Addition of a polyhistidine tag alters the regioselectivity of carbonyl reductase S1 from: Candida magnoliae

Article abstract of DOI:10.1039/c7ob02666h

Studying enzymatic reductions of substrates with more than a single keto group is challenging, as the carbonyl reduction can create a vast array of regio- and stereoisomers. If used as reference compounds, regio- and stereopure hydroxy ketides could facilitate the characterization of reductases with unclear regio- and stereoselectivity. We have combined nonenzymatic and enzymatic reduction and oxidation steps to obtain all four regio- and stereoisomers of tert-butyl hydroxyoxohexanoates in high optical purity (enantiomeric ratio (er) of 99:1 for the δ-hydroxy-β-keto isomers; er of >97:3 for the β-hydroxy-δ-keto isomers). Furthermore, we have prepared seven of the eight possible regioisomers and diastereomers of γ-methylated hydroxyoxohexanoates. These 11 compounds allowed unraveling the complex stereoselectivity of β,δ-diketo ester reductions catalyzed by carbonyl reductase S1 from Candida magnoliae (CMCR-S1). Our analysis shows that the regio- and stereoselectivity of CMCR-S1-catalyzed reductions is highly sensitive toward modifications at the C-terminus of CMCR-S1: in addition to the expected δ-hydroxy product, the variant with a C-terminal His-tag also led to formation of β-hydroxy by-products with high optical purity.

Full text of DOI:10.1039/c7ob02666h

View Article Online
View Journal
Organic &
Biomolecular
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: J. Haas, M. Häckh,
V. Justus, M. Muller and S. Lüdeke, Org. Biomol. Chem., 2017, DOI: 10.1039/C7OB02666H.
This is an Accepted Manuscript, which has been through the
Volume 14 Number
1 7 January 2016 Pages 1–372
Royal Society of Chemistry peer review process and has been
accepted for publication.
Organic &
Biomolecular
Chemistry
www.rsc.org/obc
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-0520
COMMUNICATION
Takeharu Haino et al.
Solvent-induced emission of organogels based on tris(phenylisoxazolyl)
benzene
rsc.li/obc

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 1 of 10
View Article Online
DOI: 10.1039/C7OB02666H
Journal Name
ARTICLE
Addition of a polyhistidine tag alters the regioselectivity of carbonyl reductase
S1 from Candida magnoliae
Julian Haas,a Matthias Häckh,a Viktor Justus,a Michael Müllera and Steffen Lüdeke*a
Received 00th January 20xx,
Accepted 00th January 20xx
Studying enzymatic reductions of substrates with more than a single keto group is challenging, as the carbonyl reduction
can create a vast array of regio- and stereoisomers. If used as reference compounds, regio- and stereopure hydroxy
ketides could facilitate the characterization of reductases with unclear regio- and stereoselectivity. We have combined
nonenzymatic and enzymatic reduction and oxidation steps to obtain all four regio- and stereoisomers of tert-butyl
hydroxyoxohexanoates in high optical purity (enantiomeric ratio (er) of 99:1 for the δ-hydroxy-β-keto isomers; er of >97:3
for the β-hydroxy-δ-keto isomers). Furthermore, we have prepared seven of the eight possible regioisomers and
diastereomers of γ-methylated hydroxyoxohexanoates. These 11 compounds allowed unraveling the complex
stereoselectivity of β,δ-diketo ester reductions catalyzed by carbonyl reductase S1 from Candida magnoliae (CMCR-S1).
Our analysis shows that the regio- and stereoselectivity of CMCR-S1-catalyzed reduction is highly sensitive toward
modifications at the C-terminus of CMCR-S1: in addition to the expected δ-hydroxy product, the variant with a C-terminal
DOI: 10.1039/x0xx00000x
www.rsc.org/
His-tag
also
led
to
formation
of
β-hydroxy
by-products
with
high
optical
purity.
the number of possible product isomers, since the alkyl side
chain can be either syn- or anti-orientated with respect to the
newly formed hydroxyl group. Highly stereoselective
enzymatic reductions of 1a and 1b to the corresponding δ-
hydroxy-β-keto esters 2a and 2b4-6 allow for nonenzymatic
diastereoselective reduction at the β-position.7 Alternatively,
several biocatalytic reactions lead to β,δ-dihydroxy products.8-
Introduction
When a single carbonyl group of a polycarbonyl precursor is
reduced, there are numerous possibilities for stereo- and/or
regioisomers of the resulting alcohols. This concept is, for
example, exploited in the biosynthesis of polyketides to create
diversity in this class of highly complex natural products with
alternating keto or chiral hydroxyl groups, the latter formed by
enzymatic keto reduction.
11
In contrast, precedence for exclusive direct enzymatic β-
reduction of triketide-derived substrates is rare.4 Therefore,
when trying to synthesize triketides selectively, chemists face a
dilemma: i) the lack of catalysts (including enzymes) allowing a
versatile and highly regio- and stereoselective synthesis of β-
hydroxy-δ-keto esters; ii) the lack of enantiopure reference
compounds hampering the identification and characterization
of selective catalysts such as enzymes with β-reductase
activity.
Keto reductases are generally categorized as R- or S-selective
according to the observed stereochemical outcome in a model
reaction. However, studying the transformation of short-chain
model substrates does not provide a complete picture of
enzymatic selectivity and specificity: an enzyme may show
opposite selectivity for small substrates, such as keto esters (a
diketide in the context of polyketide biosynthesis), relative to
longer-chain derivatives (triketides, tetraketides, etc.).3 In the
latter case, the additional carbonyl groups even allow different
regioselectivities. Therefore, longer-chain substrate models
could help to unravel hitherto unknown specificities of an
enzyme.
