Ketoreductase catalyzed stereoselective bioreduction of α-nitro ketones

Article abstract of DOI:10.1039/c9ob00051h

We report here the stereoselective bioreduction of α-nitro ketones catalyzed by ketoreductases (KREDs) with publicly known sequences. YGL039w and RasADH/SyADH were able to reduce 23 class I substrates (1-aryl-2-nitro-1-ethanone (1)) and ten class II substrates (1-aryloxy-3-nitro-2-propanone (4)) to furnish both enantiomers of the corresponding β-nitro alcohols, with good-to-excellent conversions (up to >99%) and enantioselectivities (up to >99% ee) being achieved in most cases. To the best of our knowledge, KRED-mediated reduction of class II α-nitro ketones (1-aryloxy-3-nitro-2-propanone (4)) is unprecedented. Select β-nitro alcohols, including the synthetic intermediates of bioactive molecules (R)-tembamide, (S)-tembamide, (S)-moprolol, (S)-toliprolol and (S)-propanolol, were stereoselectively synthesized in preparative scale with 42% to 90% isolated yields, showcasing the practical potential of our developed system in organic synthesis. Finally, the advantage of using KREDs with known sequence was demonstrated by whole-cell catalysis, in which β-nitro alcohol (R)-2k, the key synthetic intermediate of hypoglycemic natural product (R)-tembamide, was produced in a space-time yield of 178 g L^?1 d^?1 as well as 95% ee by employing the whole cells of a recombinant E. coli strain coexpressing RasADH and glucose dehydrogenase as the biocatalyst.

Full text of DOI:10.1039/c9ob00051h

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Ketoreductase catalyzed stereoselective
bioreduction of α-nitro ketones†
Zexu Wang,a,b Xiaofan Wu,c Zhining Li,a,b Zedu Huang
Cite this: DOI: 10.1039/c9ob00051h
*
a,b and Fener Chen*a,b
We report here the stereoselective bioreduction of α-nitro ketones catalyzed by ketoreductases (KREDs)
with publicly known sequences. YGL039w and RasADH/SyADH were able to reduce 23 class I substrates
(1-aryl-2-nitro-1-ethanone (1)) and ten class II substrates (1-aryloxy-3-nitro-2-propanone (4)) to furnish both
enantiomers of the corresponding β-nitro alcohols, with good-to-excellent conversions (up to >99%) and
enantioselectivities (up to >99% ee) being achieved in most cases. To the best of our knowledge, KRED-
mediated reduction of class II α-nitro ketones (1-aryloxy-3-nitro-2-propanone (4)) is unprecedented. Select
β-nitro alcohols, including the synthetic intermediates of bioactive molecules (R)-tembamide, (S)-temb-
amide, (S)-moprolol, (S)-toliprolol and (S)-propanolol, were stereoselectively synthesized in preparative scale
with 42% to 90% isolated yields, showcasing the practical potential of our developed system in organic
synthesis. Finally, the advantage of using KREDs with known sequence was demonstrated by whole-cell
catalysis, in which β-nitro alcohol (R)-2k, the key synthetic intermediate of hypoglycemic natural product (R)-
tembamide, was produced in a space–time yield of 178 g L−1 d−1 as well as 95% ee by employing the whole
cells of a recombinant E. coli strain coexpressing RasADH and glucose dehydrogenase as the biocatalyst.
