Cascading g-C3N4 and Peroxygenases for Selective Oxyfunctionalization Reactions

Article abstract of DOI:10.1021/acscatal.9b01341

Peroxygenases are very interesting catalysts for specific oxyfunctionalization chemistry. Instead of relying on complicated electron transport chains, they rely on simple hydrogen peroxide as the stoichiometric oxidant. Their poor robustness against H₂O₂ can be addressed via in situ generation of H₂O₂. Here we report that simple graphitic carbon nitride (g-C₃N₄) is a promising photocatalyst to drive peroxygenase-catalyzed hydroxylation reactions. The system has been characterized by outlining not only its scope but also its current limitations. In particular, spatial separation of the photocatalyst from the enzyme is shown as a solution to circumvent the undesired inactivation of the biocatalyst. Overall, very promising turnover numbers of the biocatalyst of more than 60.000 have been achieved.

Full text of DOI:10.1021/acscatal.9b01341

Subscriber access provided by Nottingham Trent University
Article
3
4
Cascading g-CN and Peroxygenases for
Selective Oxyfunctionalization Reactions
Morten van Schie, Wuyuan Zhang, Florian Tieves, Da Som Choi, Chan Beum Park, Bastien Oliver
Burek, Jonathan Bloh, Isabel W.C.E. Arends, Caroline Emilie Paul, Miguel Alcalde, and Frank Hollmann
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.9b01341 • Publication Date (Web): 09 Jul 2019
Downloaded from pubs.acs.org on July 10, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
Cascading g-C N and Peroxygenases for Selective
3
4
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Oxyfunctionalization Reactions
1
1
1
2
Morten M.C.H. van Schie,* Wuyuan Zhang, Florian Tieves, Da Som Choi, Chan
2
3
3
4
Beum Park Bastien O. Burek, Jonathan Bloh, Isabel W.C.E. Arends, Caroline E.
1
5
1
Paul, Miguel Alcalde, Frank Hollmann*
1 Department of Biotechnology, Delft University of Technology, van der Maasweg 9,
2629HZ Delft, The Netherlands
2 Department of Materials Science and Engineering, Korea Advanced Institute of
Science and Technology (KAIST), 335 Science Road, Daejeon 305-701, Republic of
Korea
3
DECHEMA Forschungsinstitut, Theodor-Heuss-Allee 25, 60486 Frankfurt am Main,
Germany
ACS Paragon Plus Environment
1

ACS Catalysis
Page 2 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
University of Utrecht, Faculty of Science, Budapestlaan 6, 3584 CD Utrecht, The
Netherlands
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
Department of Biocatalysis, Institute of Catalysis, CSIC, 28049 Madrid, Spain.
* e-mail: m.m.c.h.vanschie@tudelft.nl, f.hollmann@tudelft.nl
ACS Paragon Plus Environment
2

Page 3 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
ABSTRACT
Peroxygenases are very interesting catalysts for specific oxyfunctionalization chemistry.
Instead of relying on complicated electron transport chains, they rely on simple hydrogen
peroxide as stoichiometric oxidant. Their poor robustness against H O can be addressed
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
via in situ generation of H O . Here we report that simple graphitic carbon nitride (g-C N )
2
2
3
4
is a promising photocatalyst to drive peroxygenase-catalyzed hydroxylation reactions.
The system has been characterized outlining its scope but also its current limitations. In
particular, spatial separation of the photocatalyst from the enzyme is shown as solution
to circumvent the undesired inactivation of the biocatalyst. Overall, very promising
turnover numbers of the biocatalyst of more than 60.000 have been achieved.
KEYWORDS
Cascade reactions, Enzyme catalysis, Oxidation, Oxyfunctionalization, Photocatalysis
ACS Paragon Plus Environment
3

