Extended Catalytic Scope of a Well-Known Enzyme: Asymmetric Reduction of Iminium Substrates by Glucose Dehydrogenase

Article abstract of DOI:10.1002/cbic.201700261

NADP(H)-dependent imine reductases (IREDs) are of interest in biocatalytic research due to their ability to generate chiral amines from imine/iminium substrates. In reaction protocols involving IREDs, glucose dehydrogenase (GDH) is generally used to regenerate the expensive cofactor NADPH by oxidation of d-glucose to gluconolactone. We have characterized different IREDs with regard to reduction of a set of bicyclic iminium compounds and have utilized ¹H NMR and GC analyses to determine degree of substrate conversion and product enantiomeric excess (ee). All IREDs reduced the tested iminium compounds to the corresponding chiral amines. Blank experiments without IREDs also showed substrate conversion, however, thus suggesting an iminium reductase activity of GDH. This unexpected observation was confirmed by additional experiments with GDHs of different origin. The reduction of C=N bonds with good levels of conversion (>50 %) and excellent enantioselectivity (up to >99 % ee) by GDH represents a promiscuous catalytic activity of this enzyme.

Full text of DOI:10.1002/cbic.201700261

DOI: 10.1002/cbic.201700261
Communications
Extended Catalytic Scope of a Well-Known Enzyme:
Asymmetric Reduction of Iminium Substrates by Glucose
Dehydrogenase
Sebastian Roth+,[a] Andreas Prꢀg+,[a] Cindy Wechsler,[a] Marija Marolt,[a] Sascha Ferlaino,[a]
Steffen Lꢁdeke,[a] Nicolas Sandon,[b] Dennis Wetzl,[b] Hans Iding,[b] Beat Wirz,[b] and
Michael Mꢁller*[a]
NADP(H)-dependent imine reductases (IREDs) are of interest in
biocatalytic research due to their ability to generate chiral
amines from imine/iminium substrates. In reaction protocols
involving IREDs, glucose dehydrogenase (GDH) is generally
used to regenerate the expensive cofactor NADPH by oxida-
tion of d-glucose to gluconolactone. We have characterized
different IREDs with regard to reduction of a set of bicyclic imi-
acids, with most such proteins existing as homodimers or
homotetramers.[3–5]
Although pluripotency/promiscuity can be regarded as a typ-
ical feature of some SDRs involved in the metabolism of xeno-
biotics, many biological functions are controlled by SDRs that
are assumed to be highly substrate-specific and monofunction-
al. Glucose dehydrogenase (GDH), a prototypical member of
the SDR family, for example, was previously thought solely to
catalyze NAD(P)-dependent oxidation of d-glucose to glucono-
lactone.[6] Until recently, GDH was not known to be active with
non-sugar substrates.[7]
1
nium compounds and have utilized H NMR and GC analyses
to determine degree of substrate conversion and product
enantiomeric excess (ee). All IREDs reduced the tested iminium
compounds to the corresponding chiral amines. Blank experi-
ments without IREDs also showed substrate conversion, how-
ever, thus suggesting an iminium reductase activity of GDH.
This unexpected observation was confirmed by additional ex-
periments with GDHs of different origin. The reduction of C=N
bonds with good levels of conversion (>50%) and excellent
enantioselectivity (up to >99% ee) by GDH represents a pro-
miscuous catalytic activity of this enzyme.
