Analytica Chimica Acta 815 (2014) 22–32

Contents lists available at ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

Design of functional guanidinium ionic liquid aqueous two-phase systems for the efficient purification of protein Xueqin Ding, Yuzhi Wang ∗ , Qun Zeng, Jing Chen, Yanhua Huang, Kaijia Xu State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082, PR China

h i g h l i g h t s

g r a p h i c a l

a b s t r a c t

• A series of novel cationic functional

•

• • •

hexaalkylguanidinium ionic liquids and anionic functional tetraalkylguanidinium ionic liquids have been synthesized. Functional guanidinium ionic liquid aqueous two-phase systems have been first designed for the purification of protein. Mechanisms and performances of the process were researched. Simple, green, safety and presents better purified ability than ordinary process. A potential efficient platform for protein purification and related studies.

a r t i c l e

i n f o

Article history: Received 15 November 2013 Received in revised form 7 January 2014 Accepted 9 January 2014 Available online 19 January 2014 Keywords: Functional guanidinium ionic liquid Functional guanidinium ionic liquid aqueous two-phase systems Protein Purification Aggregation

a b s t r a c t A series of novel cationic functional hexaalkylguanidinium ionic liquids and anionic functional tetraalkylguanidinium ionic liquids have been devised and synthesized based on 1,1,3,3-tetramethylguanidine. The structures of the ionic liquids (ILs) were confirmed by 1 H nuclear magnetic resonance (1 H NMR) and 13C nuclear magnetic resonance (13C NMR) and the production yields were all above 90%. Functional guanidinium ionic liquid aqueous two-phase systems (FGIL-ATPSs) have been first designed with these functional guanidinium ILs and phosphate solution for the purification of protein. After phase separation, proteins had transferred into the IL-rich phase and the concentrations of proteins were determined by measuring the absorbance at 278 nm using an ultra violet visible (UV–vis) spectrophotometer. The advantages of FGIL-ATPSs were compared with ordinary ionic liquid aqueous two-phase systems (IL-ATPSs). The proposed FGIL-ATPS has been applied to purify lysozyme, trypsin, ovalbumin and bovine serum albumin. Single factor experiments were used to research the effects of the process, such as the amount of ionic liquid (IL), the concentration of salt solution, temperature and the amount of protein. The purification efficiency reaches to 97.05%. The secondary structure of protein during the experimental process was observed upon investigation using UV–vis spectrophotometer, Fourier-transform infrared spectroscopy (FT-IR) and circular dichroism spectrum (CD spectrum). The precision, stability and repeatability of the process were investigated. The mechanisms of purification were researched by dynamic light scattering (DLS), determination of the conductivity and transmission electron microscopy (TEM). It was suggested that aggregation and embrace phenomenon play a significant role in the purification of proteins. All the results show that FGIL-ATPSs have huge potential to offer new possibility in the purification of proteins. © 2014 Elsevier B.V. All rights reserved.

∗ Corresponding author. Tel.: +86 731 88821903; fax: +86 731 88821848. E-mail address: [email protected] (Y. Wang). 0003-2670/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.aca.2014.01.030

X. Ding et al. / Analytica Chimica Acta 815 (2014) 22–32

1. Introduction In recent years, ionic liquids (ILs) are attracting a mass of attention as electrolyte materials, reaction solvents, and extraction solvents attributed to their unique properties, such as excellent thermal stability, negligible vapour pressure, high conductivity stability and designable structures, etc. [1–3]. ILs can be designed to be environmentally benign, with large potential benefits for sustainable chemistry [4]. Nowadays research on ILs is booming, but research on the functional ILs is still in the initial stage. Functional ionic liquids are the ILs with functional cations or with functional anions [5]. In functional IL, the performances of IL and functional group are both changed obviously [6]. Functional IL has common performance of IL, meanwhile, it has its significant advantages in viscosity, melting point or other aspects [7]. Functional IL is mainly made up of the centre ion which has the functional group and other corresponding ions. Some centre ions are the independent atoms while some are heterocycles. The majority of the syntheses for functional IL are based on quaterisation [8] or ion exchange reaction of amine [9], imidazole [10] and other heterocycles [11–14]. Functional IL could be divided into cationic functional IL, anionic functional IL and dual functional IL. Cationic functions of IL including hydroxyl function, carboxyl function, ammonio function and so on. And anionic function processes of IL are usually implemented via the introduction of OH− , CF3 SO3− , (CF3 SO2 )N− , CH3 CH(BF3 )CH2 CN− and CN− [7]. Aqueous two-phase system (ATPS) is a highly promising technology of novel separation and pretreatment [15,16]. Compared with the traditional protein purification methods such as ammonium sulfate precipitation, salting out, electrophoresis, ionexchange chromatography and affinity chromatography, ATPS is considered to be the method which can save time and cost, and have no influence on the activity of proteins. ATPS is usually formed by combining either two incompatible polymers or a polymer and a salt in water above a certain critical concentration [17,18]. The two phases are mostly composed of water and non-volatile components, thus making ATPS be a kind of environmentally friendly technology. ATPS has been widely used in biotechnological applications for the separation, purification and recovery of proteins, antibodies, enzymes, nucleic acids and antibiotics. However, most of phase-forming polymers in ATPS are highly viscous, form an opaque solution, and sometimes interfere with the analytes [19]. Ionic liquid aqueous two-phase system (IL-ATPS), which combines the advantages of IL and ATPS, has many unique performances that traditional aqueous two-phase system cannot match [20]. Therefore, the IL-ATPS is widely applied in the biological separation and purification fields. The ionic liquid aqueous two-phase system was first found in 2002 by Dupont [21]. Then Rogers’ team mixed hydrophilic ionic liquid [C4 mim]Cl with K3 PO4 to form an aqueous two-phse system [22]. In previous works, medicinal compounds and proteins have been extracted by ordinary imidazolium IL-ATPSs [23–25]. IL-ATPS is a promising technology of purification and pretreatment [26]. But existing IL-ATPSs are few and most have poor designability and purification result. Therefore, it is urgent to devise a series of new IL-ATPS with stronger functionality and higher purification efficiency. Guanidinium IL is a new member of ILs family. Functional guanidinium ionic liquid not only has common performances of guanidinium IL but also has its unique potential in purification field. Owing to the higher thermal and chemical stability, better catalytic activity, and stronger biological activity, guanidine compounds have attracted a multitude of attention of pharmaceutical scientists and chemists [27]. Especially in life support system, guanidine group is an emphasis of the bioscientific research because it embraces unique molecular recognition