Received 9th January 2019,
Accepted 19th March 2019
DOI: 10.1039/c9ob00051h
rsc.li/obc
well-recognized synthetic intermediates that can undergo a
variety of useful transformations,3,5 such as conversion to
Introduction
Ketoreductases (KREDs) are NAD(P)H-dependent enzymes dis- β-amino alcohols, which are privileged structural elements
playing a broad substrate spectrum for the reduction of pro- present in many pharmaceuticals and bioactive natural pro-
chiral ketones to the corresponding chiral alcohols.1 ducts (Fig. 1).6 β-Amino alcohols are also widely used as chiral
Undoubtedly KREDs are some of the most common and soph- ligands and auxiliaries in organic syntheses.7
isticated biocatalysts utilized in the industrial syntheses of
When we initiated our project, the bioreduction of α-nitro
pharmaceuticals, mainly due to their excellent stereo- ketones was exclusively performed using whole cells of baker’s
selectivity, good stability and environmental compatibility.1a,2 yeast, Comamonas testosterone, and Candida parapsilosis
Ketones with
a substituent α
to the carbonyl group (Scheme 1).8 However, most of those studies, only demon-
(α-substituted ketones) represent an important class of sub- strated limited substrate scope (mainly aliphatic α-nitro
strates for KREDs, because the resulting alcohols are bifunc- ketones) and only accessed one enantiomer of the β-nitro
tional.3 In this regard, KREDs catalyzing the asymmetric alcohol products. These low stereoselectivities can be attribu-
reduction of α-azido ketones, α-nitrile ketones, α-hydroxy ted to the presence of multiple KREDs in the whole cell
ketones and α-halo ketones are well-documented.4 In contrast, system, some with opposite stereochemical preferences.8d
the bioreduction of α-nitro ketones has been scarcely studied, Therefore, we decided to systematically investigate the biore-
which is surprising given that the resulting β-nitro alcohols are
aEngineering Center of Catalysis and Synthesis for Chiral Molecules, Department of
Chemistry, Fudan University, 220 Handan Road, Shanghai, 200433, P. R. China.
E-mail: huangzedu@fudan.edu.cn, rfchen@fudan.edu.cn
bShanghai Engineering Research Center of Industrial Asymmetric Catalysis of Chiral
Drugs, 220 Handan Road, Shanghai, 200433, P. R. China
cCollege of Chemical Engineering, Fuzhou University, 2 Xueyuan Road, Fuzhou,
350100, P. R. China
†Electronic supplementary information (ESI) available: Supplementary figures,
molecular biology process, enzyme purifications, and synthetic procedures for
the preparation of substrates and standards and their spectroscopic characteriz-
ation. See DOI: 10.1039/c9ob00051h
Fig. 1 Select examples of bioactive natural products and pharma-
ceutical molecules containing vicinal amino alcohol moiety.
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
viously shown to be capable of stereoselectively reducing
α-nitrile ketones.4c To our delight, YGL039w indeed was able
to convert 1a to the desired alcohol 2a (Fig. S3 and Table S4†).
We found that citric acid buffer of pH 5 was the best buffer, in
which 2a was formed in 83% conversion as measured by 1H
NMR, alongside with 9% 1a and 8% benzoic acid (3a) (see
Fig. S3 and S4† for a detailed discussion of calculating 1H
NMR conversions). The formation of benzoic acid presumably
arises from the instability of 1a under the reaction con-
ditions.9,11 Due to the strong acidity of its α-proton, 1a likely
exists predominantly as its enol(ate) form under higher pHs,
accounting for the relatively slow reactions observed at pHs 6,
7, and 8. On the other hand, YGL039w and/or glucose dehydro-
genase used for NADPH regeneration might have decreased
catalytic abilities at pH 4, which again would result in lower
conversions. We were pleased to find that the complete conver-
sion of 1a to 2a could be realized by increasing the enzyme
loading and reaction time, and importantly no benzoic acid
was detected under such conditions (entry 6, Table S4†).
Moreover, the biosynthesized 2a possessed excellent enantio-
meric purity (99% ee, (S)-configuration). Aiming to uncover an
stereocomplementary enzyme, an in-house collection of
13 KREDs with publicly known sequences were further evalu-
ated on the bioreduction of 1a under the optimal reaction con-
ditions (Table S5†). Gratifyingly, RasADH, a KRED originating
from Ralstonia species,12 produced the desired (R)-2a with
98% ee.