GDH is a component of a well-established method for nico-
tinamide cofactor regeneration.[8] Thus, this enzyme is often
applied to provide NAD(P)H for the characterization of, for ex-
ample, imine reductases (IREDs), a recently discovered group
of enzymes identified as catalyzing the asymmetric reduction
of cyclic or (generated in situ) open-chain imines.[9,10] IREDs
have a broader substrate range than imine- or iminium-reduc-
ing enzymes involved in primary metabolism or alkaloid bio-
synthesis and have therefore been investigated as potential
biocatalytic tools for an approach to chiral amines.[11]
In general, a promiscuous enzyme can be defined as one “that
does things it is not expected to do”.[1] In this context, the
term “pluripotent” describes members of the family of short-
chain dehydrogenases/reductases (SDRs) that are capable of
accepting more than their physiological substrate.[2] Spanning
several EC classes, the SDR superfamily constitutes one of the
largest enzyme families, with more than 46000 members.[3,4]
Unrelated SDR members usually share very low protein se-
quence similarity, often as low as 15–30% in pairwise compari-
son. Even so, SDR proteins possess highly similar three-dimen-
sional structures, an a/b-folding pattern (Rossmann-fold motif)
for nucleotide binding, and chain lengths of 250–350 amino
Here we describe the serendipitous identification of GDHs
from different organisms as catalysts for the asymmetric re-
duction of iminium salts. Prerequisite for this finding was the
characterization of purified IREDs, thus excluding background
reactions, and the selection of appropriate substrates.
Although IREDs have been extensively characterized for the
reduction of cyclic imines, there are few reports on the reduc-
tion of iminium salts.[12,13] We aimed to characterize R- and S-
selective IREDs for the reduction of cyclic iminium salts. Start-
ing from the amino acid sequences of known IREDs as a
template, two putative IREDs from Streptomyces virginiae were
chosen, in addition to 17 already described IREDs from differ-
ent microorganisms.[14,15] The genes were purchased as syn-
thetic codon-optimized genes with an additional N-terminal
His-tag and overexpressed in Escherichia coli BL21-Gold (DE3)
cells (see the Supporting Information). Subsequently, all IREDs
were purified by Ni-NTA affinity chromatography. IRED activity
of the two newly identified enzymes from S. virginiae was
confirmed by the reduction of 2-methyl-4,5-dihydro-1-pyrroline
(data not shown). The substrate range and reaction stereo-
selectivity of the 19 bacterial IREDs towards a set of isoquinoli-
nium substrates 1–3 were determined with use of d-glucose/
GDH for NADPH regeneration (Scheme 1).
[a] S. Roth,+ Dr. A. Prꢀg,+ Dr. C. Wechsler, M. Marolt, S. Ferlaino, Dr. S. Lꢁdeke,
Prof. Dr. M. Mꢁller
Institut fꢁr Pharmazeutische Wissenschaften
Albert-Ludwigs-Universitꢀt Freiburg
Albertstrasse 25, 79104 Freiburg (Germany)
E-mail: michael.mueller@pharmazie.uni-freiburg.de
[b] Dr. N. Sandon, Dr. D. Wetzl, Dr. H. Iding, Dr. B. Wirz
Process Chemistry and Catalysis, F. Hoffmann–La Roche, Ltd.
4070 Basel (Switzerland)
[+] These authors contributed equally to this work.
Supporting information and the ORCID identification numbers for the
authors of this article can be found under https://doi.org/10.1002/
cbic.201700261.
ChemBioChem 2017, 18, 1 – 5
1
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
Scheme 1. Reduction of the iminium compounds 1–3 by IRED or GDH. Re-
action conditions: 0.6 mgmLÀ1 purified enzyme, 10 mm substrate (1, 2, or 3),
20 mm d-glucose, 0.5 mm NADP+, 2 mm MgCl2, 0.26 mgmLÀ1 GDH in
100 mm HEPES buffer, pH 7.5, 308C, 20 h.
A 1H NMR-based assay was used for determination of the en-
zymatic activity and product formation immediately after re-
moval of the proteins by filtration (10 kDa cutoff). This method
facilitated the simultaneous identification of the hydrophilic
substrates and lipophilic products in the crude reaction mix-
ture. The aromatic signals of the substrates and products, and
the signals of the C-1 methyl group (in the case of products
4–6 a doublet in the aliphatic region), were clearly separated
from each other and from those of other contaminants (see
the Supporting Information). Furthermore, this procedure en-
abled concurrent qualitative and semiquantitative determina-
tion of the degree of conversion.
Figure 1. Experimentally measured IR and VCD spectra in comparison with
spectra calculated for (R)-4 at the B3LYP/6–31+G(d,p) level in Gaussian 09.[16]
The excellent agreement in positions and signs of the experimentally mea-
sured and calculated VCD bands confirms the R configuration.
