- Email: [email protected]
Metal chelate affinity precipitation: Purification of BSA using poly(N-vinylcaprolactam-co-methacrylic acid) copolymers
Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
Contents lists available at SciVerse ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Metal chelate affinity precipitation: Purification of BSA using poly(N-vinylcaprolactam-co-methacrylic acid) copolymers Yuan-Qing Ling a , Hua-Li Nie a,b , Christopher Brandford-White c , Gareth R. Williams c , Li-Min Zhu a,∗ a b c
College of Chemistry, Chemical Engineering and Biotechnology, Donghua University, 2999 North Renmin Road, Songjiang University City, Shanghai 201620, PR China Key Laboratory of Textile Science & Technology, Ministry of Education, Donghua University, Shanghai 201620, PR China Institute for Heath Research and Policy, London Metropolitan University, London N7 8D8, UK
a r t i c l e
i n f o
Article history: Received 10 November 2011 Received in revised form 31 January 2012 Accepted 2 February 2012 Available online 10 February 2012 Keywords: Metal chelate Affinity precipitation Thermo-sensitive PNVCL BSA
a b s t r a c t This investigation involves the metal chelate affinity precipitation of bovine serum albumin (BSA) using a copper ion loaded thermo-sensitive copolymer. The copolymer of N-vinylcaprolactam with methacrylic acid PNVCL-co-MAA was synthesized by free radical polymerization in aqueous solution, and Cu(II) ions were attached to provide affinity properties for BSA. A maximum loading of 48.1 mg Cu2+ per gram of polymer was attained. The influence of pH, temperature, BSA and NaCl concentrations on BSA precipitation and of pH, ethylenediaminetetraacetic acid (EDTA) and NaCl concentrations on elution were systematically probed. The optimum conditions for BSA precipitation occurred when pH, temperature and BSA concentration were 6.0, 10 ◦ C and 1.0 mg/ml, respectively and the most favorable elution conditions were at pH 4.0, with 0.2 M NaCl and 0.06 M EDTA. The maximum amounts of BSA precipitation and elution were 37.5 and 33.7 mg BSA/g polymer, respectively. It proved possible to perform multiple precipitation/elution cycles with a minimal loss of polymer efficacy. The results show that PNVCL-co-MAA is a suitable matrix for the purification of target proteins from unfractionated materials. © 2012 Elsevier B.V. All rights reserved.
1. Introduction The selective isolation and purification of proteins by affinity chromatography using highly specific ligands has proven to be highly successful, but also complicated in terms of scale-up, and sample-treatment, suffering from low throughput of the products at high cost [1]. Immobilized metal affinity chromatography (IMAC) uses chelating compounds bound to polymeric supports to immobilize metal ions, and the latter act as affinity ligands for various proteins [2,3]. The technique has many advantages over other systems in terms of ligand stability, binding capacity, protein recovery and matrix regeneration [4]. Metal chelate affinity precipitation is a nonchromatographic process which combines specific affinity with precipitation [5,6]. The application of affinity precipitation often employs “intelligent” or “smart” polymers [7]. These materials can undergo reversible transitions between water soluble and insoluble phases in response to changes in environmental conditions (temperature, pH, light or solvent). Thermo-sensitive polymers have been extensively used in bioseparation and bioprocessing. They can very easily
∗ Corresponding author. Tel.: +86 21 67792659; fax: +86 21 67792655. E-mail address: [email protected] (L.-M. Zhu). 0927-7765/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2012.02.004
be separated from aqueous solutions by heating above the lower critical solution temperature (LCST) of the polymer [8–10]. Poly(Nvinylcaprolactam) (PNVCL), a water-soluble thermo-responsive polymer, has a LCST of about 35 ◦ C, and so lies in the physiological temperature range [11]. Galaev and Mattiasson [12,13] have studied the use of PNVCL in affinity precipitation with promising results reported. PNVCL is stable to hydrolysis, non-toxic, and has potential applications in bioseparation, medicine and pharmacology [14]. To ensure that precipitation is both predictable and selective, affinity ligands should be attached to the polymer chain [5]. Zinc and copper ions are known to form stable complexes with histidine and cysteine amino acids in proteins [15]. Hence, they could be used to enhance the strength of interactions between a polymer and protein. PNVCL has no reactive groups that could be used to couple with affinity ligands, and hence reactive groups need to be introduced through copolymerization of N-vinylcaprolactam (NVCL) with monomers containing reactive groups (e.g. methacrylic acid) [16]. Galaev et al. [17] copolymerized NVCL with N-vinylimidazole for metal ion loading. In the present study, we have synthesized a thermosensitive copolymer, poly(N-vinylcaprolactam-co-methacrylic acid) (PNVCL-co-MAA), comprising NVCL and methacrylic acid (MAA). MAA contains carboxylic groups, which were used to load copper ions onto the polymer. The cation-loaded copolymer was used as an affinity macroligand for the precipitation of bovine serum albumin (BSA). This affinity precipitation process with
