Easy separation of optically active products by enzymatic hydrolysis of soluble polymer-supported substrates

Article abstract of DOI:10.1016/j.tetlet.2007.09.141

The easy separation of optically active compounds from enzymatic kinetic resolution products by simple precipitation using poly(ethylene glycol)(PEG)-supported carbonates is disclosed. The water-soluble substrate was prepared by the immobilization of (±)-1-phenylethanol onto a middle-molecular weight (av M_w 5000) monomethoxy PEG (MPEG) through a carbonate linker. The enantioselective hydrolysis using Lipase from porcine pancreas (PPL; Type II, Sigma) in a mixed solvent (hexane/buffer = 9:1) proceeded to afford the corresponding optically active compounds. In this system, the separation of the products was achieved by a simple procedure without laborious column chromatography. A hydrophobic spacer between the MPEG moiety and the carbonate linker affected both the reactivity and enantioselectivity, and the substrate with a phenylethyl spacer was hydrolyzed with the highest enantioselectivity (E value = 270).

Full text of DOI:10.1016/j.tetlet.2007.09.141

Available online at www.sciencedirect.com
Tetrahedron Letters 48 (2007) 8540–8543
Easy separation of optically active products by enzymatic
hydrolysis of soluble polymer-supported substrates
Masayuki Okudomi, Megumi Shimojo, Masaki Nogawa, Ayae Hamanaka,
Natsue Taketa and Kazutsugu Matsumoto*
Department of Chemistry, Meisei University, Hodokubo 2-1-1, Hino, Tokyo 191-8506, Japan
Received 18 August 2007; revised 17 September 2007; accepted 20 September 2007
Available online 26 September 2007
Abstract—The easy separation of optically active compounds from enzymatic kinetic resolution products by simple precipitation
using poly(ethylene glycol)(PEG)-supported carbonates is disclosed. The water-soluble substrate was prepared by the immobiliza-
tion of (± )-1-phenylethanol onto a middle-molecular weight (av M 5000) monomethoxy PEG (MPEG) through a carbonate linker.
w
The enantioselective hydrolysis using Lipase from porcine pancreas (PPL; Type II, Sigma) in a mixed solvent (hexane/buffer = 9:1)
proceeded to afford the corresponding optically active compounds. In this system, the separation of the products was achieved by a
simple procedure without laborious column chromatography. A hydrophobic spacer between the MPEG moiety and the carbonate
linker affected both the reactivity and enantioselectivity, and the substrate with a phenylethyl spacer was hydrolyzed with the highest
enantioselectivity (E value = 270).
ꢀ
2007 Elsevier Ltd. All rights reserved.
The kinetic resolution of racemic alcohols and esters
using hydrolytic enzymes is one of the practical methods
for the preparation of optically active alcohols, and a
analysis of the PEG-supported substrates, and the easy
separation of the products by an extraction procedure
was achieved. However, for the isolation of the
MPEG-supported compounds through all processes of
the substrate syntheses, the column chromatography
steps were still needed because the compounds were
liquids. In this Letter, we report the enzyme-mediated
kinetic resolution of higher-molecular weight (av Mw
5000) MPEG-supported carbonates, which are solids
and easier to handle. Furthermore, the introduction of
an appropriate hydrophobic spacer between the MPEG
moiety and the carbonate linker affects both the reactiv-
ity and enantioselectivity.
1
significant number of examples have been published.
During the reaction process, the enantiomers, the
remaining substrate and the resulting product, could
be mainly separated by column chromatography. How-
ever, the tedious and wasteful purification step is a
bottleneck to an easy operation and a sustainable synthe-
sis. Although several studies of easy separation processes
2
–5
have been published,
facile and efficient procedures
are still desired. Recently, poly(ethylene glycol) (PEG)
has been recognized as an inexpensive and convenient
6
–9
soluble polymer,
and we noticed that a PEG-sup-
ported strategy could be suitable for an enzymatic trans-
formation and potentially useful for the easy isolation of
the products. We have already succeeded in the kinetic
resolution of a low-molecular weight monomethoxy
PEG (MPEG, av Mw 550 or 750)-supported substrate
with a carbonate linker using a hydrolytic enzyme
We chose the carbonate (± )-1 and 2 as the substrates
with and without a spacer, respectively. The carbonate
is a typical linker used in organic synthesis on a polymer
support, and (± )-1 was easily prepared by the coupling
of the racemic 1-phenylethanol ((± )-3) with
1
1,12
MPEG5000–OH (Scheme 1).
