Stereocomplementary bioreduction of α,β-unsaturated dicarboxylic acids and dimethyl esters using enoate reductases: Enzyme- And substrate-based stereocontrol

Article abstract of DOI:10.1021/ol7019185

Asymmetric bioreduction of α,β-unsaturated dicarboxylic acids, such as 2-methylmaleic/fumaric and 2-methylenesuccinic acid, as well as the corresponding dimethyl esters, using three cloned enoate reductases furnished 2-methylsuccinic acid or dimethyl 2-methylsuccinate, respectively. Opposite stereoisomeric products were obtained in up to >99% ee either by choice of the enzyme or by using E/Z-configurated substrates. Cofactor-recycling systems (NADH/FDH/formate, NADH/GDH/glucose or NADPH/G6PDH/glucose-6-phosphate) only worked in presence of a divalent metal ion, such as Ca²⁺, Mg ²⁺, or Zn²⁺.

Full text of DOI:10.1021/ol7019185

ORGANIC
LETTERS
2
007
Vol. 9, No. 26
409-5411
Stereocomplementary Bioreduction of
-Unsaturated Dicarboxylic Acids and
5
r,â
Dimethyl Esters using Enoate
Reductases: Enzyme- and
Substrate-Based Stereocontrol
†
†
‡
‡
Clemens Stueckler, M e´ lanie Hall, Heidemarie Ehammer, Eva Pointner,
†
‡
,†
Wolfgang Kroutil, Peter Macheroux, and Kurt Faber*
Department of Chemistry, Organic and Bioorganic Chemistry, UniVersity of Graz,
Heinrichstrasse 28, 8010 Graz, Austria, and Institute of Biochemistry, Graz UniVersity
of Technology, Petersgasse 12, 8010 Graz, Austria
kurt.faber@uni-graz.at
Received September 14, 2007
ABSTRACT
Asymmetric bioreduction of
r,â-unsaturated dicarboxylic acids, such as 2-methylmaleic/fumaric and 2-methylenesuccinic acid, as well as the
corresponding dimethyl esters, using three cloned enoate reductases furnished 2-methylsuccinic acid or dimethyl 2-methylsuccinate, respectively.
Opposite stereoisomeric products were obtained in up to >99% ee either by choice of the enzyme or by using E/Z-configurated substrates.
Cofactor-recycling systems (NADH/FDH/formate, NADH/GDH/glucose or NADPH/G6PDH/glucose-6-phosphate) only worked in presence of a
divalent metal ion, such as Ca2 , Mg2 , or Zn2
+
+
+
.
The asymmetric conjugate reduction of CdC bonds leads
to the creation of up to two stereogenic centers and is thus
a key transformation in asymmetric synthesis. Stereoselective
bioreduction of activated alkenes is catalyzed by flavin-
CR from the opposite side. As a consequence of the
stereochemistry of this mechanism, the overall addition of
[H ] proceeds in a strict trans-fashion. Enoate reductases are
2
able to reduce a wide variety of substrates, such as conjugated
1-3
dependent enoate reductases [EC 1.3.1.X],
members of
enals, enones, R,â-unsaturated carboxylic acids, imides,
4
1,3,6,7
the “old yellow enzyme” family at the expense of a
nicotinamide cofactor.1 The flavin cofactor transfers a
hydride onto Câ of the substrate while an essential Tyr
residue conserved along the OYE family adds a proton onto
nitroalkenes, and ynones.
,5
The asymmetric bioreduction of R,â-unsaturated carboxy-
lic acids was predominantly investigated by the group of H.
Simon using isolated enzymes or whole cells of strict (or
facultative) anaerobes, such as Clostridium spp., Proteus
mirabilis, Enterobacter agglomerans, Acetobacterium woo-
†
University of Graz.
Graz University of Technology.
‡
8-11
dii, and Sporomusa termitida.
Although the stereoselec-
(
1) Williams, R. E.; Bruce, N. C. Microbiology 2002, 148, 1607-1614.
(2) Steinbacher, S.; Stumpf, M.; Weinkauf, S.; Rohdich, F.; Bacher, A.;
tivities reported were impressive, practical application of
Simon, H. In FlaVins and FlaVoproteins; Chapman, S. K.; Perham, R. N.;
Scrutton, N. S. Weber: Berlin, 2002; pp 941-949.
(6) Hall, M.; Stueckler, C.; Kroutil, W.; Macheroux, P.; Faber, K. Angew.
Chem., Int. Ed. 2007, 46, 3934-3937.
(
3) Stuermer, R.; Hauer, B.; Hall, M.; Faber, K. Curr. Opin. Chem. Biol.
2
007, 11, 203-213.
(7) M u¨ ller, A.; St u¨ rmer, R.; Hauer, B.; Rosche, B. Angew. Chem., Int.
Ed. 2007, 46, 3316-3318.
(
(
4) Warburg, O.; Christian, W. Biochem. Z. 1933, 266, 377-411.
5) Karplus, P. A.; Fox, K. M.; Massey, V. FASEB J. 1995, 9, 1518-
(8) Tischer, W.; Bader, J.; Simon, H. Eur. J. Biochem. 1979, 97, 103-
112.
1
526.
1
0.1021/ol7019185 CCC: $37.00
© 2007 American Chemical Society
Published on Web 11/22/2007