282
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
thermo-sensitive copolymers is found to be a system with the ability to significantly enhance bioseparation. 2. Material and methods 2.1. Chemicals and reagents N-vinylcaprolactam (NVCL) and bovine serum albumin were purchased from the Sigma Chemical Corp., Shanghai, China. Methacrylic acid, ammonium persulfate (APS), and copper sulfate were obtained from Sinopharm Chemical Reagent Co., Ltd., Shanghai, China. All chemicals were analytic grade reagents, and were used without purification. 2.2. Synthesis of PNVCL-co-MAA An aqueous solution of NVCL and MAA was used for copolymerization. A reaction mixture containing different amounts of the two monomers was degassed under vacuum for 30 min. APS (0.04 g) was then added as an initiator and the pH was adjusted to 7.0 with NaOH. The reaction was allowed to proceed under a nitrogen atmosphere for 3 h at 65 ◦ C. The supernatant was removed and the precipitate dissolved in water. This process was repeated three times, with 0.2 M NaCl being added for the final precipitation phase (see Fig. 1a). 2.3. Polymer characterization Fourier transform infrared spectroscopy (FT-IR) spectra of PNVCL-co-MAA, NVCL and MAA were obtained using a FT-IR spectrophotometer (NEXUS-670, Thermo Nicolet Corp., USA). The LCST of the copolymer for a 5.0 mg/ml aqueous solution was determined at 540 nm using a UV–vis spectrophotometer (Lambda 35, PerkinElmer Inc., USA) by heating from 0 to 50 ◦ C at a heating rate of 1 ◦ C/min. The molecular weight of the polymer was determined from viscosity measurements [18]. 2.4. Loading of copper Cu2+ loading onto PNVCL-co-MAA was carried out by adding copper sulfate solution to 5 ml of a 10% (w/v) PNVCL-co-MAA copolymer aqueous solution [4,7]. CuSO4 (10 ml) was added slowly under stirring at room temperatures. The cation-loaded copolymers were stirred for 1 h and then precipitated by adding 2 ml 0.2 M NaCl and heating at 50 ◦ C for 15 min. The supernatant was decanted, and the precipitates dissolved in water. This procedure was repeated 3 times, and finally the copolymers were dissolved in water to give a 2% (w/v) solution. A summary of the process is provided in Fig. 1b. The metal ion loading was calculated from inductively coupled plasma optical emission spectroscopy data (Prodigy, Leeman, USA). 2.5. Affinity precipitation Protein affinity precipitation experiments were carried out by adding 0.5 g PNVCL-co-MAA-Cu to aqueous BSA solutions (0.25, 0.5, 1.0, 1.5, 2.0 or 4.0 mg/ml) to give a final copolymer content of 2.5% (w/v). Precipitation of BSA was studied at various ionic strengths (in 0, 0.1, 0.2, 0.3, 0.4, 0.5 or 0.6 M NaCl solution) and pH values (4.0, 5.0, 6.0, 7.0, or 8.0) in a 0.05 M Tris–HCl buffer. Affinity precipitation experiments were conducted for 1 h at different temperatures (0, 10, 20 or 30 ◦ C), and were followed by a heat treatment at 40 ◦ C for 10 min (to raise the polymer above its LCST and render it insoluble). The amount of precipitated BSA was determined by measuring the initial and final concentrations of BSA in the adsorption medium
using UV spectroscopy at 280 nm (UV-2102PC spectrophotometer, Unico, China)[19]. The amount of precipitated BSA was calculated using Eq. (1): qp =
(Ci − Ct )Vs m
(1)
where qp is the amount of BSA precipitated per unit mass of copolymer (mg/g), Ci and Ct are the concentrations of BSA in the solutions before and after precipitation (mg/ml), Vs is the volume of BSA solution (ml), and m is the mass of the copolymer (g). 2.6. Elution profiling Elution was carried out after precipitation of BSA at pH 6.0, with 1.0 mg/ml BSA and at 10 ◦ C. The copolymer/protein complex was added to a 0.05 M Tris–HCl buffer to extract the precipitated protein. Experiments were performed at different pH values at 4 ◦ C. Thereafter, various concentrations of NaCl and ethylenediaminetetraacetic acid (EDTA) were added. The solution was heated at 40 ◦ C for 10 min and the supernatant withdrawn and analyzed by UV spectroscopy. The amount of eluted BSA was calculated using Eq. (2): qe =
(Ci − Cw )Vs m
(2)
