The use of l-lysine decarboxylase as a means to separate amino acids by electrodialysis

Article abstract of DOI:10.1039/c0gc00611d

Amino acids (AA's) are interesting materials as feedstocks for the chemical industry as they contain chemical functionalities similar to conventional petrochemicals. This offers the possibility to circumvent process steps, energy and reagents. AA's can be obtained by the hydrolysis of potentially inexpensive voluminous protein streams derived from biofuel production. However, isolation of the preferred AA is required in order to carry out further transformation into the desired product. Theoretically separation may be achieved using electrodialysis. To increase efficiency, specific modification to a product of industrial interest and removes charged groups of AA's with similar isoelectric points is required. Here, the reaction of l-lysine decarboxylase (LDC) was studied as a means to specifically convert l-lysine (Lys) to 1,5-pentanediamine (PDA) in the presence of l-arginine (Arg) to produce products with different charge thus allowing isolation of products by electrodialysis. Immobilization of LDC in calcium alginate enhanced the operational stability and conversion in mixtures of amino acids was highly specific. At 30 °C the presence of Arg had little effect on the activity of the enzyme although inhibition by the product PDA could be observed. Volumetric productivity was calculated and raw material and transformation costs were estimated for a potential process using a mixture of Arg and Lys. The Royal Society of Chemistry.

Full text of DOI:10.1039/c0gc00611d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 624
www.rsc.org/greenchem
PAPER
The use of L-lysine decarboxylase as a means to separate amino acids by
electrodialysis
Yinglai Teng,a Elinor L. Scott,*a Albert N. T. van Zeelandb and Johan P. M. Sandersa
Received 27th September 2010, Accepted 14th January 2011
DOI: 10.1039/c0gc00611d
Amino acids (AA’s) are interesting materials as feedstocks for the chemical industry as they
contain chemical functionalities similar to conventional petrochemicals. This offers the possibility
to circumvent process steps, energy and reagents. AA’s can be obtained by the hydrolysis of
potentially inexpensive voluminous protein streams derived from biofuel production. However,
isolation of the preferred AA is required in order to carry out further transformation into the
desired product. Theoretically separation may be achieved using electrodialysis. To increase
efficiency, specific modification to a product of industrial interest and removes charged groups of
AA’s with similar isoelectric points is required. Here, the reaction of L-lysine decarboxylase (LDC)
was studied as a means to specifically convert L-lysine (Lys) to 1,5-pentanediamine (PDA) in the
presence of L-arginine (Arg) to produce products with different charge thus allowing isolation of
products by electrodialysis. Immobilization of LDC in calcium alginate enhanced the operational
stability and conversion in mixtures of amino acids was highly specific. At 30 ◦C the presence of
Arg had little effect on the activity of the enzyme although inhibition by the product PDA could
be observed. Volumetric productivity was calculated and raw material and transformation costs
were estimated for a potential process using a mixture of Arg and Lys.
If chemicals could be synthesized from biomass fractions, the
Introduction
carbon content, process energy and the extra chemicals required
Diminishing reserves, environmental impacts, fluctuating prices
to functionalize the base petrochemicals might be reduced. For
and geopolitical problems are forcing people to find alternatives
example, carbohydrates can be fermented to bioethanol and
to fossil resource.1 The application of biomass and the associated
dehydrated to ethylene, while lignin is a potentially interesting
biorefining is gaining more interest.2 For example, The European
source to make aromatic compounds.5 Currently the use of AA’s
Union has estimated that by 2030 up to one quarter of its
as a raw material for the production of chemicals is less well
transport fuel could be replaced by biofuels.3 However, most
studied. This is surprising given that they contain functional
research focuses on bioenergy and biofuels but the application
groups, such as primary amine and carboxylic acid groups, which
of biomass to produce industrial chemicals has received less
attention.
makes them similar to many chemicals. For example, L-aspartic
acid can be converted to b-alanine that could be an intermediate
The traditional route to produce chemicals is heavily de-
for producing acrylamide or acrylonitrile,6 L-glutamic acid can
pendant on fossil resources as the carbon and energy source.