23

function. Moreover, the high dispersive degree of the cationic parts, the adjustability of the three nitrogen-atoms and other characteristic properties making the synthesis and the application of guanidine salt increasingly popular [28]. It is demonstrated that guanidinium IL is an environmental friendly solvent which has good designability, and has more potential than traditional imidazolium IL. But no one had researched the purification of proteins by functional guanidinium ionic liquid aqueous two-phase system (FGIL-ATPS). In this paper, we functionalized guanidinium IL based on its unique structure and properties, and synthesized a series of novel cationic functional hexaalkylguanidinium ILs and anionic functional tetraalkylguanidinium ILs. Then functional guanidinium ionic liquid aqueous two-phase systems (FGIL-ATPSs) have been first designed with these functional guanidinium ILs and phosphate solution for the purification of protein (Scheme 1). After phase separation, proteins had transferred into the IL-rich phase and the concentrations of proteins were determined by measuring the absorbance at 278 nm using a UV2450 UV–vis spectrophotometer. The secondary structure of protein during the experimental process was observed upon investigation using UV–vis, FTIR and CD. The mechanisms of purification were researched and confirmed by DLS, determination of the conductivity and TEM.

2. Experimental 2.1. Materials and apparatus Bovine serum albumin (BSA), lysozyme, ovalbumin and trypsin were obtained from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). 1,1,3,3-tetramethylguanidine (TMG, >99%) was purchased from Haohua Chemical Co., Ltd. (Shanghai, China). 1-bromobutane (≥98%), 1-bromohexane (≥98%) and Tetrabutylammonium bromide (≥99%) were supplied by Energy Chemical Reagent Co. (Shanghai, China). Formic acid (≥99%), acetic acid (≥99.5%) and propionic acid (≥99%) were obtained from BoDi Chemical Reagent Co. (Tianjin, China). Lactic acid (≥99%) was bought from XiLong Chemical Reagent Co. (Guangdong, China). 4-Chloro-1-butanol (98%) and 6-chloro-1-hexanol (98%) were purchased from XinYuan Chemical Reagent Co. (Tianjing, China). Potassium carbonate (≥99%) was obtained from FengChuan Chemical Reagent Co. (Beijing, China). Other reagents were supplied by Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). Deionized water was used throughout the experiment. Ionic liquids were dried by a 101-0E Ventilated drying oven (Beijing, China) and DZF-6051 vacuum drying oven (Shanghai, China). A Thermostats cultivating shaker (Shanghai, China) was used to provide a certain temperature and rotation speed in the experiment. A TGL-16C High-speed centrifuge (Shanghai, China) was employed to accelerate the phase separation process. Ultraviolet detection and determining the absorbance of sample were carried out on a UV-2450 UV–Vis Spectrophotometer (SHIMADZU, Japan). A DDS-2A conductivity (Shanghai, China) and a Zetasizer Nano-ZS90 dynamic light scattering (Malvern, Britain) were used for the measurement of the cluster phenomenon. A Mos-500 circular dichroism (CD) spectrometer was used to determine secondary structure of protein. A JEM-3010 transmission electron microscope (JEOL, Japan) was used to examine the microstructures of samples before and after extraction. FT-IR spectra were recorded in the range of 500–4000 cm−1 on a Spectrum One FT-IR spectrometer (PerkinElmer, USA). A Varian Inova 400 NMR Spectrometer (Varian, USA) was employed to record 1 H NMR and 13 C NMR spectra of the synthetic ILs.

24

X. Ding et al. / Analytica Chimica Acta 815 (2014) 22–32

Scheme 1. FGIL-ATPS for purification of protein.

