Disubstituted dialkoxysilane precursors in binary and ternary sol-gel systems for lipase immobilization

Article abstract of DOI:10.1016/j.procbio.2011.11.023

The disubstituted dimethyldiethoxysilane (DMDEOS), methyl(phenyl) diethoxysilane (MPDEOS) and diphenyldiethoxysilane (DPDEOS) were used in binary silane precursor systems in combination with tetraethoxysilane (TEOS) for the immobilization of lipase from Pseudomonas fluorescens (Lipase AK). In addition, ternary silane precursor systems with TEOS and octyltriethoxysilane (OTEOS) or phenyltriethoxysilane (PTEOS) were also studied for encapsulation. The best performing ternary sol-gel preparations (418-736% activity yields of the immobilized enzyme with 1-phenylethanol rac-1a as compared to the native form) were tested as biocatalysts for kinetic resolutions of rac-1a, 1-phenylpropan-2-ol rac-1b and 4-phenylbutan-2-ol rac-1c. Because the catalytic properties and the operational stability of the DMDEOS-containing preparations proved to be superior to all the tested free and sol-gel entrapped Lipase AK biocatalysts in batch mode, the kinetic resolutions of rac-1a and rac-1b were performed with the TEOS/PTEOS/DMDEOS 4:1:1 Lipase AK in a continuous-flow reactor as well.

Full text of DOI:10.1016/j.procbio.2011.11.023

Process Biochemistry 47 (2012) 428–434
Contents lists available at SciVerse ScienceDirect
Process Biochemistry
journal homepage: www.elsevier.com/locate/procbio
Disubstituted dialkoxysilane precursors in binary and ternary sol–gel systems for
lipase immobilization
Diána Weiser, Zoltán Boros, Gábor Hornyánszky, Alexandra Tóth, László Poppe∗
Department of Organic Chemistry and Technology and Research Group for Alkaloid Chemistry, Budapest University of Technology and Economics, H-1111 Budapest,
Mu˝ egyetem rkp. 3, Hungary
a r t i c l e i n f o
a b s t r a c t
Article history:
The disubstituted dimethyldiethoxysilane (DMDEOS), methyl(phenyl)diethoxysilane (MPDEOS) and
diphenyldiethoxysilane (DPDEOS) were used in binary silane precursor systems in combination with
tetraethoxysilane (TEOS) for the immobilization of lipase from Pseudomonas fluorescens (Lipase AK).
In addition, ternary silane precursor systems with TEOS and octyltriethoxysilane (OTEOS) or phenyl-
triethoxysilane (PTEOS) were also studied for encapsulation. The best performing ternary sol–gel
preparations (418–736% activity yields of the immobilized enzyme with 1-phenylethanol rac-1a
as compared to the native form) were tested as biocatalysts for kinetic resolutions of rac-1a, 1-
phenylpropan-2-ol rac-1b and 4-phenylbutan-2-ol rac-1c. Because the catalytic properties and the
operational stability of the DMDEOS-containing preparations proved to be superior to all the tested
free and sol–gel entrapped Lipase AK biocatalysts in batch mode, the kinetic resolutions of rac-1a and
rac-1b were performed with the TEOS/PTEOS/DMDEOS 4:1:1 Lipase AK in a continuous-flow reactor as
well.
Received 18 October 2011
Received in revised form
20 November 2011
Accepted 25 November 2011
Available online 6 December 2011
Keywords:
Lipase
Immobilization
Sol–gel
Disubstituted dialkoxysilane
Kinetic resolution
Continuous-flow reactor
© 2011 Elsevier Ltd. All rights reserved.
1. Introduction
and glycerol [15] at the lipid–water interface [16,17]. In syn-
thetic processes, lipases can catalyze a wide range of enantio- and
Nowadays enzyme catalyzed transformations have an ever
increasing importance in modern synthetic processes [1–4]. Syn-
thetic application of isolated enzymes being native proteins suffers
several drawbacks. Proteins are relatively unstable in water and
the recovery of the enzymes from aqueous reaction mixtures may
be difficult due to the high solubility of proteins in water. Conse-
quently the loss or the tedious recovery of the enzymes may render
the process uneconomical. Immobilized biocatalysts were defined
as ‘enzymes, cells or organelles (or combinations of these) which
are in a state that permits their reuse’ [5]. Immobilized enzymes
may show enhanced stability, activity and selectivity as compared
to their native form. Because the immobilized biocatalysts are recy-
clable, storable and easy to handle, immobilization of enzymes is
an important trend in biotechnology [1,6–11].
By far the most often used biocatalysts are lipases (triacyl-
glycerol hydrolases, EC 3.1.1.3). They are exploited as cheap and
versatile biocatalysts in several applications [12,13], among them
in bakery and as laundry detergents or even in alternative energy
strategies to convert vegetable oil into biofuel [14]. In nature,
lipases catalyze the hydrolysis of triglycerides into fatty acids
regioselective reactions such as hydrolysis, esterification, transes-
terification, aminolysis and ammoniolysis [18–20].
Among the multitude of immobilization methods, including
adsorption, covalent attachment to solid supports and entrap-
ment within polymers [7], the entrapment of enzymes in
inorganic/organic hybrid polymer matrices has received in recent
years much attention and has provided new opportunities in
the field of material science [21–25]. Sol–gel encapsulation has
proven to be a particularly easy and effective way to immobi-
lize purified enzymes and other proteins or even whole cells
[23,26–28]. The main advantages of sol–gel immobilization are
enhanced thermostability and long-term operational stability
[29,30].
In a typical sol–gel immobilization process following acid-
or base-catalyzed hydrolysis of a tetraalkoxysilane precursor
[Si(OR)4], polycondensation can create a matrix in which the
enzyme can be encapsulated [24]. Using organically modified
silanes RꢀSi(OR)3 in admixture with Si(OR)4, with increasing the
alkyl chain length (Rꢀ) and the amount of the hydrophobic silanes
activity enhancement of the immobilized lipases could be observed
[31–34]. Furthermore, a comparison of binary sol–gel systems
consisting Si(OR)4 and trialkoxysilanes RꢀSi(OR)3 with ternary
sol–gel systems consisting Si(OR)4 and two different trialkoxysi-
lanes RꢀSi(OR)3 revealed advantages of the latter [35].
