Preparation of (S)-mandelic acids by enantioselective degradation of racemates with a new isolate Pseudomonas putida ECU1009

Article abstract of DOI:10.1016/j.tetasy.2005.05.022

An enantioselective (R)-mandelate degrading bacterium, Pseudomonas putida ECU1009, was newly isolated from soil. The degradation activity of the bacterial cells was significantly enhanced by supplementing an optimal amount of racemic mandelic acid (0.4%), benzoylformic acid (0.4%), or benzoic acid (0.2%) to the culture medium as the enzyme inducer. Using the resting cells as a biocatalyst, three kinds of (S)-mandelic acids 1-3 were prepared with high isolated yields and enantiomeric excesses. Moreover, in a one-pot fermentation-transformation process using 1.25% (RS)-mandelic acid as the sole carbon/energy source for cultivation of the bacterium, (S)-mandelic acid 1 was accumulated after 48 h of bioconversion with 46.5% yield and >99% ee.

Full text of DOI:10.1016/j.tetasy.2005.05.022

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 16 (2005) 2113–2117
Preparation of (S)-mandelic acids by enantioselective
degradation of racemates with a new isolate
Pseudomonas putida ECU1009
Han-Rong Huang,a Jian-He Xu,a,* Yi Xu,a Jiang Pana and Xiang Liub
aLaboratory of Biocatalysis and Bioprocessing, State Key Laboratory of Bioreactor Engineering,
East China University of Science and Technology, Shanghai 200237, PR China
bKey Laboratory of Industrial Biotechnology, Ministry of Education, Southern Yangtze University, Wuxi, Jiangsu 214036, PR China
Received 5 April 2005; accepted 17 May 2005
Abstract—An enantioselective (R)-mandelate degrading bacterium, Pseudomonas putida ECU1009, was newly isolated from soil.
The degradation activity of the bacterial cells was significantly enhanced by supplementing an optimal amount of racemic mandelic
acid (0.4%), benzoylformic acid (0.4%), or benzoic acid (0.2%) to the culture medium as the enzyme inducer. Using the resting cells
as a biocatalyst, three kinds of (S)-mandelic acids 1–3 were prepared with high isolated yields and enantiomeric excesses. Moreover,
in a one-pot fermentation–transformation process using 1.25% (RS)-mandelic acid as the sole carbon/energy source for cultivation
of the bacterium, (S)-mandelic acid 1 was accumulated after 48 h of bioconversion with 46.5% yield and >99% ee.
Ó 2005 Published by Elsevier Ltd.
1. Introduction
microorganisms to produce (S)-1 from the racemic
mixture.
Mandelic acid and its derivatives are a class of chiral
synthons for the production of various pharmaceuticals,
such as semi-synthetic penicillins, cephalosporins, and
antiobesity agents.1–3 Many methods have been re-
ported for the preparation of enantiomerically pure
(S)- or (R)-mandelic acid (1).4–10 For example, (S)-1
has been separated from racemic mixtures by means of
high-pressure crystallization.11 It has also been prepared
by the enzyme-catalyzed synthesis of (S)-cyanohydrins
and subsequent hydrolysis into (S)-a-hydroxy-carbox-
ylic acids,12 from benzoylformic acid by fermentation
with Lactobacillus genus (e.g., Lactobacillus bulgaricus
and Lactobacillus plantarum) or Pediococcus genus
(e.g., Pediococcus parvulus and Pediococcus pentosac-
eus),13 or from methyl mandelate by lipase-catalyzed
hydrolysis.14 However, little is known about the biocata-
lytic conversion of ( )-1 into the (S)-enantiomer,9 in
spite of many reports regarding the bioproduction of
(R)-mandelate.4,8–10 This prompted us to screen for
The enzymes involved in the mandelate metabolic path-
way allow various Pseudomonas species to utilize either
one or both of the mandelate enantiomers as its sole car-
bon source.15–18 Usually, the catabolism of the (R)-man-
delate requires the presence of mandelate racemase to
equilibrate the enantiomers and generate (S)-mandelate,
which can then be oxidatively degraded to benzoate by
the remaining enzymes in the pathway. Though Asper-
gillus sydowi IFO 4284 and Fusarium oxysporum IFO
5942 have previously been reported as being capable
of selectively oxidizing (R)-1 from its racemate, the
activity of these two fungal strains were rather low.9
Herein, we report a newly isolated bacterial strain, Pseu-
domonas putida ECU1009, which is capable of degrading
the (R)-enantiomer of racemic mandelate and affording
(S)-mandelate in high enantiomeric excess (Scheme 1).