2.2. Synthesis of the ionic liquids Four kinds of anionic functional tetraalkylguanidinium-based ILs which have –OH functional groups, two kinds of cationic functional hexaalkylguanidinium-based ILs which have –OH functional groups and two kinds of ordinary hexaalkylguanidinium-based ILs were prepared with minor modification according to the literatures [29–31] and the reaction conditions were optimized systematically in our work (Scheme 2). 2.2.1. Synthesis of anionic functional tetraalkylguanidinium-based ionic liquids In this work, we prepared the ILs consisting of 1,1,3,3tetramethylguanidinium cation and different anions, which were directly synthesized by neutralization of 1,1,3,3,tetramethylguanidine (TMG) and acids. In the experiment, 2.30 g TMG (20.0 mmol) and 100 mL ethanol were loaded into a 250-mL flask in a water bath of 25 ◦ C. Then 20.0 mmol CH3 CH(OH)COOH in 40 mL ethanol was charged into the flask under stirring. The reaction lasted for 2 h and the resulting mixture was evaporated under reduced pressure. The crude oily residue was dissolved in 100 mL ethanol, treated with active carbon, filtered, and evaporated under vacuum for 24 h, and then 3.7 g product was obtained with a yield of 93.7%. [TMG]COOH, [TMG]CH3 COOH and [TMG]CH3 CH2 COOH were prepared under the same procedure, and the corresponding production rate are 93.3%, 91.8% and 95.9%, respectively. 2.2.2. Synthesis of cationic functional hexaalkylguanidinium-based ionic liquids A mixture of TMG (20 mmol), 4-chloro-1-butanol (44 mmol), tetrabutylammonium bromide (2% molar fraction) and potassium carbonate (60 mmol) were refluxed in acetonitrile (50 mL) for 6 h with stirring at 60 ◦ C, then cooled to room temperature, filtered, and added 80 mL of H2 O. The residue was treated with 5 mL of 30% NaOH. The reaction mixture was extracted with 80 mL of petroleum ether to remove unreacted 1-bromobutane and

pentaalkylguanidine. The petroleum ether phase was back extracted with 30 mL of H2 O. A saturated aqueous NaBr solution of 20 mL was added to the combined aqueous phase. The hexaalkylguanidinium ILs was extracted from the saturated aqueous phase with 3 × 50 mL CH2 Cl2 . The organic layer was dried over MgSO4 and the solvent was removed under reduced pressure. Then N,N,N,Ntetramethyl-N , N -butanol-guanidinium ILs ([diBOHTMG]Cl) was composed and the production rate is 94.6%. [diHOHTMG]Cl was prepared under the same procedure, and the corresponding production yield is 91.4%. 2.2.3. Synthesis of ordinary hexaalkylguanidinium-based ionic liquids The synthetic route of ordinary hexaalkylguanidinium-based ILs was the same as that of cationic functional hexaalkylguanidiniumbased ILs except the change of 4-chloro-1-butanol to 1bromobutane. The production yields of N,N,N,N-tetramethyl-N dibutyl-guanidinium ILs ([diBTMG]Br) and [diHTMG]Br were 93.1% and 92.7%, respectively. 2.3. Characterization of the ionic liquids All of the synthetic ILs were dried for 24 h at 70 ◦ C under vacuum before usage and their structures were confirmed by 1 H NMR and MS. The properties of ILs were shown in Table 1. 2.4. Preparation of phase diagrams Phase diagrams were determined by the cloud-point method [32]. An IL solution of known mass fraction was added in a tube, and a known concentration of salt solution was then added dropwise until the mixture became turbid or cloudy, then a known volume of water was added to make the mixture clear again. The procedure above was repeated to obtain sufficient data to construct a liquid-liquid equilibrium binodal curve. The temperature of the

X. Ding et al. / Analytica Chimica Acta 815 (2014) 22–32

25

Scheme 2. The synthetic routes and structures of ionic liquids: (a) [TMG]COOH, (b) [TMG]CH3 COOH, (c) [TMG]CH3 CH2 COOH, (d) [TMG]CH3 CH(OH)COOH, (e) [diBOHTMG]Cl, (f) [diHOHTMG]Cl, (g) [diBTMG]Br, (h) [diHTMG]Br

systems was controlled to 25 ± 0.1 ◦ C by a B-220 water thermostat (Shanghai Yarong Biochemical Instrument Factory, China). 2.5. Purification of the proteins An appropriate amount (3.5 mmol) of ionic liquid, 1 mL K2 HPO4 aqueous solution (pH = 7 by adding H3 PO4 or K3 PO4 solution) with a concentration of 0.5 g mL−1 and 1 mL aqueous protein solution were added into a graduated tube. The mixture was shaken vigorously in a thermostats cultivating shaker for 30 min to attain equilibrium, and the temperature of the system was controlled at 25 ± 0.1 ◦ C. The phase separation quickly occurred after cessation of the shaking process. Then, a high-speed centrifuge operated at 3000 rpm was used to run for a period of 8 min in each test to ensure complete phase separation. After experiment, the concentration of protein in the top phase and bottom phase were determined by measuring the absorbance at 278 nm using a UV2450 UV–vis spectrophotometer. Protein standards were run in every solvent before the measurement in order to make the results more accurate. To avoid interference from the phase components, the samples were analyzed against the blanks containing the same phase components but without protein. Partition coefficients (K) and purification

