Enzymatic synthesis of optically active lactones via asymmetric bioreduction using ene-reductases from the old yellow enzyme family

Article abstract of DOI:10.1002/adsc.201500094

In contrast to the widely studied asymmetric bioreduction of α,β-unsaturated carboxylic acid esters catalyzed by ene-reductases, the reaction applied to lactones remains unexplored. A broad set of ene-reductases was found to reduce various α-, β- and γ-substituted α,β-unsaturated butyrolactones to yield the corresponding saturated non-racemic lactones. Substitution patterns greatly influenced activities and stereoselectivities and lactone products were obtained in moderate to excellent yields; importantly, enzyme-based stereocontrol allowed access to both enantiomers in up to >99% ee. Chiral recognition of a distant γ-center led to kinetic resolution with remarkable enantioselectivities (E values up to 49). An unprecedented case of dynamic kinetic resolution was observed with 3-methyl-5-phenylfuran-2(5H)-one, whereby spontaneous racemization of the substrate furnished the product in up to 73% conversion and >99% ee and 96% de.

Full text of DOI:10.1002/adsc.201500094

FULL PAPERS
DOI: 10.1002/adsc.201500094
Very Important Publication
Enzymatic Synthesis of Optically Active Lactones via
Asymmetric Bioreduction using Ene-Reductases from the Old
Yellow Enzyme Family
Nikolaus G. Turrini,a MØlanie Hall,a and Kurt Fabera,*
a
Department of Chemistry, University of Graz, Heinrichstrasse 28, 8010 Graz, Austria
Fax: (+43)-316-380-9840; phone (+43)-316-380-5332; e-mail: kurt.faber@uni-graz.at
Received: February 13, 2015; Revised: April 10, 2015; Published online: May 7, 2015
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201500094.
Abstract: In contrast to the widely studied asymmet- both enantiomers in up to >99% ee. Chiral recogni-
ric bioreduction of a,b-unsaturated carboxylic acid tion of a distant g-center led to kinetic resolution
esters catalyzed by ene-reductases, the reaction ap- with remarkable enantioselectivities (E values up to
plied to lactones remains unexplored. A broad set of 49). An unprecedented case of dynamic kinetic reso-
ene-reductases was found to reduce various a-, b- lution was observed with 3-methyl-5-phenylfuran-
and g-substituted a,b-unsaturated butyrolactones to 2(5H)-one, whereby spontaneous racemization of the
yield the corresponding saturated non-racemic lac- substrate furnished the product in up to 73% conver-
tones. Substitution patterns greatly influenced activi- sion and >99% ee and 96% de.
ties and stereoselectivities and lactone products were
obtained in moderate to excellent yields; important- Keywords: asymmetric bioreduction; dynamic kinetic
ly, enzyme-based stereocontrol allowed access to resolution; ene-reductases; optically active lactones
Introduction
Marchantia polymorpha and reductases p83 and p51
from Glycine max (soy) were shown to catalyze the
Optically active lactones are important synthons pres- stereoselective reduction of substituted 2-butenolides
ent in a broad range of natural products, such as anti- at the expense of NAD(P)H.[10] Recently, a chemoen-
biotics, plant hormones, fragrances and flavor com- zymatic one-pot, two-step synthesis combining two re-
pounds.[1–3] Derivatives of g-butyrolactones, for in- duction reactions was developed: The asymmetric re-
stance, are responsible for the flavor of many fruits, duction of the C=C bond of g-keto-a,b-unsaturated
such as raspberries, strawberries, peaches, apricots or carboxylic acid esters by ene-reductases, followed by
mangos.[1–3] The lactone moiety is also a common the ADH-catalyzed reduction of the ketone moiety,
structural component of various macrocyclic antibiot- yielded the corresponding g-hydroxy carboxylic acid
ics[4,5] (e.g., erythromycin and clarithromycin), while esters, which underwent spontaneous lactonization.
lovastatin is a potent inhibitor of HMG-CoA-reduc- An analogous sequence initiated by ADH was ap-
tase and is used to treat hypercholesterolemia.[6] At plied to d-keto esters.[11]
present, only few biocatalytic strategies have been de-
Ene-reductases of the Old Yellow Enzyme (OYE)
veloped for the synthesis of optically active g- or d- family of flavoproteins are known for their broad sub-
lactones. For the most part, they rely on lipase-cata- strate scope. They are able to reduce a wide range of
lyzed kinetic resolution of g- or d-hydroxycarboxylic activated alkenes, such as aldehydes, ketones, nitroal-
esters or amides, followed by (spontaneous) lactoniza- kenes, carboxylic acids and esters, imides and nitriles,
tion.[7] In contrast, fewer asymmetric synthesis proto- employing nicotinamide [NAD(P)H] as hydride
cols exist: the Baeyer–Villiger monooxygenase-cata- source.[12] Due to their exquisite chemo-, regio- and
lyzed oxidation of cyclic ketones to asymmetric lac- stereoselectivities, ene-reductases have been imple-
tones is well established,[8] while bakerꢀs yeast cells mented in the synthesis of several industrially rele-
have been used in bioreduction approaches.[9] Overall, vant molecules.[13] While a,b-unsaturated carboxylic
bioreduction protocols of a,b-unsaturated lactones acid esters have been thoroughly investigated,[14] their
with reductases isolated from plants or microorgan- cyclic counterparts have not been studied with isolat-
isms remain underexploited. Reductases p68 from ed enzymes. Aiming to extend the product repertoire
Adv. Synth. Catal. 2015, 357, 1861 – 1871
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1861

FULL PAPERS
Nikolaus G. Turrini et al.
Results and Discussion
Bioreduction of a,b-Unsaturated Lactones
Several a,b-unsaturated lactones bearing various sub-
stituents in the a-, b-, and/or g-position were tested as
substrates in the asymmetric bioreduction reaction
with ene-reductases of the OYE family (Scheme 1).
Initial screenings were performed with stoichiometric
amounts of NADH. For selected substrates, attempts
to improve conversions/selectivities were undertaken
using cofactor-recycling based on glucose/glucose de-
hydrogenase (GDH).
The non-substituted candidate furan-2(5H)-one
(1a) was broadly accepted by all OYE homologues
tested, except for OPR1 (Table 1). Excellent conver-
sion levels were obtained with OYE 1 and OYE 2 (>
99%, Table 1, entries 1 and 2) and NCR (81%,
Table 1, entry 4). Reduced activities of YqjM and
XenA in buffer (19% and 34%, respectively, see
Table S1 in the Supporting Information) could be im-
proved by addition of toluene (20% v/v) as cosolvent
(27% and 56%, respectively, Table 1, entries 5 and
Scheme 1. Bioreduction of a,b-unsaturated lactones with
ene-reductases and a panel of accepted substrates.
to optically active lactones, a broad set of a,b-unsatu- 8).[15] In a previous study on the ene-reductase-cata-
rated lactones displaying various substitution patterns lyzed isomerization of a-methylene-g-butyrolactone
was tested as substrates with purified OYE homo- (exo-2a) to 3-methylfuran-2(5H)-one (2a),[16] the
logues and the influence of reaction conditions on ac- latter was found to be slowly reduced by few ene-re-
tivities and stereoselectivities was studied (Scheme 1). ductases. Consequently, conversions of 2a remained
low (max. 20% with OYEs 1 and 2, Table 1, entries 1
and 2), regardless of the reaction conditions. Conver-
Table 1. Bioreduction of a,b-unsaturated butyrolactones and coumarins with ene-reductases.[a]
Entry Enzyme
c. [%] c. [%] eeP [%]
c. [%] eeP [%]
c. [%] eeP [%] c. [%][b] eeS [%][c] c. [%]
c. [%]
1
2
3
4
5
6
7
8
9
OYE1
OYE2
OYE3
NCR
YqjM
OPR1
OPR3
XenA
NerA
>99
>99
27
20
32 (R)
<1%
42[h]
54
–
<1%
–
17
2
3
17
<1
5
84 (S)
95 (R)
18 (R)
88 (S)
rac
16 (R)
52 (R)
<1
>99[i]
78[i]
7[i]
87
8
<1
<1
97
15
40
2
97
12
2
11
84
n.d.