The crude oil products are cracked into hydrocarbon building
be converted to g-aminobutyric acid which can further be
converted to N-methylpyrrolidone.7,8 Other examples are also
blocks and functionalized into end products by reaction with
known.9 Potentially, AA’s could be obtained by the hydrolysis of
other chemicals such as ammonia. Thus the production route
proteins from a number of agriculture waste streams or the rest
is very energy intensive. However, many products derived from
streams of biofuel production, therefore the raw material will be
biomass, such as amino acids (AA’s), carbohydrates and lignin
potentially inexpensive and sufficiently available.1
contain the functional groups needed by the chemical industry.4,5
The hydrolysis of bio-derived proteins would results in a
mixture of up to 20 AA species. As previously mentioned
some have the potential to be converted into chemicals, while
others, such as methionine, may be more useful as animal
feed.10 To make the best use of protein, or amino acid rest
streams, it is necessary to separate them in order to make
them applicable for their different applications. A promising
technology to achieve this is electrodialysis (ED), which removes
aValorization of Plant Production Chains, Wageningen University,
Bornse Weilanden 9, 6708 WG, Wageningen, The Netherlands.
E-mail: Yinglai.Teng@wur.nl, Elinor.Scott@wur.nl; Fax: +31 317
483011; Tel: +31 317 480150
bBiobased Products, Wageningen UR Food and Biobased Research,
Bornse Weilanden 9, 6708 WG, Wageningen, The Netherlands.
E-mail: alniek.vanzeeland@wur.nl; Tel: +31 317 485244
624 | Green Chem., 2011, 13, 624–630
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
ions from a dilute solution through ion-exchange membranes to
another concentrate solution by applying an electrical potential
as the driving force. In principle, it could separate the protein
hydrolysate into acidic, basic and neutral AA streams according
to their charge behavior.11,12 However, a further ED separation
within these streams is challenging due to the similarity of their
isoelectric points (IP). To overcome this specific modification
of the charge behavior of one of these molecules is required.
This may be done by removing charged groups such as the
carboxylic acid group. In this case, enzymatic modification may
be considered. If such a modification directly leads to an end
product, the production of a chemical can be coupled with the
separation process. Therefore a sustainable chemical production
and separation route may be achieved (Fig. 1).
and urea, which have application in the plastics and fertilizer
industries.9,13 Lys can be used to prepare a monomer for the
synthesis of nylon-6 or used to make other aliphatic a,w-
diamines for nylon such as 1,5-pentanediamine (PDA).14,15 To
tackle the separation, specific modification to a product of
interest, coupled with change in charge behavior, should be
attempted. In this case, the enzyme L-lysine decarboxylase
(LDC) was applied. LDC specifically converts Lys to PDA and
CO2.16 To the best of our knowledge, no literature has reported
that the enzyme has activity towards Arg. After reaction, a
mixture of Arg (potentially with some unreacted Lys) and the
product PDA are obtained. At a pH > 11, Arg (and residual Lys)
are negatively charged while the PDA is neutral, thus allowing
further separation by ED. The potential ED separation and
reaction configuration is given in Fig. 2. The pH of the AA
mixture obtained from protein hydrolysis is altered to ca. pH 6
so the basic and acidic AA’s can be separated from the neutral
amino acids. Lys and Arg are isolated in the ED cell, which is
connected to a reactor with a constant volume. As the enzymatic
conversion of Lys to PDA proceeds, Lys and Arg are also
being continuously transported through the cation exchange
membrane. As a net result, PDA and Arg are concentrated while
Lys concentration is constant. This is comparable to a fed batch
reaction which has (solid) substrate added into the reactor.
In order to achieve optimal operating conditions of the
enzyme, and therefore have maximum conversion of Lys, a
number of parameters (such as pH, time and temperature) were
studied as well as the conditions to improve the operational
stability of the process (enzyme immobilization). To simulate the
possibility of carrying out a specific modification in a substrate
solution of Lys and Arg fed with a stream of Lys and Arg, a fed
batch method was investigated. Finally, the process costs were
estimated and a possible production route discussed.
Fig. 1 New route for chemical production based on AA’s from natural
protein as the feedstock.
Here we focus on the basic AA stream containing L-arginine
(Arg) and L-lysine (Lys) (IP pH 10.76 and pH 9.74, respectively)
obtained from a potential ED process. Both of them are inter-
esting feedstocks and thus is it would be advantageous if they
(or products there of) could be separated from one another. Arg
could be a raw material for the production of 1,4-butanediamine
Fig. 2 ED of AA’s and the combination of an enzymatic modification reactor for a basic AA stream. CEM stands for cation exchange membrane
and AEM stands for anion exchange membrane.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 624–630 | 625
©

View Article Online
the pH controlled. This method was also applied in subsequent
experiments where Arg was present in the system.