efficiencies (W%) of the proteins were calculated by using the equations: K=

Ct Cb

W% =

(1) Ct Vt × 100% C0 V0

(2)

where Cb and Ct represent the protein concentration in bottom and top phases, respectively. V0 and C0 are the added volume and concentration in prepared protein solution. Vt on behalf of the volume in the top phase. 3. Results and discussion 3.1. Phase diagrams of the functional guanidinium ionic liquid aqueous two-phase systems Phase diagrams were determined to investigate phase forming conditions by the cloud-point method. It is interesting to find that these ILs are all able to form FGIL-ATPSs with K2 HPO4 , and the top phase of each system is IL-rich phase while the bottom phase is salt-phase. It is known that the closer to the origin of

X. Ding et al. / Analytica Chimica Acta 815 (2014) 22–32

MS (m/z)

the binodal curve, the lower the IL concentration required for the formation of two phases and then the stronger the phaseforming ability of the IL. It can be seen from Fig. 1 that the phase-forming abilities of the ILs decrease in the order: [diHOHTMG]Cl > [diBOHTMG]Cl > [diHTMG]Br > [diBTMG]Br > [TMG] CH3 COOH ≈ [TMG]CH3 CH(OH)COOH ≈ [TMG]CH3 CH2 COOH > [TMG]COOH. These different ILs have different curves, and the phase-forming abilities would be determined by many complex factors. Hydration capacity [33–35], for instance, is one of the main factors affecting the phase-forming abilities. IL with higher charge density has stronger hydration capacity than those with lower charge densities, resulting in the decrease in the number of water molecules available to hydrate the salt in the systems [36,37]. Moreover, –OH can form hydrogen bond with water molecule in water solution, which will promote the hydration capacity of ILs. In addition, the larger the carbon chain, the easier the two phases formed.

Positive ion: 116(M + H+ ), negative ion: 44(M − H− ) Positive ion: 116(M + H+ ), negative ion: 59(M − H− ) Positive ion: 116(M + H+ ), negative ion: 73(M − H− ) Positive ion: 116(M + H+ ), negative ion: 89(M − H− ) Positive ion: 229(M + H+ ), negative ion: 79(M − H− ) Positive ion: 261(M + H+ ), negative ion: 34.5(M − H− ) Positive ion: 285(M + H+ ), negative ion: 79(M − H− ) Positive ion: 317(M + H+ ), negative ion: 34.5(M − H− )

26

H NMR (300MHz, D2 O) 1

93.3 91.8 95.9 93.7 93.1 94.6 94.6 91.4 [TMG]COOH [TMG]CH3 COOH [TMG]CH3 CH2 COOH [TMG]CH3 CH(OH)COOH [diBTMG]Br [diBOHTMG]Cl [diHTMG]Br [diHOHTMG]Cl

Form (25 ◦ C) Productivity (%) ILs

Table 1 Properties of ionic liquids.

3.2.2. Effects of different proteins In order to study the different purification effects among different proteins, [TMG]COOH and [TMG]CH3 CH(OH)COOH were selected to purify lysozyme, trypsin, ovalbumin and BSA. It can be seen from Fig. 2 the purification efficiencies decrease in the order: lysozyme > trypsin > ovalbumin > BSA. The result can be attributed to the size effect of proteins because the transfer of protein to the ILrich phase requires the breaking of the interactions between phase components to create a cavity where the protein can be included [39,40]. In the FGIL-ATPSs the larger the protein size, the greater the energy required. So a smaller sized protein is easier to be extracted to the IL-rich phase than a larger one. For the proteins studied in this work, lysozyme (MW = 14 400) has a smaller molecular

Solid Liquid Liquid Liquid Solid Liquid Liquid Liquid

3.2.1. Effects of different ionic liquid aqueous two-phase systems In the purification work, BSA was chosen as model protein to research the purification ability about FGIL-ATPSs, and the partition behaviour in the ATPSs of Guanidinium IL ([TMG]COOH, [TMG]CH3 COOH, [TMG]CH3 CH2 COOH, [diBOHTMG]Cl, [diHOHTMG]Cl, [TMG]CH3 CH(OH)COOH, [diHTMG]Br, or [diBTMG]Br) + K2 HPO4 has been examined at 25 ◦ C. After purification, the concentrations of proteins were determined by measuring the absorbance at 278 nm using a UV–vis spectrophotometer. Table 2 shows the values of the partition coefficients for BSA in the IL-ATPSs change from 2.56 to 46.23. The purification efficiency of all the functional ILs were higher than the two ordinary ILs, and the value of [diHOHTMG]Cl was the highest (97.05%). These phenomena can be attributed to the –OH and carbon chain. With the introduction of –OH, IL molecule can form hydrogen bond with water molecule, and then produce large cavities to accept proteins. Moreover, we can see from the table that the longer the carbon chain, the higher the purification efficiency. For instance, [TMG]CH3 CH2 COOH > [TMG]CH3 COOH > [TMG]COOH. These three ILs have the same cations and different anions, so the main effects of purification came from anions. From COOH− to CH3 CH2 COOH− , hydrophobicities of ILs become stronger as the carbon chain become longer. In IL-ATPS, hydrophobic effects are the main force of the partition force of protein according to the literature [38]. Aromatic amino acids exist in the surface of proteins (tryptophan, phenylalanine and tyrosine), and they account for around 10% of amino acid residues of the protein. All of these amino acids have hydrophobic aromatic group. When IL coexisted with protein, the hydrophobic alkyl side chain of IL can produce hydrophobic interaction with amino acid residues of protein’s surface. This interaction is also one of the main driving forces to extract proteins in FGIL-ATPS. Similar hydrophobic interaction exists in many PEG-ATPSs.