where qe is the amount of BSA eluted per unit mass of copolymer (mg/g), Ci and Cw are the concentrations of BSA in the solutions before and after elution (mg/ml), Vs is the volume of BSA solution (ml), and m is the mass of the copolymer (g). 2.7. Recycling of PNVCL-co-MAA-Cu To determine the reusability of the copolymer, the precipitation and elution cycle was repeated five times using the optimal conditions for each process. The precipitation of BSA was at pH 6.0, 10 ◦ C, with 1.0 mg/ml. Elution was carried out after precipitation at pH 4.0, 0.2 M NaCl and 0.06 M EDTA. Since the elution buffer contained the chelating agent EDTA, which would have caused some copper ions to be removed from the copolymer during recycling, Cu2+ was supplemented by adding copper sulfate solution (pH 5.0, containing 2.0 mg/ml Cu2+ ) before each precipitation process. The summary of metal-chelate affinity precipitation including the process of precipitation, elution and recycling is provided in Fig. 2. 3. Results and discussion 3.1. Synthesis and characterization of PNVCL-co-MAA PNVCL-co-MAA was synthesized using free radical polymerization of NVCL and MAA, with APS acting as an initiator. Fig. 1a presents the reaction scheme for the introduction of carboxylic groups using MAA and Table 1 shows the various conditions investigated for the synthesis of PNVCL-co-MAA. To confirm the successful preparation of PNVCL-co-MAA, FT-IR and viscosity measurements were carried out. Fig. 3 shows the FT-IR spectra of NVCL (Fig. 3a), MAA (Fig. 3b), and PNVCL-co-MAA (Fig. 3c). The C C stretching band at 1654 cm−1 in the spectrum of the NVCL monomer and at 1641 cm−1 in the spectrum of MAA monomer disappeared in PNVCL-co-MAA. The amide I and II band of NVCL were observed in the spectrum of PNVCL-co-MAA at 1626 and 1481 cm−1 , respectively. In addition to the NVCL bands, an absorption band is visible at 1709 cm−1 in the PNVCL-co-MAA spectrum. This corresponds to the stretching of C O in carboxylic acids. By changing the MAA: NVCL molar ratio as listed in Table 1, various PNVCL-co-MAA products with different molecular weights
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
283
Fig. 1. Reaction schemes for the synthesis of (a) the polymer PNVCL-co-MAA; and, (b) metal chelated PNVCL-co-MAA-Cu.
Fig. 2. Diagram of metal-chelate affinity precipitation using PNVCL-co-MAA.
(MW) were obtained. The MW of PNVCL-co-MAA increased with an increasing MAA: NVCL molar ratio initially due to low steric hindrance between NVCL and MAA, but decreased when the ratio rose to 0.20:1. This is possibly due to the high MAA content in the latter copolymer.
As PNVCL is a thermo-sensitive polymer exhibiting LCST in aqueous solution, PNVCL-co-MAA was also expected to possess this property. The LCST of the water-soluble polymer solutions was determined using the temperature at which a sharp decrease in transmittance measured by a spectrophotometer at 540 nm was
Table 1 Synthesis conditions, molecular weight, recovery, and LCST of PNVCL-co-MAA. Sample code
NVCL:MAA:APS (mol ratio)
Molecular weight (Da)a
Recovery (%)
LCSTb (◦ C)
PNVCL-co-MAA(1) PNVCL-co-MAA(2) PNVCL-co-MAA(3) PNVCL-co-MAA(4) PNVCL-co-MAA(5)
1:0.04:0.0244 1:0.08:0.0244 1:0.12:0.0244 1:0.16:0.0244 10.20:0.0244
4.95 × 105 5.76 × 105 6.13 × 105 6.76 × 105 6.43 × 105
86.13 87.47 88.23 88.31 86.13
16.6 21.7 25.5 32.1 27.9
a b
Determined by viscosity method. The LCST was defined as the temperature where the transmittance was 50%.
± ± ± ± ±
0.2 0.2 0.2 0.2 0.2
284
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
the ionization of carboxylic groups, leading to electrostatic attraction between copolymers and Cu2+ ion, which helps the loading of copper ions. The decline was found when pH was higher than 5.0 due to the decrease in solubility of copper ion. Therefore, the optimal conditions for effective coupling were at pH 4.5–5.5. It was found that the maximum Cu2+ loading was 48.1 mg/g at pH 5.0, indicating that it could be effectively coupled to PNVCL-co-MAA. Fig. 5b shows the influence of the initial concentration of copper on its coupling to the copolymer. When the concentration was increased from 0.5 to 10.0 mg/ml, the Cu2+ densities rose from 9.5 to 49.6 mg/g. Further increases in copper concentration caused a decline in the loading efficacy as a result of the fixed number of binding sites on the copolymer and high ion concentration decreasing the solubility of PNVCL-co-MAA. The optimal copper ion concentration of metal ion loading was determined to be 2.0 mg/ml. 3.3. Affinity precipitation Fig. 3. FT-IR spectra of (a) NVCL; (b) MAA; (c) PNVCL-co-MAA.
Fig. 4. Influence of temperature on the transmittance at 540 nm of PNVCL-co-MAA with different contents of NVCL and MAA. Polymers with NVCL: MAA ratios of 1:0.04 (), 1:0.08 (䊉), 1:0.12 (), 1:0.16 (), 1:0.20 () were explored.
