Efficient synthesis of a chiral precursor for angiotensin-converting enzyme (ace) inhibitors in high space-time yield by a new reductase without external cofactors

Article abstract of DOI:10.1021/ol300397d

A new reductase, CgKR2, with the ability to reduce ethyl 2-oxo-4-phenylbutyrate (OPBE) to ethyl (R)-2-hydroxy-4-phenylbutyrate ((R)-HPBE), an important chiral precursor for angiotensin-converting enzyme (ACE) inhibitors, was discovered. For the first time, (R)-HPBE with >99% ee was produced via bioreduction of OPBE at 1 M without external addition of cofactors. The space-time yield (700 g·L^-1·d ^-1) was 27 times higher than the highest record.

Full text of DOI:10.1021/ol300397d

ORGANIC
LETTERS
2012
Vol. 14, No. 8
1982–1985
Efficient Synthesis of a Chiral Precursor
for Angiotensin-Converting Enzyme (ACE)
Inhibitors in High Space-Time Yield by a
New Reductase without External
Cofactors
Nai-Dong Shen, Yan Ni, Hong-Min Ma, Li-Juan Wang, Chun-Xiu Li, Gao-Wei Zheng,
Jie Zhang, and Jian-He Xu*
Laboratory of Biocatalysis and Synthetic Biotechnology, State Key Laboratory of
Bioreactor Engineering, East China University of Science and Technology,
130 Meilong Road, Shanghai 200237, People’s Republic of China
jianhexu@ecust.edu.cn
Received February 18, 2012
ABSTRACT
A new reductase, CgKR2, with the ability to reduce ethyl 2-oxo-4-phenylbutyrate (OPBE) to ethyl (R)-2-hydroxy-4-phenylbutyrate ((R)-HPBE), an
important chiral precursor for angiotensin-converting enzyme (ACE) inhibitors, was discovered. For the first time, (R)-HPBE with >99% ee was
produced via bioreduction of OPBE at 1 M without external addition of cofactors. The space-time yield (700 g Lꢀ1
the highest record.
d
ꢀ1) was 27 times higher than
3
3
Chiralalcoholshavewideapplicationsinthesynthesisof
numerous pharmaceuticals.1 Ethyl (R)-2-hydroxy-4-phe-
nylbutyrate ((R)-HPBE), for instance, is an important
building block for the production of various angiotensin-
converting enzyme (ACE) inhibitors, which are used to
treat hypertension and congestive heart failure, such as
benazepril, cilazapril, enalapril, and ramipril.2 A series of
technical processes for the synthesis of (R)-HPBE have
been explored, including resolution of the corresponding
racemate,3 reduction of 2-oxo-4-phenylbutanoic acid or
its derivates,4 and a chemical multistep asymmetric syn-
thesis.5 Among these routes, the asymmetric synthesis of
(R)-HPBE with reductases has attracted much attention
due to its significant advantages such as high conversion,
eco-friendliness (neutral pH and room temperature), and
remarkable stereoselectivity making it preferable to other
€
(1) (a) Hummel, W.; Abokitse, K.; Drauz, K.; Rollmann, C.; Groger,
H. Adv. Synth. Catal. 2003, 345, 153–159. (b) Huisman, G. W.; Liang, J.;
Krebber, A. Curr. Opin. Chem. Biol. 2010, 14, 122–129. (c) Xie, Y.; Xu,
J. H.; Xu, Y. Bioresour. Technol. 2010, 101, 1054–1059.
(2) (a) Kaluzna, I.; Andrew, A. A.; Bonilla, M.; Martzen, M. R.;
Stewart, J. D. J. Mol. Catal. B: Enzym. 2002, 17, 101–105. (b) Chen,
Y. Z.; Lin, H.; Xu, X. Y.; Xia, S. W.; Wang, L. X. Adv. Synth. Catal.
2008, 350, 426–430. (c) Zhang, W.; Ni, Y.; Sun, Z. H.; Zheng, P.; Lin,
W. Q.; Zhu, P.; Ju, N. F. Process Biochem. 2009, 44, 1270–1275.
(3) (a) Chadha, A.; Manohar, M. Tetrahedron: Asymmetry 1995, 6,
651–652. (b) Liese, A.; Kragl, U.; Kierkels, H.; Schulze, B. Enzyme
Microb. Technol. 2002, 30, 673–681. (c) Huang, S. H.; Tsai, S. W. J. Mol.
