Promotion of multipoint covalent immobilization through different regions of genetically modified penicillin G acylase from E. coli

Article abstract of DOI:10.1016/j.procbio.2009.10.013

A novel approach is proposed to prepare a set of immobilized derivatives of a enzyme covalently rigidified through different regions of its surface. Six different variants of penicillin G acylase (PGA) from Escherichia coli (which lacks Cys) were prepared

Full text of DOI:10.1016/j.procbio.2009.10.013

Process Biochemistry 45 (2010) 390–398
Contents lists available at ScienceDirect
Process Biochemistry
journal homepage: www.elsevier.com/locate/procbio
Promotion of multipoint covalent immobilization through different regions
of genetically modified penicillin G acylase from E. coli
a,e
a
a
a
b
Valeria Graz u´ , Fernando L o´ pez-Gallego , Tamara Montes , Olga Abian , Ram o´ n Gonz a´ lez ,
c
d
a
a,
Juan A. Hermoso , Jos e´ L. Garc ı´ a , C e´ sar Mateo , Jos e´ M. Guis a´ n *
a
Departamento de Biocat a´ lisis, Instituto de Cat a´ lisis, CSIC, Campus UAM, Cantoblanco, 28049 Madrid, Spain
b
Departamento de Microbiolog ı´ a de los Alimentos, Instituto de Fermentaciones Industriales CSIC, C/Juan de la Cierva 3, 28006 Madrid, Spain
c
Departamento de Cristalograf ı´ a, Instituto de Qu ı´ mica-F ı´ sica ‘‘Rocasolano’’ (CSIC), Serrano 119, 28006 Madrid, Spain
d
Departamento de Microbiolog ı´ a Molecular, Centro de Investigaciones Biol o´ gicas, CSIC, Ramiro de Maeztu 9, 28040 Madrid, Spain
e
Grupo de Biofuncionalizaci o´ n de Nanopart ı´ culas y Superficies (bioNanoSurf), Instituto Universitario de Nanociencia de Arag o´ n, Campus R ı´ o Ebro, Edificio I+D, Mariano Esquillor, s/n,
5
0018 Zaragoza, Spain
A R T I C L E I N F O
A B S T R A C T
Article history:
A novel approach is proposed to prepare a set of immobilized derivatives of a enzyme covalently
rigidified through different regions of its surface. Six different variants of penicillin G acylase (PGA) from
Escherichia coli (which lacks Cys) were prepared by introducing a unique Cys residue via site-directed
mutagenesis in six different enzyme regions which were rich in Lys residues. All variants exhibited a
similar activity and stability compared to those of the native enzyme. Each variant was immobilized on
supports having a low concentration of reactive disulfide moieties and a high concentration of poorly
reactive epoxy groups. After immobilization at pH 7.0 by site-directed thiol-disulfide intermolecular
exchange, derivatives were further incubated at pH 10.0 for 48 h to promote an additional
intramolecular reaction between Lys residues of enzyme and epoxy groups of the support. The
establishment of at least three covalent attachments per PGA molecule was determined for all
immobilized enzyme variants. The different derivatives exhibited diverse stability against several
distorting agents and different selectivity in two interesting reactions. The derivative of the PGA variant
obtained by replacement of GlnB380 by Cys was the most stable against heat and organic cosolvents: it
preserved 90% of the initial activity and was 30-fold more stable than soluble PGA. This derivative also
exhibited an improved enantioselectivity in the hydrolysis of chiral esters (E was improved from 8 to 16)
and in kinetically controlled synthesis of amides (synthetic yields were increased from 31 to 49%).
ß 2009 Elsevier Ltd. All rights reserved.
Received 23 March 2009
Received in revised form 4 October 2009
Accepted 21 October 2009
Keywords:
Directed immobilization
Multipoint covalent attachment
Directed rigidification
Enzyme stabilization
Enzyme selectivity
Thiol-epoxy supports
1
. Introduction
Multipoint covalent enzyme immobilization may strongly
tion. Rigidification of surface areas place opposite tothe active site of
the enzyme could improve stability as well as enzyme activity
towards macromolecular substrates (zone c) [19]. Rigidification of
surfaces regions placed next to the active site could alter
conformational changes that occur during catalysis (zone d) [1,7,8].
The best way to obtain a very intense multipoint interaction
between the enzyme and the support is through the reaction
between Lys residues on the enzyme surface and suitable active
groups on highly activated supports [1,13,20,21]. During the last
years, several attempts using these supports had been done in
order to preferentially rigidified enzymes through different zones
of their surface [14,15,21–24]. However, none of the developed
strategies are flexible enough to allow the oriented rigidification of
any region of the enzyme surface.
improve enzyme stability, activity and selectivity [1]. In fact, many
industrial enzymes have been highly stabilized via multipoint
covalent immobilization [1–6]. In addition to that, some enzymes
(e.g., penicillin G acylase from Escherichia coli (PGA) and several
lipases) also exhibit dissimilar selectivities after multipoint covalent
immobilization [1,7–12]. Obviously, the promotion of intense
multipoint covalent attachments on different regions of the enzyme
surface could promote diverse functional effects on the enzyme
(
Fig. 1) [1]. In this context, rigidification of surface areasrich in lysine
moieties (zone a) [13–15] or containing instable loops (zone b) [16–
8] could reduce conformational changes involved in the inactiva-
1
In this work we propose a new multidisciplinary approach that
takes advantage of site-directed mutagenesis not to directly
improve enzyme stability but to orient the covalent rigidification
of enzymes via Lys residues. Taking advantage that the natural
*
Corresponding author. Tel.: +34 91 585 48 09; fax: +34 91 585 47 60.
E-mail address: jmguisan@icp.csic.es (J.M. Guis a´ n).
1
359-5113/$ – see front matter ß 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.procbio.2009.10.013

V. Graz u´ et al. / Process Biochemistry 45 (2010) 390–398
391
Fig. 1. Schematic representation of different zones of an enzyme surface with
special interest for site-directed rigidification. Oriented multipoint covalent
attachment of zones (a) and (b) of the protein surface could reduce the
conformational changes involved in inactivation and this lead to a stabilizing
effect against denaturing agents. Rigidification of zones (c) and (d) could improve
enzyme activity and/or selectivity of an enzyme.
occurrence of surface cysteine residues in proteins is rare [25], the
introduction of Cys residues by site-directed mutagenesis enables to
prepare a library of enzyme derivatives where the orientation
obtained during immobilization on disulfide-supports is fully
determined by the position of the cysteine residue [16,17,26,27].
