Pegylated enzyme entrapped in poly(vinyl alcohol) hydrogel for biocatalytic application

Article abstract of DOI:10.1016/S0014-827X(01)01032-1

A procedure for enzyme entrapment into matrices suitable for biocatalytic applications is reported. The method, which takes advantage of the stable formation of polyvinyl alcohol (PVA) hydrogels by freezing and thawing PVA aqueous solutions, was assayed using lipase as model enzyme. The leakage of lipase was minimised by using high molecular weight PVA and by previous conjugation of the enzyme to PEG. The immobilised PEG enzyme maintained its catalytic activity in organic solvents also, thus allowing enzymatic activity towards water insoluble substrates. The activity was largely increased reducing the diffusional constrain by cutting the matrices into slices of micron size. Matrix-entrapped lipase-PEG, when used in the hydrolysis of acetoxycoumarins, showed a conversion rate of about 10 times lower than the enzyme-PEG in the free form, and maintained regioselectivity when a diacetylated product was used as substrate.

Full text of DOI:10.1016/S0014-827X(01)01032-1

www.elsevier.com/locate/farmac
Il Farmaco 56 (2001) 541–547
Pegylated enzyme entrapped in poly(vinyl alcohol) hydrogel for
biocatalytic application
Francesco Maria Veronese a,*, Cristina Mammucari a, Franco Schiavon a,
Odolone Schiavon a, Silvano Lora a, Francesco Secundo b, Adriana Chilin b,
Adriano Guiotto b
a Department of Pharmaceutical Sciences, Uni6ersity of Padua, Center of Chemical In6estigation of Drugs CNR, Pado6a Via Marzolo n. 5,
35131 Padua, Italy
b Institute of Biocatalysis and Molecular Recognition CNR, Via Mario Bianco n. 9, 20131 Milan, Italy
Received 21 March 2000; accepted 18 July 2000
Abstract
A procedure for enzyme entrapment into matrices suitable for biocatalytic applications is reported. The method, which takes
advantage of the stable formation of polyvinyl alcohol (PVA) hydrogels by freezing and thawing PVA aqueous solutions, was
assayed using lipase as model enzyme. The leakage of lipase was minimised by using high molecular weight PVA and by previous
conjugation of the enzyme to PEG.
The immobilised PEG enzyme maintained its catalytic activity in organic solvents also, thus allowing enzymatic activity towards
water insoluble substrates. The activity was largely increased reducing the diffusional constrain by cutting the matrices into slices
of micron size. Matrix-entrapped lipase-PEG, when used in the hydrolysis of acetoxycoumarins, showed a conversion rate of
about 10 times lower than the enzyme-PEG in the free form, and maintained regioselectivity when a diacetylated product was used
as substrate. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: PEG; PVA; Poly(ethylene glycol); Poly(vinyl alcohol); Enzyme immobilisation; PVA hydrogels; Lipase; Biocatalysis; Coumarins;
Regioselectivity
thus allowing a further catalytic step; this possibility is
achieved by linking the enzymes to an insoluble matrix.
In the majority of cases, the enzyme is covalently linked
to an organic polymer, either natural or synthetic, but
important examples of physical entrapment have also
been reported [7]. The entrapment into polysaccharides,
proposed by Chribata [8], is an example of successful
industrial application of the latter technology. Unfortu-
nately, the immobilisation by covalent binding is gener-
ally accompanied by a loss of enzyme activity, and it is
always a delicate and tedious process. On the other
hand, physical entrapment is limited to aqueous operat-
ing conditions and is often accompanied by enzyme
leakage.
1. Introduction
Organic synthesis received an unexpected accelera-
tion from the use of enzymes as catalysts, due to their
high rate of conversion, specificity, regio- and, fre-
quently, stereo-selectivity [1]. Moreover, the use of en-
zymes in solid state finely dispersed in organic solvents,
as first described by Klibanov [2,3], widened its applica-
tion to the conversion of water insoluble compounds.
More recently, solubility and activity of enzymes in
organic solvents was also achieved by covalent linking
to amphiphilic polymers, poly(ethylene glycol) (PEG) in
particular [4–6]. The extensive use of enzymes as cata-
lysts was related to the possibility of their easy recovery
from the reaction mixture at the end of the reaction,
Poly(vinyl alcohol) (PVA) presents the unusual prop-
erty of affording stable hydrogels, through extremely
mild freezing and thawing conditions of its aqueous
solutions, a property that was exploited for the prepa-
ration of new biomaterials and for drug release [9–14].