Catal. B: Enzym. 2004, 28, 65–69.
(4) (a) Chadha, A.; Manohar, M.; Soundarajan, T.; Lokeswari, T. S.
Tetrahedron: Asymmetry 1996, 7, 1571–1572. (b) de Lacerda, P. S. B.;
Ribeiro, J. B.; Leite, S. G. F.; Coelho, R. B.; da Silva Lima, E. L.;
Antunes, O. A. C. Biochem. Eng. J. 2006, 28, 299–302.
(5) (a) Herold, P.; Indolese, A. F.; Studer, M.; Jalett, H. P.; Siegrist,
U.; Blaser, H. U. Tetrahedron 2000, 56, 6497–6499. (b) Lin, W. Q.; He,
Z.; Jing, Y.; Cui, X.; Liu, H.; Mi, A. Q. Tetrahedron: Asymmetry 2001,
12, 1583–1587.
r
10.1021/ol300397d
Published on Web 04/05/2012
2012 American Chemical Society

methods.6 Chen et al. employed Candida boidinii CIOC21
for the production of (R)-HPBE, yielding 99% ee in
aqueous medium with 20 mM substrate under the optimal
conditions.2b Zhang et al. reported the production of (R)-
HPBE by whole cells of Candida krusei SW2026 in 99.7%
ee and 95.1% yield by reducing 2.5 g/L of OPBE.2c
However, the ee of the product decreased to 97.4% in
the presence of 20 g/L of OPBE. Despite the reported
enzyme-mediatedasymmetricreductiontoyield(R)-HPBE
with excellent stereoselectivity, there were still drawbacks
including low substrate loading, low space-time yield, and
external addition of expensive cofactors, which restricted
the industrial applications. Interestingly, in our recent
work, the biosynthesis of (S)-HPBE was observed at very
high substrate loading (620 g/L) with >99% ee in the
absence of external cofactors.7 Although (S)-HPBE is not
the normally employed enantiomer used as an intermediate
of ACE inhibitors, we still believed that the production of
(R)-HPBE by the same pathway might also be achieved by
developing new reductases.
Figure 1. Specific activities and enantioselectivity of the
screened reductases. (b) CgKR2; (O) other reductases with
lower activity or selectivity.
CAG60239.1, CAG60240.1, CAG58278.1, CAG58834.1,
EDK38638.2, CAH02579.1, EEQ44752.1, and EEQ47109.1)
were cloned and heterologously expressed. Among the
total 13 recombinant reductases obtained, CgKR2 was
identified to be the best target biocatalyst because of its
highest specific activity (2.4 U/mg of protein) and enan-
tioselectivity (>99%) (Figure 1).
To discover and develop promising and versatile reduc-
tases, many new routes have been adopted.8 Among them,
a genome mining method is an effective strategy employed
for the discovery of novel biocatalysts. For example, a
novel NADH-dependent reductase newly discovered by
genome mining from Streptomyces coelicolor shows nota-
ble catalytic performance in the reduction of prochiral
ketones.8c
To investigate its basic catalytic properties, the recom-
binant reductase CgKR2 was purified to homogeneity by a
Ni-NTA column from the cell free extract of E. coli
containing the recombinant CgKR2 (E. coli/pCgKR2).
The specific activity of the purified CgKR2 was 10.6 U/mg
of protein, which was 4.3-fold higher compared to the
crude extract. The optimum pH of purified CgKR2 was
determined at pHs ranging from 4.0 to 11.0. CgKR2
displayed the maximum activity at pH 6.0 in sodium
phosphate buffer. The effect of temperature on the activity
of purified CgKR2 was examined from 25 to 55 °C. The
optimal activity was detected at 45 °C. The thermostability
of purified CgKR2 was also studied. Purified CgKR2
retained 48% of initial activity after incubation for 84 h
at 30 °C and only 10% of activity after 36 h of preservation
at 40 °C. The enzyme was unstable at 50 °C with a half-life
of 2.6 min. The results indicate that CgKR2 is more stable
under moderate reaction conditions.
