Lipase dissolution and stabilization in ether-functionalized ionic liquids

Article abstract of DOI:10.1039/b905388c

Ionic liquids (ILs) with various structures and properties have been extensively investigated in many biocatalytic reactions and processes. However, although hydrophobic ILs tend to stabilize the insoluble (suspended) enzymes, they usually have low hydrogen-bonding basicity with solutes, which limits the solubility of many substrates (such as D-glucose, ascorbic acid, and cellulose). In contrast, hydrophilic ILs (such as those based on chloride, acetate and dicyanamide) are able to dissolve many of these substances that are not quite soluble in common organic solvents. Unfortunately, enzymes are not always active in these hydrophilic media due to strong interactions (such as H-bonding) between proteins and ILs. To resolve this dilemma, we recently synthesized new acetate-based ILs carrying a long alkyloxyalkyl chain in their cations, and found that these ether-functionalized solvents are lipase-compatible and can dissolve considerable amounts of D-glucose and cellulose (Green Chem., 2008, 10, 696). In this study, we further observed that these ILs could dissolve high concentrations of lipase B from Candida antarctica (CALB) (> 5 mg/mL at 50 °C), as well as other substrates including amino acids and betulinic acid. Therefore, these novel media offer new opportunities for carrying out homogeneous enzymatic reactions, which is practically important for large substrate molecules. In this article, we further confirmed the lipase compatibility of these ILs through the transesterification between ethyl butyrate and 1-butanol. The second derivative infrared spectra of CALB suggest the conservation of secondary structures of proteins in these ILs. We further investigated these ether-functionalized ILs in two important biocatalytic reactions: enzymatic synthesis of methyl-phthalate of betulinic acid, and CALB-catalyzed synthesis of D-glucose fatty acid esters. These substrates are not very soluble in conventional organic solvents, but very soluble in ILs, which improved the catalytic efficiency of these reactions. Moderate to high conversions were achieved in both reactions.

Full text of DOI:10.1039/b905388c

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/greenchem | Green Chemistry
Lipase dissolution and stabilization in ether-functionalized ionic liquids†
Hua Zhao,* Cecil L. Jones and Janet V. Cowins
Received 17th March 2009, Accepted 12th May 2009
First published as an Advance Article on the web 4th June 2009
DOI: 10.1039/b905388c
Ionic liquids (ILs) with various structures and properties have been extensively investigated in
many biocatalytic reactions and processes. However, although hydrophobic ILs tend to stabilize
the insoluble (suspended) enzymes, they usually have low hydrogen-bonding basicity with solutes,
which limits the solubility of many substrates (such as D-glucose, ascorbic acid, and cellulose). In
contrast, hydrophilic ILs (such as those based on chloride, acetate and dicyanamide) are able to
dissolve many of these substances that are not quite soluble in common organic solvents.
Unfortunately, enzymes are not always active in these hydrophilic media due to strong interactions
(such as H-bonding) between proteins and ILs. To resolve this dilemma, we recently synthesized
new acetate-based ILs carrying a long alkyloxyalkyl chain in their cations, and found that these
ether-functionalized solvents are lipase-compatible and can dissolve considerable amounts of
D-glucose and cellulose (Green Chem., 2008, 10, 696). In this study, we further observed that these
ILs could dissolve high concentrations of lipase B from Candida antarctica (CALB) (> 5 mg/mL
at 50 ◦C), as well as other substrates including amino acids and betulinic acid. Therefore, these
novel media offer new opportunities for carrying out homogeneous enzymatic reactions, which is
practically important for large substrate molecules. In this article, we further confirmed the lipase
compatibility of these ILs through the transesterification between ethyl butyrate and 1-butanol.
The second derivative infrared spectra of CALB suggest the conservation of secondary structures
of proteins in these ILs. We further investigated these ether-functionalized ILs in two important
biocatalytic reactions: enzymatic synthesis of methyl-phthalate of betulinic acid, and
CALB-catalyzed synthesis of D-glucose fatty acid esters. These substrates are not very soluble in
conventional organic solvents, but very soluble in ILs, which improved the catalytic efficiency of
these reactions. Moderate to high conversions were achieved in both reactions.
However, these substrates were still suspended in solutions as
Introduction
heterogeneous systems. Therefore, an ideal solution would be a
Biocatalysis in organic solvents undoubtedly offers numerous
solvent system that can dissolve both the enzyme and substrates,
advantages and intriguing opportunities over aqueous media.1,2
while the enzyme maintains a high activity in the solvent.
However, even with the modifications of enzymes and solvent
As alternatives to conventional organic solvents, ionic liquids
systems, enzymes are typically less active in organic media
(ILs) belong to a young family of non-aqueous solvents. In addi-
than in water.3,4 In particular, when the substrate molecules
tion to their low-volatility, ILs have tunable physical properties
are sparingly soluble in organic solvents, the biotransformation
through the judicious selection and combination of different
becomes extremely challenging since enzymes are also insoluble
cations and anions. Therefore, ILs have found a wide range
in low-water systems. To overcome this hurdle, a common
of applications in organic reactions,10 biocatalysis,11 chemical
strategy is to solubilize the enzyme in organic media by either
processes,12 and other areas.12,13 Pure hydrophobic ILs (typically
chemical modifications or the formation of reversed micelle of
-
consisting of PF6 and Tf2N-) do not dissolve appreciable
enzymes, but both methods have some serious disadvantages.5
amounts of enzymes; on the other hand, hydrophilic ILs (such
The Dordick group6,7 reported a method in dissolving enzymes
-
-
as those based on NO3 , lactate, EtSO4 , and CH3COO-) and
their aqueous solutions may dissolve some enzymes,‡ however,
most of them tend to strongly interact with proteins (such as
via hydrogen bonding), resulting in enzyme inactivation.14–20 At
present, only a few enzyme-dissolving ILs are known enzyme-
compatible. For example, choline dihydrogen phosphate (m.p.
119◦C) containing 20% (wt) water was observed capable of
dissolving and stabilizing cytochrome c (cyt c).21,22 Another IL,
in organic solvents with low concentrations of surfactants. Based
on this technique, the transesterifications of solid b-cyclodextrin,
amylose, hydroxy ethyl cellulose, and cellulose were conducted
in isooctane or pyridine using soluble subtilisin Carslberg.5,8,9
Chemistry Program, Savannah State University, Savannah, GA 31404,
USA. E-mail: huazhao98@gmail.com, zhaoh@savannahstate.edu
† Electronic supplementary information (ESI) available: NMR charac-
terization of three new ionic liquids; NMR characterization of glucose
laurate; infrared spectra of free CALB in ILs; fluorescence spectra of
free CALB in ILs; and 13C NMR spectra of glucose in ILs. See DOI:
10.1039/b905388c
-
‡ There are also some exceptions. For example, BF4 based ILs are
hydrophilic but do not dissolve the enzyme.14
1128 | Green Chem., 2009, 11, 1128–1138
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
triethylmethylammonium methyl sulfate ([Et3MeN][MeSO4])§
was reported able to dissolve > 1.2 mg/mL Candida antarctica
lipase B (CALB) and retain its catalytic capability.14,23 Thus,
to enable the homogeneous biocatalysis in ILs, a common
route is to modify the enzyme molecules with crown ether
or poly(ethylene glycol) (PEG), both of which are soluble in
ILs.14,24,25 However, the preparation of PEG-modified enzymes
could be cumbersome, and the enzyme catalytic properties may
vary in different immobilization batches.