2.91(s, 12H), 8.41(s, 1H) 1.757 (s, 3H), 2.754 (s, 12H) 1.58 (s, 3H), 2.72 (s, 12H), 3.61 (m, 2H) 1.09 (d, 4.6 Hz, 3H), 2.70 (s, 12H), 3.85 (q, J = 4.6 Hz, 1H) 0.89 (t, J = 7.2 Hz, 6H), 1.16–1.56 (m, 12H), 3.08 (s, 12H) 1.39–1.48 (m, 16H), 2.99 (s, 12H), 3.040 (m, 2H) 0.92 (t, J = 7.2 Hz, 6H), 1.26–1.63 (m, 20H), 2.90 (s, 12H) 1.23–1.65 (m, 24H), 2.92 (s, 12H), 3.10–3.16 (m, 2H)

3.2. Purification of the proteins

X. Ding et al. / Analytica Chimica Acta 815 (2014) 22–32

27

Fig. 1. Phase diagram of FGIL-ATPSs.

weight than trypsin (MW = 23 800), ovalbumin (MW = 45 000) and BSA (MW = 65 000), thus the reversed order for the purification efficiency has been observed.

ability of IL phase to capture and accommodate protein is far stronger than that of water phase, so there will be more protein transfer into the IL phase as more IL added in the system.

3.2.3. Effects of the amount of ionic liquid Added amount of IL in the system is a significant factor affecting the purification efficiency, we used the [diHOHTMG]Cl-ATPS and [TMG]CH3 CH(OH)COOH-ATPS as representations to find the optimum amount, 20 mg mL−1 as the concentration of BSA and 50% (w/w) K2 HPO4 as the salt solution. The results are shown in Fig. 3(a). When the ILs amounts varied from 1.5 mmol to 4 mmol, the purification yields increased rapidly. That is because with the concentration of IL increased, the formed micelles gradually increased which facilitated protein purification. Moreover, the

3.2.4. Effects of the concentration of salt solution From the graph (Fig. 3(b)), we can see that the purification efficiencies increased significantly as the concentration of salt solution varied from 0.3 g mL−1 to 0.6 g mL−1 in the two IL-ATPS systems. At relatively high salt concentration, the space of salt phase appeared to be tighter and more structured so that the protein (BSA) could be preconcentrated into the top IL phase. However, the purification was decreased when the concentration of salt was higher than 0.6 g mL−1 . That is because salt solution with too high concentration can contend with the top IL phase to capture water molecule,

Table 2 Purification results of different IL-APTSs (n = 3). Purification solvent

Ct (mg mL−1 )

Cb (mg mL−1 )

C (mg mL−1 )

Vt (mL)

V (mL)

W%

K

[diHOHTMG]Cl [diBOHTMG]Cl [diHTMG]Br [diBTMG]Br [TMG]CH3 CH(OH)COOH [TMG]CH3 CH2 COOH [TMG]CH3 COOH [TMG]COOH

17.33 16.18 11.56 11.42 14.35 12.83 12.70 11.90

0.41 0.35 3.84 4.46 1.16 3.29 3.67 4.42

20 20 20 20 20 20 20 20

1.12 1.15 1.16 1.17 1.27 1.20 1.17 1.15

1 1 1 1 1 1 1 1

97.05 92.68 67.05 66.81 91.12 76.98 74.30 68.43

42.27 46.23 3.01 2.56 12.37 3.90 3.46 2.69

Note: K symbolizes the partition coefficient, which was calculated from Eq. (1); W% represents the purification efficiency, which was calculated from Eq. (2); Vt is the volume of top phase; Ct and Cb are concentrations of top phase and bottom phase, respectively; C is concentration of added protein solution; V on behalf of volume of added protein solution.

28

X. Ding et al. / Analytica Chimica Acta 815 (2014) 22–32

Fig. 2. Purification efficiency of the proteins by sample FGIL-ATPSs at 25 ◦ C: [TMG]COOH-ATPS (black), [TMG]CH3 CH(OH)COOH-ATPS (striated).

rendering the water content of top IL phase decreased and the protein degenerated. Meanwhile, the higher concentration of salt solution, the stronger the electrostatic interaction between protein and salt will be and this will decrease the purification efficiency.