We are also interested
(
porcine pancreas lipase (PPL), lipase Type II from
in the affect of a hydrophobic spacer between the MPEG
moiety and the carbonate linker. The substrates (± )-2
were also synthesized in 4 steps from MPEG5000–
1
0
Sigma). The broad solubility of PEG facilitated the
1
2
OH. By making use of the MPEG5000, which has been
utilized in many studies, it allows us to isolate and purify
the substrates (± )-1 and 2 by a simple precipitation pro-
Keywords: Carbonates; Enantioselective hydrolysis; Hydrolase; Poly-
(ethylene glycol)-supported substrates; Spacer.
*
Corresponding author. Tel./fax: +81 42 591 7360; e-mail: mkazu@
chem.meisei-u.ac.jp
7
cedure from diethyl ether.
0040-4039/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2007.09.141

M. Okudomi et al. / Tetrahedron Letters 48 (2007) 8540–8543
8541
O
Ph
Me
N
N
O
O
Ph
Me
(
±)-4
DMAP, DMF
00 °C
MPEG5000-OH
O
O
1
(±)-1 (88%)
OMe
HO
n
O
a: n = 0
b: n = 1
c: n = 2
MsCl, Et N
6a-c
3
OMe
MPEG5000-OH
OMs
(98%)
O
n
CH Cl
Cs CO , DMF
2
2
2
3
O
rt
70 °C
5
7
7
7
a (97%)
b (96%)
c (97%)
O
Ph
DIBAL
CH Cl
(±)-4
DMAP, DMF
00 °C
OH
O
O
Me
O
n
O
n
2
2
1
0
°C
8
8
8
a (97%)
b (90%)
c (87%)
(±)-2a (95%)
(±)-2b (95%)
(±)-2c (99%)
Scheme 1.
At first, we examined the PPL-catalyzed reaction of (± )-
under the same reaction conditions as those for the
MPEG550-supported substrates (0.1 M phosphate buffer
were up to 0.44 and 0.30, respectively, besides the car-
bonates (± )-2a (n = 0) and 2c (n = 2) were hydrolyzed
with very high enantioselectivities and E values were
also up to 151 and 74, respectively. These results indi-
cate that the substrates would more favorably fit to
the enzyme active site. On the other hand, the phenyl-
ethyl spacer in the substrate 2b (n = 1) gave a different
effect (entry 8). The hydrolysis of (± )-2b proceeded with
the highest enantioselectivity (E value = 270) to afford
the almost optically pure (R)-3, although the conversion
was slightly improved against that of the original sub-
1
1
0
(
pH 6.5) medium; 24 h; 30 ꢁC). As expected, it was
found that the hydrolysis of (± )-1 proceeded to afford
the optically active (S)-1 (56% ee) and (R)-3 (82% ee)
(
conv. = 0.41, E value = 18),13,14 although the reactivity
and enantioselectivity were lower than those of the
corresponding substrate with MPEG (conv. = 0.55,
5
50
1
0
E value = 29). In this case, the extraction process
was also necessary for recovering the unreacted 1, and
there were not many advantages for the easy prepara-
tion versus using MPEG550. We then tried to examine
the reaction in a mixed solvent (organic solvent/
buffer = 9:1). In a typical experiment, 125 mg of (± )-1
1
6
strate (± )-1. The result suggests that the introduction
of the phenylethyl spacer could drastically lower the
affinity of the slow reactive enantiomer of 2b for the
enzyme.
(
sub. concn 5 mM) and 10 mg of PPL were added to a
hexane–buffer (4.5 mL:0.5 mL), and stirred at 30 ꢁC
Table 1, entry 1). Fortunately, the enantioselective
The use of the MPEG5000 and the two-phase system also
enabled us to achieve an easier work-up procedure
(Scheme 2). Only the resulting alcohol (R)-3 was
extracted in the hexane layer, and was isolated after
evaporation. On the other hand, the aqueous layer was
diluted with CH Cl , and the dehydration was per-
(
hydrolysis smoothly proceeded even in this case. While
the use of the organic solvent did not improve the
enantioselectivity, the conversion was up to 0.58
(
conv. = 0.58, E value = 13). Changing the organic
2
2
solvent to toluene and i-Pr O (entries 2 and 3) decreased
formed with anhydrous Na SO . After evaporation,
2
2 4
both the reactivity and enantioselectivity. For the kinetic
controlled reaction, the temperature could be one of
the important factors in many cases. We then investi-
gated the temperature effect of the reaction (entries
the residue was poured into Et O to precipitate the
2
mixture of the MPEG5000-supported carbonate and
MPEG5000–OH as a white solid, which was collected
by simple filtration. The chemical hydrolysis with
4
–6). As expected, lowering the temperature apparently
NaOH in MeOH–H O gave the optically active (S)-3.