Scheme 2 Structure of AMMOENGTM 110 ([Amm110]Cl).
suggested that the immobilized CALB retained high synthetic
activities in this IL.15,20 The same IL was also used to form
aqueous biphasic systems for the biocompatible purification of
active enzymes.41 Other similar ammonium ILs in the Ammoeng
family carrying alkyloxyalkyl groups have also been used in
stabilizing lipases for the glycerolysis of fats and oils.42,43
Very recently, we synthesized a new type of acetate-based ILs
(Scheme 1) through crafting oligoethylene (or oligopropylene)
glycol units into the cations; these ILs exhibit a number of
favorable properties including high lipase compatibility, low
viscosity, and ‘super’ solvents for a variety of substrates such
as cellulose and sugars.15 In addition, we noticed that these
ILs could dissolve a considerable amount of enzymes (data
in a later section). Therefore, these ether-functionalized ILs
have great potentials for homogeneous enzymatic reactions in
ILs. Historically, it has been known that poly(ethylene oxide)s
(PEOs) can be incorporated into cationic or anionic units to
produce the liquid state of ion conductive polymers.13,26 In
particular, the liquid state of ionic polymers has been synthesized
by grafting alkyloxy substituents (ether or alcohol groups)
onto the imidazolium ring.27 Following these studies, ether-
functionalized ILs based on imidazolium28–37 and pyridinium38
have been synthesized and their physical properties were deter-
mined. The inclusion of alkyloxy or alkyloxyalkyl groups lowers
the melting-points and viscosities of the organic salts, resulting in
room-temperature ILs in most cases. The low viscosity resulting
from the incorporation of alkoxy chains was justified by the
molecular dynamic simulations as the less effective assembly
between more flexible alkoxy chains (vs. more rigid alkyl
chains).39 In addition, the molecular simulation also suggested
the reduction of intermolecular correlation (particularly tail-tail
segregation) and cation-anion specific interactions due to the
incorporation of ether chain, which is responsible for the faster
dynamics in ether-functionalized imidazolium ILs compared to
alkyl substituted ILs.40
Although our previous study has suggested the high enzyme-
R
compatibility of immobilized CALB (Novozymꢀ 435) in ether-
functionalized ILs,15 we have not fully understood the underly-
ing mechanisms via the free enzyme. It is the aim of this study
to examine the dissolution of lipase in these promising ILs, to
investigate the catalytic capability of free/immobilized lipase
in these media, and to understand how IL structures affecting
the enzyme stabilization. We also demonstrate the application
of these ether-functionalized ILs in two important reactions:
enzymatic synthesis of methyl-phthalate of betulinic acid, and
CALB-catalyzed synthesis of D-glucose fatty acid esters.
Experimental
General
The following chemicals were purchased from Sigma-Aldrich
(St. Louis, MO, USA): ethyl butyrate, butyl butyrate, 1-
butanol, 1-ethyl-3-methylimidazolium bromide ([EMIM]Br),
1-butyl-3-methylimidazolium bromide ([BMIM]Br), 1-methyl-
3-octylimidazolium bromide ([OMIM]Br), 1-ethyl-3-methyl-
imidazolium ethylsulfate ([EMIM][EtSO4]), 1,3-dimethylimid-
azolium methylsulfate ([MMIM][MeSO4]), 2-methylimidazole,
bis(trifluoromethane)sulfonimide lithium salt (Li[Tf2N]),
sodium dicyanamide (Na[dca]), sodium acetate, phosphorus
pentoxide (P2O5), L-ascorbic acid, vinyl laurate, D-glucose, 13C6-
labeled D-glucose, betulinic acid, dimethyl phthalate and glu-
cose HK assay. 1-Ethyl-3-methylimidazolium trifluorometha-
nesulfonate ([EMIM][OTf] or [EMIM][CF3SO3]) was obtained
from the Alfa Aesar Company (Ward Hill, MA, USA).
AMMOENGTM 110 (short as [Amm110]Cl, Scheme 2) was
obtained from Solvent Innovation GmbH◦(Nattermannallee,
Germany) as a colorless liquid (mp < -65 C, density = 1.03
g/cm3 at 20 ◦C, viscosity = 495 mPa s at 20 ◦C, pH = 3.83 and
conductivity = 0.090 mS/cm).
Scheme 1 Imidazolium and ammonium-based ILs consisting of
alkyloxyalkyl-substituted cation and acetate anion ((a) [Me(OEt)n-
Et-Im][OAc], (b) [Me(OEt)n-Me-Et-Im][OAc], and (c) [Me(OEt)n-
Et3N][OAc] respectively) (n = 2, 3, . . .).
The free form lipase B from Candida antarctica (CALB) was
also obtained from Sigma-Aldrich (product # 62288, recom-
binant from Aspergillus oryzae, powder, beige, ~9 units/mg).
A few groups have reported the uses of these ether-
functionalized ILs in the enzyme stabilization. ‘Ammoeng 110’ is
a commercial quaternary ammonium mixture containing oligo-
propylene glycol units (Scheme 2). Our previous investigations
R
Novozymꢀ 435 from Sigma-Aldrich is a thermal stable CALB
immobilized on acrylic resin (product # L4777, ≥10,000 U/g,
recombinant, expressed in Aspergillus oryzae).
IL preparations
§ This IL was reported as a liquid at room temperature,14 however, it
tends to form liquid/solid co-existing state at room temperature and
50 ◦C (see Experimental).
The
preparation
of
[BMIM][Tf2N],
[BMIM][dca],
[BMIM][HCOO], [EMIM][OAc] and [OMIM][OAc], as
This journal is
The Royal Society of Chemistry 2009
Green Chem., 2009, 11, 1128–1138 | 1129
©

View Article Online
well as the alkyloxyalkyl substituted ILs (Scheme 1) was
IRsolution 1.30 software following the Savitsky-Golay method
(9 data point window).
described in our recent paper.15 All ILs were dried in an oven
◦
R
at 100 C over 4 h before use. Acrosꢀ 3A molecular sieves
were added into ILs during storage. The characterization of
these ILs were reported previously15 except three new ILs (see
ESI† for NMR characterization of [Me(OEt)3-Et3N][HCOO],
[Me(OEt)3-Me-Et-Im][OAc] and [Me(OEt)3-Et-Im][dca]).
The preparation of triethylmethylammonium methylsulfate
([Et3MeN][MeSO4]) followed a literature method.14 The purified
product became semi-solid after cooling to room temperature
(and at 50 ◦C). 1H NMR (300 MHz, CDCl3, [ppm]): d = 1.37 (t,
9H, CH2CH3, J = 2.1 Hz), 3.06 (s, 3H, SO4CH3), 3.43 (m, 6H,
CH2CH3), 3.68 (s, 3H, NCH3).
Enzymatic synthesis of methyl-phthalate of betulinic acid
Betulinic acid (52 mg, 0.11 mmol) and dimethyl phthalate
(25.8 mg, 0.13 mmol) were fully dissolved in 1.0 mL [Me(OEt)3-
Et-Im][OAc] (3) after a gentle agitation. Novozym 435 (40 mg,
undried by P2O5) was added into the reaction mixture in a
capped glass-vial reactor. The reaction was kept at 50 ◦C. Upon
completion of reaction, the reaction mixture was diluted with
10 mL methanol and the immobilized enzyme was removed
through filtration. After evaporation of methanol under reduced
pressure, the crude product was dissolved in 200 mL ethyl
acetate, and washed with distilled water three times to remove
the IL and trace dimethyl phthalate (slightly soluble in water).
The product was collected after drying the ethyl acetate with
sodium sulfate, filtering off the salt, and evaporating the solvent
under reduced pressure. The isolated product weighed 60 mg
(yield 85%). IR n (KBr) cm-1: 3475, 2949, 2870, 1732, 1687, 1450,
1435, 1125, 1076. 1H NMR (JEOL ECX-300 MHz, DMSO-d6,
[ppm]): 0.61 (3H, s), 0.72 (3H, s), 0.83 (3H, s), 0.89 (3H, s), 1.61
(3H, s), 3.28 (3H, s), 3.78 (3H, s), 4.20 (3H, m), 4.55 (3H, m),
4.66 (3H, m), 7.65 (4H, m), 12.05 (1H, s). 13C NMR (DMSO-d6)
showed two new signals for the product at 129.2 ppm and
132.2 ppm, indicating the incorporation of two ester carbons
(-COOR) from phthalate.
Measuring CALB transesterification activity
Organic solvents were dried by anhydrous MgSO4 before use.
The free CALB and Novozym 435 were dried over P2O5 for at
least 2 days. A typical reaction procedure is as the following:
14 mL ethyl butyrate (0.1 M) and 46 mL 1-butanol (0.5 M) were
added into a glass reactor containing organic solvent or IL. The
final mixture volume was 1.0 mL. The reaction was started
by adding 5.0 mg free CALB, or 25 mg Novozym 435. The
reaction mixture was sealed and stirred at 50 ◦C in a water-bath.
The reaction mixture was periodically withdrawn (50 mL each
sample) and diluted with 100 mL methanol. After centrifugation,
the clear supernatant was injected into a LC-10AT Schimadzu
HPLC equipped with a SPD-10Avp UV-visible dual wavelength
detector and a Schimadzu RID10 refractive index detector. The
injection loop volume was 20 mL. The HPLC eluent consisted
of 65% (v/v) MeOH and 35% (v/v) aqueous acetate buffer
(0.05 M, pH 4.5). The flow rate was 1.0 mL min-1. The column
is a Schimadzu Premier C18 column (150 mm ¥ 4.6 mm, particle
size 5 m). The UV detection wavelength was 215 nm. The product
concentration was determined by comparing the sample’s peak
area with the standard curve of butyl butyrate. All experiments
were run at least in duplicate. The percent errors were less than
5%.
