Facile synthesis of enantiopure 4-substituted 2-hydroxy-4butyrolactones using a robust Fusarium lactonase

Article abstract of DOI:10.1002/adsc.200900628

A facile chemo-enzymatic process has been developed for producing stereoisomers of 4substituted 2-hydroxy-4-butyrolactones with good to excellent enantioselectivity. This process involves an easy separation of the diastereoisomers by column chromatography

Full text of DOI:10.1002/adsc.200900628

FULL PAPERS
DOI: 10.1002/adsc.200900628
Facile Synthesis of Enantiopure 4-Substituted 2-Hydroxy-4-
butyrolactones using a Robust Fusarium Lactonase
Bing Chen,a Hai-Feng Yin,b Zhen-Sheng Wang,b Jian-He Xu,a,* Li-Qiang Fan,a
and Jian Zhaoa
a
State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, 130 Meilong Road,
Shanghai 200237, Peopleꢀs Republic of China
Fax : (+86)-21-6425-2250; e-mail: jianhexu@ecust.edu.cn
Department of Chemistry, PepTech Corporation (Shanghai), Building D, 388 Yindu Road, Shanghai 200231, Peopleꢀs
b
Republic of China
Received: September 10, 2009; Published online: November 18, 2009
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.200900628.
Abstract: A facile chemo-enzymatic process has 200 mM) were examined, giving a substrate to cata-
been developed for producing stereoisomers of 4- lyst ratio of up to 26:1. This general and efficient en-
substituted 2-hydroxy-4-butyrolactones with good to zymatic process provides access to stereoisomers of
excellent enantioselectivity. This process involves an 4-substituted 2-hydroxy-4-butyrolactones readily and
easy separation of the diastereoisomers by column cost-effectively. The stereochemical assignments
chromatography and efficient enzymatic resolution were conducted systematically based on NMR, X-ray
by whole cells of Escherichia coli JM109 expressing diffraction and circular dichroism, leading to further
Fusarium proliferatum lactonase gene. This biocata- understanding of the enzymeꢀs stereoselectivity.
lyst shows strong tolerance towards different sub-
strate structures and at least three out four possible Keywords: chemo-enzymatic process; circular di-
isomers could be obtained in excellent enantiomeric chroism; Fusarium proliferatum lactonase; stereose-
purity. Different substrate concentrations (10 mM– lectivity; 4-substituted 2-hydroxy-4-butyrolactones
Introduction
and b share common problems, such as expensive
starting materials or auxiliaries and complicated syn-
4-Substituted 2-hydroxy-4-butyrolactones comprise a thetic routes.[4,7] On the other hand, asymmetric catal-
vast family of compounds with great importance. As ysis has the advantage that it can produce large quan-
chiral building blocks, they have been widely used in tities of optically pure compounds from simple start-
the pharmaceutical industry for the production of ing materials using relatively small amount of cata-
ACE inhibitors[1a] such as Enalapril, Cilazapril, Bena- lysts.[8]
zepril etc. They are also used in the synthesis of natu-
However, the development of a suitable asymmet-
ral products with antitumor or cytotoxic activities, in- ric catalytic method is a challenging task. A successful
cluding Apoptolidin,[1b] Epothilones,[1c] Salicylihal- asymmetric catalysis should meet several criteria in-
amides[1d] and Bengazoles.[1e] Some of these lactones cluding ready availability and reusability of the cata-
have been found to be human hunger modulators[2] or lysts and the method should have general applicability
exhibit significant antitumor and cytotoxic activities and afford both enantiomers in high ee.
against cultures of P388 cells.[3]
Indeed, none of the existing catalytic methods
Due to the great interest in their four stereoiso- (strategy c) could be considered ideal or economical.
mers, different strategies to make these chiral lactones Casy et al.[5] reported the enzymatic reduction of the
have been developed. These include: a) chiral pool a-keto group of 2,4-diketo acids and subsequent non-
synthesis, using glucose,[1d] malic acid,[1c] xylose[3] and enzymatic syn- or anti-selective reduction, but the
so on; b) chiral auxiliaries, such as hydrazine SAEP[4] bio-reduction process requires a very low substrate-
and c) asymmetric catalysis, including enzymatic[5] to-catalyst ratio (from 1/20[9] to 8/1[10]) and prolonged
and non-enzymatic routes.[6] However, strategies a reaction time (from 2 days to 12 days, respectively).
