Rationalisation of the stereochemical outcome of ene-reductase-mediated bioreduction of α,β-difunctionalised alkenes

Article abstract of DOI:10.1016/j.molcatb.2013.12.020

The OYE1-3-mediated reductions of some α,β-difunctionalised alkenes, showing on the double bond a nitrile and ester group, are submitted to a careful stereochemical analysis, in order to identify which of the two electron-withdrawing groups (EWGs) is resp

Full text of DOI:10.1016/j.molcatb.2013.12.020

Journal of Molecular Catalysis B: Enzymatic 101 (2014) 67–72
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Rationalisation of the stereochemical outcome of
ene-reductase-mediated bioreduction of ␣,␤-difunctionalised alkenes
Elisabetta Brennaa,∗, Michele Crottia, Francesco G. Gattia, Alessia Manfredia,
Daniela Montib, Fabio Parmeggiania, Andrea Pugliesea, Davila Zampieria
a Politecnico di Milano, Dipartimento di Chimica, Materiali, Ingegneria Chimica, Via Mancinelli 7, I-20131 Milano, Italy
b Istituto di Chimica del Riconoscimento Molecolare-CNR, Via M. Bianco 9, I-20131 Milano, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
The OYE1-3-mediated reductions of some ␣,␤-difunctionalised alkenes, showing on the double bond a
nitrile and ester group, are submitted to a careful stereochemical analysis, in order to identify which
of the two electron-withdrawing groups (EWGs) is responsible for the activation of the C C double
bond towards reduction and for establishing hydrogen bond interactions within the binding pocket of
the enzymes. The results show that for most of these substrates the activating EWG is the CN moiety
linked to the prostereogenic olefinic carbon atom. The final stereochemical outcome can be explained
through the empirical model which has been recently developed for difunctionalised alkenes activated
by carbonyl/carboxyl containing EWGs.
In a single case the activation is due to the COOR group linked to the less substituted olefinic carbon
atom: an alternative empirical model is established for this kind of substrates, taking into consideration
the OYE-catalysed reductions of ␤,␤ꢀ-disubstituted-␣-monofunctionalised alkenes.
© 2014 Elsevier B.V. All rights reserved.
Received 24 October 2013
Received in revised form
17 December 2013
Accepted 21 December 2013
Available online 7 January 2014
Keywords:
Ene-reductase
Old Yellow Enzyme
Enantioselectivity
Alkene reduction
Deuterium
1. Introduction
(FMNH2) prosthetic group which imparts a yellow colour to puri-
fied protein samples. It has been established that the C C double
The use of ene-reductases (ERs) for the enantioselective reduc-
tion of activated alkenes is currently receiving great interest [1],
because of the efficacy shown by this kind of transformation in
the synthesis of chiral building blocks for organic chemistry appli-
cations [2]. The effects due to the stereochemistry of the starting
alkene, to the steric and electronic characteristics of the activating
electron-withdrawing groups (EWGs), as well as of other sub-
stituents are to be carefully investigated, in order to define the
limits and potential of this kind of reaction. The information is
essential for including this bioreduction into the pool of synthetic
tools, into which chemists can delve to select the best strategy for
the preparation of their target molecules, just as they are accus-
tomed to do when they search among conventional reagents of
organic chemistry. The availability of robust enzymatic procedures
can facilitate the introduction of biocatalysed steps in modern pro-
duction processes, bringing along all the advantages of enzymes for
sustainability.
bond can only be reduced by these enzymes if it is made susceptible
to the nucleophilic attack of a hydride (delivered by the reduced
flavin mononucleotide prosthetic group) by the presence of an
EWG, which is also able to establish hydrogen bonds within the
binding pocket of the enzyme [3]. Investigations are to be devoted
to define which EWGs can activate alkenes towards OYE-catalysed
reductions by themselves and which ones are to be combined
with other groups. Up to now, it has been established that ␣,␤-
unsaturated aldehydes and ketones, nitroolefins and maleimides
are good substrates for this kind of reactions, whereas the OYE-
mediated reduction ␣,␤-unsaturated esters is only possible when
another EWG is present on the double bond, e.g. a halogen atom
linked to the same carbon atom bearing the ester function [4], or
an ester [5] or a nitrile [6] group in ␤ position.