∗
Corresponding author. Tel.: +36 1 463 3299; fax: +36 1 463 3697.
E-mail address: poppe@mail.bme.hu (L. Poppe).
1359-5113/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.procbio.2011.11.023

D. Weiser et al. / Process Biochemistry 47 (2012) 428–434
429
Table 1
Table 3
Biocatalytic properties of Lipase AK entrapped in binary sol–gel systems (1:1 molar
ratio of the silane precursor/TEOS; with rac-1a, reaction time 4 h in n-hexane:THF
2:1).
Change of the selectivity of the various Lipase AK biocatalysts in the course of
repeated kinetic resolutions of rac-1a (results of the 1st and 7th 4 h long runs in
n-hexane:MTBE 2:1 are shown).
Entry
Silane precursors
c (%)
ee(R)-2a (%)
E
UB (U g−1
)
YA (%)
Mode of immobilization
Run 1
Run 7
a
1
2
3
4
5
6
7
–
29.7
27.5
30.6
49.4
14.4
28.9
47.2
97.9
98.3
98.3
98.3
98.8
98.8
99.3
>100
10.0
9.1
10.5
17.0
5.0
100
350
360
374
750
900
994
ee(R)-2a (%)
E
ee(R)-2a (%)
E
TEOS
ꢁ200
>100
ꢁ200
>100
>200
ꢁ200
a
–
92.1
99.3
99.4
99.4
98.6
>100
>200
>200
88.9
99.2
99.0
99.4
98.4
32
>200
>200
TEOS/PTEOS
TEOS/OTEOS
TEOS/DPDEOS
TEOS/MPDEOS
TEOS/DMDEOS
iBTMOS/TMOS 5:1
TEOS/DMDEOS 1:1
TEOS/PTEOS/DMDEOS 4:1:1
TEOS/OTEOS/DMDEOS 4:1:1
ꢁ200
ꢁ200
8.2
16.3
ꢁ200
>200
a
Crude Lipase AK powder.
a
Crude Lipase AK powder.
For
enzyme
immobilization
mainly
mixtures
of
(iBTMOS), phenyltriethoxysilane (PTEOS), n-octyl-triethoxysilane (OTEOS),
dimethyldiethoxysilane (DMDEOS), methyl(phenyl)diethoxysilane (MPDEOS),
diphenyldiethoxysilane (DPDEOS), 2-propanol (IPA) and vinyl acetate were com-
mercial products of Alfa Aesar, Fluka, Sigma–Aldrich or Merck. Solvents [n-hexane,
acetone, dichloromethane, methyl t-butyl ether (MTBE)] from Molar Chemicals
were freshly distilled prior to use.
dimethyldimethoxysilane (DMDMOS) and tetramethoxysilane
(TMOS) in binary sol–gel systems were investigated [36–39]. In
such systems methanol is liberated, harmful for many enzymes
[40]. Although dimethyldiethoxysilane (DMDEOS) was used to pre-
pare biosensors based on sol–gel entrapment of enzymes [41,42],
disubstituted diethoxysilanes in combination with tetraethoxysi-
lane (TEOS) resulting in milder polymer network formation due
to ethanol liberation were never applied for the preparation of
sol–gel immobilized biocatalysts.
As fillings in packed-bed bioreactors crude lipase powders were
applied in several cases [43] but in continuous-flow systems mostly
immobilized lipases were used [20,43–47]. In packed-bed reac-
tors the native enzymes proved, however, to be less stable than
the robust immobilized lipases. Therefore, the use of immobilized
enzymes in continuous-flow mode is economically attractive [43].
It was expected that disubstituted diethoxysilane precursors
may significantly affect porosity and hydrophobicity of the result-
ing polymeric network. Therefore, in this study our goal was to
prepare novel immobilized lipase biocatalysts entrapped in binary
and ternary sol–gel systems containing disubstituted diethoxysi-
lane precursors and investigate the biocatalytic properties of the
different preparations in kinetic resolutions of secondary alcohols
both in batch and continuous-flow mode.
2.2. Sol–gel immobilization of Lipase AK using binary or ternary silane precursor
systems
Tris–HCl buffer (0.05 M, pH 7.5, 390 ␮L), PEG solution (4%, w/v, 200 ␮L) and IPA
(200 ␮L) were added to a 20 mL glass vial and the mixture was shaken at 1000 rpm
at room temperature for 10 min. Lipase AK (50 mg), the mixture of silane precur-
sors (3 mmol, compositions see in Tables 1–3) and NaF solution (1 M, 100 ␮L) were
added while continuous shaking was maintained. To complete the polymerization
process the mixture was shaken for 12 h at room temperature. The solid formed was
washed with IPA (7 mL), distilled water (5 mL), IPA (5 mL) and n-hexane (5 mL). The
resulting white powder was dried at room temperature for 24 h and stored at room
temperature.
