Immobilisation of ω-transaminase for industrial application: Screening and characterisation of commercial ready to use enzyme carriers

Article abstract of DOI:10.1016/j.molcatb.2015.04.005

Despite of the advantages that enzyme immobilisation can bring to industrial biocatalysis, its utilisation is still limited to a small number of enzymes and processes. Transaminase catalysed processes are a good example where immobilisation can be of major importance and even decisive for economic feasibility. This work presents results obtained for screening of enzyme carriers for immobilisation of ω-transaminase for industrial application. A total of 6 commercial enzyme carriers (polymeric resins) were screened and two suitable enzyme carriers were selected for immobilisation of both (S)- and (R)-selective ω-transaminases. These carriers allowed the re-use of the immobilised enzyme for 8 cycles of 24 h each, under relevant process conditions, corresponding to approximately 250 h of operation, with more than 50% of the initial activity retained. Likewise the stability towards higher temperatures and possibility to store the biocatalyst for more than 70 days (at room temperature) were obtained as result of the immobilisation on the selected supports.

Full text of DOI:10.1016/j.molcatb.2015.04.005

Accepted Manuscript
Title: Immobilization of ␻-transaminase for industrial
application: screening and characterization of commercial
ready to use enzyme carriers
¨
Author: Watson Neto Martin Schurmann Lavinia Panella
Andreas Vogel John M. Woodley
PII:
S1381-1177(15)00103-4
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcatb.2015.04.005
MOLCAB 3145
To appear in:
Journal of Molecular Catalysis B: Enzymatic
Received date:
Revised date:
Accepted date:
9-1-2015
5-4-2015
7-4-2015
¨
Please cite this article as: W. Neto, M. Schurmann, L. Panella, A. Vogel, J.M. Woodley,
Immobilization of rmomega-transaminase for industrial application: screening and
characterization of commercial ready to use enzyme carriers, Journal of Molecular
Catalysis B: Enzymatic (2015), http://dx.doi.org/10.1016/j.molcatb.2015.04.005
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Graphical Abstract
1
Page 1 of 32

Highlights
.Screening and characterisation of enzyme carriers is presented.
.A library containing 6 enzyme carriers (hydrophobic and ionic exchangers) have been
used.
.The selected carriers improved the enzymes stability towards increased temperature.
.The selected carriers allowed increased storage stability at room temperature.
.It was possible re-use the immobilized enzyme for 8 cycles of 24 hours each.
2
Page 2 of 32

Immobilization of ω-transaminase for industrial application: screening and
characterization of commercial ready to use enzyme carriers.
Watson Netoa, 1, Martin Schürmannb, Lavinia Panellab, Andreas Vogelc, and John M.
Woodleya
a Department of Chemical and Biochemical Engineering, Technical University of Denmark,
Lyngby, DK-2800 Denmark;
b DSM Innovative Synthesis BV, Urmonderbaan 22, NL-6167RD Geleen, The Netherlands;
c c-LEcta GmbH, Perlickstraße 5, 04103 Leipzig, Germany;
Corresponding author:
Name:
John M. Woodley
Address:
Department of Chemical and Biochemical Engineering, Technical University of
Denmark, Building 229 Søltofts Plads, DK-2800 Kgs. Lyngby, Denmark
Phone:
Fax:
+45 4525 2885
+45 4593 2906
jw@kt.dtu.dk
Email:
Keywords: ω-transaminase, immobilization, polymeric resins, enzyme carrier
Journal: Journal of Molecular Catalysis B: Enzymatic
1 Present address: Purolite Ltd, Unit D, Llatrisant Business Park, South Wales, CF72 8LF, UK
3
Page 3 of 32

CFESol
CFEImm
WC
Soluble cell-free extract
Immobilized cell-free extract
Whole cells
ω-TAm
MBA
PYR
ω-Transaminase
Methylbenzylamine
Pyruvate
PLP
Pyridoxal-5'-phosphate
Potassium phosphate buffer
Isopropylamine
PPB
IPA
IPAc
iPrOH
APH
BA
Isopropyl acetate
Isopropyl alcohol
Acetophenone
Benzylacetone
APB
OD
3-Amino-1-phenylbutane
Optical density
4
Page 4 of 32

ABSTRACT:
Despite of the advantages that enzyme immobilisation can bring to industrial biocatalysis, its
utilisation is still limited to a small number of enzymes and processes. Transaminase
catalysed processes are a good example where immobilisation can be of major importance
and even decisive for economic feasibility. This work presents results obtained for screening
of enzyme carriers for immobilisation of ω-Transaminase for industrial application. A total of
6 commercial enzyme carriers (polymeric resins) were screened and two suitable enzyme
carriers were selected for immobilization of both (S)- and (R)-selective ω-transaminases.
These carriers allowed the re-use of the immobilized enzyme for 8 cycles of 24 hours each,
under relevant process conditions, corresponding to approximately 250 hours of operation,
with more than 50% of the initial activity retained. Likewise the stability towards higher
temperatures and possibility to store the biocatalyst for more than 70 days (at room
temperature) were obtained as result of the immobilization on the selected supports.
1 INTRODUCTION
Considering the advantages that the use of immobilized biocatalysts offer compared to soluble
formulations (easy recovery and reuse, improved stability and possibility for continuous
operation in packed bed reactors, to name a few), one might expected many processes
running using immobilized biocatalysts as well as a large number of commercially available
immobilized enzymes. However, this is not the reality. Lipase B from Candida antarctica (Cal
B) is probably the most successful case of a commercially available immobilized enzyme
which is supplied by several companies such as Novozymes A/S, Denmark (Novozym®435),
5
Page 5 of 32