We have recently reported [6a] on the reduction of cyano esters
(E)- and (Z)-1 (Scheme 1), precursors of ␥2-amino acids for foldamer
applications, by means of OYE1-3, and we have carried out a
detailed analysis of the stereochemical course of the reaction by
means of deuterium labelling. Contemporaneously, a paper has
been published [6b] describing the reduction of the carbon–carbon
double bond of a class of regioisomeric compounds, i.e. (E)- and
(Z)-2 (Scheme 1), by means of isolated OYEs with the aim of pro-
viding a biocatalytic route to precursors of GABA analogues, such
as pregabalin.
Most of the ERs that have been identified in the last decades
belong to the well-known family of Old Yellow Enzymes (OYEs),
which are characterised by the presence of a flavin mononucleotide
∗
Corresponding author. Tel.: +39 02 23993077; fax: +39 02 23993180.
E-mail address: elisabetta.brenna@polimi.it (E. Brenna).
1381-1177/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2013.12.020

68
E. Brenna et al. / Journal of Molecular Catalysis B: Enzymatic 101 (2014) 67–72
Scheme 1. Cyano-substituted unsaturated esters subjected to ER-mediated reduc-
tion (Ref. [6a] for (E)- and (Z)-1; Ref. [6b] for (E)- and (Z)-2).
On the basis of our experience on substrates 1, we decided to
investigate in more detail the stereochemistry of the bioreduction
of compounds 2 and to establish whether the nitrile or the ester
function acted as the activating group of the bioreduction, i.e. to
determine which of the two EWGs was involved in the formation
of hydrogen bonds with the amino acid residues in the active site
of OYE1-3.
As a matter of fact, this information is essential to define the
binding mode of the substrate, and identify the stereoheterotopic
face of the alkene moiety on which the H+ or H− is delivered for
the formation of the new stereogenic centre. Under the practi-
cal point of view, it is useful to identify the structural requisites
influencing this arrangement, in order to achieve the control on
the final stereochemical outcome of the reaction through a guided
substrate-engineering strategy.
The results obtained from this study have been also used to bet-
ter define the empirical model we have recently developed [6a] to
rationalise the experimental data on OYE1-3 mediated reductions
of activated alkenes, and to describe the structural prerequisites for
the optimal arrangement of the substrates within the binding site
of the enzymes. A distinction could be made between trisubstituted
alkenes for which the activating EWG was linked to the prostere-
ogenic carbon atom, and those activated by the other functional
group.
Scheme 2. Synthesis of cyanoesters (E)-2. Reagents and conditions: (i) Me3SiCN,
DMSO-water 5:1; (ii) SOCl2, Py, 100 ◦C; (iii) cyanoacetic acid, cat. NH4OAc in toluene;
(iv) NaBH4 in water and NaHCO3; and (v) CHOCOORꢀ, Ac2O, Py.
2.3. Bioreduction procedure for the preparation of
monodeuterated samples
A solution of the substrate in i-PrOH (100 ␮L, 500 mM) was
added to potassium phosphate buffer solution (5.0 mL, 50 mM in
D2O, pH 7.0) containing glucose (20 ␮mol), NADH (75 ␮mol) and
the required OYE (1.11 ␮M). The mixture was incubated for 24 h in
an orbital shaker (160 rpm, 30 ◦C). The solution was extracted with
EtOAc (3× 5.0 mL), centrifuging after each extraction (3000 × g,
1.5 min), and the combined organic solutions were dried over anhy-
drous Na2SO4.
3. Results and discussion
2. Experimental
The known data [6b] of the bioreductions of compounds 2 can
be summarised as follows. The best values of conversion and enan-
tioselectivity were achieved by using OYE1-3. (E)-stereoisomers
were generally converted into (S)-reduction products with high
enantioselectivity, whereas the conversion of (Z)-cyano esters into
the corresponding (R)-enantiomers decreased with the increasing
steric demand of the substrates. Reducing the size of the ester
moiety from ethyl to methyl had a strong positive impact on con-
versions, thus suggesting activation due to the ester group. An
interesting case of exceptional behaviour of OYE3 was observed
as for the enantioselectivity of the reduction of the cyano ester
precursor of pregabalin.