Results and discussion
Characterization of the native enzyme
LDC immobilization
As the decarboxylation reaction leads to an increase in pH due
to loss of CO2 and production of amines, an effective pH control
is important for the enzyme assay. Here a pH-stat method was
applied thereby maintaining the optimum pH of the reaction by
titrating with HCl.17 From our experimental data the optimal
pH was found to be pH 5.8 at 37 ◦C which is in accordance with
the literature.16 Between pH 5.5 and pH 6.0 similar initial rates
are observed but this decreases sharply out with this pH range.
At constant pH and temperature (pH 5.8 and 30 ◦C) the
enzyme was found to have a Km value of 0.78 mM for Lys (lit.,16
ca. 0.38 mM) and that competitive product inhibition by PDA
(Ki 9.68 0.78 mM) takes place. No activity of LDC for Arg is
observed, which is in accordance with the literature,18 however
if the concentration of Arg (in the form of Arg·HCl) exceeds
30 mM, the relative activity towards Lys starts to decrease. As
NaCl, which is generated during protein hydrolysis, could be
introduced into the potential separation–reaction process (Fig.
2), its effect on the enzyme was also studied. Again in the range
of 20–30 mM, NaCl has little or no effect on activity, although
activity is seen to decrease at higher concentrations (Fig. 3).
In both these cases the reduction in activity is suggested to be
related to the increasing ionic strength of the solution. Other
reports also show reduction in the activity of LDC as a function
of ionic strength.19 This is possibly attributed to the formation
of (partially unfolded) aggregates that are less active.20
Immobilization is considered to be an effective way to simplify
the downstream processing due to the ease of separation
of the enzyme from the products of the reaction and of
enhancing the stability of the enzyme.21 This will result in a
cost reduction of the enzyme. Covalent immobilization of LDC
on Sepabeads EC-HFA, according to the literature procedure,13
was attempted but no activity was obtained. Due to the large
molecular weight of LDC (106 Da),16 immobilization by physical
entrapment in calcium alginate was carried out. This method has
been previously reported for the successful use of L-glutamic
acid a-decarboxylase (GAD).7 From several immobilization
experiments, the supernatant consistently showed no residual
protein, as indicated by use of a Bradford test and activity
assay. This indicates that the immobilization yield was very
close to 100%. To characterize the behavior of the immobilized
◦
enzyme, fed batch reactions were carried out at 37 C and pH
5.8 using Lys·2HCl as the titrating solution as aforementioned.
The immobilized enzyme gave the same initial rate as the native
enzyme, while activity was maintained over a prolonged reaction
time (Fig. 4). No protein was found in the reaction solution,
suggesting no enzyme leakage from the beads occurs during
operation. Similar conclusions were also obtained when the
reactions were carried out at 30 ◦C and 45 ◦C, though at a
higher temperature both the native and immobilized enzyme
have higher initial rates but decay more rapidly. The effect
on stability during storage of the enzyme was also examined.
Enzyme beads stored in acetate buffer at 4 ◦C for 5 weeks
showed only a loss of ca. 10% of the initial rate. This loss is
similar to the native enzyme suspended in water and stored for
4 h at 4 ◦C. Thus, it can be concluded that calcium alginate
can completely entrap the enzyme while retaining activity and
improving stability during operation and storage.
Fig. 3 The effect of Arg and NaCl on the LDC activity. The error bars
represent standard deviation.
The effect of batch and fed batch reactor configurations
on activity were determined. When the pH of a solution of
30 mM Lys·HCl was maintained using an aqueous solution
of L-lysine dihydrochloride (Lys·2HCl, 100 mM), compared to
HCl(aq) (100 mM), comparable initial rates but higher activities
were obtained over longer periods of time. This is due to the
substrate being continuously added therefore maintaining a
substrate concentration above the Km value. The volume of the
Lys·2HCl solution added with time was monitored. This should
indicate the progress of the reaction and this was confirmed by
analysis using Ultra High Pressure Liquid Chromatography (U-
HPLC) of the PDA concentration with time. Therefore the use
of the pH-stat titration can be a reliable online method to assay
the initial rate and operational behavior of LDC while having
Fig. 4 Effect of immobilization on the operational stability of LDC
over a 16 h period at pH 5.8 and 37 ◦C.