* Corresponding author. F.M. Veronese, Dipartimento di Scienze
Farmaceutiche via Marzolo, 5, 35131 Padova.
E-mail address: Francesco.Veronese@unipd.it (F.M. Veronese).
0014-827X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0014-827X(01)01032-1

542
F.M. Veronese et al. / Il Farmaco 56 (2001) 541–547
We recently demonstrated that the release of
biomolecules from PVA hydrogels can be modulated by
the polymer molecular weight, and/or, most impor-
tantly, by linking hindered PEG chains to the en-
trapped products [15]. This method allowed the
preparation of new biosensors based on PVA with
entrapped glucose oxidase, choline oxidase, acethylcol-
ine esterase, enzymes that by themselves would escape
from the gels [16,17].
On the basis of these prerequisites, we are now
reporting the preparation of PVA hydrogels with en-
trapped lipase, previously conjugated to PEG, for the
bioconversion of a monoacetoxycoumarin and a diace-
toxycoumarin to the corresponding hydroxycoumarins,
which are important synthons in the coumarin synthe-
ses [18,19]. A diacetylated coumarin was chosen as
substrate, in which two esterified functions with differ-
ent steric surroundings are present. Organic solvents
were used to solubilise the substrates, thanks to the
possibility to equilibrate the PVA hydrogels with n-hex-
ane without losing their physical properties and by
conveying lipophilicity to lipase by PEG conjugation.
This study further demonstrated the great potential of
PEG to overcome physical and chemical problems in
enzyme technology [20–22].
from Sigma (St. Louis, USA). n-Hexane was obtained
from Lab-Scan (Dublin, Ireland). 7-Hydroxy-cumarin
(1), 7-acetoxy-coumarin (2), 5,7-dihydroxy-4-methyl-
coumarin (3), 5,7-diacetoxy-4-methylcoumarin (4), 5-
acetoxy-7-hydroxy-4-methylcoumarin (5) and 7-
acetoxy-5-hydroxy-4-methylcoumarin (6) (see Scheme
1) were synthesised in our laboratory and their synthe-
sis will be reported elsewhere (Guiotto et al., personal
communication). SDS-page reagents were purchased
from Biorad (Richmond, CA, USA), solvents and salts
were Merck products (Darmstadt, Germany).
2.2. Equipment
Spectrophotometric measurements were carried out
with a Perkin–Elmer (Norwalk, CT, USA) LAMBDA
5 Spectrophotometer. The ultrafiltration were per-
formed by an ultrafiltration apparatus, Amicon (Passir-
ana di Rho, Milan, Italy), equipped with a 30 kDa
cut-off membrane. A gel filtration column operated
with a Pharmacia FPLC system was used for the conju-
gates purification. For the activity measurements of
immobilised lipase, the matrices were maintained under
stirring at room temperature, using a floating stirrer of
Bicasa (Milano, Italy). HPLC analyses were performed
on a Shimadzu LC-10 AT instrument, equipped with
SPD-10 A UV–Vis detector. A Lichrocart RP18
column was used, eluting with H2O/CH3CN, 60:40;
2. Experimental
flow rate 0.3 ml min−1
.
2.1. Reagents
2.3. Lipase modification by PEG
Lipase B from Candida Antarctica (MW 35 000) [23]
was kindly supplied by Dr. G. Carrea, CNR, Milan.
Poly(vinyl alcohol) (PVA) 31–50 kDa, 124–186 kDa
(99% hydrolyzed) were purchased from Aldrich Chemie
(Stenheim, Germany). Monomethoxypoly(ethylene gly-
col)-norleucine (PEG-Nle) 5 kDa hydroxysuccinimide
ester was obtained from Shearwater Polymers
(Huntsville, USA). p-Nitrophenylbutyrate was obtained
Lipase was dissolved in 0.2 M borate buffer at pH
8.0 and PEG-Nle 5 kDa activated as hydroxysuccin-
imidylester was added as dry powder in two portions.