We decided to start a systematic investigation of the stereo-
chemical course of these reactions, by preparing other substrates
of type 2, showing (E) stereochemistry at the double bond and
increasing steric hindrance either at the ester moiety (2a–d) or
at the prostereogenic olefinic carbon atom (2e–h). Compound 2i
was also prepared to investigate the peculiar enantioselectivity of
OYE-mediated reductions of isobutyl substituted cyanoesters.
A different synthetic approach was chosen with respect to
that described in ref. 6b, in order to optimise the preparation of
(E)-stereoisomers of cyano esters 2. Derivatives (E)-2a–e were syn-
thesised starting from the corresponding ␤-keto ester according to
route 1 (Scheme 2) which afforded only the desired configuration of
the C C double bond. Route 2 was employed to obtain compounds
(E)-2f–i, which were recovered from the reaction mixture and sep-
arated from the (Z)-stereoisomer by column chromatography.
The absolute configuration of the reduced products was deter-
mined by conversion either in the dimethyl ester or in the diacid
derivatives (by treatment with refluxing methanol and a catalytic
2.1. General procedure for OYE1–3 mediated biotransformations
of substrates (E)-2a–i (screening)
A solution of the substrate in DMSO (10 ␮L, 500 mM) was
added to a potassium phosphate buffer solution (1.0 mL, 50 mM,
pH 7.0) containing glucose (20 ␮mol), NADP+ (0.1 ␮mol), GDH
(4 U) and the required OYE (0.89 ␮M). The mixture was incubated
for 24 h in an orbital shaker (160 rpm, 30 ◦C). The solution was
extracted with EtOAc (2× 250 ␮L), centrifuging after each extrac-
tion (15,000 × g, 1.5 min), and the combined organic solutions were
dried over anhydrous Na2SO4. Two replicates were performed for
each biotransformation: no significant differences were observed
for conversion and enantiomeric excess values.
2.2. General procedure for OYE1–3 mediated biotransformations
(50 mg scale)
For substrates (E)-2a–h a similar protocol was followed on a
larger scale, employing the OYE which provided the best conversion
and/or ee, in order to isolate and characterise the correspond-
ing reduced product. A solution of the suitable cyano ester in
i-PrOH (1 mL, 250 mM) was added to a potassium phosphate buffer
solution (25 mL, 50 mM, pH 7.0) containing the required OYE
(0.89–2.66 ␮M), GDH (100 U), glucose (1 mmol, 180 mg) and NADP+
(5 ␮mol, 3.7 mg). The reaction was monitored by GC until complete
conversion. The mixture was then extracted with EtOAc (3× 10 mL)
and submitted to column chromatography (n-hexane with increas-
ing amount of EtOAc), when conversion was not complete in order
to purify the reduced product from the starting material.

E. Brenna et al. / Journal of Molecular Catalysis B: Enzymatic 101 (2014) 67–72
69
Scheme 4. Results of OYE1-mediated biotransformations of derivatives (E)-1a and
(Z)-1b.
Scheme 3. Empirical model for the ER-mediated reductions of olefins activated by
carbonyl/carboxyl moieties linked to the prostereogenic carbon atom (G is the sub-
stituent linked to the C O moiety of the EWG; ␣ and ␤ are the substituents in
alpha and beta position with respect to the EWG; S and L stand for small and large,
respectively).
Conversely, when G is large and ␣ is small, the “classical” binding
mode (Scheme 3b) is preferentially adopted. The optimal stereo-
chemistry of the double bond is the one depicted in Scheme 3b
with the large substituent at C␤ (L) on the same side of the EWG,
i.e. (Z) stereochemistry for most substrates.
amount of sulphuric acid, or by reaction with refluxing concen-
trated hydrochloric acid, respectively), or by comparison with
known specific optical rotation data.
For example, for substrates (E)-1a and (Z)-1b (Scheme 4) the
activating group was identified to be the ester moiety by deuter-
ation experiments: as a matter of fact, the use of D2O as a solvent
of the OYE-mediated reduction, in the presence of a stoichiomet-
ric quantity of NADH is known [9] to promote the incorporation
of a deuterium atom at C␣ with respect to the activating group.