Effect of Arg
As described earlier (Fig. 2), it is envisaged that Lys and Arg
will be fed to a reactor containing Lys and Arg during an ED
process. While the behavior of Lys in both substrate and feed
are known, the effects of the presence of Arg in the substrate
solution and in the feed remain to be explored.
626 | Green Chem., 2011, 13, 624–630
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 1 Moles of Arg, Lys and PDA at 0 h and 24 h at different reaction
temperatures (Fig. 5). The results were derived from U-HPLC results.
The standard deviation is based on duplicate experiments
in a similar way to GAD. According to the literature, the
deactivation of GAD is possibly due to the conversion of PLP
to pyridoxamine 5¢-phosphate, which then dissociates from the
enzyme to form the apoenzyme.24
30 ◦
C
37 ◦
C
45 ◦
C
Interestingly, experiments carried out at 45 ◦C appear to
show an initial increase of activity. At even higher reaction
temperatures e.g. at 50 ◦C, this phenomenon is more pronounced
(data not shown). To examine if the increase in activity was
due to a temperature gradient between the internal part of the
alginate beads and solution, the immobilized enzyme was pre-
incubated for 1 h at 45 ◦C before addition of the substrate.
However the initial increase in the activity remained. This would
suggest that changes in temperature between the bead and
the solution are not responsible for the phenomenon. Another
possible explanation could be diffusional limitation of the
immobilization material. According to the literature, diffusion
within alginate can depend on electrostatic forces as the alginate
matrix is negatively charged.25 At the pH of the reaction (pH
5.8), Arg, Lys and PDA are all positively charged. Under these
conditions it may be possible that they interact with the alginate
matrix resulting in problems with diffusion. Since the calcium
ion is responsible for the crosslinking of the immobilization
material, we added CaCl2 (30 mM) to the substrate solution
(45 ◦C) and observed that the initial increase became more
pronounced. As this phenomenon was not observed when CaCl2
was used with the native enzyme, it is suggested that the initial
increase at higher temperatures was due to interaction of charged
amino acids and PDA with the immobilization material.
Arg, 0 h (mmol)
Arg, 24 h (mmol)
Lys, 0 h (mmol)
Lys, 24 h (mmol)
PDA, 0 h (mmol)
PDA, 24 h (mmol)
0.59 0.01
0.58 0.01
0.53 0.01
0.55 0.01
0
0.59 0.01
0.60 0.02
0.57 0.01
0.59 0.03
0
0.60 0.01
0.64 0.01
0.57 0.01
0.64 0.01
0
0.37 0.01
0.61 0.01
0.61 0.05
The presence of Arg in the substrate
A substrate solution of 30 mM of Lys and Arg was titrated
with 100 mM of Lys·2HCl. The specificity of the reaction, was
confirmed by U-HPLC analysis before and after reaction which
showed that the number of moles of PDA increased with time
while the number of moles of Lys and Arg remained constant
(Table 1).
From Fig. 5, it is seen that when the substrate solution
contained Lys or Lys and Arg, an increased temperature can
increase the initial rate and decrease the operational stability.
Such behavior is typical in the use of (immobilized) enzymes.7
Substrate solutions containing Arg show similar rates when
operated at lower temperatures e.g. 30 ◦C, while at higher
temperatures the initial rates are lower compared to solutions
containing only Lys. This suggests that inhibition by Arg
takes place. Currently no literature reports that Arg has an
inhibitory effect on LDC, but it has been reported that other
decarboxylases (e.g. ornithine decarboxylase) can be inhibited by
Arg,22 thereby similar behaviour may also occur with LDC. This
indicates that a lower temperature is preferred when applying
LDC in the presence of Arg.
Deactivation of LDC can also be observed in Fig. 5. The
deactivation of LDC is believed to be linked with the loss of
PLP from the enzyme.23 However, in the current experiments
additional PLP is added during titration. As seen from Fig. 5
the deactivation of immobilized LDC does not appear to be
first order. The deactivation curves can be fitted to a double
exponential decay model (residual activity = Ae-at ¥ Be-bt) which
was applied to describe the deactivation of L-glutamic acid a-
decarboxylase (GAD).7 This indicates that LDC may deactivate
Lys and Arg in substrate and feed
Ultimately, a Lys and Arg mixture would be fed to a solution
containing Lys and Arg. As both the substrate and feed streams
are a mixture of Arg and Lys, this has been called “self titration”.
