Translate this page into:
Environmentally-friendly strategy for separation of α-lactalbumin from whey by aqueous two phase flotation
⁎Corresponding author. fengzhibiao@neau.edu.cn (Zhibiao Feng)
-
Received: ,
Accepted: ,
This article was originally published by Elsevier and was migrated to Scientific Scholar after the change of Publisher.
Peer review under responsibility of King Saud University.
Abstract
Aqueous two-phase flotation (ATPF) consisting of 1000 g/mol polyethylene glycol (PEG 1000)/trisodium citrate was developed for the separation of α-Lactalbumin (α-La) from whey. The flotation efficiency (E) and purification factor (PF) of α-La were evaluated using the reversed-phase high-performance liquid chromatography (RP-HPLC). The effects of pH, concentration of trisodium citrate, flow velocity, flotation time and whey loading on the E and PF of α-La were investigated. An efficient separation of α-La from whey was achieved using ATPF with pH of 8.20, 5 mL of 0.50 g/mL PEG 1000 solution, 35 mL of 0.40 g/mL trisodium citrate solution and whey (20%, v/v), 30 mL/min of flow velocity and 42 min of flotation time. Under the optimal conditions, E and PF of α-La could reach 87.54% and 5.33, respectively. In addition, the kinetic process of the separation of α-La by ATPF of PEG 1000/trisodium citrate was investigated. The results showed that there were two stages in the separation process. Both stages obeyed the first-order kinetic equation, and the first stage was faster than the second one.
Keywords
ATPF
Whey
α-Lactalbumin
Flotation efficiency
RP-HPLC
Kinetic
1 Introduction
Aqueous two-phase flotation (ATPF) is a novel technique which integrates the principles of aqueous two-phase system (ATPS) and mass transfer mode of solvent sublation. Both ATPS and ATPF are liquid-liquid extraction techniques, but compared to ATPS, the unidirectional rising bubbles mass transfer mode is subjoined in ATPF (Lee et al., 2015). In addition, the dissolution losses of polymer in the aqueous phase in ATPF are less than that in ATPS due to unbalanced dissolution. Therefore, the ATPF method can effectively reduce the amount of polymer required and increase the enrichment of the target on the basis of retaining the ATPS advantages (milder, low cost and easy to industrialization).
Applications of ATPF have been reported so far in separation and enrichment of antibiotics, active ingredient in Chinese herbal medicine and enzymes (Lee et al., 2015). In recent years, ATPF have been used in the separation of antibiotics, such as penicillin (Bi et al., 2011), chloramphenicol in food (Han et al., 2014) and tetracyclines in water (Wang et al., 2009). In addition, Lu et al. (2016) used ATPF to separate and enrichment trace lomefloxacin and ciprofloxacin from food samples, and the flotation efficiency of lomefloxacin and ciprofloxacin were 94.50% and 98.23%, respectively. Separation and enrichment of total flavonoids in phellinus igniarius using ATPF was reported by Ge et al. (2012), and the method was suitable for the separation and enrichment of the active ingredient in the Chinese herbal medicine. Furthermore, geniposidic acid from leaves of Eucommia ulmoides (Zhao et al., 2015), flavonol glycosides from Solanum rostratum Dunal (Chang et al., 2017) and ortho-phenylphenol from biodesulfurization of 4-methyl dibenzothiophene (Padilha et al., 2017) were successfully separated by ATPF. Lipase was recovered and purified using different ATPF system, for example polyethylene glycol (PEG)/sodium citrate (Tan et al., 2014), alcohol/salt (Mathiazakan et al., 2016) and 2-propanol/potassium phosphate (Show et al., 2013). Pakhale et al. (2013) reported ATPF system of 1500 g/mol PEG (PEG 1500)/potassium phosphate was studied for the separation and partial purification of bromelain from the pineapple fruit, in result, maximum yield of 91.47% and purification fold of 4.26 were obtained. However, the separation and enrichment of protein from whey with ATPF has not been studied.
Whey is a by-product of the manufacture process of cheese or casein that is normally treated as a waste. Whey protein is a mixture of proteins that consists mainly of β-Lactoglobulin (β-Lg), α-Lactalbumin (α-La), Albumin from bovine serum (BSA), Immunoglobulin and Lactoferrin. Bovine α-La is the second major protein in whey whose content in bovine whey is about 1.2 g/L, and it is a relatively small (Mr 14200), acidic and Ca2+ binding protein (Konrad and Kleinschmidt, 2008). Containing a high proportion of tryptophan and cysteine, α-La is an excellent source of essential amino acids (Lonnerdal and Lien, 2003). The high degree of sequence homology (more than 72%) is found between human and bovine α-La (Zhang et al., 2012). However, in human milk, there is no β-Lg, which easily causes to infant allergic reactions (Savilahti et al., 2010). Various physiologic and nutritional functionalities of α-La were reported. It could boost the immune system and regulate the production of lactose (Stănciuc and Râpeanu, 2010). Moreover, it could inhibit the rising of blood glucose (Blouet et al., 2007) and growth of some cancer cell (Puthia et al., 2015), because cysteine was precursor material of reduced glutathione (GSH) in vivo. Therefore, there is considerable technical interest in its isolation if costs can be kept low. The current methods for separating α-La from whey are chromatography, membrane and isoelectric precipitation (El-Sayed and Chase, 2011). In addition, ATPS is also used for separating α-La.
