Practical application of aqueous two-phase ... - Wiley Online Library

Journal of Chemical Technology and Biotechnology

J Chem Technol Biotechnol 76:1273±1280 (online: 2001) DOI: 10.1002/jctb.507

Practical application of aqueous two-phase systems for the development of a prototype process for c-phycocyanin recovery from Spirulina maxima Marco Rito-Palomares,* Lilia Nun˜ez and Daniel Amador Centro de Biotecnologı´a, Instituto Tecnolo´gico y de Estudios Superiores de Monterrey (ITESM), Ave Eugenio Garza Sada 2501-Sur, Monterrey, NL 64849, Me´xico

Abstract: A novel process for the recovery of c-phycocyanin from Spirulina maxima exploiting aqueous two-phase systems (ATPS), ultra®ltration and precipitation was developed in order to reduce the number of unit operations and bene®t from an increased yield of the protein product. The evaluation of system parameters such as PEG molecular mass, concentration of PEG as well as salt, system pH and volume ratio was carried out to determine under which conditions the c-phycocyanin and contaminants concentrate to opposite phases. PEG1450±phosphate ATPS proved to be suitable for the recovery of c-phycocyanin because the target protein concentrated in the top phase whilst the cell debris concentrated in the bottom phase. A two-stage ATPS process with a phase volume ratio (Vr) equal to 0.3, PEG1450 7% (w/w), phosphate 20% (w/w) and system pH of 6.5 allowed c-phycocyanin recovery with a purity of 2.4 (estimated as the relationship of the 620 nm to 280 nm absorbances). The use of ultra®ltration (with a 30 kDa membrane cut-off) and precipitation (with ammonium sulfate) resulted in a recovery process that produced a protein purity of 3.8 0.1 and an overall product yield of 29.5% (w/w). The results reported here demonstrated the practical implementation of ATPS for the design of a prototype recovery process as a ®rst step for the commercial puri®cation of c-phycocyanin produced by Spirulina maxima. # 2001 Society of Chemical Industry

Keywords: aqueous two-phase systems; c-phycocyanin; Spirulina maxima; protein recovery

1 INTRODUCTION

Colouring compounds used in the food, cosmetic, detergent and molecular genetics industries are products of great commercial signi®cance.1,2 The potential production of these substances by microorganisms opens a very interesting opportunity for biotechnological processes. In this context, the production of c-phycocyanin (a blue-coloured protein) by Spirulina maxima represents a very interesting case because both the industrial application and commercial value of this product are considerable.1 The commercial value of food grade c-phycocyanin (purity of 0.7, de®ned as the relationship of 620 nm to 280 nm absorbances) is around $0.13 USD per mg, whilst that of reactive grade c-phycocyanin (purity of 3.9) varies from $1 to 5 USD per mg.3 In contrast, the commercial value of analytical grade c-phycocyanin (purity greater than 4.0) can be as high as $15 USD per mg.4 c-Phycocyanin is one of the two main biliproteins obtained from the photosynthetic systems of Spirulina

maxima. It is formed by two sub-units, a and b, of 20.5 and 23.5 kDa molecular weight, respectively5 and its isoelectric point has been reported6 to be around 5.8. The recovery of c-phycocyanin from Spirulina maxima has been attempted previously. Herrera et al 3 reported a protocol that involves stages of harvesting, drying, milling, extraction with salts (NaNO3 and CaCl2), adsorption, ultra®ltration, precipitation, dialysis, gel ®ltration and ion exchange chromatography. This procedure resulted in c-phycocyanin of two types: food grade (purity of 0.74) and reactive grade (purity of 3.9). However, this protocol has disadvantages, primarily the negative effect of the excessive number of unit operations (ie ten unit operations) upon product yield. Furthermore, the scale-up of the procedure described by Herrera et al 3 raises complications associated with the operations used, for example the use of milling by hand and the use of chromatography. In the latest case, it has been reported7 that the scaling up of chromatography processes is limited

* Correspondence to: Marco Rito-Palomares, Centro de Biotecnologı´a, Instituto Tecnolo´gico y de Estudios Superiores de Monterrey (ITESM), Ave Eugenio Garza Sada 2501-Sur, Monterrey, NL 64849, Mexico E-mail: [email protected] Contract/grant sponsor: International Foundation for Science; contract/grant number: E2620-2 Contract/grant sponsor: ‘Romulo Garza’ Foundation - Mexico (Received 2 February 2001; revised version received 22 June 2001; accepted 22 July 2001)

