Cholinium-based deep eutectic solvents and ionic liquids for lipase-catalyzed synthesis of butyl acetate

Article abstract of DOI:10.1016/j.molcatb.2015.09.005

The aim of this study was to analyze the advantages and limitations of cholinium-based ionic liquids (ILs) and deep eutectic solvents (DESs) used as green solvents for immobilized Candida antarctica lipase B-catalyzed synthesis. The reaction of acetic anhydride with 1-butanol to give short chain ester butyl acetate was chosen as a model reaction. Results showed that selected ILs (choline glycinate, choline alaninate, choline asparaginate, choline malate) and DESs (choline chloride mixtures with glycerol Gly, ethylene glycol EG, and urea U as hydrogen bond donors in molar ratio 1:2) are poor media for tested reaction if applied as pure solvents (yield <5%). This happens probably due to the substrate "entrapment" within DES structure through H-bonding. The addition of water (as a protic solvent) to the DESs strongly enhances both enzyme activity and reaction yield. Among different DES:water mixtures tested, the one with EG as hydrogen bond donor at water content of 5% (w/w) proved to be the most effective, resulting in esterification yield 80% (higher than yield obtained in a reference solvent n-heptane). Kinetic parameters of Ping-Pong Bi-Bi reaction mechanism were estimated for reactions in ChCl:EG:water mixtures and the chosen model seems to fit reaction mechanism description. Furthermore, in all ChCl:EG:water tertiary mixtures about 50-70% lipase activity decrease was observed after 3 days incubation. The enzyme activity stayed unchanged even after extending the incubation time to 15 days. Finally, the used DESs were assayed for cytotoxicity in fish (CCO) and human (MCF-7) cell lines, where they exhibited low cytotoxicity (EC₅₀ >5 mM) for both cell lines. Obtained results suggest that DESs are promising candidates for green biocatalysis.

Full text of DOI:10.1016/j.molcatb.2015.09.005

Accepted Manuscript
Title: Cholinium-based deep eutectic solvents and ionic
liquids for lipase-catalyzed synthesis of butyl acetate
ˇ
Author: Marina Cvjetko Bubalo Ana Jurinjak Tusek Marijana
´
ˇ
´
ˇ
ˇ
Vinkovic Kristina Radosevic Visnja Gaurina Srcek Ivana
ˇ ´
´
Radojcic Redovnikovic
PII:
S1381-1177(15)30066-7
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcatb.2015.09.005
MOLCAB 3237
To appear in:
Journal of Molecular Catalysis B: Enzymatic
Received date:
Revised date:
Accepted date:
6-5-2015
9-9-2015
10-9-2015
ˇ
Please cite this article as: Marina Cvjetko Bubalo, Ana Jurinjak Tusek,
´
ˇ
´
ˇ
ˇ
ˇ ´
Marijana Vinkovic, Kristina Radosevic, Visnja Gaurina Srcek, Ivana Radojcic
´
Redovnikovic, Cholinium-based deep eutectic solvents and ionic liquids for lipase-
catalyzed synthesis of butyl acetate, Journal of Molecular Catalysis B: Enzymatic
http://dx.doi.org/10.1016/j.molcatb.2015.09.005
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Cholinium-based deep eutectic solvents and ionic liquids for lipase-catalyzed synthesis of
butyl acetate
Marina Cvjetko Bubalo1* mcvjetko@pbf.hr, Ana Jurinjak Tušek1, Marijana Vinković2,
Kristina Radošević1, Višnja Gaurina Srček1, Ivana Radojčić Redovniković1
1Faculty of Food Technology and Biotechnology, University of Zagreb, Pierottijeva 6, HR-
10000 Zagreb, Croatia
2NMR Centre, Ruđer Bošković Institute, Bijenička cesta 54, HR-10000 Zagreb, Croatia
*Corresponding author: Tel.: + 385 1 46 05 136; Fax: + 385 1 46 05 078.
1

Graphical Abstract
2

Highlights
1. Pure cholinium-based ILs and DESs are poor medium for tested esterification.
2. The addition of water to DES promotes enzyme activity and esterification yield.
3. Ping-Pong Bi-Bi mechanism describes the esterification in DES very well.
4. The DESs exhibited low cytotoxicity for fish (CCO) and human (MCF-7) cell lines.
5. The DESs are promising candidates for green biocatalysis.
3

Abstract
The aim of this study was to analyze the advantages and limitations of cholinium-based ionic
liquids (ILs) and deep eutectic solvents (DESs) used as green solvents for immobilized
Candida antarctica lipase B-catalyzed synthesis. The reaction of acetic anhydride with 1-
butanol to give short chain ester butyl acetate was chosen as a model reaction. Results showed
that selected ILs (choline glycinate, choline alaninate, choline asparaginate, choline malate)
and DESs (choline chloride mixtures with glycerol Gly, ethylene glycol EG, and urea U as
hydrogen bond donors in molar ratio 1:2) are poor media for tested reaction if applied as pure
solvents (yield < 5 %). This happens probably due to the substrate “entrapment” within DES
structure through H-bonding. The addition of water (as a protic solvent) to the DESs strongly
enhances both enzyme activity and reaction yield. Among different DES:water mixtures
tested, the one with EG as hydrogen bond donor at water content of 5% (w/w) proved to be
the most effective, resulting in esterification yield 80% (higher than yield obtained in a
reference solvent n-heptane). Kinetic parameters of Ping-Pong Bi-Bi reaction mechanism
were estimated for reactions in ChCl:EG:water mixtures and the chosen model seems to fit
reaction mechanism description. Furthermore, in all ChCl:EG:water tertiary mixtures about
50–70% lipase activity decrease was observed after 3 days incubation. The enzyme activity
stayed unchanged even after extending the incubation time to 15 days. Finally, the used DESs
were assayed for cytotoxicity in fish (CCO) and human (MCF-7) cell lines, where they
exhibited low cytotoxicity (EC50>5 mM) for both cell lines. Obtained results suggest that
DESs are promising candidates for green biocatalysis.
Keywords: butyl acetate; cytotoxicity; ionic liquids; esterification; eutectic solvents; lipase;
Ping-Pong Bi-Bi reaction mechanism
4

