Stabilized and immobilized bacillus subtilis arginase for the biobased production of nitrogen-containing chemicals

Article abstract of DOI:10.1002/adsc.201000034

L-Ornithine could serve as an intermediate in the biobased production of 1,4-diaminobutane from L-arginine. Using the concept of biorefinery, Larginine could become widely available from biomass waste streams via the nitrogen storage polypeptide cyanophycin. Selective hydrolysis of L-arginine to L-ornithine is difficult to perform chemically, therefore the stabilization and immobilization of Bacillus subtilis arginase (EC 3.5.3.1) was studied in a continuously stirred membrane reactor system. Initial pH of the substrate solution, addition of L-aspartic acid and reducing agents all appeared to have an effect on the operational stability of B. subtilis arginase. A remarkably good operational stability (total turnover number, TTN = 1. 13-10⁸) at the pH of arginine free base (pH 11.0) was observed, which was further improved with the addition of sodium dithionite to the substrate solution (TTN > 1.10⁹). B. subtilis arginase was successfully immobilized on three commercially available epoxy-activated supports. Immobilization on Sepabeads EC-EP was most promising, resulting in a recovered activity of 75% and enhanced thermostability. In conclusion, the stabilization and immobilization of B. subtilis arginase has opened up possibilities for its application in the biobased production of nitrogen-containing chemicals as an alternative to the petrochemical production.

Full text of DOI:10.1002/adsc.201000034

FULL PAPERS
DOI: 10.1002/adsc.201000034
Stabilized and Immobilized Bacillus subtilis Arginase for the
Biobased Production of Nitrogen-Containing Chemicals
Paul M. Kçnst,a,b,* Pedro M. C. C. D. Turras,a,b Maurice C. R. Franssen,b,*
Elinor L. Scott,a and Johan P. M. Sandersa
a
Valorisation of Plant Production Chains, Wageningen University, Bornse Weilanden 9, 6708 WG Wageningen,
The Netherlands
Fax : (+31)-(0)317-483-011; phone: (+31)-(0)317-486-030; e-mail: paul.konst@wur.nl
Laboratory of Organic Chemistry, Wageningen University, Dreijenplein 8, 6703 HB Wageningen, The Netherlands
b
Fax : (+31)-317-484-914; phone: (+31)-317-482-976; e-mail: maurice.franssen@wur.nl
Received: January 15, 2010; Revised: April 22, 2010; Published online: June 8, 2010
Abstract: l-Ornithine could serve as an intermediate was further improved with the addition of sodium di-
in the biobased production of 1,4-diaminobutane thionite to the substrate solution (TTN>1·109). B.
from l-arginine. Using the concept of biorefinery, l- subtilis arginase was successfully immobilized on
arginine could become widely available from bio- three commercially available epoxy-activated sup-
mass waste streams via the nitrogen storage polypep- ports. Immobilization on Sepabeads EC-EP was
tide cyanophycin. Selective hydrolysis of l-arginine most promising, resulting in a recovered activity of
to l-ornithine is difficult to perform chemically, 75% and enhanced thermostability. In conclusion,
therefore the stabilization and immobilization of Ba- the stabilization and immobilization of B. subtilis ar-
cillus subtilis arginase (EC 3.5.3.1) was studied in a ginase has opened up possibilities for its application
continuously stirred membrane reactor system. Ini- in the biobased production of nitrogen-containing
tial pH of the substrate solution, addition of l-as- chemicals as an alternative to the petrochemical pro-
partic acid and reducing agents all appeared to have duction.
an effect on the operational stability of B. subtilis ar-
ginase. A remarkably good operational stability Keywords: amino acids; biotransformations; enzyme
(total turnover number, TTN=1.13·108) at the pH of catalysis; hydrolysis; immobilization; metalloen-
arginine free base (pH 11.0) was observed, which zymes
Introduction
biorefinery concept, it would be more energy efficient
and eventually more economic to start from function-
Global warming, depletion of fossil resources and alized compounds that already have nitrogen incorpo-
geo-political instability are forcing us to make the rated, such as amino acids, instead of starting from
transition towards CO2 neutral, sustainable and eco- naphthaꢀs plain hydrocarbons.[3] Bearing this in mind,
nomically stable alternatives for the production of our attention has been drawn to the polypeptide cya-
chemicals and transportation fuels. In this respect, nophycin (CGP).
biomass has great potential. Currently the emphasis is
CGP consists of a poly(l-aspartic acid) backbone
on the application of biomass as a feedstock for with l-arginine side chains in equimolar amounts and
energy and transport fuels. However, its application is naturally produced in cyanobacteria as a nitrogen
as a feedstock for functionalized bulk chemicals re- storage polymer.[4] Heterologous expression in indus-
ceives far less attention. This is remarkable consider- trially relevant bacteria[5–8] has opened up the possibil-
ing the enormous amounts of functionalized com- ity to produce CGP in considerable amounts. Even
pounds (e.g., amino acids, organic acids and lignins) more interestingly, recent work on Saccharomyces cer-
that are generated as side streams of the biofuel pro- evisiae[9,10] gives rise to possibilities for the simultane-
duction.[1]
Nowadays, most nitrogen-containing bulk chemicals rich waste streams from the agro-industries.[11–13]
ous production of CGP and ethanol from amino acid-
are produced from naphtha. In these processes large
Since CGP is insoluble under physiological condi-
amounts of energy and ammonia are used, for exam- tions, it can be easily isolated[14–16] which, together
ple, in the ammoxidation of olefins.[2] Following the with its defined composition, its integrated production
Adv. Synth. Catal. 2010, 352, 1493 – 1502
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1493

Paul M. Kçnst et al.