20[e]
11[f]
18*
12*
48 (R)[e]
78 (R)[f]
13 (R)*
>99 (R)[h] 93
>99 (R)
96 (R)
54 (S)
–
52 (S)
12 (R)
>99 (R)
66
81
>99[g] 81 (R)[g]
>99
<1%
7
3
12
27[d]
>99 (R)* <1%
>99 (R)* <1%
–
–
–
–
<1% 4*
16
56[d]
4
<1%
4*
3
–
<1%
>99 (R)[f] <1%
n.d. >99
23
>99 (S) <1
83 (R) 95
>99 (R)
33
82 (R)
n.d.
[a]
Screening experiments were performed with stoichiometic amounts of NADH, selected experiments (marked with *)
were done with GDH-based NADH-recycling system.
Based on recovered material after extraction from the aqueous phase.
Product always racemic.
20% toluene.
30% n-hexane.
10% i-PrOH.
[b]
[c]
[d]
[e]
[f]
[g]
[h]
10% EtOAc.
30% EtOAc.[i] 10% t-BuOMe; n.d.=not determined.
1862
asc.wiley-vch.de
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 1861 – 1871

FULL PAPERS
Enzymatic Synthesis of Optically Active Lactones via Asymmetric Bioreduction
Table 2. Product distribution from the biotransformation of 3a with ene-reductases (Scheme 2).
Entry
Enzyme
Conditions
3a [%]
exo-2a [%]
2a [%]
2b [%]
1
2
3
4
5
6
7
8
OYE2
OYE2
OYE2
OYE3
OYE3
OYE3
NCR
1.5 equiv. NADH
NADH-GDH[a]
3 equiv. NADH
1.5 equiv. NADH
NADH-GDH[a]
3 equiv. NADH
1.5 equiv. NADH
NADH-GDH[a]
3 equiv. NADH
1.5 equiv. NADH
3 equiv. NADH
1.5 equiv. NADH
1.5 equiv. NADH
1.5 equiv. NADH
1.5 equiv. NADH
1.5 equiv. NADH
<1
<1
15
<1
46
18
3
<1
18
<1
35
32
5
<1
48
41
8
91
84
25
95
24
27
11
15
16
10
61
85
58
51
13
70
6
16
53
5
36
40
49
50
55
49
7
10
27
1
<1
4
<1
<1
4
<1
<1
<1
<1
<1
<1
<1
NCR
NCR
9
10
11
12
13
14
15
16
NerA
NerA
XenA
OPR1
OPR3
YqjM
control
34
36
35
29
41
[a]
2 equiv. glucose.
Scheme 2. Biotransformation cascade of 3a mediated by ene-reductases via C=C reduction of 3a, followed by spontaneous
HBr elimination. Isomerization of exo-2a to 2b was shown to be thermodynamically favored and the reverse reaction cannot
take place.[16]
sions could be improved by employing a glucose de- gested: (i) bioreduction of 3a to 3b catalyzed by ene-
hydrogenase (GDH)-based NADH-recycling system, reductase is followed by (ii) spontaneous dehydroha-
while excellent stereoselectivities were conserved: logenation of 3b (owing to the acidity of the a-pro-
both YqjM and OPR1 yielded (R)-2b in 99% ee with ton),[14a,b] leading to the formation of exo-2a; (iii) sub-
12% and 18% conversion, respectively (Table 1, en- sequent enzyme-catalyzed isomerization, which only
tries 5 and 6). The b-methyl-analogue 17a (see requires catalytic amounts of NADH,[16] yields 2a,
Figure 4) was not a substrate for ene-reductases, which iv) upon further bioreduction delivers the end
going in line with previously reported activities on product 2b. The product distribution depends on the
alkyl-substituted a,b-unsaturated carboxylic acid catalyst used (Table 2). With OYE2, the small amount
esters, which are considered as weakly activated bor- of end product 2b obtained using 1.5 equiv. of NADH
derline substrates.[14] In an attempt to enhance the re- (7%, Table 2, entry 1) could be increased by employ-
activity of lactones, additional (halogenated) electron- ing a GDH-based NADH-recycling system or 3 equiv.
withdrawing groups were introduced:[14b] 3-(Bromo- of NADH (10% and 27%, respectively, Table 2, en-
methyl)furan-2(5H)-one (3a) and its b-substituted an- tries 2 and 3). This strategy allowed us to shift the cas-
alogue 4-(bromomethyl)furan-2(5H)-one (4a) were cade towards formation of end product 2b and dimin-
therefore synthesized. In addition, compound 5a, ished the accumulation of exo-2a (Table 2, entries 4–
bearing a phenylthiomethyl-substituent in the b-posi- 9). Interestingly, starting the cascade from 3a gave
tion was also investigated, as it was previously shown substantially more product 2b with OYE2 when com-
to be reduced by bakerꢀs yeast.[17] Bioreduction of 3a pared to the reduction of 2a (27% and 20%, respec-
did not result in formation of the expected saturated tively, Table 2, entry 3 and Supporting Information,
product 3b; instead, formation of 3-methylfuran- Table S1). In contrast, NerA, OPR1-3, XenA and
2(5H)-one (2a) as major product was observed, along YqjM were poorly active. Spontaneous dehydrohalo-
with various amounts of its bioreduction product 2b genation of 3b was dominant and could not be sup-
and exo-2a (Table 2 and Scheme 2). Based on the pressed in the presence of a biphasic system (data not
mechanism of reductive dehalogenation of haloacrylic shown).
esters,[14a,b] a-halomethyl enones[18a] and OYE-mediat-
ed C=C-isomerization,[16] the following cascade is sug- tion drastically reduced substrate acceptance by all
Shifting the substitution from the a- to the b-posi-
Adv. Synth. Catal. 2015, 357, 1861 – 1871
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1863

FULL PAPERS
Nikolaus G. Turrini et al.
Figure 1. Kinetic resolution of rac-5-oxo-2,5-dihydrofuran-2-yl acetate (6a) via enantioselective bioreduction and reductive
dehalogenation of rac-7a.
ene-reductases.
4-(Bromomethyl)furan-2(5H)-one distant g-chiral center was observed with rac-5-oxo-
(4a) was accepted by only four enzymes, and quantita- 2,5-dihydrofuran-2-yl acetate (6a) (Figure 1). Except
tive conversion was obtained with NCR and NerA for OPR1 (5% conversion, Table 3, entry 15), all en-
(Table 1, entries 4 and 9), while OYE2 and OYE3 zymes were able to reduce 6a to a varying extent and
were only moderately active (21% and 45% conver- product 6b could be recovered with low enantiomeric
sion, respectively, Table 1, entries 2 and 3). OYE2, excess (up to 28% ee with NerA, Table 3, entry 16).