In this work, we solve this problem by presenting an enzyme-
aided stereoselective route to triketides. We employ regio-
and stereoselective oxidation catalyzed by NAD(P)+-dependent
alcohol dehydrogenases (ADH) to obtain different
stereoisomers of tert-butyl 3-hydroxy-5-oxohexanoate (3a
)
and tert-butyl 3-hydroxy-4-methyl-5-oxohexanoate (3b). The
stereochemistry of the desired products was controlled by the
choice of ADH, and the absolute configuration of the different
stereoisomers was confirmed by vibrational circular dichroism
(VCD). To demonstrate the value of selective triketide
β,δ-Diketo esters, such as tert-butyl 3,5-dioxohexanoate (1a),
are the simplest examples of triketides. A single keto reduction
gives rise to four possibilities: R- or S-selective reduction at
either the β- or δ-position. In tert-butyl 4-methyl-3,5-
dioxohexanoate (1b), the alkylation at the γ-position doubles
synthesis, we used 11 stereoisomers of 2a, 2b, 3a, and 3b as
reference compounds for studying the regio- and
stereoselectivity of diketo substrate reductions catalyzed by
carbonyl reductase S1 from Candida magnoliae (CMCR-S1).
While recombinant enzymes with an N-terminal His-tag
(CMCR-S1N-His) catalyzed δ-selective reductions, catalysis by
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 2 of 10
View Article Online
DOI: 10.1039/C7OB02666H
ARTICLE
Journal Name
the C-terminal His-tag variant (CMCR-S1C-His) delivered both δ- which uses molecular oxygen from ambient air in presence of a
and β-hydroxy products, each with high stereoselectivity.
quinone mediator as the final electron acceptor.12 We
analyzed both oxidation products by VCD spectroscopy and
comparison to spectra from quantum chemical calculations, a
straightforward method for the determination of the absolute
Results
configuration of chiral molecules18,19
. The VCD analysis
Selective route to δ,β- and β,δ-hydroxyketo esters
confirmed the rationale of corresponding selectivities in
oxidation and reduction direction and allowed the assignment
The two δ-hydroxy-β-keto ester enantiomers 2a were
synthesized via stereoselective keto reductions of 1a, as
described previously.13,14 Thus, reduction with ADH from
Lactobacillus brevis (LbADH) gave (R)-2a and the use of ADH
YGL157w from baker’s yeast resulted in (S)-2a with high
enantiomeric ratios (Scheme 1A). The selective synthesis of
the corresponding β-hydroxy-δ-keto ester regioisomer 3a is
less trivial and usually comprises multiple steps, including
enzyme catalysis.4,15 We aimed to develop a straightforward
and versatile synthesis employing well-characterized or even
commercially available ADHs. As a substrate for the crucial
enzymatic oxidation step, the racemic erythro-diol§ tert-butyl
of the LbADH-product to (R)-3a and the ADH-A-product to (S)
-
3a (Figure 1). Enantiomeric ratios (er) as given in the charts in
Scheme 1B were determined by chiral-phase HPLC† (Figure
S5).
Stereoselective synthesis of δ,β- and β,δ-hydroxyketo-γ-methyl
esters
Some ADHs show a strong preference for reducing either (R)-
or (S)-1b in biocatalytic syntheses.5,6 The continuous
racemization of substrate 1b via keto–enol tautomerism allows
dynamic kinetic resolution (DKR) upon ADH-catalyzed
reduction of the δ-keto-group. Previously, we described the
DKR of 1b via ADH reduction of the δ-keto group, giving access
to three of the four possible diastereomers of tert-butyl 5-
hydroxy-4-methyl-3-oxohexanoate (2b; Scheme 1C).6
3,5-dihydroxyhexanoate (erythro-4a) was synthesized by
diastereoselective reduction of rac-2a16,17 (obtained by aldol
addition of acetaldehyde to tert-butyl acetoacetate).6 To
obtain (R)-3a and (S)-3a from enzymatic kinetic resolution of
erythro-4a, we used LbADH (the same enzyme employed for
the stereoselective reduction of 1a, see above), and ADH-A
from Rhodococcus ruber. These enzymes were chosen
according to their known opposite stereoselectivities in
reduction direction. To generate a thermodynamic driving
force, we combined the ADH-catalyzed oxidation reactions
with a previously developed regeneration system for NAD(P)+,
For the synthesis of the different stereoisomers of the β-
hydroxy species 3b
,
we assumed that an ADH’s
stereospecificity was the same in oxidation and reduction
direction. According to the observations made for the DKR of
1b,6 R-specific LbADH should preferably catalyze the oxidation
of
a (4S,5R)-configured substrate, while ADH-A should
specifically convert the (4R,5S)-configured enantiomer. As
Scheme 1 Stereoselective synthesis of hydroxy keto esters involving ADHs for stereoselective reduction (LbADH, YGL157w, ADH-A) or
stereospecific oxidation (LbADH, ADH-A). Reaction conditions: i) 1. B(OMe)Et2 (–70 °C), 2. NaBH4, 3. H2O2;16,17 ii) (CH3)4N(CH3CO2)3BH (–
20 °C).33 The regio- and stereochemical composition of the resulting hydroxy keto esters are shown in pie charts, as previously introduced by
Kaluzna et al.13 A) δ-Hydroxy-β-keto esters. B) β-Hydroxy-δ-keto esters. C) δ-Hydroxy-β-keto-γ-methyl esters. D) β-Hydroxy-δ-keto-γ-methyl
esters.
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 3 of 10
View Article Online
DOI: 10.1039/C7OB02666H
Journal Name
ARTICLE
substrates, we synthesized racemic erythro-4b and threo-4b Stereo- and regioselectivity of CMCR-S1-catalyzed reductions
via diastereoselective reduction of 2b (Scheme 1).16,17,33 This
With an array of different regio- and stereoisomers of
was the same strategy to control the configuration of the β-
methylated and non-methylated hydroxyketo esters of
hydroxy group (here relative to the γ-methyl group) as applied
moderate to high optical purity in hand, it was possible to
for the non-methylated substrates.
examine the stereoselectivity behavior of keto reductases
ADH-A-catalyzed oxidation of erythro-4b and threo-4b
toward triketide-derived substrates. CMCR-S1, an enzyme
exhibited the expected regioselectivity and stereospecificity,
whose physiological substrate is unknown, was chosen as a
to afford syn-(3R,4S)-3b (24%, er = 96:4; drsyn:anti = 75:25) and
model.