1. Introduction
Within a green chemistry concept a growing area of research is devoted to the design of new,
more environment friendly solvents, indicatively called green solvents. Various tunable and
smart solvents have been evaluated for this purpose, and so far ionic liquids (ILs) and deep
eutectic solvents (DESs) stand out as the most promising candidates in the field of solvent
innovation for food technology, biotechnology and chemical technology [1]. Properties that
attribute ILs the title “environment friendly” are non-volatility, non-flammability and
excellent stability. Conventional imidazolium- and pyridinium-based ILs so far showed
excellent technological properties, however, they are expensive, based on non-sustainable
feedstock and possess relatively high environmental impact [2]. This led to development of
ILs from cheap and carefully selected naturally-derived materials (e.g. cholinium salts, sugars,
organic acids and amino acids) that are suitable for CO2 separation processes [3] and
pretreatment of lignocellulosic biomass [4]. As for their environmental impact, choline-amino
acid ILs for example displayed low toxicity toward the bacteria [5], aquatic crustacean A.
salina and HeLa cell culture [6], and were also classified as ‘readily biodegradable’ [5].
However, further ILs structure-toxicity studies have to be conducted before claiming them as
truly eco-friendly [2]. The DESs represent a new generation of liquid salts and are generally
based on mixtures of cheap and readily available components: nontoxic quaternary
ammonium salts (e.g. choline chloride) and a naturally-derived uncharged hydrogen-bond
donor (HBD) (e.g. sugars, alcohols, amides and carboxylic acids) [7]. They are liquid
supramolecules usually composed of common metabolites at specific molar ratios, including
water in some cases, and are characterized by extensive intermolecular interactions. Since
their emergence in broad use, these solvents have attracted attention in synthesis,
electrochemistry, nanomaterials, biochemistry, separation, and analysis [8]. The assumption
5

that these solvents are environmentally friendly is based on the benign property of their
starting materials, however, recent studies indicate that these assumptions have to be carefully
examined [2,8].
Lipases, as catalysts in synthetic (ester synthesis, interesterification and transesterification
reactions) or hydrolytic reactions, are the most industrially applied enzymes in food,
cosmetic, and pharmaceutical industry, as well as in, surfactant synthesis and preparation of
biodiesel [9]. These enzymes tolerate non-natural environments surprisingly well, and have
been established as promising catalysts for biotransformations in organic solvents, certain ILs
and DESs [10-12]. So far, hydrophobic imidazolium- and pyridinium-based ILs with long
alkyl chains attached to cation core and anions, such as bis(trifluoromethylsulfonyl)imide and
hexafluorophosphate, proved to be excellent media for various lipase-catalyzed synthetic and
hydrolytic reactions, as well as for regio- and stereoselective transformations [13,14], whereas
the potential of ILs from natural sources (e.g. cholinium-based ILs) have been reported only
by Deive et al. [15]. The DESs were at first considered as a poor choice for biocatalysis due to
denaturing properties of HBD like urea. However, Gorke et al. [16] were the first to find that
many hydrolytic enzymes retain excellent activity in such environment and associated this to
the fact that strong H-bonds between DES components lowered reactivity of individual
components. To date, several publications have pointed out the use of DESs based on choline
chloride or ethylammonium chloride with a HBD such as alcohols, acids or amides as media
for lipase-catalyzed esterification [17,18], transesterification [16,19], alcoholysis [12,20],
lipophilization of phenolic acids [21], epoxidation of fatty acids and phenolic compounds [22]
and acylation reactions [23].
Esterification reactions are commercially interesting for lipase-catalyzed preparation of short
chain flavour esters by a reaction between short chain acid and alcohol. These esters are
6

commonly used as flavour additives in food-, pharma- and cosmetic industry [14,25-26]. A
number of papers regarding their enzymatic synthesis in nonconventional media, such as n-
heptane [24], supercritical CO2 [25] and imidazolium ILs [26,14] have been published so far.
Herein the potential of chlolinium-based solvents as media for lipase-catalyzed preparation of
short chain esters have been investigated. The synthesis of butyl acetate, as a bench reaction
for proposed estrification, was conducted in a series of cholinium-based ILs and DESs using
immobilized lipase. Additionally, to evaluate environmental properties of DESs, cytotoxicity
in fish (CCO) and human (MCF-7) cell lines have been investigated.
2. Material and methods
2.1. Biological and chemical materials
Novozym 435 (lipase B from Candida antarctica; immobilized on macro-porous polyacrylic
resin beads), 1-butanol, acetic anhydride, butyl acetate, n-heptane and all chemicals for ILs
and DESs syntheses were purchased from Sigma Aldrich, Germany (purity of ≥ 99%), and
used without further purification.
The CCO fish cell line (ATCC No. CRL-2772) and MCF-7 human tumor cell line (ATCC
No. HTB-22) were used in this work. WST-1 assay was purchased from Roche, Germany.
Dulbecco’s Modified Eagle’s Medium and fetal bovine serum were purchased from Gibco,
UK.
To confirm structures of obtained ILs, 1H NMR spectra from DMSO-d6 were recorded with a
Bruker AV-300 spectrometer at NMR Centre on Ruđer Bošković Institute (Zagreb, Croatia).
7

2.2. Deep eutectic solvents
Choline chloride (ChCl) and HBD (glycerol Gly, ethylene glycol EG and urea U) were dried
in the vacuum concentrator (Savant SPD131DDA SpeedVac Concentrator, Thermo scientific,
USA) at 60°C for 24 hours before use. The ChCl and HBD in molar ratio 1:2 were directly
weighed in a flask and the mixture was placed on the rotary vacuum evaporator (250 mbar
and 60°C) until a colourless liquid was formed (typically 2 h). The DESs synthesised were as
follows: choline chloride:glycerol (ChCl:Gly), choline chloride:ethylene glycol (ChCl:EG)
and choline chloride:urea (ChCl:U) (Table 1). The water content of the DESs and ILs was
determined by Karl Fischer titration (Metrohm, 787KF Titrino, Switzerland) at 20°C, and was
found to be <1% (w/w). Tertiary mixtures ChCl:HBD:water were also prepared. The ChCl,
HBD and water in the corresponding molar ratio (1:2:0.5-2.5, depending on HBD used) were
directly weighed in a flask and the mixture was stirred at 60°C until a colourless liquid was
formed (typically 1h).
2.3. Ionic liquids
Imidazolium-based IL 1-pentyl-3-methylimidazolium (bistrifluormethylsulfonyl)imide
[c5mim][Tf2N] was prepared as published previously [27]. Preparation of cholinium-based
ILs (Table 1) was performed according to standard procedures [28,29] or their modifications.
Briefly, an aqueous solution of amino acid (glycine, alanine or asparagine) (1.2 equiv) or
organic acid (malic acid) (1.2 equiv) was added dropwise into the choline hydroxide (1 equiv)
solution in water. The mixture was stirred for 24 h, followed by the removal of water under
reduced pressure. For amino acid-based ILs a mixed solution of acetonitrile and methanol
(9:1, v/v) was added into the crude product and the mixture was then stirred vigorously to
dissolve the IL and the excess amino acid (precipitate) was filtered off, and the solvents were
8

