Highly enantioselective separation using a supported liquid membrane encapsulating surfactant-enzyme complex

Article abstract of DOI:10.1021/ja049378d

We developed a highly enantioselective separation system for the optically active compounds, (S)-ibuprofen and L-phenylalanine, from their racemic mixtures by employing a supported liquid membrane (SLM) encapsulating a surfactant-lipase complex (or a surfactant-α-chymotrypsin complex). In the present system, enzymes encapsulated in the liquid-membrane phase effectively drove the enantioselective transport of optically active compounds through the SLM. This novel SLM allowed high enantioselectivity (ee over 91%) in the optical resolution of racemic ibuprofen and phenylalanine. Copyright

Full text of DOI:10.1021/ja049378d

Published on Web 06/25/2004
Highly Enantioselective Separation Using a Supported Liquid Membrane
Encapsulating Surfactant-Enzyme Complex
Eijiro Miyako, Tatsuo Maruyama, Noriho Kamiya, and Masahiro Goto*
Department of Applied Chemistry, Graduate School of Engineering, Kyushu UniVersity, PRESTO, JST,
Fukuoka 812-8581, Japan
Received February 4, 2004; E-mail: mgototcm@mbox.nc.kyushu-u.ac.jp
There is an increasing demand for optically pure enantiomers in
the chemical industry.1 Organic acids and amino acids represent a
large portion of this market, but for these useful organic compounds,
only one enantiomer is known to be normally biologically active.
Many researchers have attempted the separation of optically active
compounds.2 Some enzymes, such as lipase and protease, catalyze
highly enantioselective hydrolysis or esterification of drugs and
amino acids. By employing the enantioselectivity of these enzymes,
optical resolutions of a number of racemic mixtures have been
reported.3 These studies demonstrated the selective production of
optically pure compounds from racemic mixtures; however, this
enzymatic technique requires further processing (crystallization,
solvent extraction, etc.) to separate the target enantio-derivative from
another enantiomer.
Supported liquid membranes (SLMs), in which the organic liquid
is entrapped in a porous membrane, have been widely studied as a
selective separation technique.4 Application of this technique has,
however, been limited mainly to the separation of metal ions and
amines to utilize the crown ethers. Rethwisch et al. and we
previously demonstrated that lipase-catalyzed reactions (esterifi-
cation and hydrolysis) drove highly selective separation of organic
acids through the bulk liquid membrane and SLM, in which the
selectivity of the organic acids separation was based on the substrate
specificity of the lipases.5 In the previous system, the transport
efficiency for a targeted organic acid and the enantioselectivity were
not satisfactory; we presumed that the low transport efficiency and
the low enantioselectivity were due to the low esterification activity
of native lipase deposited in the aqueous phase.
To date, several research groups, along with ours, have demon-
strated that a surfactant-enzyme complex, which was soluble in
organic solvents, effectively catalyzed esterification reaction in
organic media.6 In the present study, we developed a novel SLM
encapsulating the surfactant-enzyme complex in the liquid mem-
brane phase and succeeded in an efficient and highly enantiose-
lective separation for the optically active compounds (S)-ibuprofen
and L-phenylalanine from their racemic mixtures.
Figure 1 gives a schematic diagram of the enantioselective
separation system for the racemic mixtures through the SLM
encapsulating the surfactant-enzyme complex. The surfactant-
enzyme complex can be solubilized in the thin organic membrane
and effectively catalyzes the esterification reaction in the thin film.
The surfactant-enzyme complex is good at catalyzing enantiose-
lective esterification in the liquid membrane phase, but another
enzyme is used as an ester hydrolysis catalyst in the receiving phase.
Therefore, the (S)-isomer is selectively esterified by the surfactant-
enzyme complex at interface 1 in the SLM phase, and the resulting
ethyl ester of (S)-isomer dissolves into the organic phase of the
SLM and diffuses across the SLM. At interface 2 in the receiving
phase, another enzyme catalyzes the ester hydrolysis to produce
the initial (S)-isomer and ethanol, which are water-soluble. Thus,
Figure 1. Concept of enantioselective separation of racemic mixtures
through the SLM encapsulating the surfactant-enzyme complex.
the (S)-isomer is selectively transported to the receiving phase
through the SLM, based on the enantioselectivity of the enzymes.
The SLM encapsulating the surfactant-enzyme complex7 was
prepared by immersing a hydrophobic poly(propylene) film (Cel-
gard 2500; the thickness of the film was 25 µm, and the maximum
pore size was 0.2 × 0.05 µm) into isooctane containing the
surfactant-enzyme complex. Enantioselective transport experiments
through the SLM were performed at 37 °C using a pair of glass
cells (each cell had a volume of 55 mL and a 5 cm2 cross-section).
The SLM encapsulating the surfactant-enzyme complex separated
the two aqueous phases. The feed phase consisted of McIlvaine
buffer (pH 6.3) containing 10 mM racemic substrate and ethanol.