With a pair of stereocomplementary KREDs identified, we
next evaluated the performance of these two enzymes in redu-
cing 23 class I α-nitro ketones: 1-aryl-2-nitro-1-ethanone (1)
(Scheme 2). In YGL039w-catalyzed reductions of ketones with
a mono-substituted phenyl ring (1a to 1q), the desired alcohols
(2a to 2q) were generated in moderate-to-excellent conversions
(up to >99%), with good-to-excellent enantiomeric purities in
most cases (up to >99% ee). Apparently, steric effect played an
important role with regard to reaction conversion. For
example, moderate conversions (55% and 35%) were seen for
substrates 1o and 1q which bear bulky p-NHC(O)CH3 and p-
tBu groups, respectively. In contrast, >95% conversions were
achieved for substrates 1j and 1k with a p-Me and a p-OMe
group, respectively. Steric effects were also reflected in the
reduction of substrates with the same substituents, located at
different positions. For instance, while ketone 1i with a
p-chloro group was transformed to alcohol 2i in >99% conver-
sion, its positional isomers 1e (m-chloro) and 1b (o-chloro)
were reduced in 73% and 63% conversions, respectively.
Electronic effect also has a strong influence on the reactivity,
as evidenced by the superior conversions observed for ketones
with electron-donating groups (EDGs) (e.g., 1j and 1k) over
those with electron-withdrawing groups (EWGs) (e.g., 1l and
1p). Bis-substituted ketones 1r and 1s are also substrates of
YGL039w, affording the corresponding alcohols 2r and 2s in
80% and 45% conversions, respectively. Replacement of the
Scheme 1 Different approaches to the stereoselective bioreduction of
α-nitro ketones.
duction of α-nitro ketones using isolated KREDs. Very recently,
Brenna and co-workers reported an elegant study using com-
mercial KREDs for the stereoselective reduction of α-nitro
ketones (class I, Scheme 1), affording both enantiomers of
β-nitro alcohols with good-to-excellent ee values in most
cases.9 Although the use of commercial KREDs is convenient,
genome mining or genetic manipulation with these KREDs
becomes difficult for most researchers, given that the gene/
protein sequences of most of commercial KREDs are not pub-
licly available.10 Herein, we report the stereoselective bioreduc-
tion of α-nitro ketones (both class I and class II, Scheme 1),
using isolated KREDs with publicly known sequences, to
β-nitro alcohols with good-to-excellent conversions (up to
>99%) and stereoselectivities (up to >99% ee) in most cases.
We also succeeded in the preparative scale synthesis of several
important β-nitro alcohols, including the synthetic intermedi-
ates of bioactive molecules (R)-tembamide, (S)-tembamide, (S)-
moprolol, (S)-toliprolol and (S)-propanolol. Moreover, using
whole cells of an E. coli strain recombinantly coexpressing
RasADH and GDH as the biocatalyst, alcohol (R)-2k was syn-
thesized in a highly stereoselective fashion with a space–time
yield of 178 g L−1 d−1, underscoring the superior applicability
of KREDs with publicly known sequence.
Results and discussion
We first examined the ability of YGL039w, a KRED originating phenyl ring with a naphthalenyl ring probably rendered the
from Saccharomyces cerevisiae, in the reduction of the model carbonyl group less accessible, which resulted in poor conver-
substrate α-nitro acetophenone 1a, since this enzyme was pre- sions observed for 1t and 1u. Nevertheless, thus formed
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
Scheme 2 KRED-catalyzed stereocomplementary synthesis of β-nitro alcohols 2. Unless otherwise stated, the reaction was carried out with 10 mM
ketone substrate, 20 mM glucose, 0.2 mM NADP+, 1 mg mL−1 KRED (24.7 μM YGL039w or 34.6 μM RasADH) and 1 mg mL−1 glucose dehydrogenase
(GDH) (24.3 μM) in 1 mL of 50 mM citric acid, pH 5.0 at 30 °C and 180 rpm for 1 h. [a] The reaction conversion was determined using 1H NMR (see
Table S6† for details). [b] The ee was determined using chiral HPLC. [c] The absolute configuration was determined by comparing the elution order
in chiral HPLC with known data. [d] The absolute configuration was not determined (N.D.). [e] The reaction was performed with 5 mg mL−1 KRED
(123.7 μM YGL039w or 173.1 μM RasADH) and 1 mg mL−1 GDH (24.3 μM) for 2 h. [f] The reaction was performed with 3 mg mL−1 KRED (74.2 μM
YGL039w or 103.8 μM RasADH) and 1 mg mL−1 GDH (24.3 μM) for 2 h. [g] The reaction was performed with 1 mg mL−1 KRED (24.7 μM YGL039w or
34.6 μM RasADH) and 1 mg mL−1 GDH (24.3 μM) for 2 h.