ACS Catalysis
Page 4 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
Selective oxyfunctionalization of non-activated sp carbon-hydrogen bonds is a
challenge in organic synthesis. Catalysts exhibiting both high oxidation potential and high
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
selectivity are rarely found. In this respect, so-called unspecific peroxygenases (UPOs),
next to the established P450 monooxygenases and some other non-heme-
2
3-5
monooxygenases, have attracted considerable interest. Monooxygenases generate
their catalytically active species through reductive activation of molecular oxygen. For
this, monooxygenases rely on complex electron transport chains delivering the reducing
equivalents from NAD(P)H to the monooxygenases’ active sites. This complicates their
practical use. Furthermore, quite frequently, a significant portion of the reducing
equivalents is uncoupled into a futile reaction with molecular oxygen thereby wasting
valuable reducing equivalents and generating hazardous reactive oxygen species.6
UPOs do not rely on the aforementioned electron transport chains by rather use simple
H O as oxidant. This makes UPOs very attractive from a preparative point-of-view. H O ,
2
2
2
2
however, is also a potent inhibitor of peroxygenases as already small excesses
7
oxidatively inactivate the prosthetic heme group. Generating H O in situ through
2
2
ACS Paragon Plus Environment
4

Page 5 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
8
catalytic reduction of O is the most common approach to alleviate the inactivation issue.
2
In essence, these systems provide H O at rates that enable efficient peroxygenase
2
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
activity while minimizing the H O -related inactivation. These methods comprise a range
2
2
of chemical,9-15 electrochemical, 16-21 enzymatic22-26 and photocatalytic27-34 approaches.
The latter is particularly interesting as it, in principle, enables using light energy to
access simple sacrificial electron donors to drive the reduction of O to H O . To make
2
2
2
light energy available for this reaction, a photosensitizer (or photocatalyst) is necessary.
Both, homogeneously dissolved molecular and semiconductor-based solid material
photocatalysts have been evaluated. While the first often suffer from issues of
photobleaching and inactivation, the latter excel by their high robustness and reusability.
So far, mainly TiO -based semiconductor photocatalysts have been evaluated to promote
2
peroxygenase-catalyzed oxyfunctionalization reactions.
Therefore, we set out to investigate a broader set of (in)organic photocatalyst systems35
to promote peroxygenase-catalyzed reactions (Scheme 1).
ACS Paragon Plus Environment
5

ACS Catalysis
Page 6 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Photoenzymatic hydroxylation of ethyl benzene combining heterogeneous photocatalysts for
the reductive activation of O
2
to H
2
2
O with a peroxygenase-catalyzed oxyfunctionalization reaction.
We chose the selective hydroxylation of ethyl benzene to (R)-1-phenyl ethanol catalyzed
by the peroxygenase from Agrocybe aegerita (rAaeUPO) as the model reaction.36-39
In a first set of experiments, we evaluated several reported and/or commercially available
heterogeneous photocatalysts (Figure 1) with respect to their ability to form H O ;
2
2
particularly, we investigated their performance in the envisioned photoenzymatic cascade
transforming ethyl benzene to (R)-1-phenyl ethanol (Figure 1). It is worth mentioning that
in the absence of the photocatalysts or light, no product formation was observed. In the
ACS Paragon Plus Environment
6

Page 7 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
absence of rAaeUPO, upon prolonged reaction times, traces of racemic product and the
overoxidation product were observed in some cases.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Performance of several heterogeneous photocatalysts to promote rAaeUPO-catalyzed
oxyfunctionalization, forming phenyl ethanol (blue) and the over-oxidation product acetophenone (red), in
-1
absence (left) or presence (right) of methanol. Conditions: 5 mg ml heterogeneous catalyst, 50 mM ethyl
benzene, 0 or 250 mM methanol and 100 nM rAaeUPO in a 100 mM phosphate buffer at pH 7, 30 °C and
stirring at 300 rpm. Illumination by a Osram 200W light bulb for 30 minutes. Reactions were performed in
;40 2: Co
(quantum dots);41 3: Co
;42 4: Pt-TiO
;47 10: g-C
independent duplicates. 1: Au-BiVO
4
3
O
4
4
(H
2
O)
2
(W
9
O
34
)
2
2
(
Rutile);43 5: MnO (on Faujasite);44 6: Co-TiO
2
;45 7: MnO (nanowires); 8: Ir@SiO
46
2
; 9: Fe
2
O
3
3
N
4
;
11: ZnO (nanoclusters).48
ACS Paragon Plus Environment
7