Roche Molecular Biochemicals, Germany (Chirazyme L-2), Purolite Ltd, United Kingdom (CalB
ImmoPlusTM) as well as similar preparations from SPRIN Technologies S.p.A (Italy), c-LEcta
GmbH (Germany) and CLEA Technologies B.V (The Netherlands) [1].
The usual loss of activity commonly observed in the immobilisation process could be one
reason for this observation. On the other hand, another explanation could be the lack of a
generally applicable, and simple to use, method for immobilization. Today, most usually a trial
and error approach is required to immobilize enzymes. Moreover, the loss in activity due to
introduction of mass transfer limitations and loss of active enzyme by leaching or irreversible
denaturation, which are commonly observed in immobilization of enzymes, can also be a
limiting factor. Another contribution could be related to the increase in biocatalyst cost
contribution to the production as a result of the introduction of an immobilization step in the
biocatalyst formulation. However, the cost contribution of the immobilized enzyme for an
implemented full scale process should in principal be lower than for a soluble enzyme, since
the immobilized enzyme may be re-used for many reaction cycles [2].
For challenging processes where several processing strategies need to be applied in order to
achieve industrially relevant yields, and where the cost associated with the biocatalyst
production or purchase are high (which often is associated with the small scale of production
of the biocatalyst), immobilizing the enzyme can be instrumental in determining process
feasibility.
The asymmetric synthesis of chiral amines using ω-transaminases (ω-TAms) (EC 2.6.1.1) is a
good example of such a process [2]. ω-TAm can be used as suitable catalysts for the
production of optically pure chiral amines, which are important building blocks in the (agro)-
chemical and pharmaceutical industries [2-10]. The enzyme transfers an amine group from a
6
Page 6 of 32

donor molecule to an acceptor (pro-chiral ketone), yielding a chiral amine and a carbonyl co-
product (Figure 1). The reaction can either be run as a kinetic resolution [3, 5, 10, 11] (limited
to a maximum return yield of 50%) or in an asymmetric synthesis [4, 6, 7, 12], which is
preferable since in principle it can give 100% yield. However, despite these promising
features, ω-TAms are known to be often severely inhibited both by the ketone substrate and
the chiral amine product and frequently affected by an unfavourable equilibrium (Keq<<1) in
the synthetic direction [2]. Thus, several process strategies often need to be implemented to
help shift the equilibrium and alleviate substrate and product inhibition in order to allow
acceptable yield and product concentration [2]. Some of these strategies make use of external
agents such as organic solvents or polymeric resins which, in principle, can decrease the
biocatalyst stability and activity or interfere with its availability in the reactor (by binding
onto a resin column for example). In such cases, immobilizing the biocatalyst can be
advantageous, especially when operating with cell-free extract (CFE). Additionally, ω-TAm
catalysed reactions are relatively new processes and therefore production is frequently at
small scale. This means that the cost for production and/or purchase of the enzyme is
potentially high. In principle, immobilization of the enzyme can reduce these costs since it
allows biocatalyst to be re-used for several cycles (increasing the biocatalyst yield –
gproduct/gbiocatalyst) and also decreasing the cost associated with the downstream processing,
since immobilization simplifies the separation of the biocatalyst from the products.
There are already a few reports in the scientific literature regarding the use of immobilized ω-
TAm both formulated as WC and CFE. Interestingly most of the reports using immobilized
WCs show reduced performance, mostly due to diffusional limitations [5, 13]. This has been
overcome in another unrelated work where WC and permeabilized cells have been
7
Page 7 of 32

immobilized by entrapment in PVA-gel (Lentikats®) resulting in low diffusional limitations
[14]. In another study, immobilization of E. coli by crosslinking with glutaraldehyde and by
entrapment with carrageenan and polyacrylamide were reported unsuccessful, with more
than 50% of activity being lost in the case of entrapment in polyacrylamide, while the
entrapment in carrageenan resulted in a mechanically unstable preparation [15]. In the same
work, the authors reported immobilization of WC using hydrous titanium oxide (surface
adsorption), calcium alginate (entrapment) and chitosan (by cell flocculation). The former
was reported to show very poor loading capacity (less than 0.1 gWC/gcarrier, on a dry weight
basis) and consequently decreased immobilization yield, while the preparation using calcium
alginate showed decreased activity due to mass transfer limitations already at low loadings of
0.2 gWC/ gcarrier. On the other hand, the preparation using chitosan allowed loadings of up to
3.2 gWC/ gcarrier (dry weight) and more than 60% residual activity.
Immobilization of CFE of ω-TAm has been achieved both by covalent linkage to different solid
support materials and by entrapment in sol–gel matrices [16, 17] with reported
immobilization yields of 20–50% (immobilized protein/total protein) and with less than 20%
of activity remaining [18], respectively. Other studies have reported low immobilization yields
as well as poor residual activity (<50%) after immobilization by covalent attachment to
carriers [19-21]. On the other hand, the same studies reported increased storage stability,
which could be associated with conformational stabilisation of the enzyme as result of the
immobilization.
Enzyme recycle studies using immobilized ω-TAm enzymes have also been sought and
reported by serval or the above mentioned authors [14, 17, 18].
8
Page 8 of 32