removed from the filtrate through a rotary evaporator. For organic acid-based IL product was
washed thoroughly with diethyl ether, dried under reduced pressure at 70°C. The IL product
was dried at 90°C for 24 h. 1H NMR spectra of obtained ILs matched previously reported
data [23-25].
2.4. GC analysis
Butyl acetate, 1-butanol and acetic anhydride concentrations in the n-heptane phase were
determined by a gas chromatograph equipped with a Rxi_5Si/MS column (30 m x 0.25 mm
i.d. x 0.25 mm) and with electrospray ionization mass spectrometry (Shimadzu, Japan).
Helium was used as a carrier gas at a flow rate of 109.6 ml min-1. The temperature of the oven
at the injection was 55°C and was kept constant during whole analysis time (3 min). Injector
and detector temperatures were set at 250°C. The ionization of the samples was achieved at
70 eV using the SCAN mode. Quantification of data was done by the calibration with
standard samples.
2.5. Esterification
The DESs and ILs were stored in vacuum oven at 60°C for 24 h prior to reaction. All
experiments were carried out in test tubes at 25°C placed on a vortex mixer at 2000 rpm. The
reaction started by adding 5 mg of Novozym 435 to 2 ml of solvents (IL, DES, tertiary
mixture ChCl:HDB:water or n-heptane) containing 0.05 mol l-1 acetic anhydride and 0.1 mol
l-1 1-butanol (Fig 1). Reactions without the enzyme were also performed. When monitoring
reactions in DES and IL, butyl acetate was recovered at specified time intervals via liquid
extraction using n-heptane. The biphasic mixture containing 50 μl of reaction aliquots and 50
μl of n-heptane was strongly shaken for 3 min and analyzed by a gas chromatograph as stated
above. For reactions performed in n-heptane, aliquots of 20 μl were directly analyzed by a gas
9

chromatograph. All experimental measurements were performed in triplicate and in 95% the
confidence range results had no statistical differences.
Concentrations of butyl acetate in DESs and ILs were calculated taking into account
coefficients of partition of butyl acetate between DES/IL and n-heptane. Briefly, 1 ml of DES
or IL containing 0.05 mol l-1 butyl acetate was incubated for 3 min with 1 ml of n-heptane at
1500 rpm at 25°C. The concentration of butyl acetate in the n-heptane phase was determined
by gas chromatography as stated above. The partitioning coefficient was evaluated as the ratio
of butyl acetate concentration in n-heptane phase to its concentration in DES or IL, where the
latter was calculated from the initial butyl acetate concentration.
2.6. NMR spectroscopy
1H NMR spectra from DMSO-d6 were recorded with a Bruker AV-300 spectrometer. One-
1
13
and twodimensional homo- and heteronuclear H and C NMR spectra of pure mixtures were
recorded with a Bruker AV-600 spectrometer, operating at 600.133 MHz for the 1H nucleus and
150.917 MHz for the 13C nucleus. Samples were measured as pure viscous mixtures with
external C6D6 and TMSsealedin capillary as standard at 25 °C in 5 mm NMR tubes.
1
The following mesurement techniques were used: standard H, APT, COSY, NOESY, HMQC
and HMBC. The NOESY spectra were measured in phase-sensitive mode, with mixing time
of 0.50 s and 64 scans per each increment.
2.7. Kinetic modelling of esterification reaction
The enzymatic esterification was described with Ping-Pong Bi-Bi mechanism according to
Siong et al [30]. According to the proposed mechanism, esterification process of acetic
anhydride and 1-butanol is occurring via two reactions:
10

Reaction 1 (main reaction):
aceticanhydride + 1- butanol→ aceticacid +butyl acetate
(A)
(B)
(P)
(Q)
Reaction 2 (secondary reaction):
aceticacid + 1- butanol→ water + butylacetate
(P) (B) (R) (Q)
Reaction mechanism can be described by equations (1-5) taking into account acetic anhydride
and acetic acid inhibition:
k1
k2
E+ A←⎯→ EA ⎯⎯→ E* +P
(1)
(2)
(3)
(4)
(5)
k3
k4
E* + B←⎯→ E*B ⎯⎯→ E +Q
k5
k6
E + P←⎯→ EP ⎯⎯→ E* +R
k7
E + A←⎯→ EAi
k8
E + P←⎯→ EP
i
During the esterification process E, EA, E*, E*B, EP, EAi and EPi behave as reactive
intermediates and according to Pseudo Steady State Hypothesis their rate of formation is
equal to zero (Equations 6-12):
d
[
E
]
= k1
[
E
][
A
]
+k−1
[
EA
]
− k5
[
E
][
P
]
+ k−5
[
EP
]
+ k4
[
E *B
]
− k7
[
E
][
A +
]
(6)
(7)
dt
k−7
[
EAi
]
− k8
[
E
][
P
]
+ k8
[
EP
]
= 0
i
d
[
EA
]
=k1
[
E
][
A
]
−k−1
[
EA
]
- k2
[
EA =0
]
dt
11

d
[
E *B
]
=k3
[
E *][
B
]
− k−3
[
E *B
]
- k4
[
E*B
]
=0
(8)
(9)
dt
d
[
EP
]
= k5
[
E
][
P
]
− k−5
[
EP
]
- k6
[
EP =0
]
dt
d
[
E*
]
= k2
[
EA
]
+ k6
[
EP
]
- k3
[
E*][
B
]
+ k−3
[
E*B
]
= 0
(10)
(11)
(12)
dt
d
[
EAi
]
=k7
[
E
][
A
]
− k−7
[
EAi
]
= 0
dt
d
[
EP
]
=k8
i
[
E
][
P
]
-k−8
[
EP = 0
]
i
dt
Enzyme conservation through reactions is expressed by Equation 13:
ET EA EP EAi EP E * E *B
[
]
=
[E
]
+
[
]
+
[
]
+
[
]
+
[
]
+
[
]
+
[
]
(13)
i
The expressions for the reaction rates for acetic anhydride, 1-butanol, acid and ester are
obtained rearranging Equations 6-13 using procedure described by Ishikawa et al. [28] in WR
Mathematica 10.0. The obtained expressions are given in equations 14-17:
d
[
A
]
= k−1
[
EA
]
−
(
k1 + k7
)
[
E
][
A
]
+ k−7
[
EA
]
i
dt
[
A
][
B
]
k1 k2 k3 k4
k3
k−1 k4 k7 + k2 k4k7 + k1 k2 k−7 + k1 k4k−7 ))k−8
...
k−5 k−8 + k6 k−8 + k−5 k8
(
k5 + k6
)
k−7 k−8 ET
(14)
= −
[
A
]
(
k5 + k6 )(k1 k2
(
k−3 + k4
)
k−7
+
[
B
]
(
+
(
k−1 + k2
)
k−7
+
((k−3 + k4
)
k5 k6 k−8
[
P
]
[
+
[
B
]
k3 k5 k6 k−8
(
[
P
]
+ k4
(
[
P
]
+ k6 k8
[
P
]
+ k5 k−8
[ ])) )
P
d
[
B
]
= −k4
[
E * B
]
dt
B
]
k3 k4 k−7 k−8
k−7
(
[
A
]
k1 k2
k−1 k4 k7 + k2 k4k7 + k1 k2 k−7 + k1 k4k−7 ))k−8
...
(
k−5 + k6
)
+
(
k−1 + k2
)
k5 k6
[
P
])
ET
= −
[
A
]
(
k5 + k6 )(k1 k2
(k−3 + k4
)
+
[
B
]
k3
(
+
(
k−1 + k2
)
k−7
+
((k−3 + k4
)
k5 k6 k−8
[P
]
+
[B
]k3
(
k5 k6 k−8
[P
]
+ k4
(
k−5 k−8 + k6 k−8 + k−5 k8
[P
]
+ k6 k8
[P
]
+ k5 k−8
[ ]
P
)) )
(15)
12