Herein, we also employed a genome mining method to
discover enzymes for stereoselective synthesis of (R)-
HPBE. Three reductases from Saccharomyces cerevisiae
(YDL124w, YDR368w, YGL185c) are known to possess
excellent enantioselectivity toward OPBE.9 A series of pre-
dicted putative carbonyl reductases were selected as the
candidates on the basis of pBLAST searching with the
three reported reductases from Saccharomyces cerevisiae
as query sequences. Five genes (Genbank accession numbers:
CAG61069.1, ABN67667.2, CAG57781.1, CAG62011.1,
and ABN65769.2) bearing 29ꢀ67% amino acid identities
with the three probe reductases were cloned and expressed
in Escherichia coli BL21 (DE3). After one round of screen-
ing, CgKR2 from Candida glabrata (Genbank accession:
CAG61069.1) was identified as the one with the highest
activity and enantioselectivity toward OPBE, giving optically
pure (R)-HPBE from the bioconversion of 10 mM sub-
strate. Then a second round of mining based upon
pBLAST searching was carried out with CgKR2 as the query
sequence. As a result, 8 genes (Genbank accession numbers:
The kinetic parameters of the purified CgKR2 were deter-
mined, giving Vmax and Km of 18.5 μmol minꢀ1 mgꢀ1 of
protein and 0.1 mM, respectively. CgKR2 has a lower Km
toward OPBE than the carbonyl reductase from Candida
krusei SW2026 toward OPBE10 and YiaE toward 2-oxo-
4-phenylbutanoic acid.11
(6) (a) Nakamura, K.; Yamanaka, R.; Matsuda, T.; Harada, T.
Tetrahedron: Asymmetry 2003, 14, 2659–2681. (b) Tao, J. H.; Xu,
J. H. Curr. Opin. Chem. Biol. 2009, 13, 43–50. (c) de Carvalho, C. C.
C. R. Biotechnol. Adv. 2011, 29, 75–83.
(7) Ni, Y.; Li, C. X.; Zhang, J.; Shen, N. D.; Bornscheuer, U. T.; Xu,
J. H. Adv. Synth. Catal. 2011, 353, 1213–1217.
(8) (a) Knietsch, A.; Waschkowitz, T.; Bowien, S.; Henne, A.; Daniel,
R. J. Mol. Microbiol. Biotechnol. 2003, 5, 46–56. (b) Zhu, D. M.; Malik,
H. T.; Hua, L. Tetrahedron: Asymmetry 2006, 17, 3010–3014. (c) Wang,
L. J.; Li, C. X.; Ni, Y.; Zhang, J.; Liu, X.; Xu, J. H. Bioresour. Technol.
2011, 102, 7023–7028.
(9) Kaluzna, I. A.; Matsuda, T.; Sewell, A. K.; Stewart, J. D. J. Am.
Chem. Soc. 2004, 126, 12827–12832.
The substrate specificity and stereoselectivity of CgKR2
for various R- and β-ketoesters as well as aromatic ketones
were also explored. The specific activity of CgKR2 toward
the reduction of ketones and ketoesters was determined
spectrophotometrically. The stereoselectivity of CgKR2
(10) Li, N.; Ni, Y.; Sun, Z. H. J. Mol. Catal. B: Enzym. 2010, 66, 190–
197.
(11) Yun, H.; Choi, H. L.; Fadnavis, N. W.; Kim, B. G. Biotechnol.
Prog. 2005, 21, 366–371.
Org. Lett., Vol. 14, No. 8, 2012
1983

Table 1. CgKR2-Catalyzed Reduction of Various Ketones and Ketoesters
a The enzyme activity was measured using the standard assay protocol. To calculate the relative activity, the activity for 2 mM OPBE was taken as
100%. b Measured by chiral GC analysis. c The ee values were measured by chiral GC or HPLC analysis. d Determined by chiral GC analyses after
acetylation of the product. e Not detected.
was assayed byGC or HPLC after the reduction of ketones
and ketoesters with a NADPH recycling system containing
glucose and glucose dehydrogenase (GDH). As described
Table 2. Asymmetric Reduction of 1 M OPBE with Cells of
E. coli/pCgKR2 and GDHa
in Table 1, R-ketoesters served as good substrates with
CgKR2
(kU/L)
NADPþ
(mM)
time
(h)
conv
(%)b
yield
(%)c
ee
(%)b
relative activities of more than 60% except for ethyl
benzoylformate, 3a. CgKR2 exhibited higher activity to-
ward aromatic R-ketoesters with an ortho-chloro substi-
tuent on the benzene ring, and methyl esters gave higher
activities than ethyl esters. The highest activity was ob-
served when using methyl o-chlorobenzoylformate, 5a.