Lipase-catalyzed synthesis of D-glucose fatty acid ester
D-Glucose (0.0396 g, 0.22 mmol) and vinyl laurate (0.075 g,
0.33 mmol) were added into 1.0 mL IL in a capped glass-vial
reactor. After both substrates fully dissolved in the IL under
a gentle stirring at 50 ◦C, Novozym 435 (40 mg, dried over
P2O5) was carefully added into the mixture. The reaction was
kept at 50 ◦C and monitored over 30 h. Periodically, 30 mL
reaction mixture was withdrawn and mixed with 970 mL distilled
water. After centrifugation (or filtration) to settle (or remove)
the precipitated acid and ester, the glucose concentration in
supernatant (or filtrate) was determined by the glucose HK
assay.44 Upon completion of reaction, the reaction mixture was
diluted with 10 mL methanol and the immobilized enzyme
was removed through filtration. After evaporation of methanol
under reduced pressure, the crude product was precipitated from
IL by the addition of distilled water. The crude product was
collected into chloroform45 (or dichloromethane) and washed
several times with water to remove IL. The glucose fatty acid
ester was purified through washing with hexane to remove
the remaining fatty acid.46 After drying the chloroform, the
solvent was removed under vacuum. The structures of product
CALB stability in ILs
Novozym 435 (25 mg, dried over P2O5) was incubated in 1.0 mL
◦
IL under a gentle agitation at 50 C. After certain incubation
time, 14 mL ethyl butyrate (0.1 M) and 46 mL 1-butanol (0.5 M)
were added to the enzyme suspension in IL. The reaction was
maintained at 50 ◦C for 24 h, and the product concentration was
determined by the HPLC analysis (see above).
FT-IR spectra of CALB in ILs
Free CALB was dissolved in ILs (2 mg/mL) or water
(10 mg/mL) and incubated in water bath at 50 ◦C. The enzyme
solution was taken periodically and placed between two CaF2
windows. The FT-IR spectra were measured using a Shimadzu
IRPrestige-21 spectrophotometer through averaging 32 scans at
2 cm-1 resolution (Happ-Genzel apodization). The respective
pure IL or water between CaF2 windows was scanned as the
background before each measurement. The original spectra were
smoothed based on the 9-point Savitsky-Golay algorithm. The
second derivatives of spectra were calculated via the Shimadzu
1
(see ESI†) were confirmed by H and 13C NMR (13C6-labeled
D-glucose was used as reactant for the analysis of 13C NMR of
product).
Results and discussion
Selection of assay for measuring lipase’s synthetic activity
A number of transesterification reactions are routinely carried
out to evaluate the enzyme’s synthetic activity in non-aqueous
1130 | Green Chem., 2009, 11, 1128–1138
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
systems. It is very common to use vinyl ester as acyl donor
because the transesterification between vinyl ester and an
alcohol is irreversible and fast, although it is known that the
by-product acetaldehyde may lead to some degree of enzyme
inactivation.47,48 Therefore, the enzyme-catalyzed acyl transfer
reactions have been frequently examined in ILs.11,49 However,
when conducting the lipase-catalyzed transesterification be-
tween vinyl butyrate and benzyl alcohol in formate- and acetate-
based ILs (i.e. [Bu4N][HCOO] (2), [Me(OEt)3-Et-Im][OAc] (3),
and [Me(OEt)3-Et3N][OAc] (4)), we observed comparably fast
reaction rates even in the absence of CALB (see Table 1). The
reaction did not proceed in [BMIM][Tf2N] (1), [BMIM][dca]
(6), [EMIM][EtSO4] (9), and [MMIM][MeSO4] in the absence of
enzyme. This observation suggests that formate- and acetate-
type ILs can catalyze the acylation reaction since the vinyl
ester is an activated substrate and the reaction is irreversible.
For this reason, the transesterification reaction involving vinyl
ester is not suitable for evaluating the lipase’s synthetic activity
in formate- or acetate-based ILs. In fact, the acetate-catalyzed
transesterification (or esterification) was used in the industrial
production of PET (polyethylene terephthalate), where dimethyl
terephthalate (or tetrephthalic acid) reacts with an excess of
ethylene alcohol catalyzed by Cu, Co, or Zn acetate at 100–
◦
150 C and 10–70 bar.50 It is also known that ILs based on
acetate, formate and dicyanamide are Lewis bases with various
hydrogen-bonding basicity (proton acceptors), and may act as
base catalysts in reactions such as acetylation of alcohols.51,52
Alternatively, another transesterification reaction between
ethyl butyrate and 1-butanol was chosen because no reaction
was detected in formate- and acetate-based ILs (2–4) with-
out CALB. This reaction is relatively fast when catalyzed
by immobilized◦CALB, and ◦can reach equilibrium within a
few hours at 40 C20,53 or 100 C.54 However, the same reaction
catalyzed by free CALB is much slower, and it usually takes
◦
◦
24 h to reach completion at 40 C14,23,53 or 100 C.55 It is
known that the immobilized lipases usually induce higher
reaction rates than free enzyme particles in organic media.56,57
The same improvement effect was seen with other enzymes
such as proteases, thermolysin, liver alcohol dehydrogenase,
mushroom polyphenol oxidase and carboxyesterase.57,58 Thus,
we conducted this model reaction in t-butanol and a few ILs
catalyzed by both free and immobilized CALB (Table 2). Our
data confirm that the reaction rates were higher in all ILs when
using the immobilized CALB instead of free lipase. However,
most reactions in ILs with immobilized CALB still took 24 h
to complete, although most of the product (butyl butyrate)
was formed within the first 5 h. Based on the estimation that
Novozym 435 contains ~20% (wt) of CALB,59 the protein
Table 1 Initial rate (mmol/min mg) of transesterification between vinyl
butyrate and benzyl alcohol at 50 ◦C a,b
IL
Free CALB
Control (no enzyme)
1
2
3
4
[BMIM][Tf2N]
[Bu4N][HCOO]
[Me(OEt)3-Et-Im][OAc]
[Me(OEt)3-Et3N][OAc]
0.21
1.25
0.89
0.90
0
1.23
0.54
0.88
a Reaction conditions: 0.1 M vinyl butyrate (13 mL), 0.5 M benzyl alcohol
(103 mL), 2.0 mg free CALB or no enzyme (as control) in 1.0 mL IL
at 50 ◦C. b HPLC conditions: 65% (v/v) acetonitrile and 35% (v/v)
aqueous acetate buffer (0.05 M, pH 4.5); both vinyl butyrate and benzyl
butyrate were detected at 215 nm.
Table 2 CALB activity in ILs at 50 ◦C (percent yield of butyl butyrate) a
Immobilized
CALB
Free CALB
Solubility of free
Solvent
5 h
24 h
5 h
24 h
CALB (mg/mL) d
[Anion] (M) e
b
5
t-BuOH
[BMIM][Tf2N]
[BMIM][dca]
[EMIM][OAc]
[OMIM][OAc]
[EMIM][EtSO4]
[Bu4N][HCOO]
[BMIM][HCOO]
70
62
33
17
56
57
—
33
43
60
42
55
58
42
57
34
32
25
41
37
30
73
62
48
44
70
86
41
15
51
56
61
81
78
48
72
62
40
36
51
65
33
11
35
4
11
11
38
23
4
37
6
21
6
26
60
17
22
40
67
25
14
54
17
46
18
43
26
21
22
55
31
46
35
20
—
—
—
b
1
2.86
5.85
7.05
4.72
5.08
4.17
6.52
—
4.65
3.97
3.63
2.46
3.31
3.48
3.79
3.88
4.56
3.90
4.09
4.61
6
5
5
5
2
2
5
1
> 5
> 5
4
1
1
> 5
1
5
7
8
9
c
2
2a
10
13
3
[Amm110]Cl
[Me(OEt)2-Et-Im][OAc]
[Me(OEt)3-Et-Im][OAc]
[Me(OEt)3-Bu-Im][OAc]
[Me(OEt)7-Et-Im][OAc]
[Me(OEt)3-MeOEtOMe-Im][OAc]
[Me(OPr)3-Et-Im][OAc]
[Me(OEt)3-Me-Et-Im][OAc]
[Me(OEt)3-Et-Im][dca] f
[Me(OEt)2-Et3N][OAc]
[Me(OEt)3-Et3N][OAc]
[Me(OEt)3-Et3N][HCOO]
[EMIM][OTf]
14
15
16
17
18
19
11
4
14
13
8
5
38
10
17
12
3
4
> 5
4
12
20
b
—
a Reaction conditions: 0.1 M ethyl butyrate, 0.5 M 1-butanol, 5.0 mg free CALB or 25 mg Novozym in 1.0 mL solvent at 50 ◦C. b solubility is too low
to be observed. c The HPLC yield was not obtained due to difficulty in integration. d The solubility was determined by an incremental addition of
0.5 mg CALB into 1.0 mL IL. e The anion molar concentration was calculated from eqn 1; Ammoeng 110 is an IL mixture and its molecular weight
is unknown; f A low viscosity (50 mPa s) was obtained for this dca-based IL (using method in Ref. 15).