Thus, combining this genetically modification with a subsequent
multipoint covalent attachment on heterofunctional disulfide-
epoxy supports [28,29], it would be possible to obtain a set of
immobilized derivatives rigidified through any region of the protein
surface (Fig. 2).
The genetic introduction of thiols with the purpose of improving
the stability of an enzyme has been reported previously [16–18].
However, the supports proposed contained only thiol-reactive
groups and hence the degree of multipoint covalent attachment was
low even when more than one thiol was introduced in the target
region. This is therefore the first attempt to combine two strategies
useful for protein stabilization but that had been used till now in a
separate way: genetically introduction of thiols for protein
orientation and additional protein rigidification through multi-
puntual covalent immobilization between Lys group on the protein
and epoxy groups on the support. In order to make a first trial of this
Fig. 3. 3D structure of PGA showing the distribution of lysine residues on its surface.
Space-filled model where lysine residues are represented in dark grey. The 3D
structures where obtained using Grasp version 1.2-1994.
new site-specific protocol for multipoint covalent immobilization,
penicillin G acylase from E. coli (PGA) was selected as model protein.
The enzyme has a known 3D structure [30,31] (Protein Data Bank
[PDB] code: 1AI4) and it has no Cys in its sequence. This enzyme is
veryrichinsurface Lys residues(Fig. 3). Therefore, asa first approach
and proof-of-concept, immobilized enzyme derivatives site-directly
rigidified through different already Lys-rich regions of the protein
surface were prepared (Fig. 4). The different PGA derivatives were
evaluatedregardingtheirstabilityagainstdifferentdistortingagents
(heat or the presence of organic cosolvents), also as selective
biocatalysts for kinetically controlled synthesis [32–35] and as
biocatalysts for the resolution of racemic mixtures of chiral esters by
enantiospecific hydrolysis [36–40].
2. Materials
Polyethyleneimine (PEI, 25 kDa), streptomycin sulfate, sodium dodecyl sulfate
0
(
SDS), 2,2 -dipyridyl disulfide (2-PDS), sodium sulfide (Na
2
S), mercapthoethanol,
dithiotreitol (DTT) glycinamide, phenylacetic acid, glycerol, and isopropyl-beta-D-
thiogalactopyranoside (IPTG), and R(ꢀ)-mandelic acid methyl ester were supplied by
Sigma–Aldrich (St. Louis, MO, USA). R(ꢀ) S(+)-(2)-phenylpropionic acid ethyl ester
was a gift of M. Terreni (University of Pavia, Pavia, Italy). Q-Sepharose Fast-flow was
purchased from Amersham Biosciences (Uppsala, Sweden). Eupergit C was from
Rohm Pharma GmbH (Darmstadt, Germany). Penicillin G was kindly provided by
Antibi o´ ticos S.A. (Le o´ n, Spain). All other reagents were of analytical grade. The
different disulfide and disulfide-epoxy supports were prepared as described in [28]
(Fig. 5).
3. Methods
All experiments were carried out at least in triplicate, and standard errors were
never over 5%.
.1. Bacterial strains, plasmids and culture media
E. coli strains DH5 (laboratory stock) and One Shot Top 10 (Invitrogen) were
3
1
a
used for routine cloning procedures. Overproduction of PGA was performed on E.
1
coli BL21 Star (DE3) (Invitrogen). The E. coli strains were routinely cultured at
Fig. 2. Mechanism of oriented multipoint covalent immobilization of Cys-enzymes
on heterofunctional supports containing thiol-reactive disulfides and epoxide
groups (HDE supports). (a) At neutral pH, Cys-mutated enzymes are firstly linked to
the support via thiol-disulfide exchange, (b) a further incubation under alkaline
conditions promotes an intense intramolecular multipoint attachment between
vicinal Lys group and a high density of epoxy groups on the support surface.
37 8C in Luria-Bertani (LB) broth, either in liquid or solid media containing
ampicillin (150
mg/ml). For overproduction experiments, cells were cultured at
lower temperatures (20 8C) to avoid the formation of inclusion bodies [41]. Plasmid
pET101/D-TOPO was used for cloning purposes according to specifications provided
by the supplier (Invitrogen, Paisley, UK). Plasmid pOAF containing the wild-type pac
gene encoding the PGA from E. coli ATCC 11105 was previously described [14].

3
92
V. Graz u´ et al. / Process Biochemistry 45 (2010) 390–398
Fig. 4. Different Cys-variants of PGA: different positions where a cysteine residue was introduced by site-directed mutagenesis. These figures were obtained with Grasp
version 1.2-1994, space-filled model where lysine residues are represented in dark blue. The - and -subunit of PGA are shown in yellow and grey, respectively. Residues
a
b
selected for replacement with cysteine are represented in green: (A) Ser A86, (B) Ser B9, and (C) Ser B201, which are localized at the upper, back or right-lateral face of the
protein respectively, (D) Gln B112, (E) Ala B361, and (F) Gln B380 are in the face of the protein containing the active site. The localization of the different areas is referred to the
face of the PGA containing the active site.
3
.2. Site-directed mutagenesis of PGA
unwanted mutations, and were named as pA86C, pB9C, pB112C, pB201C,
pB361C and pB380C, respectively. Expression and purification of native and
mutant enzymes was performed using E. coli BL21 cells (DE3) as described in
The plasmid pOAF was used as template for the construction of the SA86C,
SB9C, QB112C, SB201C, AB361C and QB380C single mutants by a site-directed
mutagenesis PCR technique [42]. Primers used for mutagenesis provided
by Genotek (Madrid, Spain) are shown in Table 1. Plasmids containing the
desired mutations were completely sequenced to verify the absence of
[
41].
Protein concentration was measured using Bradford’s reagent [43]. The purity of
proteins was checked on 12% (w/v) SDS-PAGE gels and Coomassie staining as
described by Laemmli [44].
Fig. 5. Schematic representation of all the different supports used in this paper for enzyme immobilization. Monofunctional disulfide (control) supports (MD),
monofunctional epoxy supports (ME) and heterofunctional thiol-reactive disulfide/epoxy supports (HDE).

V. Graz u´ et al. / Process Biochemistry 45 (2010) 390–398
393
Table 1
PCR primers used for site-directed mutagenesis of pac gene. The residues replaced by cysteine are outlined in bold. See further
details in Section 3.