alcohol 2u possessed remarkably high enantiomeric purity Contrary to YGL039w, RasADH transformed α-nitro ketones
(99% ee). α-Nitro ketones 1v and 1w, containing one and two equipped with substituents at the ortho-position more readily
carbons, respectively, between the phenyl ring and the carbo- than those bearing substituents at the meta-position (1b, 1c,
nyl group, were subjected to YGL039w-catalyzed reduction, fur- 1d versus 1e, 1f, 1g). With the exception of ketone 1q (bearing
nishing the desired alcohols 2v and 2w in 82% and 48% con- a p-tBu substituent), all the mono-para-substituted ketones
versions, respectively. Interestingly, the configuration of turned out to be excellent substrates for RasADH, and the
product 2v was identified as (R), opposite to that of 2a, desired alcohols were generated in 85%-to->99% conversions.
suggesting multiple enzyme–substrate binding modes exist. Similar to YGL039w, RasADH reduced bis-substituted ketones
Due to the instability of the α-nitro ketones under reaction (1r and 1s) and ketones with a naphthalenyl ring (1t and 1u)
conditions, variable amounts (4% to 40%) of benzoic acid with diminished efficiency, probably due to steric hindrance.
derivatives 3 appeared as side products in nine out of 23 biore- Having one and two carbons, respectively, between the phenyl
duction reactions (see Table S6† for details).
ring and the carbonyl group, ketone 1v and 1w were both
The majority of the examined substrates were reduced by reduced by RasADH in moderate conversions but with fairly
RasADH in excellent stereoselectivities (up to >99% ee). different stereoselectivities (17% versus 90% ee). Finally, reac-
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
tion conversions could be improved by increasing enzyme
loadings or prolonging reaction time (e.g., 1c and 1u).
Given that enantioenriched 1-aryloxy-3-nitro-2-propanol
derivatives (5) are key synthetic intermediates to β-adrenergic
receptor blocking agents,6b we were keen to find out that if
they could be prepared through KRED-catalyzed stereoselective
reduction of their ketone precursors: 1-aryloxy-3-nitro-2-propa-
none (4). We first re-examined the same library of KREDs used
for the above reduction of 1 for the reduction of the model
substrate 4a (Table S8†). Again, YGL039w and RasADH stood
out as the most selective enzymes with opposite stereochemi-
cal preferences, affording 5a in 96% ee and 58% ee, respect-
ively. To find an enzyme with the same stereochemical prefer-
ence as RasADH but with improved stereoselectivity in redu-
cing 4a, we screened 10 more KREDs. Out of these enzymes,
SyADH performed the best, yielding the intended isomer of 5a
in 96% ee. Originating from Sphingobium yanoikuyae,13 SyADH
was previously used in the stereoselective reduction of α-chloro
ketones to β-chloro alcohols.
