Biochemical Engineering Journal 34 (2007) 156–164

Aqueous two phase extraction for purification of C-phycocyanin

Ganapathi Patil, K.S.M.S. Raghavarao ∗
Department of Food Engineering, Central Food Technological Research Institute, Mysore 570020, India
Received 8 May 2006; received in revised form 22 November 2006; accepted 28 November 2006

Abstract
Aqueous two phase extraction is employed for the purification of C-phycocyanin from Spirulina platensis. A systematic approach is suggested
to arrive at the optimal process parameters of aqueous two phase extraction by considering a case study of C-phycocyanin. The influence of various
process parameters such as type of aqueous two phase systems, phase forming salt, molecular weight of the phase forming polymer, system pH,
phase composition, phase volume ratio, and type and concentration of neutral salts on differential partitioning of C-phycocyanin is evaluated.
Desirable conditions for the purification of C-phycocyanin are found in polymer–salt systems especially in polyethylene glycol (4000)/potassium
phosphate system. Increase in purity of C-phycocyanin to 3.52 from initial purity of 1.18 is achieved at pH 6, tie line length of 35.53% with a phase
volume ratio of 0.8 in a single step of aqueous two phase extraction. Multiple extractions resulted in further increase in purity of C-phycocyanin
without loosing the yield and a maximum purity of 4.05 is achieved in third aqueous two phase extraction. The integration of ultrafiltration with
aqueous two phase extraction facilitated the selective removal of polyethylene glycol from the purified C-phycocyanin. Finally, C-phycocyanin is
freeze dried to obtain in powder form.
© 2006 Elsevier B.V. All rights reserved.

Keywords: Aqueous two phase extraction; C-phycocyanin; Purification; Differential partitioning; Ultrafiltration

1. Introduction Major drawback of these protocols is the large number of steps

Phycobiliproteins are the major photosynthetic pigments in
cyanobacteria. Phycobiliproteins are water-soluble proteins,
having covalently attached tetrapyrroles. C-phycocyanin
(C-PC) is the major component of the phycobiliprotein family.
C-PC exhibits a strong red fluorescence when it is present in
native and concentrated form. It is not only used as nutrient
ingredient and natural color for food and cosmetics [1] but also
used as potential therapeutic agent in oxidative stress-induced
diseases [2] and as fluorescent markers in biomedical research
[3]. The purity of C-PC is generally evaluated based on the
absorbance ratio of A620 /A280 . C-PC of purity 0.7 is considered
as food grade, 3.9 as reactive grade and greater than 4.0 as
analytical grade [4]. Various researchers have developed several
methods for the purification of C-PC [5–10]. However, almost
all these methods of purification of C-PC involve number of
steps wherein precipitation, centrifugation, dialysis are used
in initial purification, while ion-exchange chromatography and
gel filtration chromatography are used in final purification.
∗ Corresponding author. Tel.: +91 821 2513910; fax: +91 821 2517233.
E-mail address: fed@cftri.res.in (K.S.M.S. Raghavarao).
involved, and it is known that higher the number of steps higher
is the loss of product yield [11]. Furthermore, the scale-up of
these methods is difficult and also expensive. It may be noted
that 50–90% of the production cost resides in the purification
steps [12]. Hence, there is a need for efficient and economical
large-scale bioseparation methods, which will achieve high
purity as well as high yield, while maintaining the biological
activity of the molecules. One such purification method
that meets all these criteria is aqueous two phase extraction
(ATPE).
ATPE in many cases offers a better alternative to exist-
ing methods, especially in the early processing stages, with
regard to scale of operation, low processing time, enrichment
of the product and continuous operation for the separation and
purification of desired enzymes/proteins from a complex mix-
ture [11,13–16]. In ATPE, selective partitioning of the desired
enzymes or proteins to one phase and contaminant proteins to
the other phase not only purify the enzymes/proteins but also
concentrate them in one of the phases [17,18]. In view of this it
was thought prudent to use ATPE for the purification of C-PC.
The main objective of the present study is to develop a simple
and more efficient downstream process for the purification and
concentration of C-PC. More emphasis is given to arrive at the