these oxygen-sensitive enzymes/organisms was severely
impeded because they required growth and handling of
anaerobes and complex electro-microbial or -enzymatic
cofactor-recycling systems, based on toxic mediators, such
as Paraquat. Overall, these drawbacks rendered industrial
applications utopian. During our search for oxygen-stable
and durable enoate reductases, we investigated the asym-
metric bioreduction of R,â-unsaturated dicarboxylic acids and
diesters using 12-oxophytodienoate reductase isoenzymes
deactivation might be caused by removal of the essential
metal through complexation by diacid 1a acting as strong
chelating agent. In order to neutralize this effect, the medium
2
+
2+
2+
was supplemented with Ca , Mg , or Zn equimolar to
the substrate, which completely restored the activity of the
recycling enzymes (entries 3-5). Since the nature of the
bivalent metal did not have any influence on the outcome
of the reaction regarding conversion and/or stereoselectivity,
this is a strong hint that the deactivation of the recycling
enzyme occurs via the aforementioned mechanism. The
6,12
OPR1 and OPR3 from Lycopersicum esculentum (tomato)
and YqjM from Bacillus subtilis.13
3+
addition of Fe as supplementing metal failed; no conversion
In order to gain insight into the substrate-selectivity
relationship of these enzymes, a structurally related set of
diacids, 2-methylmaleic acid (“citraconic acid”, 1a), 2-me-
thylfumaric acid (“mesaconic acid”, 2a), and 2-methylene-
succinic acid (“itaconic acid”, 3a), were chosen, since it was
expected that they would yield the same reduction product,
was observed (data not shown). The simplicity of this system
is in strong contrast with the complex conditions required
for Clostridia.
A striking influence of the substrate configuration was
found for the E-configurated substrate analog 2a, which
proved to be unreactive with all three enzymes (entry 6). A
related (however, opposite) trend was reported with Clostrid-
ium formicoaceticum, which reduced (E)-2a to (S)-1b while
the (Z)-counterpart 1a turned out to be unreactive.10
9
2-methylsuccinic acid (1b). The corresponding dimethyl
esters (4a-6a) were tested to compare the activating effects
of the carboxyl- versus the alkoxycarbonyl groups (Scheme
1, Table 1).
The exo-methylene analog 3a proved to be a difficult
substrate, which was only converted to (R)-1b by YqjM in
low conversion, albeit with excellent stereoselectivity (entries
8
and 9).
Scheme 1. Asymmetric Bioreduction of Dicarboxylic Acids
Although R,â-unsaturated esters were suspected to be good
1a-3a and Dimethyl Esters 4a-6a
3
substrates for enoate reductases, it turned out that ester
hydrolysis occurred first when whole cells were used, which
rendered the corresponding R,â-unsaturated carboxylic acids
1
0,15
as the actual substrates.
This drawback was successfully
circumvented by using isolated enzymes (entries 10-21).
Overall, diesters 4a-6a proved to be superior in comparison
to the corresponding diacids 1a-3a. In contrast to the
dicarboxylic acids 1a-3a, diesters 4a-6a had no adverse
effects on the cofactor-recycling systems.
The cis-configurated diester 4a was reduced by all
enzymes with perfect stereoselectivity furnishing (R)-4b in
>99% ee (entries 10-13). Among them, OPR3 displayed
the highest activity (c ) 99%). In contrast, the exo-methylene
analogue 5a turned out to be more cumbersome (entries 14-
17): it was not accepted by OPR3 and also OPR1 showed
moderate activity (up to 66% conversion); only YqjM gave
sufficient conversion (up to 91%).
Substrate 1a was not converted by OPR3 but was
quantitatively reduced to (R)-1b in >99% ee by OPR1 and
YqjM using NADH or NADPH in molar amounts (entries 1
and 2). Much to our surprise, cofactor recycling using well-
established systems (NADH/FDH/formate, NADH/GDH/
glucose, NADPH/G6PDH/glucose-6-phosphate) completely
failed (data not shown). Since FDH, GDH, and G6PDH
depend on essential metal ions,14 we suspected that this
Depending on the enzyme, the (E/Z)-configuration of the
diester substrate had a strong impact on the stereochemical
outcome of the reaction. Whereas OPR1 converted both (Z)-
(9) Simon, H.; G u¨ nther, H.; Bader, J.; Tischer, W. Angew. Chem. 1981,
4
a and (E)-6a with similar activities yielding (R)-4b, OPR3
9
3, 897-898.
accepted only the (Z)-substrate 4a and was completely
inactive on (E)-6a. A related behavior was observed for
dimethyl maleate reductase isolated from Clostridium for-
micoaceticum, which was inactive on the (E)-configurated
(
10) Eck, R.; Simon, H. Tetrahedron 1994, 50, 13631-13640.
(11) Rohdich, F.; Wiese, A.; Feicht, R.; Simon, H.; Bacher, A. J. Biol.
Chem. 2001, 276, 5779-5787.
12) (a) Schaller, F. J. Exp. Bot. 2001, 52, 11-23. (b) Strassner, J.;
F u¨ rholz, A.; Macheroux, P.; Amrhein, N.; Schaller, A. J. Biol. Chem. 1999,
(
2
74, 35067-35073. (c) Breithaupt, C.; Kurzbauer, R.; Lilie, H.; Schaller,
10
substrate dimethyl fumarate.
A.; Strassner, J.; Huber, R.; Macheroux, P.; Clausen, T. Proc. Natl. Acad.
Sci. U.S.A. 2006, 103, 14337-14342.
Although YqjM showed comparable activities on both
stereoisomers, it produced opposite enantiomers with perfect
stereoselectivity (ee >99%). This substrate-based stereocon-
trol is remarkable in magnitude since the stereochemical
outcome of the reaction could be completely controlled by
(13) Kitzing, K.; Fitzpatrick, T. B.; Wilken, C.; Sawa, J.; Bourenkov,
G. P.; Macheroux, P.; Clausen, T. J. Biol. Chem. 2005, 280, 27904-27913.
(
14) The following essential metals were reported. (i) NADH-dependent
2+
6+
formate dehydrogenase: Fe , Mo (http://www.brenda.uni-koeln.de/php/
result_flat.php4?ecno)1.2.1.2). (ii) NADP-dependent glucose 6-phophate
2
+
2+
2+
dehydrogenase: Ca , Mg , Mn (http://www.brenda.uni-koeln.de/php/
result_flat.php4?ecno)1.1.1.49). (iii) NAD(P)H-dependent D-glucose de-
2
+
2+
2+
+
+
2+
2+
hydrogenase: Ca
,
Mg
,
Mn
,
K , Na , Ni
,
Zn
(http://
(15) Ferraboschi, P.; Grisenti, P.; Casati, R.; Fiecchi, A.; Santaniello, E.
www.brenda.uni-koeln.de/php/result_flat.php4?ecno)1.1.1.47).
J. Chem. Soc., Perkin Trans. 1 1987, 1743-1748.
5410
Org. Lett., Vol. 9, No. 26, 2007