Fig. 4. UV–vis spectra before and after purification.

3.2.5. Effects of temperature The influence of temperature on the activity of BSA was also investigated. The line graph (Fig. 3(c)) shows when the temperature is below 25 ◦ C, the purification efficiencies of BSA increase obviously. But the efficiencies of both decrease gradually after

Fig. 3. Effect factors of purification process: [diHOHTMG]Cl-ATPS (black), [TMG]CH3 CH(OH)COOH-ATPS (red). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

X. Ding et al. / Analytica Chimica Acta 815 (2014) 22–32

29

Fig. 5. FT-IR spectra before and after purification.

reaching the top point at 25 ◦ C, and then the downward trends continued but more noticeable until temperature go up to 55 ◦ C. The possible reason for this phenomenon is that the hydrophobic interaction enhances with the increased temperature, but when the temperature is enough high to destroy the hydrogen bonding interaction between the surface water of protein and amino acid residue, the formation of hydrogen bond is less influenced by hydrophobic interaction so that hydrophobic effect weakened. When the temperature is higher than 65 ◦ C, all of the proteins are denatured and the extraction efficiencies are both 0%. Therefore, the optimum temperature of the process is 25 ◦ C. 3.2.6. Effects of the amount of protein The amount of protein added into experimental system is an important factor affecting the accuracy and stability of experimental result. In order to discuss the effect on purification efficiency of the amount of protein in our study, [diHOHTMG]Cl-ATPS and

Fig. 6. CD spectra before and after purification.

[TMG]CH3 CH(OH)COOH-ATPS (3.5 mmol IL and 1 mL 0.5 g mL−1 K2 HPO4 ) were adopted and the results illustrated in Fig. 3(d). It can be seen from Fig. 3(d) that the purification efficiencies were both more than 90% when the amount of BSA was between 12–24 mg, while out of this range, the purification efficiencies ware decreased. Therefore, too few or too many protein was adverse to protein purification. It is obvious that the number of protein aggregates is growing with the increasing amount of BSA. So it is believed that in the beginning, the purification efficiencies increases quickly before 12 mg. But the purification efficiencies show little change between 12 mg and 24 mg, which is the best scope of BSA purification. However, a purification system has a limited ability of purification, so the system is near saturation when BSA addition is 24 mg. When BSA addition is more than 24 mg, the purification quantities may have a little increase, but the purification efficiencies decreased rapidly.

Fig. 7. Conductivities of ionic liquids.

30

X. Ding et al. / Analytica Chimica Acta 815 (2014) 22–32

Fig. 8. Size distribution before and after purification.

3.2.7. Effects of pH It is known that the pH is a crucial parameter to influence the purification efficiency of protein in IL-ATPS. In this work, we used the [diHOHTMG]Cl-ATPS and [TMG]CH3 CH(OH)COOH-ATPS as representations to investigate the effects of pH on BSA purification. Because the phosphate buffer solution below the pH value of 7 cannot form an ATPS with an IL, we only investigate pH values above 7. The determination of the pH value was performed from 7.0 to 13.0 by adding H3 PO4 or K3 PO4 solution, as suitable. However, we did not obtain reliable results. Because the BSA investigated here have high purification efficiencies in the FGIL-ATPSs, more than 90% BSA have been enriched into the IL-rich phase. Therefore, the purification efficiency of BSA is not sensitive to the change of pH values in the studied pH range.

precision was investigated by the analysis of the top phase solution for three times by UV detection under the same conditions. The result indicates that the relative standard deviation (RSD) obtained was 1.08% (n = 3), explaining the precision of the UV–vis spectra is excellent. Stability experiment was performed by monitoring a sample continuously in three days under the same conditions. The value of RSD is 1.75% (n = 3), showing the good effectiveness of the method. Repeatability experiment was carried out by taking three copies of the same sample measured, respectively under the same conditions. The results show the RSD is 1.59% (n = 3), proving the method has good repeatability.

3.3. Method validation

4.1. Ultra violet spectroscopy

Method in this study was validated by the precision experiment, stability experiment and repeatability experiment. Apparatus

In order to study the conformation of protein before and after purification, UV–vis was investigated for BSA. Fig. 4 shows the

4. Mechanism of the ionic liquid aqueous two-phase systems

X. Ding et al. / Analytica Chimica Acta 815 (2014) 22–32

31

Fig. 9. TEM image of purification process: (A) IL-phase before purification, (B) protein solution before purification, (C) IL-phase after purification.