1369-703X/$ – see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.bej.2006.11.026
 G. Patil, K.S.M.S. Raghavarao / Biochemical Engineering Journal 34 (2007) 156–164
optimal process parameters of ATPE by considering a case study
of C-PC and is applicable for other biomolecules.
2. Materials and methods
2.1. Chemicals
Polyethylene glycol (PEG) was procured from Sisco
Research Laboratories, Mumbai, India. Potassium phosphate,
sodium phosphate, ammonium sulphate, magnesium sulphate,
sodium sulphate and sodium citrate were procured from Ran-
baxy Chemicals, S.A.S. Nagar, India.
2.2. Preparation of crude extract of phycocyanin
Freshly harvested Spirulina biomass (obtained from Depart-
ment of Plant Cell Biotechnology, CFTRI) was washed with
distilled water for about two to three times to remove the culture
media components. The biomass was homogenized at a pres-
sure range of 200–400 kg/cm2 for about 5 min to break the cells
and centrifuged at 10,000 rpm for about 10 min to separate the
released phycocyanin from cell debris and stored at 4–5 ◦ C and
used for the experiments.
2.3. Aqueous two phase extraction
ATPE was carried out by adding predetermined (based on
phase diagram from literature, [13,19–22]) weighed quantities of
polyethylene glycol and salt to a given quantity of crude extract
of C-PC making the total weight of the system 100% on w/w
basis. The mixture was stirred thoroughly for about an hour to
equilibrate and allowed for phase separation. Concentrations of
C-PC and total protein in both the phases were analyzed for
estimating the purity and yield of C-PC.
2.3.1. Tie
 line length (TLL)
TLL = CT2 + CB2
where ‘CT ’ (w/w, %) is the difference in concentration of
the predominant top phase forming polymer between top and
bottom phases, and ‘CB ’ is defined similarly for the predom-
inant bottom phase forming component. Concentrations of the
phase forming polymer–salts were selected from the literature
[13,19–22].
2.3.2. Free volume (VFV )
1 1
VFV = −
ρ ρo
where ‘ρ’ is density of one of the phases and ‘ρ0 ’ is density of
the respective reference phase solution (blank phase).
2.4. Spectroscopic measurements
Absorption spectra of C-PC were measured using UV–vis
spectrophotometer (Double beam spectrometer, Shimadzu,
model UV-200S Japan). The ratio of A620 to A280 gives the purity
157
of C-PC, wherein A620 is the maximum absorbance of C-PC and
A280 is the absorbance of total proteins.
2.5. Polyacrylamide gel electrophoresis
Sodium dodecyl sulphate-polyacrylamide gel electrophore-
sis (SDS-PAGE) was carried out by following the method as
described in Methods in Enzymology [23], using 30% poly-
acrylamide slab gel. Electrophoresis was run at 50 V, 12.5 mA,
for about 3–4 h. The gel was stained with a Coomassie Brilliant
Blue R250 of 0.05% (w/v), 50% (v/v) methanol and 12% (v/v)
acetic acid. The gel was destained using the same buffer without
Coomassie Brilliant Blue.
2.6. Ultrafiltration
Ultrafiltration units (Amicon Ultra, 15 ml, 10 and 30 kDa)
were procured from Millipore, Bangalore, India and used for
the separation of PEG from C-PC.
3. Results and discussion
In aqueous two phase extraction, the purification of most
of the proteins is mainly due to the differential partitioning of
the target protein to one phase and the contaminant proteins to
the other phase. The basis of partitioning of proteins is com-
plex and depends mainly on size, charge and hydrophobicity
of the biomolecule. Most of these properties change whenever
there is a change in their environment (phase system), which
is dependent on many factors such as: (1) type of aqueous two
phase system; (2) phase forming salt; (3) molecular weight of the
phase forming polymer; (4) pH of the system; (5) phase volume
ratio; (6) type and concentration of neutral salts. Due to the mul-
tiple factors, the selection of suitable phase system is a search
in a multidimensional space requiring a number of experiments
to achieve optimization efficiently. If each of these factors were
analyzed at least for three different values, over 2000 (37 = 2187)
experiments would be necessary [24]. The optimized condition
thus obtained is suitable only for a given biomolecule and does
not hold good for the other biomolecule.
A properly designed systematic approach is required in order
to reduce this large number of experiments. Selection of key
parameters is an important task and requires thorough knowl-
edge of solute to be studied. In view of this, experiments were
planned through a systematic approach as shown in Fig. 1,
wherein all the parameters, that are expected to affect the par-
titioning of proteins and enhance the purification of C-PC, are
considered in a logical sequence. Details of selection of each
of these parameters and their effect on protein partitioning have
been discussed in the following sections.
3.1. Selection of type of aqueous two phase system
In order to know the suitable type of phase system
(polymer–polymer or polymer–salt), both the polymer–polymer
(PEG/maltodextrin) and polymer–salt (PEG/potassium phos-
phate) phase systems were prepared using crude extract of
158 G. Patil, K.S.M.S. Raghavarao / Biochemical Engineering Journal 34 (2007) 156–164
Fig. 1. Systematic approach to arrive at the optimal process parameters ofATPE.
phycocyanin. The results indicated that the protein purification
in polymer–salt system is better (1.5-fold) when compared to
that of the polymer–polymer system.
The polymer–salt phase systems are economical when com-
pared to the polymer–polymer phase systems. Polymer–salt
systems have the additional advantage of lower phase viscos-
ity, thus needing a shorter time for phase separation. This has
favored the selection of PEG/salt systems. After selecting the
type of phase system (polymer–salt), the next most important
step is the selection of type of phase forming salt for partitioning
of the target solute (C-PC).
3.2. Selection of phase forming salt
In order to identify the suitable salt for purification of C-PC,
ATPE experiments were carried out by adding predetermined
weighed quantity of different molecular weight polyethylene
glycol (MW 1500, 4000 and 6000) and different phase form-
ing salts (sodium phosphate, potassium phosphate, magnesium
sulphate, sodium sulphate, ammonium sulphate, and sodium cit-
rate) to a given quantity of crude extract of C-PC making the
total weight of the system 100% on w/w basis. The results are
shown in Table 1. It is observed that, the majority of the target
protein partitioned to top phase, which implies that C-PC exhib-
ited a strong preference to top-phase. Selective partitioning of
the C-PC to the top phase posed problems in the calculation
of its partition coefficient (K = CT /CB , where ‘CT ’ and ‘CB ’
are the concentration of solute in the top and bottom phases,
respectively) in most of the systems studied. As a consequence,
selection of phase forming salt has become difficult. Hence, the
purity of C-PC in top phase was considered as the response
variable to evaluate the effect of different phase forming salts.
Different salts studied for the same molecular weight of PEG at
the same composition have resulted in different pH (Table 1).
Phosphate salts gave the pH of 6.7–7.3, sulphate salts from 4.3
to 5.0 and citrate salt had shown a pH of 8.0. Such change in
the phase system environment with respect to different salts had
resulted in a change in partitioning of the protein. In most of the
PEG/sulphate phase systems, C-PC precipitated to the interface.
In case of PEG/citrate system, fluorescence of C-PC disappeared
indicating the denaturation of the protein. Since phycocyanin is
quite stable in a pH range of 5–8, it was thought prudent to use
phosphate salts for the purification of C-PC. From Table 1, it can
be observed that phosphate salts have shown maximum purity
of C-PC for all the molecular weights of PEG when compared to
the other salts. This is mainly due to the favorable biocompatible
phase environment by the phosphate salts. Further, among phos-
phate salts, potassium phosphate showed better results in terms
of purity and overall yield when compared to that of sodium
phosphate. In view of this PEG/potassium phosphate system
was chosen for further experiments.
3.3. Selection of polymer molecular weight
In order to make selection of suitable molecular weight of
phase forming polymer for purification of C-PC, ATPE was per-
formed with different molecular weights of PEG (MW 1500,
4000, 6000 and 20,000). Other parameters such as phase forming
salt (potassium phosphate), phase composition, phase volume
ratio, temperature and pH of system were kept constant and the
results are shown in Table 2. It can be observed that the parti-
tion coefficient of the total protein decreased with an increase
in molecular weight of the PEG. Similar results were observed
for most of the other proteins [22,25,26] and could be attributed
to two different effects. The most obvious is the increase in the
top phase hydrophobicity. Although the composition of PEG
and salt added is same for different molecular weight of the
PEG (MW 1500, 4000, 6000 and 20,000), the amount of PEG
to be added to the system reduces as the molecular weights of
polymer increases. In fact as PEG chain length increases (with
increasing molecular weight) there will be less hydroxyl groups
for the same concentration of the polymer and so the hydropho-
bicity of the polymer-rich (top) increases. On the other hand, the
increase in the chain length will also cause the reduction of the
free volume, meaning less space available for the protein. The
increase in molecular weight of PEG had facilitated the parti-
tioning of the total proteins to the bottom phase. A maximum
partition coefficient of total protein of 2.53 was observed in case
of lower molecular weight of PEG (1500), correspondingly in C-
PC purity of 2.58. A slight increase in purity of C-PC (2.75) was
obtained in MW of PEG 4000. Further increase in MW of PEG
(above 4000) resulted in precipitation of C-PC to the interface.
Hence the PEG 4000 was chosen for subsequent experiments.
 G. Patil, K.S.M.S. Raghavarao / Biochemical Engineering Journal 34 (2007) 156–164 159