Compounds 1–3 were accepted by all tested IREDs (Table S7
in the Supporting Information). Several IREDs, such as IR_4
from Kribbella flavida (WP_012921542.1) and IR_20 from Strep-
tomyces tsukubaensis (WP_006347397.1), showed acceptable
levels of conversion (27–74%) for each substrate. The absolute
configuration of product 4 was determined by vibrational cir-
cular dichroism (VCD). For this purpose, the transformation of
1 in the presence of IR_20 was performed on a preparative
scale (50 mmol), yielding oily 4. A VCD spectrum of the neat
liquid was recorded and compared with the calculated spec-
trum of (R)-4, thus establishing the absolute R configuration
(Figure 1). With the assumption of an identical reduction
mechanism and, additionally, from the uniform retention time
order in the chiral-phase gas chromatograms (later eluting
enantiomer; see the Supporting Information), the R configura-
tion is postulated for the products of IR_20. Accordingly, IR_4
gave access to the S enantiomers of products 4–6 (earlier elut-
ing enantiomers; Tables S8 and S9).
ee values provide evidence of asymmetric reduction of the pro-
chiral substrates.
To elucidate the enantioselective background reaction, addi-
tional control experiments with substrate 3, which showed the
highest “background reaction” in the NMR assay, were per-
formed. The presence of equimolar or 2 equivalents of NADPH,
without any enzyme, gave 4 and 7% conversion, respectively.
As expected, the product 6 was obtained in almost racemic
form in both transformations (ee<20%). Hence, the noted
asymmetric induction must arise from enzyme control.
Accordingly, GDH-catalyzed transformation of substrates 1–3
was tested with increased amounts of GDH (0.023 mgmLÀ1
,
0.23 mgmLÀ1, 1.15 mgmLÀ1), in the absence of any IRED. The
results (Table 1) clearly showed a correlation between enzyme
concentration and conversion along with product ee. The de-
creased ee values of the products at lower enzyme concentra-
tion can be explained in terms of a higher contribution of non-
enzymatic reduction by NADPH.
The enantiomeric excesses (ee values) of the enzymatic
products 4–6 were determined by chiral-phase GC analysis
(Figure S11 and Table S9). Many of the transformations with
high levels of conversion also revealed high ee values for the
products. Surprisingly, significant levels of conversion (9 and
20%) were also obtained for substrates 2 and 3, as well as
high ee values for the products 5 and 6 (83 and 84% ee), in
the blank experiments without IRED enzymes. Although non-
specific reduction by the (chiral) cofactor NADPH could, per se,
explain product formation in the blank experiments, this
should result in products with only slight ee. Instead, the high
With the highest enzyme concentration tested, the reduc-
tion of 3 led to virtually enantiopure product (R)-6 (>99% ee),
with 57% conversion. Moreover, this result explains the moder-
ate to low ee values (<50%) obtained for products 5 and 6 in
assays with S-selective enzymes, as the accumulation of the R-
configured products from the more or less pronounced GDH
background reaction would partially cancel the gain from IRED
catalysis. Correspondingly, replacement of the GDH/glucose
regeneration system should result in higher ee values of the
&
ChemBioChem 2017, 18, 1 – 5
www.chembiochem.org
2
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Communications
tested enzyme, such as an IRED. Therefore, experiments relat-
ing to the reduction of reactive iminium compounds in which
IREDs are applied alongside GDH should be validated with re-
spect to the cofactor regeneration system, particularly in cases
in which crude cell IRED extracts or whole-cell cofactor regen-
eration are used.
Table 1. Conversion of 1, 2, and 3 by glucose dehydrogenase (GDH) and
resulting enantiomeric excesses (ee values) of the products.[a]
Substrate
Product
GDH[b]
conc.
[mgmLÀ1
Conv. ee [%]
[%]
(absolute
]
config.)