Adv. Synth. Catal. 2009, 351, 2959 – 2966
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2959

Bing Chen et al.
FULL PAPERS
In addition, only one substrate, 4-phenyl-2,4-diketo- mers (Scheme 1) as substrates of reFPL. Crude com-
butyric acid, was studied in this method. Carpentier et pound 2, prepared according to a previous literature
al.[6] reported asymmetric hydrogenation of readily method,[5] with 80%–90% purity, was subjected to
available ethyl 2,4-diketo acids using a series of NaOH hydrolysis without further purification. After
metal/ligand systems. The efficiency and selectivity of acidification of the reaction mixture, the precipitate
the catalysts were highly dependent on the substrate of keto acid 3 was collected in nearly quantitative
structures and only two [(2R,4R)- and (2R,4S)-lac- yield as colourless crystals. It was reduced with
tones] of the four stereoisomers of 4-substituted 2-hy- NaBH4 in methanol, and subsequent in situ cycliza-
droxy-4-butyrolactones could be prepared.
tion afforded a mixture of cis- and trans-5 in 80–90%
Aiming for a practical catalytic approach to prepare yield, which was separated by column chromatogra-
the four stereoisomers of 4-substituted 2-hydroxy-4- phy on silica gel. We also examined the effects of
butyrolactones, we developed a novel procedure via temperature on the diastereoselectivity in NaBH4 re-
separation of 4-substituted 2-hydroxy-4-butyrolac- duction. Reduction of 2,4-dioxo-4-(o-fluorophenyl)-
tones into cis- and trans-isomers and subsequent bio-
resolution by Fusarium proliferatum lactonase.
ACHTUNGTRENNUNG
Fusarium lactonases had been shown to readily cat- using 19F NMR. Moderate diastereoselectivity of
alyze the enantioselective hydrolysis of 2-hydroxy-g- 1.25:1–1:1.5 (anti-4d/syn-4d) was observed when the
lactones to optically pure hydroxy acids twenty years temperature was varied between ꢀ408C and 608C.[15]
ago,[11] since then recombinant E. coli cells expressing We did not further optimize this ratio, but Casy[5] re-
Fusarium lactonase have gradually become an easy- ported that it was possible to obtain a ratio from 6:1
to-make biocatalyst.[12] However, most of the applica- to 1:50 (cis-5c/trans-5c), which would help to improve
tions were focused on the production of the (R)- our theoretical yield from 50% to 100%. In this pro-
isomer of 2-hydroxy-g-lactones, especially d-pantolac- cess, only one column chromatography was needed to
tone.[13] In our group, we had cloned Fusarium prolif- separate the cis- and trans-isomers, which makes the
eratum lactonase gene[14] and expressed it into E. coli synthesis of racemic lactones more cost effective.
JM109. In this paper, we report for the first time the
successful application of this biocatalyst in the enan-
tioselective synthesis of 4-substituted 2-hydroxy-4-bu- Enzymatic Resolution
tyrolactones with two chiral centres and four stereo-
isomers.
After cis-5 and trans-5 had been separated, we sub-
mitted them to enzymatic resolution (Scheme 2).
Both racemic cis-5 and trans-5 were readily recog-
nized (except 5g, possibly because the p-pentylphenyl
group is too bulky and could not be accepted by the
active site of the enzyme) by recombinant E. coli
JM109 cells harboring plasmid pET28a carrying Fusa-
Results and Discussion
Synthesis of the Racemic Substrates
The present study began with the preparation of cis- rium proliferatum lactonase (reFPL) gene. No side re-
and trans-4-substituted 2-hydroxy-4-butyrolactone iso- action was observed. E. coli JM109 harboring plasmid
Scheme 1. Synthesis of 4-substituted 2-hydroxy-4-butyrolactones.
