Embed Size (px)
Citation preview
Integration of production and aqueous two-phase systemsextraction of extracellular Fusarium solani pisi
cutinase fusion proteins
M.T. Cunha, M.J.L. Costa, C.R.C. Calado, L.P. Fonseca,M.R. Aires-Barros *, J.M.S. Cabral
Centro de Engenharia Biologica e Quımica, Instituto Superior Tecnico, Av. Rovisco Pais, 1049-001 Lisbon, Portugal
Received 25 October 2001; received in revised form 27 March 2002; accepted 8 July 2002
Abstract
Genetic engineering was integrated with the production and purification of Fusarium solani pisi cutinases, in order to
obtain the highest amount of enzyme activity units, after purification. An aqueous two-phase system (ATPS) of
polyethylene glycol 3350, dipotassium phosphate and whole broth was used for the extraction of three extracellular
cutinases expressed in Saccharomyces cerevisiae . The production/extraction process was evaluated regarding cutinases
secretion in the medium, partition behaviour and extraction yields in the ATPS. The proteins studied were cutinase wild
type and two fusion proteins of cutinase with the tryptophane-proline (WP) fusion tags, namely (WP)2 and (WP)4. The
(WP)4 fusion protein enabled a 300-fold increase of the cutinase partition coefficient when comparing to the wild type.
However, the secretion of the fusion proteins was lower than of the wild type cutinase secretion. A batch extraction
strategy was compared with a continuous extraction in a perforated rotating disc contactor (PRDC). The batch and
continuous systems were loaded with as much as 60% (w/w) whole cultivation broth. The continuous extraction strategy
provided a 2.5 higher separation capacity than the batch extraction strategy. Considering the integrated process, the
cutinase-(WP)2 proved to lead to the highest product activity, enabling five and six times more product activity than the
wild type and the (WP)4 fusion proteins, respectively.
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: Aqueous-two phase systems; Cutinase fusion protein; Batch and continuous extraction
1. Introduction
The production of foreign proteins using a
selected host with the necessary post-translational
modifications is one of the key successes in
modern biotechnology. This methodology allows
the industrial production of proteins that were
otherwise produced in small quantities. However,
* Corresponding author. Tel.: �/351-21-8419065; fax: �/351-
21-8419062
E-mail address: [email protected] (M.R. Aires-
Barros).
Journal of Biotechnology 100 (2003) 55�/64
www.elsevier.com/locate/jbiotec
0168-1656/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 8 - 1 6 5 6 ( 0 2 ) 0 0 2 2 5 - 0
the separation and purification of these proteinsfrom the cultivation media constitutes a major
bottleneck for the widespread commercialisation
of recombinant proteins. The major production
costs (50�/90%) for a typical biological product
resides in the purification strategy. There is a need
for efficient, effective and economic large-scale
bioseparation techniques, to achieve high purity
and high recovery, while maintaining the biologi-cal activity of the molecule. The key now lies in
understanding downstream processing, integrating
it with upstream processing, and thus providing
better insights into improving the economics of the
‘whole’ process itself. Common sense dictates that
a large-scale process should be designed to mini-
mise the number of steps while maintaining high
yields and product purity, quality and activity.The choice of the purification scheme depends
on the location of the target protein and on the
desired purity of the product, which is also
determined by the further utilisation of the pro-
tein. It has been demonstrated that it is possible to
use liquid�/liquid extraction technology to the first
downstream purification steps enabling simulta-
neously separation and concentration of targetprotein (Kula, 1985; Albertsson, 1986; Diamond
and Hsu, 1992; Zaslavsky, 1995; Costa et al.,
2000).
Aqueous two-phase systems (ATPS) offer dif-
ferent physical and chemical environments, which
allow for the partitioning of solutes such as
proteins (Cunha et al., 2000), cell particles and
nucleic acids. The phases of the system have a highcontent of water (between 80 and 90%) and have
been shown to provide a protective environment
for biological materials (Albertsson, 1986).