This method gives a theoretical yield of 50%, as do other
classical kinetic resolutions. However, the method pos-
sesses several advantages over the other processes men-
tioned above: (R)-1 and its intermediates can be
completely degraded, therefore only a simple extraction
step is needed to gain the final product in fairly high pur-
ity. Hence, the product isolation and purification is
*
Corresponding author. Fax: +86 21 6425 2250; e-mail:
jianhexu@ecust.edu.cn
0957-4166/$ - see front matter Ó 2005 Published by Elsevier Ltd.
doi:10.1016/j.tetasy.2005.05.022

2114
H.-R. Huang et al. / Tetrahedron: Asymmetry 16 (2005) 2113–2117
OH
O
COOH
COOH
COOH
β-Ketoadipate
Pathway
(R)-mandelic acid
Benzoyl formic acid
Scheme 1. Proposed mandelate pathway of Pseudomonas putida ECU1009.
5
4
3
2
1
0
simple and efficient. Although the theoretical yield of
(S)-1 is limited to 50%, an enantioselective two-step ste-
reoinversion8 of ( )-1 to (S)-1 can be considered to over-
come this disadvantage.
30
20
10
0
2. Results and discussion
2.1. Screening of microorganisms
Bacteria with mandelate-degrading activities were iso-
lated from soil samples through a two-step enrichment
cultivation procedure on a minimal salt medium
(MSM) containing benzoylformic acid (0.2%) as the sole
carbon source with the first round of screening for ben-
zoylformic acid-degrading activity, and the second
round of screening for selective degrading activity on
(S)- or (R)-1. In the first round, benzoylformic acid-
degrading strains were isolated after being grown at
30 °C on MSM for 3 days; about 140 bacterial strains
among 200 isolates were considered to be significantly
active on benzoylformic acid. In the second round of
screening, about 10 strains gave >98% enantiomeric
excess of mandelic acid, of which nine strains could
selectively degrade (S)-1. Finally, a strain marked as
ECU1009 was chosen for further study due to its high
activity and enantioselectivity toward (R)-1. This strain
was later identified to be P. putida by the Shanghai Cen-
ter for Disease Control based on its taxonomical charac-
teristics (data not shown) and thus designated as P.
putida ECU1009. To the best of our knowledge, this is
the first example of the enantioselective degradation of
(R)-1 from racemate by the redox system of P. putida.
0
12
24
36
48
Bioconversion time (h)
Figure 1. Time course of enantioselective degradation of ( )-mandelic
acid by resting cells of Pseudomonas putida ECU1009. The reaction
was performed at 30 °C and 160 rpm in 80 mL of 100 mM potassium
phosphate buffer (pH 6.0) containing 60 mM of ( )-1 with 8.47 g
DCW/L of resting cells. Symbols: ( ) (S)-1; ( ) (R)-1; ( ) benzoyl-
formate; ( ) benzoate.
An average enantiomeric ratio (E-value) of about
130 30 was obtained according to the equation of
Chen et al.:19 E = ln[(1 ꢀ c)(1 ꢀ ees)]/ln[(1 ꢀ c)(1 + ees)].