Similar
to
other
short
chain
dehydrogenases/reductases (SDRs) from Candida species,20,21
CMCR-S1 displays high selectivity in the reduction of carbonyl
substrates, and therefore has a high potential for use as a
catalyst in the synthesis of chiral building blocks.22 This
potential has been demonstrated for whole cell
transformations of ethyl 4-chloroacetoacetate and ethyl
acetoacetate with overexpressed CMCR-S1 in E. coli, yielding
high product concentrations.23,24
anti-(3S,4S)-3b (12%; er = 92.5:7.5; drsyn:anti
=
13:87),
respectively (Scheme 1D; absolute configurations confirmed by
VCD, see Figures S14 and S15). Interestingly, the LbADH-
catalyzed oxidations delivered two diastereomers each for
erythro-4b and threo-4b (Scheme 1D), suggesting a lower
stereospecificity for the configuration of the γ-methyl group in
the oxidation direction than in LbADH-catalyzed reductions.
Nevertheless, as the retention times of syn-(3R,4S)-3b and
anti-(3S,4S)-3b could be assigned, it was possible to deduce
the elution order and determine the dr of all four
diastereomers of 3b by chiral-phase HPLC (Figure S8).
The codon-optimized sequence for CMCR-S1 was cloned into
the pET-19b vector (for an N-terminal His-tag) or the pET-
22b(+) vector (for C-terminal His-tag) and the corresponding
fusion proteins CMCR-S1N-His and CMCR-S1C-His were produced
in E. coli. The His-tag was introduced to allow purification by
affinity chromatography prior to enzymatic conversions, to
avoid background reactions from cell lysate enzymes. To
obtain isolated CMCR-S1wt, the N-terminal His-tag from
purified CMCR-S1N-His was cleaved with enterokinase. The
circular dichroism (CD) spectra of the three variants (Figure S1)
indicate a high α-helix content superimposed with a β-sheet
pattern25,26 as expected for a properly folded SDR with a
Rossmann-fold motif.27-29 The three variants also had a similar
thermal denaturation behavior (Figure S2), with melting
temperatures of 56 °C (CMCR-S1wt), 52 °C (CMCR-S1N-His), and
50 °C (CMCR-S1C-His). Slightly decreased thermostability after
addition of a His-tag is not unusual, and has already been
reported for other proteins.30,31 All three variants showed
activity for the conversion of ethyl 4-chloroacetoacetate, albeit
with a significantly lower specific activity (<1%) of CMCR-S1C-His
(0.8 U mg–1) relative to CMCR-S1N-His (109.2 U mg–1). This can
be explained by the C-terminus being generally closer to the
active site in SDRs than the N-terminus,32 and thus possible
interference with the enzyme–substrate interaction. The
specific activity of CMCR-S1wt (12.0 U mg–1) was also lower
than that of CMCR-S1N-His, most probably due to a loss in
activity after the enterokinase treatment. Still, the specific
activity of CMCR-S1wt was 15-fold higher than that of CMCR-
S1C-His
.
The enzymatic reduction of diketo substrate 1a catalyzed by
CMCR-S1N-His afforded predominantly the δ-hydroxy product
2a. VCD analysis (Figure S11) and comparison of the HPLC
chromatogram of the CMCR-S1N-His reduction product with
those of the reference compounds (Figure S6) allowed the
assignment of the R-configuration (CMCR-S1N-His; er = 98.5:1.5;
Scheme 2A). This is consistent with the stereoselectivity
previously observed for the conversion of ethyl
4-chloroacetoacetate.‡22 In contrast, the conversion of 1a in
the presence of CMCR-S1C-His afforded two products: the
regioisomers 2a and 3a (84:16; assigned by NMR spectroscopy,
Figure 1 Experimental VCD and IR spectra of tert-butyl 3-
hydroxy-5-oxohexanoate (3a) from the oxidation of tert-butyl
syn-3,5-dihydroxyhexanoate (erythro-4a) catalyzed by ADH-A
(red) and LbADH (blue) in comparison to calculated spectra for
the enantiomers (R)-3a (black) and (S)-3a (gray).
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 4 of 10
View Article Online
DOI: 10.1039/C7OB02666H
ARTICLE
Journal Name
Figure S17). The availability of reference compounds for all latter still occurring with high stereoselectivity and
four possible isomeric reduction products allowed a qualitative stereospecificity.
and quantitative stereochemical analysis by chiral-phase HPLC
Bispecificity of CMCR-S1
(Figure S6). Reduction at the δ-position catalyzed by CMCR-
S1C-His occurred with the same selectivity as for CMCR-S1N-His
,
According to these findings, CMCR-S1 provides two
yielding (R)-2a (er = 98.5:1.5). In addition, formation of the β- independent and highly selective reaction paths. Variation of
hydroxy product (R)-3a (16%, er = 95:5) was observed (Scheme the C-terminus, here serendipitously undertaken by the
2A). In comparison to CMCR-S1N-His, CMCR-S1C-His seems to introduction of
a His-tag, was sufficient to reveal the
provide an alternative reaction path for β,δ-diketo ester previously unknown β-reduction pathway.
reductions with marked stereoselectivity for reduction of Two hypotheses can be advanced to explain the observed
diketone 1a at the β-position. The high stereopurity of the by- bispecific behavior of CMCR-S1C-His: 1) The C-terminal His-tag
product suggests that β-reduction is not a random, but rather invokes the formation of an additional binding mode for
a selective process involving the existence of a highly specific substrates 1a and 1b, which coincidentally leads to high
binding mode in addition to the one responsible for selective stereoselectivity and stereospecificity for keto reductions at
δ-reduction.
the β-position. In this regard, it is known that the C-terminal
NMR spectroscopy and GC-MS analysis revealed that CMCR- loop of SDRs is involved in recognition, binding and positioning
S1N-His reduces the methylated diketone 1b solely at the δ- of the substrate.34-40 2) More likely, binding modes for δ-
position. HPLC analysis (Figure S9) and comparison to the δ- specificity and for β-specificity are already present in the wild-
hydroxy reference compounds 2b indicated R-selectivity and type enzyme. Here, the δ-reducing binding mode is strongly
high diastereoselectivity for syn-(4S,5R)-2b (er = 98:2; favored, thereby rendering contributions from the β-specific
drsyn:anti = 85:15; Scheme 2B).