2
improved the enantioselectivity. In particular, for the
reaction at 0 ꢁC (entry 6), the E value was up to 58
and (R)-3 with 96% ee was obtained, although the con-
The concept of this reaction was applicable to the prep-
aration of not only the optically active 3, but also other
optically active compounds. For example, the reaction
of substrate (± )-9, which was constructed from
MPEG5000 and (± )-4-benzyloxy-2-butanol (10), also
proceeded with enantioselectivity to afford the corres-
ponding optically active compounds, although the reac-
tion conditions were not necessarily optimized (Scheme
1
5
version apparently decreased to 0.16. We considered
that a suitable spacer could increase the affinity to the
active site of the enzyme, which had many hydrophobic
amino residues inside the binding pocket. Beyond our
expectation, the introduction of the spacer drastically
increased not only the conversion, but also the enantio-
selectivity. For the cases of the phenylmethyl and
phenylpropyl spacers (entries 7 and 9), the conversions
1
5
3). The work-up was performed as the same procedure
as that mentioned above.

8542
M. Okudomi et al. / Tetrahedron Letters 48 (2007) 8540–8543
a
Table 1. Enantioselective hydrolysis of MPEG5000-supported carbonates (± )-1 and 2
b
c
d
e
Entry
Substrate
Organic solvent
Temp (ꢁC)
ee of carbonate (%)
ee of alcohol (%)
Conv.
E value
1
2
3
4
5
6
7
8
9
1
Hexane
Toluene
30
30
30
20
10
0
0
0
0
88
15
18
50
41
18
76
31
42
65
78
78
88
93
96
97
99
96
0.58
0.17
0.19
0.36
0.31
0.16
0.44
0.24
0.30
13
9
10
26
39
1
1
i-Pr
2
O
1
Hexane
Hexane
Hexane
Hexane
Hexane
Hexane
1
1
58
2a
2b
2c
151
270
74
a
The reaction was performed using 5 mM of the substrates with PPL in a mixed medium (organic solvent/0.1 M phosphate buffer (pH 6.5) = 9:1) for
4 h.
Determined by GLC analysis after the hydrolysis of the unreacted substrates.
Determined by GLC analysis.
Calculated by ee(carbonate)/[ee(carbonate) + ee(3)].
Calculated by ln[(1 ꢀ conv.)(1 ꢀ ee(carbonate))]/ln[(1 ꢀ conv.)(1 + ee(carbonate))].
2
b
c
d
e
(
1) Separation
(2) Precipitation
1
2
3
) dillution
Hexane layer
with CH Cl2
Aqueous layer
Ph
2
Solid
O
) dehydration
OH
O
over Na SO4
2
Ph
) evaporation
Me
Ph
O
O
Me
2
CH Cl2
O
O
Me
+
Et O
+
2
OH
OH
Scheme 2.
OH
Me
OBn
O
(R
)-10
3%, 91% ee
PPL
hexane/buffer = 9/1
O
O
O
Me
OBn
0
°C, 24 h
OH
NaOH
MeOH-H O
(
±)-9
Conv. = 0.07
E
value = 23
2
Me
OBn
(
S
)-10
7
2%, 7% ee
Scheme 3.
In conclusion, we have demonstrated the enzyme-
mediated kinetic resolution of soluble polymer
moiety and the carbonate linker. We anticipate that
the use of a soluble polymer as the matrix of the
substrates will provide an operationally simple and
eco-friendly protocol.
(
MPEG5000)-supported carbonates to afford optically
active secondary alcohols (3 and 10). In our method,
the separation and isolation of the reaction products
were achieved by a simple precipitation technique
without the time- and solvent-consuming column chro-
matography. Furthermore, we have disclosed that the
reactivity and enantioselectivity can be controlled using
a suitable hydrophobic spacer between the MPEG
Acknowledgment
We thank Material Science Research Center (Meisei
University) for NMR analysis.