observed (see Fig. 4). Different LCSTs were obtained owing to the varying contents of MAA in the PNVCL-co-MAA macromolecule. The LCST generally rises with increasing MAA content as the balance between hydrophilic and hydrophobic groups altered [20]. It should be noted that even polymers with a high content of the hydrophilic monomer (MAA) exhibit clear transmittance changes. PNVCL-co-MAA (4) was used for further studies as it was felt to be the most effective polymer for affinity precipitation owing to its relatively high LCST. 3.2. Metal ion loading Cu2+ ions can be easily linked to the polymer by the addition of copper sulfate to PNVCL-co-MAA. The reaction scheme for copper cation-loading is given in Fig. 1b. A range of various conditions for metal ion loading were explored, and are shown in Fig. 5. The pH is a critical parameter for ion loading, since it affects the specificity of copper binding, surface charge, concentration of counter ions on the copolymer, and the degree of ionization of the adsorbate during the reaction. The influence of pH on the coupling of Cu2+ to the copolymer is depicted in Fig. 5a. The coupling efficiency initially increases with pH because of
For affinity precipitation of BSA, various factors (pH, temperature, concentrations of NaCl and BSA) were investigated. Fig. 6a illustrates the effect of pH on the precipitation of BSA. The maximum amount of protein was precipitated at pH 6.0. Changes in the charges at the surface of the protein occur with varying pH. Under acidic conditions the specific binding between metal ions and protein molecules increases with rising pH because of the deprotonation of histidine [21,22]. However, in alkaline conditions the copper ion on the copolymer will precipitate leading to the decrease in the amount of BSA absorbed in polymer. The optimal value of pH for affinity precipitation was pH 6.0. As depicted in Fig. 4, PNVCL-co-MAA displays a LCST in aqueous solution, and hence its solubility will be affected by temperature. This parameter was investigated, and the results shown in Fig. 6b. These experiments demonstrated that the amount of precipitated BSA increased as the temperature rose from 0 to 10 ◦ C, and then declined with further increases in temperature. Therefore, the optimum temperature for affinity precipitation appears to be 10 ◦ C. Although the LCST is greater than 10 ◦ C, the balance between hydrophilic and hydrophobic groups in the copolymer changes prior to the onset of LCST. The effect of NaCl concentration on affinity precipitation of BSA was also studied, and the results are presented in Fig. 6c. A decline in BSA precipitation occurs in response to an increasing NaCl concentration. This reduction in affinity precipitation arises as a result of a number of facts. First, at higher NaCl concentrations, ions shield the surface charge of the protein, resulting in reduced electrostatic interactions between charged protein groups and the copper loaded onto the polymer. Second, the rise in salt concentration will affect the hydrophobic interaction between proteins [23]. The initial BSA concentration was also studied, this is an important consideration because of the fixed number of binding sites on the copolymer. As shown in Fig. 6d, the maximum BSA precipitation level reached at 1.0 mg/ml initial BSA concentration. The binding of protein was at 37.5 mg BSA per gram of PNVCL-co-MAA-Cu. Increasing the BSA concentration further did not result in increased binding owing to saturation of the binding sites on the copolymer. Therefore, the optimized conditions for BSA precipitation were observed when the pH, temperature and BSA concentration were 6.0, 10 ◦ C and 1.0 mg/ml. These optimized conditions were subsequently adopted in BSA elution studies. 3.4. Elution The influence of three factors (pH, NaCl concentration and EDTA concentration) on the elution of BSA from protein/copolymer composite was investigated. EDTA competitively binds to copper ions. Since the interaction between EDTA and Cu2+ is much stronger
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
285
Fig. 5. The effect of processing parameters on Cu2+ loading (mean ± SD, n = 3): (a) pH (2.0 mg/ml Cu2+ at room temperature); and (b) the initial concentration of Cu2+ (pH 5.0 at room temperature). The ligand density is the amount of Cu2+ (mg) loaded per gram of PNVCL-co-MAA.
than that between BSA and the cation, the protein molecules can be effectively eluted by adding EDTA [24]. Fig. 7a confirms this: the amount of eluted BSA increases with EDTA concentration. The maximum BSA elution occurs with 60 mM EDTA. The influence of pH on BSA elution is quite different to that in affinity precipitation (see Fig. 7b). The amount of eluted protein decreases as pH increases. The pH profoundly affects the surface charge of the protein. At low pH conditions, the protein surface
carries a positive charge, leading to electrostatic repulsion between the proteins and the Cu2+ ion, which helps the elution of protein. An increase in pH decreases the electrostatic repulsion, and therefore the BSA was harder to elute. The pH yielding optimal elution of BSA was determined to be 4.0. A popular approach to achieve effective elution is to change the ionic strength of the medium. The data in Fig. 7c show that the amount of eluted BSA rises in response to increasing NaCl
Fig. 6. The influence of different parameters on BSA precipitation (mean ± SD, n = 3): (a) pH (10 ◦ C, 1.0 mg/ml BSA); (b) temperature (pH 6.0, 1.0 mg/ml BSA); (c) NaCl (pH 6.0, 10 ◦ C, 1.0 mg/ml BSA); and (d) initial BSA concentration (pH 6.0, 10 ◦ C).
286
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
Fig. 7. The influence of environmental conditions on BSA elution (mean ± SD, n = 3): (a) EDTA concentration (pH 4.0, 0.2 M NaCl); (b) pH (0.06 M EDTA, 0.2 M NaCl); and (c) NaCl concentration (0.06 M EDTA, pH 4.0).
concentration. Solvated Na+ and Cl− ions reduce the electrostatic interactions between protein molecules and the Cu2+ ions on the copolymer, and hence the amount of eluted BSA increases. However, when the amount of NaCl was greater than 0.2 M, the hydrophobic interactions between copolymer molecules increased, leading to condensation and reducing the amount of eluted BSA.