Arg in the substrate solution was shown earlier not to influence
the specific enzyme activity at 30 ◦C and indeed presence in the
feed also shows very similar activity (Fig. 6A). By using the
aforementioned double exponential model, the residual activity
and the accompanying PDA production for this process were
derived (Fig. 6B).7 As indicated in the figure, most of the Lys
conversion takes place over a 4 day period (corresponding to a
residual enzyme activity of 5%). U-HPLC results reflected that in
Fig. 5 Effect of temperature on the operational stability of immobilized LDC in a fed batch setup with Lys or Arg and Lys in the substrate solution
fed with Lys. The titration was done at pH 5.8 at 30, 37 and 45 ◦C for 24 h.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 624–630 | 627
©

View Article Online
Fig. 6 (A) Comparison of titrated by Lys·2HCl and self titration at 30 ◦C and pH 5.8 for 24 h and (B) Estimated residual activity curve and the
PDA production curve corresponding to the self titration setup.
the self titration setup Arg and PDA were accumulated from the
feed stream and the conversion, respectively, and the number
of moles of Lys with respect to Arg and PDA were reduced.
This increases the purity of the product and simplifies the ED
separation.
enzyme cost and alginate costs would be ca. 12 and 3 € per ton
of PDA, respectively.
Since 1,4-butanediamine and ethanediamine analogues of
PDA, are estimated to cost ca.1600 € per ton,1 then the reactor
and immobilized enzyme costs are only a small fraction and
the margin for protein hydrolysis and ED separation could be
around 1000 € per ton of product, which is considered to be
reasonable.
Process costs
Process costs can be divided into amino acid, vessel, enzyme and
other (hydrolysis and ED) costs.
Conclusion
In the case of amino acid cost, one could assume that the cost
of protein is ca. 200 € per ton.1,4 If all AA’s from the protein
could be isolated and have a specific added value application
then the cost of each amino acid per ton could be similar to the
protein source. Considering the molecular weights of Lys and
PDA, the Lys cost would be ca. 300 € per ton of PDA.
In this study an enzyme with a very low activity was used
(0.0294 U mg-1). However, the activity of pure LDC has been
The aim of this article was to design a fed batch bio-process
of Lys to PDA in a mixture of basic AA’s potentially obtained
from an ED process. This was carried out to modify the charge
allowing product formation (to PDA) to be coupled with a
potential separation from the remaining basic AA’s by ED. The
reaction using LDC was found to be specific but was inhibited
by high concentrations of Arg (>30 mM), NaCl (>30 mM) and
PDA (Ki ca. 10 mM). It was possible to overcome the product
inhibition by maintaining a sufficiently high Lys concentration.
Immobilization of the enzyme did not result in a change in
the activity and improved the operational stability over a 16 h
period. The presence of Arg had little effect on the activity when
the operating temperature was 30 ◦C. At higher temperatures
(37 and 45 ◦C) it was shown that a significant decrease in
the specific activity and a reduction in operational stability
took place. Based on process design and cost estimation, the
associated cost in using immobilized LDC is not a barrier in such
a product formation and separation process using amino acids.
In conclusion it is considered that such a route offers a more
techno-economically efficient method for the production of
chemicals. This is attributed to the use of biomass raw materials,
which eliminates the use of ammonia and extreme reaction
conditions, to produce amines while offering the possibility to
couple production with separation by electrodialysis.
◦
reported to be 85.7 U mg-1 (30 C, pH 5.8).16 From Fig. 6B,
1 kg of enzyme produces 2.85 kg of PDA in 48 h, but this could
increase to ca. 8.3 tons using 1 kg of pure LDC. However, the use
of a pure enzyme in a large scale application is unlikely but an
enzyme extract could have an achievable activity of 50%, which
would result in the use of 0.24 kg of LDC for the production of
1000 kg of PDA. A weight ratio of enzyme : alginate of 0.5 in
immobilized enzymes has been reported for other enzymes and
this was also used in the current work.26 The solution used to
prepare the beads contained 2 w/v % alginate and this would
result in a total bead volume of 0.024 m3. As the volume ratio
of the beads : reactor is 0.1, a volumetric productivity of 87.5 kg
PDA m-3 h-1 is achievable. However such a potentially high
productivity would lead to inhibition problems which would
affect the rate but this may be overcome by increasing the volume
of the vessel or using a (chemostat) cascade of reactors to ensure
lower PDA concentrations are maintained.27 By the choice or
design of reactor to maintain maximum productivity, the costs
for the operation of the reactor can be lowered and should not
contribute to more that 10% of the cost of the product.