Ion-exchange chromatography (Machado et al., 2015) and hydrophobic interaction chromatography (Muca et al., 2017; Zhang et al., 2010) were mainly chromatographic separation of α-La. However, chromatographic separation of whey protein was considered too expensive, even if this method had an ideal separation effect (Pedersen et al., 2003). Ultrafiltration was a common method for whey protein fractionation (Konrad and Kleinschmidt, 2008; Metsämuuronen and Nyström, 2006). Arunkumar (Arunkumar and Etzel, 2013) used positively charged tangential flow ultrafiltration membranes to fractionate α-La and β-Lg on the basis of isoelectric point (4.4 vs. 5.2, respectively). Isoelectric precipitation was also used for α-La fractionation (Lucena et al., 2007). However, the ultrafiltration and isoelectric precipitation process are cumbersome and time consuming. Therefore, the search for a simple, economical and efficient method is one of the goals of the current α-La separation. Furthermore, ATPS began to be favored by researchers. The successful separation of α-La and β-Lg from WPI with ATPS formed by ethylene oxide–propylene oxide (EOPO) co-polymer (UCON, Mn ≈ 3900)/phosphate was reported by Zhang (Zhang et al., 2016). Alcântara (Alcântara et al., 2014) found that PEG phase contained only α-La in exploring the partition of α-La and β-Lg from cheese whey in ATPS containing PEG and sodium polyacrylate. However, ATPS exists salt consumption, especially in most cases the use of phosphate or sulfate, easy to cause the pollution. Sodium citrate is safe and non-toxic (Zafarani-Moattar et al.,2004), due to its basic raw materials are derived from food. In addition, sodium citrate is biodegradable (Han et al., 2010). After a lot of natural water dilution, some become citric acid, which is prone to biodegradation (Shan et al., 2015), and eventually get carbon dioxide and water. Food and Agriculture Organization of the United Nations does not make any restrictions on sodium citrate daily intake (Hotha et al., 2014), and according to GB1886.25-2016, sodium citrate usage based on normal production needs. In 2014, Sivakumar (Sivakumar and Iyyaswami, 2014) achieved maximum partition coefficient (k) of 16.67 for α-La in ATPS composed of PEG 1000/trisodium citrate. Afterwards, Kalaivani (Kalaivani and Regupathi, 2015) combined with the Response Surface Methodology (RSM) to optimize the extraction conditions, and α-La was separated with recovery of 89% and purity of 96% in the PEG phase. So far, there has been no report about the separation of α-La from whey with ATPF.
The aim of this study was to develop a simple, inexpensive, pollution-free and efficient separation process for the separation of α-La from whey. With this goal, the separation of α-La from whey by PEG 1000/trisodium citrate ATPF was studied for the first time, and flotation efficiency, purification and kinetic of α-La from whey in ATPF were investigated.
2 Experimental
2.1 Materials
Milk samples were purchased from the local supermarket. 1000 g/mol polyethylene glycol (PEG 1000) was purchased from Aladdin, Shanghai. Acetonitrile and trifluoroacetic acid (TFA) were chromatographically pure (Dikma, USA). Foamdoctor®F1075 Defoamer was obtained by Dongguan zhong di mei fine chemicals import and export co., Ltd. The α-La and β-Lg standard sample was purchased from Sigma, USA. Sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) gel preparation kit (Solarbio, Beijing) was used for characterization of protein. Folin–Ciocalteu phenol reagent (Beijing Solarbio Science & Technology Co., Ltd.) was used for determining the concentration of protein. Trisodium citrate, citric acid, NaCl and other reagents were analytical reagent grade.
2.2 Apparatus
Chromatographic analysis was carried out on a Waters e2695 liquid chromatograph equipped with a 2998 PDA detector and a ChemStation (Waters, USA). The lyophilizer (LyoQuest-85 Plus) was used to lyophilize the sample (Telstar, Spain). Solution pH was measured by a PHS-3C pH meter (Shanghai Instrument Electric Scientific Instrument Co., Ltd.). An AL-04 electronic analytical balance (Mettler Toledo Instruments Co., Ltd., Shanghai) was used to measure the weights of samples. The sample was dialyzed with a 7000 Da MWCO filter (Spectrum Labs, USA). A CT14D desktop high speed centrifuge (Shanghai Techcomp Scientific Instrument Co., Ltd.) and a SC-3610 low speed centrifuge (Hefei Zhongke Zhongjia Scientific Instrument Co., Ltd) were used for the sample treatment. Biorad Mini-PROTEAN Tetra Cells 4-Gel 165–8004 was from Bio-Rad Co., Ltd. (USA). The ultraviolet–visible spectrophotometer was from Beijing Purkinje General Instrument Co., Ltd. An 50.0 mL flotation column was obtained from Zhenjiang Gas Chemical Glass Plant Co., Ltd. The flotation device was made by ourselves according to a reference (Lee et al., 2015).
2.3 Preparation of whey
To minimize the loss and modification of the sample protein, a simple method- isoelectric precipitation for the preparation of whey samples was used (Chen et al., 2014). The fresh milk was centrifuged at 4 °C (3000g, 10 min) to remove the upper fat, and skimmed milk was obtained. The pH of the skimmed milk was adjusted to 4.6 with 1 mol/L hydrochloric acid, and the mixture was continuously stirred while adjusting the pH value. After centrifugation at 5000g at 4 °C for 20 min, the pH of the supernatant was brought to 7 for getting whey. Whey was stored at 4 °C in the refrigerator for use.
2.4 Procedure
2.4.1 Flotation process
35 mL salt solution containing the whey and 50 ppm defoamer was transferred into 50.0 mL flotation column, and then 5.0 mL PEG 1000 solution was spread on the top of the sample solution. The nitrogen gas bubbling was through the bottom of the column at a certain flow velocity and flotation time. The α-La was carried to PEG phase by nitrogen gas. The top phase was transferred to a new centrifugal tube for the α-La determining (Fig. 1).
The schematic diagram of separation α-La in ATPF (Lee et al., 2015; Show et al., 2013). (1) Nitrogen gas cylinder; (2) buffer; (3) rotameter; (4) sintered glass disk (G4 porosity); (5) flotation column.
2.4.2 Purification
Whey and protein in both phases in ATPF were dialyzed against deionized water for 24 h at 4 °C by dialysis (dialysis bag with 7000 Da molecular weight cutoffs), with 4–5 changes. The salt and PEG 1000 were removed and then lyophilized to obtain a purified protein.