This journal is
The Royal Society of Chemistry 2009
Green Chem., 2009, 11, 1128–1138 | 1131
©

View Article Online
contents in 25 mg Novozym 435 and 5 mg free CALB are about
the same.
Et-Im][OAc] (3), [Me(OEt)3-Bu-Im][OAc] (14), [Me(OEt)7-Et-
Im][OAc] (15), [Me(OPr)3-Et-Im][OAc] (17), and [Me(OEt)3-
Me-Et-Im][OAc] (18). Free CALB is more active in [Me(OEt)3-
Et-Im][dca] (19) than the immobilized lipase. Fig 1 illustrates the
relationship between the immobilized-CALB activity and the
anion molar concentration for a homologous series of acetate
ILs (differing in numbers of -OEt- and/or -CH2- groups). The
general trend confirms that a lower anion concentration leads
to a higher lipase activity. Of course, as shown in Table 2, the
lipase activity bears no simple linear relationship with the anion
concentration as the cation and overall IL properties also play
critical roles.
Lipase dissolution and stabilization in ILs
The lipase solubility data in Table 2 suggest that the CALB is
not quite soluble in t-butanol (5) and hydrophobic ILs such as
[BMIM][Tf2N] (1).¶ The lipase has a low solubility (1 mg/mL
or less) in [Amm110]Cl (10), [EMIM][OTf] (20), [Me(OEt)7-
Et-Im][OAc] (15), [Me(OEt)3-MeOEtOMe-Im][OAc] (16), and
[Me(OEt)3-Me-Et-Im][OAc] (18), and a modest solubility
(2 mg/mL) in [EMIM][EtSO4] (9) and [Bu4N][HCOO] (2). On
the other hand, very high protein solubilities (> 4–5 mg/mL)
were observed in other dca-, HCOO- and OAc- based ILs
(especially in 3 and 4, the free CALB solubility could be as
high as 10 mg/mL in some reaction mixtures60). The general
trend is that higher molar concentrations of anions (such as
acetate and formate) lead to higher lipase solubility. Meanwhile,
the structure of cations and the overall IL properties also affect
the lipase solubility as suggested by Table 2. It is interesting
to mention the dissolved lipase in ILs can be precipitated out
strongly by the addition of methanol, ethanol or acetone, and
weakly by 2-propanol, but not by the addition of DMSO and
acetonitrile. This allows the possible use of IL mixtures with
organic solvents as media for homogeneous biocatalysis.
The anion molar concentration (M or mol/L) of a pure IL
can be estimated from its density (d) and molecular weight (Mw)
based on eqn 1 (assuming IL is monoanion),
Fig. 1 Effect of anion (acetate) molar concentration on the lipase
activity (plot from data of Novozym 435-catalyzed percent yield at 24 h
in Table 2; the anion concentrations were estimated from eqn 1; the
number near each data point represents the IL number in Table 2).
(1)
where m is the mass of IL (g), V is the volume of IL (mL), and
d is assumed to be 1.2 mg/mL for all ILs (based on density
data of ether-functionalized ILs29,34). Our previous study15
suggested that a higher anion (acetate) molar concentration
(i.e. lower Mw and a smaller cation of IL) is beneficial to
the dissolution of substrates, whereas a lower anion molar
concentration (i.e. higher Mw and a larger cation) is advan-
tageous for the enzyme stabilization. Therefore, the IL structure
can be tailored to exhibit a balanced H-bonding capability:
strong enough to dissolve substrates (at high temperatures if
necessary) and enzymes, but not too strong to disrupt the
protein structures and interact with enzyme’s active sites (at
moderate or low temperatures). This explanation has been
confirmed by the lipase activity data in Table 2. Lower
CALB activities were seen in enzyme-denaturing ILs such as
[BMIM][dca] (6), [EMIM][OAc] (7) and [Bu4N][HCOO] (2) due
to the strong hydrogen-bonding basicity of anions and high
anion concentrations,15–18,52 which is consistent with previous
findings of low enzyme activities in these ILs.15,17,20 With the
increase in cation size, we observed higher lipase activity (both
free and immobilized forms) in ILs such as [OMIM][OAc]
(8), [Amm110]Cl (10), [Me(OEt)2-Et3N][OAc] (11), [Me(OEt)3-
Et3N][OAc] (4), [Me(OEt)3-Et3N][HCOO] (12), [Me(OEt)3-
It is important to address two particular ILs (18 and 20) in
Table 2. 18 is different from 3 by having an additional methyl
group at the C-2 position of imidazolium ring. It is known
the C-2 hydrogen of imidazolium is acidic (pKa = 21–23),61–63
and replacing this hydrogen by a methyl group is expected to
reduce the H-bonding acidity. Based on Table 2, the activities
of immobilized CALB in 3 and 18 are comparable, however,
free CALB seems more active in 3 than in 18. The free CALB
solubility is reduced from > 5 mg/mL in 3 to 1 mg/mL in 18.
◦
In addition, the viscosity of 18 (239 mPa s at 20 C, based on
method in Ref. 15) is significantly higher than 3 (92 mPa s15). The
other IL, [EMIM][OTf] (20), seems lipase denaturing based on
the activity data (Table 2), although several studies achieved the
CALB-catalyzed synthesis of sugar esters in [EMIM][OTf] and
[BMIM][OTf],64–66 as well as CALB-catalyzed biodiesel synthesis
in [EMIM][OTf].67 In addition, [EMIM][OTf] (20) has a poor
dissolution power towards many substances (see Table 3 and
discussion in next section).
To address issues of the lipase stability and reusability,
immobilized CALB was incubated in 3 at 50 ◦C for 72 h,
and no loss in catalytic capability was observed (Fig 2). In
addition, ether-functionalized ILs (3, 4, 11, 12, 13 and 19) have
low viscosities (50–150 mPa s at 20 ◦C), and can dissolve a
variety of substrates (see Ref. 15 and Table 3). High CALB
activities (both free and immobilized) were also obtained in
[EMIM][EtSO4] (9), although a low activity was reported previ-
¶ When measuring Circular Dichroism (CD) spectra of enzyme in ILs,
the CALB concentration in two Tf2N- based ILs was 0.22 mg/mL.79
The a-chymotrypsin in [EMIM][Tf2N] was 0.03 mg/mL for fluorescence
spectroscopy, and 0.1–0.15 mg/mL for CD spectroscopy.77
ously for the immobilized CALB.14 The anion (EtSO4 ) has a low
-
1132 | Green Chem., 2009, 11, 1128–1138
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
Table 3 Solubility of different substances in ILs
IL
Cellulose wt% (110 ◦C)
D-glucose wt% (60 ◦C)
DOPA (20 ◦C)a
Betulinic acid (25 ◦C)
7
[EMIM][OAc]
[EMIM][EtSO4]
[Amm110]Cl
[Me(OEt)3-Et-Im][OAc]
[Me(OEt)3-Et3N][OAc]
[Me(OEt)3-Et3N][HCOO]
[Me(OEt)3-Et-Im][dca]
[EMIM][OTf]
15%
< 0.5%
0.5%
12%
10%
10%
< 0.5%
< 0.5%
60% (~4000 mM)
30% (~2000 mM)
20% (~1300 mM)
80% (~5300 mM)
16% (~1100 mM)
70% (~4700 mM)
20% (~1300 mM)
1% (~65 mM)
60 mM
< 50 mM
< 70 mM
420 mM
280 mM
200 mM
< 50 mM
< 50 mM
40 mM
< 20 mM
< 20 mM
640 mM
300 mM
50 mM
9
10
3
4
12
19
20
< 20 mM
< 20 mM
a DOPA is 3,4-dihydroxy-DL-phenylalanine.