Ten α-nitro ketones of this second class were then chosen
for the study of substrate scope (Scheme 3). In general, steric
and electronic effects exhibited by the phenyl ring R group
were less significant here than in the bioreduction of class I
α-nitro ketones, probably because the phenyl ring is not
directly connected to the carbonyl group. Except ketone 4j, all
the ketones were readily reduced by YGL039w and SyADH to
give the desired alcohols in good-to-excellent conversions (up
to >99%), along with excellent stereoselectivities (up to 99%
ee) being achieved in most cases. Similar as in the above dis-
cussion, the steric hindrance introduced by the naphthalenyl
ring likely caused the low conversions observed for ketone 4j.
Nevertheless, the bioreduction of 4j catalyzed by YGL039w
yielded the corresponding alcohol (S)-5j with 98% ee. As a
result of the instability of α-nitro ketones, α-aryloxy acetic acid
(6) was detected as side products in about half of the bioreduc-
tion reactions in amounts ranging from 5% to 41%
(see Table S9† for details).
Scheme 3 KRED-catalyzed stereocomplementary synthesis of β-nitro
alcohols 5. The reaction was carried out with 10 mM ketone substrate,
20 mM glucose, 0.2 mM NADP+, 1 mg mL−1 KRED (24.7 μM YGL039w or
33.2 μM SyADH) and 1 mg mL−1 glucose dehydrogenase (GDH) (24.3 μM)
in 1 mL of 50 mM citric acid, pH 5.0 at 30 °C and 180 rpm for 1 h. [a] The
reaction conversion was determined using 1H NMR. [b] The ee was
determined using chiral HPLC. [c] The absolute configuration of 5c was
determined by 1H NMR spectroscopy using Mosher’s reagent (see
Fig. S5 and S6, and the associated discussion in ESI† for details); the
absolute configuration of the rest of compounds (5a, 5b, 5d to 5j) was
assigned by analogy.
To demonstrate the synthetic potential of our developed
system, the bioreduction of select α-nitro ketones (1a, 1i, 1k,
4a to 4j) was carried out in preparative scale (250 mg ketone
substrates, 5 g L−1 concentration). DMSO was added to 10%
(v/v) to help dissolve the hydrophobic substrates. Using cell-
free extracts (CFE) of KREDs as biocatalysts, ketone substrates respectively (entry 1, Table S15†). In contrast, substantially
were completely consumed, and the desired β-nitro alcohols decreased conversions were seen in reactions with higher sub-
were isolated in 42%-to-90% yields with 32% ee-to-99% ee strate loadings (<71% and <21% for 20 and 50 g L−1, respect-
(Tables S12–S14†). Notably, enantioenriched alcohols (R)-2k, ively). We conjectured that using other organic solvents might
(S)-2k, (S)-5c, (S)-5d, and (S)-5j thus prepared are key synthetic combat potential substrate or product inhibition;15
intermediates to the hypoglycemic natural product (R)-temba- 16 additional organic solvents (10% v/v), both water-miscible
mide, the antiviral natural product (S)-tembamide, and the and water-immiscible, were examined in bioreduction
β-adrenergic receptor blocking agents (S)-moprolol, (S)-tolipro- reactions carried out at a substrate concentration of 50 g L−1
lol, and (S)-propanolol (Fig. 1).6b,14
(Table S16†). Disappointingly, only limited improvement was
In order to further demonstrate the practical utility of these realized, with the highest conversion being 47% (entry 5,
KRED-catalyzed bioreductions, we sought to increase the sub- Table S16†).
strate loadings of the above processes. At a substrate concen-
An alternative form of biocatalysis,16 the use of resting
tration of 10 g L−1 (ketone 1k), 89% and >99% conversions E. coli whole cells coexpressing RasADH and GDH was also
were achieved for YGL039w- and RasADH-catalyzed reactions, explored for the reduction of 1k (Table S17†). Both two gene-
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Paper
two plasmid and two gene-one plasmid approaches were
adopted for the construction of the required recombinant
Acknowledgements
E. coli strains. We first evaluated the biocatalyst prepared using We thank Dr Abraham Wang (MIT) for proofreading this
the former approach. In the presence of 10% DMSO (v/v), the manuscript. We thank Prof. Ping Lu and Mr Changxu Zhong
reaction, conducted at a substrate concentration of 50 g L−1
,
for the help of determining the ee of compound 2o. Financial
reached 54% conversion after 5 h (entry 1, Table S17†). The supports from the National Natural Science Foundation of
inclusion of other organic solvents (20% v/v) positively influ- China (no. 21801047) and Shanghai Sailing Program
enced the reaction conversions (up to 80%) (entries 4 to 11, (18YF1402100) are greatly appreciated.