2.4.3 Kinetic of ATPF (Bi et al., 2011)
The α-La was separated by ATPF of PEG 1000/trisodium citrate at different flow velocities (25, 30 and 35 mL/min). Simultaneously, the timer was started. The samples of bottom phase were taken periodically to determine and monitor the process. The samples were analyzed by RP-HPLC.
2.4.4 Determining
The Bradford (Kalaivani and Regupathi, 2015) was used to determine the concentration of total protein in whey and the top phase. RP-HPLC was used to examine the concentration of α-La in whey and in the top/bottom phases before and after purification. 15 mg purified sample was dissolved in 10 mL distilled H2O for the determination of concentration of α-La by RP-HPLC. The conditions for RP-HPLC were as follows: Agilent C8 column (150 mm × 4.6 mm, 5 μm), 0.1% aqueous TFA as mobile phase A, 0.1% TFA in acetonitrile as mobile phase B, flow rate of 1.00 mL/min and detection wavelength of 280 nm. Gradient elution was programmed as follows: the concentration of mobile phase B was increased from 30% to 35% in 5 min, and then from 35% to 55% over 20 min. Purified protein from both phases and whey were subjected to SDS–PAGE analysis in a gel of 12% resolving gel and 5% stacking gel. Samples were boiled for 10 min in boiling water bath with equal volumes of sample loading buffer. Separation process was carried out at a constant voltage of 80 V for stacking gel and 120 V till the dye front reaches the bottom of the gel, respectively. After the end of the electrophoresis, the gel was washed with distilled water for several times and then stained with Coomassie Brilliant Blue R-250 for visualization of protein bands, and decolorized with eluent.
2.5 Determination of the process parameters of ATPF
The flotation efficiency (E) and purification factor (PF) of α-La were used to evaluation the separation and enrichment of α-La with ATPF, and were calculated by the following equations:
The purity (P) of α-La after purification was used to measure the standard of purification effect, and was calculated by the following equations:
2.6 Phase diagram determination
The phase diagrams were obtained by the cloud point method (Zhang et al., 2016). A dry conical flask was first weighed (ma). Citrate solution with a concentration (c) of 12% (w/w) was added and then the flask was reweighed (m0). Polymer solution was added drop wise to the citrate solution in the flask, while shaking at room temperature. The addition was continued until the cloud point of the solution was reached and the turbidity did not disappear on further shaking. The weight of this system was recorded (m1). Water was added to make the system clear and the system was weighed (m2). Then the above operation was repeated and weights at multiple cloud points were recorded as (m3, …, mn, n was even, n ≥ 2). According to formulae 4 and 5, the mass fractions of the salts (Y1) and polymers (Y2) in the system were calculated and the phase diagrams were plotted.
2.7 Statistical analysis (Jiang et al., 2017)
The range of optimized conditions of the ATPF for the separation and enrichment of α-La from whey was obtained by the one-variable-at-a-time experiments. Response Surface Methodology (RSM) was performed to study the interactive effect of the multiple independent variables. In this experiment, factorial design of RSM was performed by a Box-Behnken experimental design (BBD) in Design-Expert 8.0.6 with four factors and three levels. The effects of four factors were studied and the experimental design was shown in Table 1. Separation conditions of α-La from whey by ATPF were optimized to achieve the maximum of E and PF (Table 1).
Variables
Coded variable levels
−1
0
1
X1 pH
7
8
9
X2 concentration of trisodium citrate (g/mL)
0.35
0.40
0.45
X3 flow velocity (mL/min)
25
30
35
X4 flotation time (min)
30
40
50
Statistical analysis of the responses was carried out using Analysis of variance (ANOVA). The responses (E and PF) were fitted into the following quadratic equation
3 Results and discussion
3.1 Analysis of one-variable-at-a-time experiments
Many factors may affect the ATPF, such as pH, bubble size, bubble dispersant, ionic strength, composition and concentration of flotation system, flow velocity and flotation time (Lu and Zhu, 2007). The bubbles are the carrier of the material transfer during the flotation process, and the amount of the gas-liquid interface directly determines the flotation efficiency of the target. The size of the bubbles can be adjusted from both the porosity of the sintered glass disk and the nature of the flotation solution, for instance, adding the bubbles dispersant or adjusting the ionic strength. This article used the sintered glass disk of G4 porosity. Therefore, in our pre-experiment, the foaming properties of proteins seriously affect the flotation process, and it was found that adding a small amount of siloxane and silica defoamer (Foamdoctor® F1075, food grade) could achieve the purpose of rapid defoaming without affecting the flotation efficiency of α-La. In addition, the inorganic salt in the flotation system itself were able to meet the requirement of the ionic strength of the system, and there was no need to add the additional inorganic salt (NaCl) (Han et al., 2011). The results generated by pre-experimental also showed that adding NaCl had no effect on the flotation efficiency of α-La.
3.1.1 The choice of polymer and salt for ATPF
In general, the increase of molecular weight of the polymer resulted in the decrease of flotation efficiency of the target in ATPF (Lee et al., 2015). The targets of ATPF were generally biomolecules, such as antibiotics, proteins and enzymes. As the molecular weight of the polymer increased, the affinity of the biomolecules for the polymer phase decreased due to the repulsive interactions between the polymer and the biomolecules (Hatti-Kaul, 2000). Furthermore, the polymer molecular weight and concentration affected the viscosity of polymer solution and thus indirectly influenced the mass transfer efficiency of biomolecules as well as bubble rupturing time at the polymer phase (Show et al., 2011). However, low molecular weight might cause difficulties in phase separation. Based on the above, we selected the polymer PEG 1000 to build the ATPF (Lee et al., 2015).