Table 1. Asymmetric Bioreduction of R,â-Unsaturated Dicarboxylic Acids 1a-3a and Corresponding Diesters 4a-6a Using Enoate
Reductases OPR1, OPR3, and YqjM
a
Medium supplemented with Ca2+, Mg2+, or Zn2+ equimolar to the substrate; for details see the electronic Supporting Information. b nc ) no conversion;
nd ) not determined.
choosing an (E)- or (Z)-configurated substrate. Whereas (Z)-
a furnished (R)-4b, (E)-6a gave the mirror image (S)-4b
Bacillus subtilis with virtually absolute stereoselectivites (ee
up to >99%). In contrast to enoate reductases from anaer-
obes, standard protocols for cofactor recycling could be used
with these oxygen-stable enzymes and undesired ester
hydrolysis was entirely eliminated. Most remarkably, the
position of the CdC double bond within the substrate and
its (E/Z)-configuration played a crucial role in the substrate
recognition. Whereas exo-methylene substrates were rather
difficult to reduce, stereochemical control was achieved by
the choice of enzyme or the (E/Z)-configuration of the
substrate.
4
counterpart in >99% ee. A related phenomenon was
observed during the reduction of (E/Z)-2-chloroacrylic acids
16
using baker’s yeast and (E/Z)-enals using enoate reductases
from yeast and Zymomonas mobilis.1 In addition, an
enzyme-based stereocontrol was also observed during the
reduction of 6a, where the absolute configuration of the
product could be controlled by the choice of biocatalyst:
whereas OPR1 delivered (R)-4b (eemax 80%), YqjM furnished
7
(S)-4b (ee >99%).
In conclusion, we have shown that R,â-unsaturated dicar-
boxylic acids and their corresponding diesters were stereo-
selectively reduced by 12-oxophytodienoate reductase isoen-
zymes OPR1 and OPR3 from Lycopersicon esculentum
Acknowledgment. This project was performed in coop-
eration with BASF-AG (Ludwigshafen) and financial support
is gratefully acknowledged.
(tomato) and the “old yellow enzyme” homolog YqjM from
Supporting Information Available: General synthetic
and analytical methods. This material is available free of
charge via the Internet at http://pubs.acs.org.
(
16) Principle of stereochemical control: Utaka, M.; Konishi, S.;
Mizuoka, A.; Ohkubo, T.; Sakai, T.; Tsuboi, S.; Takeda, A. J. Org. Chem.
1
989, 54, 4989-4992.
17) M u¨ ller, A.; Hauer, B.; Rosche, B. Biotechnol. Bioeng. 2007, 98,
2-29.
(
2
OL7019185
Org. Lett., Vol. 9, No. 26, 2007
5411