d
[ ]
P
= k2
[
EA
]
−
(
k5 + k8
)
[
E
][
k3 k4 k−7 k−8
k−3 + k4 k−7
P
]
+ k−5
[
EP
]
+ k−8
[
EP
]
i
dt
[
B
]
(
[
A
]
k1 k2
(
k−5 + k6
)
−
(
k−1 + k2
)
k5 k6
[
P ET
]
)
(16)
(17)
=
[
A
]
(
k5 + k6 )(k1 k2
(
)
+
[
B
]
k3
(
k−1 k4 k7 + k2 k4k7 + k1 k2 k−7 + k1 k4k−7 ))k−8
+
(
k−1 + k2
)
k−7
)) )
...
+
((k−3 + k4
)
k5 k6 k−8
[
P
]
+
[
B
]
k3
(
k5 k6 k−8
[
P
]
+ k4
(
k−5 k−8 + k6 k−8 + k−5 k8
[
P
]
+ k6 k8
[
P
]
+ k5 k−8
[
P
]
d
[
Q
]
= k4
[
E * B
]
dt
[
B
]
k3 k4 k−7 k−8
k−7
(
[
A
]
k1 k2
(
k−5 + k6
)
+
(
k−1 + k2
)
k5 k6
[P ET
]
)
=
[
A
]
(
k5 + k6 )(k1 k2
(
k−3 + k4
)
+
[
B
]
k3
(
k−1 k4 k7 + k2 k4k7 + k1 k2 k−7 + k1 k4k−7 ))k−8
+
(
k−1 + k2
)
k−7
)) )
...
+
((k−3 + k4
)
k5 k6 k−8
[
P
]
+
[
B
]
k3
(
k5 k6 k−8
[
P
]
+ k4
(
k−5 k−8 + k6 k−8 + k−5 k8
[
P
]
+ k6 k8
[
P
]
+ k5 k−8
[
P
]
Kinetic parameter values were estimated by fitting the experimental data for acetic anhydride,
1-butanol, acetic acid and butyl acetate to Equations 14-17 using ParametricNDSolveValue
algorithm implemented into WR Mathematica 10.0. The importance of individual kinetic
parameter describing the esterification process was analyzed using global sensitivity analysis.
FAST (Fourier Amplitude Sensitivity Test) method was used [32]. Application of FAST
provides sensitivities to large and simultaneous change of the complete set of the model
parameters.
2.8. Lipase stability
Novozym 435 (2.5 mg) was incubated in tertiary mixtures ChCl:EG:water (986 μl), n-heptane
and or [c5mim][Tf2N] at 25°C. At selected intervals within 15 days of incubation, enzyme
reactions were initiated by adding acetic anhydride and 1-butanol into the lipase-solvent
mixture in amounts to yield final concentrations 0.05 mol l-1 and 0.1 mol l-1, respectively, and
the reactions were monitored as stated above. The relative activity (%) was calculated from
the initial reaction rate obtained by the enzyme after incubation in certain solvent, compared
to the one obtained without previous exposure.
13

2.9. Cytotoxicity assay
In vitro toxic effects of tested DESs were determined on fish CCO and human tumor MCF-7
cells which were grown in Dulbecco’s Modified Eagle’s Medium (DMEM,) supplemented
with 10% (v/v) fetal bovine serum (FBS). Cells were cultured at 30°C (CCO cells) and 37 °C
(MCF-7 cells) in the humidified atmosphere with air (95%) and CO2 (5%). The cytotoxic
effect of tested DESs on cell proliferation was examined by the WST-1 assay (Roche,
Germany), according to the recently reported procedure [33].
3. Results and Discussion
3.1. Screening of pure DESs and ILs for esterification
Within this work cholinium-based DESs and ILs (table 1) have been evaluated as media for
the Candida antarctica lipase B-catalyzed butyl acetate synthesis accomplished by
esterification of 1-butanol with acetic anhydride (Fig 1). Reaction was also performed in
organic solvent (n-heptane) and hydrophobic imidazolium-based IL 1-pentyl-3-
methylimidazolium bis(trifluoromethylsulfonyl)imide [c5mim][Tf2N], both known as suitable
medium for lipase-catalyzed reactions [14,24]. Blank experiments without the enzyme were
also performed in all tested ILs and DESs, and no conversion was observed within reaction
times used in this study.
In cholinium-based ILs ([Ch][Gly], [Ch][Ala], [Ch][Asp] and[Ch][MA]), which are
considered as a green alternative to rather toxic imidazolium-based ILs [2], poor reaction
yields were observed (<5%) (Fig S1). It is possible that due to their high water affinity, these
compounds remove tightly bound water from the enzyme molecule and consequently cause its
14