This information, combined with the absolute configuration of the
final products, i.e. (S)-3a and (R)-3b respectively, allowed then the
binding mode to be defined: a flipped arrangement for (E)-1a and a
classical one for (Z)-1b. In order to give the most consistent repre-
sentation of the active site, we adopted the convention of drawing
the substrates in such a way that the activating EWG is always
shown on the left and highlighted in boldface character. According
to this representation, the hydride is always delivered from below
the plane, while the proton (or in this particular case the D+) is
captured from above.
The bioreduction of the corresponding stereoisomers, (Z)-1a
and (E)-1b, occurred with lower conversions and enantioselectiv-
ities, because the stereochemistry of the double bond was not the
optimal one for the binding mode induced by the steric hindrance
of G and ␣.
In order to make analogous considerations on the most reac-
tive binding mode of substrates 2a–i it was necessary to establish
whether the activating functional group was the nitrile or the ester
moiety. With this aim, derivatives (E)-2b, g and i were submitted
to deuteration experiments: the analysis of the 1H NMR spectra of
the products (Fig. 1b, and S1b and S2b in the Supplementary Data)
showed that in the OYE2-mediated reduction of these substrates,
affording the (S) enantiomer of the corresponding monodeuter-
ated reduced products 4b, g, and i, the activating group was the
CN moiety linked to the most substituted olefinic carbon atom
(Scheme 5).
Unexpectedly, the nitrile function resulted to be involved in the
hydrogen bond interaction within the enzyme active site. The bind-
ing of the CN group to the amino acid residues is preferred to the
coordination of the carboxylic moiety, in spite of the peculiar linear
geometry of the carbon nitrogen triple bond.
The substrate binding modes established for the OYE2-mediated
reduction of compounds of (E)-2b, g, and i are in accordance with
the empirical model we have proposed (Scheme 3): the activating
CN moiety has no G group and the steric hindrance of the ␣ sub-
stituent determines the arrangement. Thus, a flipped binding mode
results to be the most reactive one and the (E)-stereochemistry
of the substrate is the most suitable one for this binding
mode.
The results of OYE 1–3 biotransformations are reported in
Table 1. The analysis of these data showed that a decrease in enan-
tioselectivity readily occurred when the steric hindrance of the
alkyl substituent linked to the stereogenic carbon atom increased,
except for compound (E)-2i. A decrease of conversion values was
observed only in the reduction of the branched derivatives (E)-2g
and i, probably as a consequence of a reduced reaction rate. As for
substrates 2a–d, characterised by a methyl group on the more sub-
stituted olefinic carbon atom, the increasing length of the alkoxy
chain perturbed neither the enantiomeric excess, nor the conver-
sion values.
We considered these data with regard to the stereochemi-
cal analysis we had previously performed on the bioreduction of
regioisomeric cyano esters 1. We were also interested in the inves-
tigation of the reactions of substrate (E)-2g, for which a change
in enantiopreference had been already observed [6b] by compar-
ing the results obtained with OYE 1 and 2 with those collected
with OYE3. For this reason we prepared also the corresponding
propyl ester 2i which was described to afford invariably the (S)-
enantiomer by OYE1-3 mediated reductions.If the activating EWG,
i.e. the one which is involved in the formation of hydrogen bonds
with His191 and/or Asn194 in the active site of OYE1-3 (numbering
according to PDB structure code 1OYB), contains a C O group and
it is linked to the prostereogenic olefinic carbon atom, two binding
modes have been described in the literature till now: a “classical”
and a “flipped” one (Scheme 3) [3c,7]. Recently [8], a modifica-
tion of the arrangement of the substrate in the enzyme active site
induced by rotation of the alkenyl fragment on the C␣-CO single
bond has been hypothesised. This conformational change produces
a different orientation of the substrate in the so-called “flipped”
binding mode: the stereoheterotopic faces exposed to H+ and H−
addition are indeed the same.Our past experience on regioisomeric
cyano esters 1 had allowed us to draw the conclusions which are
depicted in Scheme 3 for trisubstituted alkenes, activated by an
EWG containing a carbonyl/carboxyl moiety. In the drawing G rep-
resents the group linked to the C O moiety of the EWG, and ␣ and
␤ are the substituents of the double bond in alpha and beta posi-
tion with respect to the EWG. The so-called “flipped” binding mode
(Scheme 3a) is preferred when G is small (S) and ␣ is large (L). The
favoured stereochemistry of the double bond is the one shown in
the picture with the large substituent at C␤ (L) on the opposite side
of the activating group, i.e. (E) stereochemistry for most substrates.