Citrate is biodegradable and non-toxic and could be discharged into biological wastewater treatment plants (Su et al., 2014; Tubío et al., 2006). In order to avoid environmental pollution caused by inorganic salts, sodium citrate and potassium citrate were selected as the study objects. By changing the adding amount of PEG 1000, the phase diagrams of PEG 1000/potassium citrate and PEG 1000/sodium citrate were drawn by the cloud point method. As shown in Fig. 2, sodium citrate had the stronger phase separation ability. Therefore, PEG 1000/sodium citrate ATPF was chosen (Fig. 2).
Phase diagrams for PEG 1000/potassium citrate and PEG 1000/sodium citrate.
3.1.2 Effect of pH on the flotation efficiency
The pH of the system was an important factor affecting the ATPF process. It largely determined the existential form of the flotation target which directly affected the adsorption capacity of the substance on the bubbles surface and the solubility in the polymer phase, hence affected the flotation efficiency (Lu and Zhu, 2007; Wang et al., 2010). Therefore, pH range of 5–9 was chosen to study the pH dependence on E and PF by adding suitable amounts of citric acid in this study. The variation of E and PF with pH over the range 5–9 was shown in Fig. 3a. With the increasing of pH value, E and PF increased firstly and then decreased, reaching the maximum at pH 8. It indicated a strong dependence of pH on flotation of α-La from whey using ATPF of PEG1000/sodium citrate. The system pH influences the ionizable groups of protein and alters the protein surface charges (Pakhale et al., 2013). The isoelectric point of α-La is pH 4–5 (Zhang et al., 2012). At the pH range of 4–5 (isoelectric point), the solubility of α-La is significantly reduced, which leads to decrease in flotation efficiency. When the pH is lower than 4, strongly acidic environment may lead to protein denaturation. According to the pre-experiment, when the pH was 4, the top phase contained α-La and high content of β-Lg, which meant that α-La could not be separated from the whey. For the purpose of reducing pollution, citrates were chosen because it could be degraded. When pH value was 5, which was close to the isoelectric point of α-La, α-La was electrically neutral. In addition, α-La was in a premolten globule state where in the tryptophan residues were buried inside the hydrophobic core, which resulting that the hydrophobic interaction between polymer and protein were prevented by the highly compact structure (Sivakumar and Iyyaswami, 2014). Therefore, lower E and PF were observed for α-La at lower pH. More α-La was separated into the top phase when the system pH was maintained above the isoelectric point of α-La. With the increase of pH, the four hydrophobic side chains of tryptophan residues exposed on the surface of the protein which readily forms hydrophobic interaction with PEG. The stronger the hydrophobic effect was, the easier the protein was adsorbed on the surface of the rising bubbles, thereby increasing the flotation efficiency (Zhao et al., 2015). The E and PF of α-La were found to be decreased above pH 8, which may be due to instability of the protein at high pH (Pakhale et al., 2013). Generally, the pH of the citric acid-sodium citrate buffer is less than 9. To raise the pH of the solution above 9, additional NaOH is required, which increases the separation cost. In general, the protein has the highest solubility at pH 8–9, and the solubility decreases above 9. We choose pH 5–9 as the range of exploring the effect of pH on α-La flotation efficiency. Hence, the system pH 8 was selected for further experimentation (Fig. 3).
Comparison of single factor results. (a) effect of pH on the flotation efficiency; (b) effect of concentration of trisodium citrate on the flotation efficiency; (c) effect of flow velocity on the flotation efficiency; (d) effect of flotation time on the flotation efficiency; (e) effect of flotation time on the flotation efficiency (where E: the flotation efficiency of α-La; PF: the purification factor of α-La).
3.1.3 Effect of concentration of trisodium citrate on the flotation efficiency
In ATPF, a high concentration of salt is generally required to sustain a stable immiscible two-phase via salting-out effect, due to the high volume ratio of salt phase to polymer phase (Lee et al., 2015). It was found the effect of concentration (0.30–0.45 g/mL) of trisodium citrate on the E and PF of α-La in Fig. 3(b). It could be observed that the E and PF increased as the concentration of trisodium citrate increased till the concentration reached 0.40 g/mL. This may be due to the salting-out effect. With the increase of trisodium citrate concentration in the bottom phase, the solubility of protein in the bottom phase decreased, resulting in the transfer of protein to the upper PEG phase. When trisodium citrate concentration continuously increased, the high viscosity of salt solution would weaken the mass transfer of protein to air–water interface of rising bubbles, which resulted in the decrease in the E and PF (Han et al., 2014). The E and PF of α-La were 87.42% and 5.33 at 0.40 g/mL trisodium citrate concentration. In addition, at this salt concentration, the volume of the top phase did not change much before and after flotation with 0.50 g/mL PEG 1000 in PEG 1000/sodium citrate ATPF.
The polymer concentration affects the viscosity of polymer solution and thus indirectly influences the mass transfer efficiency of biomolecules between aqueous-rich phases. High concentration leads to high viscosity in the top, which is not conducive to the bubble entering the top phase through the interface. As a result, the amount of α-La in the top phase is reduced. Moreover, unsuitable high concentration of polymer in top phase will cause significant changes in the volume of the top phase before and after flotation, which is detrimental to repeated operations. While the interface of two-phase is unstable with a low concentration of polymer and it is easily destroyed during the flotation process. So 0.40 g/mL trisodium citrate and 0.50 g/mL PEG 1000 were selected.