Recombinant DNA technology allows the fu-
sion of affinity tags to the original protein. This
genetic modification of the target protein can
greatly increase the purification efficiency in the
purification of proteins in PEG/salt ATPS (Kohleret al., 1991a,b; Berggren et al., 1999, 2000; Costa et
al., 2000; Bandmann et al., 2000). Its strong
affinity for the PEG rich phase, was attributed to
the high tryptophan content.
The technology for the large-scale recovery,
with these systems, has been, mostly, based on
batch wise mode. An alternate scheme to the batch
extraction is the use of a column type extractor toimprove extraction efficiency using ATPS. The
selection of a particular contacting column, de-
pends upon the needs of the operation, the proper-
ties of the biomolecules, and the type of ATPS
involved (Laddha and Degaleesan, 1976; Cunha
and Aires-Barros, 2002). The perforated rotating
disc contactor (PRDC) is well suited for systems
with low interfacial tension.The model system used for this study was a
recombinant Fusarium solani pisi cutinase, which
consisted of a single polypeptide chain with one
disulphide bridge, which is essential for the activity
of the enzyme. The X-ray structure of cutinase
revealed an a/b hydrolase fold with an active site
serine accessible to the solvent. This serine is a
member of the Ser-Asp-His catalytic triad (Marti-nez et al., 1992). Mutants of this recombinant
cutinase have been constructed with a fusion
peptide composed of tryptophan residues inter-
spersed with prolines, which enabled the increase
of the protein hydrophobicity.
The aim of this work was to compare the effect
of the fusion peptide (WP) on the secretion,
partitioning and extraction yields. The geneticengineering was integrated with the production
and purification scheme, i.e. the integration is here
considered in an economic sense of the ‘whole’
process. The goal is to find a fusion protein, which
leads to the highest amount of enzyme activity
units, after purification, with such ATPS with
whole broth. Two different extraction strategies
(batch and continuous mode operation) were alsoevaluated.
2. Materials and methods
2.1. Recombinant cutinases
F. solani pisi cutinase was cloned and expressed
in S. cerevisiae strain MM01. Cutinase exhibitsmolecular weights of 20 605 (wild type) and an
isoelectric point of 7.8. Fusion proteins of cutinase
with affinity tags composed of Tryptophane (W)
peptides interspersed with proline (P) were con-
structed and provided by the Unilever Research
Laboratory, Vlaardingen, The Netherlands. The
M.T. Cunha et al. / Journal of Biotechnology 100 (2003) 55�/6456
tags (WP)2 and (WP)4 have been fused to the C-terminus of cutinase. The molecular weights of the
resulting cutinases were 21 172 and 21 739 for
cutinase-(WP)2 and cutinase-(WP)4, respectively.
2.2. Cutinase production
A two fed-batch cultivation was performed in a
5 l Braun bioreactor (Biostat† MD B. Braun). Thefirst phase was a batch growth phase initiated by
transfer of inoculum cells (1.2 g dcw l�1) to a 2.0 l
culture medium. After 18 h of the batch growth
phase D(�/)-galactose was added for induction of
cutinase expression. Following induction, an ex-
ponential feeding phase was started by addition of
glucose and yeast extract, considering a constant
specific growth rate of m�/0.14 h�1.The composition and cultivation conditions of
the pre-cultivation for inoculum growth and of the
cutinase production were performed according to
Calado et al. (2002).
The culture broth was added directly to the
ATPS without any pre-treatment.
2.3. Dry cell weight
Optical density was measured at 600 nm after
appropriate dilution with 0.8% (v/v) NaCl. Dry
cell weights were obtained by filtrating the culture
and subsequently drying the filter in a microwave
oven at 105 8C for 5 min until constant weight.
Dry cell weight measurements were correlated to
turbidity measurements and subsequently con-verted to dry cell weight.