deactivation by unfolding, as previously proposed for imidazolium-based hydrophilic ILs
[34]. In addition, Van Rantwijk et al. [35] suggested that carboxylate anions can form strong
and stable H-bonds with the enzyme polypeptide backbone and may dissociate the H-bonds
that maintain the structural integrity of the α-helices and β-sheets, causing the protein to
unfold wholly or partially.
When synthesis was performed in pure DESs (ChCl:Gly, ChCl:EG, ChCl:U), regardless of
DES used, low conversions were also observed (<5%), which is not in agreement with several
papers reporting on lipase activity within cholinium-based DESs [16,19].
The poor activity of lipase within DES choline chloride:urea (molar ratio 1:2) and laccase
within DES choline chloride:malic acid (molar ratio 1:1) was also noticed by Duran et al. [21]
and Choi et al. [36], respectively. Authors explained these observations by the protein
stiffening in low water content medium and/or substrate “entrapment” within DES structure
through H-bonding. Zhao et al. [37] also showed that activity and selectivity of proteases in
glycerol-based DESs is dependent on the water content, whereby a small amount of water of
about 5% can increased the enzymatic activity. As for our study, 1-butanol could act as both
H-donor and acceptor while acetic anhydride could act as H-bond acceptor, meaning they
could interact with the solvent (DES) and thus be unavailable for binding to the enzyme.
3.2. NMR study of interactions between substrates and DES
Nuclear magnetic spectroscopy (NMR) is a quite feasible tool to study molecular interaction
1
involved in function phenomenon DESs. Fig 2a shows the H NMR spectra of ChCl:EG,
abovementioned with 0.05 mol l-1 of acetic anhydride (mixture 1), with 0.1 mol l-1 of 1-
butanol (mixture 2) and mixture of all four compounds (mixture 3).
15

1
The H chemical shifts in mixture of ChCl and EG (1:2) are broad singlet at 3.29 ppm
(integrated 9 H, 3 x CH3 groups of ChCl), multiplet at 3.51-3.63 ppm (integrated 10 H, two
CH2 groups of EG and one of ChCl), broad singlet at 3.99 ppm (integrated 2 H, CH2 group of
ChCl), broad singlet at 4.84 ppm (integrated 4 H, OH groups of EG) and triplet at 5.26 ppm
(integrated 10 H, J = 9.96 Hz, OH group of ChCl). The presence of halide ion-H-bond donor
supramolecular complexes that characterize DESs ontaining ChCl and EG was mostly
reflected in the multiplet signal at 3.51-3.63. It is integrated 10 H and resulted from the
overlapping of methylene groups of EG and ChCl. Slightly broad linewidth of signals is
mainly due to the viscosities of DESs. Inter- and intra-dipolar interactions causes broadening
effect on the lineshapes of the NMR spectrum.
1
The H chemical shifts of compounds in mixtures 1, 2 and 3 are almost the same in addition
of chemical shifts of acetic anhydride in mixture 1 (singlet at 2.29 ppm, CH3 groups) and of
1-butanol in 2 (triplet at 0.92 ppm, CH3 group; sextet at 1.36 and quintet at 1.50 ppm, CH2
groups). There is presence of small amount of the free acetic acid and acetic acid ester of
ChCl and/or EG. 1H NMR spectrum of mixture 3 showed the characteristic signals of all four
present compounds. All the structures were confirmed by the COSY, APT, HMQC and
HMBC spectra. Nuclear overhauser enhancement spectroscopy (NOESY) is very useful
method for searching hydrogen bonding and the other molecular interactions of DESs. The
NOESY spectrum of mixture 0 shows strong H-bonding between hydroxyl groups of EG and
ChCl. The other NOE interactions are very weak OH (ChCl) – CH3 (ChCl) and weak OH
(EG) – CH3 (ChCl). Mixture 1 shows the same, without any other expected interactions
between mixture EG and ChCl and added acetic anhydride. In mixture 2 interactions are the
same except one new weak hydrogen bond between OH (ChCl) – OH (1-butanol) and one
16

new strong interactions, OH (EG) – OH (1-butanol). NOESY spectrum of mixture 3 shows all
the abovementioned hydrogen bondings and the other NOE interactions (Fig 2b and 2c).
3.3. The effect of water on esterification course within DESs
Given that NMR study implied that poor reaction yield could be a consequence of interactions
between substrate (1-butanol) and ChCl:EG (interactions of 1-butanol with ChCl:Gly and
ChCl:U could be expected accordingly) approach for substrate release from the net of H-
bonds was considered. The improvement in lipase-catalyzed catalysis by addition of certain
amount of protic co-solvent (water) to the system was successfully presented by Duran et al.
[21] when studying lipophilization of phenolic acids. Therefore, the influence of water
content in DESs on the course of esterification was herein tested. For this purpose
esterification in tertiary mixtures ChCl:HBD:water containing 1 mol ChCl and 2 mol HBD
(Gly, U, EG) as a function of water content (0.5-2.5 mol, depending on HBD used) have been
monitored, and results are given in Fig 3. Esterification yield for all DESs tested gradually
increased with the increase of water content in DES, however, at certain point optimum was
reached and further addition of water negatively impacted the yield. In general, the water
content optimum could be explained by the compromise between two different effects
occurring during reaction: when the water content is too low 1-butanol is associated with DES
through H-bonds and thus barely available for the reaction; whilst at water content above
optimum value the excess of water molecules near the active site generates a competitive
hydrolysis reaction, resulting in lower yields. At optimal water content not only substrate is
released from 3D H-bond net, but DES could also “tie up” water molecules formed during
esterification in the same 3D H-bond net (hydrolysis prevented) [37]. Differences in optimal
water content considering HBD used could be explained by the difference in the strength of
H-bonds formed in DESs. The Gly contains three strong HBD sites (hydroxyl groups)
17

compared to hydroxyl groups in EG and two amide groups in U which typically have weaker
H-bond donor properties than hydroxyl groups. This means that more water is required to
“occupy” all possible sites for H-bonding in DES with Gly as HDB (optimum at 2 mol of
water, Fig. 3a) than in DES with EG (optimum at 1.5 mol of water, Fig. 3c) and finally U
(optimum at 1 mol of water, Fig. 3b), corresponding to mass fraction of water of 10, 5 and
6%, respectively.
Initial reactions rates were also calculated and are presented in Fig 4. It could be observed that
both HBD and water content exhibit strong influence on enzyme activity and esterification
yield. Initial reaction rates gradually increased with the amount of water in DES. For
example, initial reaction rate in tertiary mixture ChCl:EG:water at 1:2:2 molar ratio (1.2 µmol
min-1 mg-1) was 4-fold higher than in the corresponding tertiary mixture at 1:2:0.5 molar ratio
(0.3 µmol min-1 mg-1). This supports the assumption that by addition of protic solvents
substrate (1-butanol) is „released“from 3D net of H-bonds (replaced with water molecules)
and available for reaction [9,21]. Additionally, since polarity of the DES ChCl:EG is lower
than water [38], it should also be considered that the polarity of the latter slightly increases
with the increase in water content, enhancing the solubility of 1-butanol and hence the rate of
mass transfer and the reaction rate. The highest initial rate was observed for ChCl:Gly
containing 2.5 mol of water (2.2 µmol min-1 mg-1), however still lower than values obtained in
reference solvents n-heptane (4.5 µmol min-1 mg-1) and [c5mim][Tf2N] (4.9 µmol min-1 mg-1).
In terms of the HBD (at optimum water content), alcohols Gly and EG triggered the enzyme
to exhibit higher enzyme activity and esterification yield than the third counterpart U, which
was also observed by Huang et al. [39] and Lindberg et al [40] when studying catalytic
performance of Penicillium expansum lipase and epoxide hydrolase from potato decreased in
various DESs, respectively. The authors explained this favourable effect of Gly and EG on
18