As shown in Table 3, [EMIM][EtSO4] (9), [Amm110]Cl (10),
[EMIM][OTf] (20), and [Me(OEt)3-Et-Im][dca] (19) could barely
dissolve any cellulose, mainly because of the poor hydrogen-
bonding basicity of EtSO4 in 9 and OTf- in 20, and a low Cl-
-
concentration in 10. It is known dca- based ILs are poor in
dissolving cellulose.15,33,36 Other acetate and formate-based ILs
(3, 4, 7 and 12) in Table 3 are able to dissolve at least 10% (wt)
of cellulose. This makes the enzymatic modification of cellulose
in ILs possible.15
The solubility of D-glucose in most ILs (Table 3) ranges from
16% (wt) (~1100 mM) to 80% (~5300 mM) at 60 ◦C. Particularly,
the glucose solubility in [Me(OEt)3-Et-Im][OAc] (3) is as high
as 80% (wt) at 60 ◦C. Since D-glucose is not quite soluble
in most organic solvents, these encouraging outcomes enable
the synthesis of glucose derivatives at high concentrations via
enzymatic routes.15 In addition, we also observed that IL 3
could dissolve more than 20% (wt) L-ascorbic acid (vitamin
C) at 25 ◦C (data not shown in Table 3). On the other hand,
D-glu◦cose has a poor solubility in [EMIM][OTf] (1% or ~65 mM
at 60 ), which is consistent with literature values (30.6 mM in
[BMIM][OTf] at 40 ◦C,65 6.1 g/L or 34 mM in [EMIM][OTf] and
Fig. 2 Stability of immobilized CALB in [Me(OEt)3-Et-Im][OAc] (3)
at 50 ◦C (see Experimental section).
hydrogen-bonding strength as implied by the relatively low
solubilities of CALB (Table 2) and substrates (Table 3). It is
also important to reiterate that there is some leaking of the
lipase from Novozym-435 into the ether-functionalized ILs;15
therefore, the covalent immobilization of CALB deserves a
further study in these ILs.
◦
4.8 g/L or 27 mM in [BMIM][OTf] at 25 C;64 supersaturated
concentrations in OTf- ILs range from 113–363 mM64,65). The
dca- based IL (19) dissolved a high concentration (20%) of D-
glucose, but could dissolve little other compounds (see Table 3).
Amino acids, as important Active Pharmaceutical Ingredients
(APIs), are another type of compounds that are usually soluble
in water but not soluble in most organic solvents without
proper N-protection. Therefore, a direct (chemical or enzymatic)
modification of amino acids in organic solvents is not efficient.
We selected an important amino acid known as DOPA (3,4-
dihydroxy-DL-phenylalanine) to demonstrate its solubility in
ILs (more comprehensive study of other amino acids will be
reported elsewhere). Its enantiomer, L-DOPA is able to increase
dopamine levels for the treatment of Parkinson’s disease and
DOPA-Responsive Dystonia. As shown in Table 3, DOPA has
a poor to moderate solubility in 7, 9, 10, 19 and 20, but has a
very high solubility in 3, 4 and 12 (200–420 mM).
Betulinic acid (3b-hydroxy-lup-20(29)-en-28-oic acid) (21)
(Scheme 3) is a pentacyclic lupane-type triterpene found in
the bark of some trees (such as birch). This compound and
its derivatives have exhibited a variety of biological properties
such as anticancer, anti-HIV, antibacterial, anti-malarial, anti-
inflammatory, anthelmintic activities.73,74 However, betulinic
acid has very low solubilities in water and common organic
solvents (such as alcohols, ethers and esters), which creates a
Solubility of different substrates in ether-functionalized ILs
Previously, we observed a low cellulose solubility in ILs 8, 10,
14–17 due to their low anion concentrations.15 In contrast,
ILs 3, 4 and 11–13, have ‘super’ dissolution power towards
many substances that are difficult to dissolve in conventional
organic solvents and/or aqueous solutions (Ref. 15 and Table 3).
In this paper, we selectively report the solubilities of several
representative substrates in ILs (Table 3). Cellulose is the most
abundant natural biomass on earth. However, many cellulose-
related processes are hampered by the low solubility of cellulose
in most organic solvents. In the past several years, it was
discovered that certain ILs can dissolve ~10% (wt) or more
cellulose; the main dissolution mechanism is known as the high
hydrogen-bonding basicity of anions (such as chloride, formate,
acetate or alkylphosphonate).15,68,69 Experimentally, 13C and
35/37Cl NMR relaxation measurements on [BMIM]Cl confirmed
that chloride ions interact with the cellulose -OH groups in a 1:1
stoichiometry (-OH ◊ ◊ ◊ Cl).70 Molecular dynamics simulation of
b-D-glucose in 1,3-dimethylimidazolium chloride ([MMIM]Cl)
also reached the same conclusion about the hydrogen bonding
between chloride anions and hydroxyl groups (-OH ◊ ◊ ◊ Cl).71,72
This journal is
The Royal Society of Chemistry 2009
Green Chem., 2009, 11, 1128–1138 | 1133
©

View Article Online
Scheme 3 Synthesis of methyl-phthalate of betulinic acid (22) via
enzymatic transesterification in [Me(OEt)3-Et-Im][OAc] (3).
huge barrier for a further chemical or enzymatic modification
of this compound. For example, its solubility in water is only
about 0.02 mg/mL at room temperature.75 We also measured
◦
the solubility in some organic solvents at 25 C: 1% (wt/v) in
ethanol, 5% (wt/v) in DMSO, and > 23% (wt/v) in pyridine.
The high solubility in pyridine is due to the high basicity of the
solvent; however, many chemical and enzymatic reactions can
not be carried out in pyridine. On the other hand, betulinic acid is
very soluble in ILs 3 and 4 (640 and 300 mM respectively at 25 ◦C,
Table 3). This creates a new opportunity for derivatizing this
naturally occurring compound in ionic media at high concen-
trations. The enzymatic conversions of some of these molecules
in ether-functionalized ILs are illustrated in a later section.
FT-IR and fluorescence determination of protein structures in ILs
The infrared (IR) spectra of proteins can reveal characteristic
amide bands from different vibrations of the peptide moiety.
Among different amide modes of the peptide group, the most
=
common mode is the amide I, which is the C O stretching
vibration of the amide group between the region of 1600 and
1700 cm-1. The second derivatives of IR spectra in this region are
often used to study the protein’s secondary structures in terms
of structural elements such as a-helix, b-sheets, b-turns, and
non-ordered or irregular structures.76 The a-helical structures
are normally shown between 1650 and 1658 cm-1, and b-sheets
are between 1620 and 1640 cm-1.76 The assignments of infrared
amide I band components with various secondary protein struc-
tures are summarized from the literature (Table S.1 in ESI†).
As shown in Fig 3(a) and (b), after incubated in ILs 3 and
4 respectively for 24 h at 50 ◦C, free CALB has essentially
the same second derivative spectra as that in water, suggesting
these two ILs do not disrupt the protein’s secondary struc-
tures. A similar conclusion was observed in two other ILs
[OMIM][OAc] (8) and [EMIM][EtSO4] (9) (see ESI,† Fig. S.1).
However, second derivative spectra of free CALB incubated
in [EMIM][OAc] (7), [BMIM][HCOO] (2a) and [Me(OEt)3-
Et3N][HCOO] (12) suggest moderate to severe structural
changes: band shifting and uncharacteristic structures at 1623
(b-sheets) and 1678 cm-1 (b-turns) in IL 7 (Fig 3c), loosing
characteristics at 1623 (b-sheets), 1664 (b-turns) and 1671 cm-1
(b-turns) in IL 2a (Fig 3d), and disappearing a-helix and
b-sheets structures in IL 12 (Fig S.1). These structural changes
are generally in agreement with the lipase activities in ILs
(Table 2): high enzyme activity in ILs 3, 4, 8 and 9, but low
activity in 2a and 7. It also confirms our previous finding
that lower anion molar concentration of IL is more enzyme-
stabilizing (such as 3 and 8 are more enzyme-compatible than
2
Fig. 3 Second derivative (d2A/dn ) spectra of free CALB in water
(dashed line) and in ILs (solid lines, with different incubation times at
50 ◦C).
7 although they all contain acetate anions; 4 is more enzyme
compatible than 11). Based on the infrared spectrum (Fig S.1c),
IL 12 is a protein structure breaker and enzyme denaturant. The
free lipase in this IL is only moderately active, but it seems that
the immobilized enzyme could tolerate this IL (Table 2).
1134 | Green Chem., 2009, 11, 1128–1138
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
In summary, the presence of low molar concentrations of
denaturing anions (such as acetate) does not cause much change
of protein’s secondary structures. The ether-functionalized ILs
3 and 4 are lipase-compatible based on both activity data and
IR secondary derivatives.
On the other hand, it is also important to realize that the
enzyme may not be catalytically active even if its secondary
structures are well preserved in the solvent. [BMIM][dca] (6)
is known an enzyme-denaturing IL (Table 2 and Refs. 15,17),
however, the second derivative spectrum of CALB incubated
in 6 for 4 days is very similar to that of native lipase
(Fig S.1d).
