Chemo- and Regioselective Dihydroxylation of Benzene to Hydroquinone Enabled by Engineered Cytochrome P450 Monooxygenase

Article abstract of DOI:10.1002/anie.201812093

Hydroquinone (HQ) is produced commercially from benzene by multi-step Hock-type processes with equivalent amounts of acetone as side-product. We describe an efficient biocatalytic alternative using the cytochrome P450-BM3 monooxygenase. Since the wildtype enzyme does not accept benzene, a semi-rational protein engineering strategy was developed. Highly active mutants were obtained which transform benzene in a one-pot sequence first into phenol and then regioselectively into HQ without any overoxidation. A computational study shows that the chemoselective oxidation of phenol by the P450-BM3 variant A82F/A328F leads to the regioselective formation of an epoxide intermediate at the C3=C4 double bond, which departs from the binding pocket and then undergoes fragmentation in aqueous medium with exclusive formation of HQ. As a practical application, an E. coli designer cell system was constructed, which enables the cascade transformation of benzene into the natural product arbutin, which has anti-inflammatory and anti-bacterial activities.

Full text of DOI:10.1002/anie.201812093

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201812093
German Edition: DOI: 10.1002/ange.201812093
Benzene Dihydroxylation
Chemo- and Regioselective Dihydroxylation of Benzene to
Hydroquinone Enabled by Engineered Cytochrome P450
Monooxygenase
Hangyu Zhou+, Binju Wang+, Fei Wang, Xiaojuan Yu, Lixin Ma, Aitao Li,* and
Manfred T. Reetz*
Abstract: Hydroquinone (HQ) is produced commercially
from benzene by multi-step Hock-type processes with equiv-
alent amounts of acetone as side-product. We describe an
efficient biocatalytic alternative using the cytochrome P450-
BM3 monooxygenase. Since the wildtype enzyme does not
accept benzene, a semi-rational protein engineering strategy
was developed. Highly active mutants were obtained which
transform benzene in a one-pot sequence first into phenol and
then regioselectively into HQ without any overoxidation. A
computational study shows that the chemoselective oxidation
of phenol by the P450-BM3 variant A82F/A328F leads to the
formation of the intermediate alkylhydroperoxides, and
final acid catalyzed formation of the desired product together
with equivalent amounts of acetone as side-product (Fig-
ure 1A).[1]
=
regioselective formation of an epoxide intermediate at the C3
C4 double bond, which departs from the binding pocket and
then undergoes fragmentation in aqueous medium with
exclusive formation of HQ. As a practical application, an E.
coli designer cell system was constructed, which enables the
cascade transformation of benzene into the natural product
arbutin, which has anti-inflammatory and anti-bacterial activ-
ities.
Figure 1. Synthesis of hydroquinone (HQ) from benzene by the
conventional route (A), the newly developed biocatalytic procedure (B),
and, as an example of practical application, the biocatalytic cascade
synthesis of arbutin from benzene using designed E. coli cells harbor-
ing P450-BM3 mutant and arbutin synthase (C).