The optimal NaCl concentration for elution was hence determined to be 0.20 mol/l. 3.5. Recycling of PNVCL-co-MAA-Cu In order to demonstrate the ability to reuse PNVCL-co-MAA and to investigate its stability, the precipitation and elution cycle was repeated five times using the optimal conditions for both precipitation (pH 6.0, 10 ◦ C, 1.0 mg/ml BSA) and elution (pH 4.0, 0.2 M NaCl, 0.06 M EDTA). Fig. 8 outlines the results of performing the precipitation/elution cycle multiple times. There was a slight reduction in the BSA adsorption capacity after 5 cycles. This may be caused by degradation of the polymer. However, the degradation is slight and the polymer remains relatively stable for at least 3 precipitation/elution cycles. This suggests that binding to the active sites in PNVCL-co-MAA-Cu is reversible, and the polymer can be recycled, thereby providing a highly effective system for BSA precipitation. 4. Conclusions
Fig. 8. The outcome of multiple BSA precipitation/elution cycles using PNVCL-coMAA-Cu (mean ± SD, n = 3). Precipitation conditions: 1.0 mg/ml BSA solution, pH 6.0, 10 ◦ C; elution conditions: pH 4.0, 0.2 M NaCl and 0.06 M EDTA.
In this investigation, a thermo-sensitive copolymer of Nvinylcaprolactam and methacrylic acid (PNVCL-co-MAA) was copolymerized and subsequently loaded with Cu2+ . After metal ion loading, the copolymer was used as an affinity macroligand for the precipitation of BSA. The parameters that affect the affinity precipitation and elution of BSA were evaluated. The results indicated that the addition of NaCl decreased the amount of precipitated protein, and the optimal conditions for BSA precipitation
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
were with pH, temperature and BSA concentration at 6.0, 10 ◦ C and 1.0 mg/ml respectively. For the elution of the polymer-bound BSA, the optimum conditions are pH 4.0, 0.2 M NaCl, and 0.06 M EDTA. Under these conditions, maximum amounts of BSA precipitation and elution of 37.5 mg/g and 33.7 mg/g respectively were achieved. Moreover, the copper-loaded copolymer can be effectively recycled for multiple precipitation/elution uses. These results suggest that the thermo-sensitive copolymer synthesized here could be developed further for larger scale protein preparations. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (21006010) and the Fundamental Research Funds for the Central Universities. The work was also supported in part by the UK–China Joint Laboratory for Therapeutic Textiles (based in Donghua University), the Biomedical Textile Materials “111 Project” of the Ministry of Education of P.R. China (no. B07024) and the Research Fund for the Doctoral Program of Higher Education (no. 2009007512001). References [1] F.H. Arnold, J.J. Chalmers, M.S. Saunders, M.S. Croughan, H.W. Blanch, C.R. Wilke, ACS Symp. Ser. 271 (1985) 113–122.
287
[2] V. Gaberc-Porekar, V. Menart, J. Biochem. Bioph. Methods 49 (2001) 335–360. [3] A.J. Fatiadi, Crit. Rev. Anal. Chem. 18 (1987) 1–44. [4] A. Kumar, P.O. Wahlund, C. Kepka, I.Y. Galaev, B. Mattiasson, Biotechnol. Bioeng. 84 (2003) 494–503. [5] B. Mattiasson, A. Kumar, A.E. Ivanov, I.Y. Galaev, Nat. Protoc. 2 (2007) 213–220. [6] H.X. Wang, J.F. Wan, X.J. Cao, Sep. Purif. Technol. 68 (2009) 172–177. [7] A. Kumar, A.A.M. Khalil, I.Y. Galaev, B. Mattiasson, Enzyme Microb. Technol. 33 (2003) 113–117. [8] A. Kumar, A. Srivastava, I.Y. Galaev, B. Mattiasson, Prog. Polym. Sci. 32 (2007) 1205–1237. [9] A.S. Hoffman, P.S. Stayton, Prog. Polym. Sci. 32 (2007) 922–932. [10] A.S. Hoffman, P.S. Stayton, T. Shimoboji, G.H. Chen, Z.L. Ding, A. Chilkoti, C. Long, M. Miura, J.P. Chen, T. Park, N. Monji, C.A. Cole, J.M. Harris, K. Nakamae, Macromol. Symp. 118 (1997) 553–563. [11] A.S. Hoffman, T. Inoue, G.H. Chen, K. Nakamae, Polym. Gels Networks 5 (1997) 561–575. [12] I.Y. Galaev, B. Mattiasson, Biotechnol. Bioeng. 41 (1993) 1101–1106. [13] I.Y. Galaev, B. Mattiasson, Biotechnol. Technol. 6 (1992) 353–358. [14] L.Z. Meng, L.F. Zhang, Y. Liang, Polym. Adv. Technol. 21 (2010) 720–725. [15] J. Porath, J. Carlsson, I. Olsson, G. Belfrage, Nature 258 (1975) 598–599. [16] F. Garret-Flaudy, R. Freitag, J. Polym. Sci. Polym. Chem. 38 (2000) 4218–4229. [17] I.Y. Galaev, A. Kumar, R. Agarwal, M.N. Gupta, B. Mattiasson, Appl. Biochem. Biotechnol. 68 (1997) 121–133. [18] W.H. Jiang, S.J. Han, J. Polym. Sci. Polym. Phys. 36 (1998) 1275–1281. [19] L.M. Zhu, Z.Y. He, B.W. Christopher, Y.T. Zhou, H.L. Nie, Sep. Sci. Technol. 45 (2010) 525–534. [20] E.J. Lee, Y.H. Kim, Fiber Polym. 11 (2010) 164–169. [21] H. Fitzpatrick, P.F. Luckham, S. Eriksen, K. Hammond, Colloids Surf. 65 (1992) 43–49. [22] D. Gao, D.Q. Lin, S.J. Yao, Chem. Eng. Sci. 61 (2006) 7260–7268. [23] A.K. Bajpai, Indian J. Chem. A 36 (1997) 705–707. [24] S.H. Li, J. Wang, M.P. Zhao, J. Sep. Sci. 32 (2009) 3359–3363.
