. H. Lee, J.
K. Lee, C. H. Yun, K. J. Jeong, C. B. Park, Angew. Chem. Int. Ed.
2015, 54, 969–973; Angew. Chem. 2014, 127, 983-987.
S. F. Rowe, G. Le Gall, E. V. Ainsworth, J. A. Davies, C. W. J.
Lockwood, L. Shi, A. Elliston, I. N. Roberts, K. W. Waldron, D. J.
Richardson, et al., ACS Catal. 2017, 7, 7558–7566.
C. Gally, B. M. Nestl, B. Hauer, Angew. Chem. Int. Ed. 2015, 54,
12952–12956; Angew. Chem. 2015, 127, 13144-13148.
J. M. Halder, B. M. Nestl, B. Hauer, ChemCatChem 2018, 10, 178–
182.
[32]
[33]
[34]
[35]
[36]
[37]
[38]
R. E. Parales, K. Lee, S. M. Resnick, H. Jiang, D. J. Lessner, D. T.
Gibson, J. Bacteriol. 2000, 182, 1641–1649.
C. J. Seel, A. Králík, M. Hacker, A. Frank, B. König, T. Gulder,
ChemCatChem 2018, 10, 3960–3963.
W. J. Ferguson, K. I. Braunschweiger W R Braunschweiger, J. R.
Smith, J. Justin Mccormick, C. C. Wasmann, N. P. Jarvi, D. H. Bell,
N. E. Good, Anal. Biochem. 1980, 104, 300–310.
[39]
[40]
[41]
C. M. H. Ferreira, I. S. S. Pinto, E. V. Soares, H. M. V. M. Soares,
RSC Adv. 2015, 5, 30989–31003.
D. Adam, L. Bosche, L. Castaneda-Losada, M. Winkler, P. Apfel, T.
Happe, ChemSusChem 2017, 10, 894–902.
S. Langer, M. Hashimoto, B. Hobl, T. Mathes, M. Mack, J. Bacteriol.
2013, 195, 4037–4045.
[42]
[43]
Y. Suzuki, N. Koyama, Biodegradation 2009, 20, 39–44.
C. G. Hatchard, C. A. Parker, Proc. R. Soc. London. Ser. A. Math.
Phys. Sci. 1956, 235, 1203, 518-536.
[44]
[45]
R. Yamanaka, K. Nakamura, M. Murakami, A. Murakami,
Tetrahedron Lett. 2015, 56, 1089–1091.
K. Nakamura, R. Yamanaka, Tetrahedron: Asymmetry 2002, 13,
2529–2533.
This article is protected by copyright. All rights reserved.

10.1002/anie.201914519
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
Layout 1:
COMMUNICATION
Text for Table of Contents
Author(s), Corresponding Author(s)*
Page No. – Page No.
Title
((Insert TOC Graphic here))
Layout 2:
COMMUNICATION
F. Feyza Özgen, Michael E. Runda,
Peter Wied, Bastien O. Burek, Jonathan
Z. Bloh, Robert Kourist and Sandy
Schmidt*
Page No. – Page No.
Artificial light-harvesting complexes
enable Rieske oxygenase-catalyzed
hydroxylations in non-photosynthetic
cells
Illuminate me! The photo-activation of Rieske dioxygenases in the absence of
glucose or any cofactor was successfully conducted using several photosensitizers
for the bioconversion of three different substrates and hence represents the first
example of a photo-induced Rieske system.
This article is protected by copyright. All rights reserved.