FULL PAPERS
Figure 1. Route from cyanophycin towards nitrogen-containing bulk chemicals such as 1,4-diaminobutane with the enzymatic
hydrolysis of l-arginine as an intermediate step.
and its high nitrogen content, make CGP an interest- draws Mn2+ from the reaction solution which might
ing feedstock for the production of nitrogen-contain- result in deactivation.
ing bulk chemicals.[13]
Most examples of studies on the application of argi-
Our research is focused on the development of in- nase are focused on analytical purposes,[30–34] while
tegrated processes for the conversion of CGPꢀs hydro- only a few investigate its application for the produc-
lysate, l-aspartic acid and l-arginine, towards existing tion of l-ornithine.[35–37] One of the more elaborate
nitrogen-containing bulk chemicals. Earlier we inves- studies on the enzymatic l-ornithine production is by
tigated the applicability of l-aspartate a-decarboxy- Bommarius et al.[37] and investigates the operational
lase (EC 4.1.1.11) in the conversion of l-aspartic acid stability of calf liver arginase in membrane reactors.
to b-alanine (3-aminopropionic acid), a possible inter- They observed a good stabilization of arginase upon
mediate in the production of acrylamide and acryloni- addition of l-ascorbic acid, which is attributed to pre-
trile.[17] Starting from l-arginine, we envision a route venting the loss of Mn2+ as a result of oxidation. De-
towards 1,4-diaminobutane, one of the monomers in activation as a result of mechanical stress by stirring,
nylon-4,6 which is commercially available under the however, remained a problem.
name Stanylꢂ (Figure 1).[18,19]
Immobilization might enhance the enzymeꢀs stabili-
In this route l-arginine is hydrolyzed yielding l-or- ty towards conditions that may be encountered in an
nithine and urea. Methods for the non-enzymatic hy- industrial process.[38] On an industrial scale, the immo-
drolysis of l-arginine have limitations due to either bilization of enzymes is desirable, as this enables
slow reaction or to the production of a number of reuse and overcomes problems with product recovery.
side products.[20–23] On the other hand, the enzyme ar- There are several example of arginase immobilization,
ginase (l-arginine amidinohydrolase, EC 3.5.3.1) from either by entrapment in a matrix[32,35] or covalent
the bacterium Bacillus subtilis is capable of selectively binding to a support.[30,31,34,36] However, immobiliza-
hydrolyzing l-arginine to l-ornithine and urea under tion of arginase on commercially available and indus-
mild conditions.[24]
trially relevant epoxy-activated supports was not
For its catalytic activity arginase is dependent on tested until now. A suitable support would be Euper-
two manganese ions (Mn2+), which form a binuclear git, which was one of the first epoxy-activated sup-
cluster in the active site in order to ionize the water ports introduced on the market.[39] Another suitable
molecule that attacks the guanidinium carbon of l-ar- epoxy-activated support would be Sepabeads which,
ginine in the catalytic cycle.[25–27] However, at the pH due to their ease of operation and their robustness
optimum (pH 9–10) of B. subtilis arginase and many under physical stress, are very suitable for enzyme im-
other arginases,[24,28] Mn2+ is readily oxidized by air mobilization for industrial purposes.[40,41]
(Scheme 1).[29]
In this paper, we present the results of a study on
Mn2+ oxidation poses problems for the application the stabilization and immobilization of Bacillus subti-
of arginase, since the formation of precipitates may lis arginase for the production of l-ornithine as an in-
cause operational problems, but most of all it with- termediate for biobased, nitrogen-containing, bulk
chemicals. For the soluble enzyme the conditions for
optimal operational stability were determined using a
continuously stirred membrane reactor system. Fur-
thermore, B. subtilis arginase was immobilized on
three different epoxy-functionalized supports: Sepa-
beads EC-EP and EC-HFA, and Eupergit C 250L.
Scheme 1. Base-induced oxidation of Mn2+ towards MnO2(s)
.
1494
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1493 – 1502

Stabilized and Immobilized Bacillus subtilis Arginase for the Biobased Production
The thermal and operational stabilities of the immobi- otherwise. Furthermore, the total turnover number
lized and soluble enzymes were compared.
(TTN) (mol l-ornithine·molÀ1 arginase) was deter-
mined from the linear relation between the amount of
l-ornithine formed and the amount of arginase spent.
The results are summarized in Table 1 and discussed
below.
Results and Discussion
Operational Stability of Soluble Arginase
The influence of pH, l-aspartic acid and reducing Effect of pH
agents on the operational stability of B. subtilis argi-
nase was investigated, since they affect the enzyme as The operational stability of arginase was tested with
such and the availability of Mn2+ for the enzyme. The substrate solutions at three different pH values: pH
operational stabilities were determined using a con- of non-adjusted substrate solution (pH 11.0), optimum
tinuously stirred membrane reactor system, which was pH as previously determined for B. subtilis arginase
fed through a nitrogen gas pressurized substrate solu- (pH 9.5)[24] and a pH at which Mn2+ oxidation is negli-
tion reservoir. The outflow of the reactor was collect- gible (pH 8.5). Additionally, a run was included with
ed in tubes over set time intervals. After a steady substrate solution at pH 9.5, but without additional
state had been reached the specific reaction rate r Mn2+. The results are depicted in Figure 2.
(mmol·minÀ1·mgÀ1) in the course of time was calculat-
Comparing the results of the experiments with ini-
ed from the flow rate Q (mL·minÀ1), the product con- tial pH 9.5 with and without additional Mn2+, it be-
centration Cp (mmol·mLÀ1) and the amount of enzyme comes clear that the addition of Mn2+ is crucial for
applied Xenz (mg) [Eq (1)].[42]
both activity and stability. While arginase shows a
single exponential deactivation in the presence of
Mn2+ (kdeact =5.88·10À2 hÀ1), deactivation in the ab-
sence of Mn2+ occurs much faster following a model
involving two deactivation constants (kdeact,1
=
4.51·10À1 hÀ1, kdeact,2 =0.36·10À2 hÀ1).
From the data obtained, the extrapolated initial re-
Although no signs of Mn2+ oxidation were ob-
action rate rinit (mmol·minÀ1·mgÀ1) and the deactiva- served, the use of substrate solution with initial
tion constant kdeact (hÀ1) were determined using a pH 8.5 still led to arginase deactivation (kdeact
=
single exponential deactivation model unless stated 3.26·10À2 hÀ1). On the other hand, application of the
Table 1. Summary of results obtained with soluble B. subtilis arginase using a continuously stirred membrane reactor system.
The values are the result of duplicate experiments ÆS.E.
[a]
pH of substrate solution Additional to 0.5 mM MnCl2
rinit [mmol·minÀ1·mgÀ1][b] kdeact [hÀ1][b]
TTN [mol·molÀ1][c]
not adjusted (11.0)
9.5
–
462Æ9
341Æ9
215Æ12
2.50 (Æ0.26)·10À2 1.13 (Æ0.13)·108
0.50 mM DTN
without MnCl2
stable[d]
stable
4.50 (Æ0.43)·10À1 n.d.