Table 1
Effect of different salts on partitioning of C-PC
Salts PEG (MW) Purity of C-PCa Yield of C-PC (%) K of C-PC pH of phase systema Visual observation

1500 2.50 ± 0.02 96.07

Sodium phosphate 4000 2.61 ± 0.01 85.17 CND 6.7 No precipitation
6000 2.54 ± 0.04 55.55
1500 2.58 ± 0.06 96.59 21.70
Potassium phosphate 4000 2.75 ± 0.04 90.57 10.47 7.3 No precipitation
6000 2.57 ± 0.07 87.40 281.56
1500 2.33 ± 0.02 84.31
Magnesium sulphate 4000 2.45 ± 0.05 73.55 CND 4.3 Precipitation
6000 2.29 ± 0.04 74.55
1500 1.98 ± 0.05 84.46
Sodium sulphate 4000 2.33 ± 0.03 68.53 CND 5.0 Precipitation
6000 1.89 ± 0.08 57.37
1500 2.05 ± 0.07 96.16 –
Ammonium sulphate 4000 2.55 ± 0.03 98.53 58.92 4.6 Precipitation
6000 2.23 ± 0.05 75.73 101.52
1500 2.52 ± 0.05 98.85
Fluorescence
Sodium citrate 4000 2.56 ± 0.04 92.19 CND 8.0
disappeared
6000 2.47 ± 0.06 81.84

Initial purity of C-PC is 1.18;CND:could not be determined due to near complete partitioning of C-PC to the top phases.
a Values represents mean ± S.D.