2.4. Cutinase extraction
2.4.1. Polymers and chemicals
Polyethylene glycol (PEG) 3350, p-nitrophenyl
butyrate (PNPB), cholic acid and Coomassie
Brilliant Blue G, dye content of approximately98% were obtained from Sigma. Tetrahydrofuran
and phosphate salts were of analytical grade and
were supplied by Merck.
2.4.2. Batch extraction
ATPS were prepared from stock solutions by
weighting appropriate amounts of 50% (w/w) PEG
3350, 55% (w/w) K2HPO4 and culture broth. Theequilibrium systems were done in 10 ml centrifugal
test tubes. The tubes were manually mixed by
turning them up side down several times. The
separation was achieved with a low speed centri-
fugation step (3000 rpm for ca. 10 min). After the
complete settling of the phases, their volumes were
noted and samples of each phase were taken. At
least two control systems were prepared for eachset of conditions.
The blank was a system without culture broth.
The experiments were performed at 269/2 8C. The
top phase pH was measured and found to be 8.59/
0.5.
2.4.3. Continuous extraction
A schematic representation of the experimental
set-up is shown in Fig. 1.
The PRD column was made of glass and had an
internal cross section area of 804 mm2 with an
expanded cross section area of 5542 mm2. The
expansion was designed to enable higher input
flows, without the occurrence of flooding. Sevenperspex discs distanced 15 mm from each other
with 24% free area each, 1 mm height and 30 mm
Fig. 1. Schematic diagram of the experimental set-up, (A) light
phase feed; (B) heavy phase feed; (C) rotameter; (D) heavy
phase inlet; (E) light phase outlet; (F) heavy phase outlet; (G)
light phase inlet.
M.T. Cunha et al. / Journal of Biotechnology 100 (2003) 55�/64 57
diameter were used. The rotor was driven by anelectric motor, at a speed of 170 rpm, which was
measured by a portable tachometer. The shaft had
a diameter of 8.5 mm.
Both phases were fed to the column by gravity,
in a counter-current operation mode, from over-
head tanks. The flow-rates to the column were
indicated by rotameters, fine control being
achieved by means of stainless steel needle valves.The rotameters were used to indicate the con-
stancy of the flow and the rates were determined
by timing the flow of known volumes of the phases
to the nearest 1/10 of the second. Room was kept
at 209/1 8C. The column had a cooling jacket,
which was kept at 20 8C.
Before starting the run, the column was filled
with 235 ml of clear bottom phase and 250 ml ofclear top phase. The bottom phase consisted of
17% (w/w) dipotassium phosphate and 0.05% (w/
w) PEG 3350. The top phase was composed of
30% (w/w) PEG 3350 and 3% (w/w) dipotassium
phosphate.
The agitation was started, and the top phase
flow rate was fixed at the desired flow rate. After,
the bottom phase including the 83% (w/w) wholebroth (ca. 2.2% (w/w) dry cell weight and approxi-
mately 0.4 g l�1 total protein) was fed into the
column. Samples of the outlet and inlet streams
were taken and checked for protein and activity
content. The steady state was achieved when a
constant activity in the top outlet-stream samples
was observed.
The control systems were done in 10 mlcentrifugal test tubes. Analogous volumes to the
inlet streams of top and bottom phases were added
and mixed by turning the tubes up side down
several times. The tubes were subjected to the same
treatment as in the batch extraction experiments.
2.5. Cutinase activity assay
The cutinase esterolytic activity was determined
spectrophotometrically, following the hydrolysis
of PNPB at 400 nm. Twenty microlitres of sample
were added to 980 ml of a 0.56 mM PNPB solution
in 50 mM potassium phosphate buffer pH 7 with
11.3 mM sodium cholate and 300 mM tetrahy-
drofuran. The reactions were followed for 1 minagainst the blank solution.
One unit of activity was defined as the amount
of enzyme required to convert 1 mmol of PNPB in
p-nitrophenol in 1 min, under the specified condi-
tions. The extinction coefficient of p-nitrophenol
was considered to be 1.84�/104 M�1 cm�1, from
the supplier Sigma. Each sample was analysed at
least twice. The samples were all diluted withdistilled water.