H
ydroquinone (HQ; 1,4-dihydoxy benzene) is an important
intermediate used in the commercial preparation of many
different compounds for the chemical, pharmaceutical, and
polymer industries.[1] HQ is also a pheromone in termites,[2a]
and has been found in higher plants in the mono-glucosylated
form called arbutin, a natural product with anti-oxidant, anti-
inflammatory, and anti-bacterial properties, often used as
a skin-lightener.[2b,c] The current industrial production of HQ
utilizes benzene as an oil-derived feedstock, which is sub-
jected to a double Hock-type procedure involving two
Friedel–Crafts isopropylations, followed by O2-mediated
In terms of green chemistry, hydroxylation of benzene
followed by regioselective introduction of the second hydrox-
yl group at the para-position would be preferred. This is
a long-standing challenge because the introduction of hy-
droxyl groups in aromatic compounds such as benzene results
in uncontrollable oxidative susceptibility.[1,3] It is interesting
[*] H. Zhou[+,++], F. Wang, X. Yu, Prof. L. Ma, Prof. A. Li
State Key Laboratory of Biocatalysis and Enzyme Engineering, Hubei
Collaborative Innovation Center for Green Transformation of Bio-
Resources, Hubei Key Laboratory of Industrial Biotechnology, School
of Life Sciences, Hubei University
32 West 7th Avenue, Tianjin Airport Economic Area, Tianjin 300308
(China)
and
Max-Planck-Institut fꢀr Kohlenforschung
Kaiser-Wilhelm-Platz 1, 45470 Muelheim (Germany)
and
Wuhan, 430062 (P. R. China)
E-mail: aitaoli@hubu.edu.cn
Prof. B. Wang[+,++]
State Key Laboratory of Physical Chemistry of Solid Surfaces and
Fujian Provincial Key Laboratory of Theoretical and Computational
Chemistry, College of Chemistry and Chemical Engineering, Xiamen
University
Department of Chemistry, Philipps-University
Hans-Meerwein-Strasse 4, 35032 Marburg (Germany)
E-mail: reetz@mpi-muelheim.mpg.de
[+] These authors contributed equally to this work.
++] co first authors.
[
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.201812093.
Xiamen 360015 (P. R. China)
Prof. M. T. Reetz
Tianjin Institute of Industrial Biotechnology, Chinese Academy of
Sciences
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
Table 1: P450-BM3 catalyzed hydroxylation of phenol to hydroquinone
(HQ) by E. coli cells containing P450-BM3 and its mutants, and turnover
rate and coupling efficiency measured using corresponding purified
enzymes.
to note that HQ was reported to be one of several products, in
the mammalian metabolism of benzene.[4] We now report the
preparative one-pot regioselective dihydroxylation of ben-
zene with formation of HQ without any overoxidation,
catalyzed by a semi-rationally evolved mutant of cytochro-
me P450-BM3 (Figure 1B). We have also devised a cascade
conversion of benzene to arbutin by constructing E. coli
designer cells harboring the P450 monooxygenase and
a glucosyltransferase (Figure 1C).
We considered the self-sufficient cytochrome P450 mono-
oxygenase from Bacillus megaterium, P450-BM3,[5] which
catalyzes the oxidative hydroxylation of long-chain fatty acids
as natural substrates, but is essentially inert to small molecules
such as propane, cyclohexane or benzene. When adding per-
fluoro fatty acids as decoys,[6] benzene is hydroxylated with
formation of phenol at a rate higher than the activity observed
when using a P450-BM3 mutant alone.[7] Knowing that mono-
substituted benzene derivatives are hydroxylated by WT
P450-BM3 and mutants solely at the ortho-position,[8] we
chose this enzyme without resorting to additives.
Catalysts Conv. Product distribu- Rate
[%][a] tion[b] [%] [minꢀ1 enzymeꢀ1 [c]
HQ Catechol
TTN[d] CE
[%][e]
]
WT
V78F
A82F
8
95
100
99
100
99
5
0
1
0
1
n.d.
n.d.
3
1910 33
150 8.0
5900 60
n.d.
n.d.
10
67
54
99
0.3ꢁ0.3
191ꢁ1
15ꢁ2
A328F
A82F/
A328F
V78F/
A82F
A78F/
A328F
A82F/
V78F/
A328F
590ꢁ15
33
44
85
100
100
98
0
0
2
19ꢁ2
190
110
13
12
11ꢁ1
348ꢁ15
3480 62
[a] Conditions with cells: 10 mm substrate, 308C, 200 rpm, 5 h. Con-
version was determined by HPLC analysis. For detailed conditions, see
the Supporting Information. [b] Relative amounts based on the concen-
trations of products formed as detected by HPLC analysis. Mean values
are given, standard deviation for conversion is ꢁ5% and for product
distribution ꢁ1%. [c] Conditions with purified enzymes: WT or
P450BM3 mutants (0.5 mm), phenol (10 mm), NADPH (5 mm), 308C,
200 rpm, 10 min. The reaction rate was determined by HPLC analysis
based on the amount of HQ formed. For detailed conditions see the
Supporting Information, mean values and standard errors from
triplicates. [d] Total turnover unmber (TTN): [Hydroquinone formatio-
n]/[Enzyme]. [e] Coupling efficiency (CE): ([Hydroquinone formation]/
[NADPH consumption])ꢁ100. n.d. not detectable.