2.3. Evaluation of the catalytic properties of the sol–gel Lipase AK biocatalysts
The free or sol–gel immobilized Lipase AK (50 mg) was added to the solu-
tion of the racemic test alcohol (rac-1a–c, 50 ␮L) and vinyl acetate (100 ␮L) in
n-hexane:THF 2:1 (1 mL), and the resulting mixture was shaken at 30 ◦C in a
sealed glass vial at 1000 rpm. For GC analyses, samples were taken directly from
the reaction mixture (sample size: 10 ␮l, diluted with CH2Cl2 to 100 ␮L) after
2, 4 and 8 h. The reaction mixtures were analyzed on an Agilent 4890 instru-
ment equipped with flame ionization detector (FID, 250 ◦C) and an Hydrodex
␤-6TBDM column [25 m × 0.25 mm × 0.25 ␮m film of heptakis-(2,3-di-O-methyl-
6-O-t-butyldimethylsilyl)-␤-cyclodextrin], injector temperature 250 ◦C, carrier gas
2. Materials and methods
H2, head pressure 12 psi, split ratio 1:50. Oven programs: for rac-1a and 2a – 120 ◦
C
2.1. Materials
isothermal, tR (min): 4.18 [(S)-2a], 4.64 [(R)-2a], 6.02 [(R)-1a], 6.36 [(S)-1a]; for rac-
1b and 2b – 80 ◦C (5 min), 80–120 ◦C (2 ◦C min−1), tR (min): 20.63 [(S)-2b], 20.87
[(S)-1b], 21.42 [(R)-1b], 22.61 [(R)-2b]; for rac-1c and 2c – 170 ◦C isothermal, tR
(min): 2.25 [rac-1c], 2.45 [(S)-2c], 2.55 [(R)-2c]. Data on conversion and enantiomeric
composition of the products are collected in Tables 1–4.
Lipase AK (from Pseudomonas fluorescens, Sigma–Aldrich Cat. No 534730),
racemic 1-phenylethanol (rac-1a), 1-phenylpropan-2-ol (rac-1b), and 4-
phenylbutan-2-ol (rac-1c), sodium fluoride (NaF), polyethylene glycol 1000 (PEG),
tetraethoxysilane (TEOS), tetramethoxysilane (TMOS), i-butyltrimethoxysilane
Table 2
Biocatalytic properties of Lipase AK entrapped in different ternary sol–gel systems tested by kinetic resolution of rac-1a in n-hexane:THF 2:1 at 4 h.
Entry
1
Molar ratio
c (%)
ee(R)-2a (%)
E
UB (U g−1
)
YA (%)
a
–
30.0
97.9
>100
10.0
100
Panel A
TEOS/OTEOS/DMDEOS
2
3
4
5
1:1:1
4:1:1
1:4:1
1:1:4
46.9
48.2
51.8
50.4
96.4
98.4
90.3
95.1
>200
>200
>100
>200
16.1
16.7
17.8
17.3
304
578
250
278
Panel B
TEOS/PTEOS/DMDEOS
6
7
8
9
1:1:1
4:1:1
1:4:1
1:1:4b
50.7
46.1
50.7
94.0
98.3
93.8
>100
>200
>200
17.4
16.2
17.4
329
736
245
b
b
b
b
b
–
–
–
–
–
Panel C
10
11
12
13
TEOS/OTEOS/PTEOS
1:1:1
4:1:1
1:4:1
1:1:4
52.4
44.9
52.3
50.6
89.7
97.8
89.5
95.4
>100
>200
>100
>200
18.0
15.4
18.0
17.4
264
418
325
244
a
Crude Lipase AK powder.
No polymer matrix formation.
b

430
D. Weiser et al. / Process Biochemistry 47 (2012) 428–434
Table 4
Biocatalytic properties of Lipase AK entrapped in the best ternary sol–gel systems tested by kinetic resolution of rac-1b (at 2 h: panel A) and of rac-1c (at 4 h: panel B) in
n-hexane:THF 2:1.
Substrate/entry
Silane precursors
Molar ratio
c (%)
ee(R)-2b (%)
ee(S)-1b (%)
E
UB (U g−1
)
YA (%)
rac-1b
a
1
2
3
4
–
36.5
7.1
7.2
99.1
99.2
99.6
98.5
55.6
6.3
6.7
>200
>200
ꢁ200
>100
25.2
4.9
4.9
100
32
75
TEOS/OTEOS/PTEOS
TEOS/OTEOS/DMDEOS
TEOS/PTEOS/DMDEOS
4:1:1
4:1:1
4:1:1
2.0
1.1
1.4
20
Substrate/entry
Silane precursors
Molar ratio
c (%)
ee(R)-2c (%)
ee(S)-1c (%)
E
UB (U g−1
)
YA (%)
rac-1c
a
b
5
6
7
8
–
42.3
10.9
12.2
3.3
98.5
97.8
99.4
98.6
–
>200
>100
>200
>100
14.6
3.8
4.2
100
41
104
29
b
TEOS/OTEOS/PTEOS
TEOS/OTEOS/DMDEOS
TEOS/PTEOS/DMDEOS
4:1:1
4:1:1
4:1:1
–
b
–
b
–
1.1
a
Crude Lipase AK powder.
No enantioseparation.
b
Conversion (c), enantiomeric excess (ee) and enantiomer selectivity
this work, dimethyldiethoxysilane (DMDEOS) in combination with
TEOS was studied in binary and ternary sol–gel systems for immo-
bilization of Lipase AK. In addition, the behavior of two additional
disubstituted diethoxysilanes (MPDEOS and DPDEOS) in binary
sol–gel systems were tested.
(E) were determined for all test reactions. In case of rac-1a,b enan-
tiomer selectivity (E) was calculated from eeS and eeP using the equation
E = ln[(1 − eeS)/(1 + eeS/eeP)]/ln[(1 + eeS)/(1 + eeS/eeP)] [48]. For the kinetic resolution
of rac-1c enantiomer selectivity (E) was determined from c and eeP using the equa-
tion E = ln[1 − c(1 + eeP)]/ln[1 − c(1 − eeP)] [49]. Due to sensitivity to experimental
errors, E values calculated in the 100–200 range are given as >100, in the 200–500
range as >200 and above 500 as ꢁ200.
The specific biocatalyst activity (UB) and activity yield (YA
) were deter-
3.1. Comparison of disubstituted diethoxysilane/TEOS binary
sol–gel systems to monosubstituted triethoxysilane/TEOS binary
systems
mined in the batch reactions using the equations UB = (nrac × c)/(t × mB) and
YA = 100 × UB/[(nrac × c)/(t × mE)] (where nrac is moles of racemic alcohol, c the con-
version, mB the mass of the immobilized biocatalyst and mE the mass of enzyme
preparation subjected to immobilization) [35]. To characterize the productivity of
the biocatalysts in continuous-flow systems, the specific reaction rates (rflow) were
also calculated using the equation rflow = [P] × v/mB (where [P] is the molar con-
centration of the product, v the flow rate and mB the mass of the immobilized
biocatalyst) [43].