Primer
Oligonucleotide sequence
0
0
SA86C
Sense: 5 -CTATCCGGGCGCAAATTGCTGCCCTTTGCCCAGAGGATATGTCCATTC-3
0
0
Antisense: 5 -GAATGGACATATCCTCTGGGCAAAGGGCAGCAATTTGCGCCCGGATAG-3
0
0
SB9C
Sense: 5 -GCAATATGTGGGTGATCGGCAAATGCAAAGCCCAGGATGCGAAAGC-3
0
0
Antisense: 5 -GCTTTCGCATCCTGGGCTTTGCATTTGCCGATCACCCACATATTGC-3
0
0
QB112C
SB201C
AB361C
QB380C
Sense: 5 -CCATTACGGTGAAAAATGGTTGCGCAGAGACCTTTACTGTCTGG-3
0
0
Antisense: 5 -CCAGACAGTAAAGGTCTCTGCGCAACCATTTTTCACCGTAATGG-3
0
0
Sense: 5 -GCTTATCCAGATCGTCAATGCGGCCATGATCCGCG-3
0
0
Antisense: 5 -CGCGGATCATGGCCGCATTGACGATCTGGATAAGC-3
0
0
Sense: 5 -GAAGCGTACCGTAGTGTGCGCCGTACCTATGCCATTTG-3
0
0
Antisense: 5 -CAAATGGCATAGGTACGGCGCACACTACGGTACGCTTC-3
0
0
Sense: 5 -GTGGCTACGAAACAACCTGCGACGGCCCAACTGGTTCGC-3
0
0
Antisense: 5 -GCGAACCAGTTGGGCCGTCGCAGGTTGTTTCGTAGCCAC-3
3
.3. Determination of penicillin G acylase activity
3.9. Kinetically controlled synthesis of R(ꢀ)-glycinamide mandelate
Enzyme amidase activity was determined using NIPAB [45]. The assay was
10 ml of 10 mM R(ꢀ)-mandelic acid methyl ester and 50 mM of glycinamide in
50 mM sodium phosphate at pH 7.5, were incubated with the different PGA
preparations (2 UI measured with NIPAB as substrate). The reactions were
performed at 25 8C, and the pH value of the solutions was kept constant by
automatic titration. Aliquots of the reaction mixtures were withdrawn at different
reaction times and analyzed by HPLC.
performed with 0.15 mM NIPAB in 50 mM sodium phosphate pH 7.5 and 25 8C. The
initial rate of enzymatic hydrolysis was measured continuously following the
increase of absorbance at 405 nm for 5 min. One International Unit (IU) of PGA
activity was defined as the amount of enzyme that hydrolyzes 1
per minute under the specified conditions.
mmol of substrate
The chromatographic conditions used were: column Kromasil C18
50 ꢁ 4.6 mm (An a´ lisis V ı´ nicos, S. L, Madrid); temperature: 25 8C; eluent: 30%
methanol in 10 mM ammonium phosphate at pH 3.0; flow rate: 1 ml/min;
5 mm
1
3
.4. Immobilization of native PGA and its Cys-mutants
Pure PGA (native PGA or Cys-mutants) (10 ml), containing 15 IU of activity
l
:
254 nm. Retention times: glycinamide 2.0 min, R(ꢀ)-glycinamide mandelate
3.5 min, R(ꢀ)-mandelic acid 5.0 min and R(ꢀ)-mandelic acid methyl ester
against NIPAB in 50 mM sodium phosphate at pH 7.0, were added to 1 g of
different supports. The immobilization mixtures were gently stirred at 25 8C.
Periodically, samples of supernatants were taken and analyzed for enzyme
activity. The immobilized enzymes were then thoroughly washed with 50 mM
sodium phosphate pH 7.0. A control with soluble enzyme preserves full activity
under the immobilization conditions; therefore the decrease in activity in the
supernatant of the ‘‘immobilization suspension’’ can be directly correlated to the
amount of the enzyme immobilized. In some cases, enzymes immobilized on
disulfide/epoxide supports at pH 7.0 were additionally incubated in 0.1 M
sodium bicarbonate at pH 10 containing 20% of glycerol and 100 mM
phenylacetic acid, for 48 h at 25 8C (in order to promote enzyme-epoxide
multiinteraction) [46]. Finally, the immobilized enzymes were incubated in 3 M
glycine at pH 8.5 for 24 h at room temperature to block any remaining epoxy
group [46].
1
13 min. Vs/Vh was calculated when 20% of the ester had been consumed, as
the ratio between concentration of amide and concentration of acid.
3.10. Hydrolysis of R or S-2-phenylpropionic acid ethyl ester
The enantiospecificity of the enzyme derivatives in the hydrolysis of 2-
phenylpropionic acid ethyl ester was evaluated by measuring separately the initial
rate of hydrolysis of the R and S enantiomers of the ester. Reactions were performed
at pH 7.0 and 25 8C using 1 mM of ester. The pH was kept constant by automatic
titration during the reaction. Aliquots of the reaction suspensions were withdrawn
at different reaction times, and the hydrolysis rates were evaluated monitoring the
acid formation when less than 20% of ester had been consumed.
The reaction was followed using HPLC, employing a Kromasil column C18 5
50 ꢁ 4.6 mm (An a´ lisis V ı´ nicos, S. L, Madrid); mobile phase was: 50% acetonitrile in
10 mMsodium phosphateatpH3.1;flow rate: 1 ml/min; : 210 nm. Retention times:
-phenylpropionic acid: 2.7 min, (2)-phenylpropionic acid ethyl ester: 11 min.
One International Unit (IU) of PGA activity was defined as the amount of enzyme
that hydrolyzes 1 mol of substrate per minute under the specified conditions.
mm
1
3
.5. Incubation of the immobilized enzyme with DTT
l
2
The insoluble enzyme derivatives obtained (1 g) were incubated in 10 ml of
4
00 mM DTT in 50 mM sodium phosphate at pH 8.0 at 25 8C for 1 h. Afterwards, the
m
protein and enzyme activity of the supernatants were determined. As DTT is a
disulfide reducing agent, enzyme molecules covalently linked to the support only
through disulfide bonds are released from the support [28].
Enantiospecificity (E) was defined as the ratio between the activity of the enzyme
preparations in the hydrolysis of the enantioisomers R and S.