The receiving phase consisted of McIlvaine buffer (pH 6.3)
containing hydrolysis biocatalyst. The concentrations of (S)- and
(R)-isomer of substrates in the feed and receiving phase were
determined by HPLC analysis.
Figure 2 depicts the enantioselective transport of (S)-ibuprofen
through the SLM encapsulating surfactant-lipase complex. The
surfactant-lipase CRL (lipase from Candida rugosa) complex was
encapsulated in the SLM, and native lipase PPL (lipase from porcine
pancreas) was dissolved in the receiving phase (see Figure 2 caption
for more detail). The (S)-ibuprofen concentration in the receiving
phase increased with time, with that in the feed phase correspond-
ingly decreasing. In contrast, the (R)-ibuprofen concentration in
the receiving phase did not notably increase, while that in the feed
phase did not decrease. A control experiment performed without
the lipases resulted in no ibuprofen transport through the SLM,
because ibuprofen was insoluble in isooctane, indicating that the
lipase-catalyzed reactions drove the transport of (S)-ibuprofen
through the SLM, as shown in Figure 1. A high enantiomeric excess
(ee ) 91 (%)) value for (S)-ibuprofen was obtained at the end of
the operation (48 h). These results indicate that the enantioselectivity
of the lipases induced the difference between the transport behavior
9
8622
J. AM. CHEM. SOC. 2004, 126, 8622-8623
10.1021/ja049378d CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Figure 2. Lipase-facilitated transport of (S)-ibuprofen through the SLM.
(b) (S)-Ibuprofen in the receiving phase, (O) (S)-ibuprofen in the feed phase,
(9) (R)-ibuprofen in the receiving phase, (0) (R)-ibuprofen in the feed phase.
The feed phase consisted of McIlvaine buffer (pH 6.3) containing 10 mM
racemic ibuprofen and 50 vol % ethanol. The SLM encapsulated 5 mg/mL
CRL complex. The receiving phase consisted of McIlvaine buffer (pH 6.3)
containing 8 mg/mL native PPL. Dioleyl-L-glutamate ribitol was used as
the surfactant for the surfactant-lipase complex.
Figure 3. R-Chymotrypsin-facilitated transport of L-phenylalanine through
the SLM. (b) L-Phenylalanine in the receiving phase, (O) L-phenylalanine
in the feed phase, (9) D-phenylalanine in the receiving phase, (0)
D-phenylalanine in the feed phase. The feed phase consisted of McIlvaine
buffer (pH 6.3) containing 10 mM racemic phenylalanine and 40 vol %
ethanol. The SLM encapsulated 5 mg/mL R-chymotrypsin complex. The
receiving phase consisted of McIlvaine buffer (pH 6.3) containing 0.5 mg/
mL R-chymotrypsin complex. The R-chymotrypsin complex was prepared
with the same surfactant employed for the surfactant-lipase formulation.
of (S)- and (R)-ibuprofen through the SLM. Indeed, the lipase CRL
has been reported to be a useful biocatalyst for the enantioselective
esterification reaction of (S)-ibuprofen.8 The effect of different kinds
of lipase on the enantioselective transport of (S)-ibuprofen was
examined. In these results, the maximum ee (91%) was obtained
when the surfactant-CRL complex was used in the SLM. The
maximum permeate flux of (S)-ibuprofen (0.58 [mol/(m2‚h)]) was
obtained when using 5 mg/mL surfactant-CRL complex in the
SLM and 8 mg/mL PPL in the receiving phase. In our previous
lipase-facilitated SLM, a large quantity of lipases (total 2750 mg
(CRL ) 1650 mg, PPL ) 1100 mg) was required to transport
organic acids.5c,d In the present study, the net amount of lipase CRL
was just 1 mg, which is 1/1650 that of our previous system, and
the amount of lipase PPL was 440 mg, which is 2/5 that of our
previous system. The high esterification activity of the surfactant-
CRL complex in the SLM would contribute to drastically decreasing
the amounts of lipases.
Figure 3 shows the selective transport of L-phenylalanine through
the SLM encapsulating the surfactant-R-chymotrypsin complex.
The R-chymotrypsin complex was entrapped in the SLM phase and
also dispersed in the receiving phase (see Figure 3 caption for more
detail). The L-phenylalanine concentration in the receiving phase
increased with time. On the other hand, D-phenylalanine was not
transported through the SLM. It is noteworthy that the R-chymo-
trypsin-facilitated SLM system achieved ee >99% for L-phenyl-
alanine by the end of the operation (48 h) and a maximum permeate
flux of L-phenylalanine (0.18 [mol/(m2‚h)]). Phenylalanine was not
transported through the SLM at all in a control experiment without
the surfactant-R-chymotrypsin complex. R-Chymotrypsin has also
been studied as an enantioselective biocatalyst for various amino
acids,9 suggesting that the R-chymotrypsin-catalyzed reactions drove
the enantioselective transport of L-phenylalanine based on the
enantioselectivity of R-chymotrypsin. The permeate flux of L-
phenylalanine was less than that of (S)-ibuprofen; differences in
this transport behavior could be explained by the difference in the
solubilities of each esterified substrate in the SLM phase10 and
differences between enzymatic activities in the organic solvent and
at the interfaces.11
In conclusion, the SLM encapsulating the surfactant-enzyme
complex enabled highly enantioselective separation of racemic
ibuprofen and phenylalanine. It can be envisioned that the arrange-
ment of appropriate enzymes in the SLM system will allow
enantioselective separations of various useful organic compounds.