The use of disubstituted dialkoxysilanes with different sub-
stituents combined with TEOS as silane precursors may signif-
icantly influence the porosity and the hydrophobicity of the
sol–gel matrices obtained which may play a significant role in
the final biocatalytic properties of the entrapped lipases. Higher
hydrophobicity may better stabilize the catalytically active open
conformation of the lipase, whereas higher porosity of polymer
matrix may cause a lower diffusion limit for the substrate and prod-
uct transfer. Thus the access of substrate to the active site of the
lipase may be facilitated and therefore biocatalytic efficiency of
the immobilized enzyme can be enhanced. To test these effects,
three different disubstituted dialkoxysilanes [dimethyldiethoxysi-
lane (DMDEOS), methyl(phenyl)diethoxysilane (MPDEOS) and
diphenyldiethoxysilane (DPDEOS)] in admixture with TEOS at a 1:1
molar ratio were tried for forming entrapment matrices to immobi-
lize Lipase AK and compared to the corresponding binary systems
with PTEOS and OTEOS (Table 1). As expected, the activity of Lipase
AK improved in all of these sol–gel matrices (YA > 100%), while the
high degree of selectivity was retained. Particularly, the activity
yield of the biocatalysts containing disubstituted silane precur-
sors (entries 5–7: YA = 750–994%) was significantly higher than
those with monosubstituted precursors (entry 3, 4: YA = 360–374%)
while the high enantiomer selectivity (E > 100) was maintained.
Taking into account also the specific activity, the DMDEOS con-
taining sol–gel matrix was much more efficient (UB = 16.3 U g−1 for
DMDEOS/TEOS system vs. UB = 5.0 and 8.2 U g−1 for DPDEOS/TEOS
and MPDEOS/TEOS systems, respectively). Although the activity
yields of entrapped Lipase AK in biocatalysts from disubstituted
silane precursors containing one or two phenyl groups (YA = 900%
and 750%) are comparable to the system with DMDEOS (YA = 994%),
the selectivities of the phenyl-containing disubstituted silane sys-
tems were lower (E < 500 for DPDEOS and MPDEOS vs. E > 500 for
DMDEOS). In addition, the gelation behavior of phenyl-containing
precursor systems was much inferior to those with DMDEOS since
bulkiness of DPDEOS and MPDEOS may hinder polymerization. Tak-
ing into account also the higher price of DPDEOS and MPDEOS, only
the DMDEOS-containing sol–gel systems were studied further.
2.4. Reuse of the biocatalysts
Biocatalyst (50 mg from the native Lipase AK, or 50 mg from the sol–gel prepa-
ration) was added into the solution of racemic 1-phenylethanol (rac-1a, 50 ␮L) and
vinyl acetate (100 ␮L) in n-hexane:MTBE 2:1 (1 mL), and the resulting mixture was
shaken at 30 ◦C in a sealed glass vial at 1000 rpm for 4 h. The reactions with each
biocatalyst were repeated 7 times with the same enzyme preparation, under iden-
tical conditions (biocatalyst concentration: 50 mg mL−1; substrate and vinyl acetate
concentrations: 50 ␮L mL−1 and 100 ␮L mL−1, respectively). After each cycle the
products were analyzed as described in Section 2.3. The catalyst was washed with
the same amount of solvent (2× 1 mL of n-hexane:MTBE 2:1) and dried for 24 h at
RT.
2.5. Kinetic resolutions in continuous-flow bioreactors
The continuous-flow kinetic resolutions were performed in a laboratory flow
reactor [Knauer K-120 isocratic HPLC pump attached to CatCart® columns (stainless
steel, inner diameter: 4 mm; total length: 70 mm; packed length: 65 mm; inner vol-
ume: 0.816 mL) filled with immobilized sol–gel Lipase AK (double Lipase-AK loading
was used in a TEOS/PTEOS/DMDEOS 4:1:1 precursor system [50], mB = 367 mg)].
Before use, the columns were washed with an n-hexane:MTBE 2:1 mixture. The
runs were performed at 30 ◦C and 0.2–0.6 mL min−1 flow rate in 0.1 mL min−1 steps.
Samples were collected during stationary operation (30 min after changing the
parameters) and analyzed as described in Section 2.3.
3. Results and discussion
Although it can be expected that disubstituted dialkoxysi-
lanes RꢀRꢀꢀSi(OR)2 as precursors of sol–gel matrices may result
in higher hydrophobicity and/or higher porosity, so far only the
DMDMOS/TMOS sol–gel system was tested for the immobiliza-
tion of hydrolases [36–39]. A recent study indicated the beneficial
properties of binary and ternary sol–gel systems of TEOS and
different triethoxysilane precursors for the immobilization of
Celite® 545 supported Lipase AK [35] but the usefulness of disub-
stituted diethoxysilanes in formation of binary or ternary sol–gel
matrices for the entrapment of lipases has not been reported. In

D. Weiser et al. / Process Biochemistry 47 (2012) 428–434
431
3.2. Using ternary sol–gel biocatalysts in kinetic resolution of
racemic 1-phenylethanol (rac-1a)
last cycle (Fig. 2). The enantiomer selectivity was also signif-
icantly reduced from E > 100 to E = 32 (Table 3). On the other
hand, all of our sol–gel preparations exhibited much better dura-
bility in repeated reaction cycles. Not only were the average
mass losses of the sol–gel-AK biocatalysts significantly lower
(∼10% in each cycle), but they retained much more of their
original specific activity. In contrast to the UB values of a “con-
ventional” sol–gel Lipase AK (iBTMOS/TMOS 5:1 [34]) dropped
from 6.3 U g−1 to 5.5 U g−1 (Fig. 2), the UB values of the DMDEOS
based catalysts were originally higher (TEOS/DMDEOS 1:1 Lipase
Our recent study with Celite® 545 supported Lipase AK immo-
bilized in sol–gel systems indicated that catalytic properties of the
ternary sol–gel systems of TEOS and two different triethoxysilane
precursors were superior to any of the binary sol–gel systems stud-
ied so far [35]. Consequently, it was thought worth to study the
DMDEOS precursor in three-component systems as well (Table 2).