Our finding for GDH is in line with a recent example of pro-
miscuous SDR activity: Kutchan and co-workers discovered a
C=C-bond-reducing SDR with an imine reductase side activi-
ty.[19] Moreover, in Catharanthus roseus a medium-chain alcohol
dehydrogenase homologue that possesses iminium-reducing
activity has been found.[20] Conversely, Nestl and co-workers
most recently proposed a carbonyl reductase (side) activity by
two IREDs exposed to a highly reactive ketone substrate.[21]
Nevertheless, the observation that both GDH and IREDs
reduce iminium compounds cannot easily be explained. Mem-
bers of both enzyme types are oxidoreductases and share the
Rossmann fold for cofactor binding as a common structural
feature. Comparison of GDH with known and structurally ana-
lyzed IREDs (data not shown), however, does not reveal any
major similarity. Sequence analysis indicated that 17 of the 19
tested IREDs exhibit the active-site motif suggested by Pleiss
and co-workers.[22] Nonetheless, the exact role of these resi-
dues or even their participation in the reduction of imine/imi-
nium substrates has not been fully elucidated. In contrast,
GDH has the SDR-typical catalytic triad Ser-Tyr-Lys,[23] whereas
IREDs lack a lysine residue in their active site, highlighting the
differences between these enzyme types. Moreover, the postu-
lated mechanism of reduction by dihydrofolate reductase[24] in-
cludes activation of the imine through protonation by a water
molecule, which might be another clue that, in general, the
main function of imine-reducing enzymes is to provide close
proximity for hydride transfer.[22] Thus, reduction of the tested
iminium compounds by IREDs, as well as by GDH, cannot be
explained in terms of simple structural characterization and/or
sequence alignment without implementation of other factors,
such as dynamic contributions or substrate–enzyme interac-
tions. Rather, our results and the literature examples noted
indicate that the focus on IREDs should not lead to the neglect
of SDRs and other NAD(P)H-dependent oxidoreductases as
templates for the identification of new imine-reducing en-
zymes.
0.023
0.23
1.15
0
0
9
–
–
n.d.[c]
0.023
0.23
1.15
4
20
28
28 (R)
68 (R)
86 (R)
0.023
0.23
1.15
9
16
57
86 (R)
97 (R)
>99 (R)
[a] Reaction conditions: 10 mm substrate (1, 2, or 3), 20 mm d-glucose,
0.5 mm NADP+, 2 mm MgCl2 in 100 mm HEPES buffer, pH 7.5, 308C, 20 h.
[b] GDH-105 (Codexis). [c] Not detected.
S products under such IRED catalysis conditions. IR_10, which
showed the most prominent S selectivity of the tested IREDs,
was chosen for further biotransformations with a stoichiometric
supply of NADPH, instead of the regeneration system. The sub-
stantially improved ee values of the formed products (S)-5 and
(S)-6 confirmed our hypothesis (Table S13).
To verify the unexpected results, purified GDH from Bacillus
subtilis (evocatal, Monheim am Rhein) was tested (see the Sup-
porting Information). These experiments confirmed the asym-
metric induction and enzymatic transformation of the iminium
salts by GDH, resulting in high ee values of product (R)-6 (86–
99% ee) with up to 57% conversion. In experiments with high
concentrations of commercial GDH preparations [2.6 mgmLÀ1
GDH-105 (Codexis, Redwood City, CA) or GDH-2 (Roche)] the
reduction of imines (non-N-methylated) was observed as well
(see the Supporting Information).
In summary, we have shown that all tested IREDs, including
the two newly identified enzymes from S. virginiae, are capable
of reducing the bicyclic iminium compounds 1–3. The high re-
activity of the iminium salts towards reduction was probably
helpful for elucidation of the entirely unexpected asymmetric
reduction of a C=N bond by GDH, usually regarded as a “classi-
cal” NADPH-dependent SDR.[17] Hence, we have demonstrated
that the substrate scope of GDH is broader than initially sus-
pected, underscoring our previous finding of GDH-catalyzed
reduction of naphthoquinone derivatives.[7,18] The enantioselec-
tivity of the iminium reduction is strikingly high, even being
higher than for many IRED-catalyzed transformations.