In our earlier study,[9] exploratory NNK-based saturation
mutagenesis at 8 residues lining the binding pocket of P450-
BM3 harboring a structurally small substrate (cyclohexanone)
revealed a mutational fingerprint for each position. Mutations
at residues A78, A82, and V328 signaled enhanced activity,
especially when phenylalanine was introduced. In the present
study, we learned from directed evolution and generated
mutants in which phenylalanine, a relatively large hydro-
phobic amino acid, is placed at these “hot” spots, namely
A78F, A82F, V328F, A82F/A328F, V78F/A328F, and A82F/
A328F/V78F. Such mutations were expected to reduce the
size of the large binding pocket of P450-BM3, a hypothesis
which we first tested using phenol as substrate. Gratifyingly, in
whole cell reactions, single mutant A82F, double mutant
A82F/A328F, and triple mutant V78F/A82F/A328F showed
notable turnover rates with pronounced regioselectivity in
favor of HQ, with only traces of the regioisomer catechol
being detected (Table 1), and with no overoxidation.
Table 2: Selective dihydroxylation of benzene with almost exclusive
formation of HQ using the respective P450-BM3-containing E. coli
resting cells.
Catalysts
Conv. [%][a]
Product distribution[b] [%]
HQ
Phenol
Catechol
[c]
[c]
[c]
[c]
WT
–
–
–
–
The mutant enzymes were purified so that the turnover
rate and coupling efficiency of phenol hydroxylation could be
determined. The results are noteworthy for several reasons
(Table 1). Unprecedented activity was achieved for variants
A82F, A82F/A328F, and A82F/V78F/A328F, here as well with
no overoxidation. The double mutant A82F/A328F has the
best profile, resulting in a turnover rate of 590 minꢀ1 enzymeꢀ1
with unusually high coupling efficiency (60%). For compar-
ison, the phenol hydroxylation by WT P450-BM3 enabled by
decoy molecules produced a mixture of hydroquinone and
catechol as products at very low turnover rates amounting to
only 70 minꢀ1 enzymeꢀ1 and 15 minꢀ1 enzymeꢀ1, respec-
tively.[6]
We then turned to the crucial question of whether the
mutants also catalyzes the hydroxylation of benzene by
testing mutants A82F, A82F/A328F, and V78F/A82F/A328F.
Indeed, they catalyzed the dihydroxylation of benzene with
formation of HQ with no overoxidation. Among them,
double mutant A82F/A328F shows 97% conversion, account-
ing for > 90% of HQ formation (Table 2). The best double
mutant A82F/A328F was then employed for scale-up reac-
A82F
A82F/A328F
V78F/A82F/A328F
87
97
92
86
93
93
10
6
4
4
1
3
[a] Conditions: 10 mm substrate, 308C, 200 rpm, 5 h. Conversion was
determined by HPLC analysis and is based on the amount of converted
substrate. For detailed conditions, see the Supporting Information.
[b] Relative amounts based on the concentrations of products formed as
analyzed by HPLC analysis. [c] Not determined due to low activity. Mean
values are given, standard deviation for conversion is ꢁ5% and for
product distribution ꢁ1%.
tions starting from benzene or phenol as substrates for the
preparation of HQ, which led to 55–76% product yields.
The mechanism of P450-catalyzed hydroxylation of aro-
matics involves de-aromatizing formation of an epoxide,
followed by its rearrangement into phenol derivatives.[5]
However, nothing was known about the mechanism of
P450-catalyzed selective oxidation of phenol to HQ. To gain
some insight, combined classical molecular dynamics (MD)
simulations and calculations were performed.