3.1.4 Effect of flow velocity on the flotation efficiency
The gas flow velocity is one of the most important factors in the ATPF process, which directly affects the separation efficiency and separation rate (Lu and Zhu, 2007). Mass transfer of solute to air–water interface of rising bubble in the aqueous phase is the dominant transport process strongly affecting the flotation efficiency in ATPF. The flotation efficiency with low flow velocity is lower than that with a high one. Fig. 3(c) showed the effect of the nitrogen flow velocity (20–35 mL/min) on the E and PF of α-La. The E and PF of α-La slowly increased when the flow velocity changed from 20 mL/min to 30 mL/min. The E and PF reached the maximum (87.42% and 5.33, respectively) at the flow velocity of 30 mL/min, however, the E and PF began to decrease when the flow velocity was more than 30 mL/min. At a low flow velocity, the buoyancy of air bubbles was less than the interfacial tension, which made bubbles unable to go through the interface of the two phases (Lu et al., 2016). It led to failure of the flotation. Therefore, the increasing of flow velocity of nitrogen made it easier for the bubbles to enter the polymer phase through the phase interface. Furthermore, the increasing of flow velocity within a certain range was beneficial for improving the contact area and contact time of protein molecules with bubbles, which caused E and PF increasing. Under invariable mean radius of bubbles, air–water interface area of the same air volume at high flow velocity is larger than that at the low one. As a result, more solute would be adsorbed or attached to the interface and entrained to the top phase with high flow velocity, resulting in higher flotation efficiency. Excessive high flow velocity might destroy the interface between two phases and the solution in two phases mixed into one phase. In addition, excessive high flow velocity caused the bubbles rise too fast, resulting in short contact time and unstable adsorption of the protein molecules with the bubbles, or the protein adsorbed on the surface of the bubbles fell back into the water phase again. Therefore, E and PF decreased when the flow velocity exceeded a critical value. It could be concluded that too high or too low flow velocity was harmful to the ATPF process. To sum up, the nitrogen flow velocity of 30 mL/min was optimum flotation conditions in this experiment.
3.1.5 Effect of flotation time on the flotation efficiency
The effect of flotation time on the E and PF of α-La in the range of 10–50 min was investigated Fig. 3(d). Within 10–40 min, the E and PF increased with flotation time. Before mass transfer process reached the thermodynamic equilibrium, longer flotation time made the target object (α-La) thoroughly contact with the bubbles, and the possibility that α-La collides with the bubble and loads on the surface of the bubble increased, which resulted in that the α-La was well transferred to the top phase while the E and PF raised. With the flotation time of 40 min, the E and PF reached the maximum (87.42% and 5.33, respectively), and the ATPF process basically reached the balance. No significant change in E or PF was observed with the increasing of flotation time after it reached 40 min. Due to the long processing time, the interface between the two phases would be destroyed, and the efficiency of flotation would be reduced. Therefore, in a multi-factor experiment, 40 min was selected as optimal flotation time to balance time and flotation efficiency.
3.1.6 Effect of whey loading on the flotation efficiency
The effect of whey loading (5–25%, v/v) on the E and PF of α-La was investigated, which was shown in Fig. 3(e). It could be found that too much whey loading (more than 20% (v/v)) could reduce the E and PF. It might be attributed to the increase in the amount of contaminants and impurities in the system as the whey loading increases. Excess α-La and other contaminants in the system caused a decrease in ATPF performance. In addition, the increase in whey loading changed the volume ratio between the two phases, reducing the phase volume on the PEG and affecting the distribution of α-La to top phase. Similarly, an increase in whey loading resulted in the formation of a precipitate at the interface of the two phases, resulting in low purification factor. Therefore, the maximum whey loading (20%, v/v) with the high E and PF was selected, and was used to carry out further experiments.
3.2 Analysis of the RSM
3.2.1 Analysis of variance
Based on the results of a series of one-variable-at-a-time experiments, according to the response surface experimental design (Table 1), the experimental results of response surface design and were shown in Table 2. Regression analysis was performed on the experimental data obtained using Design-Expert 8.0.6 software, the equation for predicting the E and PF of α-La was obtained, which was given as follows:
Number
A: pH
B: trisodium citrate (g/mL)
C: flow velocity (mL/min)
D:flotatiom time (min)
E (%)
PF
1
0
0
0
0
85.89
5.33
2
0
0
1
−1
79.01
4.95
3
0
0
0
0
87.37
5.35
4
0
1
−1
0
65.51
5.04
5
0
−1
1
0
64.50
4.88
6
0
1
0
−1
73.77
5.11
7
0
1
0
1
75.03
5.11
8
0
0
−1
−1
70.80
5.11
9
0
1
1
0
72.80
4.83
10
−1
0
0
−1
69.30
5.13
11
0
0
0
0
87.23
5.36
12
0
0
1
1
82.97
4.97
13
1
0
0
1
79.20
5.23
14
0
−1
−1
0
54.17
5.04
15
1
−1
0
0
58.19
5.09
16
−1
0
0
1
72.62
5.07
17
1
0
0
−1
75.70
5.11
18
1
1
0
0
71.44
5.07
19
0
0
−1
1
73.54
5.16
20
0
−1
0
−1
61.54
5.13
21
−1
1
0
0
62.22
4.98
22
0
0
0
0
87.56
5.35
23
−1
0
1
0
69.57
4.84
24
−1
0
−1
0
62.69
5.00
25
1
0
1
0
78.01
4.90
26
1
0
−1
0
67.24
5.08
27
0
−1
0
1
66.60
5.18
28
0
0
0
0
87.02
5.31
29
−1
−1
0
0
54.47
5.04
It was found that the equations were two quadratic equations with the negative quadratic coefficients, thus, the paraboloid openings of the equation images were downward and had the maximum value for optimization analysis. Analysis of variance (ANOVA) for the models E and PF were shown in Tables 3 and 4. The regression models were highly significant (p1 < 0.01, p2 < 0.01), while the lack-of-fit tests were not significant (p1 = 0.9974 > 0.05, p2 = 0.9967 > 0.05). Comparisons of true and predicted values for the E and PF of α-La yielded the small deviations Fig. 4(a) and (b), and the correlation coefficient (R12 and R22) for the models E and PF was 0.9992 and 0.9967, respectively, indicating a high degree of correlation between the true and predicted values. Therefore, this model could be used to analyze and predict the optimal flotation conditions of α-La (Tables 2–4 and Fig. 4).