OYE3 and NerA displayed perfect stereoselectivity The combination of apparent full conversion with
forming (R)-4b (ee >99%). Poor stereorecognition measurable enantioenrichment of the product could
with NCR in aqueous buffer (28% ee, Supporting In- be attributed to substrate hydrolysis [also observed
formation, Table S1) could be significantly improved with 5-hydroxyfuran-2(5H)-one
– 5-hydroxyfuran-
via medium engineering by adding EtOAc (10% v/v) 2(5H)-one was unsuitable as substrate due to sponta-
as co-solvent (81% ee, Table 1, entry 4) without af- neous hydrolysis in buffer leading to poor recovery
fecting the conversion. Spontaneous dehalogenation from the aqueous solution (data not shown)], causing
yielding 4-methylfuran-2(5H)-one (also observed in incomplete material recovery from the aqueous
the control reaction in up to 15%) was less dominant phase. In an attempt to suppress the hydrolytic side
compared to 3a.
reaction and increase the enzymatic selectivity at the
Inspired by a report on the stereoselective reduc- same time, organic cosolvents were employed.[19] Ad-
tion of 5a by bakerꢀs yeast,[17] which yielded (R)-5b in ditionally, time studies were conducted in order to
moderate yield and 99% ee, isolated ene-reductases monitor the kinetic resolution process. Shorter reac-
were tested. OYE2 and OYE3 (both from yeast) and tions times and use of a biphasic system allowed the
NCR were the most active enzymes (up to >99% isolation of substrate and/or product in high ee
conversion with NCR, Table 1, entry 4) and stereo- values: (S)-6b was obtained in 94% ee and 31% con-
complementary activities could be identified: Where- version after 1 h reaction time in 30% t-BuOMe with
as OYE2 afforded (R)-5b in >99% ee, XenA yielded OYE3 (Table 3, entry 4), while application of XenA
the opposite (S)-enantiomer in >99% ee, albeit with in 30% t-BuOMe allowed the recovery of 40% of un-
lower conversion (93% and 12%, respectively, reacted (R)-6a in >99% ee (Table 3, entry 10). The
Table 1, entries 2 and 8). Compared to whole yeast enantiopreference of OYE3 could be greatly im-
cells,[17] the use of purified OYE2 provided significant- proved by variation of the cosolvent: The enantiomer-
ly improved yield and reduced reaction time (93% ic ratio (E value[20]) was increased from 18 (in 10% i-
versus 41% conversion, and 24 h versus 5 days, respec- PrOH) to 49 in 30% t-BuOMe. This effect is in line
tively).
with the observation that (less polar) ethers are gen-
erally more biocompatible compared to (more polar)
alcohols.[21]
Kinetic Resolution of g-Substituted a,b-Unsaturated
Lactones
Double substitution at both the g-position and on
the C=C bond potentially leads to kinetic resolution
combined with asymmetric reduction in a single trans-
g-Substituted a,b-unsaturated lactones were of partic- formation. Therefore, 3-bromo-5-ethoxyfuran-2(5H)-
ular interest in view of the possibility to perform a ki- one (7a), bearing an additional halogen as activating
netic resolution, which, to the best of our knowledge, group, was tested (Figure 1). Although conversion
has not been reported yet for ene-reductases (some was observed with most enzymes, the expected satu-
kinetic preference was observed with bakerꢀs yeast on rated product 7b was not detected and the corre-
racemic d-2-decenolide[18b]). Stereorecognition of the sponding dehalogenated a,b-unsaturated compound
1864
asc.wiley-vch.de
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 1861 – 1871

FULL PAPERS
Enzymatic Synthesis of Optically Active Lactones via Asymmetric Bioreduction
Table 3. Kinetic resolution of 5-oxo-2,5-dihydrofuran-2-yl acetate (rac-6a) via enantioselective bioreduction.
Entry
Enzyme
Conditions
Conv. [%][a]
eeS [%]
eeP [%]
E[20]
1
2
3
4
5
6
7
8
OYE3
OYE3
OYE3
OYE3
OYE3
XenA
XenA
XenA
XenA
XenA
OYE1
OYE2
NCR
24 h, buffer
>99
17
27
32
52
60
16
52
39
–
rac
~1
39
18
49
47
34
38
32
33
23
~1
~1
~1
~1
~1
~1
~1
1 h, 10% i-PrOH
2 h, 10% i-PrOH
1 h, 30% t-BuOMe
3 h, 30% t-BuOMe
24 h, buffer
1 h, 10% i-PrOH
2 h, 10% i-PrOH
1 h, 30% t-BuOMe
3 h, 30% t-BuOMe
24 h, buffer
24 h, buffer
24 h, buffer
24 h, buffer
24 h, buffer
24 h, buffer
24 h, buffer
18 (R)
34 (R)
42 (R)
94 (R)
>99 (R)
16 (R)
87 (R)
54 (R)
>99 (R)
–
–
–
–
rac
–
–
94 (S)
86 (S)
94 (S)
86 (S)
74 (S)
94 (S)
84 (S)
90 (S)
65 (S)
18 (R)
10 (S)
18 (S)
10 (S)
8 (S)
9
10
11
12
13
14
15
15
16
61
>99
>99
>99
>99
5
YqjM
OPR1
OPR3
NerA
>99
>99
10 (S)
28 (S)
[a]
Max. 6% deviation from conversion calculated by [eeS/(eeS +eeP)].[20b]
Table 4. Bioreduction of rac-8a (Figure 2) and rac-exo-8a (Figure 3).
Entry
Enzyme
Conv. [%][a]
eeS [%]
Substrate rac-8a
eeP [%]
deP [%]
1
2
3
4
5
6
7
8
OYE1
OYE2
OYE3
NCR
YqjM
OPR1
OPR3
XenA
60
6
5
73
5
6
78 (4R)
17 (4S)
7 (4S)
76 (4R)
rac
rac
rac
rac
>99 (2R,4S)
44 (2R,4S)
54 (2R,4S)
>99 (2R,4S)
20 (2S,4R)
24 (2S,4R)
42 (2R,4S)
>99 (2R,4S)
92 (trans)
65 (trans)
64 (trans)
96 (trans)
48 (trans)
26 (cis)
4
2
>99 (trans)
>99 (trans)
Substrate rac-exo-8a
9
10
OYE2[b]
OYE2[c]
50
66
74 (4S)
82 (4S)
>99 (2R,4R)
>99 (2R,4R)
98 (cis)
97 (cis)
[a]
Based on total recovered products using dichloromethane spiked with limonene as internal standard for extraction.
NCR showed <5% conversion, ee and de values not determined.
2 equiv. NADH and double enzyme amount compared to standard conditions; n.d.=not determined.
[b]
[c]
5-ethoxyfuran-2(5H)-one (7c) was isolated as sole of product 7c to the corresponding carboxylic acid,
product in racemic form. The dehalogenation most which caused low recovery and thereby mimics low
likely proceeds via bioreduction yielding intermediate conversion, as observed with 5-hydroxyfuran-2(5H)-
7b, which undergoes spontaneous HBr elimination to one (see above). Addition of cosolvents led to re-
furnish 7c in analogy to the sequence suggested for duced conversion levels without any influence on se-
substrate 3a. Most enzymes showed enantioprefer- lectivity (data not shown).
ence for the (S)-enantiomer of 7a and the recovered
unreacted (R)-substrate was obtained in 95% ee with
OYE2, while the opposite stereopreference was ob- Dynamic Kinetic Resolution
served with NCR yielding (S)-7a in 88% ee (Table 1,
entries 2 and 4). These high ee values for the non-con- The reduction of substrates rac-8a and rac-exo-8a,
verted substrate are only achievable in a kinetic reso- which are isomeric regarding the position of the C=C
lution near or beyond 50% conversion. However, this bond, potentially leads to the formation of enantio-
was not the case (measured value <20%). This dis- meric pairs of cis/trans-diastereoisomers. Both sub-
crepancy may be explained by spontaneous hydrolysis strates underwent enzyme-catalyzed kinetic resolu-
Adv. Synth. Catal. 2015, 357, 1861 – 1871
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1865

FULL PAPERS
Nikolaus G. Turrini et al.
Figure 2. Dynamic kinetic resolution of 8a catalyzed by ene-reductases.