70
E. Brenna et al. / Journal of Molecular Catalysis B: Enzymatic 101 (2014) 67–72
Table 1
Results of OYE1-3-mediated reductions of substrates (E)-2a–i.
Substrate
OYE 1
ca (%)
OYE2
ca (%)
OYE3
ca (%)
R
COORꢀ
eeb (%)
eeb (%)
eeb (%)
Me
Me
Me
Me
COOMe
COOEt
(E)-2a
(E)-2b
(E)-2c
(E)-2d
(E)-2e
(E)-2f
(E)-2g
(E)-2h
(E)-2i
99
99
99
84
99
99
46
94
56
99 (S)
99 (S)
99 (S)
99 (S)
99 (S)
60 (S)
50 (S)
44 (S)
99 (S)
99
99
99
98
99
99
53
78
57
99 (S)
99 (S)
99 (S)
99 (S)
99 (S)
70 (S)
76 (S)
69 (S)
99 (S)
99
99
99
99
91
92
60
91
27
99 (S)
99 (S)
99 (S)
99 (S)
66 (S)
20 (S)
88 (R)
rac
COOnBu
COOnHex
COOMe
COOMe
COOMe
COOMe
COOnPr
Et
n-Bu
i-Bu
n-Pent
i-Bu
99 (S)
a
c = conversion percentage, calculated by GC analysis of the crude mixture after 24 h reaction time (these values are not significantly different from percentage yields, as
neither degradation nor side-product formation was observed; isolation yields are reported in the Supplementary Data).
b
Calculated by GC analysis on a chiral stationary phase.
Scheme 5. Results of deuteration experiments on substrates (E)-2b, g and i per-
formed in the presence of OYE2.
(R)-4g with c = 60% and ee = 88%. Deuteration experiments (c = 90%
and ee = 91%) highlighted that in this latter reaction the activating
group was no longer the nitrile, but the ester function (Fig. 1c and
Scheme 6a).
In the series of cyanoesters of type 1 we had observed an
exchange of the activating EWG between the ester and the
nitrile moieties in the OYE1-mediated reduction of substrate
(Z)-1a, to afford (R)-3a with ee = 72% (Scheme 6b). To our knowl-
Fig. 1. (a) 1H NMR spectrum of rac-4g; (b) 1H NMR spectrum of (S)-4g-d1 obtained
from (E)-2g by OYE2-mediated reduction in D2O in the presence of stoichiometric
NADH (c = 99%, ee = 71%); (c) 1H NMR spectrum of (R)-4g-d1 obtained from (E)-2g by
OYE3-mediated reduction in D2O in the presence of stoichiometric NADH (c = 90%,
ee = 91%).
edge, in the literature there are only two more examples of
␣,␤-difunctionalised alkenes for which it was experimentally
demonstrated that the activating EWG was the one on the less
substituted olefinic carbon atom: the two amido fumarates (Z)-5a
and b in OYE3 reductions [9] (Scheme 6c), and the nitroacrylates
(Z)-6a-c treated with OYE1 [10] (Scheme 6d).
In all these cases the substituent ␣ linked to the carbon atom
bearing the activating EWG is only a hydrogen atom, and the steric
hindrance of the two groups at ␤ position controls the arrangement
of the substrate in the binding pocket. A flipped binding mode is
assumed by alkenes (E)-2g and (Z)-5a,b in the active site of OYE3,
and a classical arrangement is adopted by (Z)-1a and (Z)-6a–c in
the binding pocket of OYE1.