As the phase components may enhance the
activity, the specific activity of three broth dilu-
tions in contact with the phases (diluted as the
samples) were checked. The correction factor was
given by the ration of the average specific activities
between the top and bottom phases.
The total activity recovered in both phases wascompared with the one of the initial preparation,
taking into account the activation due to the phase
components.
The partitioning coefficient (K ), the yield of
cutinase (Y ), the purification factor and the
concentration factor were defined as follows:
K�[U ]top
[U ]bot
Volume ratio�Top phase volume
Bottom phase volume
PF�U=mg proteinsample
U=mg proteinfeed
Ytop�Utop
Uadded
�100
Ulost�Uadded � Utop � Ubot
Uadded
�100
CF�[U ]top
[U ]broth
2.6. Protein determination
Total protein was quantified using the Bradford
(1976) assay. Coomassie reagent had a dye content
M.T. Cunha et al. / Journal of Biotechnology 100 (2003) 55�/6458
of 98% and absorbance was measured at 595 nm.Bovine serum albumin was used as standard (using
a volume ratio of sample/coomassie mixture of 30/
150).
Total protein recovered in both phases was
compared with the protein initially introduced.
Like in the activity cutinase assay, as the phase
components may change the results obtained, the
total protein of three different broth dilutions inthe phases (diluted as the samples) were checked
and corrected when necessary.
3. Results and discussion
The integrated approach described in this work
pretends to find the optimal strategy which
combines high cutinase production and extraction
yields using ATPS. The purification tags were
designed to enable high recoveries in such extrac-
tion systems.
3.1. Cutinase production
A fed-batch cultivation was performed in a two-stage culture comprising one batch yeast growth
phase followed by an exponential feed phase for
cutinase production, in order to achieve high yield,
high volumetric productivity and high product
concentration. This strategy enabled high cellular
density for the three MM01 S. cerevisiae recombi-
nant strains (between 38 and 40 g dcw l�1).
However, the increased hydrophobic length ofthe peptide (WP)n fused to cutinase had a negative
effect on the extracellular cutinase activity. For the
cutinase wild type, cutinase-(WP)2 and cutinase-
(WP)4 producing strains with extracellular activ-
ities of 162, 88 and 2.2 U ml�1, and specific
activities of 266, 200 and 15 U mg�1 protein, were,
respectively, obtained (Calado et al., 2002).
The different cutinase extracellular activitiesobserved for the several yeast strains could be
due to different production levels of the cutinases
or, most probably to the impaired secretion of the
more hydrophobic ones. This hypothesis is sup-
ported by the work performed by Sagt et al.
(1998).
3.2. Cutinase extraction with PEG/phosphate
The cultivation broth obtained from the opti-
mised fed-batch strategy for cutinases extracellular
production was directly used in the ATPS both in
batch and continuous operation mode.
3.2.1. Batch extraction
The effect of the fusion peptide and its length onthe cutinase partitioning in ATPS of 5% (w/w)
PEG 3350/15% (w/w) K2HPO4 was evaluated. The
ATPS composition used was selected because of its
low volume ratio (all systems exhibited a volume
ratio of 0.2), which enables the potential concen-
tration of the protein in the top phase.
The utilisation of PEG 3350 was based on
previous work (Sebastiao et al., 1993, 1994), inwhich it was demonstrated that an increase in PEG
molecular weight generally decreased the cutinase
partition coefficient. On the other hand, using
PEG with low molecular weights leads to higher
expenses for polymer and salt, and, therefore, to
additional costs. We have opted for an average
and commercially available molecular weight.