For this reason, the so far two best performing monosubstituted
triethoxysilanes OTEOS and PTEOS [35] were selected as additional
precursors. Because the ternary sol–gel systems proved to be suit-
able for fine-tuning of the biocatalytic properties of lipases [35], two
ternary sol–gel systems of TEOS and DMDEOS combined with either
OTEOS or PTEOS (Table 2, panels A and B, respectively) were stud-
ied. For comparison, the TEOS/OTEOS/PTEOS ternary system lacking
a disubstituted silane precursor was also investigated (Table 2,
panel C). Within each ternary system, four different molar ratios of
the three silane precursors were applied. Because the disubstituted
silane precursor may have larger influence on matrix porosity than
monosubstituted trialkoxysilanes, we expected that the DMDEOS
containing systems can have a more pronounced influence on the
biocatalytic properties of the entrapped lipase than it was observed
in our previous systems without disubstituted silane precursor
[35].
In fact, all ternary sol–gel Lipase AK biocatalysts exhibited
higher specific biocatalyst activity (UB > 15.4 U g−1) than the free
Lipase AK (UB = 10.0 U g−1) (Table 2) due to the at least 2.45-fold
activity enhancements (entries 2–13: YA > 245%) as compared to
the non-immobilized Lipase AK (entry 1). In parallel to the high
specific activities, selectivities were maintained (E > 100) or even
improved (E > 200) as compared to non-immobilized Lipase AK
(E > 100). Within a given ternary system (panels A–C of Table 2),
biocatalytic properties significantly depended on the ratio of the
components. The YA value of the entrapped lipase was much depen-
dent on nature of the sol–gel matrix. Notably, the optimal molar
ratio of the TEOS and the two other “matrix modifying” silane pre-
cursors was the same within the three panels (4:1:1; entries 3,
7 and 11). However, the matrices containing DMDEOS (panels A
and B of Table 2) were even more effective than the ones con-
taining just monosubstituted triethoxysilanes (panel C of Table 2).
The advantage of DMDEOS as a silane precursor in forming the
sol–gel network for entrapment of Lipase AK was clearly demon-
strated by the mixture of TEOS/DMDEOS (entry 6, Table 1) as the
best combination among the binary systems tested and by the
TEOS/PTEOS/DMDEOS 4:1:1 or TEOS/OTEOS/DMDEOS 4:1:1 com-
binations (entries 3 and 7 in Table 2) as the best two among the
ternary sol–gel Lipase AK biocatalysts.
AK: 17.2 U g−1; TEOS/PTEOS/DMDEOS 4:1:1 Lipase AK: 17.1 U g−1
;
TEOS/OTEOS/DMDEOS 4:1:1 Lipase AK 17.0 U g−1; see Fig. 2) and
their activity remained unchanged during recycling. In contrast
to the free enzyme, the selectivity values of the sol–gel immobi-
lized preparations decreased only slightly during the reuse test
(Table 3).
Unaltered conversion (c = 60.5%) and enantiomeric compo-
sitions [ee = 76.2% for the (R)-acetate formed and ee = 99.2%
for the residual (S)-alcohol] in a continuous-flow kinetic res-
olution of racemic 1-(thiophen-2-yl)ethanol (5 mg mL−1, 30 ◦C,
n-hexane:MTBE:vinyl acetate 6:3:1 mixture at 0.2 mL min−1 flow
rate) over 24 h in a parallel study [50] indicated the durability of the
TEOS/PTEOS/DMDEOS 4:1:1 Lipase AK in continuous-flow reactor
as well.
It is obvious from the reusability tests that sol–gel immobiliza-
tion can significantly enhance the stability of lipase and thus may
be advantageous in industrial applications.
3.4. Performance of the best ternary sol–gel biocatalysts in
kinetic resolution of 1-phenylpropan-2-ol (rac-1b) and
4-phenylbutan-2-ol (rac-1c)
Because the sol–gel immobilized Lipase AK with DMDEOS as
precursor component gave promising results in kinetic resolution
of rac-1a, it was obvious to extend our studies to further secondary
alcohols i.e. rac-1b and rac-1c (Table 4). For this the best members
of the three different ternary sol–gel systems (4:1:1 molar ratio of
TEOS and the two other precursors) were chosen.
Whereas with rac-1a the phenyl- and octyl-containing ternary
sol–gel Lipase AK biocatalysts were similarly selective (E > 200
in entries 3, 7 and 11 in Table 2), in case of rac-1b the var-
ious systems exhibited different selectivities varying between
E > 100 for the TEOS/PTEOS/DMDEOS 4:1:1 system (entry 4 in
Table 4) and E ꢁ 200 for the TEOS/OTEOS/DMDEOS 4:1:1 system
(entry 3 in Table 4). Moreover, the specific biocatalyst activi-
ties were also in favor of the octyl-containing system (Table 4:
UB = 4.9 U g−1 for TEOS/OTEOS/DMDEOS 4:1:1 AK vs. UB = 1.4 U g−1
for TEOS/PTEOS/DMDEOS 4:1:1 AK). Our ternary sol–gel biocata-
lysts showed similar behavior in the kinetic resolution of rac-1c
indicating that the TEOS/OTEOS/DMDEOS 4:1:1 AK system has
the best biocatalytic properties (entry 7 in Table 4: E > 200 with
UB = 1.4 U g−1).