Source
Sum of Squares
df
Mean Square
F
p1-Value
Model
2684.06
14
191.72
1203.56
<0.0001
Residual
2.23
14
0.16
Lack of fit
0.50
10
0.050
0.11
0.9974
Pure error
2.55
4
0.64
Cor total
2686.29
28
CV%
0.55
R12
0.9992
Source
Sum of Squares
df
Mean Square
F
p2-Value
Model
0.62
14
0.044
298.39
<0.0001
Residual
2.083E−003
14
1.488E−004
Lack of fit
4.833E−004
10
4.833E−005
0.12
0.9967
Pure error
1.600E−003
4
4.000E−004
Cor total
0.62
28
CV%
0.24
R22
0.9967
The normal plot of residuals for the flotation efficiency. (a) and purification factor (b) of α-La.
3.2.2 Interactive analysis
The relationship between response and experimental levels of each variable was visualized in the three-dimensional (3D) response surface plot, which provided a method to directly observe the interactions between the two test variables (Qiao et al., 2009). The factors influencing the E and PF of α-La were shown as the response surface plots in Fig. 5.
Response surface plots for flotation efficiency (a, b) and purification factor (c, d) of α-La.
Fig. 5(a) and (b) showed the effect of the interaction between pH and the concentration of trisodium citrate on the E and PF of α-La. The reasons might be the hydrophobic interaction and salting-out in ATPF (Lee et al., 2015). It was observed that lower pH and concentration of trisodium citrate were not conducive to the movement of α-La to the top phase. With the increase of the pH and concentration of trisodium citrate, ionic strength gradually increased, which caused that more hydrophobic parts on the protein surface exposed. As the hydrophobicity increased, more protein was adsorbed on the bubbles surface and transferred to the upper PEG phase. In addition, with the increase of salt concentration, more protein removed to the top phase due to the salting-out. However, excessively high ionic strength and salting-out effect not only reduced E, but also decreased the selectivity of the top phase for α-La, resulting in a reduction of PF (Fig. 5).
The effect of the interaction between the flow velocity and the flotation time on the E and PF of α-La was shown in Fig. 5c, d. The relationship between flow velocity and flotation time was that the higher flow rate could reduce the flotation time required to the equilibrium. But the interface between two phases was destroyed at the high flow velocity (Wang et al., 2010). Therefore, it was most appropriate to select the maximum flow velocity and the shortest time for ATPF.
3.2.3 Validation of optimal extraction conditions
According to the results of BBD, the optimal conditions were obtained when the pH of the system, concentration of trisodium citrate, flow velocity and flotation time were 8.16, 0.40 g/mL, 29.85 mL/min and 42.35 min, respectively. Under these conditions, the E and PF of α-La could reach 87.67% and 5.34, respectively. In order to facilitate the operation, the predicted optimal process conditions were amended as follows: the pH of the system was 8.20, the concentration of trisodium citrate was 0.40 g/mL, the flow velocity was 30 mL/min and the flotation time was 42 min. The repeatability and accuracy of this method were validated after repeated three times, and the E and PF of α-La could reach 87.54 ± 0.76% and 5.33 ± 0.05, respectively. The RP-HPLC chromatograms of whey, the top and bottom after ATPF under optimal conditions were shown in Fig. 6. It was be found that α-La from whey was separated to the PEG phase and β-Lg was left in the bottom phase.
RP-HPLC chromatograms of whey, the top and bottom after ATPF under optimal conditions.
3.3 Purification
Separation of proteins were presented in molecular marker (lane 1), whey (lane 2), top phase (lane 4) and bottom phase (lane 6) in Fig. 7, which clearly indicated the extraction of α-La in top phase and the presence of β-Lg and BSA in bottom phase. According to Eq. (3), the purity of α-La was 96 ± 0.79%. Therefore, the method (dialysis-lyophilization) of purifying α-La from polymer was feasible (Figs. 6 and 7).
SDS–PAGE profiles of samples after purification at the optimized ATPF conditions. molecular marker (lane 1), whey (lane 2), α-La standard (lane 3), top phase (lane 4), β-Lg standard (lane 5) and bottom phase (lane 6).
3.4 Kinetic of the separation of α-La with ATPF
The kinetic process of ATPF can be described by chemical reaction rate equation (Bi et al., 2011; Lu et al., 2005), and it is shown as follows:
In this study, the ATPF process of α-La from whey was analyzed at three stabilized flow velocities of 25, 30 and 35 mL/min. The dynamic flotation efficiency (E) of α-La was shown in Fig. 8(a). In the case of a certain volume, E and c are proportional. By rewriting Eq. (10) and taking the napierian logarithm of both sides, the first-order kinetic equation can be obtained as follows:
Dynamic process of ATPF (a) and relationship of ln E vs flotation time (b).
Based on the data of Fig. 8(a), the relationship between ln E and the flotation time t was shown in Fig. 8(b). Table 5 showed the related kinetic equations and the apparent rate constant. As shown in Fig. 8(b), it was obvious that there were two stages in the separation process. Both stages obeyed the first-order kinetic equation, and the first stage was faster than the second one. The concentration of target protein in the top and bottom phases changed with flotation time (Lee et al., 2015). In addition, many small and white particles could be observed in the bottom phase in the first stage, while in the second stage, the bottom solution became clear. In the first stage, the high concentration of trisodium citrate solution effectively reduced the solubility of protein due to the salting-out effect and can also accelerate the separation process. So, the precipitation flotation was the major separation mode, and the molecular adsorption flotation was the minor process (Bi et al., 2011). In the second stage, there was only the molecular adsorption flotation. These were also the reasons why the flotation efficiency of the first stage was higher than that of the second stage (Fig. 8 and Table 5).