Figure 3. Kinetic resolution of rac-exo-8a and C=C bond isomerization as side reaction mediated by ene-reductases.
tion, but major differences were observed (see cellent enantio- and diastereoselectivity and convert-
Figure 2 and Figure 3, Table 4): Compound rac-8a was ed (4R)-exo-8a to cis-(2R,4R)-8b, by leaving unreact-
broadly accepted by several ene-reductases to yield ed (4S)-exo-8a in 82% ee behind (no racemization ob-
trans-8b (as confirmed by NOESY-NMR experiments served, Supporting Information, Figure S1). The ste-
and comparison with published coupling constants[22]
)
reochemistry of cis-(2R,4R)-8b was unequivocally
in an enantiocomplementary fashion. OYEs 1-3, identified via NOESY-NMR experiments and validat-
NCR, OPR3 and XenA preferred (4S)-8a yielding ed with published coupling constants.[22] (4R)- and
(2R,4S)-8b (Table 4, entries 1–4, 7 and 8) in up to (4S)-8a were formed in minor amounts [7% and 5%,
>99% ee and 96% de. YqjM and OPR1 showed the respectively] as a result of C=C double bond isomeri-
opposite enantiopreference and formed trans-(2S,4R)- zation catalyzed by OYE2 (Figure 3).[16] The conver-
8b, however, in low yield and optical purity (Table 4, sion could be boosted to 66% by using 2 equiv. of
entries 5 and 6). The unexpected high conversion NADH and doubling the enzyme concentration
levels observed with OYE1 and NCR (60 and 73%, (Table 4, entry 10).
respectively) indicate dynamic kinetic resolution due
Finally, coumarin-type substrates were investigated.
to spontaneous racemization of the starting material Simple derivatives, such as 9a and 10a, were accepted
8a, which is facilitated via a resonance-stabilized fura- by most enzymes with varying degrees of conversion
nol intermediate (Figure 2). This hypothesis was con- (Table 2) and cosolvent addition (10% t-BuOMe)
firmed by incubating the recovered non-converted greatly enhanced the conversion of 9a with OYE1
(4R)-substrate 8a in aqueous buffer and monitoring (from 21% in buffer to >99%, Table 1, entry 1, and
ee over time. After 24 h, the substrate was in all cases Supporting Information, Table S1) and OYE2 (from
almost racemic (Supporting Information, Tables S2 26% in buffer to 78%, Table 1, entry 2, and Support-
and S3), independent of the conditions. Such a dynam- ing Information, Table S1). Unfortunately, substrates
ic kinetic resolution process via C=C bond reduction, with more complex substitution pattern were unstable
which enables the production of single diastereoiso- and decomposed in buffer (Figure 4).
mers bearing two distant chiral centers from racemic
starting material, has not been reported with ene-re-
ductases. The high ee values of non-converted (4R)-8a Non-Substrates
indicate that spontaneous racemization is slower than
the enantioselective enzymatic reduction.
Despite the broad acceptance of substituents along
In contrast to rac-8a, isomer exo-8a was accepted the lactone ring, several substrates were found to be
only by OYE2 and NCR with 50% and <5% conver- non-reactive (Figure 4). The electronic nature of the
sion, respectively (Table 4, entry 9). OYE2 showed ex- substituents, such as methoxy or electron-rich aromat-
1866
asc.wiley-vch.de
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 1861 – 1871

FULL PAPERS
Enzymatic Synthesis of Optically Active Lactones via Asymmetric Bioreduction
termined by comparison with authentic reference ma-
terial obtained by palladium-catalyzed allylic substitu-
tion of (R)-6a to (R)-7c (see the Experimental
Section).[26] The absolute configuration of exo-8a was
assigned based on published elution order and reten-
tion times of both enantiomers on HPLC using an
OD-H column: The enantiomer reacting with OYE2
could be identified as (R)-exo-8a by using identical
analytical conditions.[22d] The absolute configuration
of 8a-enantiomers could be deduced via selective iso-
merization of exo-8a with OYE2 using a catalytic
amount of cofactor,[16] whereby (R)-exo-8a yields (R)-
8a, as the g-center is not involved in the reaction. Fi-
nally, the absolute configuration of 8b-stereoisomers
was determined by assignment of the relative configu-
ration via NMR[22] and identification of the reacting
enantiomer (exo-8a or 8a) (see Experimental Section
for further details). This was additionally confirmed
by comparing the optical rotation of isolated product
trans-8b (obtained from reaction upscale of 8a with
NCR) with literature data.[27]
Figure 4. Non-substrates for ene-reductases.
Conclusions
ic substituents, most likely led to deactivation of the
double bond by reducing or totally quenching the par- A novel biocatalytic route for the synthesis of optical-
tial positive charge on C-b, which is a prerequisite for ly active lactones was established via asymmetric bio-
effective hydride transfer from N-5 of the reduced reduction of a,b-unsaturated lactones with isolated
flavin.[12,14] Illustrative examples of electronic deacti- and purified ene-reductases from the Old Yellow
vation, combined with steric bulk, include 11a, 27a Enzyme family. Variation of the substitution pattern,
and 28a. Additionally, a highly conjugated system and which strongly influenced enzyme activities, allowed
rigid structure made compound 25a a non-substrate. the formation of products with chiral centers posi-
Finally, few substrates were unstable and decomposed tioned along the ring in a-, b- and g-positions. The re-
in buffer (e.g., anhydrides 24a and 25a as well as cou- activity of substrates and stereocontrol could be influ-
marins 18a–23a).
enced by (i) modifying the degree of C=C bond acti-
vation by introduction of electron-withdrawing sub-
stituents, (ii) varying the position of substituents, (iii)
selection of the biocatalyst and (iv) medium engineer-
ing employing a biphasic system and/or cofactor recy-
Determination of Absolute Configurations
Absolute configurations were assigned by co-injection cling system. Chiral centers in the distant g-position
with authentic reference materials, comparison of re- were well recognized and unprecedented cases of ki-
tention times and elution orders with literature data netic resolution and dynamic kinetic resolution could
(under identical experimental conditions) or by mea- be identified with racemic g-substituted substrates.
surement of optical rotations and comparison with lit- Overall, the reduction of a,b-unsaturated g-lactones
erature values. The absolute configuration of 2b was using ene-reductases offers a biocatalytic alternative
determined based on the known (R)-stereoselectivity for the production of optically active lactones and
of OYE3 in the reduction of substrate 2a.[16] The abso- non-racemic g-hydroxy carboxylic acids. The use of
lute configurations of 4b and 5b were determined by lactones successfully circumvents the non-reactivity of
performing the bioreduction of 4a and 5a with bakerꢀs a,b-unsaturated carboxylic acids with ene-reductases.
yeast with known stereochemistry.[17,23] Similarly, the
absolute configurations of 6a and 6b were determined
via kinetic resolution catalyzed by lipases according
to a published protocol.[24] For 7a, palladium-catalyzed
dehalogenation was employed to convert enantioen-
riched substrate 7a (recovered from enzymatic reac-
tion with OPR3) to 7c with retention of configura-
Experimental Section
General Remarks
Substrates 1a, 6a and 7a were kindly provided by BASF
tion.[25] The absolute configuration of 7c was then de- (Ludwigshafen). Substrates 2a and 9a as well as product 2b
Adv. Synth. Catal. 2015, 357, 1861 – 1871
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1867

FULL PAPERS
Nikolaus G. Turrini et al.
and reference exo-2a were obtained from Sigma–Aldrich.
3-Methylene-5-phenyldihydrofuran-2(3H)-one
(exo-8a):
Prepared according to the literature.[31] A yellow oil was ob-
tained; yield: 345.8 mg (1.98 mmol, 84%). 1H NMR
(300 MHz, CDCl3): d=7.3–7.4 (m, 5H), 6.3 (t, J=2.8 Hz,
1H), 5.7 (t, J=2.5 Hz, 1H), 5.5 (dd, J=7.9, 6.7 Hz, 1H),
3.45–3.35 (dd, J=17.1, 8.1 Hz, 1H), 3.0–2.9 (m, 1H);
13C NMR (75 MHz, CDCl3): d=170.3 139.9, 134.3, 129.0,
128.7, 125.5, 122.6, 78.1, 78.1, 36.4.