ACS Catalysis
Page 8 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Almost all heterogeneous photocatalysts tested enabled the desired reaction both in the
presence and absence of an extra electron donor (methanol). However, reaction rates
tended to be significantly lower in experiments where methanol was absent. In the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
49
absence of methanol, it may be assumed that water was the primary reductant. As water
oxidation is thermodynamically more challenging than methanol oxidation, this explains
the generally lower product formation rates in the absence of methanol. In some cases,
significant amounts of acetophenone were found along with the desired (R)-1-phenyl
ethanol. We attribute this to the oxidation of the desired product by the photocatalysts.50
Platinum loaded titanium oxide (Pt-TiO ) and graphitic carbon nitride (g-C N ) stood out
2
3
4
in terms of product formation rate. For reasons of ease of preparation, we further focussed
on g-C N as photocatalyst. Furthermore, g-C N has previously been reported to highly
3
4
3
4
51
selectively for H O over other (partially reduced) reactive oxygen species.
2
2
Figure 2 shows the overall product formation of g-C N using either water (blue), methanol
3
4
(red), or formate (green) as sacrificial electron donor to promote the photoenzymatic
ACS Paragon Plus Environment
8

Page 9 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
hydroxylation of ethyl benzene. Assumptions made to obtain the bar-graphs can be found
in the experimental section.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. g-C
3
4
N as photocatalyst to promote rAaeUPO-catalyzed hydroxylation of ethyl benzene in the
absence of external electron donors (blue), or 250 mM methanol (red) or 250 mM formate (green). A: time
course of (R)-1-phenyl ethanol formation and B: time course of acetophenone formation. From these time
courses, parameters such as reaction rate (C), rAaeUPO inactivation rate (D), maximal product
concentration (E) and selectivity (F) were calculated. General conditions: [rAaeUPO] = 100 nM, [ethyl
-1
benzene] = 50 mM, [g-C
3
N ] = 5 mg ml , KPi buffer pH 7.0 (100 mM), 30 °C, magnetic stirring at 600 rpm,
4
Illumination by a Osram 200W light bulb. Reactions were performed in independent duplicates.
ACS Paragon Plus Environment
9

ACS Catalysis
Page 10 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
From a conceptual and an environmental point-of-view, water would have been the most
desirable source of reducing equivalents. However, this system fell back in terms of
catalytic rate and selectivity compared with the use of methanol or formate as sacrificial
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
electron donor. In comparative experiments illuminating g-C N in phosphate buffer gave
3
4
-1
negligible H O formation rates (<1.0 µM h ), which more than doubled to 2.0 (± 0.02)
2
2
-1
µM min upon addition of formate. Similar observations have been made before for
TiO2.52 Since g-C N is not an efficient water oxidation catalyst, other molecules are
52
3
4
more likely to serve as reductants in the g-C N –catalyzed reduction of O . Both, ethyl
3
4
2
50
benzene and phenyl ethanol can be oxidized by g-C N . This was evident from the lower
3
4
optical purity of phenyl ethanol and the increased overoxidation to acetophenone.
Another interesting effect of formate was that the robustness of the overall reaction was
significantly higher than when using methanol or no sacrificial electron donor. We attribute
this to the hydroxyl radical scavenging activity of formate (vide infra). Methanol oxidation
ACS Paragon Plus Environment
10

Page 11 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
proceeds via methoxy radicals, which again may be assumed to have a detrimental
influence on the stability of the biocatalyst.29
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Overall, formate proved to be a suitable sacrificial electron donor for the reaction. We
therefore systematically investigated the factors influencing the activity, selectivity and
robustness of the formate-driven photoenzymatic reaction system.
Increasing the concentration of formate had a positive effect on the reaction rate,
selectivity and robustness of the overall system (Figure 3). Hence, we concluded that the
photocatalytic oxidation of formate is the overall rate-limiting step in the reaction scheme.
ACS Paragon Plus Environment
11

ACS Catalysis
Page 12 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Influence of the formate concentration on the performance of the photoenzymatic hydroxylation
of ethyl benzene. A: time course of (R)-1-phenyl ethanol formation and B: time course of acetophenone
formation. From these time courses, parameters such as reaction rate (C), rAaeUPO inactivation rate (D),
maximal product concentration (E) and selectivity (F) were calculated. [HCO
-] = 0 mM (black), 50 mM
2
(green), 100 mM (red), 250 mM (blue) or 500 mM (purple). General conditions: [rAaeUPO] = 100 nM, [ethyl
-1
benzene] = 50 mM, [g-C
3
N
4
] = 5 mg ml , KPi buffer pH 7.0 (100 mM), 30 °C, magnetic stirring at 600 rpm,
Illumination by a Osram 200W light bulb. Reactions were performed in independent duplicates.
ACS Paragon Plus Environment
12