Contents lists available at SciVerse ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Metal chelate affinity precipitation: Purification of BSA using poly(N-vinylcaprolactam-co-methacrylic acid) copolymers Yuan-Qing Ling a , Hua-Li Nie a,b , Christopher Brandford-White c , Gareth R. Williams c , Li-Min Zhu a,∗ a b c
College of Chemistry, Chemical Engineering and Biotechnology, Donghua University, 2999 North Renmin Road, Songjiang University City, Shanghai 201620, PR China Key Laboratory of Textile Science & Technology, Ministry of Education, Donghua University, Shanghai 201620, PR China Institute for Heath Research and Policy, London Metropolitan University, London N7 8D8, UK
a r t i c l e
i n f o
Article history: Received 10 November 2011 Received in revised form 31 January 2012 Accepted 2 February 2012 Available online 10 February 2012 Keywords: Metal chelate Affinity precipitation Thermo-sensitive PNVCL BSA
a b s t r a c t This investigation involves the metal chelate affinity precipitation of bovine serum albumin (BSA) using a copper ion loaded thermo-sensitive copolymer. The copolymer of N-vinylcaprolactam with methacrylic acid PNVCL-co-MAA was synthesized by free radical polymerization in aqueous solution, and Cu(II) ions were attached to provide affinity properties for BSA. A maximum loading of 48.1 mg Cu2+ per gram of polymer was attained. The influence of pH, temperature, BSA and NaCl concentrations on BSA precipitation and of pH, ethylenediaminetetraacetic acid (EDTA) and NaCl concentrations on elution were systematically probed. The optimum conditions for BSA precipitation occurred when pH, temperature and BSA concentration were 6.0, 10 ◦ C and 1.0 mg/ml, respectively and the most favorable elution conditions were at pH 4.0, with 0.2 M NaCl and 0.06 M EDTA. The maximum amounts of BSA precipitation and elution were 37.5 and 33.7 mg BSA/g polymer, respectively. It proved possible to perform multiple precipitation/elution cycles with a minimal loss of polymer efficacy. The results show that PNVCL-co-MAA is a suitable matrix for the purification of target proteins from unfractionated materials. © 2012 Elsevier B.V. All rights reserved.
1. Introduction The selective isolation and purification of proteins by affinity chromatography using highly specific ligands has proven to be highly successful, but also complicated in terms of scale-up, and sample-treatment, suffering from low throughput of the products at high cost [1]. Immobilized metal affinity chromatography (IMAC) uses chelating compounds bound to polymeric supports to immobilize metal ions, and the latter act as affinity ligands for various proteins [2,3]. The technique has many advantages over other systems in terms of ligand stability, binding capacity, protein recovery and matrix regeneration [4]. Metal chelate affinity precipitation is a nonchromatographic process which combines specific affinity with precipitation [5,6]. The application of affinity precipitation often employs “intelligent” or “smart” polymers [7]. These materials can undergo reversible transitions between water soluble and insoluble phases in response to changes in environmental conditions (temperature, pH, light or solvent). Thermo-sensitive polymers have been extensively used in bioseparation and bioprocessing. They can very easily
∗ Corresponding author. Tel.: +86 21 67792659; fax: +86 21 67792655. E-mail address: [email protected] (L.-M. Zhu). 0927-7765/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2012.02.004
be separated from aqueous solutions by heating above the lower critical solution temperature (LCST) of the polymer [8–10]. Poly(Nvinylcaprolactam) (PNVCL), a water-soluble thermo-responsive polymer, has a LCST of about 35 ◦ C, and so lies in the physiological temperature range [11]. Galaev and Mattiasson [12,13] have studied the use of PNVCL in affinity precipitation with promising results reported. PNVCL is stable to hydrolysis, non-toxic, and has potential applications in bioseparation, medicine and pharmacology [14]. To ensure that precipitation is both predictable and selective, affinity ligands should be attached to the polymer chain [5]. Zinc and copper ions are known to form stable complexes with histidine and cysteine amino acids in proteins [15]. Hence, they could be used to enhance the strength of interactions between a polymer and protein. PNVCL has no reactive groups that could be used to couple with affinity ligands, and hence reactive groups need to be introduced through copolymerization of N-vinylcaprolactam (NVCL) with monomers containing reactive groups (e.g. methacrylic acid) [16]. Galaev et al. [17] copolymerized NVCL with N-vinylimidazole for metal ion loading. In the present study, we have synthesized a thermosensitive copolymer, poly(N-vinylcaprolactam-co-methacrylic acid) (PNVCL-co-MAA), comprising NVCL and methacrylic acid (MAA). MAA contains carboxylic groups, which were used to load copper ions onto the polymer. The cation-loaded copolymer was used as an affinity macroligand for the precipitation of bovine serum albumin (BSA). This affinity precipitation process with