Fluorescence and CD spectroscopic methods have
been successfully applied in analyzing the structures of
a-chymotrypsin77,78 and CALB78,79 in hydrophobic ILs includ-
ing [EMIM][Tf2N] and [BuMe3N][Tf2N]. However, a recent
study80 suggests that fluorescence and CD spectroscopy is
incapable of determining the protease’s structures in hydrophilic
[MMIM][Me2PO4] (1,3-dimethylimidazolium dimethylphos-
phate) due to the interference of IL. Our fluorescence data
(Fig S.2 in ESI†) also confirm that the emission maximum
(325 nm) of tryptophan (Trp) residues observed in water could
not be detected in ILs (3, 4 and 9) even after pure IL signals
were subtracted. CALB is a known monomeric protein (33
kDa) containing five Trp, nine Tyr, and ten Phe residues and
four disulfide bridges; three Trp residues (positions 52, 113,
and 155) are found in three different a-helices, while two other
residues (positions 65 and 104) are located in two different
b-sheets.81,82 In our fluorescence measurements, it is suspected
that IL molecules interact with Trp residues, causing the Trp
emission maximum not shown. However, a further study is
needed to understand the mechanisms of such interactions.
Lipase-catalyzed synthesis of fatty acid ester of D-glucose
Fatty acid sugar esters, as biodegradable nonionic surfactants,
have many applications in cosmetic, pharmaceutical, and food
industry. However, the enzymatic synthesis of these esters is
challenged by the selection of an appropriate solvent: most
organic solvents are not able to dissolve the highly polar sugar
and the nonpolar fatty acid at the same time. A few solvents
(such as pyridine) may serve this purpose, but they often cause
enzyme denaturation. Common organic solvents for enzymatic
transformations of D-glucose include t-butanol45 and 2-methyl-
2-butanol.86 However, sugar solubilities in these solvents are
usually low (such as < 20 mM D-glucose in 2-methyl-2-
butanol at 60 ◦C) although the supersaturated sugar solution
could be prepared (50 mM D-glucose in 2-methyl-2-butanol at
60 ◦C).86 The supercritical CO2 was explored as the alternative
solvent for the CALB-catalyzed synthesis of fructose palmitate,
where improved lipase thermal stability and conversions were
observed.87,88 However, the solubility of sugars in nonpolar CO2
is usually low; therefore, a polar co-solvent (such as t-butanol)
is usually added.87 Recently, it was observed that ILs based on
BF4 , OTf-, dca- and OAc- could dissolve high concentrations of
-
sugars.15,33,89 Among these ILs, BF4- based ILs (or mixtures with
OTf-, or Tf2N-) are often used in dissolving sugars for enzymatic
synthesis25,64,65,90 because BF4 anions in low-water systems are
-
less enzyme-denaturing than other anions.20 However, the sugar
-
solubility in BF4 based ILs is moderately low.
Since our lipase-compatible ILs (3 and 4) can dissolve high
concentrations of D-glucose (Table 3), we carried out the
synthesis of fatty acid esters in these ILs or their mixtures
with t-butanol. High concentrations of D-glucose (220 mM)
and vinyl laurate (330 mM) are fully soluble in ILs (3 and 4)
or their mixtures with t-BuOH, demonstrating the advantage of
employing these ionic solvents. In general, with the increase of
t-BuOH concentration (trials 1, 2 and 4 in Table 4), the D-glu-
cose conversion decreased, suggesting IL 3 is more enzyme
stabilizing than t-BuOH. Even in 40% IL 3 and 60% t-BuOH
(trial 4), D-glucose (220 mM) and vinyl laurate (330 mM) are
fully soluble; therefore, the limitation of substrate solubility is
unlikely the reason of lower conversion in high concentrations
of t-BuOH. The use of molecular sieves to scavenge water is
a common strategy to increase the acylation yield.25,64 How-
ever, the addition of molecular sieves into our ionic solvent
(containing 60% IL 3 and 40% t-BuOH, trial 3) only resulted
in a marginal increase in D-glucose conversion (from 79% to
82%). We rationalized two possible reasons: (1) Our solvents and
Enzymatic transesterification of betulinic acid
Due to the low solubility of betulinic acid in many organic
solvents, its synthesis and derivatization are limited to several
solvents, such as dichloromethane, chloroform, acetone, THF,
DMF, pyridine, benzene, etc.83–85 Most of these solvents (except
pyridine) do not dissolve significant amounts of betulinic acid.
Therefore, the efficiency of many reactions is limited by the poor
reactant solubility. However, as discussed previously, ILs (such
as 3 and 4) can dissolve high concentrations of betulinic acid
even at ambient temperature. Unlike pyridine being basic, these
ILs are neutral and even compatible with the lipase. For these
reasons, ether-functionalized ILs (3 and 4) could be suitable
media for chemical and enzymatic modification of betulinic
acid.
Table 4 Enzymatic transesterification of D-glucose and vinyl laurate
Solvent (IL/t-BuOH
D-glucose conversion
at 24 h
To demonstrate this concept, we conducted the CALB-
catalyzed synthesis of methyl-phthalate of betulinic acid (22)
in IL 3 via the transesterification of betulinic acid (21) and
dimethyl phthalate (Scheme 3). Hemiphthalic esters have higher
cytotoxicity than betulinic acid, but the chemical synthesis of 22
was reported as a multi-step strategy.85 In our enzymatic route,
a high substrate concentration (110 mM) was used in the one-
step reaction, and a high isolated yield (85%) was achieved (see
Experimental for procedures and product characterization). The
chemical derivatization of betulinic acid in ILs is being pursued
in our laboratory.
Trial
volume ratio)
1
2
3
IL 3/t-BuOH 100 : 0
IL 3/t-BuOH 60 : 40
IL 3/t-BuOH 60 : 40
(molecular sieves, 0.2 g)
IL 3/t-BuOH 40 : 60
IL 4/t-BuOH 100 : 0
85%
79%
82%
4
5
68%
71%
Reaction conditions: 1.0 mL solvent (IL used was 3 or 4), 0.22 mmol
D-glucose, 0.33 mmol vinyl laurate, 40 mg Novozym 435 and
50 ◦C. Molecular sieves (3A) were not added except in trial 3.
This journal is
The Royal Society of Chemistry 2009
Green Chem., 2009, 11, 1128–1138 | 1135
©

View Article Online
Table 5 Anomerization ratios of D-glucose in different solvents as
it was due to the rate of mutarotation is faster than the reaction
rate in 2-methyl-2-butanol. However, in ionic solutions (3 and
4), the interconversion of two anomeric forms of D-glucose
is expected slow because D-glucose molecules form strong
hydrogen bonds with anions.69–72 The NMR spectrum of our
product shows the chemical shift of glucose H1 is about 5.50 ppm
(see ESI†), suggesting the ester is a-D-glucose laurate.45
Recent reports93,94 suggest that the imidazolium-based ILs
(such as [EMIM][OAc] and [BMIM][OAc]) may react with
glucose (and reducing end group of cellulose) at C-2 position
of the imidazolium ring. After incubation in imidazolium IL
3 for 52 days at room temperature, D-glucose-13C6 showed
two tiny new 13C peaks at 63.7 and 64.3 ppm (see ESI,† Fig
S.3c), suggesting the side reaction between D-glucose and IL
3 occurred but insignificant. IL 3 has a long and flexible
alkyloxyalkyl chain than [EMIM][OAc] and [BMIM][OAc],
which could create steric hindrance for D-glucose to react with
3. On the other hand, after dissolving in ammonium-type IL 4
for 52 days at room temperature (see ESI,† Fig S.4c), D-glucose
showed no new 13C peaks, implying no reaction between these
two substances.
determined by 13C NMR
Solvent
b-D-glucose (%)
a-D-glucose (%)
2-methyl 2-butanol86
D2O
IL 3
IL 3 (52 days at r.t.)
IL 4
IL 4 (52 days at r.t.)
40
64
76
71
83
85
60
36
24
29
17
15
substrates were dried over MgSO4 (for t-BuOH and substrates)
or molecular sieves (for ILs), and immobilized CALB was also
intensively dried over P2O5; therefore, the initial water content
was extremely low. (2) Our hydrophilic ILs absorbed water
produced by the reaction; consequently these water molecules
are not ‘freely’ available for hydrolyzing D-glucose fatty acid
ester. The same reaction in IL 4 (trial 5) showed a lower glucose
conversion than that in IL 3 (trial 1). Our recent study15 has
confirmed that the CALB-catalyzed acylation of D-glucose in
these ILs is highly regioselective and occurs at the C-6 position
of D-glucose.