Table 2
Effect of polymer molecular weight on partitioning of C-PC (at 25 ± 2 ◦ C)
PEG MW Phase composition (w/w, %) Purity of C-PC K of total protein Yield of C-PC (%) Volume ratio

PEG Salta

1,500 13 14 2.58 ± 0.02 2.53 ± 0.10 96.59 1.03

4,000 13 14 2.75 ± 0.04 2.30 ± 0.31 90.58 0.98
6,000 13 14 2.57 ± 0.06 2.15 ± 0.15 87.40 0.95
20,000 13 14 1.89 ± 0.03 1.93 ± 0.18 78.23 0.96

Values represent mean ± S.D.

a Potassium phosphate, Initial purity of C-PC is 1.18.

After selecting the type of salt and molecular weight of PEG, From Fig. 2, it can be seen that, the maximum purity of C-PC
the other process parameters like pH, phase volume ratio, phase is observed at pH 6 for all tie lines. This is mainly due to the
composition and effect of neutral salts on partitioning of C-PC partitioning of most of the contaminant proteins to the bottom
were studied. phase at pH 6 for all tie lines. This is confirmed by plotting pH

3.4. Effect of pH on protein partitioning

In order to know the influence of pH on C-PC partitioning,

experiments were performed at different pH values for different
tie lines of PEG 4000/potassium phosphate system. As C-PC
is not stable below pH 5 and above pH 8, experiments were
restricted to the pH range of 5–8 for each tie line and the results
are shown in Fig. 2. Precipitation of C-PC to the interface was
observed at pH 5 for all tie lines. The purity of C-PC at pH 5 (for
all tie lines) is low when compared to that at other pH values.
Decrease in pH values will shift critical points of binodal of
aqueous two phase systems to higher values. Such phenomena
have been attributed to the change in properties of water [27].
Such change in properties of water structure associated with
PEG and phosphate ions had resulted in a low purity of C-PC at
lower pH. Fig. 2. Effect of pH on purity of C-PC (PEG 4000/potassium phosphate).
160 G. Patil, K.S.M.S. Raghavarao / Biochemical Engineering Journal 34 (2007) 156–164

Fig. 3. Effect of pH on partitioning of total proteins (PEG 4000/potassium

phosphate).