Poor growth and a lower degrading activity of cells were
observed if aged cells were used. In order to overcome
these drawbacks, the strain was grown freshly on a slant
medium (SM) for 24–48 h at 30 °C each time before
shake-flask cultivation. The optimal temperature and
pH of the enantioselective degradation were 30 °C and
6.0, respectively. The degrading activity of (R)-1 was sig-
nificantly enhanced when an appropriate amount of
racemic mandelic acid, benzoylformic acid, or benzoic
acid was added to the culture medium (Fig. 2). The opti-
mal concentrations of these inducers were 0.4% for race-
mic mandelic acid, 0.4% for benzoylformic acid and
0.2% for benzoic acid. At 0.4% of benzoic acid, poor
growth and a lower degrading activity of cells were
observed, indicating that the benzoic acid as inducer was
more toxic to the cells than racemic mandelic acid or
benzoylformic acid. At 60 mM of the racemic mandelate
( )-1, the bacterial cells exhibited the maximum degrad-
ing activity on (R)-1. Under the optimal reaction condi-
tions, (S)-1 was recovered in 42% isolated yield and 99%
ee after 24 h of biotransformation. The substrate spec-
trum of the newly discovered biocatalyst was examined
using several derivatives of mandelic acid, as shown in
Table 1. P. putida ECU1009 also showed high enantio-
selective activities on (R)-mandelate derivatives with
substitution groups (–OH or –Cl) at the para-position
of the aromatic ring, for example, 2 and 3. However,
when a substrate with a substitution at the meta- or
ortho-position, for example, 4 and 5, either the enzy-
2.2. Enantioselective degradation of mandelic acid
derivatives
The enantioselective degradation of ( )-1 was first
examined using the resting cells of P. putida ECU1009
as the biocatalyst. The molecular structures of two key
intermediates, benzoic acid and benzoylformic acid,
were confirmed by mass spectra and NMR (see Sections
4.4.1 and 4.4.2). The time-dependant changes of the sub-
strate and product concentrations are shown in Figure
1. Compound (S)-1 was only slightly degraded after
34 h while the residual mandelic acid concentration
was 29.5 mM (analytical yield 49%; ee 99%) and the
benzoylformic acid concentration was very low during
the whole process of biotransformation. A possible
intermediate of benzaldehyde was not detected in our
case. The concentration of the accumulated benzoate
as an intermediate of (R)-1 reached the maximum at
18 h and it was further degraded completely after 44 h.

H.-R. Huang et al. / Tetrahedron: Asymmetry 16 (2005) 2113–2117
2115
synthesis, its use as a raw material for (S)-1 production
would be beneficial for simplifying the post-reaction sep-
aration of the (S)-isomer from the racemic mixture.
4.0
3.0
2.0
1.0
0.0
4.0
3.0
2.0
1.0
0.0
3. Conclusions
Using racemic mandelic acids as starting materials, three
kinds of (S)-mandelic acids 1–3 were efficiently prepared
with high isolated yields and ee. The degrading activity
of (R)-mandelic acids can be significantly enhanced by
supplementing 0.4% racemic mandelic acid, 0.4% ben-
zoylformic acid or 0.2% benzoic acid to the culture med-
ium. Moreover, using ( )-1 as the sole carbon source,
(S)-1 could also be obtained after 48 h of fermentation
with 46.5% analytical yield and >99% ee.
control
0.2% benzoic
acid
0.4%
benzoyformic
acid
0.4% mandelic
acid
Figure 2. Effect of ( )-mandelic acid, benzoylformic acid and benzoic
acid on cell growth and production of (R)-mandelate-degrading
enzyme by Pseudomonas putida ECU1009. One unit (u) of activity
was defined as the amount of enzyme needed for the degradation of
1 lg (R)-1 per minute at 30 °C and pH 6.0. Assay conditions: ( )-1,
60 mM, 20 mL; incubation for 2 h. Orange bar: cell density reflected by
optical density at 600 nm; Green bar: enzyme activity (units per
milligram of dry cell weight).
4. Experimental
4.1. General
Minimal salt medium (MSM): 0.2% benzoylformic acid,
0.2% (NH4)SO4, 0.1% yeast extract, 0.1% K2HPO4,
0.1% KH2PO4, 0.1% NaCl, 0.05% MgSO4, 1.8% agar,
pH 7.0.
matic activity or the enantioselectivity was significantly
decreased, although the reasons are as yet unclear.