Experimental Section
GDHs of different origin were tested on an analytical (0.5 mL) scale.
GDH-105 (Codexis) was applied from a freshly prepared stock solu-
tion [2.33 mgmLÀ1 in HEPES buffer (100 mm, pH 7.5)], and GDH
from Bacillus subtilis (evocatal) as a crude cell lysate or as the puri-
fied enzyme. The substrate was added from a stock solution (1m
in MeOH) to different amounts of enzyme, resulting in a 10 mm
substrate concentration. Each reaction was started by addition of
a mixture (250 mL) containing the other reaction components
dissolved in HEPES buffer (pH 7.5, 100 mm). The reaction mixture
consisted of GDH, substrate 1, 2, or 3 (10 mm), d-glucose (20 mm),
NADP+ (0.5 mm), and MgCl2 (2 mm). After incubation for 20 h at
308C and 850 rpm, the reaction was stopped by removing the
enzyme (filtration with 10 kDa cutoff or centrifugation after heat-
As the use of GDH, even in small amounts for cofactor re-
generation, resulted in the unambiguous conversion of imini-
um compounds 2 and 3 into the amine products with high ee
values, the d-glucose/GDH cofactor regeneration system has to
be considered non-innocent. Thus, it could have a significant
influence on the apparent results achieved with a primarily
ChemBioChem 2017, 18, 1 – 5
www.chembiochem.org
3
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Communications
[11] J. H. Schrittwieser, S. Velikogne, W. Kroutil, Adv. Synth. Catal. 2015, 357,
1655–1685.
[12] S. Hussain, F. Leipold, H. Man, E. Wells, S. P. France, K. R. Mulholland, G.
Grogan, N. J. Turner, ChemCatChem 2015, 7, 579–583.
[13] H. Li, G.-X. Zhang, L.-M. Li, Y.-S. Ou, M.-Y. Wang, C.-X. Li, G.-W. Zheng, J.-
H. Xu, ChemCatChem 2016, 8, 724–727.
ing). Positive and negative controls were undertaken. Degrees of
conversion were determined by 1H NMR spectroscopy after addi-
tion of D2O (10%). For chiral-phase GC analysis, the reaction mix-
tures were basified with NaOH (8m) and extracted twice with
EtOAc.
[14] N. Itoh, C. Yachi, T. Kudome, J. Mol. Catal. B 2000, 10, 281–290.
[15] D. Wetzl, M. Berrera, N. Sandon, D. Fishlock, M. Ebeling, M. Mꢁller, S. P.
Hanlon, B. Wirz, H. Iding, ChemBioChem 2015, 16, 1749–1756.
[16] Gaussian 09, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Pe-
tersson, H. Nakatsuji, X. Li, M. Caricato, A. Marenich, J. Bloino, B. G. Jane-
sko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov,
J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J.
Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzew-
ski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, T. Vreven, K. Throssell, J. A. Montgomery, Jr., J. E. Peralta, F.
Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov,
T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C.
Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C.
Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas,
J. B. Foresman, D. J. Fox, Gaussian, Inc., Wallingford, CT, 2009.
[17] Y. Kallberg, U. Oppermann, H. Jçrnvall, B. Persson, Eur. J. Biochem. 2002,
269, 4409–4417.
Acknowledgements
We thank David Conradt for providing purified GDH (evocatal),
Daniel Becker for technical support, David Conradt, Lydia Walter,
and Marcel Wilde for helpful discussions, and Dr. Kay Greenfield
for help in improving the manuscript. This research was support-
ed in part by the bwHPC initiative and the bwHPC-C5 project pro-
vided through associated computing services of the JUSTUS HPC
facility at the University of Ulm.
Keywords: biotransformations · enantioselectivity · enzyme
promiscuity · oxidoreductases · tertiary amines
[1] K. Hult, P. Berglund, Trends Biotechnol. 2007, 25, 231–238.
[2] F. Hoffmann, E. Maser, Drug Metab. Rev. 2007, 39, 87–144.