3
.6. SDS-PAGE analysis of the immobilized preparations
4. Results and discussion
The immobilized derivatives were boiled in the presence of SDS and
4
.1. Selection of six Cys-variants of PGA in surface regions which are
mercapthoethanol. This treatment releases any enzyme subunit that is not
covalently bound to the support or attached only by disulfide bonds [28].
rich in Lys residues
3
.7. Thermal stability assays of different PGA preparations
As shown in Fig. 4, PGA has at least six different surface
regions that are quite rich in Lys groups. Three of these regions
are placed relatively close to the active site. The other three
regions are located in the upper, right-lateral, and back-face of
the protein molecule, taking the active site as reference. Six
different enzyme variants were designed according to the
following criteria: (a) one Cys was introduced by site-directed
mutagenesis in each of the selected Lys-rich regions, (b) Cys was
introduced by replacing amino acid residues fairly exposed to
the medium, (c) the most conservative replacements were
preferred: (i) in three of the selected zones, Cys could be
introduced by a very conservative replacement of serine residues
Aliquots of different PGA preparations were incubated in 25 mM sodium
phosphate at pH 7.0 and 50 8C. Samples were withdrawn periodically and
enzymatic activities were calculated on hydrolysis of NIPAB as described above.
PGA residual activity was expressed as a percentage of initial activity at the given
incubation time. The stabilization factor was calculated as the ratio between the
half-life of the reference and that of the studied derivative.
3
.8. Immobilized PGA stability in the presence of cosolvents
Aliquots of different PGA preparations were incubated in 60:40 (v-v) dioxane:
2
5 mM sodium phosphate at 4 8C (final pH was 7.0). Samples were withdrawn
periodically and enzymatic activities were calculated from the hydrolysis of
NIPAB as described above. PGA residual activity was expressed as the percentage
of initial activity at the given incubation time. The stabilization factor was
calculated as the ratio between the half-life of the reference and that of the studied
derivative.
(
Ser A86, Ser B201 and Ser B9). (ii) In the other three regions less
conservative replacements had to be designed (Gln (B112), Gln
(B380) and Ala (B361)) (Fig. 4).

3
94
V. Graz u´ et al. / Process Biochemistry 45 (2010) 390–398
The mutant enzymes were produced in all cases with yields
similar to that achieved for the native enzyme (approximately
00 IU/L). Moreover, the specific activities (ranging between 28
and 32 IU/mg), the activity/pH profile (with the maximum at pH 8)
47] and the thermal stabilities (half-lives around 7 h at 50 8C and
Moreover, in all cases, more than 90% of enzymatic activity was
recovered after the whole immobilization/rigidification process. It
is important to point out, that the addition of stabilizing agents
during this additional incubation step at so high pH values
(glycerol, reversible inhibitors, etc.), allow the maintenance of the
enzymatic activity in a wide number of cases [1,23,46]. However, if
incubation at pH 10 it is not possible, this can be offset by lowering
the pH and increasing the time of incubation. As the covalent
reaction between Lys and epoxy groups it is now intramolecular
instead of intermolecular, it could happen at reasonable intervals
of time at lower pHs than the pka of Lys moieties.
3
[
pH 7) of all enzyme variants were almost identical to the values
obtained with the native enzyme.
4.2. Immobilization of Cys-variants of PGA on monofunctional
disulfide supports and heterofunctional disulfide-epoxy supports at
neutral pH
After 24 h in 25 mM sodium phosphate at pH 7 and 25 8C,
immobilization of native PGA is negligible on all the employed
supports. On the other hand, the 6 variants of Cys-PGA are very
slowly immobilized on ME supports (less tan 5% at pH 7.0 after
4.4. SDS-PAGE analysis of multipoint covalent immobilization
A complementary analysis of immobilized PGA was performed
based in the existence of two different polypeptide chains in PGA
(24 and 62 kDa) [31]. The boiling of the immobilized enzymes in
the presence of SDS and mercaptoethanol (a disulfide reducing
agent), produces the release of all protein subunits that are not
covalently bound to the support via epoxy chemistry. As an
example, Fig. 6 shows the SDS-PAGE analysis of the proteins
presented in the supernatants obtained after applying this
treatment to the different derivatives obtained with CysB380-
PGA variant, being the results with the other enzyme variants very
similar. As expected, the boiling of all MD derivatives promotes the
release to the medium of all PGA molecules (both enzyme
subunits). PGA derivatives immobilized on HDE supports but
not incubated at pH 10.0 exhibited a partial elution of both
subunits after boiling. However, the release of any PGA subunit
was not observed in any case when the enzyme was immobilized
on HDE supports at pH 7 for 24 h and then incubated at pH 10 for
48 h at 25 8C. All these results clearly demonstrate that all Cys-PGA
variants immobilized following the optimized rigidification
protocol presented at least three enzyme-support linkages: one
epoxy bond per enzyme subunit, and one disulfide bridge per
2
4 h) but all variants were immobilized quite rapidly on MD
supports (around 85–90% at pH 7 and 8 h) with a very similar rate.
The thiol-disulfide exchange between the thiol group of each
inserted Cys residue and the low concentration (5
mmols/g) of
disulfide groups in MD supports is much more rapid than direct
nucleophilic reaction of Cys with the epoxy groups present in high
concentration in ME support (100
mmols/g). Considering the rapid
and similar immobilization rates on these supports, the new Cys
residues genetically introduced on the PGA surface seem to be
really fairly exposed to the medium.
When the mutant Cys-PGA enzymes were immobilized on HDE
support immobilization rates were very similar to those obtained
with the control MD support (85–90% of immobilization after 8 h
at pH 7.0). When the enzyme variants immobilized on this support
were incubated in the presence of DTT after 8 h of incubation at pH
7
.0, more than 80% of the enzyme activity was desorbed from the
support (this reagent should break disulfide bonds). The percen-
tage of released enzyme was 100% using MD supports. Thus, the
formation of a single disulfide bridge by thiol-disulfide exchange
between the inserted Cys and the reactive disulfide groups on the
heterofunctional support seems to be the main mechanism of
immobilization of the enzymes on disulfide-epoxy supports at pH
7
.0. On the other hand, the percentage of eluted enzyme by
incubation with DTT decreased when the incubation time
increased. After 24 h of incubation at pH 7.0, around 60% of PGA
was released from the support, while 40% of the PGA remained
immobilized on the support. These enzyme molecules should be
additionally attached to the support via irreversible covalent
linkages between nucleophiles of the protein surface and the
epoxy groups of the support [28]. These results showed that epoxy
groups are much more reactive for intramolecular reactions than
for intermolecular ones [23,28,48,49]: less than 3% of native PGA
was immobilized at pH 7.0 after 24 h on highly activated ME
supports via intermolecular reaction while 40% was immobilized
via intramolecular reaction of PGA immobilized on HDE supports
with epoxy groups after the same time period.