Table S17†). The three best-performing solvents (acetone, di-
ethylene glycol dimethyl ether (DGDE), and dioxane) were then
examined at a concentration of 50% (v/v). To our delight, the
use of dioxane at 50% (v/v) resulted in further increased con-
Notes and references
version (86%) (entry 14, Table S17†). Under such optimized
conditions, 1k (250 mg) was readily converted to (R)-2k
(111 mg, 44% isolated yield) in a 5 mL reaction after 3 h,
giving a space–time yield of 178 g L−1 d−1. Finally, the whole-
cell catalysts prepared through the two gene-one plasmid
method were subjected to the reduction of 1k as well. As only
moderate conversions were made at substrate concentration of
75 g L−1 (entries 15 and 16, Table S17†), no further efforts
were attempted. It is worth noting that such coexpression
approaches are not applicable to most of commercial KREDs,
at least for most researchers, since the associated genetic
manipulation requires knowing the gene/protein sequences.
1 (a) F. Qin, B. Qin, T. Mori, Y. Wang, L. Meng, X. Zhang,
X. Jia, I. Abe and S. You, ACS Catal., 2016, 6, 6135–6140;
(b) W. Kroutil, H. Mang, K. Edegger and K. Faber, Curr.
Opin. Chem. Biol., 2004, 8, 120–126; (c) R. Zhang, Y. Xu and
R. Xiao, Biotechnol. Adv., 2015, 33, 1671–1684.
2 (a) S. K. Ma, J. Gruber, C. Davis, L. Newman, D. Gray,
A. Wang, J. Grate, G. W. Huisman and R. A. Sheldon, Green
Chem., 2010, 12, 81–86; (b) J. Liang, J. Lalonde, B. Borup,
V. Mitchell, E. Mundorff, N. Trinh, D. A. Kochrekar,
R. N. Cherat and G. G. Pai, Org. Process Res. Dev., 2010, 14,
193–198.
3 S. Zhang, Y. Li, Y. Xu and Z. Wang, Chin. Chem. Lett., 2018,
29, 873–883.
4 (a) H. Ankati, Y. Yang, D. Zhu, E. R. Biehl and L. Hua,
J. Org. Chem., 2008, 73, 6433–6436; (b) G. Xu, H. Yu,
Z. Zhang and J. Xu, Org. Lett., 2013, 15, 5408–5411;
(c) R. W. Nowill, T. J. Patel, D. L. Beasley, J. A. Alvarez,
E. Jackson III, T. J. Hizer, I. Ghiviriga, S. C. Mateer and
B. D. Feske, Tetrahedron Lett., 2011, 52, 2440–2442;
(d) Y. Geng, R. Zhang, Y. Xu, S. Wang, C. Sha and R. Xiao,
Biocatal. Biotransform., 2011, 29, 172–178; (e) G. Xu, H. Yu,
X. Zhang and J. Xu, ACS Catal., 2012, 2, 2566–2571.
5 J. Otevrel and P. Bobal, J. Org. Chem., 2017, 82, 8342–8358.
6 (a) J. H. Schrittwieser, F. Coccia, S. Kara, B. Grischek,
W. Kroutil, N. d’Alessandro and F. Hollmann, Green Chem.,
2013, 15, 3318–3331; (b) J. D. White and S. Shaw, Org. Lett.,
2012, 14, 6270–6273.