Within this class of compounds, the increase of the bulkiness
of the ␣ group induces a decrease in enantioselectivity, that had
not been observed for derivatives 1: e.g. (E)-1a was converted by
OYE 1–3 into (S)-3a with ee = 99%, whereas (E)-2f was converted
into poorly enriched (ee = 20–70%) compounds (S)-4f. This decrease
in enantioselectivity reflects a less definite preference for a cer-
tain binding mode and the possibility of achieving more than one
arrangement within the binding pocket of the enzyme, affording
opposite enantiomers upon reduction.
An interesting behaviour was shown by the two isobutyl deriva-
tives (E)-2g and i. Compound 2g was converted into (S)-4g in
modest conversion yields (c = 46–53%) and ee values (ee = 50–76%)
by OYE1 and 2, whereas the OYE3-mediated reaction afforded the
We tried to rationalise these results by taking also into account
the literature data for OYE1-3 mediated reductions of trisubstituted
monofunctionalised alkenes showing no substituents at the carbon
atom linked to the activating EWG. The available data are shown

E. Brenna et al. / Journal of Molecular Catalysis B: Enzymatic 101 (2014) 67–72
71
Scheme 8. Empirical model for the ER-mediated reductions of olefins activated by
an EWG linked to the less substituted olefinic carbon atom.
which the largest group at the prostereogenic carbon atom is on
the opposite side with respect to the EWG (Scheme 7).
The position of the most hindered group at the ␤ carbon atom
is maintained also in the arrangement of the (Z)-stereoisomers, as
it has been shown for (Z)-␤-methylcinnamaldehyde and for (Z)-
2-methyl-1-nitrooct-1-ene, causing a flipped arrangement to be
adopted (Scheme 8), and the same absolute configuration of the
resulting reduced product to be obtained (Scheme 7). However, the
optimal stereochemistry of the double bond, affording the high-
est values of enantioselectivity, seems to be the one in which the
largest group in ␤ position is on the opposite side to the activating
EWG.
In the case of substrate (E)-2i, OYE1-3-mediated reductions gave
the (S)-reduced product in modest yields and very high enantio-
selectivity (ee = 99%). The deuteration experiments were repeated
also with OYE3 and highlighted that the CN moiety was still the
activating group. The distribution of substituents on the double
bond has here two effects: (i) it limits the conformational free-
dom of the substrate, and (ii) it is however suitable for finding a
preferred arrangement. Enantioselectivity values are related to the
preference for a particular enzyme-substrate complex. In the case
of compound 2i the steric hindrance of substituents makes the for-
mation of the complex a more energy demanding step, but it also
limits the possible arrangement.
Scheme 6. Difunctionalised alkenes activated by the EWG on the less substituted
carbon atom (this work, Refs. [6a,9,10]).
in Scheme 7, some of which have been collected employing whole
cells of baker’s yeast (BY, Saccharomyces cerevisiae, which expresses
OYE2 and OYE3) rather than isolated ERs. BY and OYE1-3-mediated
reductions of (E)-1-nitro-3-aryl-1-propene derivatives afforded the
(R)-enantiomer of the reduced products [11], and BY converted
(E)-3-arylbut-2-enals into the corresponding (S)-saturated primary
alcohols [12].
As for derivatives 2a–d, their reduction data are in agreement
with the fact that being the nitrile the activating group, no effect is
caused by the increasing steric hindrance at the ester moiety.
For these (E)-substrates the only activating EWG (NO2 or CHO)
is linked to less substituted olefinic carbon atom, and the reac-
tions occur through the same preferred classical binding mode, in
Scheme 7. Literature data of OYE1-3-mediated reductions of ␤,␤ꢀ-disubstituted-␣-monofunctionalised alkenes (Ref. [11,12]).

72
E. Brenna et al. / Journal of Molecular Catalysis B: Enzymatic 101 (2014) 67–72
4. Conclusions
York) is kindly acknowledged for the gift of plasmid pT7-OYE1. Prof.
Sven Panke and Christian Femmer (Department of Biosystems Sci-
ence and Engineering, ETH, Basel) are kindly acknowledged for the
help provided in the preparation of the plasmids and the overex-
pressing strains. D.Z. thanks CNPq (Brazilian Agency) Number of
Process 236786/2012-1 for financial support.