The pH value of all systems studied was 8.5 (9/
0.5), a pH optimal for the activity of the cutinase,
and a pH value at which cutinase is negatively
charged (pI�/7.8), therefore, enabling to take
advantage of electrostatic repulsion between the
phosphate ions, enriched in the bottom phase, and
cutinase. We recall that we are interested in
directing the target protein, cutinase, to the top
phase whereas the remaining proteins and cellsshould remain in the bottom phase.
The effect of the fusion peptide on the partition
coefficient and recovery yields is presented in
Table 1. The wild type cutinase preferably parti-
tioned to the salt rich phase, while the mutant
cutinases preferred the PEG top phase. The cells
remained in the bottom phase for all cases, as
expected. The fusion peptide enabled approxi-mately, a 300-fold increase of the cutinase parti-
tion coefficient, comparing the wild type with the
(WP)4 cutinases. For this mutant a yield of 100%
and a concentration factor over five were ob-
tained. Comparing the two fusion cutinases
((WP)2 and (WP)4), an increase of the extraction
M.T. Cunha et al. / Journal of Biotechnology 100 (2003) 55�/64 59
results was observed, due to the duplication of the
hydrophobic amino acids.
The batch extraction systems were loaded with
different amounts of whole broth, varying from 5
to 63% (w/w) (Table 1). The results obtained show
the independence of the cutinases extraction of the
whole broth. This allows the integration of the
cultivation step with the isolation and purification
steps. Furthermore, the extraction and isolation of
the fused cutinase from whole cultivation broth
was obtained in a single step, with its concentra-
tion in the PEG rich phase.
Regarding the purification factor, although it is
an important parameter in extraction evaluation,
in this case it was of minor importance. In fact,
cutinase constitutes around 95�/99% of the total
protein content of the broth, therefore, high
purification factors cannot be expected.
With the PEG/salt system selected and by using
a fused cutinase, integration of the first two steps
of the purification scheme into one step was
achieved, which allowed simultaneously separa-
tion and concentration of the target protein.
Although, the introduction of a hydrophobic tag
increased cutinase partition to the PEG rich phase
(Fig. 2A), when looking to the expression levels by
the microrganism the secreted activity and total
protein, decreased with increased hydrophobicity
of the fusion protein (Fig. 2B). As a consequence,
in spite of the better partition behaviour of the
cutinase-(WP)4 fusion protein the activities ob-
tained in the top phase (8.2 U ml�1) were lower
than the ones obtained for the wild type (23 U
ml�1).
3.2.2. Continuous extraction
Based on the cutinase extracellular activity and
on the partition results (around 125 U ml�1 in the
top phase), the fusion protein cutinase (WP)2 was
selected for the cutinase continuous extraction
studies.
A PRDC was applied for cutinase extraction
from the whole cultivation broth. This contactor is
well suited for systems with low interfacial tension.
The use of extraction columns is justified when the
selectivity of the system is not extreme and since
the cells partition to the bottom phase, the
cutinase partition coefficient should be higher
than one. Both these conditions are met with
cutinase-(WP)2 extraction in PEG/phosphate
ATPS.Like for the batch strategy, the continuous
extraction experiments were restricted to one
system, the ATPS PEG 3350 and K2HPO4. The
tie-line used was fixed (TLL�/31.5% (w/w)) and
was chosen to allow a safe operation, i.e. as the
cultivation broth conditions have an effect on the
equilibrium phases composition we wanted to
assure that the system would not fall in the one
phase region of the phases diagram.
The cutinase content of the feed salt rich, was
dependent on the cultivation productivity, and
varied between 20 and 30 Activity Units ml�1.
Table 1
Experimental results of cutinase extractions in ATPS of 5% (w/w) PEG and 15% (w/w) K2HPO4 at room temperature: partition
coefficient; top phase yield, purification and concentration factors
Cutinase Broth (% w/w) K Y PF CF
Wild type 63 0.249/0.03 4.389/0.03 1.79/0.3 0.199/0.01
Cutinase-(WP)2 5 4.09/0.1 409/1 n.d. 0.199/0.02
15 4.29/0.5 409/3 n.d. 0.659/0.09
30 4.49/0.2 429/1 n.d. 1.279/0.02
63 5.19/0.7 369/4 2.69/0.3 1.39/0.1
Cutinase-(WP)4 20 1089/6 949/8 n.d. 1.239/0.02
40 809/6 959/2 n.d. 2.29/0.1
63 629/9 1009/5 5.49/0.9 5.39/0.1
All systems exhibited a volume ratio of 0.2.