Conclusions
In summary, YGL039w and RasADH/SyADH were identified as
stereocomplementary KREDs through a campaign of screening
24 KREDs with publicly known sequences on the bioreduction
of α-nitro ketones. We show that 23 class I ketones (1-aryl-2-
nitro-1-ethanone (1)) and ten class II ketones (1-aryloxy-3-nitro-
2-propanone (4)) were successfully bioreduced to generate
both enantiomers of the corresponding β-nitro alcohols, with
good-to-excellent conversions and stereoselectivities being
achieved in most cases. To the best of our knowledge, this is
the first report of a ketoreductase-catalyzed reduction of
1-aryloxy-3-nitro-2-propanone (4). The synthetic potential of
our system was showcased by the highly stereoselective synth-
eses of key intermediates to (R)-tembamide, (S)-tembamide,
(S)-moprolol, (S)-toliprolol, and (S)-propanolol in preparative
scales. Finally, by employing whole cells of a recombinant
E. coli strain coexpressing RasADH and GDH as the biocatalyst,
the β-nitro alcohol (R)-2k, the key synthetic intermediate to
hypoglycemic natural product (R)-tembamide, was synthesized
with a space–time yield of 178 g L−1 d−1 as with 95% ee, under-
scoring the superior applicability of KREDs with publicly
known sequences. We believe that our catalytic system will
find wide usage in the preparation of valuable chiral β-nitro
alcohols.
7 D. J. Ager, I. Prakash and D. R. Schaad, Chem. Rev., 1996,
96, 835–876.
8 (a) T. Fujisawa, H. Hayashi and Y. Kishioka, Chem. Lett.,
1987, 16, 129–132; (b) E. C. S. Brenelli, M. de Carvalho,
M. T. Okubo, M. Marques, P. J. S. Moran, J. A. R. Rodrigues
and A. E. P. M. Sorrilha, Indian J. Chem., 1992, 31B, 821–
823; (c) S. R. Wallner, I. Lavandera, S. F. Mayer, R. Öhrlein,
A. Hafner, K. Edegger, K. Faber and W. Kroutil, J. Mol.
Catal. B: Enzym., 2008, 55, 126–129; (d) S. Venkataraman
and A. Chadha, RSC Adv., 2015, 5, 73807–73813.
9 F. Tentori, E. Brenna, D. Colombo, M. Crotti, F. G. Gatti,
M. C. Ghezzi and G. Pedrocchi-Fantoni, Catalysts, 2018, 8,
308–318.
10 H. Li, D. Zhu, L. Hua and E. R. Biehl, Adv. Synth. Catal.,
2009, 351, 583–588.
11 R. G. Pearson, D. H. Anderson and L. L. Alt, J. Am. Chem.
Soc., 1955, 77, 527–529.
Conflicts of interest
There are no conflicts to declare.
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
12 I. Lavandera, A. Kern, B. Ferreira-Silva, A. Glieder, S. de 15 (a) M. A. F. Delgove, M. T. Elford, K. V. Bernaerts and
Wildeman and W. Kroutil, J. Org. Chem., 2008, 73, 6003–
6005.
13 A. S. de Miranda, R. C. Simon, B. Grischek, G. C. de Paula,
B. A. C. Horta, L. S. M. de Miranda, W. Kroutil, C. O. Kappe
and R. O. M. A. de Souza, ChemCatChem, 2015, 7, 984–992.
S. M. A. De Wildeman, J. Chem. Technol. Biotechnol., 2018,
93, 2131–2140; (b) S. K. Au, B. R. Bommarius and
A. S. Bommarius, ACS Catal., 2014, 4, 4021–4026;
(c) F. G. Mutti and W. Kroutil, Adv. Synth. Catal., 2012, 354,
3409–3413.
14 D. J. Buchanan, D. J. Dixon, M. S. Scott and D. I. Lainé, 16 R. A. Sheldon and J. M. Woodley, Chem. Rev., 2018, 118,
Tetrahedron: Asymmetry, 2004, 15, 195–197.
801–838.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2019