However, the acylation reactions in ILs 3 and 4 were relatively
slow and 24 h is usually needed even with vinyl laurate as
the acyl donor. On the other hand, the same reaction was
observed faster in organic solvents (such as 2-methy-2-butanol86)
and in other ILs (such as [BMIM][OTf]64 or its mixture with
[BMIM][Tf2N]65), where most of the product was obtained in
10 h (although these solvents could dissolve much less glucose
than 3 and 4). Previously, we also reported low enzymatic con-
versions of D-glucose when reacting with methyl methacrylate
in 3 and 4.15 We suspected that since D-glucose is very soluble in
these ionic media, the substrate ground-state stabilization might
be the reason behind the slower reaction rates. To articulate
this speculation, we dissolved 13C-labeled D-glucose (13C6) in
D2O, IL 3 and 4 respectively, and then determined the anomer
ratios of D-glucose (see Figs. S.3 and S.4 in ESI,† and Table 5)
in these solutions from 13C spectrum integrations (based on
C1). When dissolving in 2-methyl-2-butanol, D-glucose has 40%
b-anomer and 60% a-anomer.86 In water (D2O), our NMR
measurements suggest there are more b-anomer (64%) than a-
anomer (36%), which is consistent with the known literature
values.91 In two ionic solutions (3 and 4), there are about 80%
b-D-glucose anomers and only 20% a-anomer (after 52 days of
incubation at room temperature, there are only slight changes
of the ratios; see Table 5). It is considered that strong solvation
effects account for the high abundance of b-anomers in water,92
and b-glucose has a lower free energy and thus more stable than
a-anomer.91 Based on this theory, D-glucose is highly solvated
in ILs 3 and 4, presumably due to the hydrogen bond network
between IL molecules and D-glucose. We also suspected that
during the enzymatic (trans)esterification, a-anomer is more
reactive due to the higher substrate ground-state energy. This
speculation is supported by the CALB-catalyzed esterification of
D-glucose with dodecanoic acid in t-butanol, where much more
a-anomer ester was produced than the b-form (a/b = 84/14
from GC analysis).45 Although Flores et al.86 observed both
anomeric forms of glucose were consumed at the same rate in the
enzymatic acylation with dodecanoic acid, they rationalized that
Conclusions
Ether-functionalized ILs were found capable of dissolving free
CALB, and stabilizing the enzyme at the same time. The infrared
spectra of lipase confirmed that these ILs do not disrupt the
secondary structures of proteins. These new ionic solvents
have ‘super’ dissolution ability towards a variety of substrates
including sugars, cellulose, L-ascorbic acid, betulinic acid and
amino acids. To demonstrate the potential biotransformation of
these substrates in ionic media, we carried out two different
lipase-catalyzed reactions to synthesize methyl-phthalate of
betulinic acid and D-glucose fatty acid esters, respectively. The
success of these reactions can be attributed to the high substrate
solubilities in ILs and the enzyme-stabilizing nature of ionic
media. However, we should cautiously examine if the high
substrate solubility in ILs may also cause the ground-state
stabilization of reactants (resulting in less-reactive substrates),
and if there are side reactions between ILs and substrates (such
as the case of imidazolium ILs reacting with D-glucose and
cellulose).
Acknowledgements
Acknowledgement is made to the Donors of the American
Chemical Society Petroleum Research Fund (46776-GB1) for
partial support of this research. The authors are very grateful
for constructive comments from two anonymous reviewers.
Notes and references
1 A. Zaks and A. M. Klibanov, Proc. Natl. Acad. Sci. USA, 1985, 82,
3192–3196.
2 G. Bell, P. J. Halling, B. D. Moore, J. Partridge and D. G. Rees, Trends
Biotechnol., 1995, 13, 468–473.
3 A. M. Klibanov, Trends Biotechnol., 1997, 15, 97–101.
4 M. N. Gupta and I. Roy, Eur. J. Biochem., 2004, 271, 2575–2583.
5 J. Xie and Y.-L. Hsieh, J. Polym. Sci. Part A: Polym. Chem., 2001,
39, 1931–1939.
1136 | Green Chem., 2009, 11, 1128–1138
This journal is
The Royal Society of Chemistry 2009
©

View Article Online
6 V. M. Paradkar and J. S. Dordick, J. Am. Chem. Soc., 1994, 116,
5009–5010.
7 V. M. Paradkar and J. S. Dordick, Biotechnol. Bioeng., 1994, 43,
529–540.
42 Z. Guo, B. Chen, R. L. Murillo, T. Tan and X. Xu, Org. Biomol.
Chem., 2006, 4, 2772–2776.
43 B. Chen, Z. Guo, T. Tan and X. Xu, Biotechnol. Bioeng., 2008, 99,
18–29.
8 F. F. Bruno, J. A. Akkara, M. Ayyagari, D. L. Kaplan, R. Gross, G.
Swift and J. S. Dordick, Macromolecules, 1995, 28, 8881–8883.
9 F. F. Bruno, J. S. Dordick, D. L. Kaplan and J. A. Akkara, in Enzymes
in Polymer Synthesis, Vol. Chapter 9 (Eds.: R. Gross, D. L. Kaplan,
G. Swift,), American Chemical Society, Washington, DC, 1998,
pp. 167–174.
44 R. J. L. Bondar and D. C. Mead, Clin. Chem., 1974, 20, 586–590.
45 P. Degn, L. H. Pedersen, J. ø. Duus and W. Zimmermann, Biotechnol.
Lett., 1999, 21, 275–280.
46 S. Park, F. Viklund, K. Hult and R. J. Kazlauskas, Green Chem.,
2003, 5, 715–719.
47 B. Berger and K. Faber, J. Chem. Soc. Chem. Commun., 1991, 1198–
10 P. Wasserscheid and T. Welton, Ionic Liquids in Synthesis, Vol. 1 & 2,
1200.
Wiley-VCH, Weinheim, 2008.
48 T. Itoh, E. Akasaki and Y. Nishimura, Chem. Lett., 2002, 31, 154–
155.
49 S. H. Krishna and N. G. Karanth, Catal. Rev, 2002, 44, 499–591.
50 K. Weissermel and H.-J. Arpe, Industrial Organic Chemistry, VCH
Verlagsgesellschaft, Weinheim, 1993.
51 S. A. Forsyth, D. R. MacFarlane, R. J. Thomson and M. von Itzstein,
Chem. Commun., 2002, 714–715.
52 D. R. MacFarlane, J. M. Pringle, K. M. Johansson, S. A. Forsyth
and M. Forsyth, Chem. Commun., 2006, 1905–1917.
53 R. M. Lau, F. van Rantwijk, K. R. Seddon and R. A. Sheldon, Org.
Lett., 2000, 2, 4189–4191.
11 F. van Rantwijk and R. A. Sheldon, Chem. Rev., 2007, 107, 2757–
2785.
12 H. Zhao, Chem. Eng. Commun., 2006, 193, 1660–1677.
13 H. Ohno, Electrochemical Aspects of Ionic Liquids, John Wiley &
Sons, Hoboken, NJ, 2005.
14 R. M. Lau, M. J. Sorgedrager, G. Carrea, F. van Rantwijk, F. Secundo
and R. A. Sheldon, Green Chem., 2004, 6, 483–487.
15 H. Zhao, G. A. Baker, Z. Song, O. Olubajo, T. Crittle and D. Peters,
Green Chem., 2008, 10, 696–705.
16 M. B. Turner, S. K. Spear, J. G. Huddleston, J. D. Holbrey and R. D.
Rogers, Green Chem., 2003, 5, 443–447.
54 B. Re´jasse, S. Lamare, M.-D. Legoy and T. Besson, Org. Biomol.
Chem., 2004, 2, 1086–1089.
55 B. Re´jasse, T. Besson, M.-D. Legoy and S. Lamare, Org. Biomol.
Chem., 2006, 4, 3703–3707.
17 A. R. Toral, A. P. de los R´ıos, F. J. Herna´ndez, M. H. A. Janssen,
R. Schoevaart, F. van Rantwijk and R. A. Sheldon, Enzyme Microb.
Technol., 2007, 40, 1095–1099.
18 A. P. de los R´ıos, F. J. Herna´ndez-Ferna´ndez, F. A. Mart´ınez, M.
Rubio and G. V´ıllora, Biocatalysis Biotransform., 2007, 25, 151–156.
19 M. Moniruzzaman, N. Kamiya and M. Goto, Langmuir, 2009, 25,
977–982.
20 H. Zhao, G. A. Baker, Z. Song, O. Olubajo, L. Zanders and S. M.
Campbell, J. Mol. Catal. B: Enzym., 2009, 57, 149–157.
21 K. Fujita, D. R. MacFarlane and M. Forsyth, Chem. Commun., 2005,
4804–4806.
56 P. J. Halling, Biocatalysis Biotransform., 1987, 1, 109–115.
57 C. Laane, S. Boeren, R. Hilhorst and C. Veeger, in Biocatalysis in
organic media (Eds.: C. Laane, J. Tramper, M. D. Lilly,), Elsevier,
Amsterdam, 1987, pp. 65-84.
58 D.-J. van Unen, J. F. J. Engbersen and D. N. Reinhoudt, Biotechnol.
Bioeng., 2001, 75, 154–158.