UV–vis spectra of IL-rich phase, BSA in pure water and BSA in IL-rich phase after purification. It is clear that the figures of BSA before and after purification were similar, and its maximum absorption peak after purification was still existed at 278 nm. At the meantime, there was no interference peak came from IL-rich phase in the studied range. This indicated that IL does not interfere with the measurement of protein, and there are no chemical bonds between BSA molecules and ILs. 4.2. Fourier-transform infrared spectra Attributed to the unique energy absorption bands for specific bonding environments or interactions, FT-IR spectra can provide useful information to identify the presence of certain functional groups or chemical bonds in a molecule of an interaction system [41]. Amide is the basic unit of the protein: amideI is assigned to both C=O stretching vibration and ring stretching vibrations, and amideII is assigned to C–N stretching vibrations. Fig. 5 shows the FT-IR spectra of pure ionic liquid, pure BSA and BSA in IL-rich phase after purification. It can be seen from the graph that before purification, the absorption band of ionic liquid (1572 cm−1 ) and the absorption bands of BSA (the amideI band at 1638 cm−1 and the amideII band at 1557 cm−1 ) were identifiable in the spectra. After purification, the two absorption bands have changed to be 1633 cm−1 and 1564 cm−1 in the IL-BSA complexity. There was no disappearance or little shift of the peak, suggesting the conformation of the protein was not changed after purification. 4.3. Circular dichroism spectra CD spectra have been utilized to determine the secondary structure of BSA. BSA water solution of 0.3 mg mL−1 has been selected to perform the experiment and the scanned range of CD spectra was 195–250 nm. The image resolution was 0.1 nm and the response time was 1 s. All of the data were the average value of three times of scan. In CD spectra, the double negative peaks of far ultraviolet (wavelengths of which are 208 nm and 222 nm, respectively) are the typical shape of ␣spiral of protein. We can see from Fig. 6 that the line shape of BSA in two different ATPSs are similar to that of BSA before purification, and the peaks before and after purification are the same (several slight differences of Y axis came from the influence of solvent). In addition, Fig. 6 manifests that the average molar ellipticities of BSA residue in different ATPSs have not changed compared to the one of BSA residue before purification. In general, all of the evidences suggest that the secondary structures of BSA in our study were not changed. 4.4. Determination of the conductivity In order to verify the existence of the cluster phenomenon in the IL-rich phase, a study of the microstructure of the top phase

is necessary. The experiment measured a series of IL solution with different concentrations at 25 ◦ C. As Fig. 7 shows, the conductivities went up rapidly at first, but then the values tended stable. The critical aggregation concentration (CAC) is the intersection of two tangent curves which were drawn from the rapidly and slowly changed points. In our work, concentrations of ILs were all greater than the CAC, so surfactant aggregation occurred in the IL-rich phase. 4.5. Dynamic light scattering detection The particle size distributions of the IL-rich phase without protein, aqueous protein solution and IL-rich phase after purification were determined by DLS. It is clear from Fig. 8 that the particle size of IL-rich phase was about dozens of nanometers before the purification and which of aqueous protein solution was several hundred nanometers (there are several interference peaks from water molecule). After purification, the particle size became larger which ranged from 600 nm to 8000 nm. This phenomenon illustrates that the interaction may occur between the surfactant and the protein. 4.6. Characterization of the structure by transmission electron microscopy In this work, TEM was used to characterize the appearance of the cluster. From Fig. 9, it can be seen that the IL particles and protein particles were all dispersive existed before the purification. While after purification these particles (IL and BSA) were aggregated together (as in the schematic diagram of the extraction process Scheme 1). This conclusion is consistent with the results of DLS. 5. Conclusion In summary, functional guanidinium ionic liquid aqueous twophase system, which combines the advantages of functional guanidinium IL and ATPS, has many unique performances that traditional aqueous two-phase system cannot match. Compared to the traditional imidazolium IL-ATPS, the proposed functional guanidinium IL-ATPSs has shorter synthetic time of ionic liquid and better designability. And these systems showed their unique advantages not only in the two-phase forming but also in protein purification. The purification efficiencies of the FGIL-ATPSs were higher than the ordinary IL-ATPSs. Lysozyme, trypsin, ovalbumin and BSA could be well enriched in IL-rich phase by FGIL-ATPSs, and single factor experiments proved that the experimental efficiency was influenced by the amount of IL, the concentration of salt solution, temperature and the amount

32

X. Ding et al. / Analytica Chimica Acta 815 (2014) 22–32

of protein. No degeneration of protein during the purified process proved FGIL-ATPS is an accepted method for protein purification. The successful design of the FGIL-ATPS opened up a piece of new fields of IL-ATPS. These FGIL-ATPSs can shorten the synthetic time of ionic liquids and improve the phase forming speed, and has stronger purification ability than ordinary IL-ATPS meanwhile has no influence on activity and structure of protein. It is certain that the FGIL-ATPS can not only be successfully applied in purification of protein but also has huge application value in purification of other biomacromolecules and some small molecules. Acknowledgements The authors greatly appreciate the financial supports by the National Natural Science Foundation of China (No. 21175040; No. 21375035; No. J1210040) and the Foundation for Innovative Research Groups of NSFC (Grant 21221003). References

[12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]