2960
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2959 – 2966

Facile Synthesis of Enantiopure 4-Substituted 2-Hydroxy-4-butyrolactones
FULL PAPERS
Scheme 2. Fusarium lactonase catalyzed chemo-enzymatic synthesis of four isomers of 4-substituted 2-hydroxy-4-butyrolac-
tones (R=aromatic or heteroaromatic rings).
pET28a alone showed no hydrolytic activity, indicat- 2R instead of 2S). For (2R)-5, good to excellent enan-
ing that the asymmetric hydrolysis was totally due to tioselectivity (84.3% to >99.0% ee) could also be ob-
the activity of reFPL.
tained using the same biocatalyst. Most of the cis-5
In a typical resolution process, we chose cis-5c as isomers, except 5a, could be crystallized, which facili-
the model substrate. To prepare (2S)-5c, a 10 mM so- tates the isolation and purification of these com-
lution of racemic cis-5c in 10 mL potassium phosphate pounds. Therefore at least three of the four possible
buffer (KPB, 50 mM, pH 6.4) was treated with an ap- stereoisomers could be prepared in optically pure
propriate amount of the biocatalyst (wet cells of E. form by using one single robust biocatalyst. For trans-
coli JM109). MeCN (200 mL) was chosen as co-solvent (2R,4S)-5, good (76.5% ee) to excellent (95.3% ee)
to enhance the solubility of 5c in water. The biocata- enantioselectivities were achieved.
lytic hydrolysis took one hour, and by the end of the
This enzyme showed higher hydrolytic activity to-
resolution, (2S)-isomer of cis-5c was isolated by wards 5a than other lactones with more bulky aromat-
simple extraction in near theoretical yield and >99% ic or heteroaromatic ring substituents, possibly due to
ee (Table 1).[15,16]
the smaller size of the ethyl group in 5a which may fit
When (2R)-5c was the target compound, undesired better in the enzymeꢀs active site. For 5b–5f, the
(2S)-5c needs to be extracted completely, and the resi- enzyme showed higher hydrolytic activity towards
due was acidified with 2N HCl to pH 2.0 and heated trans-5 than the corresponding cis-5 with the same ar-
for one hour at 608C.[15] The cyclized (2R)-5c was omatic ring substituent but different conformations,
then isolated (>99% ee) by extraction.
indicating the conformations of substrates also play
an important role in the process of enzymatic hydrol-
ysis. The enzymatic activity was also affected by the
substituents on the aromatic ring. Substrate 5e with
an electron-donating group on the phenyl ring inhibit-
Relaxed Substrate Specificity of reFPL
A series of lactones were then tested in the resolution ed the activity of reFPL.
process described above. Various substrate structures
We also tested the biocatalystꢀs hydrolytic activity
could be accepted by this robust catalyst, indicating a towards other substrates (Figure 1). As reported pre-
great generality of this method (Table 1). Excellent viously, lactones 7a and 7b could be readily recog-
enantioselectivity (>99% ee) and isolated yield nized by this enzyme. When the hydroxy group was
(50.8%–57.8%) were obtained for all of the (2S)-5 alkylated (8), the enzyme could no longer hydrolyze
substrates in the process of enzymatic resolution (in this compound. Compounds 6a, 6b and 9 are not suit-
the case of 5e, the isolated isomer turned out to be able substrates for this enzyme either. It is well
Adv. Synth. Catal. 2009, 351, 2959 – 2966
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2961

Bing Chen et al.
Conversion [%]
FULL PAPERS
Table 1. Biocatalyzed asymmetric hydrolysis of 4-substituted 2-hydroxy-4-butyrolactones.
Entry
Biocatalyst (IU)[a]
3
Configuration
ee [%]
1
2
3
4
5
6
7
8
cis-(2R,4S)-5a
cis-(2S,4R)-5a
cis-(2R,4R)-5b
cis-(2S,4S)-5b
trans-(2S,4R)-5b
cis-(2R,4R)-5c
cis-(2S,4S)-5c
trans-(2R,4S)-5c
trans-(2S,4R)-5c
cis-(2R,4R)-5d
cis-(2S,4S)-5d
trans-(2R,4S)-5d
trans-(2S,4R)-5d
cis-(2S,4S)-5e
cis-(2R,4R)-5e
cis-(2R,4R)-5f
cis-(2S,4S)-5f
96.5[c]
>99[c]
84.3[d]
>99[d]
>99
34.5[f]
51.3 (0.3 h)[g]
25.4[f]
10[b]
57.3 (2 h)[g]
52.1 (1.5 h)[g]
30.0{f]
10
10
10
10
15
10
10
>99[d]
>99[d]
95.3[e]
>99[e]
>99[e]
>99[e]
86.5[d]
>99[d]
80.3[d]
>99[d]
98.3[e]
>99[e]
76.5[e]
>99[d]
51.4 (1.5 h)[g]
36.9[f]
9
53.4 (1.0 h)[g]
37.8[f]
10
11
12
13
14
15
16
17
18
19
50.8 (1.5 h)[g]
32.1[f]
53.6 (1.2 h)[g]
34.2[f]
53.4 (3 h)[g]
34.6 (1 h)[g]
52.1 (1.5 h)[g]
40.2[f]
trans-(2R,4S)-5f
trans-(2S,4R)-5f
55.4 (1.4 h)[g]
Reaction conditions: 308C; 10 mL KPB buffer (pH 6.4, 50 mM) containing 2% (v/v) MeOH was mixed with 10ꢀ4 mol sub-
strate (10 mM); E. coli wet cells were added in various amounts; details for enzyme assay are summarized in the Support-
ing Information.