versus total protein (Fig. 3), wherein the partition coefficient of

total protein at pH 6 is lowest. Hence, increase in purity of the
C-PC at pH 6 was observed. As the pH of the system changes,
individual proteins show partition behavior, which is broadly
related to their net (surface positive or negative) charge. Under Fig. 4. Effect of pH on free volume: (a) top phase and (b) bottom phase.
these conditions, the partition coefficient of a protein can be
divided into two components [28]. One component depends in bottom phase is more significant than that of top phase. This
only on the net charge of the protein; the other depends on has resulted in partitioning of most of the contaminant proteins
the surface properties other than the charge. At pH 6, the to the bottom phase, which in turn increased the purity of the
partitioning of C-PC is mainly due to the surface properties C-PC at pH 6. Further increase in pH (from 6 to 7) of the phase
rather than net charge. This is because the isoelectric point of system showed no change in free volume in bottom phase
C-PC is 5.8, so at pH 6 either the charge of the protein is almost whereas the free volume in top phase increased. This increase
neutralised or very less. On the other hand, many amino acids in free volume in top phase resulted in a decrease in purity of
are essentially neutral in pH solutions of range 5.5–6 [29]. For C-PC at pH 7 compared to that at pH 6. This is mainly due to
this reason, it can be inferred that any effect due to charge will the increased partitioning of the contaminant proteins into the
be negligible in comparison to the effect due to other surface top phase. C-PC purity decreased with further increase of pH
properties. from 7 to 8. From Fig. 5 it can be observed that the yield of
Above isoelectric point (at pH 7 and 8), C-PC is negatively C-PC is maximum at pH 6 in the range of 80–97% for all tie
charged and PEG behaves as positively charged and thereby lines.
polyanions of protein are attracted by PEG-rich phase [27,28]. From all these analysis it was decided to carry out further
As a result C-PC gets partitioned to the top phase and contam- experiment at pH 6 as an optimum pH for the purification of
inant proteins to the opposite phase at pH 7 and 8. This in turn C-PC.
resulted in a slight decrease in purity of C-PC compared to that
at pH 6 and much higher than that at pH 5. It may be noted that
the purity observed at pH 6 is highest.
The effect of pH on differential partitioning of C-PC and
the contaminating proteins were confirmed by the estimation
of the free volume of both the phases as shown in Fig. 4. The
free volume in top phase increased with an increase in pH, for a
given tie line length and this trend (extent of increase) remained
almost constant for all tie lines (Fig. 4a), where as the free vol-
ume in bottom phase increased initially (pH 5–6) and remained
almost constant with further increase in the pH for all the tie line
lengths (Fig. 4b). From both these figures, it can be observed
that at pH 5, the available free volume is less in top as well as in
bottom phase; hence, precipitation of C-PC occurred at pH 5.
An increase in free volume in both the phase is observed with
an increase in pH from 5 to 6, but the increase in free volume Fig. 5. Effect of pH on yield of C-PC.
 G. Patil, K.S.M.S. Raghavarao / Biochemical Engineering Journal 34 (2007) 156–164
Table 3
Effect of TLL on protein partitioning (PEG 4000/potassium phosphate, at pH 6)
TLL (%) K of total proteina
13.33 0.60 ± 0.12
21.42 0.80 ± 0.24
28.20 0.67 ± 0.18
33.53 0.88 ± 0.25
39.70 0.97 ± 0.14
a Values represents mean ± S.D.
3.5. Effect of tie line lengths (TLL) on protein partitioning
Although the experiments were carried out for all the
TLL of PEG 4000/potassium phosphate at different pH val-
ues (Figs. 2 and 3), the explanation of effect of different TLL
on protein partitioning is restricted to the PEG 4000/potassium
phosphate system of pH 6 since the purity of C-PC obtained was
maximum at this pH.
From Table 3, it can be observed that increase in %TLL
increased the partition coefficient of total proteins as well as
C-PC. However, the increase in partition coefficients of C-PC
is higher when compared to that of total proteins. Changes in
partition coefficient of these proteins can be attributed to the
change in free volume of the phase system. From Fig. 6, it can
be observed that the free volume in the bottom phase decreased
when the TLL is increased. As a result, the C-PC and other con-
taminant proteins partitioned more and more to the top phase
with increasing TLL. Increase of contaminant proteins in the
top phase with increasing TLL is supposed to decrease the purity
of C-PC. However, in present study, the purity of C-PC was not
affected by the partitioning of the contaminant proteins to the top
phase. This is mainly due to the migration of C-PC from bottom
phase to the top phase was higher when compared to that of the
contaminant proteins (Table 3). As a result the concentration of
C-PC in the top phase increased with an increasing TLL, whereas
the purity of C-PC remained almost constant (purity ≈ 3.0).
Increase in yield of C-PC was observed with an increase in
%TLL; a maximum yield of 97.47% was obtained at a TLL of
Fig. 6. Change in free volume at different TLL (PEG 4000/potassium phosphate,
pH 6).
161
K of C-PC Yield of C-PC (%) Purity of C-PC