M.T. Cunha et al. / Journal of Biotechnology 100 (2003) 55�/6460
Table 2 shows the experimental results of the
column extraction runs, after attaining the sta-
tionary state, and of the controls (10 ml test tube
batch extractions), obtained for the fused cutinase
(WP)2. When adding the salt to the culture broth,
some cell rupture occurs, which leads to an
increase of the cutinase and protein concentrations
compared with the initial broth. These values
ranged from 11 to 40% for cutinase concentration
and from 30 to 50% for protein concentration. The
yields presented in Table 2 were determined with
respect to the inlet stream, i.e. already taking into
account the concentrations after the cell rupture.
However, the purification and concentration fac-
tors are determined with respect to the cultivation
broth conditions (before adding the salt).
The run 2/2 enabled a higher extraction effi-
ciency than the one step procedure, with a extrac-
tion yield of 83%, a purification factor of 3.6 and a
concentration factor of 0.97.
Comparing the continuous extraction yield with
the one of the control system of run 2/2 (Table 2),if a two step batch extraction would be applied to
this system (with a volume ratio of 1 and a
partition coefficient of 1.89/0.1), a final yield of
78% would be achieved. Only a three-step extrac-
tion would enable a recovery of 90%. The global
efficiency (based on the Murphee efficiency for the
continuous phase) was found to be 42%.
3.3. Integration of cutinase production and
extraction
In order to integrate the production with the
extraction of cutinase using PEG/phosphate
ATPS, a comparison of the batch versus contin-
Fig. 2. Effect of fusion peptide on the partition in an ATPS of 5% PEG, 15% K2HPO4 and 63% whole broth (% w/w) (A) and, on the
secreted protein (grey) and activity (black) achieved from fed-batch cultivations (B).
Table 2
Experimental results of the column extraction runs, after attaining the stationary state, and of the control systems
Run D.C.W (g l�1) Cutinase (U ml�1) Protein (mg l�1) Y PF CF
Column Control Column Control Column Control
1/1 25.5 23 333 55 519/1 1.8 2.119/0.04 0.70 0.829/0.02
2/2 25.5 23 403 83 519/1 3.6 2.559/0.05 0.97 0.829/0.02
3/3 26.1 24 333 61 609/11 3.4 69/2 0.84 1.09/0.2
2/4 27.0 26 276 51 569/1 4.1 1.919/0.05 1.1 1.459/0.05
4/2 9.30 40 358 80 869/4 1.6 3.89/0.9 0.21 0.449/0.01
Conditions of the cultivation broth used in each run are given. The run designation refers to the continuous phase volumetric flow
rate (ml min�1) to dispersed phase volumetric flow rate (ml min�1). Ten millilitre test tube batch extractions.
M.T. Cunha et al. / Journal of Biotechnology 100 (2003) 55�/64 61
uous extraction, as well as different fusion pro-
teins, with regard to extraction performance, was
made (Fig. 3). This figure comprises the optimal
cutinase production using a fed-batch strategy, the
extraction results using ATPS in a batch and
continuous mode and a rough economic evalua-
tion based in laboratory and scale-up extraction
data. To be able to make these comparisons a
common base of 2.88 l of cultivation broth
processed in 1 day was selected, as this is the
total amount of broth that the selected column
is able to process in 1 day, at the flow rate of
2 ml min�1.Comparing the concentration of cutinases in the
top phase, cutinase-(WP)2 extraction in batchwise
mode led to the higher product concentration (125
U ml�1). As previously mentioned, in spite of the
extreme yields obtained for the (WP)4 fusion
protein, the product concentration in the top
phase was lower than the one obtained with the
wild-type due to its very low expression levels.