The experimental data show that trisubstituted ␣,␤-
difunctionalised alkenes, such as cyano esters 2, and, as we
have previously shown, cyano esters 1 [6a] and diesters [5], have a
preference for binding in the enzyme active site through the EWG
linked to the prostereogenic olefinic carbon atom. The most reac-
tive binding mode does not seem to be influenced by the electronic
and geometric characteristics of the activating functional group:
no preference is shown for trigonal functionalities, and even the
linear CN can afford a suitable hydrogen bond interaction within
the enzyme active site.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.molcatb.
2013.12.020.
All the information so far collected on the factors controlling the
stereochemical outcome of OYE1-3-mediated reactions has been
summarised in figure S3 (see Supplementary Data). When the acti-
vating group contains a carbonyl/carboxyl moiety and it is linked to
the prostereogenic carbon atom, the steric hindrance of the ␣ and
G fragments control the arrangement of the substrate. A flipped or
classical arrangement is adopted according to the relative dimen-
sions of the two groups: bulky G groups promote a switching of
the binding mode and a subsequent change of the absolute config-
uration of the stereogenic centre. The two binding modes, defined
according to the spatial position of the activating EWG which estab-
lishes on which stereoheterotopic face of the alkene the H+ addition
occurs, are characterised by a preference for a certain stereochem-
istry of the C C double bond. This empirical model (Figure S3) can
be extended also to alkenes for which the activating EWG is a CN
moiety with no G group.
If the activating group is linked to the less substituted carbon
atom, then ␣ is a hydrogen atom, and the absolute configuration
of the new stereogenic centre is controlled by the structural requi-
site of positioning the most sterically demanding group at C␤ in a
well-established spatial arrangement (Figure S3). According to this
condition, (E) and (Z) stereoisomers afford the same enantiomer by
adopting a classical and flipped binding mode, respectively, with
the (E)-substrates usually giving the highest enantioselectivity val-
ues.
References
[1] F.G. Gatti, F. Parmeggiani, A. Sacchetti, in: E. Brenna (Ed.), Synthetic Methods
for Biologically Active Molecules – Exploiting the Potential of Bioreductions,
Wiley-VCH, Weinheim, 2014, p. 49 (Chapter 3);
E. Brenna, C. Fuganti, F.G. Gatti, S. Serra, Chem. Rev. 111 (2011) 4036–4072;
M. Hall, C.K. Winkler, G. Tasnádi, K. Faber, in: J. Whittall, P.W. Sutton (Eds.),
Practical Methods for Biocatalysis and Biotransformations, vol. 2, John Wiley &
Sons, West Sussex (UK), 2012, pp. 87 (Chapter 3.1);
E. Brenna, F.G. Gatti, F. Parmeggiani, in: J. Whittall, P.W. Sutton (Eds.), Practical
Methods for Biocatalysis and Biotransformations, vol. 2, John Wiley & Sons,
West Sussex (UK), 2012, pp. 96 (Chapter 3.2).
[2] (a) E. Brenna, C. Fuganti, F.G. Gatti, F. Parmeggiani, Tetrahedron: Asymmetry
20 (2009) 2594–2599;
(b) E. Brenna, C. Fuganti, F.G. Gatti, F. Parmeggiani, Tetrahedron: Asymmetry
20 (2009) 2694–2698;
(c) D. Acetti, E. Brenna, C. Fuganti, F.G. Gatti, S.S. Serra, Eur. J. Org. Chem. (2010)
142–151;
(d) G. Tasnádi, M. Hall, in: E. Brenna (Ed.), Synthetic Methods for Biologically,
Wiley-VCH, Weinheim, 2014, p. 339 (Chapter 13);
(e) H. Toogood, N.S. Scrutton, Catal. Sci. Technol. 3 (2013) 2182–2194.
[3] (a) K.M. Fox, P.A. Karplus, Structure 2 (1994) 1089–1105;
(b) B.J. Brown, Z. Deng, P.A. Karplus, V. Massey, J. Biol. Chem. 273 (1998)
32753–32762;
(c) R.M. Kohli, V. Massey, J. Biol. Chem. 273 (1998) 32763–32770.