The differences in kinetic resolutions of the three racemic
test alcohols (rac-1a–c) can be rationalized by considering the
nature of the sol–gel matrix. With 1-phenylethanol (rac-1a) having
the most aromatic character, TEOS/PTEOS/DMDEOS 4:1:1 Lipase
AK preparation containing an aromatic silane precursor exhib-
ited the best biocatalytic properties, whereas with the more
aliphatic 1-phenylpropan-2-ol (rac-1b) or 4-phenylbutan-2-ol
(rac-1c) TEOS/OTEOS/DMDEOS 4:1:1 Lipase AK system containing
an aliphatic silane-precursor seemed to be the best. Thus a phenyl
group containing sol–gel matrix was favorable for the more aro-
matic substrates while a sol–gel matrix containing only aliphatic
chain was more efficient for the substrates of more pronounced
aliphatic character.
3.3. Reusability of Lipase AK biocatalysts
For practical applications of immobilized enzymes, reusability
is of foremost importance. For testing the recyclability of our best
DMDEOS containing sol–gel biocatalysts, i.e. the TEOS/DMDEOS
1:1; TEOS/PTEOS/DMDEOS 4:1:1 and TEOS/OTEOS/DMDEOS 4:1:1
Lipase AK biocatalysts were compared to native Lipase AK
and a conventional binary sol–gel Lipase AK (iBTMOS/TMOS
5:1 [34]) in repeated kinetic resolutions of rac-1a. Each bio-
catalyst was reused
7 times for 4 h and after regeneration
the change of their specific biocatalyst activity (UB, results in
Fig. 2) and selectivity (ee(R)-2a and E, results in Table 3) was
recorded.
We found that native Lipase AK not only suffered from the
highest mass lost (∼25% average mass lost in each cycle), the
specific activity of native Lipase AK decreased also continuously
from UB = 12.0 U g−1 in the first cycle to UB = 9.8 U g−1 in the

432
D. Weiser et al. / Process Biochemistry 47 (2012) 428–434
OH
vinyl acetate
p
n
rac-1a,b
rac-1a-c
sol-gel AK
vinyl acetate
rac-1a-c
T
O
O
n
(R)-2a-c
+
n
1,2
OH
a
b
c
0
1
2
n
(S)-1a-c
Fig. 1. Kinetic resolutions for testing of immobilized Lipase AK preparations.
3.5. Application of TEOS/PTEOS/DMDEOS 4:1:1 Lipase AK in a
continuous-flow reactor
Fig. 3. Effect of flow rate (v) on conversion [c (%): ᭹] and productivity [rflow (U g−1):
ꢀ] of kinetic resolutions of secondary alcohols in a continuous-flow reactor filled
with sol–gel Lipase AK (TEOS/PTEOS/DMDEOS 4:1:1). (A) Kinetic resolution of rac-1a
(E > 100 for all points); (B) kinetic resolution of rac-1b (E > 100 for all points).
Since
in
batch
mode
the
lipase
entrapped
in
TEOS/PTEOS/DMDEOS 4:1:1 system exhibited optimal prop-
erties in the kinetic resolution of rac-1a–c, this system was
selected for use in continuous-flow reactor (Fig. 1). In order to
obtain an immobilized lipase with higher specific activity, we dou-
bled the enzyme loading [50]. Since with TEOS/PTEOS/DMDEOS
4:1:1 Lipase AK sufficient enantiomer selectivity (E > 100) could
only be realized for rac-1a (entry 7, Table 2) and rac-1b (entry 4,
Table 3), our studies in the continuous-flow system were restricted
to these two substrates. In accordance with the previous results
[43], the high enantiomer selectivities observed in batch mode
were retained also in the continuous-flow processes. This was the
case in n-hexane:MTBE 2:1 solvent system [E > 100 for rac-1a and
E > 200 for rac-1b, allowing the preparation of (R)-2a and (R)-2b
in >97%ee, see Fig. 3], which is industrially acceptable and more
18
U B
(U g-1)
16
14
12
10
8
6
4
2
0
1
2
3
4
5
6
7
# Run
Fig. 2. Reusability of the various Lipase AK biocatalysts in batch kinetic resolutions of rac-1a over a week. Bars representing UB (U g−1) of the various Lipase AK biocatalysts in
subsequential 4 h reaction cycles: (ꢁ) free powder; (ꢂ) iBTMOS/TMOS 5:1 [34]; ( ) TEOS/DMDEOS 1:1; ( ) TEOS/PTEOS/DMDEOS 4:1:1; ( ) TEOS/OTEOS/DMDEOS 4:1:1.

D. Weiser et al. / Process Biochemistry 47 (2012) 428–434
433
compatible with the usual pump sealing than the THF containing
solvents. The dependence of conversion (c) and productivity (r)
of the biocatalyst on flow-rate (v: between 0.2 and 0.6 mL min−1
was investigated with the substrates rac-1a and rac-1b (Fig. 3).
In case of the acylation of rac-1a, best conversion (c = 46%) was
reached at the lowest flow rate (v= 0.2 mL min−1). Therefore, it is
understandable that in the continuous-flow acylation of rac-1a
a steeper decrease of productivity (r) with increasing flow was
observed than in case of rac-1b for which a conversion of 29% at
v= 0.2 mL min−1 could be realized.
The fact, that all continuous-flow tests with the two substrates
rac-1a,b including washing between substrate switches (∼24 h)
and an additional 24 h long run with a third substrate [50] were
performed with the same TEOS/PTEOS/DMDEOS (4:1:1) Lipase AK-
filled bioreactor indicated the operational stability of this efficient
sol–gel biocatalyst.
Appendix A. Supplementary data
)
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.procbio.2011.11.023.
References
[1] Poppe L, Novák L. Selective biocatalysis: a synthetic approach. Weinheim, New
York: Wiley-VCH; 1992.
[2] Rehm HJ, Reed G, Pühler A, Stadler P, Kelly DR. Biotechnology: biotransforma-
tions I and II, vols. 8a and 8b, 2nd ed. Weinheim: Wiley-VCH; 1998.