15.23 90.42 2.95
24.12 92.44 2.72
27.55 94.03 3.06
780.29 97.47 3.06
– 81.61 3.06
33.53% (Table 3). Further increase in TLL (39.70%) had resulted
in precipitation of proteins to the interface and decrease in yield
of C-PC. This could be mainly due to the increase in concen-
tration of salt in bottom phase at higher TLL, which effectively
salts out proteins as their solubility limits are reached. Increased
partitioning of proteins to the top phase occurs, provided that
the solubility limit is not exceeded in the top phase. The precip-
itation (at the interface) observed at TLL 39.70%, indicates that
the top phase has reached the solubility limit. Based on all these
results, it was decided to study the further experiments with the
system of PEG 4000/potassium phosphate at pH of 6 and TLL
of 33.53%, which had resulted purity of C-PC of 3.05 with a
yield of 97.47%.
3.6. Effect of volume ratio on protein partitioning
The primary need of any purification process is to achieve
high purity, concentration and yield of the target molecule. In
ATPE this can easily be achieved by changing phase volume
ratio (VT /VB , where, VT and VB are the volumes of top and
bottom phases, respectively) along a given TLL, wherein the
target protein can be concentrated in one of the phases while
the contaminant proteins in the other phase. This also results in
simultaneous purification of the biomolecule. To test the feasi-
bility of this approach, the volume ratio was varied from 0.3 to
4.4 at TLL of 33.53% and the results are shown in Figs. 7 and 8.
It can be observed that, as the volume ratio increases the yield of
C-PC and partition coefficient of total protein (KTP) increased
whereas the concentration of C-PC decreased. This is due to
increase in the volume of the top phase. Higher concentration
Fig. 7. Effect of volume ratio on protein partitioning (PEG 4000/potassium
phosphate, pH 6, TLL of 33.53%).
162 G. Patil, K.S.M.S. Raghavarao / Biochemical Engineering Journal 34 (2007) 156–164
Fig. 8. Effect of volume ratio on purity and concentration of C-PC (PEG
4000/potassium phosphate, pH 6, TLL of 33.53%).
of C-PC (1.62 mg/ml) with a lower yield (63.75%) and a purity
of 3.02 was obtained at volume ratio of 0.3. Increase in volume
ratio from 0.3 to 0.8 had resulted in an increase in purity (3.52)
and a yield of C-PC (85.68%). The increase in purity and yield
might be due to the migration of C-PC from bottom phase to top
phase. The increase in partition coefficient of the total proteins
at this volume ratio is marginal indicating that small amount
of the contaminant proteins have partitioned to the top phase,
which has not affected the purity of C-PC. Maximum yield of
97.4% with a purity of 3.04 was observed at a volume ratio of
1.1. Further increase in volume ratio showed decrease in purity
as well concentration of C-PC. The decrease in purity can be
attributed to the partitioning of contaminant proteins to the top
phase (Fig. 7). At a volume ratio of 1.7 and above, the yield of
C-PC remained almost constant. In view of this, it was decided
to carryout the further experiments at a volume ratio of 0.8, since
the obtained purity of the C-PC was high (3.52).
3.7. Effect of neutral salts
In order to know the effect of neutral salt on partitioning of
C-PC and the contaminant proteins, ATPE was carried out with
different neutral salts such as LiCl, NaCl, KCl, Li2 SO4 , Na2 SO4
and K2 SO4 in the concentration range of 0 to 1% on w/w basis.
The results are shown in Fig. 9. It can be observed that, the
partition coefficient of total proteins and C-PC decreased with
an increase in concentration of salts in case of LiCl, NaCl and
KCl, whereas the reverse is true in case of Li2 SO4 , Na2 SO4
and K2 SO4 . This is because the partition coefficient of nega-
tively charged proteins decreases in the presence of chloride
salts and increases in presence of sulphates salts [13,14]. This
indicates that at pH 6 not only C-PC and also the other con-
taminant proteins behave as negatively charged proteins. It was
observed from the literature that the effect of the cation on the
partition coefficient of negatively charged proteins is in the order
of Li > Na > K [13,14,24]. However, in the present study the
negatively charged proteins have showed decreasing partition
coefficients in the order of K > Na > Li with change in cationic
series. This change in order of cation series is mainly due to
the presence of another salt (potassium phosphate) in both the
phases as one of the phase forming component in PEG/salt sys-
Fig. 9. Effect of neutral salts on partitioning of proteins (PEG 4000/potassium
phosphate, pH 6, TLL of 33.53%, VR of 0.8).
tem, whereas the literature trend represents the effects of cation
series on partition coefficient of proteins in PEG/dextran system
(wherein both the polymers are uncharged). From our results, it
can be said that, the uneven distribution of neutral salts in top and
bottom phases of PEG/salt system (aqueous two phase system
made with PEG/potassium phosphate) have different affinity of
the salt for the bottom and top phases, which creates two dis-
tinct ionic atmospheres in both phases. This might have been
changed in presence of cation series of Li, Na, K and resulted
in a different trend, however, a detailed study is needed in this
regard.
It has been reported that the addition of neutral salts either