2
www.angewandte.org
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
The phenol was first docked into the active site of variant
probably diffuses out into the bulk water where it fragments
P450-BM3 A82F/A328F, which has the highest activity. As the
with formation of HQ.
porphyrin p-cation radical ferryl-species (PorC + FeIV O, Cpd
In order to study the possible transformation of the
epoxide intermediate in water alone, hybrid cluster-contin-
uum (HCC) model calculations were performed.[11] Figure 3
shows the calculated relative free energy profile for the
fragmentation of the C3–C4–epoxide intermediate. Starting
from the reactant complex (RC2), the heterolytic cleavage of
=
I) is the active species responsible for P450-catalyzed
epoxidation and other monooxygenation reactions,[5,10] the
advanced QM(B3LYP)/MM MD simulations were carried
out in the Cpd I state in the presence of phenol substrate.
Figure 2A shows QM(B3LYP)/MM MD simulation-pre-
Figure 2. A) QM(B3LYP)/MM MD simulations-predicted phenol bind-
ing in the active site of variant P450-BM3 A82F/A328F. B) QM-
(B3LYP)/MM metadynamics-calculated free energy (in kcalmolꢀ1
)
profile for epoxide formation from phenol. The reaction coordinate is
defined as the distance between Ox and C4, while the width of the
Gaussian shaped potential hills was set to 0.1 ꢂ. The representative
structures of QM region are shown along the reaction pathway. Spin-
up isodensity surfaces are shown in yellow, while spin-down isodensity
surfaces are plotted in red.
Figure 3. BMK/6–311+ +G(d,p) relative free energies (in kcalmolꢀ1
for the fragmentation of phenol C3–C4–epoxide intermediate into
)
hydroquinone in aqueous solution (no enzyme) with five explicit water
molecules in the HCC calculations.
dicted phenol binding in the active site of variant A82F/
A328F. The model suggests that phenol is “barricaded” in the
active site by three phenylalanine residues (F328, F87, and
F82) (Figure 2A), while the precise positioning of phenol for
epoxidation is controlled mainly by the H-bonding interaction
between the phenol hydroxyl group and residue T260, forcing
ꢀ
C3 Ox bond via TS2a leads to the formation of an unusual
zwitterion intermediate of IC2a, in which the negatively
charged Ox group is stabilized by a strong H-bond (Figure S3
in the Supporting Information), while the positive charge is
=
=
ꢀ
the C3 C4 double bond of phenol to lie closest to the Fe O
moiety (Ox) throughout the QM/MM MD simulations (Fig-
ure S1 in the Supporting Information). This suggests that the
well delocalized on the phenyl and OH groups. The C3 Ox
cleavage pathway is quite facile, with a free energy barrier of
only 10.2 kcal/mol. Then the H2O-assisted proton transfer
from C4 to Ox via TS2b leads to fragmentation and formation
of HQ with a barrier of 7.2 kcalmolꢀ1 relative to IC2a.
Alternatively, IC1a may isomerize to IC2b via an intra-
molecular proton transfer from C4 to C3 (TS2c), which is
similar to the so-called NIH shift mechanism of arene
oxidation.[12] The calculated barrier for intramolecular
proton transfer is only 0.8 kcalmolꢀ1, indicating that the
intramolecular proton transfer from C4 to C3 is much more
favorable than that from C4 to Ox. Afterwards, the water-
assisted proton transfer from C3 to Ox via TS2d leads to the
HQ product. The calculated overall barrier is 15.6 kcalmolꢀ1
=
=
C3 C4 rather than the C2 C3 double bond will be selectively
epoxidized by Cpd I. It is this unique characteristic which
distinguishes the present scenario from reactions of other
mono-substituted benzene derivatives that lead to ortho-
products.[8] This is crucial for regioselective HQ production,
and not catechol formation.