Due to the unstable nature of trans-5c, we did not prepare trans-(2S,4R)-5c, and the absolute configuration was postulat-
ed according to the enantioselectivity of reFPL toward cis-5c.
[a]
[b]
[c]
[d]
[e]
[f]
Analyzed by chiral GC b-cyclodextrin column.
Analyzed by chiral HPLC with an AD-H column.
Analyzed by chiral HPLC with an OJ-H column.
Conversion was calculated from formula: c=ees/ACTHNUTRGNEUNG
(ees+eep).[16]
[g]
Isolated yield is given.
amount of 0.13 g of wet cells (about 0.02 g of dry
cells) in a 10 mL reaction system (Figure 2). When
the substrate concentration was increased to 300 mM,
the percentage of MeCN had to be increased to 40%
(v/v) leading to complete loss of the enzymatic activi-
ty of the whole cells. Moreover, this biocatalyst could
Figure 1. Compounds tested in the enzymatic hydrolysis.
known that the 2-hydroxy group is essential in this en-
zymatic hydrolysis. Our results on 2-methoxylactone
suggest that the hydrogen in the hydroxy group may
also play an important role in this process.
Increased Substrate Load
A high ratio of substrate-to-catalyst (nearly 26 g of
substrate/1 g of biocatalyst in dry cell weight) could
also be achieved in this strategy, suggesting high effi-
ciency of this biocatalyst. When cis-5c was again
chosen as the model substrate, the substrate concen-
tration could be increased to 100–200 mM and the re-
action is complete within 1–3 h with conversion of ca.
Figure 2. ReFPL-catalyzed asymmetric hydrolysis of cis-5c
65% [ee of cis-(2S)-5c >99.0%] using a catalytic at different substrate concentrations.
2962
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2959 – 2966

Facile Synthesis of Enantiopure 4-Substituted 2-Hydroxy-4-butyrolactones
FULL PAPERS
trans lactones show the proton signal at the 2-position
as a dd peak with the same coupling constants of
~7.8 Hz (J2,3 and J2,3’) while the proton signal at the 4-
position is a dd peak with the coupling constants of
~4.5 Hz and ~7.5 Hz. NOE experiments also validate
these findings.
The absolute structure (AS) assignment of chiral 4-
substituted 2-hydroxyl-4-butyrolactones was a chal-
lenge. Ito et al. assigned[3] the AS of natural product
(+)-harzialactone A (4-benzyl 2-hydroxy-4-butyrolac-
tone) using the Mosher ester method but the results
were challenged by Mereyala et al.[3] Kiegiel et al. es-
tablished the configuration of chiral 4-methyl 2-hy-
droxy-4-butyrolactone unambiguously by transform-
ing this compound into its derivative which could
crystallize to afford the single crystals suitable for X-
ray analysis.[20] However, this method is limited to
crystalline compounds (or its derivatives) only. Al-
Figure 3. Reuse of whole cells of recombinant E. coli JM109
expressing Fusarium proliferatum lactonase gene.
be reused for at least three times at a substrate con- though the limitation to crystalline compounds could
centration of 100 mM (20% MeCN, v/v) with accepta- be overcome by synthesis from chiral building blocks,
ble conversion rate (Figure 3) in less than 8 h, which it could be time-consuming to do so.
could improve the ratio of substrate to catalyst from
Circular dichroism (CD), which has been widely
8.8 g substrate/1 g catalyst to 26 g substrate/1 g cata- used in the determination of absolute configurations
lyst. About 10% of the initial enzyme activity re- of natural products and small molecule drugs,[21] is our
mained[15] in the fourth use and it was necessary to choice for the assignment of the absolute configura-
prolong the reaction time to more than 8 h. We are tion of 2-hydroxy-4-butyrolactones in a facile and
currently working on the immobilization of the whole direct way. We examined the CD spectra of nine com-
cells that may improve the enzymeꢀs stability against pounds[16] and divided them into two types. Type a in-
high concentration of MeCN to further improve the cludes trans-(2S,4R)-5c, 5d, 5f with positive C.E.
ratio of substrate-to-biocatalyst and reduce the cost of curves in the n!p* transition region at ca. 223.5–
this process.