Acknowledgment. This work was partly supported by a Grant-
in-Aid for Scientific Research (No. 13129205) from the Ministry
of Education, Science Sports and Culture of Japan.
References
(1) Crosby, J. Tetrahedron 1991, 47, 4789.
(2) (a) Lee, S. B.; Mitchell, D. T.; Trofin, L.; Nevanen, T. K.; So¨derlund, H.;
Martin, C. R. Science 2002, 296, 2198. (b) Chen, C.; Wu, S.; Girdaukas,
G.; Sih, C. J. J. Am. Chem. Soc. 1987, 109, 2812.
(3) (a) Kirchner, G.; Scollar, M. P.; Klibanov, A. M. J. Am. Chem. Soc. 1985,
107, 7072. (b) Lalonde, J. J.; Govardhan, C.; Khalaf, N.; Martinez, A.
G.; Visuri, K.; Margolin, A. L. J. Am. Chem. Soc. 1995, 117, 6845.
(4) Reviews: (a) Jo¨nsson, J. Å.; Mathiasson, L. Trends Anal. Chem. 1999,
18, 318. (b) Jo¨nsson, J. Å.; Mathiasson, L. J. Chromatogr., A. 2000, 902,
205. (c) de Gyves, J. de Miguel, E. R. Ind. Eng. Chem. Res. 1999, 38,
2182.
(5) (a) Rethwisch, D. G.; Subramanian, A.; Yi, G.; Dordick, J. S. J. Am. Chem.
Soc. 1990, 112, 1649. (b) Rethwisch, D. G.; Parida, S.; Yi, G.; Dordick,
J. S. J. Membr. Sci. 1994, 95, 83. (c) Miyako, E.; Maruyama, T.; Kamiya,
N.; Goto, M. J. Biosci. Bioeng. 2003, 96, 370. (d) Miyako, E.; Maruyama,
T.; Kamiya, N.; Goto, M. Chem. Commun. 2003, 2926.
(6) (a) Okahata, Y.; Ijiro, K. J. Chem. Soc., Chem. Commun. 1988, 1392. (b)
Noda, S.; Kamiya, N.; Goto, M. Biotechnol. Lett. 1997, 19, 307. (c)
Kamiya, N.; Goto, M. Biotechnol. Prog. 1997, 13, 488. (d) Paradkar, V.
M.; Dordick J. S. J. Am. Chem. Soc. 1994, 116, 5009.
(7) (a) Goto, M.; Matsumoto, M., Kondo, K.; Nakashio, F. J. Chem. Eng.
Jpn. 1995, 11, 893. (b) Goto, M.; Noda, S.; Kamiya, N.; Nakashio, F.
Biotechnol. Lett. 1996, 18, 839. (c) Goto, M.; Sumura, H.; Abe, K.;
Nakashio, F. Biotechnol. Tech. 1995, 9, 101.
(8) Klibanov, A. M. Acc. Chem. Res. 1990, 23, 114.
(9) (a) Oshima, T.; Inoue, K.; Furusaki, S.; Goto, M. J. Membr. Sci. 2003,
217, 87. (b) Nakamoto, Y.; Karube, I.; Kobayashi, I.; Nishida, M.; Suzuki,
S. Arch. Biochem. Biophys. 1979, 193, 117.
(10) (a) Branco, L. C.; Crespo, J. G.; Afonso, C. A. M. Angew. Chem., Int.
Ed. 2002, 41, 2771. (b) Branco, L. C.; Crespo, J. G.; Afonso, C. A. M.
Chem. Eur. J. 2002, 17, 3865.
(11) (a) Narayan, V. S.; Klibanov, A. M. Biotechnol. Bioeng. 1993, 41, 390.
(b) Filho, M. V.; Stillger, T.; Mu¨ller, M.; Liese, A.; Wandrey, C. Angew.
Chem., Int. Ed. 2003, 42, 2993. (c) Maruyama, T.; Nakajima, M.;
Ichikawa, S.; Nabetani, H.; Furusaki, S.; Seki, M. J. Am. Oil Chem. Soc.
2000, 77, 1121. (d) Maruyama, T.; Nakajima, M.; Seki, M. J. Biosci.
Bioeng. 2001, 92, 242.
JA049378D
9
J. AM. CHEM. SOC. VOL. 126, NO. 28, 2004 8623