Substrate 10a was a kind gift from Wolfgang Stadlbauer
(Graz). GC-MS analyses were performed on an HP Agilent
Technologies 6890 Series GC system equipped with a 5973
mass selective detector and a 7683 Series injector. GC-FID
measurements were carried out on an HP Agilent Technolo-
gies 7890A GC system equipped with an FID detector and
a 7693 Autosampler or a 7683B Injector in combination
with a 7683 Series Autosampler. For HPLC analysis, a Shi-
madzu LC-20AD HPLC system with a DGU-20 A5 degas-
ser, a SIL-20AC autosampler, SPD-M20A diode array de-
tector and a CTO-20AC column oven, equipped with a Chir-
alcel OD-H column (25 cm”0.46 cm) was used. NMR spec-
tra were measured on a Bruker Avance III 300 MHz NMR
spectrometer. Chemical shifts are reported relative to TMS
(d=0.00 ppm); coupling constants J are given in Hz. All en-
zymes were overexpressed and purified as reported[28] and
used as purified protein (>90% purity). Optical rotations
were measured on a Perkin–Elmer Polarimeter 341. HR-MS
were measured on a Thermo Q-Exactive mass spectrometer
instrument.
3-Methyl-5-phenylfuran-2(5H)-one (8a): Prepared accord-
ing to the literature,[32] the product was obtained as a yellow
oil; yield: 3.08 g (17.7 mmol, 79%). 1H NMR (300 MHz,
CDCl3): d=7.3–7.4 (m, 3H), 7.2–7.3 (m, 2H), 7.1 (m, 1H),
5.9–5.8 (m, 1H), 2.0 (t, J=1.8 Hz, 3H); 13C NMR (75 MHz,
CDCl3): d=174.4, 148.5, 135.3, 129.6, 129.1, 127.0, 126.6,
82.2, 10.7.
Synthesis of Reference Materials
Rac-2b and exo-2a were purchased from Sigma–Aldrich.
Racemic products 1b and 6b were prepared by catalytic hy-
drogenation with Pd/C (10%) and H2 (1 atm) in dry THF
and/or MeOH as described.[33]
Standard Procedure for Bioreduction Reactions
Dihydrofuran-2(3H)-one (1b):[34] Yield: 31 mg (0.35 mmol,
To a solution of Tris-HCl buffer (800 mL, 50 mM, pH 7.5)
containing the substrate (10 mM) and the cofactor NADH
(15 mM) was added an aliquot of the respective enzyme
(final protein concentration 100 mgmLÀ1). When a GDH re-
cycling system was employed, the buffer contained 100 mM
NAD+, 20 mM glucose and 10 U GDH. Substrates were
added from a 0.5M stock solution in DMSO to overcome
solubility problems (final concentration of DMSO 1%). The
mixture was shaken for 24 h at 308C and 120 rpm. After-
wards, the reaction mixture was extracted with EtOAc (2”
500 mL), the combined organic layers were dried over
Na2SO4 and transferred into GC-vials. Products were identi-
fied by comparison with authentic materials (commercially
available or synthesized according to published literature,
see below for details). (R)-Limonene (5 mM) was used as in-
ternal standard.
1
64%); colorless oil. H NMR (300 MHz, CDCl3): d=2.2 (m,
2H, CbH2), 2.5 (t, 2H, J=7.8, CaH2), 4.3 (t, 2H, J=7.0,
CgH2); 13C NMR (75 MHz, CDCl3): d=22.3 (Cb), 27.9 (Ca),
68.6 (Cg), 177.8 (C=O).
5-Oxotetrahydrofuran-2-yl acetate (6b): Yield: 46.9 mg
(0.32 mmol, 92%); colorless oil. 1H NMR (300 MHz,
CDCl3): d=2.1, (s, 3H, CH3), 2.2–2.3, (m, 1H), 2.4–2.6, (m,
2H), 2.6–2.8, (m, 1H), 6.6, s, 1H, CgH); 13C NMR (75 MHz,
CDCl3): d=21.0 (CH3), 25.9 (Cb), 27.9 (Ca), 95.0 (Cg), 169.2
(CH3-C=O), 175.6 (O=Clactone); HR-MS (ESI) m/z=
145.0493, calculated for [M+H]+: 145.04952, m/z=167.0312,
calculated for [M+Na]+: 167.03149.
5-Ethoxyfuran-2(5H)-one (7c): Prepared according to the
literature;[35] yield: 20.5 mg (0.16 mmol, 22%); orange oil.
1H NMR (300 MHz, CDCl3): d=7.2 (dd, J=5.7, 1.2 Hz,
1H), 6.2 (dd, J=5.7, 1.2 Hz, 1H), 5.9 (t, J=1.2 Hz, 1H),
4.0–3.8 (m, 1H), 3.8–3.7 (m, 1H), 1.3 (t, J=7.1 Hz, 3H);
13C NMR (75 MHz, CDCl3): d=170.7, 150.5, 125.1, 103.3,
66.3, 15.1.
Synthesis of Substrates
3-(Bromomethyl)furan-2(5H)-one (3a): Prepared according
to the procedure established by Hodgson et al.[29] After pu-
rification, the product was obtained; yield: 1.61 g
(9.15 mmol, 18%). 1H NMR (300 MHz, CDCl3): d=4.1 (s,
2H, Br-CH2), 4.9, (s, 2H, CgH2), 7.5, (s, 1H, CbH); 13C NMR
(75 MHz, CDCl3): d=20.9 (H2C-Br), 70.2 (Cg), 131.0 (Ca),
149.1 (Cb), 171.5 (C=O).
3-Methyl-5-phenyldihydrofuran-2(3H)-one (8b): Prepared
according to Huang et al.[35] The product was obtained as
a yellow-brownish oil; yield: 80 mg (0.45 mmol, 45%).
1H NMR (300 MHz, CDCl3): d=7.5–7.1 (m, 7H), 5.6 (dd,
J=7.6, 4.7 Hz, 0.4H), 5.4–5.3 (m, 0.6H), 2.9–2.7 (m, 48H),
2.5–2.3 (m, 2H), 2.2 (s, 3H), 2.1 (s, 0.5H), 1.5 (dd, J=5.4,
2.3 Hz, 0.5H), 1.4 (dd, J=8.2, 4.3 Hz, 3H); 13C NMR
(75 MHz, CDCl3): d=179.9, 179.2, 139.8, 139.1, 129.0, 128.8,
128.7, 128.5, 128.2, 128.2, 125.5, 125.3, 125.0, 40.0, 38.4, 36.4,
33.6, 15.4, 15.0, 8.8. The cis- and trans-stereoisomers of 8b
were separated by preparative TLC on silica gel (hexanes/
Et2O 2:1, detection: UV 254 nm and KMnO4 staining): Rf
(trans-8b): 0.30, Rf (cis-8b): 0.38.
4-(Bromomethyl)furan-2(5H)-one (4a): Prepared accord-
ing to the literature.[30] Yield: 486 mg (2.76 mmol, 20%);
yellow oil. 1H NMR (300 MHz, CDCl3): d=4.2, (d, J=
0.9 Hz, 2H, Br-CH2), 4.9, (m, 2H, CgH2), 6.1, (ddd, J=3.1,
1.9, 1.2 Hz, 1H, CaH); 13C NMR (75 MHz, CDCl3): d=22.6
(H2C-Br), 72.0 (Cg), 119.0 (Ca), 163.3 (Cb), 172.6 (C=O).