# 2001 Society of Chemical Industry. J Chem Technol Biotechnol 0268±2575/2001/$30.00

1273

M Rito-Palomares, L NunÄez, D Amador

by the cost of resin and the size of the process. Consequently, process problems associated with this protocol have limited its commercial application. Aqueous two-phase systems (ATPS) have been suggested as an attractive alternative for the recovery of c-phycocyanin produced by Spirulina maxima. This technique has several advantages including bio-compatibility, ease of scale-up, low cost, etc.8 The use of ATPS for the recovery of protein products from fermentation broth has been addressed before.9±13 However, no reports on the recovery of c-phycocyanin from Spriulina maxima cultures using ATPS have been published. The research presented here aims to generate knowledge on the partition behaviour of c-phycocyanin on ATPS to bene®t the production of such colouring compounds. A practical approach which exploits the known effect of systems parameters such as the concentration of polyethylene glycol (PEG) and salts (ie phosphate and sulfate), phase volume ratio (Vr), molecular weight of PEG and system pH upon protein partition was used to design a two-stage ATPS process for the fractionation of a cell homogenate from Spirulina maxima. Ultra®ltration and precipitation were also exploited for the development of a downstream process for the recovery of c-phycocyanin produced by Spirulina maxima with potential commercial application.

2 MATERIALS AND METHODS 2.1 Characterization of aqueous two-phase systems

The binodal curves were estimated by the cloud point method14 using poly(ethylene glycol) (PEG, Sigma Chemicals, St Louis, MO, USA) of nominal molecular mass of 1000 and 1450 Da (50% w/w stock solution) and sodium sulfate or di-potassium hydrogen orthophosphate/potassium di-hydrogen orthophosphate (Sigma) (30% w/w). Fine adjustment of pH was made by addition of orthophosphoric acid or sodium hydroxide. 2.2 Culture medium and cultivation conditions

Spirulina maxima was cultivated in the culture medium described by Herrera et al. 3 The algae were grown in a batch culture (500 cm3 Erlenmeyer ¯asks) at 20 °C under natural light conditions, agitation was provided using a reciprocal shaker (Lab-Line at 80 cycles min 1). The cells were allowed to grow for 12 days and were harvested by centrifugation at 12 000 rpm for 10 min (Eppendorf 5415C). After cell harvesting, cellular fragmentation was performed manually in a ceramic pot using glass beads and 0.1 mol dm 3 CaCl2 solution (10 cm3 g 1 wet biomass) for 10 min. Temperature was controlled with a dry ice bath. Complete cellular fragmentation was veri®ed using an optical microscope (Olympus CK2). Cell debris removal was achieved by centrifugation at 12 000 rpm for 10 min (Eppendorf 5415C) and the supernatant (referred to 1274

as crude extract) was introduced into the aqueous twophase system previously selected, as described below. 2.3 Influence of system parameters upon partition behaviour of c-phycocyanin in PEG–salt systems

All experimental systems used to establish the operating conditions for the ATPS process were prepared for convenience on a ®xed mass basis. Predetermined quantities of solid PEG, sodium sulfate or potassium phosphate were mixed with either a single model system (containing puri®ed phycocyanin obtained as reported before3) or a complex model system (containing 10% wet w/v cell homogenate from Spirulina maxima fermentation; referred to above as crude extract) to give a ®nal weight of 15 g. Solid components (PEG or salts) were dissolved and phases dispersed by gentle mixing for 30 min at 25 °C. Adjustment of pH was made by addition of orthophosphoric acid or sodium hydroxide. Complete phase separation was achieved by low speed batch centrifugation at 1500 g for 20 min at 25 °C. Visual estimates of the volumes of the top and bottom phases and solids were made in graduated centrifuge tubes. The volumes of the phases were then used to estimate the volume ratio (volume of the top phase/volume of the bottom phase, Vr ). Samples were carefully extracted from the phases and diluted for biochemical analysis and subsequent estimation of the phycocyanin partition coef®cient (K = concentration of solute in the top phase/concentration of solute in the bottom phase). The systems tie-line length (TLL), which represents the length of the line that connects the composition of the top and bottom phase of a de®ned ATPS, was calculated as described by Albertsson.8 Results reported are the average of three independent experiments and errors were estimated to be a maximum of 10% of the mean value. 2.4 Design of a process for c-phycocyanin recovery from Spirulina maxima