[4] (a) M. Utaka, S. Konishi, T. Ohkubo, S. Tsuboi, A. Takeda, Tetrahedron Lett. 28
(1987) (1447);
(b) M. Utaka, S. Konishi, A. Mizouka, T. Ohkubo, T. Sakai, S. Tsuboi, A. Takeda, J.
Org. Chem. 54 (1989) 4989–4992;
(c) E. Brenna, G. Fronza, C. Fuganti, D. Monti, F. Parmeggiani, J. Mol. Catal. B
Enzymatic 73 (2011) 17–21;
(d) E. Brenna, F.G. Gatti, A. Manfredi, D. Monti, F. Parmeggiani, Eur. J. Org. Chem.
(2011) 4015–4022;
(e) E. Brenna, F.G. Gatti, A. Manfredi, D. Monti, F. Parmeggiani, Org. Proc. Res.
Dev. 16 (2012) 262–268.
In the case of difunctionalised alkenes only a few examples
of switching of the activating EWG was observed, thus preven-
ting from establishing the structural requisites promoting this
exchange, which causes a variation of enantioselectivity. When the
activating EWG is on the less substituted olefinic carbon atom, the
empirical model for monofunctionalised alkenes can be employed
to rationalise the stereochemical outcome. The condition is to
accommodate the largest group at C␤ in a specific spatial region
of the pocket, thus resulting in a classical or flipped binding mode
according to the stereochemistry of the double bond.
[5] E. Brenna, F.G. Gatti, A. Manfredi, D. Monti, F. Parmeggiani, Adv. Synth. Catal.
354 (2012) 2859–2864.
[6] (a) E. Brenna, F.G. Gatti, A. Manfredi, D. Monti, F. Parmeggiani, Catal. Sci. Technol.
3 (2013) 1136–1146;
(b) C.K. Winkler, D. Clay, S. Davies, P. O’Neill, P. McDaid, S. Debarge, J. Steflik,
M. Karmilowicz, J.W. Wong, K. Faber, J. Org. Chem. 78 (2013) 1525–1533.
[7] H. Toogood, J.M. Gardiner, N.S. Scrutton, ChemCatChem 2 (2010) 892–914;
S.K. Padhi, D.J. Bougioukou, J.D. Stewart, J. Am. Chem. Soc. 131 (2009)
3271–3280.
[8] G. Oberdorfer, K. Gruber, K. Faber, M. Hall, Synlett 23 (2012) 1857–1864.
[9] C. Stueckler, C.K. Winkler, M. Hall, B. Hauer, M. Bonnekessel, K. Zangger, K.
Faber, Adv. Synth. Catal. 353 (2011) 1169–1173.
[10] M.A. Swiderska, J.D. Stewart, Org. Lett. 8 (2006) 6131–6133.
[11] (a) M. Hall, C. Stueckler, B. Hauer, R. Stuermer, T. Friedrich, M. Breuer, W. Kroutil,
K. Faber, Eur. J. Org. Chem. (2008) 1511–1516;
All the information herein collected can help in defining new
potential substrates for the optimal enantioselective reduction of
the double bond, in the aim of obtaining a control on enantiose-
lectivity and conversion through a wise structural definition of the
substrate, as a complementary approach to the optimisation of new
stereodivergent ERs.
(b) E. Brenna, G. Fronza, C. Fuganti, F.G. Gatti, Eur. J. Org. Chem. (2010)
5077–5084;
(c) H. Ohta, N. Kobayashi, K. Ozaki, J. Org. Chem. 54 (1989) 1802–1804.
[12] (a) G. Fronza, C. Fuganti, S. Serra, Eur. J. Org. Chem. (2009) 6160–6171;
(b) A. Abate, E. Brenna, C. Dei Negri, C. Fuganti, S. Serra, Tetrahedron: Asymme-
try 13 (2002) 899–904;
Acknowledgements
(c) C. Fuganti, S. Serra, J. Chem. Soc., Perkin Trans. 1 (2000) 3758–3764.
Prof. Claudio Fuganti is warmly acknowledged for fruitful dis-
cussions. Prof. Neil C. Bruce (Department of Biology, University of