CMCR-S1C-His reduction of 1b
binding kinetically negligible. However, in CMCR-S1C-His, the
,
however, yielded two hexa-His extension of the C-terminus somehow reduces
regioisomers: δ-hydroxy-β-keto ester 2b and β-hydroxy-δ-keto substrate affinity to the δ-reducing binding mode and thus the
ester 3b. Compared to the reduction of 1a, the β-hydroxy contribution from the β-reducing binding mode becomes more
product was formed in considerably higher amounts important (Figure 2). The considerably higher Michaelis
(
2b:3b = 62:38, Scheme 2). Surprisingly, HPLC analysis (Figure constant (determined for ethyl 4-chloroacetoacetate; Figure
S9) revealed different δ-hydroxy diastereomers for the CMCR- S3) for CMCR-S1C-His (Km = 19.55 mM) than for CMCR-S1wt
S1C-His and CMCR-S1N-His reductions. The CMCR-S1C-His reduction (Km = 6.7 mM) and CMCR-S1N-His (Km = 9.10 mM) suggests that
afforded predominantly anti-(4S,5S)-2b, albeit with moderate substrate binding is hindered in presence of a C-terminal His-
enantio- and diastereoselectivity (er = 89:11, drsyn:anti = 21:79). tag. However, the substrate used for Km determination only
The β-hydroxy product was identified as anti-(3S,4S)-3b, which contains a single keto group and the Km value alone cannot
was formed with high enantio- and diastereoselectivity provide information about the competition of two binding
(er = 95:5, drsyn:anti = 9:91; Scheme 2B). These findings suggest modes with different substrate affinity.
that the additional methyl group interferes more with To substantiate the competing binding modes-hypothesis
experimentally, we tested the regioisomeric product
composition after CMCR-S1-catalyzed conversion (pH 7.0, 24
hours) of diketone 1a at different substrate/enzyme ratios
(Figure 2). At high enzyme concentrations (low
substrate/enzyme ratio), the influence of binding affinities is
crucial and the substrate will preferably bind in a mode that
leads to a more stable substrate–enzyme complex. In the
model depicted in Figure 2A, this is the binding mode
responsible for δ-reduction and, accordingly, at high enzyme
concentrations there will be a large excess of the δ-reduced
regioisomer. Indeed, for the CMCR-S1N-His-catalyzed reduction
at an enzyme concentration of 0.89 mg·mL–1, virtually full
conversion of 1a into the δ-hydroxy product 2a was observed
(δ:β = 99:1, Figure 2A). With a large substrate excess at low
enzyme concentrations (high substrate/enzyme ratio), binding
affinities are less important. Correspondingly, the
regioselectivity of CMCR-S1N-His-catalyzed reduction was
decreased at enzyme concentrations of 0.09 and 0.18 mg·mL–1,
Scheme 2 Regio- and stereochemical outcome of the reduction of
β,δ-diketo esters catalyzed by CMCR-S1N-His or CMCR-S1C-His
,
giving rise to detectable amounts of the β-hydroxy regioisomer
3a, albeit at lower overall conversions after 24 hours. These
observations are consistent with the hypothesis that both
binding modes are intrinsically present in the enzyme.
shown in pie charts according to the structures given in Scheme 1.
A) Reduction of 1a. B) Reduction of 1b.
substrate specificity for δ-reduction than for β-reduction, the
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 5 of 10
View Article Online
DOI: 10.1039/C7OB02666H
Journal Name
ARTICLE
At low enzyme concentrations, the overall conversion of 1a
with CMCR-S1C-His catalysis is lower than with CMCR-S1N-His
(Figure 2B). This was expected from the low specific activity
and the high Km value for CMCR-S1C-His (see above). However,
with this variant, the proportion of the β-hydroxy product 3a is
considerably higher (up to δ:β = 62:38 at 0.09 mg·mL–1
enzyme). Similar to CMCR-S1N-His, an increase in enzyme
concentration favors the formation of 2a; however, at high
enzyme concentrations, the δ:β-ratio converges to ~80:20.
These findings indicate a lower preference for δ-specific
binding in CMCR-S1C-His (Figure 2B).
Conclusions
We have presented the highly versatile, regio- and
stereoselective synthesis of 11 of the 12 possible chiral tert-
butyl hydroxyoxohexanoates with and without methylation at
the γ-position. These esters were used as reference
compounds to characterize the selectivity of the enzymatic
keto reduction of substrates with multiple carbonyl groups.
Thus, the well-known enzyme CMCR-S1 delivers two
regioisomeric hydroxy keto esters upon reduction of the
artificial triketides 1a and 1b. Particularly, the highly
stereoselective and stereospecific reduction at the β-position
was unexpected. This β-reduction seems to be an intrinsic
property of CMCR-S1; until now, this property had remained
undetected, which underlines the necessity to include longer-
chain substrates in enzyme characterization studies. We have
confirmed a role of the C-terminus in SDRs in substrate
recognition.34-40 In the case of CMCR-S1C-His, the His-tag at the
C-terminus results in an unexpected shift from one highly
specific binding mode to the other and
a substrate-
dependence of the stereoselectivity that was not observed in
the absence of the His-tag. These findings suggest that
reductases might carry multiple specificities and selectivities,
in addition to the dominating effect that is generally regarded
as the physiological activity. The hidden activities might be
rendered more dominant by marginal modifications, such as
elongation of the C-terminus. Possibly, nature utilizes hidden
activities in enzymes as an optional ‘add-on’ feature to
generate a high diversity of natural products from a minimal
mutational effort.
Figure 2 Substrate/product composition from the reduction of
tert-butyl 3,5-dioxohexanoate (1a) catalyzed by CMCR-S1N-His (A)
and CMCR-S1C-His (B) at different enzyme concentrations. The
cartoons illustrate the two scenarios as
a model of two
competing binding modes for δ-reduction and β-reduction. In
CMCR-S1N-His, δ-specific binding is largely dominant. In CMCR-
S1C-His, the modification at the C-terminus interferes with binding
in the δ-mode, thereby reducing the overall turnover, but
increases the contribution of β-mode binding to the overall
outcome.