In a recent study, Truppo and co-workers [22] reported the development of an immobilized
ω-TAm capable of operating in organic solvents (namely isopropyl acetate, isopropanol and
toluene). However, it is not clear how much of this is a direct result of the immobilization
itself or a result of prior improvements made to the enzyme through protein engineering
techniques [23]. Nevertheless, the immobilized preparation allowed 10 times recycling in
optimized reaction conditions, corresponding to approximately 200 hours of operation.
In contrast to the mentioned study by Truppo, most of the reports mentioned earlier have in
common the fact that non-commercial supports, such as calcium alginate, hydrogels or
hydrous titanium oxide are used. These supports require labour intensive preparation steps
prior the immobilization itself. In order to be attractive and suitable for industrial application,
the procedure for immobilization should be quick, robust, scalable and reproducible, while
preserving the enzyme activity [24].
The use of commercial ready-to-use supports, such as polymeric resins can in principle avoid
these concerns. Furthermore, these supports are known to be mechanically stable, which is a
feature that needs to be considered when scaling up reactions using an immobilized
biocatalyst [25, 26].
These supports are mainly made of polystyrene, polyacrylate, polyvinyls, polyamide and
polypropylene. They are commercialized as purely adsorptive supports, ion exchangers or
with epoxy functional groups (allowing covalent immobilisation) by several companies, under
different trade names such as Amberlite, Duolite and Dowex (Dow Chemicals. USA), Lewatit
(Lanxess, Germany), Diaion, Sepabeads and Relizymes (Resindion, Italy), Purolite ECR®
(Purolite, USA), among others [27].
9
Page 9 of 32

The present study focuses on the application of organic supports of synthetic origin for
immobilisation and present step-by-step screening process for selection and
a
characterization of a suitable support for immobilization of ω-TAm. A library of supports was
selected for this study containing 6 organic supports of synthetic origin (hereafter simply
referred to as “resins” or “enzyme carriers”) establishing hydrophobic and ion exchange
interactions with the enzyme through direct interaction with the octadecyl and ethylamino
functional groups, respectively, present on the surface of the supports.
2 ENZYME CARRIERS USED IN THIS STUDY
The library of 6 resins were purchased from Resindion Srl. (Milan, Italy) (Table 1) and were
composed of two different types of mechanism of interaction with the enzyme
(hydrophobicity and ion exchange) and two distinct mean pore diameters (10-20 nm and 40-
60 nm). Finally, different lengths of functional groups were also represented in the library as
shown in Table 1.
3 SCREENING METHODOLOGY FOLLOWED
Suitable supports for immobilization can be selected based on different factors, depending on
the features desired. Often, the immobilization of enzymes is carried out in order to increase
the enzyme stability, especially in non-conventional media (e.g. organic solvent, high pH or
temperature), while in other cases the possibility of re-using the biocatalyst is sought, as
discussed earlier. In these study both these features were explored and used to screen and
select a suitable support to immobilize ω-TAm. The screening methodology followed was
divided in three steps as described below and depicted in Figure 2.
10
Page 10 of 32

.Step I – Immobilization efficiency: In this step, the resins were tested for their capacity to
immobilize the enzyme. This was evaluated through the immobilization yield, loading
capacity and residual relative activity of the immobilized enzyme. A threshold of 90% for
the residual relative activity of the immobilized enzyme was defined in this study in order to
select only the most suitable resins. Immobilized preparations showing residual activity
lower than 90% were discarded.
.Step II – Operational stability: The remaining immobilized preparations were tested for their
activity for several cycles under reaction conditions. A total of 8 cycles of 24 hours or more
were made and the immobilized preparations showing relative residual activities lower than
50% were discarded.
.Step III: Activity under alternative conditions: The selected immobilized preparations were
tested for activity at high temperatures (50 °C) and in organic solvent, as well as for activity
after a long storage period (up to 70 days) at room temperature.
4 EXPERIMENTAL SECTION
All the resins were purchased from Resindion S.R.L (Milan, Italy). The chemicals were all
obtained from Sigma-Aldrich (Buchs, Switzerland). The (S)-ω-TAm (ATA-47), formulated as
lyophilized CFE, was obtained from c-Lecta GmbH (Leipzig, Germany) and the (R)- ω-TAm
(Ate-TA) formulated as WC was obtained by DSM Innovative Synthesis (Geleen, The
Netherlands). With exception for assays regarding temperature stability, all the experiments
were run in duplicate.
11
Page 11 of 32