For the design of the prototype process to recover c-phycocyanin from Spirulina maxima the use of subsequent ATPS stages after the ®rst ATPS extraction was exploited. The subsequent ATPS extraction stages were implemented in which the top PEG-rich phase from the previous extraction was transferred to a new ATPS. The operating conditions (eg PEG and phosphate concentration, system pH, Vr ) of the subsequent process were kept constant and similar to those de®ned for the ®rst extraction. During the design of the prototype process, ultra®ltration experiments were performed to separate the target protein from the PEG of the top phase. For these experiments bench scale (50 cm3 capacity) ultra®ltration equipment (Amicon, stirred ultra®ltration cell 8050) assembled with a 30 000 Da polymeric type membrane was used. Operating conditions (ie pressure of 40 psi at 15 °C) based upon previous experience15 were used for all the experiments. As a ®nal stage of the process, precipitation by the addition of solid ammonium sulfate was J Chem Technol Biotechnol 76:1273±1280 (online: 2001)

Aqueous two-phase process for recovery of c-phycocyanin

used to treat the concentrate from the ultra®ltration unit. Following the protocol described by Herrera et al,3 the precipitate was divided into two fractions corresponding to 40% and 45% saturation at 0 °C, separated by centrifugation at 12 000 rpm for 10 min (Eppendorf 5415C) and resuspended in phosphate buffer (0.4% w/w, pH 7.0). 2.5 Analytical procedures

Protein concentration in the samples was estimated by the method of Bradford.16 The purity of phycocyanin was determined as the relationship of the 620 nm to 280 nm absorbances (ie purity of phycocyanin = 620 nm Abs/280 nm Abs). Samples taken from the different experiments were also used for SDS±PAGE (12% w/w T /2.6% w/w C) analysis, following the methods described by Laemmli.17

3 RESULTS AND DISCUSSION 3.1 Influence of system parameters upon partition behaviour of c-phycocyanin in PEG–salt aqueous two-phase systems

The poor understanding of the mechanism governing the behaviour of proteins in aqueous two-phase systems (ATPS) limits the predictive design of extraction processes using ATPS. In this paper, before designing the aqueous two-phase process the in¯uence of system parameters on the partition behaviour of c-phycocyanin was studied using single model systems. Such systems were characterized by the presence of only puri®ed c-phycocyanin in the ATPS. These systems took no account of the in¯uence upon the performance of ATPS of the whole range of proteins, contaminants and cell debris which may be present in the fermentation broth of the Spirulina maxima. In the development of an ATPS extraction process, the extent of the necessary empirical experiments to determine the process conditions can be reduced by using a practical approach which exploits the known effect of system parameters such as tie-line length (TLL), phase volume ratio (Vr), system pH and molecular weight of PEG on the protein partition

behaviour. For the design of the ATPS extraction stage to fractionate the cell homogenate from Spirulina maxima, the concentration of PEG and phosphate, system pH, the phase volume ratio (Vr) and the molecular weight of PEG were manipulated to maximize c-phycocyanin recovery from the top PEGrich phase. Initially, the effect of increasing TLL upon the partition behaviour of c-phycocyanin was evaluated. Changes in the TLL affect the free volume6 available for a de®ned solute to concentrate in the phase and as a consequence affect the partition behaviour of the solute in the ATPS. Table 1 illustrates the impact of increasing TLL upon the purity of c-phycocyanin from model and complex ATPS, when PEG of two different molecular weights (ie 1000 and 1450 Da) were used. For all these systems volume ratio and system pH were kept constant at 1.0 and 7.0, respectively. The partition experiments that used puri®ed c-phycocyanin in ATPS revealed that this protein exhibited a strong preference for the top phase (data not shown), which implies that the majority of the target protein concentrated in the top phase. The top-phase preference of the c-phycocyanin resulted in partition coef®cients greater than 100.0 and with great variations (ie from 100 to 200) for all the systems studied. Such behaviour was explained by problems associated with the detection of the presence of c-phycocyanin in the bottom phase, caused by the very low amount of the protein remaining in this phase. As a consequence, it was very dif®cult to evaluate the impact of system parameters upon the partition behaviour of c-phycocyanin by monitoring the protein partition coef®cient (K). As a result, it was decided to use the purity of c-phycocyanin (expressed as the relationship of the 620 nm to 280 nm absorbance) from the top PEG-rich phase as the response variable to evaluate the effect of system parameters on the behaviour of the protein in the ATPS. Table 1 shows that for both experimental systems, ie model (with puri®ed c-phycocyanin) and complex (cell homogeneate from Spirulina maxima), increasing TLL caused the purity of c-phycocyanin from the top PEG-rich phase to decline, when PEG of 1000 and 1450 Da were used. Such behaviour may be

Table 1. Influence of increasing TLL upon the purity of c-phycocyanin from PEG–phosphate ATPS

System 1 2 3 4 5 6 7 8

Molecular weight of PEG (g gmol 1) 1000

1450

PEG (% w/w)