4-[(Phenylthio)methyl]furan-2(5H)-one (5a): Synthesized
starting from 4a.[17] The desired product was obtained; yield:
cis-8b:[22] 1H NMR (300 MHz, CDCl3): d=7.4–7.3 (m,
5H), 5.4 (dd, J=10.7, 5.3 Hz, 1H), 2.9–2.8 (m, 2H), 1.85
(dd, J=15.6, 10.7 Hz, 1H), 1.35 (d, J=6.7 Hz, 3H).
trans-8b:[22] 1H NMR (300 MHz, CDCl3): d=7.4–7.3 (m,
5H), 5.6 (dd, J=7.6, 4.7 Hz, 1H), 2.8–2.7 (m, 1H), 2.5–2.3
(m, 2H), 1.35 (d, J=7.2 Hz, 3H).
1
30.4 mg (26%, 0.15 mmol). H NMR (300 MHz, CDCl3): d=
3.8 (s, 2H, S-CH2), 4.8 (s, 2H, Cg-H), 5.8 (s, 1H, Ca-H), 7.3
(m, 5H, Hphenyl); 13C NMR (75 MHz, CDCl3): d=32.3 (S-
CH2), 72.2 (Cg), 117.9 (Ca), 128.1–133.5 (Cphenyl), 165.3 (Cb),
173.3 (C=O).
1868
asc.wiley-vch.de
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 1861 – 1871

FULL PAPERS
Enzymatic Synthesis of Optically Active Lactones via Asymmetric Bioreduction
was complete as determined by GC analysis and 7c was ob-
tained in 35% ee. The major enantiomer was attributed the
(R)-configuration by comparison/co-injection with enan-
tioenriched reference (R)-7c prepared as described below.
Procedure for the Preparation of (R)-4b
Optically enriched (R)-4b was obtained via upscaling of the
enzymatic bioreduction of 4a with bakerꢀs yeast. Upscaling
was performed in 1.5-mL Eppendorf reaction vessels (20”),
each containing 800 mL of Tris-HCl buffer (50 mM, pH 7.5)
and a GDH recycling system (20 mM glucose, 10 U GDH
from DSM), 100 mM NAD+, 10 mM 4a, and 20 mg of
bakerꢀs yeast (Spar NaturPurꢂ BioHefe) which was rehy-
drated in buffer for 30 min prior to the reaction. After 72 h
reaction time, (R)-4b was obtained; yield: 13.0 mg
(0.07 mmol, 35%). Comparison of the optical rotation ([a]2D0:
+13, c=0.69, CHCl3) with the literature value ([a]20: +43,
c=6.40, CHCl3) assigned the absolute configurationDas (R)-
4b.[23]
Procedure for the Preparation of (R)-7c via Pd-
Catalyzed Allylic Substitution of (R)-6a[26]
51.3 mg (0.36 mmol) of (R)-6a were dissolved in 10 mL ab-
solute EtOH, 4.0 mg (0.018 mmol) Pd(OAc)2 and 18.9 mg
(0.072 mmol) PPh3 were added and the resulting orange so-
lution was heated to reflux and stirred for 18 h. After filtra-
tion of the catalyst and evaporation of the solvent, purifica-
tion by flash column chromatography (hexane/EtOAc, 1:2,
v/v) afforded (R)-7c; yield: 35.9 mg (77.9%) in 30% ee.
Procedure for the Preparation of (R)-5b
Determination of the Absolute Configuration of exo-
8a
Reference material for (R)-5b was prepared by bioreduction
of 5a with bakerꢀs yeast using the conditions described for
the synthesis of (R)-4b in 2”1.5 mL. The product was inject-
ed into HPLC (see Analytical Methods) and was found to
be optically pure (>99% ee). This protocol was adapted
from the literature and allowed assignment of absolute con-
figuration.[17]
The absolute configuration of exo-8a enantiomers was as-
signed by comparing elution orders and retention times
measured via HPLC with published data under identical an-
alytical conditions, using a Daicel OD-H column (250 mm”
4.6 mm, 5 mm), n-heptane/i-PrOH (97.5:2.5 v/v, isocratic) at
208C.[22d]
Procedure for Preparation of (S)-6a and (R)-6a via
Determination of the Absolute Configuration of 8a
Lipase-Catalyzed Kinetic Resolution[24]
OYE2 was used for selective isomerization of exo-8a using
a catalytic amount of NADH based on a published proce-
dure.[16] The isomerization proceeded very slowly (~1% con-
version), whereby only (R)-exo-8a was consumed, and con-
sequently only one enantiomer of product 8a was detected
(>99% ee). This enantiomer was assigned the (R)-configu-
ration as the g-center is not involved in the reaction.
In a 2-mL Eppendorf tube, substrate 6a (10 mg, 0.07 mmol)
was dissolved in 1 mL n-heptane/n-butanol (6:1, v/v) and
5 mg of lipase PS (Amano) were added. The reaction was
shaken at room temperature for 24 h at 650 rpm. The reac-
tion was run under the same conditions with a prolonged re-
action time (52 h) using 5 mg of lipase AY (Amano) which
displays complementary stereoselectivity. After removal of
the enzyme preparation by centrifugation, followed by filtra-
tion with Celite, the samples were analyzed by chiral GC
and the absolute configuration of non-converted substrate
was assigned by comparison with literature data.[24] Lipase
PS: 64% conversion, ee 98% (R), E=14; lipase AY: 53%
conversion, ee 68% (S), E=8.
Determination of the Absolute Configuration of
trans-8b
8a (10 mM) was converted by NCR (1 aliquot, 100 mm)
under standard conditions on
a 16-mL scale (20 mM
NADH) for 72 h. Additional batches of enzyme and NADH
were added after 24 h and 48 h. The reaction mixture was
transferred into a separating funnel, extracted with CH2Cl2
(4”5 mL), the combined organic layers were dried with
sodium sulfate and the solvent was evaporated after filtra-
tion. Purification by preparative TLC (hexane/Et2O 2:1) af-
forded trans-8b; yield: 13.8 mg (39.2%) in 99% ee and 70%
de. Comparison of optical rotation data ([a]2D0: À6.8, c=0.69,
CHCl3; literature:[27] [a]D20: À19.5, c=1.1, CHCl3) allowed as-
signment of the (2R,4S)-configuration to 8b.
The reaction was scaled up to 20 mL with lipase PS in
order to produce enough material for palladium-catalyzed
allylic substitution of (R)-6a into (R)-7c. (R)-6a was recov-
ered with 98% ee.
Conversion of Enantioenriched 7a to 7c via Pd-
Catalyzed Dehalogenation[25]
7a (10 mM) was converted by OPR3 under standard condi-
tions on a 32-mL scale. After work-up and solvent evapora-
tion, 6 mg of raw product were recovered, and GC analysis
revealed a content of 20% rac-7c and 80% (R)-7a with 55%
ee. In a flame-dried 25-mL two-neck round-bottom flask,
6 mg (0.029 mmol) of the above crude mixture were dis-
solved in 3 mL dry acetonitrile. 1.8 mg (1.55 mmol)
[Pd(PPh3)4] and 104.5 mL EtN(i-Pr)2 were added and the re-
action was heated to reflux and stirred for 20 h. The reaction
mixture was then diluted with 8 mL saturated aqueous
NH4Cl, extracted with EtOAc (3”3 mL), the combined or-
ganic layers were dried with sodium sulfate and concentrat-
ed in vacuo, yielding 4.8 mg of crude product. Conversion
Analytical Methods
Determination of Conversion: Conversions of all substrates
(except of 5a) were determined by GC-FID. The columns
used were a) Agilent J&W DB 1701 (30 m”0.25 mm,
0.25 mm film) with 1, 2 and 3; b) Agilent J&W HP-5 column
(30 m”0.32 mm, 0.25 mm film) with 4, 6, 7, 8, 9 and 10. Tem-
perature program for 1: 608C, hold for 5 min, 28CminÀ1 to
808C, hold for 5 min, 208CminÀ1 to 1108C, hold for 5 min;
retention times: 1a 20.5 min, 1b 19.8 min. Temperature pro-
gram for 2 and 3: 708C, hold for 1 min, 108CminÀ1 to
Adv. Synth. Catal. 2015, 357, 1861 – 1871
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1869

FULL PAPERS
Nikolaus G. Turrini et al.