3.58 (Æ0.69)·10À2
–
283Æ4
410Æ7
243Æ6
162Æ4
52.9Æ0.1
336Æ3
383Æ4
303Æ12
289Æ7
326Æ3
329Æ5
5.88 (Æ0.22)·10À2 3.49 (Æ0.09)·107
3.30 (Æ0.22)·10À2 9.11 (Æ0.89)·107
3.65 (Æ0.25)·10À2 4.80 (Æ0.24)·107
1.70 (Æ0.24)·10À2 6.22 (Æ0.57)·107
6.70 (Æ2.10)·10À3 6.82 (Æ0.63)·107
25 mM l-asp
0.06 mM l-asc
0.12 mM l-asc
0.25 mM l-asc
0.25 mM DTN
0.50 mM DTN
1.00 mM DTN
–
stable
stable
stable
stable
stable
stable
8.5
3.26 (Æ0.31)·10À2 6.68 (Æ0.62)·107
25 mM l-asp
0.50 mm DTN
2.75 (Æ0.13)·10À2 8.76 (Æ0.40)·107
stable
stable
[a]
l-asp=l-aspartic acid, DTN=sodium dithionite, l-asc=l-ascorbic acid.
The initial reaction rate rinit and deactivation constant kdeact were determined from fitted first-order deactivation models
unless stated otherwise.
[b]
[c]
[d]
The total turnover number TTN was determined from the linear relation between the amount of l-ornithine formed and
the amount of arginase spent as determined from the residual activity.
Stable: no significant deactivation was observed, at least 1·109 mol l-ornithine were formed per mol arginase spent.
Adv. Synth. Catal. 2010, 352, 1493 – 1502
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1495

Paul M. Kçnst et al.
FULL PAPERS
might be the result of complexation of Mn2+ with l-
aspartic acid. As an l-aspartic acid-manganese com-
plex has
a
much higher dissociation constant
(2.5 mM) compared to the arginase-manganese com-
plex (45 mM), manganese will always be available to
arginase.
In the envisioned process from cyanophycin to-
wards nitrogen-containing chemicals, l-aspartic acid
will be available in equimolar amounts to l-arginine
upon complete hydrolysis (Figure 1). The potentially
stabilizing effect of l-aspartic acid on B. subtilis argi-
nase could simply be utilized by postponing the isola-
tion of l-aspartic acid until the conversion of l-argi-
nine has been performed. Various methods for the
separation of aspartic acid from arginine/ornithine are
available,[46–49] all based on the different charges of
the two amino acids at different pH values. Nonethe-
less, it has to be taken into account that, as a result of
the acidic character of l-aspartic acid, adjustment of
the substrate solutionsꢀ pH will be required, which
will lead to increased process costs.
The stabilizing effect of l-aspartic acid on the op-
erational stability of B. subtilis arginase was tested at
the previously reported optimum pH (pH 9.5)[24] and
at the pH at which Mn2+ oxidation is negligible
(pH 8.5). The results are presented in Table 1. Addi-
tion of l-aspartic acid to the substrate solution has an
overall positive effect on the initial reaction rate, es-
Figure 2. Effect of initial pH on arginase stability. The pH
values of 0.25M arginine solutions were adjusted and as
such applied with soluble arginase in a continuously stirred
membrane reactor at 308C. The specific reaction rates were
determined using Eq. (1). The error bars represent standard
deviation; the depicted lines represent fitted first-order de-
activation models unless stated otherwise.
non-adjusted substrate solution (pH 11.0), which pecially at pH 9.5. Concerning arginaseꢀs stability, the
almost instantly turned brown upon preparation as a addition of l-aspartic acid at pH 9.5 resulted in an
result of Mn2+ oxidation, resulted in a deactivation almost two times smaller deactivation constant
constant slightly lower than that observed with initial (kdeact =3.30·10À2 hÀ1), which is in agreement with the
pH 8.5 (kdeact =2.50·10À2 hÀ1). From these results it can observations of Patchet et al.[45] As described earlier,
be concluded that the observed arginase deactivation this stabilizing effect is assumed to be the result of
is not only due to loss of Mn2+ through oxidation, but Mn2+ complexation with l-aspartic acid, which re-
moreover appears to be the result of another phe- duces the susceptibility to oxidation.
nomenon.
At pH 8.5, arginase was only slightly stabilized
In addition to the higher stability observed, the ini- upon addition of l-aspartic acid (kdeact =2.75·10À2 hÀ1).
tial reaction rate as determined for initial pH 11.0 This is not surprising since Mn2+ oxidation is negligi-
(462 mmol·minÀ1·mgÀ1) is significantly higher than that ble at pH 8.5 and therefore the stabilizing effect of
observed with initial pH 9.5 (283 mmol·minÀ1·mgÀ1), complexation would be small.
the previously reported pH optimum for B. subtilis at
378C.[24]
A similar effect of initial pH on both activity and Effect of Reducing Agents
stability is described in an article on the application
of calf liver arginase.[43] While it was previously deter- Bommarius et al. showed that l-ascorbic acid has a
mined that calf liver arginase has an optimum at stabilizing effect on calf liver arginase.[37] Initially l-
pH 9.5,[44] a significantly higher activity and stability ascorbic acid was applied with the perception that as
was observed with initial pH 11.5 (0.75M arginine) a reducing agent it functions by lowering the reaction
upon application in a membrane reactor system.
mixturesꢀ redox potential and thus prevents the oxida-
tion and concurrent removal of Mn2+.
The stabilizing effect of l-ascorbic acid on the op-
erational stability of B. subtilis arginase was tested by
adding different amounts to a substrate solution with
Effect of l-Aspartic Acid
Patchett et al.[45] reported that addition of a chelating initial pH 9.5. In addition, the stabilizing effect of the
agent such as l-aspartic acid increases the stability of more economic reductant sodium dithionite was
B. caldovelox arginase. They suggest that this effect tested. The results are summarized in Table 1.
1496
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1493 – 1502

Stabilized and Immobilized Bacillus subtilis Arginase for the Biobased Production
First, as suggested by Bommarius et al., 0.25 mM l- zation of B. subtilis arginase when applied in a sub-
ascorbic acid in combination with 0.5 mM Mn2+ was strate solution with initial pH 9.5. To further investi-
tested. Oxidation of Mn2+ was not observed and B. gate its stabilizing ability, sodium dithionite was
subtilis arginase was stabilized, but nonetheless at a tested in substrate solutions with the pH adjusted to
very low initial reaction rate of 52.9 mmol·minÀ1·mgÀ1. 8.5 and with the pH of arginine free base (pH 11.0).