The amount of PEG was in a tenfold molar excess over
lipase amino groups. After 2 h the reaction mixture was
washed with water and reduced to 5 ml by ultrafiltra-
tion (30 kDa membrane). This procedure was repeated
five times. Alternatively, the reaction mixture was frac-
tionated by GPC and ultrafiltered. The activity of
lipase-PEG was determined as follows: 100 ml of p-ni-
trophenylbutyrate solution in isopropanol was diluted
to 1 ml with buffer (0.02 M NaH2PO4, 0.9% NaCl, pH
7.0) and the reaction was initiated by addition of an
aliquot of the lipase-PEG solution. The increase of
absorbance at 410 nm against buffer for a period of
about 5 min was recorded. The lipase-PEG activity was
compared to the native lipase activity evaluated under
the same experimental conditions. The unreacted amino
groups following the PEG conjugation were evaluated
by colorimetric assay [24].
2.4. Lipase immobilisation and hydrogel preparation
An aqueous solution of 20% PVA was heated at
105°C in a closed vial for 6 h under gentle stirring and
Scheme 1.

F.M. Veronese et al. / Il Farmaco 56 (2001) 541–547
543
mixture were removed, diluted with the buffer (0.02 M
NaH2PO4, 0.9% NaCl, pH 7.0) and the activity mea-
sured by monitoring the absorbance increase at 410 nm.
The experiment was repeated by dipping the carefully
washed matrix into a fresh reaction mixture. To reduce
the diffusional constraints, the hydrogels were
lyophilised and cut with a microtome maintained at
−20°C, into 8 mm thick slices. The slices were main-
tained in n-hexane/isopropanol (10:1), under delicate
stirring. At different time intervals, portions (50 ml) of
the supernatant were diluted in the phosphate buffer
and the absorbance was measured at 410 nm as above
reported.
2.6. Matrices dimension in different sol6ents
Cylindrical matrices of 1.5 cm diameter and 0.3 cm
thickness, prepared as reported above, were dipped into
various solvents and incubated for 48 h, exchanging the
external liquid every 8 h. The solvent was then re-
moved, the matrices were gently wiped with a paper
towel, and their dimension recorded.
2.7. E6aluation of acetoxycoumarin hydrolysis
Fig. 1. FPLC gel filtration pattern of the lipase PEGylation mixture
as revealed by 280 nm absorption for lipase and lipase-PEG detec-
tion, and iodine reaction for PEG: (1) PEGylated lipase first peak; (2)
PEGylated lipase second peak; (3) unreacted lipase; (4) unreacted
PEG.
7-Acetoxycoumarin (2) and 5,7-diacetoxy-4-methyl-
coumarin (4) were dissolved in n-hexane/isopropanol,
7:1, at a 10−5 M concentration. To 5 ml of the cou-
marin solution a matrix with entrapped lipase-PEG (1.5
cm diameter×0.3 cm thickness) was added and the
mixture was shaken at 36°C. Alternatively, free lipase-
PEG was also added. From the incubating mixture, 250
ml samples were drowned at scheduled times and as-
sayed by HPLC or UV absorption. In particular, 2 was
evaluated by UV absorption only, while for 4 evalua-
tion samples were withdrawn, concentrated to dryness,
dissolved in H2O/CH3CN and analysed by HPLC un-
der the conditions reported above.
then cooled at room temperature. Lipase-PEG (2 mg,
expressed as enzyme content), was added under stirring
and the mixture was poured into a mould of the desired
shape. The mixture was frozen at −25°C for 4 h. The
mould was warmed up to room temperature for 30 min
to thaw the PVA-lipase solid. The process of freezing
and thawing was repeated five times. Hydrogel cylin-
ders of 5 mm diameter and 2.5 mm thickness were
obtained. The hydrogels were soaked many times in the
solvent used for the measurements of the enzyme activ-
ity in order to reach equilibrium with the external
solvent. For the experiments of acetoxycoumarin bio-
conversion, cylinders of 1.5 cm diameters and 0.3 cm
thicknesses, containing 0.5 mg of lipase-PEG (expressed
as enzyme content) were used.
3. Results and discussion
3.1. PEG conjugation of lipase
According to colorimetric assay, 60% of lipase amino
groups appeared to be modified by PEG conjugation.