Experimental
Materials
The prices of (an enzyme extract of) LDC for industrial pro-
duction and dry alginate are ca. €100 per kg of equivalent pure
enzyme and 6 € per kg, respectively.28 Thus the corresponding
L-lysine decarboxylase (from Bacterium cadaveris, type I),
sodium alginate (low viscosity), pyridoxal 5¢-phosphate
628 | Green Chem., 2011, 13, 624–630
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
monohydrate (PLP·H2O, ≥97%), HCl standard solution (volu-
metric solution, 0.1 M), L-arginine monohydrochloride (reagent
grade, ≥98%), L-lysine monohydrochloride (reagent grade,
≥98%) and L-lysine dihydrochloride (≥98%) were bought from
Sigma–Aldrich. Quick Start Bradford dye reagent and Bovine
Serum Albumin standard set were from Bio-Rad. Other chemi-
cals were used as received. Milli-Q (Millipore) purity water was
used to make all the solutions.
determined at 595 nm using a Multiskan Spectrum microplate
spectrophotometer (Thermo Labsystem).
Immobilization
120 mg LDC powder was suspended in 12 mL sodium alginate
solution (2 wt %) The LDC suspension was added dropwise
in portions of 2 mL to 40 mL of CaCl2 solution (1 wt%) to
form calcium alginate beads with a diameter of approximately
1.5 mm. The beads were stored in sodium acetate buffer (10 mM,
pH 6.2) containing 50 mM CaCl2 and 10 mM PLP at 4 ◦C. The
samples were wrapped in aluminium foil to protect them from
light.
Software
The pH and charge behavior estimation was done by software
R
CurTiPot Version 3.3.2 (2008) for MS-Excelꢀ 1997–2007. The
software contains a database of dissociation constants.
Model fitting of operational stability data was done by
WinCurveFit 1.1.8. A double exponential model was used.
Activity assay
The initial rate of native LDC was assayed using a pH-stat
method as described in the literature.17 The titrations were
performed in a titration vessel with a thermostatic jacket
( 0.1 ◦C) and monitored using a Methrohm 718 STAT Titrino.
The substrate solution was prepared by dissolving 109.6 mg
Lys·HCl in 17.80 mL MilliQ water, after which 200 mL 5 mM
PLP·H2O solution was added.
U-HPLC
Sample preparation: 500 mL of sample (or standard 0.5–0.025
mM) was taken, put into an Eppendorf safe-lock tube and mixed
with 400 mL methanol. Subsequently, 100 mL internal standard
solution (0.4 mM taurine) was added to correct for injection
errors of the U-HPLC measurement. After this pre-treatment
the sample was filtered over a 0.2 mm filter (Spartan 13/0.2 RC,
Whatman) and placed in a 1.5 mL vial, to be ready for analysis.
The derivatization of the AA’s occurs in the needle of the Dionex
U-HPLC (Ultra High Pressure Liquid Chromatography). The
derivatization procedure is as follows; draw 50 mL ultra pure
water, draw 1 mL air, draw 2 mL reagent A (OPA-MET), draw
2 mL sample (or standard) solution, mix in needle 6¥, wait 60 s,
wash needle externally, draw 2 mL reagent B (FMOC), mix in
needle 6¥, wait 15 s, wash needle externally, draw 4 mL injection
diluent, mix in needle 6¥ and inject on column.
At 37 ◦C, the substrate solution was added to the titration
vessel and brought to the required pH using either HCl or NaOH
solution. To initiate the reaction, 1 mL LDC solution (20.0 mg
mL-1) was added to the substrate solution, resulting in a reaction
mixture with 30 mM Lys and 50 mM PLP. A solution of 0.1 M
HCl was used as titration solution. The initial rate, expressed as
U mg-1 (mmol H+ min-1 mg-1), was determined as the function
of the slope of the linear part of the titration curve.