233 nm indicating S configuration of the hydroxy
group.[22] Type b consists of cis-5b, 5c, 5d, 5e and 5f,
among which cis-(2S,4S)-5b, 5c, 5d and 5f show a pos-
itive sign in the n!p* transition region (ca. 230–
250 nm) to confirm the S configuration of the hydroxy
group while cis-(2R,4R)-5e and cis-(2R,4R)-5f give a
Stereochemical Assignment of Chiral 4-Substituted 2-
Hydroxy-4-buyrolactones
1
We used H NMR and NOE experiments to assign negative sign in the n!p* transition region indicating
the relative configuration (Figure 4).[6,17] In cis lac- the R configuration of the hydroxy group. From these
tones H-2 exhibits a dd peak with coupling constants CD results, we could deduce that g-lactones with a
of ~8.1 Hz (J2,3’) and ~11.1 Hz (J2,3) while H-4 also (2S)-hydroxy group would give a positive Cotton
shows a dd peak with coupling constants of ~5.1 Hz effect curve in the n!p* region, obeying the Okudaꢀs
1
(J4,3’) and ~11.1 Hz (J4,3).[18,19] The H NMR spectra of rule.[23] The substituent attached to the g-carbon (C-4)
has little, if any, contribution in the sign of the C.E.
curve.
X-ray diffraction analyses of crystalline cis-(2S,4S)-
5b and cis-(2S,4S)-5f[24] gave unambiguous stereo-
chemical assignments including relative and absolute
configurations and the “equatorial” orientation of
both hydroxy and aromatic (heteroaromatic) groups
(Figure 5).
Conclusions
An example of the successful asymmetric catalysis
was described in this contribution. The readily avail-
Figure 4. Coupling constants of H-2 and H-4 in 4-substituted
2-hydroxy-4-butyrolactones.
able biocatalyst affords high ee towards both enantio-
Adv. Synth. Catal. 2009, 351, 2959 – 2966
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2963

Bing Chen et al.
FULL PAPERS
Figure 5. Crystal structure of cis-(2S,4S)-5b (top) and cis-(2S,4S)-5f (bottom).
mers with reusability and great generality. The biocat- Experimental Section
alytic resolution is highly efficient, and the optically
pure product was isolated by simple extraction. In ad- Synthesis of Intermediate 2
dition to the “elegance” of the biocatalysis, the sub-
To a slurry of newly prepared NaOEt (0.1 mol) and Et2O
(50 mL), a mixture of diethyloxalic acid (0.1 mol) and com-
pound 1 (0.1 mol) in dry Et2O (15 mL) solution was added
carefully in small portions (in the case of 1b and 1e, the tem-
perature was controlled at ꢀ208C) and kept overnight. The
mixture was then poured into dilute HCl and extracted with
ethyl acetate. The combined organic layers were dried over
anhydrous sodium sulfate, filtered, and concentrated under
vacuum. Purification of the product (compound 2) was done
by column chromatography on silica gel (petroleum ether:
ethyl acetate=5:1).
strate was prepared in a facile way from readily avail-
able starting materials. All of these could compensate
for the drawbacks in theoretical yield and contribute
to the robustness of this process.
In summary, we have developed a versatile process
towards chiral 4-substituted 2-hydroxy-4-butyrolac-
tones with a single biocatalyst. We also developed a
CD method to establish the stereochemistry of these
compounds. This process combines the ready availa-
bility of the starting materials and a highly efficient
biocatalyst and should prove to be useful in the syn-
thesis of natural products and pharmaceutical com-
pounds.