The conjugation mixture, fractionated by gel filtration,
showed the presence of two peaks, centred at fractions
80 and 87, respectively, while free lipase was eluted
within a third peak centred at fraction 102, as shown in
Fig. 1. The fractions of the two peaks, pooled sepa-
rately and examined by SDS electrophoresis, showed
the pattern reported in Fig. 2. From the migration, by
comparison with proteins standard, an apparent molec-
ular weight of 65 and 50 kDa, respectively, was calcu-
lated for the two samples of lipase-PEG. This
2.5. Acti6ity of PVA entrapped lipase-PEG
The activity of lipase-PEG entrapped in PVA hy-
drogel cylinders was determined spectrophotometri-
cally. In this way, p-nitrophenylbutyrate was dissolved
in isopropanol, the reaction was started by dipping the
lipase-PEG containing hydrogel into the substrate solu-
tion and the incubation mixture was maintained under
stirring. At different time intervals, 50 ml of the reaction

544
F.M. Veronese et al. / Il Farmaco 56 (2001) 541–547
Table 2
corresponds to a linking of approximately six and three
PEG chains. The amount of bound PEG in the first
peak fraction could also be evaluated by its norleucine
content, obtained by amino acid analysis after acid
hydrolysis [25]. A number of 7.3 polymer chains were
found, in agreement with the value obtained by colori-
Activity of PEG-lipase immobilised in PVA hydrogel matrix (1.5 cm
diameter×0.3 cm thickness), in different solvents
Reaction solvent
Activity a (%)
Buffer pH 7.00
Buffer pH 7.00/isopropanol 1:9
n-Hexane/isopropanol 10:1
100
5.66
7.66
a The activity of immobilised lipase-PEG in solvent was calculated
after six washings and is relative to that found in phosphate buffer.
The substrate was p-nitrophenylbutyrate.
metric and electrophoretic analysis. Unfortunately, the
second peak could not be examined by amino acid
analysis, since it contained an excess of unbound PEG
due to the partial overlapping of the lipase-PEG and
free PEG in the chromatogram.
For the experiments reported below, the pooled li-
pase-PEG mixture of the two peaks was employed. It
was found to possess an activity, evaluated towards the
substrate p-nitrophenylbutyrate, superior to the native
one (about 120%). This behaviour, already observed in
conjugates of the enzymes with PEG and with low
molecular weight reagents, was generally ascribed to
increase in substrate affinity or subtle positive modifica-
tions at the active site geometry [26].
Fig. 2. SDS electrophoresis of the two lipase-PEG fractions and of
standard proteins: (1) lipase reference standard; (2) unreacted lipase,
third peak of Fig. 1; (3) lipase-PEG, second fraction of Fig. 1; (4)
lipase-PEG first fraction; and (5) reference standard proteins of 175,
93, 67, 47.5, 32.5 kDa.
3.2. Lipase-PEG entrapment in PVA hydrogels and
acti6ity e6aluation
Table 1
PVA-hydrogel shrinking (%) following soaking in different solvents
Solvent
Shrinking (%)
20 (h)
Lipase-PEG entrapment into PVA was carried out
following a procedure that, with other enzymes, was
found suitable for the maintenance of the native con-
formation and activity. As a preliminary investigation,
several solvents were evaluated as potential organic
medium for PVA gels incubation. The shrinking of the
gel, after 20 and 130 h of soaking in different solvents,
was evaluated and referred to the dimension of the gel
in water (see Table 1). In n-hexane/isopropanol the
dimensions remained practically the same as in the
phosphate buffer, minor shrinking occurred in benzene,
while it was high in all the other tested solvents.
The effect of the molecular weight of PVA and of
PEG conjugated to lipase in the release of the enzyme
from the matrix was evaluated. The large influence of
both parameters in preventing lipase leakage is shown
in Fig. 3 and it was found in agreement with what has
been already reported for other proteins [15].
130 (h)
Phosphate buffer
n-Hexane/isopropanol 10:1
Phosphate buffer/isopropanol 1:9
n-Butanol
Benzene
Dichloromethane
Ethyl acetate
0
0
4.2
5
29.2
31.7
4.2
12.5
33.4
33.4
33.4
12.5
40.8
38.3
The activity of the entrapped lipase-PEG carried out
in the phosphate buffer, buffer/isopropanol, and n-hex-
ane/isopropanol as solvents is reported in Table 2. It
appears that the solvent used is critical for the enzyme
activity. Furthermore, the loss of entrapped lipase-PEG
took place at the first change of the incubation solvent,
to reach an amount of lipase-PEG permanently en-
trapped that corresponds to 37% of the starting one
(see Table 3a).
Fig. 3. Leakage from PVA hydrogel of (ꢀ) lipase, (×) lipase-PEG
from 31–50 kDa PVA, and (") lipase and (ꢁ) lipase-PEG from
124–186 kDa PVA. Mt: total amount of released enzyme at time t;
Mo: total amount of enzyme in the hydrogel.