Phosphate (% w/w)

TLL (% w/w)

Purity of c-phycocyanin (model system)

Purity of c-phycocyanin (complex system)

18.2 19.9 22.1 24.1 15.0 17.5 21.9 22.9

15.0 16.0 19.0 20.1 12.9 14.2 18.0 19.8

38 40 50 55 30 40 52 55

2.9 0.3 2.8 0.3 2.6 0.2 2.1 0.2 3.1 0.3 2.4 0.2 2.1 0.2 2.0 0.2

1.2 0.12 1.2 0.12 1.0 0.10 0.9 0.09 1.6 0.16 1.5 0.15 1.2 0.12 1.0 0.10

The tie-line lengths (TLL) of the systems were estimated from the composition of PEG and phosphate as described in Materials and Methods. The purity of c-phycocyanin is expressed as the relationship of the 620 nm to 280 nm absorbances. For all systems, volume ratio (estimated from non-biological experimental systems) and the system pH were kept constant at 1.0 and 7.0 respectively.

J Chem Technol Biotechnol 76:1273±1280 (online: 2001)

1275


explained by the increase in the contaminant proteins in the top phase caused by the rise in the TLL. It has been reported that the free volume in the bottom phase decreases when the TLL is increased.6 As a result, the solutes in the lower phase may be promoted to partition to the top phase. Therefore, an increase in the amount of contaminant proteins that concentrate in the top phase with increasing TLL is possible and as a result the purity of c-phycocyanin is negatively affected. The differences observed in the purity from top PEG-rich phases from the model and complex systems (see Table 1) are explained by the nature of the experimental vehicles (the purity from the starting material for the model system was 1.8, whilst that from the crude extract was approximately 0.7). In the case of the model systems, the sole presence of the target protein resulted in a high purity from the top phase. In contrast, for the complex system the presence of contaminants from the fermentation broth from Spirulina maxima caused an effect in the partition behaviour and purity of c-phycocyanin. For all the systems studied using cell homogenate from Spirulina maxima (or crude extract), the purity of c-phycocyanin increased in ATPS compared with that from the crude extract (ie purity from the crude extract was approximately 0.7). PEG1450±phosphate ATPS, characterized by TLLs of 30% (w/w) (PEG 15.0%

w/w and phosphate 12.9% w/w, Vr = 1.0 and system pH of 7.0) and 40% (w/w) (PEG 17.0% w/w and phosphate 14.2% w/w, Vr = 1.0 and system pH of 7.0), resulted in the maximal purity (1.6 and 1.5, respectively) from these experiments. Once the impact of increasing TLL upon the purity of c-phycocyanin from the top phase was evaluated, the effect of system pH on the purity of the protein was investigated using cell homogenate from Spirulina maxima. The in¯uence of system pH on protein partition behaviour has been discussed by several authors.11±13,18,19 In general, these reports concluded that increasing the pH (eg from 6.5 to 9.0) caused an increase in the protein concentration in the top phase and a decrease in the bottom phase. Such behaviour of proteins has been attributed to free-volume effects.13 An alternative explanation may be associated with the speciation of the phosphate salts over the pH range and to conformational changes in the structural integrity of proteins.11 Table 2 shows the in¯uence of pH on the purity of c-phycocyanin in PEG± phosphate ATPS, when PEG of two different molecular weights (ie 1000 and 1450 Da) were used. The purity of c-phycocyanin decreased with increasing pH regardless of system pH or the molecular weight of PEG. The decrease of purity can be associated with an increase in the migration of contaminant proteins to

Table 2. Influence of changing system pH upon the purity of c-phycocyanin from PEG–phosphate ATPS

PEG (% w/w)

Phosphate (% w/w)

TLL (% w/w)

System pH

Purity of c-phycocyanin

18.2

15.0

38

6.5 7.0 8.0

1.5 0.15 1.6 0.16 1.2 0.12

2

19.9

16.0

40

6.5 7.0 8.0

1.5 0.15 1.4 0.14 1.2 0.12

3

22.1

19.0

50

6.5 7.0 8.0

1.4 0.14 1.4 0.14 1.0 0.10

4

24.1

20.1

55

6.5 7.0 8.0

1.2 0.12 1.2 0.12 0.9 0.09

15.0

12.9

30

6.5 7.0 8.0

1.5 0.15 1.5 0.15 0.9 0.09

6

17.5

14.2

40

6.5 7.0 8.0

1.8 0.18 1.6 0.16 1.2 0.12

7

21.9

18.0

52

6.5 7.0 8.0

1.5 0.15 1.5 0.15 1.2 0.12

8

22.9

19.8

55

6.5 7.0 8.0

1.5 0.15 1.5 0.15 1.0 0.10

System 1

5

Molecular weight of PEG (g gmol 1) 1000

1450

System pH was adjusted as described in Materials and Methods. The purity of c-phycocyanin is expressed as the relationship of the 620 nm to 280 nm absorbances. For all systems, volume ratio (estimated from non-biological experimental systems) was kept constant at 1.0.