4.3. Long-term incubation of the immobilized PGA under alkaline
conditions
After immobilization at neutral pH values for 24 h, an additional
incubation of the disulfide-epoxy enzyme derivatives for 48 h at
pH 10 was performed [46]. This additional incubation under
alkaline conditions completely prevents the elution of the Cys-PGA
variants from the support by DTT treatment. The higher rate of
epoxy-enzyme reaction under alkaline conditions is a consequence
of the increase of reactivity of the nucleophiles of the enzyme able
to react with the epoxy groups of the support [23]. Blanks of
soluble enzyme in the presence of phenylacetic acid plus glycerol
were fully stable when incubated at room temperature at pH 10.
Fig. 6. SDS-PAGE analysis of the supernatant obtained after boiling PGA derivatives
in the presence of SDS and mercaptoethanol. The results obtained with the enzyme
derivatives of CysB380-PGA variant are shown. All enzyme derivatives used had a
similar amount of immobilized protein. Molecular weight markers: Lane 1; enzyme
derivative prepared onto control support (MD): Lane 2; enzyme derivative prepared
onto heterofunctional thiol-reactive disulfide/epoxy support (HDE): incubated 24 h
at pH 7.0 (Lane 3), and forwardly additionally incubated at pH 10 for 48 h (Lane 4).
Similar results were obtained with the other Cys-PGA variants immobilized on HDE
supports.

V. Graz u´ et al. / Process Biochemistry 45 (2010) 390–398
395
Fig. 8. Time-courses of cosolvent inactivation of Cys-mutant derivatives prepared
onto heterofunctional supports. Inactivation experiments were carried out in the
presence of 60% of dioxane in sodium phosphate 25 mM pH 7.0 at 4 8C. Enzyme
activity was determined as described in Section 3. Optimal rigidified enzyme
derivatives prepared onto heterofunctional supports (HDE) after 48 h of incubation
at pH 10: (*) CysA86-PGA; (&) CysB9-PGA; (^) CysB112-PGA; ( ) CysB201-PGA;
Fig. 7. Time-courses of thermal inactivation of Cys-mutant derivatives prepared
onto heterofunctional supports. Immobilized derivatives were incubated at 55 8C in
sodium phosphate buffer 25 mM pH 7.0. Enzyme activity was determined as
described in Section 3. Optimal rigidified enzyme derivatives prepared onto
heterofunctional supports (HDE) after incubation at pH 10 for 48 h: (*) CysA86-
PGA; (&) CysB9-PGA; (^) CysB112-PGA; ( ) CysB201-PGA; (+) CysB361-PGA; (~)
CysB380-PGA. The enzyme derivative of CysB380-PGA obtained onto control
support (MD) (~) represents the inactivation course of control derivatives of all
Cys-PGA mutants, as differences among them were not significant.
(+) CysB361-PGA; (~) CysB380-PGA. The enzyme derivative of CysB380-PGA
obtained onto control support (MD) (~) represents the inactivation course of
control derivatives of all Cys-PGA mutants, as differences among them were not
significant.
enzyme molecule. A more precise quantitative analysis of the
degree of multipoint attachment (e.g., determination of the
percentage of modified Lys residues by amino acid analysis of
the protein [20]) could not be performed in this case because of
severe interferences caused by the epoxy-acrylic composition of
the commercial support used.
4.5.2. Stability of immobilized PGA in the presence of organic
cosolvents
The stability of the immobilized enzymes was also studied in
the presence of 60% of dioxane at pH 7.0 and 4 8C (Fig. 8). Again, all
the one-point immobilized preparations presented a very similar
stability (results not shown). Once more, all the enzymes
immobilized on HDE supports following the optimal rigidification
protocol presented higher stabilities than the respective control
derivatives. The observed stabilities were also different depending
on the location of the inserted Cys residue on the enzyme surface.
In this case, the most stable immobilized preparation was the one
obtained with CysB380-PGA. Curiously, the disulfide-epoxy
derivative of CysB201-PGA (with very similar thermal inactivation
profile to the one of CysB380-PGA HDE derivative) presented now a
much lower stability against cosolvents than other HDE immo-
bilized PGAs variants. Thus, distortions of the PGA structure
promoted by heat and by organic cosolvents seem to affect
different areas of the protein.
4.5. Stability of the different immobilized derivatives of Cys-variant
enzymes
4.5.1. Thermal stability
The stabilities of all one-point immobilized (on MD supports)
and free mutant enzymes were very similar (half-life times around
h at pH 7.0 and 50 8C). This result suggested that the one-point
7
immobilization through the introduced Cys in any of the six
regions studied did not produce any stabilization of PGA.
Fig. 7 shows the inactivation courses of the optimal rigidified
derivatives of the six variants of PGA. It is clear that all immobilized
preparations using HDE supports presented a higher stability than
the corresponding preparations using the MD support. These
results suggested that a certain stabilization of the enzyme was
achieved via multipoint covalent attachment of all different
regions where the Cys residue has been introduced. However,
the stability of the different preparations using HDE supports
strongly depended on the region of the protein where the Cys was
placed; being the most stable those obtained using the enzyme
variants bearing the Cys residue in the position B201 or B380
4.6. Effects of the site-directed multipoint covalent attachment on the
enzyme catalytic properties
4.6.1. Enantiospecificity of the different enzyme preparations
The enantiospecific hydrolysis of R and S enantiomers of 2-
phenylpropionic acid ethyl ester catalyzed by all the biocatalysts
was studied (Fig. 9). Control derivatives of all variants exhibited an
enantiospecificity value (E) almost identical: the activity versus the
R isomer was 8–9-times higher than against the S isomer. This
value was almost identical to that of the native enzyme (Table 2).
The multipoint attached derivatives of 5 different variants also
exhibited the same enantiospecificity. However, one exception
was observed. The multipoint attached derivative of the CysB380
(around 30-fold more stable than the one-point immobilized
enzymes). Remarkably, both Cys residues are located in very
different regions of the enzyme surface, i.e., CysB380 is very near to
the active site, while CysB201 is placed in the right-lateral face, far
away from the enzyme active site (see Fig. 4).
Fig. 9. Scheme of the resolution of R/S 2-phenylpropionic acid ethyl ester catalyzed by immobilized PGAs.

3
96
V. Graz u´ et al. / Process Biochemistry 45 (2010) 390–398
Fig. 10. Scheme of the kinetically controlled synthesis of R(ꢀ)-glycinamide mandelate catalyzed by immobilized PGAs.