Lowering the l-ascorbic acid concentration led to in-
creased initial reaction rates, but at the expense of significantly reduced, addition of sodium dithionite,
the stability. both at initial pH 8.5 and pH 11.0, did result in stabili-
Although the initial reaction rate at pH 11.0 was
At a concentration of 0.25 mM, sodium dithionite zation comparable to that found with initial pH 9.5
showed a similar stabilizing effect as l-ascorbic acid, (Table 1). However, as observed at pH 9.5, addition
but at a much higher initial reaction rate of 336 mmol·- of 0.5 mM dithionite could not prevent oxidation of
minÀ1·mgÀ1. The initial reaction rate increased upon in- Mn2+ at pH 11.0. Emphasizing the results obtained by
creasing the sodium dithionite concentration to studying the effect of initial pH, the increased stabili-
0.50 mM (383 mmol·minÀ1·mgÀ1) and decreased again ty of arginase appears not only to be related to the
with 1.0 mM (303 mmol·minÀ1·mgÀ1), while maintaining availability of free Mn2+. It might be speculated that
the stabilizing effect. At all three tested concentrations not Mn2+ availability, but the reduced redox potential
of sodium dithionite, arginase showed a good stabiliza- that results from dithionite addition has its effect on
tion, although even 1.0 mM sodium dithionite was not the operational stability of B. subtilis arginase.
sufficient to prevent Mn2+ from oxidizing.
The observed decrease in initial reaction rate with
increased l-ascorbic acid concentration is possibly the Conditions for Optimal Operational Stability
result of enzyme deactivation. To investigate if this
decrease in initial activity is reversible, for example, The lowest operational stability determined for B.
due to Mn2+ chelation, the following experiment was subtilis arginase in this study (initial pH 9.5 and
performed.
0.5 mM Mn2+, 3.49·107 mol·molÀ1) is already consid-
Immobilized B. subtilis arginase (vide infra) was in- ered to be in the range of good biocatalyst perfor-
cubated in buffer with 0.5 mM Mn2+ and 2 mM l-as- mance for large scale processing (TTN>
corbic acid for 15 min. Subsequently, the l-ascorbic 106 mol·molÀ1)[52]. However, when not adjusting the
acid was removed by extensively washing in buffer pH of the substrate solution the TTN already in-
containing 0.5 mM Mn2+, after which the residual ac- creased to a very high 1.13·108 mol·molÀ1. This high
tivity was determined in the absence of l-ascorbic TTN was not exceeded upon addition of l-aspartic
acid. The same experiment was performed with 2 mM acid, although this had a positive effect at initial pH
sodium dithionite and the results were compared with 8.5 and 9.5.
a blank experiment in which reducing agents were
omitted.
Addition of sodium dithionite to the substrate solu-
tion at any initial pH led to an impressive stabiliza-
While 15 min of incubation with sodium dithionite tion. Over the observed time period no significant de-
had no negative effect on the initial activity, incuba- activation could be observed upon addition of sodium
tion with l-ascorbic acid resulted in an irreversible dithionite, so the TTNs are estimated to be larger
decrease in initial activity of 74%.
than 1·109 mol·molÀ1.
It is not clear how l-ascorbic acid lowers the initial
Since addition of sodium dithionite resulted in sta-
activity of B. subtilis arginase, but it is more likely bilization at all three pH values tested, the choice
due to a structural effect than a reductive effect, since whether to adjust the pH of the substrate solution de-
arginases from the genus Bacillus appear to be re- pends on the robustness of the process design towards
markably stable towards reducing agents.[50] Arginases precipitates formed upon Mn2+ oxidation. When B.
from the genus Bacillus have a low cysteine content, subtilis arginase is to be applied in a membrane reac-
which diminishes the chance of deactivation due to tor it might be beneficial to work at pH 8.5 to avoid
changes in the secondary structure of the protein as a membrane fouling, but in a CSTR with immobilized
result of disulfide bond reduction. B. subtilis arginase, enzyme it might be possible to simply work at the pH
for example, contains only one cysteine (UniProtKB of arginine free base.
accession code P39138 aligned in NCBI-BLAST).[51]
Covalent Immobilization of Arginase
Effect of Sodium Dithionite at Different Initial pH
Values
B. subtilis arginase was covalently immobilized on the
commercially available, epoxy-activated supports Se-
Although 0.5 mM sodium dithionite could not prevent pabeads EC-EP and EC-HFA, and Eupergit C 250L.
oxidation of Mn2+, it did result in an excellent stabili- The immobilization was performed by simply incubat-
Adv. Synth. Catal. 2010, 352, 1493 – 1502
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1497

Paul M. Kçnst et al.
FULL PAPERS
Table 2. Immobilization yield and recovered activities, rela-
tive to the expected activity, of the arginase preparations
using a loading of 2.6 mg arginase per g dry support.
Arginase prepa-
rations
Immobilization
yield [%]
Recovered activity
[%]
EC-EP
EC-HFA
Eupergit C 250L 99Æ0.5
100Æ0.3
98Æ0.1
75Æ1.1
84Æ2.6
44Æ2.1
ing the enzyme together with the support in a trietha-
nolamine-acetate buffer with 0.5 mM Mn2+.
After immobilization the arginase preparations
were washed and the immobilization yields were de-
termined (Table 2). In all cases leaching appeared to
be minimal. Furthermore, the recovered activity (the
activity relative to the expected activity) was deter-
mined (Table 2). Compared to studies with other en-
zymes, the recovered activities of arginase immobi-
lized on EC-HFA (84%) and EC-EP (75%) are
good.[53–56] Compared to the values obtained with Se-
pabeads the recovered activity in case of Eupergit C
250L (44%) is considerably lower.
Figure 3. Thermostability of covalently immobilized and
soluble arginase compared. Aliquots of B. subtilis arginase
preparations were incubated in 50 mM triethanolamine-ace-
tate buffer pH 8.0 with 0.5 mM Mn2+ at 608C and in the
course of time the residual activity was determined upon ad-
dition of l-arginine up to 100 mM. The error bars represent
standard deviation; the depicted lines are a guide to the eye.