[3] C. Filling, K. D. Berndt, J. Benach, S. Knapp, T. Prozorovski, E. Nordling, R.
Ladenstein, H. Jçrnvall, U. Oppermann, J. Biol. Chem. 2002, 277, 25677–
25684.
[18] S. M. Husain, M. A. Schꢀtzle, S. Lꢁdeke, M. Mꢁller, Angew. Chem. Int. Ed.
2014, 53, 9806–9811; Angew. Chem. 2014, 126, 9964–9969.
[19] M. B. Kilgore, C. K. Holland, J. M. Jez, T. M. Kutchan, J. Biol. Chem. 2016,
291, 16740–16752.
[4] B. Persson, Y. Kallberg, J. E. Bray, E. Bruford, S. L. Dellaporta, A. D. Favia,
R. G. Duarte, H. Jçrnvall, K. L. Kavanagh, N. Kedishvili, M. Kisiela, E.
Maser, R. Mindnich, S. Orchard, T. M. Penning, J. M. Thornton, J. Adam-
ski, U. Oppermann, Chem.-Biol. Interact. 2009, 178, 94–98.
[5] K. L. Kavanagh, H. Jçrnvall, B. Persson, U. Oppermann, Cell. Mol. Life Sci.
2008, 65, 3895–3906.
[20] A. Stavrinides, E. C. Tatsis, E. Foureau, L. Caputi, F. Kellner, V. Courda-
vault, S. E. O’Connor, Chem. Biol. 2015, 22, 336–341.
[21] M. Lenz, J. Meisner, L. Quertinmont, S. Lutz, J. Kꢀstner, B. Nestl, Chem-
BioChem 2017, 18, 253–256.
[22] S. Fademrecht, P. N. Scheller, B. M. Nestl, B. Hauer, J. Pleiss, Proteins
Struct. Funct. Bioinf. 2016, 84, 600–610.
[6] S. Ferri, K. Kojima, K. Sode, J. Diabetes Sci. Technol. 2011, 5, 1068–1076;
other carbohydrates are also known substrates of GDH.
[7] D. Conradt, M. A. Schꢀtzle, S. M. Husain, M. Mꢁller, ChemCatChem 2015,
7, 3116–3120.
[23] K. Yamamoto, G. Kurisu, M. Kusunoki, S. Tabata, I. Urabe, S. Osaki, J. Bio-
chem. 2001, 129, 303–312.
[24] C. M. Czekster, A. Vandemeulebroucke, J. S. Blanchard, Biochemistry
2011, 50, 367–375.
[8] C.-H. Wong, D. G. Drueckhammer, H. M. Sweers, J. Am. Chem. Soc. 1985,
107, 4028–4031.
[9] K. Mitsukura, M. Suzuki, S. Shinoda, T. Kuramoto, T. Yoshida, T. Naga-
sawa, Biosci. Biotechnol. Biochem. 2011, 75, 1778–1782.
[10] T. Huber, L. Schneider, A. Prꢀg, S. Gerhardt, O. Einsle, M. Mꢁller, Chem-
CatChem 2014, 6, 2248–2252.
Manuscript received: May 16, 2017
Version of record online: && &&, 0000
&
ChemBioChem 2017, 18, 1 – 5
www.chembiochem.org
4
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

COMMUNICATIONS
S. Roth, A. Prꢀg, C. Wechsler, M. Marolt,
S. Ferlaino, S. Lꢁdeke, N. Sandon,
D. Wetzl, H. Iding, B. Wirz, M. Mꢁller*
&& – &&
Doing it all: Glucose dehydrogenase
(GDH) was applied as a cofactor regen-
eration system in the characterization of
imine reductases (IREDs) in the reduc-
tion of bicyclic iminium compounds.
Both enzyme types were found to be
capable of reducing such substrates.
Extended Catalytic Scope of a Well-
Known Enzyme: Asymmetric
Reduction of Iminium Substrates by
Glucose Dehydrogenase
ChemBioChem 2017, 18, 1 – 5
www.chembiochem.org
5
ꢂ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