282
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
thermo-sensitive copolymers is found to be a system with the ability to significantly enhance bioseparation. 2. Material and methods 2.1. Chemicals and reagents N-vinylcaprolactam (NVCL) and bovine serum albumin were purchased from the Sigma Chemical Corp., Shanghai, China. Methacrylic acid, ammonium persulfate (APS), and copper sulfate were obtained from Sinopharm Chemical Reagent Co., Ltd., Shanghai, China. All chemicals were analytic grade reagents, and were used without purification. 2.2. Synthesis of PNVCL-co-MAA An aqueous solution of NVCL and MAA was used for copolymerization. A reaction mixture containing different amounts of the two monomers was degassed under vacuum for 30 min. APS (0.04 g) was then added as an initiator and the pH was adjusted to 7.0 with NaOH. The reaction was allowed to proceed under a nitrogen atmosphere for 3 h at 65 ◦ C. The supernatant was removed and the precipitate dissolved in water. This process was repeated three times, with 0.2 M NaCl being added for the final precipitation phase (see Fig. 1a). 2.3. Polymer characterization Fourier transform infrared spectroscopy (FT-IR) spectra of PNVCL-co-MAA, NVCL and MAA were obtained using a FT-IR spectrophotometer (NEXUS-670, Thermo Nicolet Corp., USA). The LCST of the copolymer for a 5.0 mg/ml aqueous solution was determined at 540 nm using a UV–vis spectrophotometer (Lambda 35, PerkinElmer Inc., USA) by heating from 0 to 50 ◦ C at a heating rate of 1 ◦ C/min. The molecular weight of the polymer was determined from viscosity measurements [18]. 2.4. Loading of copper Cu2+ loading onto PNVCL-co-MAA was carried out by adding copper sulfate solution to 5 ml of a 10% (w/v) PNVCL-co-MAA copolymer aqueous solution [4,7]. CuSO4 (10 ml) was added slowly under stirring at room temperatures. The cation-loaded copolymers were stirred for 1 h and then precipitated by adding 2 ml 0.2 M NaCl and heating at 50 ◦ C for 15 min. The supernatant was decanted, and the precipitates dissolved in water. This procedure was repeated 3 times, and finally the copolymers were dissolved in water to give a 2% (w/v) solution. A summary of the process is provided in Fig. 1b. The metal ion loading was calculated from inductively coupled plasma optical emission spectroscopy data (Prodigy, Leeman, USA). 2.5. Affinity precipitation Protein affinity precipitation experiments were carried out by adding 0.5 g PNVCL-co-MAA-Cu to aqueous BSA solutions (0.25, 0.5, 1.0, 1.5, 2.0 or 4.0 mg/ml) to give a final copolymer content of 2.5% (w/v). Precipitation of BSA was studied at various ionic strengths (in 0, 0.1, 0.2, 0.3, 0.4, 0.5 or 0.6 M NaCl solution) and pH values (4.0, 5.0, 6.0, 7.0, or 8.0) in a 0.05 M Tris–HCl buffer. Affinity precipitation experiments were conducted for 1 h at different temperatures (0, 10, 20 or 30 ◦ C), and were followed by a heat treatment at 40 ◦ C for 10 min (to raise the polymer above its LCST and render it insoluble). The amount of precipitated BSA was determined by measuring the initial and final concentrations of BSA in the adsorption medium
using UV spectroscopy at 280 nm (UV-2102PC spectrophotometer, Unico, China)[19]. The amount of precipitated BSA was calculated using Eq. (1): qp =
(Ci − Ct )Vs m
(1)
where qp is the amount of BSA precipitated per unit mass of copolymer (mg/g), Ci and Ct are the concentrations of BSA in the solutions before and after precipitation (mg/ml), Vs is the volume of BSA solution (ml), and m is the mass of the copolymer (g). 2.6. Elution profiling Elution was carried out after precipitation of BSA at pH 6.0, with 1.0 mg/ml BSA and at 10 ◦ C. The copolymer/protein complex was added to a 0.05 M Tris–HCl buffer to extract the precipitated protein. Experiments were performed at different pH values at 4 ◦ C. Thereafter, various concentrations of NaCl and ethylenediaminetetraacetic acid (EDTA) were added. The solution was heated at 40 ◦ C for 10 min and the supernatant withdrawn and analyzed by UV spectroscopy. The amount of eluted BSA was calculated using Eq. (2): qe =