To study the pH dependent activity, assays were performed at
pH 4.5, 5.0, 5.5, 5.8, 6.0, 6.5 and 7.0 at 37 ◦C. Similarly, to study
the KM value of Lys, activity assays were performed at 30 ◦C with
0.25, 0.50, 1.00, 5.00 and 30.00 mM of Lys. To study the Arg
and PDA inhibition at 30 ◦C, assays were performed with 0.50,
1.00, 2.00 and 30.00 mM of Lys in combination with 0, 15 and
30 mM of Arg·HCl or PDA·2HCl. Additionally, to investigate
the effect of Arg concentration, assays were performed with of
0.1, 0.2, 0.3 and 0.6 M Arg·HCl. Finally, to investigate the effect
of ionic strength, activity assays were performed with 0.03, 0.1,
0.2, and 1 M NaCl.
Chromatography: the analysis was performed by using a
Dionex U-HPLC instrument (Dionex Corporation, Sunnyvale,
CA, USA) consisting of a Ultimate 3000 RS (Rapid Separation)
pump and a Ultimate 3000 autosampler, a Ultimate 3000 column
compartment with a thermo stable column area, and a Ultimate
3000 variable wavelength detector, operating with the Dionex
ChromeleonTM 6.8 software.
R
The AA’s were separated using an Acquity UPLCꢀ BEH C18
reversed phase column (1.7 mm particle size, 2.1 ¥ 50 mm) and
R
a Acclaimꢀ 120 C18 guard column (5 mm, 2.0 ¥ 10 mm), flow
rate, 1 mL min-1. The column temperature was maintained at
65◦ C. Mobile phase A consisted of 10 mM sodium phosphate,
10 mM sodium borate and 2 mM sodium azide (buffer adjusted
with HCL to pH 7.8). Mobile phase B consisted of acetonitrile,
methanol and ultra pure water 45 : 45 : 10 (v/v). Detections were
performed simultaneously, by using the variable wavelength
detector at 338 nm and 263 nm (data collection rate: 4.0 Hz,
time constant 0.18 s).
Operational stability
The operational stability of native and immobilized enzyme was
det◦ermined by performing fed batch reactions at 30, 37 and
45 C over a period of 24 h. The reaction setup was similar to
that previously described in the Activity assay section.
The reaction vessel was filled with 20 mL 30 mM Lys·HCl and
50 mM PLP·H2O solution and subsequently the pH was adjusted
to 5.8 using a NaOH solution. At the required temperature,
20.0 mg LDC or an equivalent of immobilized enzyme was
added to initiate the reaction. The titration solution was an
aqueous solution of 0.1 M Lys·2HCl and 50 mM PLP·H2O
(50 mM). The activity (U mg-1) in time was determined as the
function of the tangent slope of the titration curve at given points
in time. The residual activity is defined as the activity at the given
point in time relative to the initial activity.
Gradient, expressed as % solvent B: 0–6.98 min: 2–57% B,
6.98–7.80 min: 57–100% B, 7.80–9.49 min: 100% B, 9.49–9.72
min: 100–2% B, 9.72–10 min 2% B.29
Bradford protein assay
Bradford protein assays were performed using the Bio-Rad
Protein Assay Kit II with BSA standard. The absorbance was
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 624–630 | 629
©

View Article Online
To study the effect of Arg on the operational stability of
immobilized LDC, the operational stability was determined in
the presence of 30 mM Arg·HCl at 30, 37 and 45 ◦C over a period
of 24 h. The standard titration solution was supplemented with
0.1 M Arg·HCl to maintain the level of Arg during the reaction.
Using U-HPLC analysis, the concentrations of Arg, Cad and Lys
were determined at the beginning and the end of the reaction.
8 T. M. Lammens, M. C. R. Franssen, E. L. Scott and J. P. M. Sanders,
Green Chem., 2010, 12, 1430–1436.
9 P. M. Ko¨nst, M. C. R. Franssen, E. L. Scott and J. P. M. Sanders,
Green Chem., DOI: 10.1039/c0gc00564a.
10 W. Leuchtenberger, K. Huthmacher and K. Drauz, Appl. Microbiol.
Biotechnol., 2005, 69, 1–8.