Medium for slant culture (SM): 1.5% glycerol, 0.5% pep-
tone, 0.5% yeast extract, 0.1% K2HPO4, 0.1% KH2PO4,
0.1% NaCl, 0.05% MgSO4, 0.4% racemic mandelic acid,
1.8% agar, pH 7.0.
Furthermore, P. putida ECU1009 was able to grow with
(R)-1 of the racemate (1.25%) as the sole carbon/energy
source, leaving (S)-1 in 46.5% analytical yield and >99%
ee after 48 h of cultivation/transformation on an opti-
mized fermentation medium. As ( )-1 is produced com-
mercially on a large scale at a very low cost by chemical
Fermentation medium (FM): the same as SM but with-
out agar and mandelic acid.
Table 1. Preparation of (S)-mandelic acids by enantioselective degradation of racemates using resting cells of Pseudomonas putida ECU1009a
O
OH
COOH
COOH
β-Ketoadipate
Pathway
COOH
X
X
X
OH
COOH
OH
COOH
X
X
+
( )-1 X=H
+
( )-2 X=p-OH
+
( )-3 X=p-Cl
+
X=m-Cl
( )-4
+
( )-5 X=o-Cl
Entry
Substrate
Concn (mM)
Time (h)
Yieldb (%)
ee (%)c
1
2
3
4
5
6
7
( )-1
( )-2
( )-3
( )-3
( )-4
( )-4
( )-5
60
60
60
40
40
40
40
24
24
48
36
36
48
48
42
28
47
35
56
45
90
>99
>99
>99
>99
43
85
3
a The reactions were carried out with 14.0 g DCW/L of resting cells in 100 mM potassium phosphate buffer (pH 6.0) at 30 °C, and 160 rpm.
b Isolated yield after purification by silica gel column chromatography.
c ee (%) were determined by chiral HPLC.

2116
H.-R. Huang et al. / Tetrahedron: Asymmetry 16 (2005) 2113–2117
Benzoylformic acid and racemic mandelic acids ( )-1–5
were purchased from Guangde Chemicals Co. Ltd,
China. All other chemicals were also from commercial
sources and of reagent grade.
7.377 (m, 2H), 7.518 (s, 1H), 7.860 (m, 2H). MS (ESI)
m/z: 150 (M+); C8H6O3 requires 150.
4.4.2. Benzoic acid. White powder. Yield 80% (isolated
at 18 h). 1H NMR (500 MHz, CDCl3): d/ppm: 7.540 (m,
3H), 8.170 (m, 2H); MS (ESI) m/z: 122 (Mꢀ); C7H6O2
requires 122.
4.2. Microorganism, growth and biotransformation
conditions
P. putida ECU1009 was maintained on SM at 30 °C.
The strain, after being grown on SM for 24–48 h, was
first pre-cultivated in 5 mL FM for 12 h, then inoculated
into 500 mL Erlenmeyer flasks containing 95 mL of the
FM and shaken at 30 °C and 160 rpm. After 12 h,
appropriate inducers were added to the FM. After
24 h of cultivation, cells were harvested from the culture
broth, washed and re-suspended in 100 mM potassium
phosphate buffer (pH 6.0) containing 60 mM ( )-1.
4.4.3. (S)-Mandelic acid (S)-1. White powder. Yield
25
D
42%. ½aꢁ ¼ þ152.0 (c 1.0, H2O), >99% ee. HPLC
conditions for its methyl ester: Chiralcel OD, hexane–
2-propanol (90:10, v/v; 0.8 mL/min), detector 228 nm;
(S) 18.7 min, (R) 25.7 min. H NMR (500 MHz, D2O):
1
d/ppm: 5.260 (s, 1H), 7.350 (m, 5H).