Table 2
variant of PGA exhibited an E value of 16 (again favouring the
R isomer), a 2-fold higher than the one of the native enzyme
Enantiospecificity of different Cys-PGA derivatives in the hydrolysis of R or S 2-
phenylpropionic acid ethyl ester. Experiments were performed as described in
Section 3. E value is expressed as activity versus R isomer/activity versus S isomer.
(
Table 2).
Enzyme
Enantiospecificity
4
.6.2. Kinetically controlled synthesis of R(ꢀ)-glycinamide mandelate
CysA86-PGA
The synthesis of R(ꢀ)-glycinamide mandelate from methyl
Control derivative
Optimal derivative
7.8 ꢂ 0.3
8.1 ꢂ 0.2
ester of R-mandelic acid and glycinamide was studied (Fig. 10). In
this reaction, the ratio between initial rate of synthesis of amide
and the initial rate of hydrolysis of the ester is a key parameter (Vs/
CysB9-PGA
Control derivative
Optimal derivative
7.6 ꢂ 0.4
7.9 ꢂ 0.2
Vh
native enzyme exhibited the same Vs/Vh
very similar maximum synthetic yield (around 30% of amide)
Table 3). Only the multipoint attached derivative of the Cys-B380
1
) [50]. Again, the derivatives of 5 variants and derivatives of
1
ratio (around 0.9) and a
CysB112-PGA
Control derivative
Optimal derivative
8.4 ꢂ 0.3
8.0 ꢂ 0.2
(
variant exhibited different and improved properties: a synthesis/
hydrolysis ratio of 1.8 and a synthetic yield of 49% (Table 3).
CysB201-PGA
Control derivative
Optimal derivative
8.3 ꢂ 0.3
8.1 ꢂ 0.3
5
. Conclusions
CysB361-PGA
Control derivative
Optimal derivative
8.1 ꢂ 0.3
8.3 ꢂ 0.2
A new procedure for the promotion of multipoint covalent
attachment of different regions of the surface of PGA from E. coli
has been developed. This procedure involves a multidisciplinary
combination of molecular biology (preparation of different Cys-
mutants of PGA in each protein region that is rich in Lys groups),
the preparation of tailor-made heterofunctional supports (with a
very low concentration of highly reactive disulfide groups and a
very high concentration of poorly reactive epoxy groups) [28] and
the development of a two-step immobilization protocol [21,23,28].
This procedure allows the multipoint covalent attachment of each
region of PGA surface that is rich in Lys groups: at least three
covalent linkages between the enzyme and the support were
formed in all cases. All these rigidified derivatives were moderately
more stable against inactivating effects promoted by heat or
organic cosolvents than the pure and soluble PGA or one-point
immobilized derivatives. However, stability and selectivity of
derivatives were different depending on the region where multi-
point attachment was promoted:
CysB380-PGA
Control derivative
Optimal derivative
7.7 ꢂ 0.4
16.8 ꢂ 0.3
Table 3
Kinetic parameters in the synthesis of R(ꢀ)-glycinamide mandelate catalyzed by
different derivatives of different Cys-PGA mutants. Vs/Vh ratio was calculated
1
when only a 20% of the ester had been consumed. Experimental conditions: sodium
phosphate buffer 50 mM pH 7.5, 25 8C, 10 mM R(ꢀ)-mandelic acid methyl ester and
5
0 mM of glycinamide. Other specifications are described in Section 3.
Enzyme
T Vs/Vh
1
Maximum yield (%)
CysA86-PGA
Control derivative
Optimal derivative
0.90 ꢂ 0.05
0.86 ꢂ 0.04
32 ꢂ 3
35 ꢂ 2
CysB9-PGA
Control derivative
Optimal derivative
0.84 ꢂ 0.03
0.89 ꢂ 0.05
34 ꢂ 3
31 ꢂ 1
CysB112-PGA
a. Multipoint attachment in the vicinity of the residue B380 gave
the best results in all experiments. This derivative was 30-fold
more stable than the corresponding one-point covalently
immobilized derivative in experiments of cosolvent inactiva-
tion. It also was the most stable in experiments of thermal
inactivation and it exhibited the highest enantioselectivity and
Control derivative
Optimal derivative
0.91 ꢂ 0.03
0.87 ꢂ 0.02
30 ꢂ 2
29 ꢂ 3
CysB201-PGA
Control derivative
Optimal derivative
0.88 ꢂ 0.04
0.85 ꢂ 0.03
30 ꢂ 2
36 ꢂ 4
CysB361-PGA
1
Vs/Vh in kinetically controlled synthesis. This region is very
Control derivative
Optimal derivative
0.90 ꢂ 0.02
0.93 ꢂ 0.03
33 ꢂ 3
31 ꢂ 2
close to the area of the active center involved in the recognition
of the acyl donor. The adsorption of acyl donors promotes
significant conformational changes during PGA catalysis [31,51]
and hence it may be logical that a certain ‘‘rigidification’’ of this
region turns out to be crucial for stability and selectivity [1].
CysB380-PGA
Control derivative
Optimal derivative
0.93 ꢂ 0.04
1.80 ꢂ 0.03
31 ꢂ 3
49 ꢂ 4

V. Graz u´ et al. / Process Biochemistry 45 (2010) 390–398
397
[
12] Wang PY, Tsai SW, Chen TL. Improvement of enantioselectivity and stability of
Klebsielia oxytoca hydrolase immobilized on Eupergit C 250L. J Chem Technol
Biotechnol 2008;83(11):1518–25.
b. Multipoint attachment in the vicinity of residue B201 (quite far
from the active center) is also quite relevant in thermal
inactivation but it is not relevant for cosolvent inactivation.
These results remark the complexity of PGA inactivation and
that the involvement of different enzyme regions during
inactivation may be different against distinct distorting agents.
c. Curiously, the region vicinal to B361 is also very close to the
enzyme active center (in the vicinity of the area involved in the
recognition of nucleophiles) [31] but it does not play a quite
relevant role in any kind of inactivation and the resulting
derivative do not exhibit altered selectivity in the reactions
studied.
[13] Mateo C, Palomo JM, Fuentes M, Betancor L, Grazu V, L o´ pez-Gallego F, et al.
Glyoxyl agarose: a fully inert and hydrophilic support for immobilization and
high stabilization of proteins. Enzyme Microb Technol 2006;39(2):274–80.
[
14] Abian O, Graz u´ V, Hermoso J, Gonz a´ lez R, Garc ı´ a JL, Fern a´ ndez-Lafuente R, et al.