Since the immobilization yield is accounted for, the
loss in recovered activity appears to be directly relat-
ed to deactivation due to covalent binding. The con-
When covalently bound, Sepabeads EC-EP and Eu-
siderable loss in activity upon immobilization on Eu- pergit C 250L have relatively short (two bond
pergit C 250L could be related to its higher density of lengths) spacer arms, while Sepabeads EC-HFA has
epoxy groups relative to Sepabeads EC-EP and EC- relatively long (fifteen bond lengths) spacer arms. It
HFA (functional group densities of 0.9%, 0.4% and has been suggested by Mateo et al.[58] and more re-
0.2%, respectively, as determined for wet support by cently by Magner et al.[59] that short spacer arms (two
Resindion S.R.L.). This would result in a very inten- to three bond lengths) are beneficial for the confor-
sive covalent multipoint attachment of the enzyme to mational stability of the covalently bound enzyme.
the support, hence altering the enzyme conformation
Binding through multiple short spacer arms pro-
vides that immobilized enzyme residues are likely to
preserve their relative positions in the event of a
heat-induced conformational change. Accordingly, Se-
pabeads EC-HFAꢀs significantly longer spacer arms
provide a higher degree of conformational freedom
and distorting its active site.[57]
Thermostability of Arginase Preparations
The thermostabilities of covalently immobilized and and therefore the enzyme is more susceptible to heat-
soluble arginase were compared by following their re- induced conformational changes leading to deactiva-
sidual activity in time at a relatively high temperature tion. It must, however, be stated that there are exam-
of 608C (Figure 3).
ples of enzymes that show an increased thermostabili-
Immobilization of B. subtilis arginase on Eupergit ty upon immobilization on Sepabeads EC-HFA[54,60]
C 250L, and to a higher extent on Sepabeads EC-EP, and that relatively short spacer arms are not always
resulted in an enhanced thermostability. On the other beneficial.[53,55,61]
hand, arginase immobilized on Sepabeads EC-HFA
Considering the recovered activity and the en-
showed a thermostability comparable to that of solu- hanced thermostability, immobilization of B. subtilis
ble arginase. Incubation of aliquots of covalently im- arginase on Sepabeads EC-EP was most successful.
mobilized arginase in an excess of glycine in order to
block unreacted epoxy groups, did not result in an im-
proved thermostability (results not shown). While this Operational Stability of Soluble and Covalently
blocking step has become a common procedure for Immobilized Arginase
immobilization on epoxy supports, it has been shown
here and in a previous study that it is not always ef- Because of its high operational stability, recycling of
fective.[17]
arginase would be more efficient as opposed to dis-
1498
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1493 – 1502

Stabilized and Immobilized Bacillus subtilis Arginase for the Biobased Production
carding it after a single use.[52] In this study two meth- three commercially available epoxy-activated sup-
ods of enzyme retention were employed: entrapment ports. Immobilization on Sepabeads EC-EP was most
of soluble arginase in a membrane reactor and cova- successful resulting in a recovered activity of 75%
lent binding on epoxy supports. Besides the purpose and enhanced thermostability. The stabilization and
of recycling, enzyme retention also facilitates product immobilization of B. subtilis arginase has opened up
recovery and allows continuous process operation.[62]
possibilities for its application in the biobased produc-
The operational stabilities of soluble arginase (en- tion of nitrogen-containing chemicals as an alternative
trapment) and arginase immobilized on Sepabeads to the petrochemical production.
EC-EP (covalent binding) were compared by apply-
ing them in the previously described continuously
stirred membrane reactor system. The reactions were Experimental Section
carried out over a 72-h period at 308C using a sub-
strate solution with 250 mM l-arginine adjusted to
pH 9.5, 0.5 mM Mn2+ and 0.5 mM dithionite.
The Bacillus subtilis arginase stock solution used in this
study was obtained from an l-arginine/urea/ammonia assay
kit (K-LARGE) from Megazyme International Ireland Ltd.
and was used as received. SDS-PAGE analysis revealed that
the arginase stock was of high purity and the enzyme con-
sisted of 38 kD subunits.
After 72 h, neither soluble nor covalently immobi-
lized arginase showed significant deactivation. B. sub-
tilis arginase appears to be a robust enzyme, since it
was not affected by the shear stress caused by stirring
with 250 rpm and the applied temperature of 308C.
Application of calf liver arginase under similar condi-
tions (pH 9.5, 200 rpm, 258C) and stabilized by
0.25 mM ascorbic acid, resulted in a deactivation con-
stant of 3.8·10À3·hÀ1.[37]
All buffers and solutions used were prepared with MilliQ
water and sterilized (0.20 mm) before use. Bradford protein
assays were performed using Bio-Rad Protein Assay Kit II
with BSA standard. l-arginine free base, used as the sub-
strate, and other chemicals used in this study were of analyt-
ical grade or better and used as received.
Covalent immobilization of arginase on Sepabeads
EC-EP does not seem to add up to the already high
operational stability of B. subtilis arginase. This result
together with the observed loss in activity (Table 2)
and the additional costs upon covalent immobilization
favors the use of soluble arginase, entrapped in a
membrane reactor. On the other hand, the capital
costs of a membrane reactor are rather high and cova-
lent immobilization on spherical beads allows the ap-
plication of a packed-bed reactor in which higher de-
grees of conversion or higher space-time yields can be
achieved than in a CSTR.[52]
Sepabeads EC-EP (epoxide) and EC-HFA (epoxide on
an amino spacer) used in this study are epoxy activated
methacrylic beads with pore sizes in the range of 30–40 nm
and diameters in the range of 150–300 mm.[63] Both were
kindly donated by Dr. Paolo Caimi from Resindion S.R.L.
(Mitsubishi Chemical, Milan, Italy). Eupergit C 250L (epox-
ide) consists of epoxy-activated acrylic beads with an aver-
age pore size of 100 nm and diameter of 250 mm and was ob-
tained through Sigma–Aldrich B.V., The Netherlands.