4.4.4. (S)-p-Hydroxyl-mandelic acid (S)-2. White
25
D
powder. Yield 28%. ½aꢁ ¼ þ140.0 (c 1.1, H2O), >99%
ee. HPLC conditions for its methyl ester: Chiralcel
OD, hexane–2-propanol (88:12, v/v; 0.8 mL/min), detec-
tor 228 nm; (S) 18.7 min, (R) 25.7 min. 1H NMR
(500 MHz, D2O): d/ppm: 5.200 (s, 1H), 6.770 (m, 2H),
7.254 (m, 2H).
4.3. Bioconversion process by resting cells
The bioconversion process was monitored with HPLC
by withdrawing 0.6 mL of samples at fixed time inter-
vals. The cells were removed by centrifugation and
0.1 mL of the supernatant was used for the determina-
tion of the product and substrate concentrations by
HPLC (Shim-pack VP-ODS, Shimadzu Co., Japan)
with an elution system of 10 mM KH2PO4–CH3OH
(87:13, v/v) at a flow rate of 0.8 mL/min and the UV
absorbance at 225 nm. The remaining 0.5 mL sample
was used for HPLC determination of the enantiomeric
excess (ee) of mandelic acid after derivation into methyl
mandelate with methanol and trimethylchlorosilane,
using a chiral column (Chiralcel OD, Daicel Co., Japan)
eluted with hexane–isopropanol (90:10, v/v, 0.8 mL/min)
and detected at 228 nm.
4.4.5. (S)-p-Chloro-mandelic acid (S)-3. White powder.
25
Yield 47%. ½aꢁ ¼ þ128.6 (c 1.0, H2O), >99% ee. HPLC
D
conditions for its methyl ester: Chiralcel OD, hexane–2-
propanol (98:2, 0.8 mL/min), detector 228 nm; (S)
1
19.0 min, (R) 22.0 min. H NMR (500 MHz, D2O): d/
ppm: 5.150 (s, 1H), 7.350 (m, 4H).
4.4.6. (S)-m-Chloro-mandelic acid (S)-4. White pow-
25
D
der. Yield 45%. ½aꢁ ¼ þ105.8 (c 1.0, H2O), 85% ee.
HPLC conditions for its methyl ester: Chiralcel OD,
hexane–2-propanol (98:2, 0.8 mL/min), detector
228 nm; (S) 18.3 min, (R) 21.3 min. 1H NMR
(500 MHz, D2O): d/ppm: 5.200 (s, 1H), 7.320 (m, 3H),
7.426 (m, 1H).
4.4. Preparation of (S)-1, its intermediates and derivatives
by resting cells
4.4.7. (S)-o-Chloro-mandelic acid (S)-5. White powder.
25
P. putida ECU1009, was grown on SM for 24–48 h at
30 °C, pre-cultivated in 5 mL liquid medium (FM) for
12 h, inoculated into 500 mL Erlenmeyer flasks contain-
ing 95 mL of FM and shaken at 30 °C and 160 rpm.
After 12 h, 0.2% benzoic acid was added to the culture
medium. After 24 h of cultivation, cells harvested from
250 mL of culture broth were re-suspended in 50 mL
of 100 mM potassium phosphate buffer (pH 6.0) con-
taining a certain concentration of ( )-1, ( )-2, ( )-3,
( )-4, or ( )-5. After a certain time of bioconversion,
the cells were removed by centrifugation and the super-
natant acidified with concentrated H2SO4 to pH 1.0 and
extracted with ethyl acetate. The ethyl acetate layer was
collected, dried over anhydrous Na2SO4 and concen-
trated under reduced pressure to obtain a crude crystal
of (S)-(+)-mandelic acids. The acids were purified by sil-
ica gel column chromatography. The eluent components
are toluene–ethyl acetate–formic acid (5:1:0.2, v/v/v) for
3, 4, 5; benzene–ethyl acetate–formic acid (5:1:0.5, v/v/v)
for 1, benzoylformic acid and benzoic acid; and benz-
ene–methanol–formic acid (5:1:0.5, v/v/v) for 2.