F.M. Veronese et al. / Il Farmaco 56 (2001) 541–547
545
Since diffusion could be the major reason for the
decreased activity of entrapped lipase, an experiment
was carried out after reducing the thickness of the
hydrogel to 8 mm size. As in the experiment reported in
Table 3b, n-hexane/isopropanol, 10:1, was used as sol-
vent. At intervals the hydrogel was removed from the
substrate solution, and soaked into a new one, before
the activity was measured again. This process was
repeated seven times. After three changes of the sub-
strate solution the activity remained constant to about
one half of the starting value (see Table 3b). The
increase in activity following matrix slicing is impres-
sive, about 100 times greater than the original. This
result demonstrates the great influence of diffusional
constraints that can be successfully minimised by reduc-
ing the thickness of the gel slices.
The use of organic solvents in these experiments
demonstrated also the suitability of the immobilisation
procedure to carry out biocatalysis in organic
environment.
3.3. Acetoxycoumarin hydrolysis
(a) Comparisons between free and gel entrapped li-
pase-PEG acti6ity. The hydrolysis rate of 7-acetoxycou-
marin (2) in n-hexane/isopropanol could be evaluated
by UV absorption since a decrease of absorption at 280
Fig. 4. Time course of 7-acetoxycoumarin (2) hydrolysis operated by
lipase-PEG hydrogels and monitored by UV absorption. 7-Acetoxy-
coumarin after (a) 45 min incubation; (b) 4 h incubation; and (c) at
the end of hydrolysis (21 h).
Table 3
Activity of PEG-lipase immobilised into PVA hydrogels (1.5 cm
diameter×0.3 cm thickness) in n-hexane/isopropanol, 10:1, following
repeated incubation before (a), and after cutting the same matrix in
micron size slices (b)
nm accompanied by an increase of absorption at 310
nm was observed (see Fig. 4). A comparison between
the rate of the free and immobilised lipase-PEG under
the same experimental conditions demonstrated that the
free enzyme is approximately 10 times more active than
the gel-entrapped one. Taking into account the loss of
enzyme that occurs during the washings needed to
reach the permanently entrapped enzyme into the ma-
trix we could in fact calculate that the specific activity
of free lipase-PEG towards 7-acetoxycoumarin (2) is
0.43×10−3 mM h−1 while that of the entrapped en-
zyme is 0.072×10−3 mM h−1. According to the exper-
iments reported above, such reduction of activity may
be mainly related to the diffusion limitations.
(b) Regioselecti6ity in the 5,7-diacetoxy-4-methylcou-
marin (4) hydrolysis reaction. 5,7-Diacetoxy-4-methyl-
coumarin (4) was incubated with the immobilised
enzyme and the reaction mixture was examined by
HPLC, since the multiplicity of the products did not
allow direct monitoring of the reaction by UV absorp-
tion. Four products were revealed by HPLC during the
time course of the catalysis, corresponding to the start-
ing product, eluted at 14.97 min, the fully deacetylated
5,7-dihydroxy-4-methylcoumarin (3), eluted at 5.75 min
Number of incubations
(a) Before cutting a First
Activity (%)
100
Second
Third
Fourth
Fifth
76.78
30.63
35.57
36.56
37.05
Sixth
(b) After cutting b
First
9061
9680
6272
4330
4128
4422
4294
Second
Third
Fourth
Fifth
Sixth
Seventh
a The activity following incubations is referred to the freshly pre-
pared matrix. After each incubation the matrix is gently washed with
fresh incubation mixture. The substrate was p-nitrophenylbutyrate
and the solvent was n-hexane/isopropanol, 10:1.
b The matrices were cut with a microtome at −20°C to obtain 8
mm thick slices. The percentage of activity of the sliced matrix is
calculated assuming 100 as the activity of the unsliced one from the
sixth incubation, when no more enzyme leakage occurs. At the
seventh washing an increase in activity of 100 fold, due to the slicing,
could be calculated.

546
F.M. Veronese et al. / Il Farmaco 56 (2001) 541–547
and the two monoacetylated coumarins 7-acetoxy-5-hy-
droxy-4-methylcoumarin (6) and 5-acetoxy-7-hydroxy-
4-methylcoumarin (5), eluted at 7.92 and 9.36 min,
respectively (see Fig. 5).