changes the phase composition of PEG/potassium phosphate
or alters the properties of the partitioning solute [17,27,30,31].
The volume ratio in the present study has remained practically
constant for all the neutral salt studied at different concentration,
indicating that the composition of the PEG/potassium phosphate
is not influenced much by the change of neutral salts and their
concentration. Hence, it can be inferred that the change in prop-
erties of the C-PC is responsible for the observed results. The
yield of C-PC decreased with an increase in the concentration
of chloride salts and increased in case of sulphate salts (Fig. 10).
The purity of C-PC remained almost constant for all the neutral
salt studied.
3.8. Influence of multiple extractions on purity of C-PC
In order to explore the possibility of further increase in purity
of C-PC, multiple ATPE were carried out. First extraction was
 G. Patil, K.S.M.S. Raghavarao / Biochemical Engineering Journal 34 (2007) 156–164
Fig. 10. Effect of neutral salts on yield of C-PC.
carried out in PEG 4000/potassium phosphate system with opti-
mum process parameters of 33.53% TLL, volume ratio 0.8, pH
6, which resulted in maximum purity of C-PC (3.52). Second
extraction was carried out by mixing the top phase of the first
ATPE with the fresh bottom phase of same composition. The
aliquots of the separated phases after the second extraction were
analyzed for the purity of C-PC. It was found that the purity of
C-PC increased from 3.52 to 3.82. The repetition of the same pro-
cedure for third extraction resulted in a purity of 4.05 (Fig. 11a).
The increase in purity of C-PC in each ATPE is due to the parti-
tioning of the contaminant proteins from top phase to the bottom
phase. Further extraction (fourth extraction) did not show any
increase in the purity of C-PC (Fig. 11b). The overall yield of the
C-PC obtained during ATPE remained almost constant at 85%.
Fig. 11. Influence of multiple extractions on purity of C-PC: (a) spectra of C-PC
(after third extraction) and (b) purity of C-PC in each extraction.
163
3.9. Integration of ATPE with membrane process
(ultrafiltration)
In ATPE, the targeted protein (C-PC) partitions to the top
phase and unwanted proteins to the other phase, hence, the
purity of the C-PC increases. In order to make C-PC suitable for
its targeted application in pharmaceutical, cosmetics and food
application, it is necessary to remove PEG. This was achieved
by ultrafiltration using centrifugal ultrafiltration units (30 kDa
membrane cut-off, at 10,000 rpm for about 10 min).
Integration of ultrafiltration with the ATPE has facilitated the
separation of PEG without any damage to the integrity of C-PC;
this has increased the concentration of C-PC whereas the purity
of the C-PC remained constant. Finally, C-PC was freeze dried
to obtain in powder form.
3.10. SDS PAGE
The purity of the C-PC obtained from ATPE was confirmed
using SDS-PAGE as show in Fig. 12. Lane 1 indicates the
molecular marker, while lane 2 the crude extract of C-PC and
lane 3 is the C-PC after ATPE. From the SDS–PAGE, it can be
observed that, the majority of the contaminant proteins present
in the crude extract were partitioned to the bottom phase during
ATPE, hence, increase in purity of C-PC is observed (top
phase). However, other bands are visible in lane 3 apart from
C-PC (Fig. 12) indicating the presence of other proteins. Further
purification of the C-PC is needed in order to remove the other
contaminant proteins, which could be achieved by integrating
the ATPE with chromatography technique like ion-exchange
chromatography [32].
Fig. 12. SDS PAGE of C-phycocyanin: lane 1: molecular marker, lane 2: crude
extract of C-PC and lane 3: C-PC after ATPE.
164 G. Patil, K.S.M.S. Raghavarao / Biochemical Engineering Journal 34 (2007) 156–164
4. Conclusions
During this study, a systematic approach was made to develop
a process for the purification of C-PC. The process parameters
responsible for the purification of C-PC and their optimization
were discussed in detail. Polymer–salt system showed higher
purification of C-PC when compared to polymer–polymer
system. Optimized conditions of the PEG 4000/potassium phos-
phate system of tie line length of 35.53% with a phase volume
ratio of 0.8 and pH 6 showed a C-phycocyanin purity of 3.52
in a single ATPE. Multiple extractions of ATPE resulted in the
C-PC purity of 4.05. The overall yield of the C-PC obtained
during extraction was 85%. Integration of ATPE with ultrafil-
tration has facilitated the removal of polymer from the purified
C-PC, which was freeze dried to obtain in powder form. Overall
results obtained here demonstrated the feasibility of ATPE for
the purification of C-PC.
Acknowledgements
The authors thank Dr. V. Prakash, Director, CFTRI, for
the encouragement and keen interest in separation processes.
Thanks are due also to Dr. G.A. Ravishankar, Head, Plant
Cell Biotechnology, CFTRI, Mysore for providing the Spir-
ulina biomass. The help by Chethana S. and Narayan A.V. in
harvesting the Spirulina is acknowledged. Authors acknowl-
edge the Department of Biotechnology, Government of India
for project grant (G-1726/04-05). Ganapathi Patil acknowledges
CSIR, Government of India for the Senior Research Fellowship.
References
[1] A. Yoshida, Y. Takagaki, T. Nishimune, Enzyme immunoassay for phy-
cocyanin as the main component of Spirulina color in foods, Biosci.