Arginase Activity Assay
To 595 mL activation buffer (0.5 mM MnCl2 in 50 mM tri-
The final choice whether to use entrapment or co-
valent immobilization will heavily depend on the ex-
pertise and the equipment available to the manufac-
turer and should be subject of studies on the pilot-
scale. Nonetheless, the overall high operational stabil-
ity achieved in this study shows the potential of B.
subtilis arginase in the large-scale conversion of l-ar-
ginine to l-ornithine and urea.
AHCTUNGTREGeNNUN thanolamine-acetate buffer pH 8.0), 5 mL arginase stock so-
lution were added. This solution was incubated in an Eppen-
dorf Thermomixer (378C, 1200 rpm). After 5 min the assay
was started by addition of 400 mL 250 mM l-arginine solu-
tion of which the pH was adjusted to 9.5 using acetic acid.
10 mL samples were drawn after 2, 4, 6 and 8 min and imme-
diately quenched with 150 mL 1M acetic acid. The samples
were diluted by adding 700 mL MilliQ water.
HPLC Analysis
Conclusions
l-Arginine and l-ornithine concentrations were determined
using the fluorescamine derivatization method described by
Udenfriend et al.[64] To 5 mL of sample, 75 mL of 100 mM
borate buffer pH 9.0 were added and both substrate and
product were derivatized by addition of 20 mL of 5 mg·mLÀ1
fluorescamine in acetonitrile followed by vigorously vortex-
ing.
The HPLC system used for quantification was based on
separation on a NovaPak C18-column (60 ꢃ, 4 mm, 3.9ꢄ
150 mm) at 308C, isocratic elution with 1 mL·minÀ1 50%
methanol, 50% 20 mM sodium acetate buffer (pH 4.50) and
detection with a Jasco 820-FP fluorescence detector (lex =
In our pursuit to make B. subtilis arginase suitable for
application in the biobased production of nitrogen-
containing chemicals as an alternative to petrochemi-
cal production, we were successful in its stabilization
and immobilization. While the pH of the substrate so-
lution and addition of l-aspartic acid have their effect
on the operational stability of B. subtilis arginase, ad-
dition of sodium dithionite to the substrate solution
resulted in very good stabilization. Furthermore, B.
subtilis arginase was successfully immobilized on 390 nm, lem =460 nm).
Adv. Synth. Catal. 2010, 352, 1493 – 1502
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1499

Paul M. Kçnst et al.
FULL PAPERS
rinsed twice with 8 mL activation buffer (0.5 mM MnCl2 in
50 mM triethanolamine-acetate buffer pH 8.0) and finally
filter-dried. The filter-dry supports were weighed and stored
at 48C.
500 mg aliquots of each filter-dried support were trans-
ferred to 10-mL screw-cap tubes and blocked by incubation
on the rotator in 5 mL of 3M glycine in activation buffer
(pH 8.0) for 18.5 h at room temperature. After incubation
the blocked supports were washed as described above,
weighed and stored at 48C.
Operational Stability Experiments
The stability experiments were performed in a stirred cell
(Model 8010, Millipore) connected (CDS20 Selector Valve,
Millipore) to an N2 pressurized reservoir (RC800 Mini-Re-
servoir, Millipore). A standard substrate solution of l-argi-
nine and 0.05% NaN3 in MilliQ water was prepared. After
adjusting the pH with glacial acetic acid, dithionite, l-ascor-
bic acid, l-aspartic acid and manganese chloride were added
or omitted to obtain the desired solution.
The freshly prepared substrate solution was transferred to
the reservoir and subsequently the headspace was flushed
with N2 gas for 10 min. Meanwhile, the cell was mounted
with a 25-mm diameter regenerated cellulose ultrafiltration
membrane with a 10,000 D cut-off (PLGC02510, Millipore).
The cell was filled with 8 mL substrate solution and upon
addition of 100 mL 1.2 mg·mLÀ1 arginase stock, the cell was
closed and the pressure was built up to 2.0 bar, allowing a
flow of 0.25 mL·minÀ1. The cell was placed in an isothermal
water-bath and stirred at 250 rpm with the by default ap-
plied stirrer. The cellꢀs outflow was collected in fractions of
10 min using a fraction collector (LKB Bromma, 2211 Su-
perrac). Upon dilution of 5 mL with 800 mL MilliQ water,
the fractions were analyzed with the previously described
HPLC analysis.
The protein contents in the supernatant during immobili-
zation and washing were determined with Bradford protein
assays. Residual activities of the enzyme preparations were
determined with 30 mg of immobilized arginase in the stan-
dard activity assay.
Thermostability of Arginase Preparations
Aliquots of soluble and immobilized arginase in 600 mL acti-
vation buffer were incubated in a 608C water-bath. At vari-
ous time intervals the aliquots were transferred to an Ep-
pendorf Thermomixer (608C, 1200 rpm). After an incuba-
tion of 5 min, activity assays were started by adding 400 mL
of 250 mM l-arginine (pH 9.5). Sampling and analyses were
performed as previously described.
Operational Stability of Soluble and Immobilized
Arginase
Effect of Preincubation with Reducing Agents
Approximately 20 mg of B. subtilis arginase immobilized on
Sepabeads EC-EP were incubated for 15 min in 600 mL of
50 mM triethanolamine-acetate buffer pH 8.0 with 0.5 mM
Mn2+ and 2 mM of reducing agent in an Eppendorf Thermo-
mixer at 378C and 1200 rpm. Subsequently, the supernatant
was removed and upon addition of 1 mL buffer without re-
ducing agent the supports were suspended by gently vortex-
ing for 20 seconds. This rinsing step was repeated once more
and finally the supernatant was removed and replaced with
600 mL buffer without reducing agent. After 5 min of incu-
bation (378C, 1200 rpm), 400 mL of 250 mM of l-arginine so-
lution were added and the activity was determined as previ-
ously described. The blank experiment was performed in a
similar way, but with the reducing agent omitted.
The reactor was set-up and operated as previously described
using a substrate solution with 250 mM l-arginine adjusted
to an initial pH of 9.5, 0.5 mM Mn2+ and 0.5 mM sodium di-
thionite. Aliquots of 100 mg immobilized arginase on Sepa-
beads EC-EP or 25 mL 1.2 mg·mLÀ1 soluble arginase stock
were used. To be able to perform a run of 72 h without
having to refill the substrate reservoir, the pressure was ad-
justed to 0.75 bar, allowing a flow of 0.10 mL·minÀ1. The
fraction interval was increased to 15 min. The collected frac-
tions were analyzed as previously described.