lipase activity to be related to the impact of these polyols on the enzyme’s catalytic
mechanism through their H-bonding with the hydroxyl group of the serine residue by
increasing the nucleophilicity of the hydroxyl group of the latter and leading to its fast attack
on the substrate [39]. Also, the chemical nature of the DES (e.g. type of HBD) can notably
affect the solvent properties such as polarity, viscosity, and surface tension, parameters that
have been known as the key determinants of the enzyme activity, stability and selectivity [40].
Overall, among tertiary mixture tested, the one with EG as HBD containing 1.5 mol of water
(5%, w/w) was shown to be the most appropriate one, resulting in esterification yield of 80%,
even higher that in n-heptane (50%) and [c5mim][Tf2N] (41%). It should also be mentioned
herein that the water added to DES should be considered as a part of eutectic solvent since
work done by Choi et al. [41] showed that water can be present as part of DES, being strongly
retained in the liquid (this water cannot be evaporated). For example, in some eutectic
mixtures of different sugars with ChCl water has to be present as part of the solvent to ensure
liquid state at room temperature (1:1:1 molar ratio), corresponding to approximately 6%
water, similar to optimal water contents observed for the esterification reaction studied herein.
3.4. Kinetic study of esterification reaction in ChCl:EG tertiary mixtures with water
After performing the esterification in different DESs and ILs, the DESs representing tertiary
mixtures Ch:EG:water were chosen as optimal reaction medium. For chosen DESs the kinetic
parameters for were estimated. The used model, based on the model proposed by Siong et al.
[27], describes the esterification reaction following Ping-Pong Bi-Bi mechanism which is also
the most accepted description of the lipase-catalyzed reactions, irrespective of the type of
reaction (e.g. hydrolysis, esterification, or transesterification) [42]. Based on the hypothesis
that the acyl enzyme intermediate forms a dead-end complex the inhibition by acetic
19

anhydride and acetic acid was considered. Proposed esterification occurs in two steps where
initially one acyl group of acetic anhydride is directly employed to form ester while the other
one leads to the formation of an acetic acid molecule (which may also react with the alcohol).
As for the reaction mechanism, it involves nucleophilic attack of active site serine residue of
the catalytic triad on the anhydride or acetic (formation of first tetrahedral intermediate)
whereupon the intermediate loses acetic acid or water molecule, respectively, to give an acyl-
enzyme complex. Nucleophilic attack by 1-butanol causes the addition of a hydroxyl group to
the carboxyl group (second tetrahedral intermediate), and after rearranging ester molecule is
released and the active site serine of lipase is regenerated. Both of the tetrahedral
intermediates possess an oxyanion that is stabilized by H-bonds to protein atoms of the
oxyanion hole [43]. The favourable enzyme activity in cholinium-based DESs containing EG
as HBD could, besides previously mentioned increase in the nucleophilicity of the hydroxyl
group of the active site serine residue, may be explained by the fact that this solvent supports
stabilization of these tetrahedral intermediates through H-bonding, leading to decrease in the
activity energy barrier and therefore increase the reactivity.
Reaction mechanism is herein described with seven reactions and 13 kinetic parameters.
ParametricNDSolveValue algorithm implemented into WR Mathematica 10.0 was used to
estimate the kinetic parameter values based on the experimental data on reactants and product
concentrations. The estimated values of kinetic parameters are given in Table 2. Kinetic
parameters estimation results show that acetic anhydride enzyme complex formation
(described with kinetic parameter k1) is the fastest in the DES with 2 mol of water and the
slowest in the DES with 0.5 mol of water. This is consistent with assumption that addition of
water promotes the reaction rate. For all analysed DESs the reaction equilibrium of enzyme-
acetic anhydride complex (EA) formation is moved to the right (kinetic parameter k-1 is
20

numerically smaller than parameter k1). The kinetic parameter k2 describes the reaction rate of
the enzyme-acyl intermediate (E*) formation and release of the acetic acid and it is the
highest for DES with 1 mol of water. Analysing the obtained results it can be also noticed that
binding of the enzyme-acyl intermediate and 1-butanol (described with k3) takes place most
rapidly in the DES with 2 mol of the water (possible due to release of 1-butanol from H-bond
net, as well as increase in polarity of solvents, in environment with high water content, as
explained earlier). The acetic acid binding to the free enzyme (described with k5) is the fastest
in the DES with 2 and isomerisation to enzyme-acyl intermediate (described with k6) is the
fastest in the DES with 1 mol of the water. Comparing the values of k4 estimated for different
amount of water in ChCl:EG it can be seen that the ternary complex enzyme-acyl-alcohol
intermediate is the fastest in the DES with 2 mol of water, while the slowest in the DES with
0.5 mol of water. The effect of the acetic anhydride and acetic acid inhibition was also
analysed calculating the inhibition constants for both substrates. Inhibition constants values
were obtained by dividing the rate of enzyme-anhydride and enzyme-acid complex with the
rate of the enzyme-anhydride and enzyme-acid complex degradation. The results show that
acetic anhydride inhibition constant is the highest for the DES with 0.5 mol (Ki7 = 45.43) and
the lowest for DES with 2 mol of water (Ki7 = 3.589). On the other hand acetic acid inhibition
constant has numerically the highest value for the DES with 2 mol of water (Ki8 = 149), and
the lowest for the DES with 0.5 mol of water (Ki8 = 17.2).
The comparison between experimental data and model results for butyl acetate concentrations
and given in Fig 5. It is observed that used model describes the experimental data very well.
The best agreement between experimental data model results is obtained between t = 90 min
and t = 150 min, while the biggest deviation between experiment and model is obtained in
first 30 min of experiment. This phenomenon could be explained with the limited number of
21