[3] Faber K. Biotransformations in organic chemistry. 5th ed. Berlin: Springer;
2004.
[4] Whittall J, Suton P, editors. Practical methods for biocatalysis and biotransfor-
mations. Chichester: Wiley; 2010.
[5] Ballesteros A, van Beynum G, Borud O, Buchholz K. Guidelines for the charac-
terization of immobilized biocatalysts. Enzyme Microb Technol 1983;5:304–7.
[6] Lee KW, Min K, Park K, Yoo YJ. Biotechnol Bioprocess Eng 2010;15:603–7.
[7] Mateo CJ, Palomo M, Fernandez-Lorente G, Guisan JM, Fernandez-Lafuente R.
Improvement of enzyme activity, stability and selectivity via immobilization
techniques. Enzyme Microb Technol 2007;40:1451–63.
[8] Rodgers LE, Knott RB, Holden PJ, Pike KJ, Hanna JV, Foster LJR, et al. Structural
evolution and stability of sol–gel biocatalysts. Physica B 2006;386:508–10.
[9] Noureddini H, Gao XJ. Immobilized Pseudomonas cepacia lipase for biodiesel
fuel production from soybean oil. Sol–Gel Sci Technol 2007;41:31–41.
[10] Hanefeld U, Gardossi L, Magner E. Understanding enzyme immobilisation.
Chem Soc Rev 2009;38:453–68.
[11] Noureddini H, Gao X, Philkana RS. Immobilized Pseudomonas cepacia lipase for
biodiesel fuel production from soybean oil. Bioresour Technol 2005;96:769–77.
[12] Schmid RD, Verger R. Lipases: interfacial enzymes with attractive applications.
Angew Chem Int Ed 1998;37:1608–33.
[13] Jaeger KE, Reetz MT. Microbial lipases form versatile tools for biotechnology.
Tibtech 1998;16:396–403.
[14] Kulkarni MG, Dalai AK. Waste cooking oil – an economical source for biodiesel:
a review. Ind Eng Chem Res 2006;45:2901–13.
[15] Borgström B, Brockmann HL, editors. Lipases. Amsterdam: Elsevier; 1984.
[16] Desneulle P. In: Boyer PD, editor. The enzymes, vol. 7, 3rd ed. New York: Aca-
demic Press; 1972.
[17] Verger R, Mieras MCF, De Haas GH. Action of phospholipase A at interfaces. J
Biol Chem 1973;248:4023–34.
4. Conclusions
Our studies with disubstituted diethoxysilanes not used so far
for lipase immobilization in binary and ternary sol–gel matri-
ces, indicated the beneficial properties of this class of silane
precursors in sol–gel entrapment networks. With racemic 1-
phenylethanol (rac-1a) as test substrate, we found that using the
dimethyl-substituted silane (DMDEOS), as a component in binary
sol–gel systems for entrapment of the Lipase AK provided the
best biocatalytic properties. Applying DMDEOS-containing ternary
precursor systems for the preparation of the sol–gel network
offered, however, more opportunities to fine-tune the porosity and
hydrophobicity of the entrapment matrix. The fact that the phenyl-
containing sol–gel matrix was favorable for 1-phenylethanol rac-1a
whereas the octyl-containing one was more efficient for the more
aliphatic substrates rac-1b,c indicated the possibility of substrate-
dependent fine-tuning of the entrapment matrix for obtaining the
most efficient biocatalyst for a given compound.
[18] Bornschauer UT, Kazlauskas RJ. Catalytic promiscuity in biocatalysis: using old
enzymes to form new bonds and follow new pathways. Weinheim, New York:
Wiley-VCH; 2004.
[19] Reetz MT. Lipases as practical biocatalysts. Curr Opin Chem Biol 2002;6:145–50.
[20] Liese A, Seelbach K, Wandrey C. Industrial biotransformations. 2nd ed. Wein-
heim, New York: Wiley-VCH; 2006.
[21] Hench LL, West JK. The sol–gel process. Chem Rev 1990;90:33–72.
[22] Avnir D, Braun S, Lev O, Ottolenghi M. Enzymes and other proteins entrapped
in sol–gel materials. Chem Mater 1994;6:1605–14.
[23] Anvir D. Organic chemistry within ceramic matrices: doped sol–gel materials.
Acc Chem Res 1995;28:328–34.
[24] Kim J, Grate JW, Wang P. Nanostructures for enzyme stabilization. Chem Eng
Sci 2006;61:1017–26.
[25] Avnir D, Coradin T, Lev O, Livage J. Recent bio-applications of sol–gel materials.
J Mater Chem 2006;16:1013–30.
In addition to the fine-tuning capabilities, all the DMDEOS-
containing sol–gel Lipase AK biocatalysts exhibited excellent
stability with regard to specific biocatalyst activity and enantiomer
selectivity in repeated kinetic resolution tests with of rac-1a.
The tests in continuous-flow reactor filled with our optimal
Lipase AK biocatalyst (TEOS/DMDEOS/PTEOS 4:1:1 with twofold
lipase loading) using rac-1a and rac-1b as substrates at various flow
rates indicated the tunability of the continuous-flow system and
the robustness of the biocatalyst.
[26] Gill I, Ballesteros A. Encapsulation of biologicals within silicate, siloxane, and
hybrid sol–gel polymers: an efficient and generic approach. J Am Chem Soc
1998;120:8587–98.
In conclusion, the DMDEOS-containing binary biocatalysts (with
TEOS) and especially the ternary sol–gel lipases (with TEOS and
OTEOS or PTEOS) turned out to be robust, fine-tunable and effective
biocatalysts both in batch and continuous-flow biotransformations.
[27] Reetz MT, Jaeger KE. Overexpression, immobilization and biotechnological
application of Pseudomonas lipases. Chem Phys Lipids 1998;93:3–14.
[28] Braun S, Rappoport S, Zusman R, Avnir D, Ottolenghi M. Biochemically active
sol–gel glasses: the trapping of enzymes. Mater Lett 2007;61:2843–6.