Stabilization of penicillin G acylase from Escherichia coli: site-directed muta-
genesis of the protein surface to increase multipoint covalent attachment.
Appl Environ Microbiol 2004;70(2):1249–51.
[
[
[
[
15] Rayn BJ, O’F a´ g a´ in C. Arginine-to-lysine substitutions influence recombinant
horseradish peroxidase stability and immobilization effectiveness. BMC Bio-
technol 2007;7:86.
16] Mansfeld J, Vriend G, Vanden Burg B, Eijsink V, Ulbrich-Hofmann R. Probing
the unfolding region in a thermolysin-like protease by site-specific immobi-
lization. Biochemistry 1999;38:8240–5.
17] Mansfeld J, Ulbrich-Hofmann R. Site-specific and random immobilization of
thermolysin-like proteases reflected in the thermal inactivation kinetics.
Biotechnol Appl Biochem 2000;32:185–95.
18] Ulbrich-Hofmann R, Arnold U, Mansfeld J. The concept of the unfolding region
for approaching the mechanisms of enzyme stabilization. J Mol Catal B Enzym
1999;7:125–31.
The best stabilization factors obtained are not as high as the
ones we had already reported for this enzyme [13,21]. Perhaps
disulfide-epoxy Eupergit are not the best supports for PGA
stabilization [1,20]. But these supports can be easily prepared
from commercial Eupergit C and they can be the best ones for the
main objective of this work: the rapid detection of key regions of
the surface of recombinant enzyme for an improvement of its
stability and selectivity. From this rapid detection a further
improvement of the stabilization factors can be done using: (i)
protein engineering (molecular evolution, etc.), and/or (ii)
immobilization methods, e.g.: by enrichment of key regions with
more Lys residues (replacement of existing Arg ones), by selection
of the best support (supports with other internal morphology:
large internal surface with good geometrical congruence with the
enzyme surface), by selection of the best immobilization-
stabilization protocol and so on. All these additional approaches
will be the matter of forthcoming papers.
[19] Pessela BCC, Torres R, Fuentes M, Mateo C, Fernandez-Lafuente R, Guis a´ n JM.
Immobilization of rennet from Mucor miehei via its sugar chain. Its use in milk
coagulation. Biomacromolecules 2004;5:2029–33.
[
20] Pedroche J, Yust MM, Mateo C, Fern a´ ndez-Lafuente R, Gir o´ n-Calle J, Alaiz M,
et al. Effect of the support and experimental conditions in the intensity of the
multipoint covalent attachment of proteins on glyoxyl-agarose supports:
correlation between enzyme-support linkages and thermal stability. Enzyme
Microb Technol 2007;40:1161–7.
[21] Mateo C, Graz u´ V, Pessela BCC, Montes T, Palomo JM, Torres R, et al. Advances
in the design of new epoxy supports for enzyme immobilization-stabilization.
Biochem Soc Trans 2007;35(6):1593–601.
[
22] Mateo C, Graz u´ V, Palomo JM, Lopez-Gallego F, Fernandez-Lafuente R, Guisan
JM. Immobilization of enzymes on heterofunctional epoxy supports. Nat
Protoc 2007;2(5):1022–33.
23] Mateo C, Abian O, Fern a´ ndez-Lafuente R, Guis a´ n JM. Multifunctional epoxy-
supports. A new tool to improve the covalent immobilization of proteins: the
promotion of physical adsorption on the supports before their covalent
linkage. Biomacromolecules 2000;1:739–45.
[
[
24] Cecchini AD, Serra I, Ubiali D, Terreni M, Albertini AM. New active site oriented
glyoxyl-agarose derivatives of Escherichia coli penicillin G acylase. BMC Bio-
technol 2007;7:54.
Acknowledgments
[
[
25] Hermanson GT, editor. Bioconjugate techniques. New York: Academic Press;
1996. p. 8–11.
The authors gratefully recognize the support from the Spanish
CICYT (project. BIO-2005-8576). We gratefully recognize MEC for a
fellowship for L o´ pez-Gallego and Montes. We thank Antibi o´ ticos
SA for the kind donation of reagents. The help and comments from
Dr. Angel Berenguer (Universidad de Alicante) are gratefully
recognized.
26] Torrance L, Ziegler A, Pittman H, Paterson M, Toth R, Eggleston I. Oriented
immobilization of engineered single-chain antibodies to develop biosensors
for virus detection. J Virol Methods 2006;134(1–2):164–70.
27] Madoz-G u´ rpide J, Abad JM, Fern a´ ndez-Recio J, Velez M, V a´ zquez L, G o´ mez-
Moreno C, et al. Modulation of electroenzymatic NADPH oxidation through
oriented immobilization of ferredoxin: NADP+ reductase onto modified gold
electrodes. J Am Chem Soc 2000;122:9808–17.
[
[
28] Graz u´ V, Abi a´ n O, Mateo C, Batista-Viera F, Fern a´ ndez-Lafuente R, Guis a´ n JM.
Stabilization of enzymes by multipoint immobilization of thiolated proteins
on new epoxy-thiol supports. Biotechnol Bioeng 2005;90:597–605.
29] Graz u´ V, Abian O, Mateo C, Batista-Viera F, Fern a´ ndez-Lafuente R, Guis a´ n JM.
Novel bifunctional epoxy/thiol-reactive support to immobilize thiol contain-
ing proteins by the epoxy chemistry. Biomacromolecules 2003;4(6):1495–
References
[
[
1] Mateo C, Palomo JM, 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.
501.
[
[
[
[
[
[
2] Gupta MN. Thermostabilization of proteins. Biotechnol Appl Biochem
[
[
[
[
[
30] Hunt PD, Tolley SP, Ward RJ, Hill CP, Dodson GG. Expression, purification and
1
991;14:1–11.
3] Gianfreda L, Scarfi MR. Enzyme stabilization: state of the art. Mol Cell Biochem
991;100:97–128.
crystallization of penicillin G acylase from E. coli ATCC 11105. Prot Eng
1990;3:635–9.
1
31] McVey CE, Walsh MA, Dodson GG, Wilson KS, Brannigan JA. Crystal structures
of penicillin acylase enzyme-substrate complexes: structural insights into the
4] Padma I. Enzyme stability and stabilization—aqueous and non aqueous envir-
onment. Proc Biochem 2007;43:1019–32.
5] Cao L, van Langen L, Sheldon RA. Immobilized enzymes: carrier-bound or
carrier-free? Curr Opin Biotechnol 2003;14:387–94.