1276



System

Molecular weight of PEG (g gmol 1)

System pH


a

1000

18.0

15.0

4.0 5.0 6.0 7.0

0.6 0.06 0.9 0.09 0.8 0.08 0.8 0.08

b

1450

18.3

11.0

4.0 5.0 6.0 7.0

1.5 0.15 1.5 0.15 1.7 0.17 1.7 0.17

the top phase with the increase in pH. Although increasing the system pH from 6.5 to 8.0 resulted in changes in the purity of c-phycocyanin from the ATPS studied, it is clear that no great differences to the previous protein purity obtained (see Table 1) were achieved. However, an ATPS with TLL of 40% (w/w) (PEG 1450 17.5% w/w, phosphate 14.2% w/w), Vr = 1.0 at pH of 6.5 was selected as the one that provided the best conditions to satisfy the needs of maximal protein purity (ie 1.8). Extractions using PEG±sulfate aqueous two-phase systems were evaluated in this research to further investigate the partition behaviour of c-phycocyanin in ATPS. Table 3 illustrates the in¯uence of changing system pH upon the purity of c-phycocyanin from PEG±sulfate ATPS. It is clear that systems with a PEG molecular weight of 1450 g gmol 1 exhibited greater protein purity compared with that from ATPS with a PEG molecular weight of 1000 g gmol 1. Although, the purity of c-phycocyanin obtained at a system pH below neutrality from the PEG±sulfate ATPS (ie 1.7; see Table 3) is similar to that from the PEG± phosphate ATPS (ie 1.8; see Table 2), it was decided to carry out the extraction with the PEG±phosphate ATPS to avoid contamination by cell debris in the top phase, since cell debris concentrated in the interface and top phase (data not shown) in the PEG±sulfate system. In the selected ATPS (TLL of 40% w/w,

Table 4. Influence of changing system Vr upon the purity of c-phycocyanin from PEG1450–phosphate ATPS

I II III IV V

Sulfate (% w/w)

System pH was adjusted as described in Materials and Methods. The purity of c-phycocyanin is expressed as the relationship of the 620 nm to 280 nm absorbances. For all systems, volume ratio (estimated from nonbiological experimental systems) and TLL were kept constant at 1.0 and 40% (w/w), respectively.

Table 3. Influence of changing system pH upon the purity of c-phycocyanin from PEG–sulfate ATPS

System

PEG (% w/w)

Volume ratio

PEG1450 (% w/w)

Phosphate (% w/w)


5.0 2.6 1.2 0.6 0.3

29.0 23.5 17.7 12.0 7.0

8.0 11.0 14.0 17.0 20.0

1.2 0.12 1.4 0.14 1.6 0.16 1.8 0.18 2.0 0.2

The volume ratio (Vr) in non-biological experimental systems along a single tie-line length (40% w/w) was estimated after phase separation in graduate centrifuge tubes. The purity of c-phycocyanin is expressed as the relationship of the 620 nm to 280 nm absorbances. For all ATPS, system pH was kept constant at 6.5.


PEG1450 17.5% w/w, phosphate 14.2% w/w, Vr = 1.0 at pH of 6.5), a decrease in the Vr caused the c-phycocyanin to rise (see Table 4). Hustedt et al 7 proposed that the protein partition behaviour remains constant for systems along the same tie-line. Such a proposal may be extended for the behaviour of c-phycocyanin in ATPS along the same tie-line. Changes in the protein purity with Vr can be attributed to a concentration effect. A decrease in the Vr implies a reduction of the volume of the top phase. Consequently, the c-phycocyanin in this phase will concentrate further and as a result an increase in the protein purity from that phase is possible. The system comprising Vr = 0.3, PEG1450 7.0% (w/w), phosphate 20.0% (w/w), TLL of 40% (w/w) and system pH of 6.5, which provided the best conditions for the maximum protein purity (ie 2.0), was selected for the evaluation of the effect of PEG molecular weight on the purity of c-phycocyanin from ATPS. The purity of c-phycocyanin from ATPS decreased when high molecular weights of PEG were used (see Table 5). The effect of increasing the molecular weight of PEG upon protein partition behaviour has been explained based upon the protein hydrophobicity20,21 and phase excluded volume.9,22,23 In the case of c-phycocyanin, the decrease in protein purity when high molecular weights of PEG were used may be explained by a migration of contaminant