Experimental
Molecular cloning, protein production and purification
All restriction, PCR and cloning steps were performed
according to standard procedures.41 The synthetic codon-
optimized genes for CMCR-S1 and ADH-A were obtained from
GenScript (Piscataway, USA) and Thermo Fisher Scientific
(Waltham, USA), respectively. The amplified CMCR-S1 gene
(see ESI for PCR primers) was cloned into the vectors pET-19b
(encodes for an N-terminal His-tag) and pET-22b(+) (C-terminal
His-tag) (both Merck KGaA, Darmstadt, Germany) using the
NdeI (5’) and XhoI (3’) restriction sites. The synthetic ADH-A
gene was delivered with a 15 base pair overlap on each side,
complementary to the ends of linearized pET-19b vector (see
ESI), which allowed cloning of the amplified gene with the In-
Fusion HD Cloning Plus kit (Takara, Kyoto, Japan). From the
resulting vector, the gene was cloned into pET-22b(+) using the
restriction sites NdeI (5’) and BamHI (3’), resulting in an
expression vector that does not encode for a His-tag. The
vectors were transformed in competent E. coli TG1 cells for
amplification and in competent E. coli BL21(DE3) cells for
heterologous overexpression (both Agilent Technologies, Santa
Clara, USA).
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 6 of 10
View Article Online
DOI: 10.1039/C7OB02666H
ARTICLE
Journal Name
After inoculation with 5 mL of an overnight starter culture butyl
(100 µg·mL–1 ampicillin-containing LB-Lennox medium; Carl methylacetamide.6,45
3-oxopentanoate
Racemic
with
tert-butyl
N-methoxy-N-
erythro-3,5-
Roth GmbH, Karlsruhe, Germany), the ADH-A gene was dihydroxyhexanoate (erythro-4a, 55%, drerythro:threo = 40:1) was
expressed in 500 mL of Studier’s auto induction medium42 (ZY- obtained by diastereoselective reduction of 2a (1740 mg,
medium, salt solution M and solution 5052 as carbon source) 8.6 mmol) as described by Thottathil et al.17 Accordingly, tert-
containing 100 µg·mL–1 ampicillin and incubated at 24 °C and butyl erythro-3,5-dihydroxy-4-methylhexanoate (erythro-4b
;
160 rpm for 24 h. For CMCR-S1N-His and CMCR-S1C-His, 400 mL of 67%; drerythro:threo = 39:1) was synthesized from 2b (2500 mg,
ampicillin-containing (100 µg·mL–1) LB-Lennox medium was 11.6 mmol). Both diols were purified by flash column
inoculated with 4 mL of an overnight starter culture. After the chromatography on silica gel (erythro-4a: cyclohexane:ethyl
culture was grown to an optical density of 0.6 (λ = 600 nm), acetate 1:1, erythro-4b: cyclohexane:ethyl acetate 2:1; see ESI
the expression was induced by addition of IPTG (0.8 mM) and for analytical data). For the synthesis of tert-butyl threo-3,5-
the protein was produced during incubation at 37 °C and dihydroxy-4-methylhexanoate (threo-4b) from 2b (230 mg,
160 rpm for 4 h. LbADH, YGL157w and NfsB were expressed 1 mmol), the anti-selective reduction method of Evans and
and produced in E. coli as described elsewhere.6,12,43 E. coli Chapman33
provided
206 mg
(86%)
of
threo-4b
cells were harvested by centrifugation (15000g, 4 °C, 15 min). (drerythro:threo = 1:58, GC-MS; see ESI for analytical data).
After resuspension in KPi buffer (50 mM, pH 7.5; 0.5 mL per
Enzymatic reductions
100 mL culture), the cells were disrupted by sonication
(6 × 15 s with 10 s intermission; Branson Sonifier II W-250; Substrates according to Scheme 1 (1 equiv), glucose (5 equiv
Heinemann, Schwäbisch Gmünd, Germany). Cell debris was and 0.05 equiv of cofactor (NADP+ for LbADH, YGL157w,
removed by centrifugation (12000g, 4 °C, 45 min).
CMCR-S1N-His and CMCR-S1C-His NAD+ for ADH-A) were
)
;
CMCR-S1N-His and CMCR-S1C-His were purified by affinity dissolved in KPi buffer (50 mM, pH 7.0; 50 mL buffer per mmol
chromatography on Ni-NTA agarose (Qiagen, Hilden, substrate). E. coli lysate containing the respective ADH was
Germany). Nonspecifically bound proteins were washed off added to a final activity of 20 (LbADH) or 1.0 U·mL–1 (YGL157w
with KPi buffer (50 mM, pH 7.5) containing 50 mM imidazole and ADH-A, see ESI for details on enzyme activity
(CMCR-S1N-His) or 25 mM imidazole (CMCR-S1C-His). CMCR-S1N- determination). CMCR-S1N-His (1.1 U·mL–1) and CMCR-S1C-His
His
and CMCR-S1C-His were eluted with KPi buffer (50 mM, (0.03 U·mL–1) were used as purified enzymes. For LbADH-
pH 7.5) containing 500 mM or 250 mM imidazole, respectively. catalyzed reductions, the mixture additionally contained 1 mM
The volume of the elution fractions was reduced to 5 mL by MgCl2. Glucose dehydrogenase (GDH; 200
U per mmol
ultrafiltration (4500g; Vivaspin 20 centrifugal concentrator, substrate) for regeneration of NAD(P)H was added to start the
10000 MWCO PES; Sartorius Stedim Biotech GmbH, Göttingen, reaction. Mixtures were stirred smoothly at room temperature
Germany) and desalted by size-exclusion chromatography (PD- and sampled periodically for TLC monitoring. Reactions were
10 Desalting Columns; GE Healthcare Bio-sciences AB, Uppsala, quenched by addition of NaCl (0.25 g per mL solvent), after
Sweden). Purified CMCR-S1N-His and CMCR-S1C-His were stored 24 h or when the monitoring indicated complete conversion.
at 4 °C. To obtain CMCR-S1wt, the His-tag was cleaved from The mixture was extracted with ethyl acetate. Precipitated
purified CMCR-S1N-His with enterokinase (New England Biolabs proteins were removed from the organic phase by filtration
GmbH, Frankfurt, Germany) according to the manufacturer’s over Celite 535. The organic extracts were dried over Na2SO4
recommendations. NfsB was purified as described previously and concentrated in vacuo. If necessary, the crude product
prior to use.12 LbADH and YGL157w were used as crude was purified by flash column chromatography. Conversion
extracts. ADH-A was used as a glycerol-containing crude yields were determined by 1H NMR spectroscopy.