[1] K. Fukumoto, M. Yoshizawa, H. Ohno, J. Am. Chem. Soc. 127 (2005) 2398–2399. [2] P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis, Wiley-VCH, Weinheim, 2003. [3] J. Dupont, R.F. de Souza, P.A.Z. Suarez, Chem. Rev. 102 (2002) 3667–3692. [4] R.D. Rogers, K.R. Seddon, Science 302 (2003) 792–793. [5] J. Tang, L. Feng, S. Gao, Chem. Bioeng. 28 (2011) 4–10. [6] Z. Mua, F. Zhoua, S. Zhang, Y. Liang, W. Liu, Tribol. Int. 38 (2005) 725–731. [7] D. Zhao, Z. Fei, T.J. Geldbach, R. Scopelliti, G. Laurenczy, P.J. Dyson, Helv. Chim. Acta 88 (2005) 665–675. [8] S. Huang, Y. Wang, Y. Zhou, L. Li, Q. Zeng, X. Ding, Anal. Methods 5 (2013) 3395–3402. [9] W.A. Herrmann, C. Kocher, L.J. Goossen, G.R.J. Artus, Chem. Eur. J. 2 (1996) 1627–1636. [10] J. Harlow, F. Hill, T. Welton, Synthesis 6 (1996) 697–698. [11] D. Demberelnyamba, S.J. Yoon, H. Lee, Chem. Lett. 33 (2004) 560–561.

[31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41]

D. Zhao, Z. Fei, R. Scopelliti, P.J. Dyson, Inorg. Chem. 43 (2004) 2197–2205. K.S. Kim, D. Demberelnyamba, H. Lee, Langmuir 20 (2004) 556–560. H. Itoh, K. Naka, Y. Chujo, J. Am. Chem. Soc. 126 (2004) 3026–3027. A.M. Azevedo, P.A.J. Rosa, I.F. Ferreira, Sep. Purif. Technol. 65 (2009) 31–39. J.S. Becker, O.R.T. Thomas, M. Franzreb, Sep. Purif. Technol. 65 (2009) 46–53. L. Li, F. Liu, X.X. Kong, Anal. Chem. Acta 452 (2002) 321–328. H. Walter, G. Johansson (Eds.), Methods of Enzymology, 228, Academic Press, New York, NY, 1994. C.X. Li, J. Han, Y. Wang, Y.S. Yan, X.H. Xu, J.M. Pan, Anal. Chem. Acta 653 (2009) 178–183. J. Han, Y. Wang, C. Yu, C. Li, Y. Yan, Y. Liu, L. Wang, Anal. Chim. Acta 685 (2011) 138–145. J. Dupont, C.S. Consorti, P.A.Z. Suarez, Org. Synth. 15 (2002) 236–243. R.D. Rogers, K.E. Gutowski, G.A. Broker, J. Am. Chem. Soc. 125 (2003) 6632–6633. Y. Yuan, Y. Wang, R. Xu, M. Huang, H. Zeng, Analyst 136 (2011) 2294–2305. H. Zeng, Y. Wang, J. Kong, C. Nie, Y. Yuan, Talanta 83 (2010) 582–590. X. Lin, Y. Wang, X. Liu, S. Huang, Q. Zeng, Analyst 137 (2012) 4076–4085. C. Li, J. Han, Y. Wang, Y. Yan, X. Xu, J. Pan, Anal. Chim. Acta 653 (2009) 178–183. Q. Zeng, Y. Wang, N. Li, X. Huang, X. Ding, X. Lin, S. Huang, X. Liu, Talanta 116 (2013) 409–416. H. Duan, X. Guo, S. Li, Y. Lin, S. Zhang, H. Xie, Chinese J. Org. Chem. 10 (2006) 1335–1343. H.X. Gao, B.X. Han, J.C. Li, T. Jiang, Z.M. Liu, W.Z. Wu, Y.H. Chang, J.M. Zhang, Synth. Commun. 17 (2004) 3083–3089. S.H. Li, Y.J. Lin, J.G. Cao, H.F. Duan, W.J. Cai, J.N. Xu, Chem. Res. Chin. Univ. 21 (2005) 158–162. A. Arce, E. Rodil, A. Soto, J. Solution Chem. 35 (2006) 63–78. K.S. Kim, S.Y. Park, S. Choi, H. Lee, J. Chem. Eng. Data 49 (2004) 1550–1553. M.G. Freire, J.F.B. Pereira, M. Francisco, H. Rodriquze, L.P.N. Rebelo, R.D. Rogers, J.A.P. Coutinho, Chem. Eur. J. 18 (2012) 1831–1839. Y. Marcus, J. Chem. Soc., Faraday Trans. 87 (1991) 2995–2999. S. Shahriari, C.M.S.S. Neves, M.G. Freire, J.A.P. Coutinho, J. Phys. Chem. B 116 (2012) 7252–7258. N.L. Abbot, D. Blankschtein, T.A. Hatton, Bioseparation 1 (1990) 191–225. ¯ J. Huddleston, A. Veide, K. Kohler, J. Flanagan, S. Enfors, A. Lyddiatt, Trends Biotechnol. 9 (1991) 381–388. S.N. Baker, T.M. McCleskey, S. Pandey, et al., Chem. Commun. 8 (2004) 940–941. Y. Pei, J. Wang, K. Wu, X. Xuan, X. Lu, Sep. Purif. Technol. 64 (2009) 288–295. Z. Li, X. Liu, Y. Pei, J. Wang, M. He, Green Chem. 14 (2012) 2941–2950. L. Jin, R.B. Bai, Langmuir 18 (2002) 9765–9770.