Table 5. Influence of molecular weight of PEG upon the purity of cphycocyanin from PEG–phosphate ATPS

Molecular PEG Phosphate weight of PEG Purity of System (% w/w) (% w/w) (g gmol 1) c-phycocyanin A B C D

7.0 7.0 7.0 7.0

20.0 20.0 20.0 20.0

1000 1450 3350 8000

2.0 0.2 2.0 0.2 1.6 0.16 1.5 0.15

The purity of c-phycocyanin is expressed as the relationship of the 620 nm to 280 nm absorbances. For all systems, volume ratio (estimated from nonbiological experimental systems) and the system pH were kept constant at 0.3 and 6.5 respectively.

1277


Figure 1. Effect of the implementation of multiple ATPS extraction stages on the purity of c-phycocyanin. The purity of c-phycocyanin was estimated as the relationship of absorbances (Abs620/Abs280) as described in Materials and Methods and is represented relative to the number of ATPS extraction stages. ATPS extraction conditions were PEG1450 7% (w/w), phosphate 20% (w/w), Vr = 0.3, TLL of 40% w/w, system pH of 6.5 for all the extraction stages.

proteins from the bottom phase or interface to the top phase. An alternative explanation involves c-phycocyanin migration from the top to the bottom phase or the interface. ATPS with a low molecular weight of PEG (ie PEG1000 and PEG1450) exhibited the best protein purity. An ATPS using PEG1450±phosphate (instead of PEG1000±phosphate) was selected for the extraction stage, since the cell debris concentrated in the bottom phase. In contrast, in ATPS with PEG1000, cell debris accumulated at the interface. Such a situation may cause contamination problems when the top PEG-rich phase is removed for further processing. 3.2 Design of a process for c-phycocyanin recovery from Spirulina maxima

From the studies of the in¯uence of system parameters upon the purity of c-phycocyanin from ATPS, optimal process conditions (ie Vr = 0.3, PEG1450 7.0% w/w, phosphate 20.0% w/w, TLL of 40% w/w and system pH of 6.5) were selected for the ATPS extraction stage. Such extraction conditions resulted in a protein product with a purity of 2.0. In order to further increase the protein purity, subsequent ATPS extraction stages were used. In these extraction stages the top

PEG rich-phase (where the c-phycocyanin was present) from the previous ATPS, was further processed by the addition of fresh phosphate to create a new ATPS extraction stage. The process conditions for the subsequent ATPS extraction stages were kept constant and equal to those used for the ®rst ATPS extraction. Figure 1 illustrates the effect of the application of multiple ATPS extraction stages on the purity of c-phycocyanin. It is evident that the use of consecutive aqueous two-phase systems proved to be effective to increase the c-phycocyanin purity (from the top PEGrich phase) by further removing (in the bottom saltrich phase) the majority of protein contaminants. However, it is clear that no signi®cant increment in protein purity was obtained when a third ATPS extraction stage was used. The purity of c-phycocyanin from the top phase of the third extraction stage remained almost constant in comparison with that from the second ATPS extraction stage. Therefore, it was decided to adopt a two-stage ATPS extraction process to fractionate the fermentation broth of Spirulina maxima to recover c-phycocyanin. Once the operating conditions for the ATPS process were de®ned, the problem of removing the PEG from the target protein was addressed. The high concentration of PEG in the top phase compared with that of the c-phycocyanin (grams of PEG vs micrograms of protein) severely questioned the potential commercial value of the product. The removal of PEG1450 from the c-phycocyanin was attempted by ultra®ltration. The successful use of ultra®ltration for the elimination of PEG in ATPS has been reported before.12,18 In the present research a bench-scale ultra®ltration unit with a 30 000 Da membrane cut-off was used for the removal of PEG. Operating conditions (ie 40 psi at 15 °C) were de®ned based upon a previous report.15 PEG from the top phase of the second ATPS extraction was easily removed by ultra®ltration without apparent damage to the integrity of c-phycocyanin. The practical implementation of the ultra®ltration stage to the recovery process was that it resulted in an increase of protein purity (see Table 6). After the use of the ultra®ltration unit, the purity of c-phycocyanin increases from 2.4 to 3.1 approximately. It is possible to assume that the elimination of PEG by ultra®ltration was characterized by the removal of low molecular

Table 6. Protein yield and purity from each unit operation of the process for the recovery of c-phycocyanin

Crude extract First Aqueous two-phase extraction Second Aqueous two-phase extraction Ultra®ltration Ammonium sulfate precipitation

Protein yield from each individual step (%)

Cumulative yield (%)


Not evaluated 99.5 87.2 70.5 48.2

100 99.5 86.8 61.1 29.5

0.7 0.07 2.1 0.2 2.4 0.2 3.1 0.15 3.8 0.10

The yield from each step represents the practical recovery of c-phycocyanin and is expressed relative to the original amount of c-phycocyanin loaded to the process step. The purity of c-phycocyanin is expressed as the relationship of the 620 nm to 280 nm absorbances.