1
extract (20% v/v) to allow storage at –20 °C. The activity of the Diastereomeric ratios (dr) correspond either to the ratio of H
produced enzymes was determined photometrically as a NMR integrals or to peak areas (AUC) in the GC-MS
function of decrease or increase in concentration of the chromatograms. Enantiomeric ratios (er) were determined by
cofactor NAD(P)H (for details, see ESI).
chiral HPLC (see ESI). The absolute configuration of optically
pure products was confirmed by VCD. Conversions at various
CMCR-S1N-His or CMCR-S1C-His concentrations (Figure 2) were
Organic syntheses
tert-Butyl 5-hydroxy-3-oxohexanoate (2a) and tert-butyl 5- carried out on an analytical scale (4 mg ≙ 0.02 mmol of 1a, 6 U
hydroxy-4-methyl-3-oxohexanoate (2b) were synthesized by GDH, 1 mL) and were quantified by GC-MS analysis directly
the aldol addition of acetaldehyde to the respective β-keto after ethyl acetate extraction.
ester, as described by Lüdeke et al.6 The NMR data (ESI) are The analytical data for (R)-2a from LbADH-catalyzed reduction
consistent
with
the
literature.6,14
tert-Butyl
3,5- of 1a (>99% conversion, er = 99:1) or from reductions
dioxohexanoate (1a) was obtained via Jones oxidation44 of 2a catalyzed by CMCR-S1N-His (89%, er = 98.5:1.5) or CMCR-S1C-His
(1010 mg, 5 mmol), as described by Bariotaki et al.,4 and (84%, er = 98.5:1.5), and for (S)-2a (YGL157w; 84%, er = 99:1)
purified by automated flash column chromatography (Isolera were consistent with those for 2a synthesized by aldol addition
Prime, SNAP KP-Sil cartridge: 50 g, solvent gradient: of acetaldehyde to tert-butyl acetoacetate (see above). The
cyclohexane:ethyl acetate from 98:2 to 78:22, flow rate: identity of the methylated species syn-(4S,5R)-2b from LbADH-
50 mL·min–1) yielding 589 mg (58%) of 1a. tert-Butyl 4-methyl- catalyzed (>99%, er >99:1, drsyn:anti = 93:7) or CMCR-S1N-His
3,5-dioxohexanoate (1b) was synthesized by acylation of tert- catalyzed reduction of 1b (61%, er = 98:2, drsyn:anti = 85:15), of
-
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 7 of 10
View Article Online
DOI: 10.1039/C7OB02666H
Journal Name
ARTICLE
syn-(4R,5S)-2b from ADH-A-catalyzed reduction of 1b (>99%, oxidation of threo-4b gave 66 mg (31%) of a mixture of tert-
er >99:1, drsyn:anti = 98:2), and of anti-(4S,5S)-2b from butyl syn/anti-(3R,4R/S)-3-hydroxy-4-methyl-5-oxohexanoate
YGL157w-catalyzed (70%, er >99:1, drsyn:anti = 6:94) or CMCR- (syn/anti-(3R,4R/S)-3b, drsyn:anti = 33:67). The analytical data for
S1C-His-catalyzed reduction of 1b (62%, er = 89:11, the oxidation products are given in the ESI.
drsyn:anti = 21:79) was confirmed by comparison to the
VCD analysis
analytical data for syn/anti-2b obtained by aldol addition of
acetaldehyde to tert-butyl 3-oxopentanoate (see above). The For the measurement of experimental IR and VCD spectra,
β-hydroxy-δ-keto esters (R)-3a, (16%, er = 95:5), and anti- samples were dissolved in anhydrous CDCl3 and placed in a
(3S,4S)-3b (38%, er = 95:5, drsyn:anti = 9:91), obtained as rotatable BaF2 cell (path length: 110 µm). The measured
byproducts from CMCR-S1C-His reduction of 1a and 1b
,
spectra represent the average of 12 h measurements at
respectively, were not separable from the corresponding δ- different angles from 0° to 360° in increments of 7.5°. All
hydroxy-β-keto esters by flash column chromatography. spectra were background-corrected for a CDCl3 spectrum. The
Therefore, and to avoid adulteration of the regio- and concentrations of the samples are specified on the individual
stereochemical composition from purification, their identity spectra. To facilitate the assignment of its configuration,
and the isomeric ratios were determined chromatographically (4S,5S)-2b was transformed into the six-membered lactone
for the crude reduction products through comparison with the trans-(5S,6S)-5,6-dimethyldihydro-2H-pyran-2,4(3H)-dione, as
reference compounds from enzymatic oxidations (see below).
described by Tabuchi et al.,46 thereby reducing conformational
freedom and increasing the signal intensity in the VCD
spectrum.