Nitrogen flow velocity (mL/min)
First stage
Second stage
Apparent rate constant k
Kinetic equation
Correlation coefficient R2
Apparent rate constant k
Kinetic equation
Correlation coefficient R2
25
0.2252
ln E = 0.2252 t − 3.9485
0.9754
0.0114
ln E = 0.0114 t − 0.8486
0.9138
30
0.2551
ln E = 0.2551 t − 3.3287
0.9408
0.0057
ln E = −0.0057 t − 0.3956
0.8668
35
0.2608
ln E = 0.2608 t − 3.0872
0.9612
0.0056
ln E = 0.0056 t − 0.3501
0.8618
3.5 Comparison of ATPF and other methods for the separation of α-La
Isoelectric point precipitation, membrane separation, enzymatic hydrolysis, hydrophobic chromatography, gel chromatography and ion exchange are the common techniques used to separate α-La from whey. In these processes, it is difficult to ensure the yield and purity at the same time. The process cost is too high, even though good purity and yield can be obtained. PEG 1000/trisodium citrate ATPF is more advantageous than traditional liquid-liquid extraction. The good yield and purity for the separation α-La from whey can be obtained by ATPF of PEG 1000/trisodium citrate. Compared to Kalaivani and Regupathi (2015), the amount of polymer is significantly reduced, and no need to add NaCl. Compared with Zhang (Zhang et al., 2016), the process costs and time are saved. More importantly, the use of safe and biodegradable sodium citrate effectively avoids environmental pollution problems.
4 Conclusions
In this study, α-La from the whey was separated by an ATPF containing PEG 1000/trisodium citrate. Under optimum conditions, including a pH of 8.20, 5 mL of 0.50 g/mL PEG 1000 solution, 35 mL of 0.40 g/mL trisodium citrate solution and whey loading (20%, v/v), 30 mL/min of flow velocity and 42 min of flotation time, E and PF of α-La could reach 87.54 ± 0.76% and 5.33 ± 0.05, respectively. After the purification by the method of dialysis-lyophilization, the purity of α-La was 96 ± 0.79%. According to RP-HPLC chromatograms and the electrophoresis diagram, the ATPF of PEG 1000/trisodium citrate and dialysis-lyophilization purification method can successfully separate and purify α-La from the whey. It has the advantages of not only reducing the polymer loss, but also to avoid the inorganic pollution of the environment. In addition, it was be found that there were two stages in the kinetic process of the separation of α-La by ATPF of PEG 1000/trisodium citrate. Both stages obeyed the first-order kinetic equation. The first stage was faster than the second one, because the precipitation flotation was the major separation mode, and the molecular adsorption flotation was the minor process in the first stage, there was only the molecular adsorption flotation in the second stage. Theoretical research on the development of kinetics models of ATPF is still going on (Lee et al., 2015).
Acknowledgements
The authors would like to acknowledge financial support from the National Natural Science Foundation of China (No. 31201366), Natural Science Foundation of Heilongjiang Province (C2018019) and Research Science Foundation in Technology Innovation of Harbin (2016RAQXJ052).
References
- Partitioning of α-lactalbumin and β-lactoglobulin from cheese whey in aqueous two-phase systems containing poly (ethylene glycol) and sodium polyacrylate. Food. Bioprod. Process.. 2014;92:409-415.
- [Google Scholar]
- Fractionation of α-lactalbumin from β-lactoglobulin using positively charged tangential flow ultrafiltration membranes. Sep. Purif. Technol.. 2013;105:121-128.
- [Google Scholar]
- Separation behavior of penicillin in aqueous two-phase flotation. Chin. J. Anal. Chem.. 2011;39:425-428.
- [Google Scholar]
- Meal cysteine improves postprandial glucose control in rats fed a high-sucrose meal. J. Nutr. Biochem.. 2007;18:519-524.
- [Google Scholar]
- Separation of four flavonol glycosides from Solanum rostratum Dunal using aqueous two-phase flotation followed by preparative high-performance liquid chromatography. J. Sep. Sci.. 2017;40:804-812.
- [Google Scholar]
- 2-DE Analysis of whey protein extracted by different isoelectric precipitations and ultracentrifugation methods from cow milk. Food. Sci.. 2014;35:180-184.
- [Google Scholar]
- Separation/richment total flavonoids in phellinus igniarius by ionic liquid/salt aqueous two-phase flotation. Chin. J. Anal. Chem.. 2012;40:317-320.
- [Google Scholar]
- Aqueous Two-phase Systems: Methods and Protocols. Humana Press; 2000.
- Optimization of separation and determination of chloramphenicol in food using aqueous two-phase flotation coupled with HPLC. J. Iran. Chem. Soc.. 2014;11:1775-1782.
- [Google Scholar]
- Separation, concentration and determination of chloramphenicol in environment and food using an ionic liquid/salt aqueous two-phase flotation system coupled with high-performance liquid chromatography. Anal. Chim. Acta. 2011;685:138-145.
- [Google Scholar]
- Liquid–liquid equilibria of ionic liquid 1-butyl-3-methylimidazolium tetrafluoroborate and sodium citrate/tartrate/acetate aqueous two-phase systems at 298.15K: experiment and correlation. Fluid Phase Equilibr.. 2010;295:98-103.
- [Google Scholar]
- Development of better-quality assay method for the citric acid and sodium citrate in ophthalmic/oral solutions and their application to deformulation studies. Am. J. Anal. Chem.. 2014;05:1249-1260.
- [Google Scholar]
- Separation and enrichment of lectin from Zihua snap-bean (phaseolus vulgaris) seeds by PEG 600-ammonium sulfate aqueous two-phase system. Molecules. 2017;22:1596.
- [Google Scholar]
- Synergistic extraction of α-Lactalbumin and β-Lactoglobulin from acid whey using aqueous biphasic system: process evaluation and optimization. Sep. Purif. Technol.. 2015;146:301-310.
- [Google Scholar]
- A new method for isolation of native α-lactalbumin from sweet whey. Int. Dairy. J.. 2008;18:47-54.
- [Google Scholar]
- Aqueous two-phase flotation for the recovery of biomolecules. Sep. Purif. Rev.. 2015;45:81-92.