After 150 h the reaction slows down and three prod-
ucts are present, the two monoacetoxy-monohydroxy-
coumarines and the fully deacetylated dihydroxy-
coumarin. In Fig. 6 the time course of the enzymatic
reaction, as obtained from the HPLC analysis, is
shown. It appears that a preferential regiospecificity
was achieved in favour of the formation of the product
deacetylated in position 5. This seems to indicate that
the hydrolysis of the second acetyl group is a slow
process, the monoacetylated compounds being worse
enzyme substrates than the diacetylated one.
4. Conclusions
The most striking information coming out from this
study was the demonstration that PEGylated lipase,
entrapped in poly(vinyl alcohol) hydrogel by freezing
and thawing, can be conveniently used in organic sol-
vent biocatalysis. This enzyme immobilisation method
is much more convenient in comparison to other type
of immobilisation, that take advantage of enzyme cova-
Fig. 6. Time course of disappearance of 5,7-diacetoxy-4-methylcou-
marin (4) and formation of the completely deacetylated product (3),
and of the products of 6 and 5.
lent linkage to insoluble matrix, since the chemical step
which is time consuming and harmful to enzyme activ-
ity, is avoided.
Furthermore, it was found that non-aqueous solvents
are compatible with gel stability and enzyme activity,
thus allowing a choice for easy dissolution of the prod-
ucts of interest.
The PEGylation step is a straightforward procedure
known for many enzymes, which is characterised by the
high maintenance of activity [14–16]. This modification
is the critical step to increase the enzyme size needed to
improve retention inside the gel and to convey at the
same time the solubility of the enzyme inside the
matrix.
As expected, the entrapment inside the gel reduces
the catalytic rate of lipase to about one order of
magnitude, due to diffusional constraints. However,
these limitations can be minimised by a proper slicing
of the gel matrix.
Finally, the application of the lipase bioconversion to
diacetoxycoumarin derivatives demonstrated the feasi-
bility of the proposed method in the hydrolysis of
products of pharmaceutical interest and to obtain re-
gioselective enrichment of one of the two monodeacety-
lated derivatives.
Fig. 5. HPLC analysis of the hydrolysis mixture of 5,7-diacetoxy-4-
methylcoumarin: 5,7-dihydroxy-4-methylcoumarin (3), 7-acetoxy-5-
hydroxy-4-methylcoumarin (6), 5-acetoxy-7 hydroxy-4methyl-
coumarin (5) and starting 5,7-diacetoxy-4-methylcoumarin (4).

F.M. Veronese et al. / Il Farmaco 56 (2001) 541–547
547
[11] N.A. Peppas, S.R. Stauffer, Reinforced uncross linked poly(vinyl
alcohol) gels produced by cyclic freezing–thawing process: a
short review, J. Controlled Release 16 (1991) 305–310.
[12] F. Urushizaki, H. Yamaguchi, K. Nakamura, S. Numajiri, K.
Sugibayashi, Y. Morimoto, Swelling and mechanical properties
of poly(vinyl alcohol) hydrogels, Int. J. Pharm. 58 (1990) 135–
142.
[13] P.R. Byron, R.N. Dalby, Effects of heat treatment on the
permeability of polyvinyl alcohol films to a hydrophilic solute, J.
Pharm. Sci. 76 (1987) 65–67.
Application of the PEGylated gel entrapped enzymes
with different specificity is presently under investigation
and will be reported separately.
Acknowledgements
The authors thank A. Moretto for his skilful techni-
cal assistance and CNR ‘‘Finalised Project of Biothec-
nology and Bioinstrumentation’’ for financial support.
[14] K. Burczak, T. Fujisato, M. Hatada, Y. Ikada, Protein perme-
ation through poly(vinyl alcohol) hydrogel membranes, Biomate-
rials 15 (1994) 231–237.
[15] F.M. Veronese, C. Mammucari, P. Caliceti, O. Schiavon, Influ-
ence of PEGylation on the release of low and high molecular-
weight proteins from PVA matrices, J. Bioact. Compat. Polym.
14 (1999) 315–330.
[16] L. Doretti, D. Ferrara, P. Gattolin, S. Lora, F. Schiavon, F.M.
Veronese, PEG-modified glucose oxidase immobilized on a PVA
cryogel membrane for amperometric biosensor applications, Ta-
lanta 45 (1998) 891–898.