Page 13 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
The concentration of the biocatalyst also had a significant influence on the robustness of
the overall reaction but hardly influenced the initial product formation rate (Figure 4). This
supports our assumption that the photocatalytic H O formation was rate-limiting and that
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
undesired rAaeUPO-inactivation by the photocatalyst represented the major undesired
side reaction. Incubation of g-C N with rAaeUPO in the darkness did not result in a
3
4
significant inactivation of the biocatalyst (Figure S10) suggesting that photocatalytically
generated reactive oxygen species are responsible for the enzyme inactivation observed
(vide infra).
ACS Paragon Plus Environment
13

ACS Catalysis
Page 14 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Influence of the rAaeUPO concentration on the performance of the photoenzymatic hydroxylation
of ethyl benzene. A: time course of (R)-1-phenyl ethanol formation and B: time course of acetophenone
formation. From these time courses, parameters such as reaction rate (C), rAaeUPO inactivation rate (D),
maximal product concentration (E) and selectivity (F) were calculated. [rAaeUPO] = 20 nM (black), 50 nM
(
red), 100 nM (blue), 200 nM (green) or 500 nM (purple). General conditions: [NaHCO ] = 250 mM, [ethyl
2
-1
benzene] = 50 mM, [g-C
3
N
4
] = 5 mg ml , KPi buffer pH 7.0 (100 mM), 30 °C, magnetic stirring at 600 rpm,
Illumination by a Osram 200W light bulb. Reactions were performed in independent duplicates.
ACS Paragon Plus Environment
14

Page 15 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
Next, we systematically varied the concentration of the photocatalyst (Figure 5). Very
much to our surprise, the anticipated positive correlation between g-C N concentration
3
4
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
and H O formation rate was less pronounced than expected. Moreover, rAaeUPO
2
2
inactivation even decreased at higher g-C N concentrations.
3
4
ACS Paragon Plus Environment
15

ACS Catalysis
Page 16 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 5. Influence of the g-C
3
N
4
concentration on the performance of the photoenzymatic hydroxylation of
ethyl benzene. A: time course of (R)-1-phenyl ethanol formation and B: time course of acetophenone
formation. From these time courses, parameters such as reaction rate (C), rAaeUPO inactivation rate (D),
-1
maximal product concentration (E) and selectivity (F) were calculated. [g-C
3
N
4
] = 1 mg ml (black), 2.5 mg
-1
-1
-1
-1
ml (red), 5 mg ml (blue), 10 mg ml (green) or 15 mg ml (purple). General conditions: [NaHCO ] = 250
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
mM, [ethyl benzene] = 50 mM, [rAaeUPO] = 100 nM, KPi buffer pH 7.0 (100 mM), 30 °C, magnetic stirring
at 600 rpm, Illumination by a Osram 200W light bulb. Reactions were performed in independent duplicates.
The most likely explanation for this observation is a self-shading effect of g-C N (already
3
4
-1
at a photocatalyst concentration of 1 mg ml the reaction mixture was turbid). Hence, due
to a decreased light penetration at elevated photocatalyst concentrations, the majority of
the photocatalyst remains inactive neither contributing to H O generation nor to
2
2
inactivating the biocatalyst.
The overall rate of the reaction was highest at neutral to slightly alkaline pH values (Figure
S6). Currently, we are lacking a plausible explanation for this observation as the H O
2
2
53
generation rate of g-C N was reported to be largely pH-independent whereas the
3
4
highest activity of rAaeUPO may be expected in slightly acidic media.39 Similar trends
have been observed before for TiO254 suggesting that the pH-dependency of the complex
54
O -reduction mechanism may play an important role here.
2
ACS Paragon Plus Environment
16

Page 17 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
Also the morphology of the g-C N catalyst had a significant influence on the overall
3
4
reaction rate as well as the rAaeUPO inactivation rate. The form of g-C N used so far
3
4
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
55
are the so-called g-C N sheets, obtained from calcination of urea. Further thermal
3
4
56
treatment process lead to so-called amorphous g-C N , exhibiting a higher surface area.
3
4
A preparation with lower specific surface area (bulk g-C N ) was obtained by starting the
3
4
synthesis from melamine instead of urea (Figure S1). In line with our previous
observations that the photocatalytic H O generation is overall rate-limiting, we also
2
2
observed a correlation between surface area and overall reaction rate (Figure 6). The
same, however was also true for the inactivation rate of rAaeUPO, which increased with
increasing surface area.
ACS Paragon Plus Environment
17