1808C, 208CminÀ1 to 2808C, hold for 2 min; retention
times: 2a 8.0 min, 2b 7.5 min, 3a 13.5 min. Temperature pro-
gram for 4, 8, 9 and 10: 1008C, hold for 1 min, 108CminÀ1 to
2808C, hold for 2 min; retention times: 4a 5.8 min, 4b
5.3 min, exo-8a 9.4 min, 8a 9.2 min, trans-8b 8.9 min, cis-8b
9.0 min, 9a 7.4 min, 9b 6.7 min, 10a 8.2 min, 10b 7.5 min.
Temperature program for 6: 808C, hold for 1 min, 58CminÀ1
to 1608C, 208CminÀ1 to 2808C, hold for 2 min; retention
times: 6a 7.1 min, 6b 8.5 min. Temperature program for 7:
708C, 58CminÀ1 to 1208C, 108CminÀ1 to 2808C, hold for
2 min; retention times: 7a 9.8 min, 7c 6.1 min. Conversion of
5a was determined by HPLC using a Chiralcel OD-H
column (25 cm”0.46 cm). Conditions: flow: 1 mLminÀ1,
258C, n-heptane/i-PrOH (70:30, v/v, isocratic). Retention
times: 5a 15.0 min, (S)-5b 20.0 min, (R)-5b 22.5 min.
Determination of Enantiomeric Excess: GC-FID was
used for the determination of ee values except for 5b
(HPLC). Columns for GC-FID: c) Astec Chiraldex B-TA
(40 m”0.25 mm, 0.12 mm film) for 2; d) Restek Rt-bDEXse
(30 m”0.32 mm, 0.25 mm film) for 4 and 8; e) Chirasil Chir-
aldex DEX-CB (25 m”0.32 mm, 0.25 mm film) for 6 and 7.
Temperature program for 2: 1008C, 58CminÀ1 to 1458C,
18CminÀ1 to 1608C, 108CminÀ1 to 1808C, hold for 2 min;
retention times: 2a 14.0 min, (R)-2b 11.0 min, (S)-2b
11.2 min. Temperature program for 4a: 1008C, hold for
0.5 min, 58CminÀ1 to 1608C, 28CminÀ1 to 1758C,
208CminÀ1 to 2008C, hold for 5 min; retention times: 4a
17.7 min, (S)-4b 15.9 min, (R)-4b 16.0 min. Temperature pro-
gram for 6: 1008C, 38CminÀ1 to 1408C, 208CminÀ1 to
1808C; retention times: (R)-6a 5.9 min, (S)-6a 6.5 min, (R)-
6b 6.9 min, (S)-6b 7.5 min. Temperature program for 7:
708C, hold for 1 min, 58CminÀ1 to 1208C, 108CminÀ1 to
1808C, hold for 2 min; retention times: (R)-7a 13.4 min, (S)-
7a: 14.0 min, (R)-7c 9.8 min, (S)-7c 10.5 min. Temperature
program for exo-8a/8a: 1008C, 58CminÀ1 to 1608C, hold for
10 min, 108CminÀ1 to 1808C, hold for 5 min; retention
times: (4S)-exo-8a 23.0 min, (4R)-exo-8a 23.2 min, (4R)-8a
22.1 min, (4S)-8a 22.3 min, trans-(2R,4S)-8b 20.6 min, trans-
(2S,4R)-8b 20.7 min, cis-(2S,4S)-8b 21.6 min, cis-(2R,4R)-8b
21.8 min. HPLC conditions for the determination of enan-
tiomeric excess of 5b, see method for determination of con-
version. The absolute configuration of exo-8a was deter-
mined via HPLC, using a Daicel OD-H column (250 mm”
4.6 mm, 5 mm), n-heptane/i-PrOH (97.5:2.5 v/v, isocratic) at
208C; retention times: (4R)-exo-8a 19.9 min; (4S)-exo-8a
23.1 min.[22d]
Kenneth Jensen for HR-MS measurements. Bettina Nestl
(Stuttgart) is warmly acknowledged for providing additional
OYE1 and OYE3 samples. BASF SE (Ludwigshafen) and
Wolfgang Stadlbauer (Graz) are kindly acknowledged for the
donation of chemicals.
References
[1] G. Rousseau, Tetrahedron 1995, 51, 2777–2849.
[2] a) E. Brenna, C. Fuganti, S. Serra, Tetrahedron: Asym-
metry 2003, 14, 1–42; b) E. Brenna, C. Fuganti, F. G.
Gatti, S. Serra, Chem. Rev. 2011, 111, 4036–4072.
[3] A. Abate, E. Brenna, C. Dei Negri, C. Fuganti, S.
Serra, Tetrahedron: Asymmetry 2002, 13, 899–904.
[4] a) R. B. Woodward, E. Logusch, K. P. Nambiar, K.
Sakan, D. E. Ward, B. W. Au-Yeung, P. Balaram, L. J.
Browne, P. J. Card, C. H. Chen, J. Am. Chem. Soc.
1981, 103, 3210–3213; b) R. B. Woodward, B. W. Au-
Yeung, P. Balaram, L. J. Browne, D. E. Ward, P. J.
Card, C. H. Chen, J. Am. Chem. Soc. 1981, 103, 3213–
3215; c) R. B. Woodward, E. Logusch, K. P. Nambiar,
K. Sakan, D. E. Ward, B. W. Au-Yeung, P. Balaram,
L. J. Browne, P. J. Card, C. H. Chen, J. Am. Chem. Soc.
1981, 103, 3215–3217.
[5] I. J. Collins, J. Chem. Soc. Perkin Trans. 1 1999, 1377–
1395.
[6] A. W. Alberts, J. Chen, G. Kuron, V. Hunt, J. Huff, C.
Hoffman, J. Rothrock, M. Lopez, H. Joshua, E. Harris,
A. Patchett, R. Monaghan, S. Currie, E. Stapley, G.
Albers-Schonberg, O. Hensens, J. Hirshfield, K. Hoogs-
teen, J. Liesch, J. Springer, Proc. Natl. Acad. Sci. USA
1980, 77, 3957–3961.
[7] a) O. Pàmies, J.-E. Bäckvall, J. Org. Chem. 2002, 67,
1261–1265; b) W. Adam, P. Groer, C. R. Saha-Mçller,
Tetrahedron: Asymmetry 2000, 11, 2239–2243; c) Y. Shi-
motori, T. Miyakoshi, J. Oleo Sci. 2006, 55, 12, 629–635.
[8] a) M. D. Mihovilovic, B. Müller, P. Stanetty, Eur. J.
Org. Chem. 2002, 3711–3730; b) J. D. Stewart, K. W.
Reed, J. Zhu, G. Chen, M. M. Kayser, J. Org. Chem.
1996, 61, 7652–7653; c) M. T. Bes, R. Villa, S. M. Rob-
erts, P. W. H. Wan, A. Willetts, J. Mol. Catal. B:
Enzym. 1996, 1, 127–134.
[9] a) M. Aquino, S. Cardani, G. Fronza, C. Fuganti, R.
Fernandez, A. Tagliani, Tetrahedron 1991, 47, 7887–
7896; b) E. Brenna, C. Dei Negri, C. Fuganti, S. Serra,
Tetrahedron: Asymmetry 2001, 12, 1871–1879; c) K.
Takabe, H. Hiyoshi, H. Sawada, M. Tanaka, A. Miyaza-
ki, T. Yamada, T. Katagiri, H. Yoda, Tetrahedron:
Asymmetry 1992, 3, 1399–1400; d) G. Fronza, C. Fugan-
ti, P. Grasselli, M. Barbeni, Tetrahedron Lett. 1992, 33,
6375–6378; e) C. Fuganti, G. Pedrocchi-Fantoni, A.