Figure 2B presents the QM (B3LYP)/MM metadynamics
calculated free energy profile for epoxide formation from
phenol in the doublet state, which in similar transformations
was demonstrated to be more reactive than the quartet.[10] It
also became clear that phenol epoxidation reaction proceeds
through a concerted mechanism, avoiding radical or cationic
ꢀ
(IC2b!TS2d). We also investigated the alternative C4 Ox
ꢀ
ꢀ
Fe Ox phenol complexes. The predicted free energy barrier
is 17.6 kcalmolꢀ1, in good agreement with experimental value
of about 16 kcalmolꢀ1. To demonstrate whether the so-
formed epoxide intermediate could undergo further trans-
formations in the binding pocket, the QM(B3LYP)/MM MD
simulations were performed on the 2IC1 species. It was found
that the epoxide intermediate is not persistent in the binding
pocket (Figure S2 in the Supporting Information). These
computational results suggest that the epoxide intermediate
cleavage pathway (Figure S4 in the Supporting Information),
which would provide 1,3-dihydoxy benzene (resorcinol).
Intriguingly, this pathway requires a high barrier of 25.3 kcal
molꢀ1, in line with the experimental observations.
With these experimental results and computational
insights in hand, numerous possibilities can be anticipated
which exploit the selective formation of HQ as a starting point
for creating designer cells as catalytic systems that convert
benzene into higher-value products. Our exemplary efforts
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
serving as substrate. An 83% yield for arbutin was obtained
within 1 h, while the glycosylated product (5) from phenol (2)
was not detected. In these experiments we observed the
accumulation of 17% HQ (3) as an intermediate that was not
converted further at prolonged reaction time (Figure S7 in the
Supporting Information), possibly indicating that the catalytic
activity of arbutin synthase (AS) towards HQ is the slow step
in the artificial cascade sequence 2 ! 3 ! 4. Indeed, our
kinetic study shows that the catalytic performance of AS
towards HQ (0.054 mmꢀ1 minꢀ1) is considerably lower than
that of the P450 mutants (480 mmꢀ1 minꢀ1 for A82F/A328F)
towards phenol (Table S1 in the Supporting Information).
Finally, the designed whole cell system was tested in the total
cascade sequence starting from benzene (5 mm), 1 ! 2! 3 !
4 (Figure 4B). As anticipated, high yields of arbutin (80–
83%) were achieved. As before, only a small amount of HQ
(3) accumulated in the reaction mixture owing to the limited
activity of AS towards this intermediate compound.
In conclusion, we have demonstrated the preparative
biocatalytic dihydroxylation of benzene with formation of
hydroquinone (HQ) in high yield and essentially complete
regioselectivity, using mutants of P450-BM3 produced by
semi-rational protein engineering. Overoxidation of HQ was
not observed. Phenol can also be used as substrate, which
likewise provides HQ as a result of exclusive para-selectivity.
To obtain an insight concerning the mechanistic intricacies of
this unusual transformation, multiscale simulations and
computations were performed. QM/MM MD simulations
Figure 4. A) Biosynthesis of arbutin with designed E. coli whole cell
catalyst for the one-pot production of arbutin (4) starting from
benzene (1). Green arrows: Natural endogeneous metabolic pathway;
blue arrows: artificial pathway. P450 mutant: Engineered P450-BM3
mutant A82F/A328F; GCK: Glucokinase; PGM1: Phosphoglucomu-
tase; UGP1: UDP-glucose pyrophosphorylase; AS: Arbutin synthase.
B) Time course for the production of arbutin (4) from benzene (1)
using resting cells of E. coli (P450A82F/A328F+AS).
=
=
suggest that the C3 C4 rather than the C2 C3 double bond
becomes selectively epoxidized by Cpd I, leading to the C3–
C4–epoxy intermediate, which departs into the aqueous
medium where H2O-assisted fragmentation occurs. For illus-
trating the practical potential of this process, designer E. coli
whole cells were created exploiting the best P450-BM3
mutant and an appropriate glucosyltransferase for the effi-
cient one-pot synthesis of arbutin as an added-value com-
pound with distinct biological activity. Further E. coli
designer cells can be envisioned for cascades,[15] leading to
other types of higher-value products starting from benzene as
the feedstock.
focused on the one-pot conversion of benzene to arbutin
(Figure 4A). The biosynthesis of this compound in plants
involves glucose as starting material, which is also the
feedstock in the case of some engineered pathways,[13] not
benzene as in our system.