Biotechnol. Biochem. 60 (1996) 57–60.
[2] V.D. Bhat, K.M. Madyastha, Scavenging of peroxynitrite by Phycocyanin
and phycocyanobilin from Spirulina platensis: protection against oxidative
damage to DNA, Biochem. Biophys. Res. Commun. 285 (2001) 262–266.
[3] A.N. Glazer, Phycobiliproteins—a family of valuable widely used fluo-
rophores, J. Appl. Phycol. 6 (1994) 105–112.
[4] M. Rito-Palomares, L. Nunez, D. Amador, Practical application of aque-
ous two-phase systems for the development of a prototype process for
C-Phycocyanin recovery from Spirulina maxima, J. Chem. Tech. Biotech.
76 (2001) 1273–1280.
[5] S. Boussiba, A.E. Richmond, Isolation and characterization of phyco-
cyanins from the blue-green alga Spirulina platensis, Arch. Microbiol. 120
(1979) 155–159.
[6] A. Herrera, S. Boussiva, V. Napoleone, A. Holberg, Recovery of c-
Phycocyanin from cyanobacterium Spirulina maxima, J. Appl. Phycol. 1
(1989) 325–331.
[7] L.F. Du, H.L. Fu, Purification and properties of phycobiliprotein from
Spirulina platensis, J. Sichuan. Univ. 31 (1994) 576–578.
[8] G. Manoj, K.S.M.S. Raghavarao, S.G. Jayaprakashan, G.A. Ravishankar, T.
Ajithkumar, L.V. Venkataraman. An improved process for the preparation
of a natural blue colorant Phycocyanin from Spirulina plantensis, Indian
patent, 2504/DEL/96, 1996.
[9] Y.M. Zhang, F. Chen, A simple method for the efficient separation and
purification of C-Phycocyanin and allophycocyanin from Spirulina platen-
sis, Biotech. Technol. 13 (1999) 601–603.
[10] P. Anamika, M. Sandhya, P. Richa, P.K. Gosh, Purification and charac-
terization of c-Phycocyanin from cyanobacterial species of marine and
freshwater habitat, Prot. Exp. Purif. 40 (2) (2005) 248–255.
[11] M.R. Kula, K.H. Kroner, H. Hustedt, Purification of enzymes by
liquid–liquid extraction, Adv. Biochem. Eng. 24 (1982) 73–118.
[12] A.D. Diamond, J.T. Hsu, Aqueous two-phase systems for biomolecule
separation, Adv. Biochem. Eng./Biotechnol. 47 (1992) 89–135.
[13] P.A. Albertsson, Partition of Cell Particles and Macromolecules, 2nd ed.,
Wiley, New York, 1986.
[14] H. Walter, in: T. Gerritsen (Ed.), Modern Separation Methods of Macro-
molecules and Particles, Wiley-Interscience, New York, NY, 1969.
[15] K.S.M.S. Raghavarao, N.K. Rastogi, M.K. Gowthaman, N.G. Karanth,
Aqueous two phase extraction for downstream processing of enzymes/
proteins, Adv. Appl. Microbiol. 41 (1995) 97–171.
[16] S. Chethana, N. Naveen, G.A. Ravishankar, K.S.M.S Raghavarao, An
Improved Process for the Concentration and Purification of Phycocyanin,
Indian Patent 390/DEL/03 (2003).
[17] N.D. Srinivas, R.S. Barhate, K.S.M.S. Raghavarao, Aqueous two-phase
extraction in combination with ultrafiltration for downstream processing
of Ipomoea peroxidase, J. Food Eng. 54 (2002) 1–6.
[18] K.S.M.S. Raghavarao, M.R. Guinn, P. Todd, Recent developments in aque-
ous twophase extraction in bioprocessing, Sep. Purif. Meth. 27 (1998)
1–50.
[19] B.Y. Zaslavasky, Aqueous Two Phase Partitioning, Physical Chem-
istry and Bioanalytical Applications, Marcel Dekkar Inc., New York,
1995.
[20] A. Salbat, The influence of salts on the phase composition in aqueous
two phase systems: experiments and predictions, Fluid. Phase. Equilibr.
187/188 (2001) 489–498.
[21] M.T. Zafarani-Moattar, R. Sadegi, Liquid–liquid equilibra of aqueous two
phase systems containing polyethylene glycol and sodium dihydrogen
phosphate or disodium hydrogen phosphate experiment and correlation,
Fluid Phase Equilibr. 181 (2001) 95–112.
[22] J.C. Marcos, L.P. Fonseca, M.T. Ramalho, J.M.S. Carbral, Partial purifi-
cation of pencillin acylase from Escherichia coli in poly (ethylene glycol)
– sodium citrate aqueous two phase system, J. Chromatogr. B. 734 (1999)
15–22.
[23] M. Deutscher, Methods in enzymology guide to protein purification; purifi-
cation procedures, Electrophoret. Meth. 182 (1990) 425–488.
[24] Hatti-Kaul. Rajni, Aqueous Two-Phase Systems Methods and Protocols,
Humana Press Inc., Totowa, New Jerseym, 1999.
[25] A.S. Schmidt, A.M. Ventom, J.A. Asenjo, Partitioning and purification of
␣-amylase in aqueous two-phase systems, Enzym. Microb. Technol. 16
(1994) 131–142.
[26] M.V. Miranda, O. Cascone, Partition behaviour of horse radish peroxi-
dase isoenzymes in aqueous two phase poly (ethylene glycol) phosphate
systems, Biotechnol. Tech. 84 (1994) 275–280.
[27] J.G. Huddleston, K.W. Ottomar, D.M. Ngonyani, A. Lyddiatt, Influence
of system and molecular parameters upon fractionation of intracellular
proteins from Saccharomyces by aqueous two-phase partition, Enzyme
Microb. Technol. 13 (1991) 24–32.
[28] S. Tanuja, N.D. Srinivas, K.S.M.S. Raghavarao, M.K. Gowthaman, Aque-
ous two-phase extraction for downstream processing of amyloglucosidase,
Process Biochem. 32 (1997) 635–641.
[29] J.A. Asenjo, A.S. Schmidt, F. Hachem, B.A. Andrews, Model for predict-
ing the partition behaviour of proteins in aqueous two phase systems, J.
Chromatogr. A 668 (1994) 47–54.
[30] J.P. Chen, Partitioning and separation of ␣-lactalbumin and ␤-lactoglobulin
in PEG/potassium phosphate aqueous two-phase systems, J. Fert. Bioeng.
73 (2) (1992) 140–147.
[31] N.L. Abbot, T.A. Hatton, Liquid–liquid extraction for protein separations,
Chem. Eng. Prog. 84 (1988) 31–41.
[32] P. Ganapathi, S. Chethana, A.S. Sridevi, K.S.M.S. Raghavarao, Method
to obtain C-phycocyanin of high purity, J. Chromatogr. A 1127 (2006)
76–81.