available experimental data. Nevertheless the used model can be useful for prediction of
concentration profiles in different DESs with slight variations in kinetic parameter values.
To get better insight into which of the kinetic parameters included into the model of
enzymatic synthesis of butyl-acetate is the most important the global sensitivity analysis was
applied. FAST (Fourier Amplitude Sensitivity Test) was used to get the insight into changes
of reactants and products concentrations as the effect of simultaneous change of value of all
model kinetic parameters. Results are given in Fig 6. It can be seen that concentration of
acetic anhydride, 1-butanol, acetic acid and butyl acetate are mostly sensitive to changes in
values of kinetic parameters k1 and k-1 included into reaction of enzyme-anhydride complex
formation and k2 describing the enzyme-anhydride complex unimolecular isomerisation.
Depending on the reactant or product the sensitivity coefficient values for mentioned kinetic
parameters varied in the range 37-57%. This is in accordance with findings of Peeters et al.
[44] who studied lipase-catalyzed ring-opening polymerizations of 4-substituted ε-
caprolactones employing Novozym 435 that the formation of the acyl–enzyme intermediate
and its deacylation as rate–determining step.
The present investigation shows that the reaction kinetics in all tertiary mixtures follows a
Ping-Pong Bi Bi mechanism with competitive inhibition by acetic anhydride and acetic acid.
Taking into account the obtained values for coefficient of determination (R2 (DES with 0.5
mol of water) = 0.979, R2 (DES with 1 mol of water) = 0.973, R2 (DES with 1.5 mol of water)
= 0.993 and R2 (DES with 2 mol of water) = 0.975) between experimental and data values of
butyl acetate concentration the obtained result can be used in future experiment planning
taking into account the importance of enzyme-substrate complex formation.
22

3.5. Stability of lipase ChCl:EG tertiary mixtures with water
To assess the suitability of tertiary mixtures ChCl:EG:water for lipase-catalyzed reactions, we
carried out a stability evaluation of Novozym 435 within these solvents. Stability was also
studied in n-heptane and imidazolium-based ILs [c5mim][Tf2N], being again taken as
reference solvents. Namely, previous studies on the stability of free and immobilized lipases
in various hydrophobic surroundings, such as n-heptane and imidazolium-based ILs with long
alkyl chains linked to imidazolium core associated with [Tf2N] anion, revealed the benefit of
their use due to coating and thus protecting the layer of essential water surrounding the
enzyme [14,45]. In this work, the lipase was incubated in each solvent for 15 days at 25°C
and at specific time intervals the initial specific activity of acylation of 1-butanol by acetic
anhydride was measured (Fig. 7). As expected, Novozym 435 showed excellent stability in
reference solvents within a 15-day period: in n-heptane no loss of activity was observed in
comparison with the non-incubated lipase, while in [c5mim][Tf2N] only a 7% decrease in
activity was observed. In all tested tertiary mixtures, about 50% activity decrease was
measured after 3 days incubation, nevertheless extending the incubation time up to 15 days
did not result in any significant enzyme activity loss. Similar trend was observed by Durand et
al. [20] in cholinium-based DES with Gly as HBD who reported no additional enzyme
activity loss after measured decrease in activity (down to 30 % of initial activity) at first 24 h
of incubation and explained it with enzyme need to adopt its final structure. In this work, after
15-day period, residual activity in ChCl:EG tertiary mixtures with water was in the range
from 27-46%, meaning that DES tested should allow to carry out long and complex reactions,
or even be used in continuously governed processes. The superior stability was observed for
ChCl:EG containing 1 mol of water, however, no clear trend between water content and
residual enzyme activity was detected. To explain such excellent stability of enzymes in DESs
23

Monhemi et al. [46] studied structure and dynamics of Candida antarctica lipase B in non-
aqueous DES (ChCl:U) with about 66% urea. The enzyme showed very good conformational
stability, in contrast to complete enzyme denaturation in water containing only 12% urea
through interrupting the intra-chain H-bonds. Namely, in ChCl:U, as a result of H-bonding, U
molecules cannot reach the protein domains due to a low diffusion coefficient. It was also
shown that U, Ch and Cl ions form H-bonds with the surface residues of the enzyme, leading
to greater enzyme stability. Wu et al. [47] also observed that, despite the destabilization effect
from U and choline acetate, horseradish peroxidise was greatly stabilized in the presence of
the DES containing them, meaning that the DESs affect the enzyme performance primarily by
virtue of the DES complex itself rather than by its individual components [48].
3.6. Cytotoxicity of DESs
To verify if the DESs used in described enzyme-catalyzed reactions are truly environment
friendly, we tested them for cytotoxicity in fish (CCO) and human (MCF-7) cell line by WST-
1 cell proliferation assay 72 h post treatment. Namely, application of cell lines as an in vitro
tool in the evaluation of toxicity induced by environmental pollutants is of the important
significance since their use satisfies a societal desire to avoid the use of whole animals in
toxicity testing. Also, such in vitro approach is highly recommended by EC Regulation
1907/2006 REACH which promotes the development of alternative methods for the substance
hazard assessment. Obtained cytotoxicity results of tested DESs are presented as EC50 values
in Table 3.
In both cell lines, ChCl:EG and ChCl:Gly and ChCl:U did not cause 50% of inhibition in the
range of tested concentrations (1-2000 mg l-1) meaning that EC50 values for this compounds
are >2000 mg l-1 (>10 mmol l-1). Comparing the obtained EC50 values to UFT Merck ILs
24

Biological Effects Database (http://www.il-eco.uft.uni-bremen.de), all tested cholinium-based
DESs exhibit low cytotoxicity (EC50>5 mmol l-1) toward both cell lines used in this study.
These data are in accordance with a general premise about DESs toxicity that non-toxic raw
materials used for DES preparation lead to formation of non-toxic solvents. However,
research of Hayyan et al. [49] emphasis the possibility of synergetic effect of combining the
compounds in the DESs which must be taken into account, because it could have significant
impact on biological properties of such mixtures. In addition, preliminary test of DESs’ effect
on the growth of Hydra sinensis conducted by Huang et al [48] implied that the DES
components (such as ChCl) may be toxic, whereby this toxicity can be weakened by
incorporating them into DES rather than by physically mixing them. Newest research from
Hayyan et al. [50] on the cytotoxic potential of ammonium-based DESs with four HBDs (Gly,
EG, triethylene glycol TEG and U) is conducted in vitro and in vivo using cell lines and
animal model. Obtained results on cytotoxic activity of DESs, their individual constituents
and aqueous solutions toward various carcinoma and normal cells are cell-line dependent and
for most cell lines IC50 values followed the sequence of TEGDES< GlyDES< EGDES< UDES
.
Although, IC50 differs significantly from our results on cholinium-based DES, overall
conclusion for example on DES containing U as HBD is that most cancer cell-lines are
resistant to it [50] what is in agreement with our observation that cholinium-based DESs
exhibit low cytotoxicity. In vivo toxicity studies using ICR mice showed that these DESs were
relatively toxic in comparison to their individual components through ROS production and
apoptosis induction [50]. Nevertheless, at further need to fully understand the biological
effects of DESs at different biological levels is of general importance, in order to provide
health and environment professionals, as well as technologists, with new data considering the
25