Enyzmatic oxidations
Substrates according to Scheme 1 (1 equiv), 0.05 equiv of the For the calculation of theoretical IR and VCD spectra,
oxidized cofactor (NADP+ for LbADH, NAD+ for ADH-A) and conformer geometries of (R)-2a, (R)-3a, (4R,5S)-2b, (3S,4S)-3b
,
lawsone (0.05 equiv) were dissolved in Tris HCl buffer (50 mM, (3R,4S)-3b and trans-(5S,6S)-5,6-dimethyldihydro-2H-pyran-
pH 8.0; 50 mL buffer per mmol substrate). E. coli lysate 2,4(3H)-dione were generated with the MMFF-based
containing either LbADH or ADH-A was added to a final activity conformer search algorithm in Spartan 08 (Wavefunction,
of
3
U·mL–1 (see ESI for details on enzyme activity Irvine, USA). Geometry optimizations and calculations of
determination). For LbADH-catalyzed oxidations, the mixture relative energies, IR, and VCD frequencies were performed at
additionally contained 1 mM MgCl2. Purified NfsB (2000 U per the DFT level in Gaussian 0947 (in the case of 2a and 3a with
mmol substrate) was added for cofactor regeneration. B3LYP/6-311++G(d,p); in the case of 2b
, 3b and 11 with
Reactions were quenched by addition of NaCl (0.25 g per mL B3LYP/6-31+G(d,p) and PCM solvent model for
a
solvent), after stirring at room temperature for 20 h. The chloroform48). The predicted frequencies were scaled by 0.981
mixture was extracted with ethyl acetate, and precipitated for B3LYP/6-311++G(d,p) and 0.983 for B3LYP/6-31+G(d,p)
enzymes were removed from the organic phase by filtration (scaling factors determined with VCDspecTech49). The shown
over Celite 535. The organic extracts were washed with spectra represent Boltzmann-weighted averages of the
aqueous NaOH solution (0.01 M) and deionized water, dried conformer spectra with Lorentzian line shapes (6 cm–1
over Na2SO4 and concentrated in vacuo. The crude products bandwidth) around calculated intensities.
were purified by automated flash column chromatography
(Isolera Prime, SNAP KP-Sil cartridge: 25 g, solvent gradient:
Acknowledgements
cyclohexane:ethyl acetate from 88:12 to 0:100, flow rate:
1
25 mL·min–1). The dr was calculated from H NMR integrals or
We thank Dr. Kay Greenfield for help in improving the
the AUCs in GC-MS chromatograms. The er was obtained from
manuscript. Financial support of this work by the DFG (IRTG
the AUCs in chiral HPLC chromatograms.
1038, RTG 1976) is gratefully acknowledged. This research was
Oxidation of erythro-4a (204 mg, 1 mmol) catalyzed by LbADH
supported in part by the bwHPC initiative and the bwHPC-C5
afforded 72 mg of tert-butyl (R)-3-hydroxy-5-oxohexanoate
project provided through associated computing services of the
((R)-3a) after purification (36%, er = 97.5:2.5), while ADH-A
JUSTUS HPC facility at the University of Ulm.
catalysis yielded 74 mg of purified tert-butyl (S)-3-hydroxy-5-
oxohexanoate ((S)-3a, 37%, er = 98.5:1.5). ADH-A oxidation of
erythro-4b (218 mg, 1 mmol) led to 52 mg (24%) of purified
Notes and references
§To facilitate the distinction from the syn/anti configuration of
hydroxyl groups relative to vicinal alkyl groups, henceforth, the
relative stereochemistry of two hydroxy groups is denoted as
erythro and threo.
tert-butyl
syn-(3R,4S)-3-hydroxy-4-methyl-5-oxohexanoate
er = 96:4, drsyn:anti = 75:25), while ADH-A
(syn-(3R,4S)-3b
,
oxidation of threo-4b gave tert-butyl anti-(3S,4S)-3-hydroxy-4-
methyl-5-oxohexanoate (anti-(3S,4S)-3b 26 mg, 12%,
,
er = 92.5:7.5, drsyn:anti = 13:87). LbADH-catalyzed oxidations
were not stereospecific with respect to the γ-methyl group.
Thus, oxidation of erythro-4b (218 mg, 1 mmol) and
purification afforded 77 mg (36%) of a diastereomeric mixture
†
To allow UV detection and to increase the resolution of the
separations, all HPLC analyses in this study were performed on
benzoylated derivatives.
of
tert-butyl
syn/anti-(3S,4R/S)-3-hydroxy-4-methyl-5-
oxohexanoate (syn/anti-(3S,4R/S)-3b, drsyn:anti = 45:55), while
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 8 of 10
View Article Online
DOI: 10.1039/C7OB02666H
ARTICLE
Journal Name
‡
The reported reduction product ethyl 4-chloro-3-
30. M. Krupka, J. Masek, L. Barkocziova, P. T. Knotigova, P. Kulich,
J. Plockova, R. Lukac, E. Bartheldyova, S. Koudelka, R.
Chaloupkova, M. Sebela, D. Zyka, L. Droz, R. Effenberg, M.
Ledvina, A. D. Miller, J. Turanek and M. Raska, PloS One, 2016,
11, 1.
31. P. Ledent, C. Duez, M. Vanhove, A. Lejeune, E. Fonzé, P.
Charlier, F. Rhazi-Filali, I. Thamm, G. Guillaume, B. Samyn, B.
Devreese, J. Van Beeumen, J. Lamotte Brasseur and J. M.
Frère, FEBS Lett., 1997, 413, 194.
32. K. Kavanagh, H. Jörnvall, B. Persson and U. Oppermann, Cell.
Mol. Life Sci., 2008, 65, 3895.
33. D. A. Evans and K. T. Chapman, Tetrahedron Lett., 1986, 27,
5939.
34. M. A. Schätzle, S. Flemming, S. M. Husain, M. Richter, S.
Günther and M. Müller, Angew. Chem. Int. Ed., 2012, 51, 2643.
35. T. Nishioka, Y. Yasutake, Y. Nishiya and T. Tamura, FEBS J.,
2012, 279, 3264.
36. T. Nishioka, Y. Yasutake, Y. Nishiya, N. Tamura and T. Tamura,
Proteins, 2009, 74, 801.
37. A. C. Freydank, W. Brandt and B. Dräger, Proteins, 2008, 72,
173.
38. D. I. Liao, J. E. Thompson, S. Fahnestock, B. Valent and D. B.
Jordan, Biochemistry, 2001, 40, 8696.
39. H. P. Zhou, F. X. Yan and H. H. Tai, Eur. J. Biochem., 2001, 268,
3368.
40. R. Albalat, M. Valls, J. Fibla, S. Atrian and R. González-Duarte,
Eur. J. Biochem., 1995, 233, 498.
41. D. W. R. J. F. Sambrook, Molecular cloning: A laboratory
manual, Cold Spring Harbor, New York, 3rd edn., 2000.