By using a PRDC with ATPS for continuous
extraction of cutinase-(WP)2, higher yields were
obtained, increasing from 36 to 83%.
Regarding the separation capacity, i.e. the total
amount of units separated cutinase-(WP)2 extrac-
tion in continuous mode led to the higher results
(252 426 Activity Units per day). It is actually 2.7
times higher than one step batch extraction (91 238
Activity Units per day).
Regarding the cost evaluation, based only on
the chemicals used, although the continuous
extraction required larger amounts of polymers
and salts, as it enabled a larger amount of units
recovered, it also led to the lowest cost of
chemicals/activity unit after extraction.
Fig. 3. Results obtained for 2.88 l cultivation broth and extraction process processed in 1 day, either in batch or continuous (dispersed
and continuous flow rate of 2 ml min�1).
M.T. Cunha et al. / Journal of Biotechnology 100 (2003) 55�/6462
4. Conclusions
A fed-batch cultivation strategy allowed high
cellular density for the three strains, although the
cutinases extracellular activity decreased with the
increased hydrophobicity of the fusion peptide.
The extraction of cutinase with PEG 3350/
K2HPO4 ATPS, proved to suit for the extraction
of the cutinase-(WP)n fusion proteins to the PEG
rich phase while the wild type had a tendency for
the salt rich phase. This different behaviour of the
fusion proteins is due to the hydrophobic tag that
directs the protein to the more hydrophobic phase
(PEG rich phase). The broth does not interfere in
the partitioning of the target protein�/cutinase at
least up to 63% (w/w). The partition coefficient of
the wild type cutinase, (WP)2 and (WP)4 was
found to be 0.24, 4.4 and around 80, respectively.Although, the partition coefficient and the yield
of the most hydrophobic fusion protein were very
high, due to its lowest secreted yields it rendered it
worse than the wild type in what matters to
separation capacity.
The fusion protein with the (WP)2 tag has
proved to be the best candidate combining expres-
sion and extraction yields. The highest separation
capacity was obtained for this protein. Comparing
the continuous with the batch extraction mode, it
was possible to achieve two times higher recovery
yields with the former. The optimum flow ratio
found was 1, with a linear velocity of 3.67�/10�5
m s�1 for both phases. However, the outlet
solution is 1.5 times more diluted in continuous
than in batch mode operation.
Acknowledgements
The authors acknowledge Dr Maarten Egmond,
Dr Arthur Fellinger and Dr Maurice Mannesse
from Unilever Research Laboratory for providing
the yeast strains. M.T. Cunha, M.J.L. Costa and
C. Calado acknowledge fellowship from Fundacao
para a Ciencia a Tecnologia, Portugal. This
investigation was supported by EU project BIO4-
CT 96-0435.
References
Albertsson, P.-A., 1986. Partitioning of Cell Particles and
Macromolecules. Wiley, New York.
Bandmann, N., Collet, E., Leijen, J., Uhlen, M., Veide, A.,
Nygren, P.-A., 2000. Genetic engineering of the Fusarium
solani pisi lipase cutinase for enhanced partitioning in PEG-
phosphate aqueous two-phase systems. J. Biotechnol. 79,
161�/172.
Berggren, K., Veide, A., Nygre, P.-A., Tjerneld, F., 1999.
Genetic engineering of protein�/peptide fusions for control
of protein partitioning in thermoseparating aqueous two-
phase systems. Biotechnol. Bioeng. 62, 135�/144.
Berggren, K., Tjerneld, F., Veide, A., 2000. Peptide fusion tags
with tryptophane and charged residues for control of
protein partitioning in PEG-potassium phosphate aqueous
two-phase systems. Bioseparation 9, 69�/80.
Bradford, M.M., 1976. A rapid sensitive method for the
quantification of microgram quantities of protein utilizing
the principle of protein-dye binding. Anal. Biochem. 72,
248�/254.