[17] L. Doretti, D. Ferrara, S. Lora, F. Schiavon, F.M. Veronese,
Acetylcholine biosensor involving entrapment of acetyl-
cholinesterase and PEG-modified choline oxidase in PVA cryogel
membrane, Enzyme Microb. Tech. (2000), in press.
[18] R.D.H. Murray, J. Mendez, S.A. Brown, The Natural Cou-
marins, Occurrence, Chemistry and Biochemistry, Wiley, New
York, 1982.
References
[1] C.H. Wong, G.M. Whitesides, Enzymes in Synthetic Organic
Chemistry, Pergamon, Amsterdam, 1994.
[2] Zaks, A.M. Klibanov, Enzymatic catalysis in nonaqueous sol-
vents, J. Biol. Chem. 263 (1988) 3194–3201.
[3] C.R. Wescott, A.M. Klibanof, The solvent dependence of en-
zyme specificity, Biochim. Biophys. Acta 1206 (1994) 1–9.
[4] P.A. Mabrouk, The use of poly(ethylene glycol)-enzymes in
nonaqueous enzymology, in: J.M. Harris, S. Zalipsky (Eds.),
Poly(ethylene glycol) Chemistry and Biological Applications,
American Chemical Society, Washington, DC, 1997, pp. 118–
133.
[5] Y. Inada, A. Matsushima, Y. Kodera, H. Nishimura,
Poly(ethylene glycol) (PEG)–protein conjugates: application to
biomedical and biotechnological processes, J. Bioact. Compat.
Polym. 5 (1990) 343–364.
[6] Y. Inada, T. Takahashi, T. Yoshimoto, A. Ajima, A. Mat-
sushima, Y. Saito, Application of poly(ethylene glycol)modified
enzyme in biotechnological process: organic solvent soluble en-
zyme, Tibtech 4 (1986) 190–194.
[7] W. Marconi, F. Bartoli, F. Morisi, A. Marani, Improved whey
treatment by immobilized lactase, in: H.H. Weetal, G.P. Royer
(Eds.), Enzyme Engineering, vol. vol. 5, Plenum Press, New
York, 1979, pp. 269–278.
[8] T. Tosa, M. Furui, N. Sakata, O. Otsuki, I. Chibata, Design of
a bioreactor using immobilized biocatalyst for the slurry reac-
tion, Ann. N.Y. Acad. Sci. 542 (1988) 440–443 (Enzyme Engi-
neering 9).
[9] N.A. Peppas, J.E. Scott, Controlled release from poly(vinyl
alcohol) gels prepared by freezing-thawing process, J. Controlled
Release 18 (1992) 95–100.
[10] N. Peppas, S.L. Borcherding, Oral delivery device from freeze/
thawed poly(vinyl alcohol) hydrogels, in: Proc. Int. Symp. Con-
trol. Rel. Bioact. Mater, vol. 23, 1996, pp. 145–146.
[19] R.D.H. Murray, Naturally occurring plant coumarins, Fortschr.
Chem. Org. Naturst. 72 (1997).
[20] C. Delgado, G.E. Francis, D. Fisher, The use and properties of
PEG linked, Crit. Rev. Ther. Drug 9 (1992) 249–304.
[21] F.M. Veronese, M. Morpurgo, Bioconjugation in pharmaceuti-
cal chemistry, Il Farmaco 54 (1999) 497–516.
[22] C. Monfardini, F.M. Veronese, Stabilization of substances in
circulation, Bioconjug. Chem. 9 (1998) 418–450.
[23] S.A. Patkar, F. Bjorking, M. Zundel, M. Schulein, Purification
of two lipase from Candida Antarctica and their inhibition by
various inhibitors, Indian J. Chem. 32 (1993) 76–80.
[24] A.F. Habeeb, Determination of free aminogroups in proteins by
trinitrobenzensulphonic acid, Anal. Biochem. 14 (1966) 328–336.
[25] L. Sartore, P. Caliceti, O. Schiavon, F.M. Veronese, Enzyme
modification by MPEG with amino acid or peptide as spacer
arm, Appl. Biochem. Biotech. 27 (1991) 55–63.
[26] F.M. Veronese, C. Monfardini, P. Caliceti, O. Schiavon, M.D.
Scraven, D. Beer, Improvement of pharmacokinetic, immunolog-
ical and stability properties of asparaginase to linear and
branched monomethoxypoly(ethylene glycol), J. Controlled Re-
lease 40 (1996) 199–206.
.