Pre-treatment of (S)-ω-TAm enzyme:
The enzyme was obtained already formulated as a lyophilized CFE powder. An enzyme
solution of 5 gCFE/L (dry weight) was gently prepared in 100 mM potassium phosphate buffer
(PPB) pH 7 containing 8 mM of pyridoxal-5’-phosphate (PLP). 1 mL of this solution was added
to each of the resins.
Pre-treatment of (R)-ω-TAm enzymes:
The enzyme was obtained as frozen centrifuged fermentation broth. The cells were re-
suspended in 100 mM PPB pH 7 containing 8 mM of PLP solution (with a buffer/WC mass
ratio of 2:1). The OD at 620 nm was measured to be approximately 0.3 (after a 30-fold
dilution). The suspension was sonicated using a Sonics vibra-cell TM, sonicator Model”CV.18
9836A (Newtown, CT, USA), set for 10 minutes with 75% of amplitude and a pulse of 10. The
OD was once again measured to be approximately 0.03 (after a 30-fold dilution). The
sonicated broth was then centrifuged (Eppendorf Model 5415 R, Hamburg, Germany) for 10
minutes at 1300 rpm and the resulting pellet discarded. The supernatant was diluted with
100 mM PPB pH 7 containing 8 mM of PLP with a ratio 1:9. The final enzyme solution had
approximately 7.9 gCFE/L (dry weight) when lyophilized.
Lyophilisation: Samples were frozen at -80˚C and left overnight prior to lyophilisation at -
54˚C for 6 hours under vacuum (10E-2 mBar) using a Heto LyoLab 3000 from Thermo
Scientific (Waltham, MA, USA).
Immobilization procedure:
Resins (50 mg for (S)-ω-TAm and 100 mg for (R)-ω-TAm) were washed with 1 mL 100 mM
PPB pH 7 at room temperature. After 1 minute the washing buffer was discarded by means of
12
Page 12 of 32

a syringe with a needle. Subsequently, 1 mL of the enzyme solution was added to each of the
resins. Preparations were placed in an orbital shaking incubator (IKA ® KS, Model 130 Basic -
Staufen, Germany) at room temperature (25-27 ˚C) for 48 h at 400 rpm.
Activity assay:
The activity assay was carried out in 4 mL vials at 30 ˚C at approximately 400 rpm (IKA ® KS,
Model 130 Basic – Staufen, Germany). All the immobilised enzyme prepared was used
(approx. 50 mg for (S)- and 100 mg for (R) ω-TAm). 1 mL of reaction mixture composed of 1
M isopropylamine (IPA), 30 mM acetophenone (APH), 2 mM PLP and 100 mM PPB pH 7 was
used for (S)-ω-TAm and 2 mL of reaction mixture composed by 50 mM DL-(-)-MBA, 10 mM
benzylacetone (BA), 2 mM PLP and 100 mM PPB pH 7 was used for (R)-ω-TAm. Samples of
100 µL were taken at minutes 1, 3 and 5 for Step I, at 0, 1.5, 3, 6 and 24 hours for Step II and at
0 and 1.5 hours for Step III. Samples were diluted into 400 µL of 1 M HCl and centrifuged for
10 minutes (Eppendorf Model 5415 R – Hamburg, Germany) at 13000 rpm, prior to analysis.
Analytical:
All samples were analysed by measuring the concentration of (S)-MBA (for reactions
catalysed by (S)-ω-TAm) and (R)-3-Amino-1-phenylbutane ((R)-APB) (for reactions catalysed
by (R)-ω-TAm) using HPLC (Agilent 1100 Series). Compounds were separated on a Grace ™
Prevail ™ C18 (250x4.6 mm), 5 µm column (Grace, Columbia, MD, USA) at a flow rate of 2
mL/min using a multi-step gradient flow of aqueous 100 mM perchloric acid and pure
acetonitrile, with the following percentage of acetonitrile: 0 min (0%), 1 min (10%), 2.5 min
(10%), 5.9 min (60%), 6 min (0%), 7 min (0%). Compounds were detected at 210 nm. The
quantitative analysis was performed based on peak areas, calibrated against external
standards.
13
Page 13 of 32

Step I of the screening methodology – Immobilization efficiency
Following immobilization, samples were taken from the supernatant and assayed for residual
activity in order to estimate the amount of residual un-adsorbed enzyme. The remaining
supernatant was discarded and the resins were washed twice (for 2 minutes and 1h30
respectively) with 100 mM PPB pH 7.0 under mild agitation (400 RPM) at room temperature
(25-27 ˚C). Samples of 100 µL were taken from the supernatant after the second washing and
analysed for activity in order to quantify the leakage from the carriers (unconjugated protein).
The remaining washing buffer was discarded using a syringe and the reaction mixture was
added to the immobilised enzyme. Activity assay was carried out to estimate the residual
activity in the immobilized enzyme. Soluble enzymes were treated under the same conditions
as a reference.
Step II of the screening methodology – Operational stability
Both (S)- and (R)-selective ω-Tam, immobilized as previously described, were re-used for 8
cycles of 24 hours each (with exception for cycle number 5 which lasted 72 hours). Samples
were taken at 1.5, 3, 6, and 24 or 72 h. After each cycle, the remaining reaction mixture was
discarded and the resins were washed once with 1 mL of 100 mM PPB pH 7. The washing
buffer was removed (by means of a syringe with a needle) and fresh reaction mixture was
added to start the new cycle.
14
Page 14 of 32