M. Okudomi et al. / Tetrahedron Letters 48 (2007) 8540–8543
8543
References and notes
8. (a) Veronese, F. M.; Morpurgo, M. Il Farmaco 1999, 54,
4
7
97–516; (b) Greenwald, R. B. J. Controlled Release 2001,
4, 159–171; (c) Vincent, M. J.; Duncan, R. Trends
1
. (a) Bornscheuer, U. T.; Kazlauskas, R. J. Hydrolases in
Organic Synthesis; Wiley-VCH: Weinheim, 1999; (b)
Bommarius, A. S.; Riebel, B. R. Biocatalysis; Wiley-
VCH: Weinheim, 2004; (c) Faber, K. Biotransformations in
Organic Chemistry: A Textbook, 5th ed.; Springer: Berlin–
Heidelberg–New York, 2004; (d) Ghanem, A.; Aboul-
Enein, H. Y. Chirality 2004, 17, 1–15; (e) Garcia-Uradales,
E.; Alfonso, I.; Gotor, V. Chem. Rev. 2005, 105, 313–354;
Biotechnol. 2006, 24, 39–47.
. L o´ pez-Pelegr ´ı n, J. A.; Wentworth, P., Jr.; Sieber, F.; Metz,
W. A.; Janda, K. D. J. Org. Chem. 2000, 65, 8527–8531.
0. (a) Shimojo, M.; Matsumoto, K.; Nogawa, M.; Nemoto,
Y.; Ohta, H. Tetrahedron Lett. 2004, 45, 6769–6773; (b)
Nogawa, M.; Shimojo, M.; Matsumoto, K.; Okudomi,
M.; Nemoto, Y.; Ohta, H. Tetrahedron 2006, 62, 7300–
9
1
7
306.
(
f) Bornscheuer, U. T. Trends and Challenge in Enzyme
1
1. The synthetic method of the imidazolide (± )-4 has already
Technology; Springer: Berlin/Heidelberg, 2005; (g)
Fogassy, E.; N o´ gr a´ di, M.; Kozma, D.; Egri, G.; P a´ lovics,
E.; Kiss, V. Org. Biomol. Chem. 2006, 4, 3011–3030; (h)
Gadler, P.; Faber, K. Trends Biotechnol. 2007, 25, 83–88;
1
0
1
been reported. Compound 1: H NMR (500 MHz,
CDCl ) d 1.59 (d, J = 6.5 Hz, 3H), 3.38 (s, 3H, CH O–
3
3
PEG), 3.50 (t, J = 5.0 Hz, 2H), 3.55 (t, J = 4.5 Hz, 2H),
.57–3.68 (m, PEG–methylenes), 3.70 (t, J = 4.5 Hz, 2H,
PEG), 3.78 (t, J = 5.0 Hz, 2H), 4.23 (td, J = 4.5 Hz,
J = 12.0 Hz, 1H), 4.28 (td, J = 5.0 Hz, J = 12.0 Hz, 1H),
3
(
i) Ghanem, A. Tetrahedron 2007, 63, 1721–1754.
2
3
. For representative techniques, see: (a) Terao, Y.; Tsuji, K.;
Murata, M.; Achiwa, K.; Nishio, T.; Watanabe, N.; Seto,
K. Chem. Pharm. Bull. 1989, 37, 1653–1655; (b) Yamano,
T.; Kikumoto, F.; Yamamoto, S.; Miwa, K.; Kawada, M.;
Ito, T.; Ikemoto, T.; Tomimatsu, K.; Mizuno, Y. Chem.
Lett. 2000, 448–449; (c) Beier, P.; O’Hagan, D. Chem.
Commun. 2002, 1680–1681; (d) Luo, Z.; Swaleh, S. M.;
Theil, F.; Curran, D. P. Org. Lett. 2002, 4, 2585–2587.