6] Sheldon RA. Enzyme immobilization: the quest for optimum performance. Adv
Synth Catal 2007;349:1289–307.
7] Terreni M, Pagani G, Ubiali D, Fern a´ ndez-Lafuente R, Mateo C, Guis a´ n JM.
Modulation of penicillin acylase properties via immobilization techniques:
one-pot chemoenzymatic synthesis of Cephamandole from Cephalosporin C.
Bioorg Med Chem Lett 2001;11:2429–32.
catalytic mechanism. J Mol Biol 2001;313:139–50.
32] Van Langen LM, Oosthoek NHP, Guranda DT, Van Rantwijk F, Sˇ vedas VK,
Sheldon RA. Penicillin acylase-catalyzed resolution of amines in aqueous
organic solvents. Tetrahedron Asymmetry 2000;11:4593–600.
33] Illanes A, Fajardo A. Kinetically controlled synthesis of ampicillin with immo-
bilized penicillin acylase in the presence of organic cosolvents. J Mol Catal B
Enzym 2001;11:587–95.
34] Jager SAW, Jekel PA, Janssen DB. Hybrid penicillin acylases with improved
properties for synthesis of
007;40:1335–44.
b-lactam antibiotics. Enzyme Microb Technol
[
8] Rocchietti S, Urrutia ASV, Pregnolato M, Tagliani A, Guis a´ n JM, Fern a´ ndez-
Lafuente R, et al. Influence of the enzyme derivative preparation and substrate
structure on the enantioselectivity of penicillin G acylase. Enzyme Microb
Technol 2002;31:88–93.
2
[
[
35] Pan S-B, Wu Q, Chen C-X, Lin X-F. Efficient enzymatic synthesis of ampicillin in
organic media. J Mol Catal B Enzym 2008;54:13–8.
36] Li D, Cheng S, Wei D, Ren Y, Zhang D. Production of enantiomerically pure (S)-
[
9] Chaubey A, Parshad R, Koul S, Taneja SC, Qazi GN. Enantioselectivity modula-
tion through immobilization of Arthrobacter sp. lipase: kinetic resolution of
fluoxetine intermediate. J Mol Catal B Enzym 2006;42:39–44.
b-phenylalanine and (R)-b-phenylalanine by penicillin G acylase from Escher-
ichia coli in aqueous medium. Biotechnol Lett 2007;29:1825–30.
37] Cabrera Z, Lopez-Gallego F, Fernandez-Lorente G, Palomo JM, Montes T, Grazu´
V, et al. Asymmetric hydrolysis of dimethyl phenylmalonate by immobilized
penicillin G acylase from E. coli. Enzyme Microb Technol 2007;40:997–1000.
38] Massolini G, Calleri E, Lavecchia A, Loiodice F, Lubda D, Temporini C, et al.
Enantioselective hydrolysis of some 2-aryloxyalkanoic acid methyl esters and
isosteric analogues using a penicillin G acylase-based HPLC monolithic silica
column. Anal Chem 2003;75:535–42.
[
[
[
10] Palomo JM, Mu n˜ oz G, Fern a´ ndez-Lorente G, Mateo C, Fuentes M, Guisan JM,
et al. Modulation of Mucor miehei lipase properties via directed immobilization
on different hetero-functional epoxy resins: hydrolytic resolution of (R,S)-2-
butyroyl-2-phenylacetic acid. J Mol Catal B Enzym 2003;21:120–201.
11] Palomo JM, Segura RL, Mateo C, Terreni M, Guisan JM, Fern a´ ndez-Lafuente R.
Synthesis of enantiomerically pure glycidol via a fully enantioselective lipase-
catalyzed resolution. Tetrahedron Asymmetry 2005;16:869–74.
[

3
98
V. Graz u´ et al. / Process Biochemistry 45 (2010) 390–398
[
[
[
39] Topgi RS, Ng JS, Landis B, Wang P, Behling JR. Use of enzyme penicillin acylase
in selective amidation/amide hydrolysis to resolve ethyl 3-amino-4-pentyno-
ate isomers. Bioorg Med Chem 1999;7:2221–9.
[46] Mateo C, Abian O, Fern a´ ndez-Lorente G, Predoche J, Fern a´ ndez-Lafuente R,
Guisan JM. Epoxy Sepabeads: a novel epoxy support for stabilization of
industrial enzymes via very intense multipoint covalent attachment. Biotech-
nol Progr 2002;18:629–34.
[47] Arroyo M, De La Mata I, Acebal C, Castill o´ n MP. Biotechnological applications
of penicillin acylases: state-of-the-art. Appl Microb Biotechnol 2003;
60:507–14.
[48] Wheatley JB. Effect of antigen size on optimal ligand density of immobilized
antibodies for a high-performance liquid chromatographic support. J Chro-
matogr A 1991;548:243–53.
[49] Wheatley JB, Schmidt DE. Salt-induced immobilizations of proteins on a high-
performance liquid chromatographic epoxide affinity support. J Chromatogr A
1993;644:11–6.
40] Fadnavis NW, Sharfuddin M, Vadivel SK. Resolution of racemic 2-amino-1-
butanol with immobilised penicillin
999;10:4495–5000.
G
acylase. Tetrahedron Asymmetry
1
41] Keilmann C, Wanner G, Bock A. Molecular basis of the exclusive low-tem-
perature synthesis of an enzyme in E. coli: penicillin acylase. Biol Chem Hoppe-
Seyler 1993;374:983–92.
[
[
42] Braman J, Papworth C, Greener A. Site-directed mutagenesis using double-
stranded plasmid DNA templates. Methods Mol Biol 1996;57:31–44.
43] Bradford MM. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein dye binding. Anal
Biochem 1972;72:248–54.
[50] Giordano RC, Ribeiro MPA, Giordano RLC. Kinetics of
b-lactam antibiotics
[
[
44] Laemmli UK. Cleavage of structural proteins during the assembly of the head
of bacteriophage T4. Nature 1970;227:680–5.
45] Kutzbach C, Rauenbusch E. Preparation and general properties of crystalline
penicillin G acylase from E. coli ATCC 11105. Hoppe-Seyler’s Z Physiol Chem Bd
synthesis by penicillin G acylase (PGA) from the view of point of the industrial
enzymatic reactor optimization. Biotechnol Adv 2006;24:27–41.
[51] Fernandez-Lafuente R, Rosell CM, Guisan JM. Modulation of the properties of
penicillin G acylase by acyl donor substrates during N-protection of amino
compounds. Enzyme Microb Technol 1998;22:583–7.
1
974;354S:45–53.