Step III of the screening methodology – Activity under alternative process
conditions
Activity at 50 °C
Soluble enzymes were incubated for 12 hours at 50 °C in 100 mM PPB pH 7 solution
containing 2 mM PLP. After the incubation, the reaction mixture was added to the enzymes
and the activity assay was initiated at 50 °C. The immobilized enzymes were incubated at
50°C in the reaction mixture used for the activity assay. After the incubation, the liquid was
discarded and the immobilized enzymes were rinsed with 100 mM PPB pH7 solution before
fresh reaction mixture was added to start the activity assay at the same temperature. These
were compared to free and immobilized enzyme assayed at 30 °C.
Activity in organic solvents
Lyophilized enzymes (5 mg for (S)-ω-TAm and 7.9 mg for (R)-ω-TAm – dry weight) and
immobilized enzymes (prepared as previously described and followed by a drying step with
Nitrogen sweeping) were tested for activity in water saturated toluene, isopropyl acetate,
cyclohexane and 50 % isopropanol. Solvents were saturated with an equal volume of aqueous
solution (100 mM PPB pH 7, containing 2 mM of PLP) for 72 hours under vigorous mixing to
ensure proper saturation of both phases. The aqueous phase was discarded and IPA and APH
(final concentrations of 1 M and 30 mM, respectively) were added to the saturated organic
phase for the reaction using (S)-ω-TAm, while DL-MBA and BA (final concentration of 50 mM
and 10 mM respectively) were added to reactions using (R)-ω-TAm. The activity assay was
carried out in 4 mL vials at 30 ˚C at approximately 400 rpm (IKA ® KS, Model 130 Basic –
15
Page 15 of 32

Staufen, Germany). Samples of 100 μL were taken at 1.5, 3, 6 and 24 h, diluted in 400 μL
acetonitrile, centrifuged and analysed by HPLC.
Activity after storage
Free and immobilized enzymes were left in the fume cabinet for 20 days at room temperature
(which was measured to oscillate between 25-27 °C). The soluble enzymes were prepared in
100 mM PPB pH 7.0, 8 mM PLP solution, while the immobilized enzymes were left in a semi
wet state (the excess of immobilization buffer was removed by means of a syringe).
In addition, the immobilized enzymes previously used in Step II were washed with 100 mM
PPB pH 7.0, 8 mM PLP solution after the 8th cycle, the excess washing buffer was then
removed by means of a syringe and the immobilized preparations were left in the fume hood
(25-27 °C ) for 60 days. After the ‘storage’ period, the biocatalysts were added to the reaction
mixtures and an activity assay was initiated.
5 RESULTS AND DISCUSSION
Step I – Immobilization efficiency
The relative residual activity of the immobilized preparation, together with the loading
capacity are of high importance, since low residual activity and/or low loading capacity are
often translated into an economically infeasible immobilization process. Often, a reference
loading capacity of 0.1 gCFEImm/gResin is used in scientific literature as ideal [1, 27]. In this
step, these parameters as well as the leakage quantification were used to evaluate the
immobilization efficiency for the different preparations. The effect of the pore diameter and
length of the functional groups were also evaluated.
16
Page 16 of 32

With exception for Sepabeads EC-OD, all the resins allowed immobilisation yield above 60%
(Figure 3).
Leakage quantification:
For all the preparations, the results suggest no significant enzyme leakage after 1.5 h stirring
in 100 mM PPB pH 7. No activity was detected in the washing buffer (data not shown).
Effect of functional group and its length on the immobilization:
The results obtained suggest an overall better immobilisation yield for resins establishing
ionic interactions with the enzyme (Sepabeads EC-EA, Relizyme EA403, Sepabeads EC-HA and
Relizyme HA403), in comparison with those establishing hydrophobic interactions
(Sepabeads EC-OD and Relizyme OD403) (Figure 3). Interestingly, within this group, the
resins having long length functional groups appeared to have slightly better immobilisation
yield. This was observed for Sepabeads EC-HA and Relizyme HA403 when compared with
Sepabeads EC-EA, Relizyme EA403 . This is normally associated with obstruction of the
enzyme’s catalytic centre upon immobilization in the case of short length functional groups
due to the proximity of the enzyme to the carrier surface, causing enzyme inactivation. This is
often overcome by the use of spacers [25, 27, 28].
Effect of pore size on the immobilization:
For all cases, resins with larger pore sizes (Relizyme: 40-60 nm) showed higher
immobilization yields than those having smaller pore sizes (Sepabeads: 10-20 nm). For
instance, immobilizations with Relizyme HA403 showed higher immobilization yield than
immobilization with Sepabeads EC-HA. The same can be observed for Relizyme EA403 in
comparison with Sepabeads EC-EA and Relizyme OD403 vs Sepabeads EC-OD (Figure 3).
17
Page 17 of 32