Synthesis of Intermediate 3
To a 250-mL round-bottom flask was added crude com-
pound 2 (42 mmol) in 50 mL THF, followed by LiOH
(50.4 mmol) in 50 mL water in one portion. Saponification
was terminated when compound 2 was consumed complete-
ly by TLC analysis (petroleum ether:ethyl acetate=1:1).
2964
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2959 – 2966

Facile Synthesis of Enantiopure 4-Substituted 2-Hydroxy-4-butyrolactones
FULL PAPERS
The mixture was concentrated under vacuum to remove the
THF and then was extracted with ethyl acetate (50 mL)
three times. The aqueous layer was acidified with 6N HCl
and a large amount of compound 3 precipitated. This pre-
cipitate was filtered and washed with water (100 mL) three
times and dried under vacuum. It can be used without fur-
ther purification.
100-mL flask was kept at 308C and shaken at 180 rpm. The
resolution process was monitored by chiral HPLC (AD-H
or OJ-H columns) and terminated at the appropriate time.
(2S)-5 was extracted with ethyl acetate directly, while the so-
lution containing (2R)-5 was worked up by acidification
(pH 2.0, 2N HCl) and heating (608C, 1 hour) after which
the product was extracted with ethyl acetate.
Scale-Up Preparation of cis-(2S,4S)-5c
Synthesis of 2-Hydroxy-4-butyrolactones 5
An appropriate amount of wet cells was resuspended in
10 mL potassium phosphate buffer (KPB, 50 mM, pH 6.4).
To 500 mL water was added an appropriate volume of this
aqueous solution and 17.8 g of cis-5c in 150 mL MeCN. This
mixture in a 1-L beaker was kept in room temperature and
stirred. The pH was automatically controlled at 6.3–6.5 with
0.1M NaOH. The resolution was monitored by HPLC (AD-
H column) and terminated when the ee value of cis-(2S,4S)-
5c reached >99%. The biocatalyst was centrifugated and
the supernant was extrated with ethyl acetate directly. Con-
centration of this organic layer afforded 5.5 g of cis-(2S,4S)-
5c as colourless crystals.
To a 250-mL round-bottom flask containing a stirrer, crude
compound 3 (47.6 mmol) in 50 mL MeOH was added, fol-
lowed by NaBH4 (190 mmol) in small portions at room tem-
perature. The mixture was stirred for another 3 h. 50 mL
water were added to decompose residual NaBH4 and the
mixture of 4 was concentrated under vacuum to remove the
solvent. Then the mixture was acidified to pH 1.0–2.0 and
heated at 608C to afford compound 5. 5 was extracted with
ethyl acetate and dried over anhydrous sodium sulfate. The
separation of the diastereomers was done by column chro-
matography on silica gel (petroleum ether:ethyl acetate=
6:1~3:1).
Scale-Up Preparation of cis-(2R,4R)-5c
Expression of Recombinant Fusarium proliferatum
(reFPL) Lactonase
An appropriate amount of wet cells was resuspended in
10 mL potassium phosphate buffer (KPB, 50 mMm, pH 6.4).
To 500 mL water was added an appropriate volume of this
aqueous solution and 17.8 g of cis-5c in 150 mL MeCN. This
mixture was kept in room temperature and stirred. The pH
was automatically controlled at 6.2–6.4 with 0.1M NH3·H2O.
The resolution was terminated when conversion reached
around 30%. The biocatalyst was centrifugated and the su-
pernant was concentrated in vacuum to remove MeCN and
then extrated with ethyl acetate to remove the cis-(2S,4S)-5c
completely. While syn-(2R,4R)-4c in this aqueous layer was
acidified and heated, after which the product was extracted
with ethyl acetate to afford 4.9 g of crystals. Recrystalliza-
tion of these crystals in petroleum ether/ethyl acetate (5:1)
gave 4.1 g of colourless crystals in >99% ee.
FPL gene sequence was ligated with Nde I and BamH I di-
gested pET28a. The ligation mixture was introduced by
transformation into E. coli JM109 and then screening of
clones containing the restriction fragment was performed.
The resultant plasmid was designated as pET28a-FPL. The
recombinant E. coli JM109 (DE3) containing pET28a-FPL
was cultured aerobically at 378C in 50 mL LB medium until
OD600 reached 0.7, and then the cultures were immediately
shifted to 238C. IPTG was added at a final concentration of
0.5 mM to induce the lac promoter. After further cultiva-
tion, the cells were harvested by centrifugation and washed
thoroughly with physiological saline, and then were used for
further studies.