Moreover, stereoselectivity of >99% ee and conversion
of >99% were achieved for the product methyl (R)-o-
chloro-mandelate, (R)-5b, a key intermediate for clopido-
grel,12 expanding the scope of possible CgKR2 applica-
tions. All R-ketoesters were reduced to the corresponding
(R)-chiral alcohols with high stereoselectivity and high
conversion.
entry
1
60
60
0.5
0.3
0
5
6.5
24
7
>99
>99
97.1
98.4
>99
91
86
>99
>99
>99
>99
>99
2
3
60
4
5d
120
120
0
0
6
84
a Reaction conditions: OPBE (1 M), D-glucose (1.5 M), lyophilized
cells of E. coli/pCgKR2 (activity was indicated above, 1200 U equals
about 0.5 g of lyophilized cells (about 2 g of wet cells)), lyophilized GDH
powders (1 equiv activity with respect to CgKR2, 1200 U equals about
0.14 g of enzyme powders), NADPþ (concentration was indicated
above), 10 mL of sodium phosphate buffer (pH 6.0, 100 mM), 30 °C.
pH was kept at 6.0 with 2 M Na2CO3. b Determined by GC analysis.
c Isolated yield of (R)-HPBE. d 100 mL of sodium phosphate buffer.
For β-ketoesters and aromatic ketones, CgKR2 was
less active, with <10% relative activity. Stereoselectivity
of >97% ee as well as conversion of >99% were obtained
only for the product of (R)-1-phenyl-2,2,2-trifluoroetha-
nol, (R)-13b, which is a synthon for liquid crystals.13
CgKR2 catalyzed the reduction of ethyl propionylacetate,
7a, with excellent stereoselectivity, while other β-ketoesters
with a halogen group were reduced with moderate to high
selectivity. It has been found that the substitution of a
halogen group at the 4-position of ethyl 3-oxo-butyrate
influences the enantioselectivity of the reported reductases,
YtbE14 and SSCR.15 For CgKR2, the effect is more
obvious and it seems that the enantioselectivity decreases
as the substituent group becomes bulkier.
To further utilize the newly discovered biocatalyst for
practical synthesis of (R)-HPBE, initial experiments were
performed at a higher substrate loading, 0.1 M, in 1 mL of
sodium phosphate buffer. Lyophilized cells of E. coli/
pCgKR2 were employed to carry out the reduction, and
lyophilized GDH (crude powders) was used to complete
the requisite regeneration of cofactor by catalyzing the
oxidation of glucose. The conversion reached 98.4% after
a 20 h reaction with >99% ee. The increase of substrate
loading to 0.4 M led to a 34% decrease of conversion but
had no effect on stereoselectivity. These results displayed
great potential for industrial applications. Then a series of
optimizitions were carried out, including autotitrating the
reaction mixture with 2 M Na2CO3, increasing the sub-
strate loading and the dosage of reductase and scaling
up the reaction. It was found that a nearly complete
(12) Ema, T.; Okita, N.; Ide, S.; Sakai, T. Org. Biomol. Chem. 2007, 5,
1175–1176.
(13) Fujisawa, T.; Ichikawa, K.; Shimizu, M. Tetrahedron: Asymme-
try 1993, 4, 1237–1240.
(14) Ni, Y.; Li, C. X.; Ma, H. M.; Zhang, J.; Xu, J. H. Appl.
Microbiol. Biotechnol. 2011, 89, 1111–1118.
(15) Zhu, D. M.; Yang, Y.; Buynak, J. D.; Hua, L. Org. Biomol.
Chem. 2006, 4, 2690–2695.