Following these results, all resins but Sepabeads EC-OD, which had very low immobilisation
yield, were selected for further steps of characterization. All these five selected resins had a
loading capacity comparable to the reference of 0.1 gCFEImm/gResin used in the scientific
literature [1, 27] (data not shown).
Step II – Operational stability
The ability to re-use the biocatalyst for several cycles is in most cases the key advantage
achieved by immobilization. In this step of the screening, the five selected resins were used to
prepare new immobilised enzymes using this time no only the (S)-ω-TAm, but a (R)-selective
ω-TAm as well. The immobilization yields for this step was above 80% for immobilisation
with (S)-ω-TAm and about 50% for immobilisation with (R)-ω-TAm. The difference in the
immobilization yields between the two enzymes is mainly due to the different initial
formulations in which these enzymes were presented resulting in different amount of protein
being immobilised. Additionally, considering that both enzymes come from different
suppliers, different level of purity and expression are to be expected.
These two set of immobilised enzymes were re-used for 8 cycles of 24 hours each under the
process conditions (with exception for cycle number 5 which lasted 72 hours). In total, the
preparations were subjected to approximately 250 hours of operation (pH 7 and temperature
of 30 ˚C with orbital agitation).
The performance for recycling of both enzymes immobilized on the selected resins were
evaluated by following the conversion over time throughout the 8 cycles (Figure 4Error!
Reference source not found.) and also by following the initial rate measurements for each
cycle (Figure 5).
18
Page 18 of 32

Based on the maximum conversion achieved at the end of each cycle, the immobilizations
using the resins Relizyme OD403, Sepabeads EC-HA and Relizyme HA403 showed the best
performance among the five for (S)-ω-TAm (Figure 4A). For (R)-ω-TAm, all the five
immobilized preparations showed maximum conversion possible after 24 hours for all cycles
(Figure 4B), suggesting an excellent activity retention despite the lower immobilization yield.
In order to have a clearer overview of the performance of the immobilized enzymes in each
cycle, it is important to look at the initial rates as well, since this dictates how much of the
activity is lost in each cycle, due to enzyme deactivation or leaching.
Due to the high immobilization yield, all the preparations for (S)-ω-TAm presented in the first
cycle an activity comparable to the free enzyme (illustrated by the dashed line in Figure 5A).
However, in the second cycle the activity decreases almost 50% for all of the preparations and
from this point the loss of activity for the following 6 cycles is not so accentuated. This can be
observed especially for the immobilizations using the resins Relizyme OD403, Sepabeads EC-
HA and Relizyme HA403 which kept approximately 50-60% of the initial activity throughout
the 8 cycles.
With respect to (R)-ω-TAm, the results obtained showed excellent activity retention
throughout the 8 cycles for all the preparations, with special emphasis to preparations using
Relizyme OD403, Sepabeads EC-EA and Sepabeads EC-HA, which allowed retention of more
than 90% of the initial activity (Figure 5B). Based on these results, the resin library was
reduced to 3: Relizyme OD403, which showed good activity retention for both enzymes;
Relizyme HA403 and Sepabeads EC-EA, which showed good performance for (S)-ω-TAm and
(R)-ω-TAm, respectively.
19
Page 19 of 32

Step III – Activity under alternative conditions
Similarly with the reusability, stability is one of the improvements often used to justify the
need for immobilization. Several authors have reported improved biocatalyst stability as a
result of immobilization using many different enzymes [29, 30].
Temperature stability is one of the improvements which can be achieved through
immobilization. A more thermo stable biocatalyst may open the possibility of carrying out
reactions at higher temperatures, which should mean higher reaction rates. In the specific
case of ω-transaminase, carrying out the reaction at higher temperature also opens the
possibility of new process options such as the evaporation of volatile co-products such as
acetone in order to shift reaction equilibrium [1].
Furthermore, the activity in non-conventional conditions (or alternative conditions), such as
in organic solvents, is often discussed and reported in the scientific literature [22, 29, 31].
Having a biocatalyst which can operate in organic solvents can make possible reaction in two
phase system or in neat organic phase which could bring advantages such as increased
substrate and product solubility [2].
Finally, the possibility to store the immobilized biocatalyst for a long period of time can also
be advantageous and has been widely discussed in the scientific literature [27, 29, 32-34]. It
can save time and allow immobilization of a large amount of enzyme in advance for future
utilization as well as opening the possibility for commercialization of the immobilized
enzyme.
20
Page 20 of 32