Acknowledgements
Covalent Immobilization of Arginase
We would like thank SenterNovem, which on behalf of the
Dutch Ministry of Economic Affairs supported us with a
grant of the Program Energie Onderzoeksstrategie Lange
Termijn, project EOSLT02034.
Two different 0.325 mg·mLÀ1 arginase solutions were pre-
pared, one without and another with 2M NaCl added to
0.5 mM MnCl2 in 50 mM triethanolamine-acetate buffer
pH 8.0. For each epoxy support duplicates of 1.000 gram dry
weight were transferred to a 10-mL screw-cap tube. To Se-
pabeads EC-EP and Eupergit C 250L 8.0 mL of the 2M
NaCl arginase solution were added and to Sepabeads EC- References
HFA 8.0 mL of the arginase solution without NaCl were
added. All tubes were mounted on a rotator (Cole-Palmer
Roto-Torque Heavy Duty Rotator) and incubated for 24 h
at room temperature. In time 80 mL samples were taken
from the supernatant which were stored at 48C.
After 24 h of incubation the supports were filtered over a
sintered glass filter with reduced pressure, twice rinsed with
8 mL of wash buffer (80 mM NaCl and 0.5 mM MnCl2 in
50 mM triethanolamine-acetate buffer pH 8.0), resuspended
in another 8 mL of wash buffer and incubated for another
hour at room temperature. After washing the supports were
[1] J. Sanders, E. Scott, R. Weusthuis, H. Mooibroek, Mac-
romol. Biosci., 2007, 7, 105–117.
[2] K. Weissermel, H.-J. Arpe, Industrial Organic Chemis-
try, 4th edn., Wiley-VCH Verlag GmbH & Co. KGaA,
Weinheim, 2003.
[3] E. Scott, F. Peter, J. Sanders, Appl. Microbiol. Biotech-
nol. 2007, 75, 751–762.
[4] R. D. Simon, P. Weathers, Biochim. Biophys. Acta 1976,
420, 165–176.
[5] I. Voss, A. Steinbꢅchel, Metab. Eng. 2006, 8, 66–78.
1500
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1493 – 1502

Stabilized and Immobilized Bacillus subtilis Arginase for the Biobased Production
[6] I. Voss, S. C. Diniz, E. Aboulmagd, A. Steinbꢅchel,
Biomacromolecules 2004, 5, 1588–1595.
[34] P. Valle-Vega, C. T. Young, H. E. Swaisgood, J. Food
Sci. 1980, 45, 1026–1030.
[7] S. C. Diniz, I. Voss, A. Steinbꢅchel, Biotechnol. Bioeng.
[35] F. M. Veronese, C. Visco, C. A. Benassi, S. Lora, M.
Carenza, G. Palma, Ann. N. Y. Acad. Sci. 1988, 542,
115–120.
[36] E. Dala, B. Szaj’ani, Appl. Biochem. Biotechnol. 1994,
49, 203–215.
[37] A. S. Bommarius, K. Drauz, Bioorg. Med. Chem. 1994,
2, 617.
[38] K. Faber, Biotransformations in organic chemistry,
Chapter 3.2 Immobilization, 4th edn., Springer-Verlag,
Berlin, 2000, pp 384–396.
2006, 93, 698–717.
[8] E. Aboulmagd, I. Voss, F. B. Oppermann-Sanio, A.
Steinbꢅchel, Biomacromolecules 2001, 2, 1338–1342.
[9] A. Steinle, K. Bergander, A. Steinbꢅchel, Appl. Micro-
biol. Biotechnol. 2009, 75, 3437–3446.
[10] A. Steinle, F. B. Oppermann-Sanio, R. Reichelt, A.
Steinbꢅchel, Appl. Environ. Microbiol. 2008, 74, 3410–
3418.
[11] Y. Elbahloul, K. Frey, J. Sanders, A. Steinbꢅchel, Appl.
Environ. Microbiol. 2005, 71, 7759–7767.
[39] E. Katchalski-Katzir, D. M. Kraemer, J. Mol. Catal. B:
[12] Y. A. K. Elbahloul, E. L. Scott, A. Mooibroek, J. P. M.
Sanders, M. Obst, A. Steinbꢅchel, U.S. Patent 2009/
0036576A1, 2009.
[13] H. Mooibroek, N. Oosterhuis, M. Giuseppin, M.
Toonen, H. Franssen, E. Scott, J. Sanders, A. Steinbu-
chel, Appl. Microbiol. Biotechnol. 2007, 77, 257–267.
[14] G. Fꢅser, A. Steinbꢅchel, Biomacromolecules 2005, 6,
1367–1374.
Enzym. 2000, 10, 157.
[40] L. Hilterhaus, B. Minow, J. Mꢅller, M. Berheide, H.
Quitmann, M. Katzer, O. Thum, G. Antranikian, A. P.
Zeng, A. Liese, Bioprocess Biosyst. Eng. 2008, 31, 163–
171.
[41] I. Ghazi, A. G. De Segura, L. Fernꢆndez-Arrojo, M. Al-
calde, M. Yates, M. L. Rojas-Cervantes, F. J. Plou, A.
Ballesteros, J. Mol. Catal. B: Enzym. 2005, 35, 19.
[42] M. Cantarella, L. Cantarella, A. Gallifuoco, R. Frezzi-
ni, A. Spera, F. Alfani, J. Mol. Catal. B: Enzym. 2004,
29, 105–113.
[15] K. M. Frey, F. B. Oppermann-Sanio, H. Schmidt, A.
Steinbꢅchel, Appl. Environ. Microbiol. 2002, 68, 3377–
3384.
[16] A. Steinle, A. Steinbꢅchel, Appl. Microbiol. Biotechnol.
2010, 85, 1393–1399.
[43] A. S. Bommarius, K. Makryaleas, K. Drauz, Biomed.
Biochim. Acta 1991, 50, S249–255.
[17] P. M. Kçnst, M. C. R. Franssen, E. L. Scott, J. P. M.
Sanders, Green Chem. 2009, 11, 1646–1652.
[18] R. J. Gaymans, E. H. J. P. Bour, (Stamicarbon BV), U.S.
Patent 4,408,036, 1983.
[44] O. A. Roholt, Jr., D. M. Greenberg, Arch. Biochem.
Biophys. 1956, 62, 454–470.