Enzyme Assay
The standard assay mixture comprised 2.5% (w/v) racemic
pantolactone and an appropriate amount of the cells with a
final volume of 10.0 mL. The reaction was performed at
308C, and the pH was automatically controlled at 6.7–6.9
with 0.1M NaOH. The hydrolysis rate of lactone was calcu-
lated based on the rate of NaOH titration. One unit of lac-
tonase (1 IU) was defined as the amount of the enzyme that
converted 1.0 mmol of pantolactone into pantoic acid per
minute under the above conditions.
Acknowledgements
This research was financially supported by National Natural
Science Foundation of China (grant nos. 20506037 and
20773038), Ministry of Science and Technology (grant nos.
2007AA02Z225 and 2009CB724706), and China National
Special Fund for State Key Laboratory of Bioreactor Engi-
neering (grant no. 2060204).
Resolution of Substrates by Recombinant E. coli
Cells
References
An appropriate amount of wet cells was re-suspended in
10 mL potassium phosphate buffer (KPB, 50 mM, pH 6.4)
and the enzyme activity of this aqueous solution was mea-
sured as 13.3 IU/100 mL.
General procedures were as following: To 10 mL potassi-
um phosphate buffer (KPB, 50 mM, pH 6.4) was added an
appropriate volume of this aqueous solution and 0.1 mmol
of cis-5 (or trans-5) in 200 mL methanol. This mixture in a
[1] a) V. Blandin, J.-F. Carpentier, A. Mortreux, Eur. J.
Org. Chem. 1999, 1787–1793, and references cited
therein; b) K. C. Nicolaou, K. C. Fylaktakidou, H.
Monenschein, Y. Li, B. Weyershausen, H. J. Mitchell,
H. Wei, P. Guntupalli, D. Hepworth, K. Sugita, J. Am.
Chem. Soc. 2003, 125, 15433–15442; c) D. Schinzer, A.
Bauer, J. Schieber, Chem. Eur. J. 1999, 5, 2492–2500;
d) G. I. Georg, Y. M. Ahn, B. Blackman, F. Farokhi,
Adv. Synth. Catal. 2009, 351, 2959 – 2966
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2965

Bing Chen et al.
FULL PAPERS
P. T. Flaherty, C. J. Mossman, S. Roy, K. Yang, Chem.
Commun. 2001, 255–256; e) M. Adamczeski, E.
Quinoa, P. Crews, J. Am. Chem. Soc. 1988, 110, 1598–
1602.
[14] The gene sequence of Fusarium proliferatum lactonase
has now been deposited in GenBank with an accession
number of EU596535.
[15] See Supporting Information.
[16] C.-S. Chen, Y. Fujimoto, G. Girdaukas, C. J. Sih, J. Am.
Chem. Soc. 1982, 102, 7294–7299.
[17] S. E. Denmark, T. Bui, J. Org. Chem. 2005, 70, 10190–
10193.
[18] The assignment of these coupling constants is support-
ed by calculation using Karplus equation. Although cal-
culations of vicinal coupling constants based on dihe-
dral angles can be tricky, some correlation can still be
found between the coupling constants (e.g., J2,3 and
J2,3’) and relative magnitudes of dihedral angles. The X-
ray diffraction of cis-5b and trans-5f (Figure 5) shows
[2] a) K. P. Puthuraya, Y. Oomura, N. Shimizu, Brain Res.
1985, 332, 165–168, and references cited therein; b) O.
Uchikawa, N. Okukado, T. Sakata, K. Arase, K.
Terada, Bull. Chem. Soc. Jpn. 1988, 61, 2025–2029, and
references cited therein; c) T. Nakano, Y. Ino, Y.
Nagai, Chem. Lett. 1989, 567–568.
[3] a) T. Amagata, Y. Usami, K. Minoura, T. Ito, A.
Numata, J. Antibiot. 1998, 51, 33–40; b) H. B. Mereya-
la, M. Joe, R. R. Gadikota, Tetrahedron: Asymmetry
2000, 11, 4071–4081.
[4] D. Enders, H. Sun, F. R. Leusink, Tetrahedron 1999, 55,
6129–6138.