different structures and molar ratios of DESs in the modulation of its features as well as in
safety of its application.
4. Conclusion
Based on the presented results, we can conclude that DESs could be excellent solvents for
lipase-catalyzed reactions in terms of enzyme performance and product yield. However, HDB
and the water content in the latter are the most curtail parameters and should be carefully
optimized for each reaction system. Ping-Pong Bi-Bi reaction mechanism was shown to be
suitable for description of reactions taking place in DES:water mixtures, consequently this
research could contribute to better understanding of reaction mechanism within these new
solvents.
Given results indicate that substrate(s) incorporation in DES H-bond net could pose a serious
problem for a variety of enzyme-catalyzed processes and should be handled individually,
depending on both enzyme and reaction type (herein a simple addition of water proved to be
effective enough). On the other hand, the possibility of product entrapment in the same H-
bond net could be very interesting from a point of chemical equilibrium, since the product
formed could be virtually removed from the enzyme surrounding, shifting the chemical
equilibrium towards product formation.
Finally, if we consider other environmental benefits of applying DESs in a role of solvents,
such as low vapour pressure (reduced air pollution), non-flammability (process safety),
nontoxicity, as well as the low cost of raw materials and techniques used to create the
solvents, DESs represent excellent candidates for environment friendly biocatalysis.
26

Authors’ contributions
MCB and IRR designed the main experimental work. MCB carried the experimental work
and drafted the manuscript. AJT performed mathematical modelling, while MV performed
NMR studies. KR participated in the cytotoxicity studies with VGS. All authors together
finalized the manuscript and approved it.
Acknowledgements
The work was supported by the University of Zagreb (Grant no. 0582261-7002223). The
authors are also thankful to dr. Aleksandra Sander and dr. Tomislav Penović for performing
Karl-Fischer titrations.
27

References
[1] M. Cvjetko Bubalo, S. Vidović, I. Radojčić Redovniković, S. Jokić, J. Chem. Technol.
Biotechnol. 2015. doi: 10.1002/jctb.4668.
[2] M. Cvjetko Bubalo, K. Radošević, I. Radojčić Redovniković, J. Halambek, V. Gaurina
Srček, Ecotox. Environ. Safe. 99 (2014) 1-12.
[3] L.C. Tomé, D.J. Patinha, R. Ferreira, H. Garcia, C. Silva Pereira, C.S. Freire, L.P.
Rebelo, I.M. Marrucho, ChemSusChem 7 (2014) 110-113.
[4] X.D. Hou, J. Xu, N. Li, M.H. Zong, Biotechnol. Bioeng. 112 (2015) 65-73.
[5] X.D. Hou, Q.P. Liu, T.J. Smith, N. Li, M.H. Zong, PLoS One 8 (2013) e591452013.
[6] W. Gouveia, T.F. Jorge, S. Martins, M. Meireles, M. Carolino, C. Cruz, T.V. Almeida,
M.E.M. Araújo, Chemosphere, 104 (2014) 51-56.
[7] B. Tang, K.H. Row, Monatsh. Chem 144 (2013) 1427-1454.
[8] P. Paiva, R. Craveiro, I. Aroso, M. Martins, R.L. Reis, A.R.C.Duarte, ACS Sustainable
Chem. Eng. 2 (2014) 1063–1071.
[9] E. Durand, J. Lecomte, P. Villeneuve, Eur. J. Lipid. Sci. Technol. 115 (2013) 379-385.
[10] R. Madeira Lau, F. van Rantwijk, K.R. Seddon, R.A. Sheldon, Org. Lett. 2 (2000) 4189-
4191.
[11] J.L. Kaar, A.M. Jesionowski, J.A. Berberich, R. Moulton, A.J. Russel, J. Am. Chem. Soc.
125 (2003) 4125–4131.
[12] E. Durand, J. Lecomte, B. Baréa, P. Villeneuve, Eur. J. Lipid. Sci. Technol. 116 (2014)
16-23.
[13] T. de Diego, P. Lozano, M.A. Abad, K. Steffensky, M. Vaultier, JL. Iborra, J.
Biotechnol. 140 (2009) 234–241.
28

[14] M. Cvjetko, J. Vorkapić-Furač, P. Žnidaršič-Plazl, Process Biochem. 47 (2012) 1344-
1350.
[15] F.J. Deive, D. Ruivo, J.V. Rodrigues, C.M. Gomes, M.Á. Sanromán, L.P.N. Rebelo, J.M.
S. S. Esperança, A. Rodríguez. RSC Adv. 5 (2015) 3386-3389.
[16] J.T. Gorke, F. Srienc, R.J. Kazlauskas, Chem. Commun. 10 (2008) 1235–1237.
[17] A. Hayyan, M.A. Hashim, M. Hayyan, F.S. Mjalli, I.M. Alnashef, J Clean Prod. 65
(2014) 246-251.
[18] A Hayyan, MA Hashim, M Hayyan, FS Mjalli, IM AlNashef, Ind Crop Prod. 46 (2013)
392-398.
[19] H. Zhao, C. Zhang, T.D. Crittle, J. Mol. Catal. B-Enzym. 85-86 (2013) 243-47.
[20] E. Durand, J. Lecomte, B. Baréa, G. Piombo, E. Dubreucq, P. Villeneuve, Process
Biochem. 47 (2012) 2081-2089.
[21] E. Durand, J. Lecomte, B. Baréa, E. Dubreucq, R. Lortie, P. Villeneuve, Green Chem. 15
(2013) 2275-2282.
[22] C. Aouf, E. Durand, J. Lecomte, M.C. Figueroa-Espinoza, E. Dubreucq, H. Fulcranda, P.
Villeneuve, Green Chem. 16 (2014) 1740-1754.
[23] A.A. Papadopoulou, M.H. Katsoura, A. Chatzikonstantinou, E. Kyriakou, A.C. Polydera,
A.G. Tzakos, H. Stamatis, Bioresource Technol. 136 (2013) 41-48.
[24] M.D. Romero, L. Calvo, C. Alba, A. Daneshfar, H.S. Ghaziaskar, Enzyme Microb.
Technol. 37 (2005) 42–48.
[25] M.D. Romero, L. Calvo, C. Alba, M. Habulin, M. Primožič, Knez, J. Supercrit. Fluids 33
(2005) 77-84.
[26] H.S. Kim, Y.-M. Koo, Korean J. Chem. Eng. 29 (2012) 1610-1614.
[27] B. Gabrić, A. Sander, M. Cvjetko Bubalo, D. Macut, Sci. World J. 2013 (2013) 1-11.
29