ACS Catalysis
Page 18 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Influence of g-C
3
N
4
morphology on the photoenzymatic hydroxylation of ethyl benzene. A: time
course of (R)-1-phenyl ethanol formation and B: time course of acetophenone formation. From these time
courses, parameters such as reaction rate (C), rAaeUPO inactivation rate (D), maximal product
concentration (E) and selectivity (F) were calculated. Amorphous g-C
3
N
4
(black), g-C
3
N
4
sheets (red) or g-
] = 5 mg
C
3
N
4
bulk (green).. Reaction conditions: [NaHCO ] = 250 mM, [ethyl benzene] = 50 mM, [g-C
2
3
N
4
-1
ml , [rAaeUPO] = 100 nM, KPi buffer pH 7.0 (100 mM), 30 °C, magnetic stirring at 600 rpm, Illumination by
a Osram 200W light bulb. Reactions were performed in independent duplicates.
To evaluate the synthetic potential of the proposed photoenzymatic oxyfunctionalization
system we performed the reaction at semi-preparative scale (Figure S7). To compensate
ACS Paragon Plus Environment
18

Page 19 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
for the high volatility of the starting material and the inactivation of the biocatalyst, both
were replenished at intervals (see SI for details of the experimental procedure). In this
experiment, 16.9 mM of essentially pure (97.8% ee) (R)-1-phenyl ethanol accumulated
in the reaction mixture. From this, 42 mg of the desired product were isolated. Admittedly,
this is not yet a practical protocol for the synthesis of (R)-1-phenyl ethanol. Especially the
high volatility of the reagents caused significant losses in the mass balance, which will
have to be addressed in future, via improved experimental setups (circumventing the
evaporation issue). Nevertheless, a respectable turnover number of more than 21000 for
the biocatalyst was obtained in this experiment.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The major bottleneck of the current system is its poor robustness; generally within 24h
accumulation of (R)-1-phenyl ethanol ceased, which we attribute to the inactivation of the
biocatalyst. As mentioned above, hydroxyl radicals may be assumed to be the primary
products of g-C N -catalyzed water photo-oxidations. Indeed, using the spin trap
3
4
57
method we could confirm the occurrence of hydroxyl radicals (Figure 7). The short life
time of ·OH in aqueous media58 suggests their predominant occurrence at the
ACS Paragon Plus Environment
19

ACS Catalysis
Page 20 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
photocatalyst surface. We therefore also investigated whether g-C N showed a tendency
3
4
to absorb the biocatalyst. Indeed we found that rAaeUPO and other proteins absorbed
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
59
significantly to the polar surface of g-C N (Figure 7). Here they are exposed to locally
3
4
high concentrations of hydroxyl radicals, which sufficiently explains the rather poor
robustness of the photoenzymatic reactions so far. As mentioned above, the absorption
per se did not result in inactivation of rAaeUPO (Figure S10).
Figure 7. Investigating the molecular reasons for the decreased rAaeUPO-stability under process
conditions. A: Detection of hydroxyl radicals formed by irradiated g-C
3
4
N using the spin-trap method.
Signals marked with an asterisk () are assigned to the oxidation product of DMPO, 5,5-dimethyl-2-
oxopyrroline-1-oxyl (DMPOX). Signals marked with triangles () belong to the spin adduct DMPO–OH; B:
Protein in solution before (blue) or after (red) incubation with g-C
3
4
N in dark, for bovine serum albumin
(BSA) or rAaeUPO.
ACS Paragon Plus Environment
20

Page 21 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
We hypothesized that spatial separation of rAaeUPO from the photocatalyst may
enhance the stability of the enzyme under reaction conditions. Therefore, we tested this
hypothesis by placing rAaeUPO into a dialysis bag thereby preventing its direct contact
with the photocatalyst (Figure 8). Here, the UPO and g-C N are contained, while both
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
4
ethyl benzene as H O could pass the membrane.
2
2
ACS Paragon Plus Environment
21