59 Y. Mei, A. Kumar and R. Gross, Macromolecules, 2003, 36, 5530–
5536.
22 K. Fujita, D. R. MacFarlane, M. Forsyth, M. Yoshizawa-Fujita, K.
Murata, N. Nakamura and H. Ohno, Biomacromolecules, 2007, 8,
2080–2086.
60 H. Zhao, Z. Song, O. Olubajo and J. V. Cowins, Appl. Biochem.
Biotechnol., 2009, submitted.
61 T. L. Amyes, S. T. Diver, J. P. Richard, F. M. Rivas and K. Toth,
J. Am. Chem. Soc., 2004, 126, 4366–4374.
62 A. M. Magill and B. F. Yates, Aust. J. Chem., 2004, 57, 1205–
1210.
23 F. van Rantwijk, F. Secundo and R. A. Sheldon, Green Chem., 2006,
8, 282–286.
24 T. Maruyama, H. Yamamura, T. Kotani, N. Kamiya and M. Goto,
Org. Biomol. Chem., 2004, 2, 1239–1244.
63 J. Dupont and J. Spencer, Angew. Chem. Int. Ed., 2004, 43, 5296–
25 F. Ganske and U. T. Bornscheuer, Org. Lett., 2005, 7, 3097–3098.
26 H. Ohno, Bull. Chem. Soc. Jpn., 2006, 79, 1665–1680.
27 M. Yoshizawa and H. Ohno, Electrochim. Acta, 2001, 46, 1723–
1728.
5297.
64 S. H. Lee, D. T. Dang, S. H. Ha, W.-J. Chang and Y.-M. Koo,
Biotechnol. Bioeng., 2008, 99, 1–8.
65 S. H. Lee, S. H. Ha, N. M. Hiep, W.-J. Chang and Y.-M. Koo,
J. Biotechnol., 2008, 133, 486–489.
66 S. H. Lee, H. M. Nguyen, Y.-M. Koo and S. H. Ha, Process Biochem.,
2008, 43, 1009–1012.
67 S. H. Ha, M. N. Lan, S. H. Lee, S. M. Hwang and Y.-M. Koo, Enzyme
Microb. Technol., 2007, 41, 480–483.
68 T. Liebert and T. Heinze, Bioresources, 2008, 3, 576–601.
69 R. C. Remsing, G. Hernandez, R. P. Swatloski, W. W. Massefski,
R. D. Rogers and G. Moyna, J. Phys. Chem. B, 2008, 112, 11071–
11078.
28 J. Pernak, A. Czepukowicz and R. Pozniak, Ind. Eng. Chem. Res.,
2001, 40, 2379–2383.
29 L. C. Branco, J. N. Rosa, J. J. Moura Ramos and C. A. M. Afonso,
Chem. Eur. J., 2002, 8, 3671–3677.
30 J. Fraga-Dubreuil, M.-H. Famelart and J. P. Bazureau, Org. Process
Res. Dev., 2002, 6, 374–378.
31 J. Pernak, A. Olszo´wka and R. Olszewski, Polish J. Chem., 2003, 77,
179–187.
32 U. Domanska and A. Marciniak, J. Chem. Thermodyn., 2005, 37,
577–585.
70 R. C. Remsing, R. P. Swatloski, R. D. Rogers and G. Moyna, Chem.
33 Q. Liu, M. H. A. Janssen, F. van Rantwijk and R. A. Sheldon, Green
Chem., 2005, 7, 39–42.
34 E. Kuhlmann, S. Himmler, H. Giebelhaus and P. Wasserscheid, Green
Chem., 2007, 9, 233–242.
35 M. Wang, X. Xiao, X. Zhou, X. Li and Y. Lin, Sol. Energy Mater.
Sol. Cells, 2007, 91, 785–790.
Commun., 2006, 1271–1273.
71 T. G. A. Youngs, C. Hardacre and J. D. Holbrey, J. Phys. Chem. B,
2007, 111, 13765–13774.
72 T. G. A. Youngs, J. D. Holbrey, M. Deetlefs, M. Nieuwenhuyzen,
M. F. C. Gomes and C. Hardacre, ChemPhysChem, 2006, 7, 2279–
2281.
36 G. Laus, G. Bentivoglio, H. Schottenberger, V. Kahlenberg, H.
Kopacka, T. Ro¨der and H. Sixta, Lenzinger Berichte, 2005, 84,
71–85.
73 P. Yogeeswari and D. Sriram, Curr. Med. Chem., 2005, 12, 657–666.
74 P. A. Krasutsky, Nat. Prod. Rep., 2006, 23, 919–942.
75 S. Ja¨ger, K. Winkler, U. Pfu¨ller and A. Scheffler, Planta Med., 2007,
73, 157–162.
76 W. K. Surewicz, H. H. Mantsch and D. Chapman, Biochem., 1993,
32, 389–394.
37 H. S. Schrekker, D. O. Silva, M. A. Gelesky, M. P. Stracke, C. M. L.
Schrekker, R. S. Gonc¸alves and J. Dupont, J. Braz. Chem. Soc., 2008,
19, 426–433.
38 J. Pernak and M. Branicka, J. Surfactants Detergents., 2003, 6, 119–
77 T. De Diego, P. Lozano, S. Gmouh, M. Vaultier and J. L. Iborra,
Biotechnol. Bioeng., 2004, 88, 916–924.
123.
39 L. J. A. Siqueira and M. C. C. Ribeiro, J. Phys. Chem. B, 2009, 113,
1074–1079.
40 G. D. Smith, O. Borodin, L. Li, H. Kim, Q. Liu, J. E. Bara, D. L.
Gin and R. Nobel, Phys. Chem. Chem. Phys., 2008, 10, 6301–
6312.
78 P. Lozano, T. De Diego, S. Gmouh, M. Vaultier and J. L. Iborra,
Biocatalysis Biotransform., 2005, 23, 169–176.
79 T. De Diego, P. Lozano, S. Gmouh, M. Vaultier and J. L. Iborra,
Biomacromolecules, 2005, 6, 1457–1464.
80 N. Wehofsky, C. Wespe, V. Cerovsky, A. Pech, E. Hoess, R. Rudolph
and F. Bordusa, ChemBioChem, 2008, 9, 1493–1499.
41 S. Dreyer and U. Kragl, Biotechnol. Bioeng., 2008, 99, 1416–1424.
This journal is
The Royal Society of Chemistry 2009
Green Chem., 2009, 11, 1128–1138 | 1137
©

View Article Online
81 J. Uppenberg, M. T. Hansen, S. Patkar and T. A. Jones, Structure,
1994, 2, 293–308.
82 J. Uppenberg, N. Oehrner, M. Norin, K. Hult, G. J. Kleywegt, S.
Patkar, V. Waagen, T. Anthonsen and T. A. Jones, Biochemistry,
1995, 34, 16838–16851.
83 H.-J. Jeong, H.-B. Chai, S.-Y. Park and D. S. H. L. Kim, Bioorg. Med.
Chem. Lett., 1999, 9, 1201–1204.
87 J.-H. Heo, S. Y. Kim, H.-S. Kim and K.-P. Yoo, Biotechnol. Lett.,
2000, 22, 995–998.
ˇ
88 S. Sabeder and M. Habulin Knez, Chem. Ind. Chem. Eng. Quart.,
2006, 12, 147–151.
89 S. Park and R. J. Kazlauskas, J. Org. Chem., 2001, 66, 8395–8401.
90 F. Ganske and U. T. Bornscheuer, J. Mol. Catal. B: Enzym., 2005,
36, 40–42.
84 N. G. Komissarova, N. G. Belenkova, L. V. Spirikhin, O. V. Shitikova
and M. S. Yunusov, Chem. Nat. Compd., 2002, 38, 58–61.
85 M. Kvasnica, J. Sarek, E. Klinotova, P. Dzubak and M. Hajduch,
Bioorg. Med. Chem., 2005, 13, 3447–3454.
86 M. V. Flores, K. Naraghi, J.-M. Engasser and P. J. Halling, Biotechnol.
Bioeng., 2002, 78, 815–821.
91 S. J. Angyal, Angew. Chem. Int. Ed., 1969, 8, 157–166.
92 S. Ha, J. Gao, B. Tidor, J. W. Brady and M. Karplus, J. Am. Chem.
Soc., 1991, 113, 1553–1557.
93 T. Liebert, Macromol. Symp., 2008, 262, 28–38.
94 G. Ebner, S. Schiehser, A. Potthast and T. Rosenau, Tetrahedron Lett.,
2008, 49, 7322–7324.
1138 | Green Chem., 2009, 11, 1128–1138
This journal is
The Royal Society of Chemistry 2009
©