Sarra, S. Servi, Tetrahedron: Asymmetry 1994, 5, 1135–
1138.
Acknowledgements
The research leading to these results has received funding
from the Innovative Medicines Initiative Joint Undertaking
under grant agreement no. 115360, resources of which are
composed of financial contribution from the European
Unionꢀs Seventh Framework Programme (FP7/2007-2013)
and EFPIA companies in kind contribution. This research is
part of project CHEM21 (https://www.chem21.eu/). The au-
thors would like to thank Barbara Grischek, Elisabeth Eger,
Bianca Grabner and Stefan Edelsbrunner for their excellent
technical assistance and Gµbor Tasnµdi for valuable advice
and discussion. Klaus Zangger and Bernd Werner are ac-
knowledged for their assistance in NMR measurements and
[10] a) K. Shimoda, N. Kubota, Tetrahedron: Asymmetry
2004, 15, 3827–3829; b) K. Shimoda, N. Kubota, T.
Hirata, Y. Kondo, H. Hamada, Tetrahedron Lett. 2007,
48, 1345–1347.
[11] a) M. Korpak, J. Pietruszka, Adv. Synth. Catal. 2011,
353, 1420–1424; b) T. Fischer, J. Pietruszka, Adv. Synth.
Catal. 2012, 354, 2521–2530; c) T. Classen, M. Korpak,
M. Schçlzel, J. Pietruszka, ACS Catal. 2014, 4, 1321–
1870
asc.wiley-vch.de
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 1861 – 1871

FULL PAPERS
Enzymatic Synthesis of Optically Active Lactones via Asymmetric Bioreduction
1331; d) E. Brenna, F. G. Gatti, D. Monti, F. Parmeg-
giani, A. Sacchetti, J. Valoti, J. Mol. Catal. B: Enzym.
2015, 114, 77–85.
[21] a) A. L. Serdakowski, J. S. Dordick, Trends Biotechnol.
2008, 26, 48–54; b) A. M. Klibanov, Nature 2001, 409,
241–246; c) G. Carrea, S. Riva, Angew. Chem. 2000,
112, 2312–2341; Angew. Chem. Int. Ed. 2000, 39, 2226–
2354.
[22] a) A. G. Myers, L. McKinstry, J. Org. Chem. 1996, 61,
2428–2440; b) A. Flores-Parra, D. M. Gutierrez-Avella,
Y. J. Guzman-Vazquez, A. Ariza-Castolo, R. Contreras,
J. Org. Chem. 1992, 57, 6067–6071; c) S. G. Davies, R.
Polywka, P. Wamer, Tetrahedron 1990, 46, 4847–4856;
d) M. Davoust, F. Cantagrel, P. Metzner, J. F. Briere,
Org. Biomol. Chem. 2008, 6, 1981–1993.
[12] a) R. Stuermer, B. Hauer, M. Hall, K. Faber, Curr.
Opin. Chem. Biol. 2007, 11, 203–213; b) H. S. Toogood,
J. M. Gardiner, N. S. Scrutton, ChemCatChem 2010, 2,
892–914; c) M. Hall, A. S. Bommarius, Chem. Rev.
2011, 111, 4088–4110.
[13] a) G. Tasnµdi, M. Hall, in: Synthetic Methods for Bio-
logically Active Molecules. Exploring the Potential of
Bioreductions, (Ed.: E. Brenna), Wiley-VCH, Wein-
heim, 2013, pp 329–374; b) C. K. Winkler, G. Tasnµdi,
D. Clay, M. Hall, K. Faber, J. Biotechnol. 2012, 162,
381–389.
[14] a) G. Tasnµdi, C. K. Winkler, D. Clay, M. Hall, K.
Faber, Catal. Sci. Technol. 2012, 2, 1548–1552; b) G.
Tasnµdi, C. K. Winkler, D. Clay, N. Sultana, W. M. F.
Fabian, M. Hall, K. Ditrich, K. Faber, Chem. Eur. J.
2012, 18, 10362–10367; c) E. Brenna, G. Fronza, C. Fu-
ganti, D. Monti, F. Parmeggiani, J. Mol. Catal. B:
Enzym. 2011, 73, 17–21; d) E. Brenna, F. G. Gatti, A.
Manfredi, D. Monti, F. Parmeggiani, Catal. Sci. Tech-
nol. 2013, 3, 1136–1146.
[15] Y. Yanto, C. K. Winkler, S. Lohr, M. Hall, K. Faber,
A. S. Bommarius, Org. Lett. 2011, 13, 2540–2543.
[16] K. Durchschein, S. Wallner, P. Macheroux, K. Zangger,
W. M. F. Fabian, K. Faber, ChemBioChem 2012, 13,
2346–2351.
[17] K. Takabe, H. Hiyoshi, H. Sawada, M. Tanaka, A.
Miyazaki, T. Yamada, T. Katagiri, H. Yoda, Tetrahe-
dron: Asymmetry 1992, 3, 1399–1400.
[18] a) B. R. S. de Paula, D. S. Zampieri, J. Augusto, R. Ro-
drigues, P. J. S. Moran, Tetrahedron: Asymmetry 2013,
24, 973–981; b) G. Fronza, C. Fuganti, P. Grasselli, Tet-
rahedron Lett. 1992, 33, 6375–6378.
[19] C. Stueckler, N. J. Mueller, C. K. Winkler, S. Glueck,
K. Gruber, G. Steinkellner, K. Faber, Dalton Trans.
2010, 39, 8472–8476.
[23] J. Schrçer, P. Welzel, Tetrahedron 1994, 50, 6839–6858.
[24] H. van der Deen, R. P. Hof, A. van Oeveren, B. L. Fer-
inga, R. M. Kellogg, Tetrahedron Lett. 1994, 35, 8441–
8444.
[25] F.-L. Xue, J. Qi, P. Peng, G.-Z. Mo, Z.-Y. Wang, Lett.
Org. Chem. 2014, 11, 64–79.
[26] H. van der Deen, A. van Oeveren, R. M. Kellogg, B. L.
Feringa, Tetrahedron Lett. 1999, 40, 1755–1758.
[27] A. K. Ghosh, K. Shurrush, S. Kulkarni, J. Org. Chem.
2009, 74, 4508–4518.
[28] C. K. Winkler, D. Clay, E. van Heerden, K. Faber, Bio-
technol. Bioeng. 2013, 110, 3085–3092.
[29] D. M. Hodgson, E. P. A. Talbot, B. P. Clark, Org. Lett.
2011, 13, 2594–2597.
[30] R. K. Boeckman, S. S. Ko, J. Am. Chem. Soc. 1982, 104,
1033–1041.
[31] H. Mattes, C. Benezra, Tetrahedron Lett. 1985, 26,
5697–5698.
[32] G. Leclerc, C.-G. Wermuth, J. Schreiber, Bull. Soc.
Chim. Fr. 1967, 4, 1302–1307.
[33] M. Hall, C. Stueckler, H. Ehammer, E. Pointner, G.
Oberdorfer, K. Gruber, B. Hauer, R. Stuermer, W.
Kroutil, P. Macheroux, K. Faber, Adv. Synth. Catal.
2008, 350, 411–418.
[34] S. Jhulki, S. Seth, M. Mondal, J. N. Moorthy, Tetrahe-
dron 2014, 70, 2286–2293.
[35] L. Huang, H. Jiang, C. Qi, X. Liu, J. Am. Chem. Soc.
2010, 132, 17652–17654.
[20] a) A. J. J. Straathof, J. A. Jongejan, Enzyme Microb.
Technol. 1997, 21, 559–571; b) tool for calculation of
enantiomeric ratio available (open access) at http://
biocatalysis.uni-graz.at/enantio/cgi-bin/enantio.pl.
Adv. Synth. Catal. 2015, 357, 1861 – 1871
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1871