ACS Catalysis
Page 22 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 8. Time course of the photoenzymatic hydroxylation of ethyl benzene to (R)-1-phenyl ethanol (black)
and overoxidation to acetophenone (red) using the dialysis bag approach. Conditions: 10 ml reaction
solution equally divided inside and outside the dialysis bag (20 kDa cutoff). Inside the bag: [NaHCO
2
] = 250
mM, [ethyl benzene] = 50 mM, [rAaeUPO] = 100 nM, KPi buffer pH 7.0 (100 mM). Outside the bag:
-1
[
NaHCO
2
] = 250 mM, [ethyl benzene] = 50 mM, [g-C
3
N ] = 5 mg ml , KPi buffer pH 7.0 (100 mM). The
4
reaction was performed once at room temperature while stirring at 600 rpm. The reaction solution was
illuminated by a lightningcure spot light (Hamamatsu) at 50% intensity with an UV filter.
To our delight, physically separating the biocatalyst from the photocatalyst had the
desired effect of stable product accumulation for more than four days. Compared to the
previous experiments, this corresponds to an improvement by more than 4 times.
ACS Paragon Plus Environment
22

Page 23 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
However, this improvement came at the expense of a significantly decreased reaction
rate, which is most likely to be attributed to diffusion limitations for the substrates over the
dialysis membrane (Figure S4).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Overall, in this contribution we have expanded the scope of photoenzymatic
oxyfunctionalization reactions to a broader range of heterogeneous photocatalysts. g-
C N appeared to be a good alternative to the established TiO -based photocatalysts.
3
4
2
Total turnover numbers of over 60.000 have been achieved for the rAaeUPO. The major
limitations identified in this study were (1) the relatively low specific H O -generation rate
2
2
of the photocatalysts and (2) the undesired inactivation of the biocatalysts at the
photocatalyst surface. The first limitation can be addressed by further optimizing the
catalyst and light intensity. The catalyst can be improved by chemically modifying the g-
60
C N e.g. by doping with donor- or acceptor type dopants may be successful. Some
3
4
61
62
preliminary results using KOH- or Co O -modified g-C N indeed demonstrated that
2
3
3
4
doping can significantly influence the H O -generation activity. Further systematic studies
2
2
will validate this approach.
ACS Paragon Plus Environment
23

ACS Catalysis
Page 24 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The second limitation, i.e. the oxidative inactivation of the biocatalyst by surface-borne
reactive oxygen species can be alleviated by physical separation. To circumvent the
massive diffusion limitations observed in this double-heterogeneous reaction system, a
linear plug-flow reactor concept may prove beneficial.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Experimental section
Catalysts preparation
rAaeUPO was produced by heterologous expression in Pichia pastoris performed
following a previously reported procedure.38
g-C N was synthesized either from urea or melamine as starting compound by heating
3
4
-1
it up in a furnace to 550 °C (heat ramp: 5 °C min ). In a typical procedure from 10 g urea
approximately 0.5 g g-C N sheets were obtained whereas from 10 g melamine
3
4
approximately 2 g of g-C N bulk were obtained. Amorphous g-C N was synthesized by
3
4
3
4
63
further calcination of g-C N sheets to 620 °C under an inert argon atmosphere.
3
4
All other photocatalysts were either commercially available of prepared following literature
procedures.
ACS Paragon Plus Environment
24

Page 25 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
Reaction setup. Unless stated differently, reactions were performed in 4 ml glass vials. 2
ml reaction mixtures were stirred at 600 rpm using a small stirring magnet (6 mm). The
vials were placed around an incandescent white light bulb (Osram, 205W Halolux Ceram)
at a distance of approximately 1 cm. The water bath was continuously cooled at 30 °C.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-1
Reaction mixtures generally contained 5 mg ml g-C N , 100 nM rAaeUPO and 50 mM
3
4
ethylbenzene in a 100 mM KPi buffer at pH 7.0. Sacrificial electron donors were added to
a concentration of 250 mM. Before use, the g-C N was dispersed via sonication in 1 ml
3
4
of the phosphate buffer. Samples were taken using a syringe and needle, keeping the
reactors closed and preventing evaporation of the ethylbenzene. 200 µl of the reaction
mixture was taken and extracted with an aliquot of ethyl acetate containing 5 mM octanol
as internal standard. The mixtures were intensively mixed for 10 s, centrifuged for 2 min
and the organic phase was dried over magnesium sulphate and subsequently analysed
via achiral CG chromatography (CP-WAX 52 CB). Optical purities were determined using
chiral GC (CP-Chirasil-Dex CB).
Scale-up reaction. A 24 ml reaction, divided over 2 ml samples, was illuminated 80 hours
for increased product titres. The reaction was performed at 30 °C in a KPi buffer (100
-1
mM, pH 7.0) containing ethyl benzene (50 mM), g-C N (5 mg ml ), rAaeUPO (100 nM)
3
4
and sodium formate (250 mM). 100 nM of rAaeUPO was added to the reaction every 12
ACS Paragon Plus Environment
25