. For enzymatic transesterification of esters with PEG as a
nucleophile, see: (a) Wallace, J. S.; Reda, K. B.; Williams,
M. E.; Morrow, C. J. J. Org. Chem. 1990, 55, 3544–3546;
1
2
1
2
1
3
5
(
7
2
.72 (q, J = 6.5 Hz, 1H), 7.23–7.39 (m, 5H); C NMR
125 MHz, CDCl ) d 22.2, 58.8, 66.7, 68.7, 70.37, 70.40,
0.6, 71.7, 76.3, 125.9, 127.9, 128.3, 140.8, 154.3; IR (KBr)
3
ꢀ
1
887, 1744, 1636, 1466, 1342, 1281, 1113, 964, 843 cm
.
1
2. The yields of the MPEG-supported compounds were
based on the weights of the starting materials. The purities
1
were determined by a H NMR analysis, and the terminal
methyl group and/or the PEG–methylenes were used as
the reference. The purities of the substrates 1, 2a, 2b, and
2
c were ca. 91%, 94%, 89%, and 93%, respectively. The
(
2
b) Whalen, L. J.; Morrow, C. J. Tetrahedron: Asymmetry
000, 11, 1279–1288.
detail of the synthetic procedure of the substrates will be
reported separately.
4
5
6
. For enzyme-mediated kinetic resolution of carboxylic
acids anchored on polystyrene based resin, see: Nanda,
S.; Rao, B.; Yadav, J. S. Tetrahedron Lett. 1999, 40, 5905–
1
3. Chen, C. S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am.
Chem. Soc. 1982, 104, 7294–7299.
4. The ee of 3 was determined by GLC analysis, and a
similar analysis of 3 derived from 1 was also performed.
GLC conditions: column, CP-Cyclodextrin-B-236-M19
1
5
908.
. For lipase-catalyzed kinetic resolution with poly(acryl-
amide)ethylene glycol (PEGA)-based resin as a nucleo-
phile, see: Ulijn, R. V.; Bisek, N.; Flitsch, S. L. Org.
Biomol. Chem. 2003, 1, 621–622.
(
Chrompack), 0.25 mm · 50 m; injection, 160 ꢁC; detec-
tion, 160 ꢁC; oven, 140 ꢁC; carrier gas, He; head pressure,
2
2
.4 kg/cm ; retention time, 8.9 (R) and 9.2 (S) min.
. (a) Gravert, D. J.; Janda, K. D. Chem. Rev. 1997, 97, 489–
1
5. For the reaction of a methyl carbonate of (± )-3 under the
same conditions, the reaction scarcely proceeded. On the
other hand, an acetate of (± )-3 was slowly hydrolyzed
5
1
09; (b) Wentworth, P.; Janda, K. D. Chem. Commun.
999, 1917–1924; (c) Dickerson, T. J.; Reed, N. N.; Janda,
K. D. Chem. Rev. 2002, 102, 3325–3344; (d) Bergbreiter,
D. E. Chem. Rev. 2002, 102, 3345–3384.
(
conv. = 0.11) without any enantioselectivity. The hydro-
lysis of a methyl carbonate of (± )-10 was also proceeded
with very low enantioselectivity (E value = 1.4).
7
. (a) Benaglia, M.; Celentano, G.; Cozzi, F. Adv. Synth.
Catal. 2001, 343, 171–173; (b) Pozzi, G.; Cavazzini, M.;
Quici, S.; Benaglia, M.; Dell’Anna, G. Org. Lett. 2004, 6,
1
0
1
1
6. The reaction was also performed using 3.6 g of the
substrate (± )-2b. In this case, we finally obtained
4
41–443; (c) Oikawa, M.; Tanaka, T.; Kusumoto, S.;
1
D
8
(
R)-3 (98% ee, ½aꢁ +32.8 (c 0.96, MeOH)) in 12%
Sasaki, M. Tetrahedron Lett. 2004, 45, 787–790; (d) Reed,
N. N.; Dickerson, T. J.; Boldt, G. E.; Janda, K. D. J. Org.
Chem. 2005, 70, 1728–1731; (e) Li, J.-H.; Hu, X.-C.;
Liang, Y.; Xie, Y.-X. Tetrahedron 2006, 62, 31–38; (f)
Hong, S. H.; Grubbs, R. H. Org. Lett. 2007, 9, 1955–1957.
2
D
5
and (S)-3 (58% ee, ½aꢁ ꢀ21.0 (c 1.06, MeOH)) in 46%
1
7
isolated yields (conv. = 0.37, E value = 179).
7. Laumen, K.; Schneider, M. P. J. Chem. Soc., Chem.
Commun. 1988, 598–600.