[29] Reetz MT, Zonta A, Vijayakrishnan V, Schimossek K. Entrapment of lipases in
hydrophobic magnetite-containing sol–gel materials: magnetic separation of
heterogeneous biocatalysts. J Mol Catal A Chem 1998;134:251–8.
[30] Bornscheuer UT. Immobilizing enzymes: how to create more suitable biocata-
lysts. Angew Chem Int Ed 2003;42:3336–7.
[31] Reetz MT, Zonta A, Simpelkamp J. Efficient heterogeneous biocatalysts by
entrapment of lipases in hydrophobic sol–gel materials. Angew Chem Int Ed
Engl 1995;34:301–3.
[32] Colton IJ, Ahmed SN, Kazlauskas RJ. A 2-propanol treatment increases the enan-
tioselectivity of Candida rugosa lipase toward esters of chiral carboxylic acids.
J Org Chem 1995;60:212–7.
[33] Reetz MT, Zonta A, Simpelkamp J. Efficient immobilization of lipases by entrap-
ment in hydrophobic sol–gel materials. Biotechnol Bioeng 1995;49:527–34.
[34] Reetz MT, Tielmann P, Wiesenhöfer W, Könen W, Zonta A. Second generation
sol–gel encapsulated lipases: robust heterogeneous biocatalysts. Adv Synth
Catal 2003;345:717–28.
[35] Tomin A, Weiser D, Hellner G, Bata Z, Corici L, Péter F, et al. Fine-tuning the
second generation sol–gel lipase immobilization with ternary alkoxysilane pre-
cursor systems. Process Biochem 2011;46:52–8.
Acknowledgements
This project was supported by the Hungarian National Office for
Research and Technology (NKFP-07-A2 FLOWREAC). This work is
also related to the scientific program of “Development of quality-
oriented and harmonized R+D+I strategy and functional model at
BME” project (TÁMOP-4.2.1/B-09/1/KMR-2010-0002), supported
by the New Hungary Development Plan. Thanks are due to Thales
Nanotechnology Inc. for the fillable CatCart® columns, Prof. Sán-
dor Kemény (Budapest University of Technology and Economics,
Hungary) for his help in experimental design and Gabriella Hell-
ner (Bunge Europe Innovation Centre, Budapest, Hungary) for her
contribution to the preparation of several sol–gel samples.
[36] Furukawa SY, Kawakami K. Characterization of Candida rugosa lipase entrapped
into organically modified silicates in esterification of menthol with butyric acid.
J Ferment Bioeng 1998;85:240–2.

434
D. Weiser et al. / Process Biochemistry 47 (2012) 428–434
[37] Gulcev MD, Goring GLG, Rakic M, Brennan JD. Reagentless pH-based biosensing
using a fluorescently-labelled dextran co-entrapped with a hydrolytic enzyme
in sol–gel derived nanocomposite films. Anal Chim Acta 2002;457:47–59.
[38] Lee C, Yi SS, Kim J, Lee YS, Kim BG. Improved immobilized enzyme systems
using spherical micro silica sol–gel enzyme beads. Biotechnol Bioprocess Eng
2006;11:277–81.
[44] Chen JC, Tsai SW. Enantioselective synthesis of (S)-Ibuprofen ester prodrug
in cyclohexane by Candida rugosa lipase immobilized on Accurel MP1000.
Biotechnol Prog 2000;16:986–92.
[45] Ujang Z, Husain WH, Seng MC, Rashid AHA. The kinetic resolution of 2-(4-
chlorophenoxy)propionic acid using Candida rugosa lipase. Process Biochem
2003;38:1483–8.
[39] Corîci LN, Frissen AE, van Zoelen DJ, Eggen IF, Peter F, Davidescu CM, et al.
Sol–gel immobilization of Alcalase from Bacillus licheniformis for application in
the synthesis of C-terminal peptide amides. J Mol Catal B Enzym 2011;73:90–7.
[40] Hara P, Hanefeld U, Kanerva LT. Sol–gels and cross-linked aggregates of lipase
PS from Burkholderia cepacia and their application in dry organic solvents. J Mol
Catal B Enzym 2007;50:80–6.
[46] Tomin A, Hornyánszky G, Kupai K, Dorkó Z, Ürge L, Darvas F, et al. Lipase-
catalyzed kinetic resolution of 2-methylene-substituted cycloalkanols in batch
and continuous-flow modes. Process Biochem 2010;45:859–65.
[47] Patel RN, Banerjee A, Szarka LJ. Biocatalytic synthesis of some chiral pharma-
ceutical intermediates by lipases. J Am Oil Chem Soc 1996;73:1363–75.
[48] Rakels LL, Straathof AJJ, Heijnen JJ.
A simple method to determine the
[41] Stanford TB, van Ast CI. Enzyme biosensor and enzyme detection method for
aerosolized bacteria in the environment. US Pat Appl, US 20040092004 A1;
2004.
[42] Stanford TB, van Ast CI, Yamagishi FG. Sensors with variable response behavior
for detecting pathogens. US Pat Appl, US 20030062263 A1; 2003.
[43] Csajági C, Szatzker G, To˝ ke ER, Ürge L, Darvas F, Poppe L. Enantiomer selec-
tive acylation of racemic alcohols by lipases in continuous-flow bioreactors.
Tetrahedron Asymmetry 2008;19:237–46.
enantiomeric ratio in enantioselective biocatalysis. Enzyme Microb Technol
1993;15:1051–6.
[49] Chen CS, Fujimoto Y, Girdaukas G, Sih CJ. Quantitative analyses of biochemical
kinetic resolutions of enantiomers. J Am Chem Soc 1982;104:7294–9.
[50] Hellner G, Boros Z, Tomin A, Poppe L. Novel sol–gel lipases by designed
bioimprinting for continuous-flow kinetic resolutions. Adv Synth Catal
2011;353:2481–91.