Activity at 50°C
The immobilized enzymes were incubated for 24 hours at 50 °C and compared to soluble
enzymes incubated at the same temperature but for 12 hours only. Afterwards, these were
compared to soluble enzymes incubated at 30 °C for 24 hours. The results illustrated in Figure
6 show that soluble enzymes incubated at 50 °C lose about 80% and 100% of initial activity
(for (S)-ω-TAm and (R)-ω-TAm respectively). None of the immobilized (S)-ω-TAm
preparations showed substantial improvements, when compared with the free enzyme
treated under the same conditions (50 °C). The best result obtained was observed for the
preparations using the resin Relizyme HA403 which showed slightly higher residual activity
in comparison to the soluble enzyme (Figure 6A).
On the other hand, the results were more prominent for the (R)-ω-TAm, which was found to
retain 80% of activity when immobilized with the Relizyme OD403 and about 30% with
Sepabeads EC-EA while the free soluble enzyme at the same temperature showed no residual
activity (Figure 6B). These results illustrate an enhanced stability at 50 °C, achieved as direct
result of immobilization.
Activity in organic solvent
Soluble and immobilized enzymes were used to run reactions in both water-immiscible and
water-miscible solvents. Similar experiments have been previously reported where the
immobilized ω-TAm shown remarkable performance well [22]. However, the results obtained
in current study suggest that the biocatalyst activity in organic phase was not enhanced as
result of immobilization. None of the immobilized preparations performed better than
lyophilised soluble enzyme in any of the solvents with exception for the assay performed
under 50% isopropanol, where almost 5% of residual activity was found in the preparation of
21
Page 21 of 32

(S)-ω-TAm immobilized on Relizyme HA403 (data not shown). These could also be related to
a poor mass transfer of the substrate from the organic phase to the biocatalyst itself, or the
absence of the cofactor (PLP) in the reaction mixture, since PLP is not soluble in organic
solvent.
Storage stability
The results obtained for the storage stability show an improved stability at room temperature
for all the immobilized preparations (Figure 7). In all cases, the immobilized preparations
retained more activity than the soluble enzymes (both (S)-ω-TAm and (R)-ω-TAm) treated
under the same conditions.
Both enzymes immobilized (using both hydrophobic and ionic exchange resins) retained
more activity after 20 and 70 days in comparison with their free formulations at room
temperature. In general, the ion exchange resins (Sepabeads EC-EA and Relizyme HA403)
allowed greater activity retention than the hydrophobic resin (Relizyme OD403) after 60
days. More than 50 % and 40% of residual activity was observed for (S)-ω-TAm and (R)-ω-
TAm immobilized on the ionic exchange support, after 60 days. These results are more
encouraging than those previously reported where losses of 30 % and 92% of activity of ω-
TAm immobilized on chitosan and Eupergit® C, respectively, after 24 days at 4 °C were
obtained [17]. It is expected that better activity retention could have be obtained if the
preparations were stored at 4-5 °C instead of room temperature.
6
CONCLUSIONS
Immobilisation of ω-TAms enzymes on commercially ready to use enzyme carrier allowed re-
use of the immobilized preparation for 8 cycles (each of 24 hours), under the process
22
Page 22 of 32

conditions, which corresponds to more than 250 hours of operation, retaining more than 50%
initial activity. The resulting preparation showed increased enzyme stability at high
temperatures in caparison to the soluble native enzymes and made it possible to store the
enzyme for more than 60 days at room temperature.
7 ACKNOWLEDGMENTS:
The author wishes to acknowledge the financial support from BIOTRAINS Marie Curie ITN,
financed by the European Union through the 7th Framework people Programme (grant
agreement
no.:
238531).
23
Page 23 of 32

8 REFERENCES
[1] P. Tufvesson, U. Törnvall, J. Carvalho, A.J. Karlsson, R. Hatti-Kaul, J. Mol. Catal. B: Enzym. 68 (2011)
200-205.
[2] P. Tufvesson, J. Lima‐Ramos, J.S. Jensen, N. Al‐Haque, W. Neto, J.M. Woodley, Biotechnol. Bioeng.
108 (2011) 1479-1493.
[3] J.S. Shin, B.G. Kim, Biotechnol. Bioeng. 55 (1997).
[4] J.S. Shin, B.G. Kim, Biotechnol. Bioeng. 65 (1999).
[5] J.S. Shin, B.G. Kim, D.H. Shin, Enzyme Microbiol. Technol. 29 (2001) 232-239.
[6] M. Hohne, S. Kuhl, R. Karen, U.T. Bornscheuer, ChemBioChem. 9 (2008) 363-365.
[7] D. Koszelewski, I. Lavandera, D. Clay, D. Rozzell, W. Kroutil, Adv. Synth. Catal. 350 (2008) 2761-
2766.
[8] M.D. Truppo, N.J. Turner, J.D. Rozzell, Chem. Commun. 16 (2009) 2127–2129.
[9] K.E. Cassimjee, C. Branneby, V. Abedi, A. Wells, P. Berglund, Chem. Commun. 46 (2010) 5569-5571.
[10] H. Yun, B.K. Cho, B.G. Kim, Biotechnol. Bioeng. 87 (2004) 772-778.
[11] J.S. Shin, B.G. Kim, A. Liese, C. Wandrey, Biotechnol. Bioeng. 73 (2001) 179-187.
[12] M.D. Truppo, J.D. Rozzell, N.J. Turner, Org. Process Res. Dev. 14 (2010) 234–237.
[13] A.R. Martin, D. Shonnard, S. Pannuri, S. Kamat, Appl. Microbiol. Biot. 76 (2007) 843-851.
[14] M. Cardenas-Fernandez, W. Neto, C. Lopez, G. Alvaro, P. Tufvesson, J.M. Woodley, Biotechnol.
Progr. 28 (2012) 693-698.
[15] G. Rehn, C. Grey, C. Branneby, L. Lindberg, P. Adlercreutz, Process. Biochem. 47 (2012) 1129-1134.
[16] C.W. Lee, S.S. Yi, J. Kim, Y.S. Lee, B.G. Kim, Biotechnol. Bioproc. E. 114 (2006) 277-281.
[17] D. Koszelewski, N. Muller, J.H. Schrittwieser, K. Faber, W. Kroutil, J. Mol. Catal. B: Enzym. 63 (2010)
39-44.
24
Page 24 of 32