11 Y. Hara, Bull. Chem. Soc. Jpn., 1963, 36, 1373–1376; J. Sandeaux,
R. Sandeaux, C. Gavach, H. Grib, T. Sadat, D. Belhocine and N.
Mameri, J. Chem. Technol. Biotechnol., 1998, 71, 267–273.
12 J. J. Krol, Ph.D. Thesis, University of Twente, 1997.
13 P. M. Ko¨nst, P. M. C. C. D. Turras, M. C. R. Franssen, E. L. Scott
and J. P. M. Sanders, Adv. Synth. Catal., 2010, 352, 1493–1502.
14 PCT, WO2005123669, 2005.
Acknowledgements
The authors would like to thank Tijs Lammens and Kees
van Kekem (WUR, The Netherlands) for providing titration
equipment. We would also like to thank Dr D. C. (Kitty)
Nijmeijer and O. M. (Olga) Katten Readi M.Sc. for their
useful discussions on electrodialysis. This project is financed by
Technology Foundation STW, The Netherlands.
15 EP1482055, 2004.
16 K. Soda and M. Moriguchi, Methods Enzymol., 1971, 17, 677–681.
17 J. Vienozˇinskiene˙, R. Janusˇevicˇiu¯te˙, A. Pauliukonis and D.
Kazlauskas, Anal. Biochem., 1985, 146, 180–183.
18 W. C. White and G. G. Guilbault, Anal. Chem., 1978, 50, 1481–1486.
19 E. F. Gale and H. M. R. Epps, Biochem. J, 1944, 38, 232–242; N. D.
Tran, J. L. Romette and D. Thomas, Biotechnol. Bioeng., 1983, 25,
329–340.
20 D. L. Sabo, E. A. Boeker, B. Byers, H. Waron and E. H. Fischer,
Biochemistry, 1974, 13, 662–670.
21 R. A. Sheldon, Adv. Synth. Catal., 2007, 349, 1289–1307.
22 A. A. Pantazaki, C. G. Anagnostopoulos, E. E. Lioliou and D. A.
Kyriakidis, Mol. Cell. Biochem., 1999, 195, 55–64.
Notes and References
1 E. Scott, F. Peter and J. Sanders, Appl. Microbiol. Biotechnol., 2007,
75, 751–762.
23 J. Mandelstam, J. Gen. Microbiol., 1954, 11, 426–437.
24 M. P. Meeley and D. L. Martin, Cell. Mol. Neurobiol., 1983, 3, 39–54.
25 A. Martinsen, I. Storrø and G. Skja˚k-Bræk, Biotechnol. Bioeng.,
1992, 39, 186–194.
2 S. Octave and D. Thomas, Biochimie, 2009, 91, 659–664.
3 Biofuels Research Advisory Council, Biofuels in the European Union
– a vision for 2030 and beyond, Final report of the Biofuels Research
Advisory Council, Office for Official Publications of the European
Communities, Luxembourg, 2006.
4 J. Sanders, E. Scott, R. Weusthuis and H. Mooibroek, Macromol.
Biosci., 2007, 7, 105–117.
5 J. Van Haveren, E. L. Scott and J. Sanders, Biofuels, Bioprod.
Biorefin., 2008, 2, 41–57.
6 P. M. Ko¨nst, M. C. R. Franssen, E. L. Scott and J. P. M. Sanders,
Green Chem., 2009, 11, 1646–1652.
7 T. M. Lammens, D. D. Biase, M. C. R. Franssen, E. L. Scott and J.
P. M. Sanders, Green Chem., 2009, 11, 1562–1567.
26 K. Won, S. Kim, K. Kim, H. W. Park and S. Moon, Process Biochem.,
2005, 40, 2149–2154.
27 P. M. Doran, Bioprocess Engineering Principles, Academic Press,
2009, ch 13, pp. 333–391.
28 P. Van Hoek, A. Aristidou, J. J. Hahn and A. Patist, Chem. Eng.
Prog., 2003, 99, 37S–42S.
29 F. Steiner, S. Fabel and F. McLeod, An automated system for at-line
process analysis of biopharmaceutical fermentation reactions, 2009,
http://www.dionex.com/en-us/webdocs/70923-lpn_2236-01.pdf.
630 | Green Chem., 2011, 13, 624–630
This journal is
The Royal Society of Chemistry 2011
©