Calado, C.R.C., Mannesse, M., Egmond, M., Cabral, J.M.S.,
Fonseca, L.P., 2002. Production of wild type and peptide
fusion cutinases by recombinant S. cerevisiae . Biotechnol.
Bioeng. 78, 692�/698.
Costa, M.J.L., Cunha, M.T., Cabral, J.M.S., Aires-Barros,
M.R., 2000. Scale-up of recombinant cutinase recovery by
whole broth extraction with PEG-phosphate aqueous two-
phase. Bioseparation 9, 231�/238.
Cunha, M.T., Cabral, J.M.S., Tjerneld, F., Aires-Barros, M.R.,
2000. Effect of salts and surfactants on the partitioning of
Fusarium solani pisi cutinase in aqueous two-phase systems
of thermoseparating ethylene oxide/propylene oxide ran-
dom copolymer and hydroxypropyl starch. Bioseparation 9,
203�/209.
Cunha, M.T., Aires-Barros, M.R., 2002. Large scale extraction
of proteins. Mol. Biol. 20, 29�/40.
Diamond, A.D., Hsu, J.T., 1992. Aqueous two-phase systems
for biomolecule separation. In: Fiechter, A. (Ed.), Advances
in Biochemical Engineering/Biotechnology, vol. 47.
Springer, Berlin, Heidelberg, pp. 89�/135.
Kohler, K., Ljungquist, C., Kondo, A., Veide, A., Nilsson, B.,
1991. Engineering proteins to enhance their partitioning
coefficients in aqueous two-phase systems. Biotechnology 9,
642�/646.
Kohler, K., Veide, A., Enfors, S.-O., 1991. Partitioning of b-
galactosidase fusion proteins in PEG/potassium phosphate
aqueous two-phase systems. Enzyme Microbiol.Technol. 13,
204�/209.
Kula, M.R., 1985. Liquid�/liquid extraction of biopolymers. In:
Cooney, C.L., Humphrey, A.E. (Eds.), Comprehensive
Biotechnology. The Principles of Biotechnology: Engineer-
ing Considerations, vol. 2. Pergamon Press, New York, pp.
451�/471.
Laddha, G.S., Degaleesan, T.E., 1976. Transport Phenomena
in Liquid Extraction. Tata McGraw-Hill, New Delhi.
M.T. Cunha et al. / Journal of Biotechnology 100 (2003) 55�/64 63
Martinez, C., De Geus, P., Lauwereys, M., Matthyssens, G.,
Cambillau, C., 1992. Fusarium solani cutinase is a lipolytic
enzyme with a catalytic serine accessible to solvent. Nature
356, 615�/618.
Sagt, C.M.J., Muller, W.H., Boonstra, J., Verkleij, A.J.,
Verrips, C.T., 1998. Impaired secretion of a hydrophobic
cutinase by Saccharomyces cerevisiae correlates with an
increased association with immunoglobulin heavy-chain
binding protein (BiP). Appl. Environ. Microbiol. 64, 316�/
324.
Sebastiao, M.J., Cabral, J.M.S., Aires-Barros, M.R., 1993.
Fusarium solani pisi recombinant cutinase partitioning in
PEG/potassium phosphate aqueous two-phase systems.
Biotechnol. Technol. 7, 631�/634.
Sebastiao, M.J., Cabral, J.M.S., Aires-Barros, M.R., 1994.
Partitioning of recombinant Fusarium solani pisi cutinase in
polyethylene glycol-aqueous salt solution two-phase sys-
tems. J. Chromatogr. A 668, 139�/144.
Zaslavsky, B.Y., 1995. Aqueous Two-Phase Partitioning.
Physical Chemistry and Bioanalytical Applications. Marcel
Dekker, New York, Basel, Hong Kong.
M.T. Cunha et al. / Journal of Biotechnology 100 (2003) 55�/6464