[45] M. L. Patchett, R. M. Daniel, H. W. Morgan, Biochim.
Biophys. Acta 1991, 1077, 291–298.
[19] DSM, DSM Engineering Plastics, http://www.dsm.com/
nl_NL/html/dep/stanyl.htm, accessed January, 2010.
[20] R. C. Warner, J. Biol. Chem. 1942, 142, 705–723.
[21] J. R. Vallentyne, Geochim. Cosmochim. Acta 1968, 32,
1353–1356.
[46] M. Senuma, O. Otsuki, N. Sakata, M. Furui, T. Tosa, J.
Ferment. Bioeng. 1989, 67, 233.
[47] Y. Hara, Bull. Chem. Soc. Jpn. 1963, 36, 1373–1376.
[48] D. Cascaval, C. Oniscu, A.-I. Galaction, Biochem. Eng.
J. , 2001, 7, 171–176.
[22] T. Ikeda, T. Yasunaga, J. Phys. Chem. 1984, 88, 1253–
1257.
[23] A. Hunter, H. Borsook, Biochem. J. 1924, 18, 883–890.
[24] N. Nakamura, M. Fujita, K. Kimura, Agr. Biol. Chem.
Tokyo 1973, 37, 2827–2833.
[49] J.-P. Le Goff, C. Gavach, R. Sandeaux, J. Sandeaux,
(SmithKline Beecham Laboratoires Pharmaceutiques),
EU Patent EP0555324B1, 1997.
[50] D. J. McGee, J. Zabaleta, R. J. Viator, T. L. Testerman,
A. C. Ochoa, G. L. Mendz, Eur. J. Biochem. 2004, 271,
1952–1962.
[25] M. C. Bewley, P. D. Jeffrey, M. L. Patchett, Z. F. Kanyo,
E. N. Baker, Structure 1999, 7, 435–448.
[26] Z. F. Kanyo, L. R. Scolnick, D. E. Ash, D. W. Christian-
son, Nature 1996, 383, 554–557.
[51] National Center for Biotechnology Information – Basic
Local Alignment Search Tool, http://blast.ncbi.nlm.nih.
gov/Blast.cgi, accessed January, 2010.
[27] D. E. Wilcox, Chem. Rev. 1996, 96, 2435–2458.
[28] Brenda, The comprehensive enzyme information
system, http://www.brenda-enzymes.info, accessed Janu-
ary, 2010.
[29] N. N. Greenwood, A. Earnshaw, Chemistry of the Ele-
ments, Chapter 24, Butterworth Heinemann, Oxford,
1998, pp 1040–1069.
[30] A. Alonso, M. J. Almendral, M. D. Bꢆez, M. J. Porras,
C. Alonso, Anal. Chim. Acta 1995, 308, 164–169.
[31] T. Bagnost, Y.-C. Guillaume, M. Thomassin, J.-F.
Robert, A. Berthelot, A. Xicluna, C. Andrꢇ, J. Chro-
matogr. B 2007, 856, 113–120.
[32] S. Karacaoglu, S. Timur, A. Telefoncu, Artif. Cells
Blood Substitutes Immobilization Biot 2003, 31, 357–
363.
[52] A. S. Bommarius, B. R. Riebel, Biocatalysis, Chapter 5,
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim,
2004, pp 91–134.
[53] F. Lꢈpez-Gallego, L. Betancor, A. Hidalgo, C. Mateo,
J. M. Guisꢆn, R. Fernꢆndez-Lafuente, J. Biotechnol.
2004, 111, 219–227.
[54] J. M. Bolivar, L. Wilson, S. A. Ferrarotti, R. Fernꢆndez-
Lafuente, J. M. Guisꢆn, C. Mateo, Enzyme Microb.
Technol. 2007, 40, 540–546.
[55] C. Mateo, R. Torres, G. Fernꢆndez-Lorente, C. Ortiz,
M. Fuentes, A. Hidalgo, F. Lꢆpez-Gallego, O. Abian,
J. M. Palomo, L. Betancor, B. C. C. Pessela, J. M.
Guisan, R. Fernꢆndez-Lafuente, Biomacromolecules
2003, 4, 772–777.
[56] L. Betancor, F. Lꢈpez-Gallego, A. Hidalgo, N. Alonso-
Morales, G. Dellamora-Ortiz, J. M. Guisꢆn, R. Fernꢆn-
dez-Lafuente, J. Biotechnol. 2006, 121, 284.
[33] R. Mira de Orduna, J. Agric. Food Chem. 2001, 49,
549–555.
Adv. Synth. Catal. 2010, 352, 1493 – 1502
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1501

Paul M. Kçnst et al.
FULL PAPERS
[57] D. Hormigo, I. De La Mata, M. P. Castillꢈn, C. Acebal,
M. Arroyo, Biocatal. Biotransform. 2009, 27, 271–281.
[58] C. Mateo, O. Abian, G. Fernꢆndez-Lorente, J. Pedro-
che, R. Fernꢆndez-Lafuente, J. M. Guisan, A. Tam, M.
Daminati, Biotechnol. Prog. 2002, 18, 629–634.
[61] A. Basso, P. Braiuca, S. Cantone, C. Ebert, P. Linda, P.
Spizzo, P. Caimi, U. Hanefeld, G. Degrassi, L. Gardossi,
Adv. Synth. Catal. 2007, 349, 877–886.
[62] A. Liese, K. Seelbach, C. Wandrey, Industrial Biotrans-
formations, Chapter 4, 1st edn., Wiley-VCH Verlag
GmbH & Co. KGaA, Weinheim, 2000.
[59] U. Hanefeld, L. Gardossi, E. Magner, Chem. Soc. Rev.
[63] Resindion S.R.L., (Mitsubishi Chemical, Milan, Italy),
http://www.resindion.com/sepabeadsec/sepabead-
sec.html.
[64] S. Udenfriend, S. Stein, P. Bohlen, W. Dairman, W.
Leimgruber, M. Weigele, Science 1972, 178, 871.
2009, 38, 453–468.
[60] R. Torres, C. Mateo, G. Fernꢆndez-Lorente, C. Ortiz,
M. Fuentes, J. M. Palomo, J. M. Guisan, R. Fernꢆndez-
Lafuente, Biotechnol. Prog., 2003, 19, 1056–1060.
1502
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1493 – 1502