1278



Figure 2. A simplified representation of the potential process for c-phycocyanin recovery from Spirulina maxima.

weight contaminant proteins. Consequently, the purity of the concentrate increased signi®cantly. As a ®nal stage for the recovery process, precipitation was considered. The operating conditions used in this latest process stage were taken from the previous work of Herrera et al,3 without further optimization. Thus, the concentrate obtained after ultra®ltration was processed by ammonium sulfate precipitation. Two fractions were obtained, which correspond to 40% and 45% ammonium sulfate saturation at 0 °C. The 45% fraction contained the c-phycocyanin of the highest purity (ie 3.8; see Table 6). The protein recovery and purity obtained after each process stage is presented in Table 6. The protein recovery values take into account the losses occurring during the handling of the different streams of the process. The general process proposed for the potential recovery of c-phycocyanin produced by Spirulina maxima is represented in a simpli®ed manner in Fig 2. This process is characterized by a ®ve-unit operation for the downstream processing to produce a c-phycocyanin with a purity of 3.8. In contrast, the analytical technique reported by Herrera et al 3 and limited to the laboratory scale only, involves a minimum of ten unit operations to obtain a product with similar characteristics. The research presented here resulted in a recovery process that produced an overall c-phycocyanin recovery of 29.5% (see Table 6). To further characterize the ®nal product from the proposed process, its absorption spectrum was determined. This absorption spectrum (data not shown) shows a c-phycocyanin purity of 3.8 and its shape was almost identical to the one previously reported for c-phycocyanin.24 Fractionation of samples (from crude extract and the puri®ed product) by SDS±PAGE (see Fig 3) suggested that the majority of contaminant proteins present in the crude extract (as represented by Coomassie Blue stained J Chem Technol Biotechnol 76:1273±1280 (online: 2001)

bands) were eliminated by the different separation stages adopted for the downstream process. Thus, it is evident that an increase in the purity of c-phycocyanin was achieved. The bands migrating at about 21.5 kDa correspond to a and b phycocyanin chains. However, contaminant proteins were still visible in the ®nal product of the recovery process (an extra band between 14.2 and 21.5 and around 45 000 kDa in track 3 of Figure 3). Further characterization of the contaminant proteins present in the puri®ed product is needed to de®ne a strategy for process optimization. It would be interesting to re-examine the operating conditions of the stages proposed here to further

Figure 3. SDS–PAGE analysis of the protein product obtained from the potential process for c-phycocyanin recovery from Spirulina maxima. Samples of identical volume from the crude extract (track 2) and the final stream of the purification process (track 3) were fractionated in a 12% T, 2.65% C SDS–PAGE gel as described in Materials and Methods. In track 3, the diffused bands migrating at about 21.5 kDa correspond to a and b c-phycocyanin chains.

1279


increase the purity of c-phycocyanin. The recovery process proposed here increases the protein purity up to 3.8 in a total of ®ve stages, which raises the potential commercial application of this process as an alternative for the practical puri®cation of c-phycocyanin produced by Spirulina maxima. 4 CONCLUSIONS

This study reports for the ®rst time the fractionation of cell homogenate of Spirulina maxima in aqueous twophase systems for the development of a process for the potential recovery of c-phycocyanin. It has been shown that tie-line length, volume ratio, molecular weight of PEG and system pH in¯uence the purity of c-phycocyanin from the top PEG-rich phase. PEG±sulfate ATPS proved to be unsuitable for the recovery of c-phycocyanin since both cell debris and the target protein partitioned to the same phase or resulted in low protein purity values (ie 0.7±0.8). The operating conditions established for the PEG1450±phosphate ATPS extraction resulted in a two-stage process for the potential recovery of c-phycocyanin from Spirulina maxima. These conditions concentrated the protein preferentially to the top phase and the cell debris to the opposite phase. The recovery process developed involved the use of the conventional techniques of ultra®ltration and precipitation after ATPS extraction which further increase the protein purity by the elimination of contaminant proteins. Overall, the results reported here demonstrated the potential application of ATPS together with ultra®ltration and precipitation for the recovery of c-phycocyanina as a ®rst step for the development of a commercial process. ACKNOWLEDGEMENTS