In order to find an appropriate glucosyltransferase for
mono- glycosylation of HQ, the UDP-glucose dependent
glucosyltransferase from Rauvolfia serpentina (arbutin syn-
thase, AS) was considered, because it shows high activity
towards HQ (3) and only 5% of activity for glucosylation of
phenol (2).[14] Therefore, knowing that the P450-BM3 mutants
exhibit very high activity for phenol hydroxylation, undesired
Acknowledgements
competitive glucosylation of intermediate
2
appeared
Aitao Li thanks the financial support from the National
Natural Science Foundation of China (Grant No. 21702052)
and State Key Laboratory of Biocatalysis and Enzyme
Engineering. M.T. Reetz thanks the Max-Planck-Society for
generous support.
unlikely. The glucosyltransferase arbutin synthase (AS) was
cloned and expressed in recombinant E. coli, and then the
whole cell activity was tested for HQ acceptance. Indeed,
more than 95% of HQ (10 mm) was converted to arbutin
within 5 h (Figure S5 in the Supporting Information). Based
on these positive results, we constructed the T7 promoter-
based E. coli expression system with one single plasmid
containing the best P450-BM3 mutant A82F/A328F and
glucosyltransferase AS (Figure S6 in the Supporting Infor-
mation), namely E. coli (P450A82F/A328F + AS). The
designed whole cells were first tested in the conversion of
phenol (2) (5 mm) to arbutin (4) (Figure S7 in the Supporting
Information). Under these conditions, the UDP-glucose is
supplied by the pathway in the E. coli host cells with glucose
Conflict of interest
The authors declare no conflict of interest.
Keywords: benzene dihydroxylation · cascade reaction ·
directed evolution · hydroquinone · P450 monooxygenase
4
www.angewandte.org
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
[8] A. Dennig, N. Lꢁlsdorf, H. Liu, U. Schwaneberg, Angew. Chem.
Int. Ed. 2013, 52, 8459 – 8462; Angew. Chem. 2013, 125, 8617 –
8620.
[9] A. Li, A. Ilie, Z. Sun, R. Lonsdale, M. T. Reetz, Angew. Chem.
Int. Ed. 2016, 55, 12026 – 12029; Angew. Chem. 2016, 128, 12205 –
12208.
[1] a) L. Krumenacker, M. Costantini, P. Pontal, J. Sentenac in Kirk-
Othmer Encyclopedia of Chemical Technology, Vol. 13 (Eds.: J. I.
Kroschwitz, M. Howe-Grant), Wiley, New York, 1995, p. 996;
b) P. M. Hudnall in Ullmannꢀs Encyclopedia for Industrial
Chemistry, Wiley-VCH, Weinheim, 2000.
[2] a) J. Reinhard, M. J. Lacey, F. Ibarra, F. C. Schroeder, M. Kaib,
M. Lenz, J. Chem. Ecol. 2002, 28, 1 – 14; b) K. Sugimoto, K.
Nomura, T. Nishimura, T. Kiso, K. Sugimoto, T. Kuriki, J. Biosci.
Bioeng. 2005, 99, 272 – 276; c) D.-H. Seo, J.-H. Jung, J.-E. Lee,
E. J. Jeon, W. Kim, C.-S. Park, Appl. Microbiol. Biotechnol. 2012,
95, 1417 – 1425.
[10] S. P. de Visser, S. Shaik, J. Am. Chem. Soc. 2003, 125, 7413 – 7424.
[11] a) J. R. Pliego, J. M. Riveros, J. Phys. Chem. A 2001, 105, 7241 –
7247; b) B. Wang, Z. Cao, J. Phys. Chem. A 2010, 114, 12918 –
12927; c) B. Wang, Z. Cao, D. Sharon, S. Shaik, ACS Catal. 2015,
5, 7077 – 7090; d) B. Wang, E. M. Johnston, P. Li, S. Shaik, G. J.