- [Google Scholar]
- Nutritional and physiologic significance of -lactalbumin in infants. Nutr. Rev.. 2003;61:295-305.
- [Google Scholar]
- α-Lactalbumin precipitation from commercial whey protein concentrates. Sep. Purif. Technol.. 2007;52:446-453.
- [Google Scholar]
- Simultaneous separation/enrichment and detection of trace ciprofloxacin and lomefloxacin in food samples using thermosensitive smart polymers aqueous two-phase flotation system combined with HPLC. Food. Chem.. 2016;210:1-8.
- [Google Scholar]
- The removal of humic acids from water by solvent sublation. J. Colloid. Interface. Sci.. 2005;283:278-284.
- [Google Scholar]
- Adsorptive behavior of α-lactalbumin on cation-exchange supermacroporous monolithic column. Fluid. Phase. Equilib.. 2015;401:64-69.
- [Google Scholar]
- Pilot-scale aqueous two-phase floatation for direct recovery of lipase derived from Burkholderia cepacia strain ST8. Sep. Purif. Technol.. 2016;171:206-213.
- [Google Scholar]
- Evaluation of six flat sheet ultrafiltration membranes for fractionation of whey proteins. Desalination. 2006;200:290-291.
- [Google Scholar]
- Effect of mass overloading on binding and elution of unstable proteins in hydrophobic interaction chromatography. J. Chromatogr. A. 2017;1492:79-88.
- [Google Scholar]
- Recovery and concentration of ortho-phenylphenol from biodesulfurization of 4-methyl dibenzothiophene by aqueous two-phase flotation. Sep. Purif. Technol.. 2017;176:306-312.
- [Google Scholar]
- Separation of Bromelain by Aqueous Two Phase Flotation. Sep. Sci. Technol.. 2013;48:984-989.
- [Google Scholar]
- Whey proteins as a model system for chromatographic separation of proteins. J. Chromatogr. B. 2003;790:161-173.
- [Google Scholar]
- P0127 Treatment and prevention of colon cancer by peroral HAMLET (human alpha-lactalbumin made lethal to tumour cells) Eur. J. Cancer.. 2015;51:e1-e36.
- [Google Scholar]
- Extraction optimized by using response surface methodology, purification and preliminary characterization of polysaccharides from Hyriopsis cumingii. Carbohyd. Polym.. 2009;76:422-429.
- [Google Scholar]
- Duration of clinical reactivity in cow's milk allergy is associated with levels of specific immunoglobulin G4 and immunoglobulin A antibodies to beta-lactoglobulin. Clin. Exp. Allergy.. 2010;40:251-256.
- [Google Scholar]
- Soft tissue regeneration after implantation of two novel citric acid-based biodegradable materials in rats. J. South. Med. Univ.. 2015;35:1184.
- [Google Scholar]
- Recovery of lipase derived from Burkholderia cenocepacia ST8 using sustainable aqueous two-phase flotation composed of recycling hydrophilic organic solvent and inorganic salt. Sep. Purif. Technol.. 2013;110:112-118.
- [Google Scholar]
- Direct recovery of lipase derived from Burkholderia cepacia in recycling aqueous two-phase flotation. Sep. Purif. Technol.. 2011;80:577-584.
- [Google Scholar]
- Recovery and partial purification of bovine α-lactalbumin from whey using PEG 1000–trisdoium citrate systems. Sep. Sci. Technol.. 2014;50:833-840.
- [Google Scholar]
- An overview of bovine α-lactalbumin structure and functionality. Fascicle VI: Food. Technol.. 2010;34:82-93.
- [Google Scholar]
- Study on the antimicrobial properties of citrate-based biodegradable polymers. Biotechnol. Bioeng.. 2014;2:23.
- [Google Scholar]
- Primary recovery of thermostable lipase 42 derived from recombinant Escherichia coli BL21 in aqueous two-phase flotation. Sep. Purif. Technol.. 2014;133:328-334.
- [Google Scholar]
- Liquid−liquid equilibria of aqueous two-phase systems containing poly (ethylene glycols) of different molecular weight and sodium citrate. J. Chem. Eng. Data. 2006;51:209-212.
- [Google Scholar]
- Pittsfield water treatment plant: once the world’s largest flotation-filtration plant. In: Flotation Technology. Humana Press; 2010. p. :485-502.
- [Google Scholar]
- Ionic liquid solvent sublation-spectrophotometric determination of trace tetracyclines antibiotic in water. Chin. J. Anal. Chem.. 2009;37:72-76.
- [Google Scholar]
- Liquid–liquid equilibria of an aqueous two-phase system containing polyethylene glycol and sodium citrate: experiment and correlation. Fluid. Phase. Equilib.. 2004;219:149-155.
- [Google Scholar]
- Separation of α-lactalbumin and β-lactoglobulin in whey protein isolate by aqueous two-phase system of polymer/phosphate. Chin. J. Anal. Chem.. 2016;44:754-759.
- [Google Scholar]
- Multiple reaction monitoring-based determination of bovine alpha-lactalbumin in infant formulas and whey protein concentrates by ultra-high performance liquid chromatography-tandem mass spectrometry using tryptic signature peptides and synthetic peptide standards. Anal. Chim. Acta. 2012;727:47-53.
- [Google Scholar]
- Separation and enrichment of geniposidic acid from leaves of Eucommia ulmoides by aqueous two-phase flotation. Chin. Tradit. Herbal. Drugs. 2015;46:2400-2406.
- [Google Scholar]
- Efficient separation of homologous alpha-lactalbumin from transgenic bovine milk using optimized hydrophobic interaction chromatography. J. Chromatogr. A. 2010;1217:3668-3673.
- [Google Scholar]
- UV-Vis spectrophotometric determination trace chloramphenicol in water by aqueous two-phase gas solvent sublation. Appl. Chem. Ind.. 2015;44:1358-1362.
- [Google Scholar]