The authors wish to acknowledge the ®nancial support of the International Foundation for Science (IFS, grant E2620-2) and the `Romulo Garza' FoundationMexico. The technical assistance of Dr Juan Manuel de la Fuente-Martinez is gratefully acknowledged. REFERENCES 1 Cohen Z, The chemical of Spirulina in Spirulina platensis (Arthrospira), in Physiology, Cell Biology and Biotechnology, Ed by Vonshak A, Taylor & Francis, London (1997). 2 Ciferri O, Spirulina, the edible microorganism. Microbiol Rev 47:551±578 (1983). 3 Herrera A, Boussiva S, Napoleone V and Holberg A, Recovery of c-phycocyanin from cyanobacterium Spirulina maxima. J Appl Phycol 1:325±331 (1989). 4 Prozyme, Catalog of products, (1999). 5 Ciferri O and Tiboni O, The biochemistry and industrial potential of Spirulina. Ann Rev Microbiol 39:503±526 (1985).

1280

6 Grossman PD and Gainer JL, Correlation of aqueous two-phase partitioning of proteins with changes in free volume. Biotechnol Prog 4:6±11 (1988). 7 Hustedt H, Kroner K-H and Kula M-R, Applications of phase partition in biotechnology, in Partitioning in Aqueous Two-phase Systems: Theory, Methods, Uses and Application in Biotechnology, Ed by Walter H, Brooks DE and Fisher D, Academic Press, Orlando, Fl. pp 529±584 (1985). 8 Albertsson PA, Partition of Cell Particles and Macromolecules, 1st edn, Wiley, NY (1986). 9 Huddleston JG, Ottomar KW, Ngonyani DM and Lyddiatt A, In¯uence of systems and molecular parameters upon fractionation of intracellular proteins from Saccharomyces by aqueous two-phase partition. Enz Microb Technol 13:24±32 (1991). 10 Boland MJ, Hesselink P and Husted H, Extractive puri®cation of enzyme from animal tissue using aqueous phase systems. J Biotechnol 11:337±352 (1989). 11 Rito-Palomares M and Hernandez M, In¯uence of systems and process parameters on partitioning of cheese whey proteins in aqueous two-phase systems. J Chromatogr 711:81±90 (1998). 12 Rito-Palomares M, Dale C and Lyddiatt A, Generic application of an aqueous two-phase process for protein recovery from animal blood. Process Biochemistry 35:665±673 (2000). 13 Rito-Palomares M and Lyddiatt A, Practical implementation of aqueous two-phase processes for protein recovery from yeast. J Chem Technol Biotechnol 75:632±638 (2000). 14 Bamberger S, Brooks DE, Sharp KA, van Alstine JM and Webber JJ, In Partitioning in Aqueous Two-Phase Systems: Theory, Methods, Uses and Application in Biotechnology, Ed by Walter H, Brooks DE and Fisher D, Academic Press, Orlando, Fl. p 86 (1985). 15 Solano-Castillo C, Kinetics of phase separation under different process and design parameters in aqueous two-phase systems, MSc thesis, ITESM, Monterrey, Mexico (1999). 16 Bradford MM, A rapid and sensitive method for the quanti®cation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248±254 (1976). 17 Laemmli UK, Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680± 685 (1970). 18 Huddleston JG, Polymer±salt aqueous two-phase systems in biochemical recovery, PhD thesis, University of Birmingham (1996). 19 Flanagan JA, Potential valorisation of brewery waste: recovery of bulk protein from waste brewer's yeast adopting established and novel technologies, PhD thesis, University of Birmingham (1994). 20 Kohler K, Ljungquist C, Kondo A, Veide A and Nilsson B, Engineering proteins to enhance their partition coef®cient in aqueous two-phase systems. Bio/Technology 9:642±646 (1991). 21 Franco T, Andrews A and Asenjo J, The use of chemically modi®ed proteins to study the effect of single protein property on partitioning in aqueous two-phase systems: effect of surface hydrophobicity. Biotechnol Bioeng 49:300±308 (1996). 22 Chen J-P, Partitioning and separation of a-lactoalbumin and b-lactoglobulin in PEG/potassium phosphate aqueous twophase systems. J Ferment Bioeng 73:140±147 (1992). 23 Creighton TE, Proteins: Structure and Molecular Properties, 2nd Edn, WH Freeman and Company, New York. pp 262±264 (1993). 24 Boussiba S and Richmond AE, Isolation and characterization of phycocyanins from the blue-green alga Spirulina platensis. Arch Microbiol 120:155±159 (1979).