Davies, P. H. Walton, C. Rovira, ACS Catal. 2018, 8, 1346 – 1351.
[12] J. W. Daly, D. M. Jerina, B. Witkop, Experientia 1972, 28, 1129 –
1264.
[13] a) B. Assoah, L. F. Veiros, C. A. M. Afonso, N. R. Candeia, Eur.
J. Org. Chem. 2016, 3856 – 3861; b) N. Ran, D. R. Knop, K. M.
Draths, J. W. Frost, J. Am. Chem. Soc. 2001, 123, 10927 – 10934.
[14] a) J. Arend, H. Warzecha, T. Hefner, J. Stçckigt, Biotechnol.
Bioeng. 2001, 76, 126 – 131; b) X. Shen, J. Wang, J. Wang, Z.
Chen, Q. Yuan, Y. Yan, Metab. Eng. 2017, 42, 52 – 58.
[15] For reviews of biocatalytic cascade reactions, see: a) N. J. Turner,
Nat. Chem. Biol. 2013, 9, 285 – 288; b) K. Faber, W. D. Fessner,
N. Turner, Adv. Synth. Catal. 2015, 357, 1565 – 1566; c) N. J.
Turner, ACS Catal. 2017, 7, 710 – 724; d) W. Kroutil, Chem. Rev.
2018, 118, 270 – 348; e) U. Bornscheuer, Nat. Catal. 2018, 1, 12 –
22; f) S. Wu, Z. Li, ChemCatChem 2018, 10, 2164 – 2178.
[3] For recent examples of transition metal catalyzed oxidation of
benzene affording product mixtures, see: X. Wang, F. Wang, B.
Chen, K. Cheng, J. Wang, J. Zhang, H. Song, Appl. Surf. Sci.
2018, 453, 320 – 329.
[4] a) R. Snyder, Drug Metab. Rev. 2004, 36, 531 – 547; b) U.
Bernauer, B. Vieth, R. Ellrich, B. Heinrich-Hirsch, G. Jꢀnig, U.
Gundert-Remy, Arch. Toxicol. 1999, 73, 189 – 196.
[5] Among many reviews of cytochrome P450 monooxygenases,
see: a) P. R. Ortiz de Montellano, Chem. Rev. 2010, 110, 932 –
948; b) S. Shaik, S. Cohen, Y. Wang, H. Chen, D. Kumar, W.
Thiel, Chem. Rev. 2010, 110, 949 – 1017.
[6] a) O. Shoji, T. Kunimatsu, N. Kawakami, Y. Watanabe, Angew.
Chem. Int. Ed. 2013, 52, 6606 – 6610; Angew. Chem. 2013, 125,
6738 – 6742; b) O. Shoji, S. Yanagisawa, J. K. Stanfield, K.
Suzuki, Z. Cong, H. Sugimoto, Y. Shiro, Y. Watanabe, Angew.
Chem. Int. Ed. 2017, 56, 10324 – 10329; Angew. Chem. 2017, 129,
10460 – 10465.
Manuscript received: October 24, 2018
[7] E. T. Farinas, M. Alcalde, F. H. Arnold, Tetrahedron 2004, 60,
525 – 528.
Revised manuscript received: November 6, 2018
Version of record online: && &&, &&&&
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Benzene Dihydroxylation
H. Zhou, B. Wang, F. Wang, X. Yu, L. Ma,
A. Li,* M. T. Reetz*
&&&& — &&&&
Chemo- and Regioselective
Dihydroxylation of Benzene to
Hydroquinone Enabled by Engineered
Cytochrome P450 Monooxygenase
Unique enzymatic one-pot dihydroxyla-
tion of benzene to hydroquinone: A
mutant of a cytochrome P450 monooxy-
genase catalyzes this challenging trans-
formation with no overoxidation, ena-
bling, among other things, the cascade
synthesis of biologically active arbutin.
6
www.angewandte.org
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2018, 57, 1 – 6
These are not the final page numbers!