ꢀ
ꢀ
ꢀ
that the dihedral angles H2 C2 C3 H3 (25.38 for cis-
ꢀ
ꢀ
ꢀ
5f and 18.98 for cis-5b) are lower than H2 C2 C3 H3’
[5] G. Casy, Tetrahedron Lett. 1992, 33, 8159–8162.
[6] V. Blandin, J.-F. Carpentier, A. Mortreux, Eur. J. Org.
Chem. 1999, 1787–1793, and references cited therein.
[7] J. M. Keith, J. F. Larrow, E. N. Jacobsen, Adv. Synth.
Catal. 2001, 343, 5–26.
[8] a) J. M. Hawkins, T. J. N. Watson, Angew. Chem. 2004,
116, 3286–3290; Angew. Chem. Int. Ed. 2004, 43, 3224–
3228; b) S. E. Denmark, E. N. Jacobsen, Acc. Chem.
Res. 2000, 33, 324.
ꢀ
ꢀ
ꢀ
(33.28 for cis-5f and 38.28 for cis-5b) and H4 C4 C3
H3 (26.18 for cis-5f and 20.78 for cis-5b) are lower than
H4 C4 C3 H3ꢀ (32.38 for cis-5f and 36.38 for cis-5b),
which indicate that J4,3 should be larger than J4,3’
ꢀ
ꢀ
ꢀ
.
[19] a) C. F. Ingham, R. A. Massy-Westropp, G. D. Rey-
nolds, Aust. J. Chem. 1974, 27, 1477–1489; b) C. D.
Nardo, L. O. Jeroncic, R. M. Lederkremer, O. Varela, J.
Org. Chem. 1996, 61, 4007–4013.
[20] K. Kiegiel, T. Balakier, P. Kwiatkowski, J. Jurczak, Tet-
[9] N. W. Fadnavis, K. R. Radhika, Tetrahedron: Asymme-
rahedron: Asymmetry 2004, 15, 3869–3878.
try 2004, 15, 3443–3447.
[10] H. Yun, H.-L. Choi, N. W. Fadnavis, B.-G. Kim, Bio-
technol. Prog. 2005, 21, 366–371.
[21] a) N. Ikemoto, L.-C. Lo, K. Nakanishi, Angew. Chem.
1992, 104, 918–919; Angew. Chem. Int. Ed. Engl. 1992,
31, 890–891; b) J. B. MacMillan, R. G. Linington, R. J.
Andersen, T. F. Molinski, Angew. Chem. 2004, 116,
6072–6077; Angew. Chem. Int. Ed. 2004, 43, 5946–
5951.
[11] a) K. Sakamoto, K. Honda, K. Wada, S. Kita, K. Tsuza-
ki, H. Nose, M. Kataoka, S. Shimizu, J. Biotechnol.
2005, 118, 99–106; b) M.-R. Yu, T.-W. Tan, Process
Biochem. 2005, 40, 2609–2614; c) Y.-X. Tang, Z.-H.
Sun, L. Hua, C.-F. Lv, X.-F. Guo, J. Wang, Process Bio-
chem. 2002, 38, 545–549; d) H. Yamada, S. Shimizu,
Angew. Chem. 1988, 100, 640–661; Angew. Chem. Int.
Ed. Engl. 1988, 27, 622–642.
[12] a) M. Kobayashi, M. Shinohara, C. Sakoh, M. Kataoka,
S. Shimizu, Proc. Natl. Acad. Sci. USA 1998, 95, 12787–
12792; b) Z.-Q. Liu, Z.-H. Sun, Chin. J. Biotechnol.
2005, 21, 390–395.
[22] a) A. M. Mathieson, Tetrahedron Lett. 1963, 4, 81–84;
b) H. Wolf, Tetrahedron Lett. 1966, 7, 5151–5156;
´
c) M. J. Milewska, M. Gdaniec, T. Polonski, Tetrahe-
dron: Asymmetry 1996, 7, 3169–3180; d) J. P. Jennings,
W. Klyne, P. M. Scopes, J. Chem. Soc. 1965, 7211–7229.
[23] T. Okuda, S. Harigaya, A. Kiyamoto, Chem. Pharm.
Bull. 1964, 12, 504–506.
[24] CCDC 736155 and 736156 contain the supplementary
crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[13] X. Zhang, J.-H. Xu, Y. Xu, J. Pan, Appl. Microbiol.
Biotechnol. 2007, 75, 1087–1094.
2966
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2959 – 2966