ACS Catalysis
Page 26 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
hours and the ethyl benzene was replenished by adding 50 mM extra substrate after 36
hours. At the end of the reaction, the compound was extracted with ethyl acetate, dried
with MgSO and purified under reduced pressure at room temperature, also evaporating
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
remaining substrate. The purity of the product was determined by H NMR, while optical
purity was determined using chiral GC (CP-Chirasil-Dex CB). NMR spectra were recorded
on an Agilent 400 spectrometer in CDCl . Chemical shifts are given in ppm with respect
3
to tetramethylsilane. Coupling constants are reported as J-values in Hz (s: singlet. d:
doublet. t: triplet. q: quartet. m: multiplet. br: broad)
1
H NMR (400 MHz, CDCl ) δ 7.40-7.26 (m, 5H), 4.91 (q, J = 6.5 Hz, 1H), 1.82 (br s, 1H,
3
-OH), 1.51 (d, J = 6.4 Hz, 3H)
13
C NMR (100 MHz, CDCl ) δ 128.5, 127.5, 125.4, 70.4, 25.10.
3
Data manipulation
As the reaction rate, the initial formation rate of phenyl ethanol was taken, considering at
least three data points. To estimate the enzyme deactivation rate, the following formula
were used:
[
푈푃푂]0
푈푃푂 푑푒푎푐푡푖푣푎푡푖표푛 푟푎푡푒 =
푅푒푎푐푡푖표푛 푡푖푚푒
ACS Paragon Plus Environment
26

Page 27 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
With:
[
푀푎푥 푝ℎ푒푛푦푙 푒푡ℎ푎푛표푙]
푅푒푎푐푡푖표푛 푡푖푚푒 =
푟푒푎푐푡푖표푛 푟푎푡푒
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The reaction selectivity was calculated by dividing the maximum amount of phenyl ethanol
reached by the total amount of product formed in that same point in time.
[
푀푎푥 푝ℎ푒푛푦푙 푒푡ℎ푎푛표푙]
푀푎푥 푝ℎ푒푛푦푙 푒푡ℎ푎푛표푙] + [퐴푐푒푡표푝ℎ푒푛표푛푒]푡
푆푒푙푒푐푡푖푣푖푡푦 =
[
ACS Paragon Plus Environment
27

ACS Catalysis
Page 28 of 44
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
AUTHOR INFORMATION
Corresponding Authors
*
m.m.c.h.vanschie@tudelft.nl, f.hollmann@tudelft.nl
SUPPORTING INFORMATION
Supporting information is available free of charge on the ACS Publications website at
DOI: XX
Details of the preparation of the catalysts, on the reaction setup, analytical data and
additional results.
ACKNOWLEDGMENT
ACS Paragon Plus Environment
28

Page 29 of 44
ACS Catalysis
1
2
3
4
5
6
7
8
9
We thank the Netherlands Organization for Scientific Research for financial support
through a VICI grant (No. 724.014.003) and by the European Union Project H2020-BBI-
PPP-2015-2-720297-ENZOX2. CEP acknowledges a VENI grant (No. 722.015.011).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
REFERENCES
1.
Roduner, E.; Kaim, W.; Sarkar, B.; Urlacher, V. B.; Pleiss, J.; Gläser, R.; Einicke,
W.-D.; Sprenger, G. A.; Beifuß, U.; Klemm, E.; Liebner, C.; Hieronymus, H.; Hsu, S.-F.;
Plietker, B.; Laschat, S., Selective Catalytic Oxidation of C-H Bonds with Molecular
Oxygen. ChemCatChem 2013, 5, 82-112.
2.
Dong, J.; Fernández-Fueyo, E.; Hollmann, F.; Paul, C.; Pesic, M.; Schmidt, S.;
Wang, Y.; Younes, S.; Zhang, W., Biocatalytic Oxidation reactions ‐ a Chemist's
Perspective. Angew. Chem. Int. Ed. 2018, 57, 9238-9261.
ACS Paragon Plus Environment
29