42. F. W. Studier, Protein Expression Purif., 2005, 41, 207.
43. W. Hummel, Adv. Biochem. Eng./Biotechnol., 1997, 58, 145.
44. K. Bowden, I. M. Heilbron, E. R. H. Jones and B. C. L. Weedon,
J. Chem. Soc., 1946, 39.
hydroxybutanoate is S-configured, as the chlorine atom switches
the CIP priority.
1. H. M. O'Hare, A. Baerga-Ortiz, B. Popovic, J. B. Spencer and P.
F. Leadlay, Chem. Biol., 2006, 13, 287.
2. K. Faber, Biotransformations in Organic Chemistry, Springer,
Berlin Heidelberg, 6th edn., 2011.
3. H. Zhou, Z. Z. Gao, K. Qiao, J. Wang, J. C. Vederas and Y. Tang,
Nat. Chem. Biol., 2012, 8, 331.
4. A. Bariotaki, D. Kalaitzakis and I. Smonou, Org. Lett., 2012, 14,
1792.
5. A. G. Ji, M. Wolberg, W. Hummel, C. Wandrey and M. Müller,
Chem. Commun., 2001, 57.
6. S. Lüdeke, M. Richter and M. Müller, Adv. Synth. Catal., 2009,
351, 253.
7. M. Wolberg, W. Hummel and M. Müller, Chem. Eur. J., 2001, 7,
4562.
8. R. N. Patel, A. Banerjee, C. G. McNamee, D. Brzozowski, R. L.
Hanson and L. J. Szarka, Enzyme Microb. Technol., 1993, 15,
1014.
9. M. Müller, M. Wolberg, W. Hummel and C. Wandrey,
Germany Pat., DE19937825 A1, 1999.
10. A. Liljeblad, A. Kallinen and L. T. Kanerva, Curr. Org. Synth.,
2009, 6, 362.
11. X. Luo, Y. J. Wang, W. Shen and Y. G. Zheng, J. Biotechnol.,
2016, 224, 20.
12. J. Haas, M. A. Schätzle, S. M. Husain, J. Schulz-Fincke, M. Jung,
W. Hummel, M. Müller and S. Lüdeke, Chem. Commun., 2016,
52, 5198.
13. W. A. Kaluzna, T. Matsuda, A. K. Sewell and J. D. Stewart, J.
Am. Chem. Soc., 2004, 126, 12827.
14. M. Wolberg, W. Hummel, C. Wandrey and M. Müller, Angew.
Chem. Int. Ed., 2000, 39, 4306.
15. S. Tartaggia, S. Fogal, R. Motterle, C. Ferrari, M. Pontini, R.
Aureli and O. De Lucchi, Eur. J. Org. Chem., 2016, 3162.
16. K. M. Chen, G. E. Hardtmann, K. Prasad, O. Repic and M. J.
Shapiro, Tetrahedron Lett., 1987, 28, 155.
45. M. Wolberg, A. G. Ji, W. Hummel and M. Müller, Synthesis,
2001, 937.
46. H. Tabuchi, T. Hamamoto, S. Miki, T. Tejima and A. Ichihara, J.
Org. Chem., 1994, 59, 4749.
47. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci,
G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P.
Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg,
M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M.
Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.
A. Montgomery Jr, J. E. Peralta, F. Ogliaro, M. Bearpark, J. J.
Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi,
J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, N. J. Millam, M. Klene, J.
E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G.
Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S.
Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J.
Cioslowski and D. J. Fox, GAUSSIAN 09 (Revision D.01),
Gaussian Inc., Wallingford CT, 2009.
17. J. K. Thottathil, Y. Pendri, W.-S. Li and D. R. Kronenthal, US
Pat., US 5 278 313, 1994.
18. T. B. Freedman, X. L. Cao, R. K. Dukor and L. A. Nafie, Chirality,
2003, 15, 743.
19. P. J. Stephens, F. J. Devlin and J.-J. Pan, Chirality, 2008, 20,
643.
20. N. Aggarwal, P. K. Mandal, N. Gautham and A. Chadha, Acta
Crystallogr, F, 2013, 69, 313.
21. S. Sudhakara and A. Chadha, Org. Biomol. Chem., 2017, 15,
4165.
22. D. M. Zhu, Y. Yang and L. Hua, J. Org. Chem., 2006, 71, 4202.
23. N. Kizaki, Y. Yasohara, J. Hasegawa, M. Wada, M. Kataoka and
S. Shimizu, Appl. Microbiol. Biotechnol., 2001, 55, 590.
24. Y. Yasohara, N. Kizaki, J. Hasegawa, M. Wada, M. Kataoka and
S. Shimizu, Tetrahedron: Asymmetry, 2001, 12, 1713.
25. N. J. Greenfield, Nat. Protoc., 2006, 1, 2876.
26. G. Holzwarth and P. Doty, J. Am. Chem. Soc., 1965, 87, 218.
27. H. Jörnvall, J. Hedlund, T. Bergman, U. Oppermann and B.
Persson, Biochem. Biophys. Res. Commun., 2010, 396, 125.
28. C. Filling, K. D. Berndt, J. Benach, S. Knapp, T. Prozorovski, E.
Nordling, R. Ladenstein, H. Jornvall and U. Oppermann, J. Biol.
Chem., 2002, 277, 25677.
48. S. Miertuš, E. Scrocco and J. Tomasi, Chem. Phys., 1981, 55,
117.
49. CDspecTech: Computer programs for calculating similarity
measures of comparison between experimental and
calculated dissymmetry factors and circular intensity
differentials.
29. M. G. Rossmann, D. Moras and K. W. Olsen, Nature, 1974, 250,
194.
https://sites.google.com/site/cdspectech1/,version
2015.
21.7,
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Please do not adjust margins
Organic & Biomolecular Chemistry
Page 9 of 10
View Article Online
DOI: 10.1039/C7OB02666H
Journal Name
ARTICLE
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Organic & Biomolecular Chemistry
Page 10 of 10
View Article Online
DOI: 10.1039/C7OB02666H
A recombinant carbonyl reductase shows different regioselectivity with a C-terminal His-tag
compared to the N-tagged enzyme toward the same triketide substrate. Highly selective
synthesis of reference triketides allowed solving this conundrum.