Yield 90%. ½aꢁ ¼ þ4.0 (c 1.0, H2O), 3% ee. HPLC con-
D
ditions for its methyl ester: Chiralcel OD, hexane–2-pro-
panol (98:2, 0.8 mL/min), detection 228 nm; (S)
18.0 min, (R) 22.0 min. H NMR (500 MHz, D2O): d/
ppm: 5.560 (s, 1H), 7.284 (m, 2H), 7.350 (m, 1H),
7.406 (m, 1H).
1
4.5. Preparation of (S)-1 by fermentation
P. putida ECU1009, grown on SM for 24 h at 30 °C, was
pre-cultivated in 10 mL FM for 12 h and inoculated into
500 mL Erlenmeyer flasks containing 100 mL of the
liquid medium composed of 1.25% racemic mandelic
acid, 0.1% NH4NO3, 0.6% K2HPO4, 0.3% KH2PO4,
0.1% NaCl, 0.05% MgSO4, and 0.05% yeast extract at
pH 7.0, 30 °C, and 160 rpm. After 48 h, the chemical
yield and ee of mandelic acid were determined by HPLC
as described above.
Acknowledgements
4.4.1. Benzoylformic acid. White powder. Yield 2%
(isolated at 6 h). 1H NMR (500 MHz, D2O): d/ppm:
This research was partially supported by the National
Natural Science Foundation of China (Grant No.

H.-R. Huang et al. / Tetrahedron: Asymmetry 16 (2005) 2113–2117
2117
8. Tsuchiya, S.; Miyamoto, K.; Ohta, H. Biotechnol. Lett.
1992, 14, 1137–1142.
9. Takahashi, E.; Nakamichi, K.; Furui, M. J. Ferment.
Bioeng. 1995, 79, 439–442.
10. Kim, B. Y.; Hwang, K. C.; Song, H. S.; Chung, N.; Bang,
W. G. Biotechnol. Lett. 2000, 22, 1871–1875.
11. Nishiguchi, N.; Moritoki, M.; Shinohara, T.; Toyokura,
K. ACS Symp. Ser. 1997, 667, 73–82.
203900506), the Ministry of Science and Technology,
PR China (Grant No. 2003CB716008) and by an open
project program of the Key Laboratory of Industrial
Biotechnology, Ministry of Education, PR China.
References
12. Franz, E.; Brigitte, H.; Siegfried, F.; Thomas, Z. Tetra-
hedron Lett. 1990, 31, 1249–1252.
1. Furlemmeier, A.; Quitt, P.; Vogler, K.; Lanz P. U.S.
Patent, 3,957,758, 1976.
2. Mill, J.; Schmiegel, K. K.; Shaw, W. N. U.S. Patent,
4,391,826, 1983.
3. Saravanan, P.; Singh, V. K. Tetrahedron Lett. 1998, 39,
167–170.
4. Yamazaki, Y.; Maeda, H. Agric. Biol. Chem. 1986, 50,
2621–2631.
13. Hamaguchi, S.; Ogura, M.; Hasegawa, J.; Kawarada, H.;
Watanabe, K. Japanese Patent, 57-198097, 1982.
14. Lee, S. H.; Choi, J. H.; Park, S. H.; Choi, J. I.; Lee, S. Y.
Enzyme Microb. Technol. 2004, 35, 429–436.
15. Lehoux, I. E.; Mitra, B. Biochemistry 1999, 38, 9948–
9955.
16. Xu, Y.; Mitra, B. Biochemistry 1999, 38, 12367–12376.
17. Tsou, A. Y.; Ransom, S. C.; Gerlt, J. A. Biochemistry
1990, 29, 9856–9862.
5. Evans, D. A.; Morrissey, M. M.; Dorow, R. L. J. Am.
Chem. Soc. 1985, 107, 4346–4348.
6. Mori, K.; Akao, H. Tetrahedron 1980, 36, 91–96.
7. Whitesides, G. M.; Patterson, M. A. K.; Szajewski, R. P.
J. Org. Chem. 1981, 46, 4682–4685.
18. Fewson, C. A. FEMS Microbiol. Rev. 1988, 54, 85–110.
19. Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am.
Chem. Soc. 1982, 104, 7294–7299.