1984
Org. Lett., Vol. 14, No. 8, 2012

Table 3. Comparison between CgKR2 and Other Reported Biocatalysts
biocatalyst
OPBE conc (M)
reaction media
reaction time (h)
ee (%)
space-time yield (g Lꢀ1 dꢀ1
)
3
3
CgKR2a
1.00
0.40
0.10
0.01
0.02
0.02
aqueous phase
6
48
16
14
24
12
>99 (R)
87.5 (R)
97.4 (R)
99.7 (R)
>99 (R)
99.0 (R)
700
17
25
4
By cellb 18
Ck cellc 2c
two-liquid phase systems
biphasic system
aqueous phase
ADHd 19
Cb celle 2b
aqueous phase
<5
8
aqueous phase
a Lyophilized cells of E. coli/pCgKR2. b Pretreated Baker’s yeast. c Wet cells of Candida krusei SW2026. d Alcohol dehydrogenase from Paracoccus
pantotrophus DSM 11072. 1 mM NADH was added. e Wet cells of Candida boidinii CIOC21.
conversion of OPBE was achieved within 5 h with >99%
ee in 10 mL of sodium phosphate buffer (Table 2, entry 1)
employing 600 U of CgKR2 while the loading of OPBE
was increased up to 1 M, which is a high concentration
from the viewpoint of industrial practice.
41.9% conversion of OPBE and 87.5% ee of (R)-HPBE.18
This is the first report of the biocatalytic synthesis of (R)-
HPBE with >99% ee via reduction of OPBE at 1 M
without the addition of any external cofactors. One can see
from Table 3, when the CgKR2-catalyzed reduction was
carried out in an aqueous monophase reaction medium,
the shortest reaction time was achieved. The space-time
As the expensive nicotinamide cofactors are a major
hindrance to the practical application of reductases,16
a
further attempt was made to reduce the amount of external
addition of NADPþ. It should be noted that even if only
the internal cofactors from the cells of E. coli/pCgKR2
were utilized to carry out the reduction (Table 2, entry 3),
CgKR2 also gave 97.1% conversion in 24 h under the same
conditions. After increasing the cells of E. coli/pCgKR2
and GDH to 120 U/mL, the reaction was almost complete
within 7 h with 98.4% conversion and >99% ee (Table 2,
entry 4). Then the bioreduction of OPBE was scaled up by
10-fold (Table 2, entry 5). After 6 h reaction and subse-
quent workup, the desired product, (R)-HPBE, was iso-
lated in 84% yield and >99% ee, indicating the potential
of CgKR2 for practical usage.
To work as an alternative to chemical catalysts, the
biocatalyst must tolerate a high substrate concentration
for practical application on an industrial scale.16a,17 To the
best of our knowledge, there is only one example of the
asymmetric reduction of OPBE at as high as 0.4 M in a
two-liquid phase system of water/organic solvent, with
yield of (R)-HPBE production reached 700 g Lꢀ1 dꢀ1
,
3
3
which was 27 times higher than the highest record reported
so far. Compared with other enzymes exhibiting excellent
enantioselectivity, CgKR2 provides several additional ad-
vantages, making it a more competitive and promising
biocatalyst. Further work including optimization of the
reaction for large scale applications is still in progress.
Acknowledgment. This work was financially supported
by the National Natural Science Foundation of China
(No. 20902023, 31071604), Ministry of Science and Tech-
nology, P.R. China (No. 2011CB710800, 2011AA02A210),
China National Special Fund for State Key Laboratory of
Bioreactor Engineering (No. 2060204), and the Innovation
Program of Shanghai Municipal Education Commission
(No. 11CXY24).
Supporting Information Available. Experimental meth-
ods and spectral data. This material is available free of
charge via the Internet at http://pubs.acs.org.
€
(16) (a) Groger, H.; Chamouleau, F.; Orologas, N.; Rollmann, C.;
(18) Shi, Y. G.; Fang, Y.; Wu, H. P.; Li, F.; Zuo, X. Q. Biocatal.
Biotransfor. 2009, 27, 211–218.
(19) Lavandera, I.; Kern, A.; Schaffenberger, M.; Gross, J.; Glieder,
Drauz, K.; Hummel, W.; Weckbecker, A.; May, O. Angew. Chem., Int.
Ed. 2006, 45, 5677–5681. (b) Zhang, J.; Witholt, B.; Li, Z. Chem.
Commun. 2006, 4, 398–400.
(17) Ma, S. K.; Gruber, J.; Davis, C.; Newman, L.; Gray, D.;
Wang, A.; Grate, J.; Huisman, G. W.; Sheldon, R. A. Green Chem.
2010, 12, 81–86.
A.; de Wildeman, S.; Kroutil, W. ChemSusChem 2008, 1, 431–436.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 8, 2012
1985