[18] S.S. Yi, C.w. Lee, J. Kim, D. Kyung, B.G. Kim, Y.S. Lee, Proc. Biochem. 42 (2007) 895-898.
[19] J.D. Rozzell, in: Mosbach K. (Ed.) Methods in Enzymology Immobilized Enzymes and Cell, New
York, Academic Press, pp. 479-497.
[20] A. Schulz, P. Taggeselle, D. Tripier, K. Bartsch, Appl. Environ. Microb. 56 (1990) 1-6.
[21] K. Bartsch, R. Schneider, A. Schulz, Appl. Environ. Microb. 62 (1996) 3794-3799.
[22] M.D. Truppo, H. Strotman, G. Hughes, ChemCatChem 4 (2012) 1071-1074.
[23] C.K. Savile, J.M. Janey, E.C. Mundorff, J.C. Moore, S. Tam, W.R. Jarvis, J.C. Colbeck, A. Krebber, F.J.
Fleitz, J. Brands, P.N. Devine, G.W. Huisman, G.J. Hughes, Science. 329 (2010), 305-309.
[24] O. Kirk, M.W. Christensen, Org. Process Res. Dev. 6 (2002) 446-451.
[25] W. Tischer, F. Wedekind, in: W.D. Fessner (Ed.) Biocatalysis from discovery to application,
Springer, Berlin, 1999, pp. 95-126.
[26] M.D. Lilly, J.M. Woodley, J. Ind. Microbiol. 17 (1996) 24-29.
[27] K. Buchholz, V. Kasche, U.T. Bornscheuer, U. T. Biocatalysts and enzyme technology, Second ed.
Wiley-Blackwell, Weinheim
[28] U. Hanefeld, L. Gardossi, E. Magner, Chem. Soc. Rev. 38 (2009) 453-468.
[29] P.V. Iyer, L. Ananthanarayan, Process Biochem. 43 (2008) 1019-1032.
[30] T. Gitlesen, M. Bauer, P. Adlercreutz, BBA-Lipid. Lipid Met. 1345 (1997) 188-196.
[31] L. Betancor, H.R. Luckarift, Trends Biotechnol. 26 (2008) 566-572.
[32] P. Monsan, D. Combes, Method. Enzymol. 137 (1988) 584-598.
[33] H. Bayraktar, M. Serilmez, T. Karkaş, E.B. Çelem, S. Önal, 49 (2011) 855-860.
[34] J.E. Přenosil, Ö.M. Kut, I.J. Dunn, E. Heinzle, in: Ullmann's Encyclopedia of Industrial Chemistry,
Wiley-VCH Verlag GmbH & Co. KGaA, 2000.
25
Page 25 of 32

NH+3
O
NH+3
CH3
O

-TAm/PLP
+
+
H3C
CH3
O
H3C
APH
MBA
ACE
IPA
NH+3
O
NH+3

-TAm/PLP
+
+
BA
MBA
APB
APH
Figure 1: Asymmetric synthesis of chiral amines catalysed by ω-Transaminase
26
Page 26 of 32

(S)-ω-TAm
6 resins
STEP I: IMMOBILIZATION EFFICIENCY
Loading
capacity
Residual
activity
Supports
discarded
Rel. residual
activity ≥ 50%?
No
Yes
(S)-ω-TAm
(R)-ω-TAm
5 resins
STEP II: OPERATIONAL STABILITY
Re-use of the
biocatalyst
Stability of
the supports
Supports
discarded
Rel. residual
activity ≥ 50%?
No
Yes
3 resins
STEP III: ACTIVITY UNDER ALTERNATIVE CONDITIONS
Activity under
Activity under
high temperature
organic solvent
Activity after long
storage period
Figure 2: Summary of screening steps.
27
Page 27 of 32

Figure 3 – Residual relative activity for the immobilized preparations. Calculated as (activity in the
immobilised)/(activity of loaded enzyme)×100. Results based on initial rate measurements for activity.
28
Page 28 of 32

Figure 4: Conversion over time for the different cycles (A: using immobilized (S)-ω-TAm; B: using immobilized
(R)-ω-TAm). Dashed line marks the maximum conversion after 24 hours achieved with free enzyme (used as
reference) under similar conditions.
29
Page 29 of 32