(Ci − Cw )Vs m
(2)
where qe is the amount of BSA eluted per unit mass of copolymer (mg/g), Ci and Cw are the concentrations of BSA in the solutions before and after elution (mg/ml), Vs is the volume of BSA solution (ml), and m is the mass of the copolymer (g). 2.7. Recycling of PNVCL-co-MAA-Cu To determine the reusability of the copolymer, the precipitation and elution cycle was repeated five times using the optimal conditions for each process. The precipitation of BSA was at pH 6.0, 10 ◦ C, with 1.0 mg/ml. Elution was carried out after precipitation at pH 4.0, 0.2 M NaCl and 0.06 M EDTA. Since the elution buffer contained the chelating agent EDTA, which would have caused some copper ions to be removed from the copolymer during recycling, Cu2+ was supplemented by adding copper sulfate solution (pH 5.0, containing 2.0 mg/ml Cu2+ ) before each precipitation process. The summary of metal-chelate affinity precipitation including the process of precipitation, elution and recycling is provided in Fig. 2. 3. Results and discussion 3.1. Synthesis and characterization of PNVCL-co-MAA PNVCL-co-MAA was synthesized using free radical polymerization of NVCL and MAA, with APS acting as an initiator. Fig. 1a presents the reaction scheme for the introduction of carboxylic groups using MAA and Table 1 shows the various conditions investigated for the synthesis of PNVCL-co-MAA. To confirm the successful preparation of PNVCL-co-MAA, FT-IR and viscosity measurements were carried out. Fig. 3 shows the FT-IR spectra of NVCL (Fig. 3a), MAA (Fig. 3b), and PNVCL-co-MAA (Fig. 3c). The C C stretching band at 1654 cm−1 in the spectrum of the NVCL monomer and at 1641 cm−1 in the spectrum of MAA monomer disappeared in PNVCL-co-MAA. The amide I and II band of NVCL were observed in the spectrum of PNVCL-co-MAA at 1626 and 1481 cm−1 , respectively. In addition to the NVCL bands, an absorption band is visible at 1709 cm−1 in the PNVCL-co-MAA spectrum. This corresponds to the stretching of C O in carboxylic acids. By changing the MAA: NVCL molar ratio as listed in Table 1, various PNVCL-co-MAA products with different molecular weights
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
283
Fig. 1. Reaction schemes for the synthesis of (a) the polymer PNVCL-co-MAA; and, (b) metal chelated PNVCL-co-MAA-Cu.
Fig. 2. Diagram of metal-chelate affinity precipitation using PNVCL-co-MAA.
(MW) were obtained. The MW of PNVCL-co-MAA increased with an increasing MAA: NVCL molar ratio initially due to low steric hindrance between NVCL and MAA, but decreased when the ratio rose to 0.20:1. This is possibly due to the high MAA content in the latter copolymer.
As PNVCL is a thermo-sensitive polymer exhibiting LCST in aqueous solution, PNVCL-co-MAA was also expected to possess this property. The LCST of the water-soluble polymer solutions was determined using the temperature at which a sharp decrease in transmittance measured by a spectrophotometer at 540 nm was
Table 1 Synthesis conditions, molecular weight, recovery, and LCST of PNVCL-co-MAA. Sample code
NVCL:MAA:APS (mol ratio)
Molecular weight (Da)a
Recovery (%)
LCSTb (◦ C)
PNVCL-co-MAA(1) PNVCL-co-MAA(2) PNVCL-co-MAA(3) PNVCL-co-MAA(4) PNVCL-co-MAA(5)
1:0.04:0.0244 1:0.08:0.0244 1:0.12:0.0244 1:0.16:0.0244 10.20:0.0244
4.95 × 105 5.76 × 105 6.13 × 105 6.76 × 105 6.43 × 105
86.13 87.47 88.23 88.31 86.13
16.6 21.7 25.5 32.1 27.9
a b
Determined by viscosity method. The LCST was defined as the temperature where the transmittance was 50%.
± ± ± ± ±
0.2 0.2 0.2 0.2 0.2
284
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
the ionization of carboxylic groups, leading to electrostatic attraction between copolymers and Cu2+ ion, which helps the loading of copper ions. The decline was found when pH was higher than 5.0 due to the decrease in solubility of copper ion. Therefore, the optimal conditions for effective coupling were at pH 4.5–5.5. It was found that the maximum Cu2+ loading was 48.1 mg/g at pH 5.0, indicating that it could be effectively coupled to PNVCL-co-MAA. Fig. 5b shows the influence of the initial concentration of copper on its coupling to the copolymer. When the concentration was increased from 0.5 to 10.0 mg/ml, the Cu2+ densities rose from 9.5 to 49.6 mg/g. Further increases in copper concentration caused a decline in the loading efficacy as a result of the fixed number of binding sites on the copolymer and high ion concentration decreasing the solubility of PNVCL-co-MAA. The optimal copper ion concentration of metal ion loading was determined to be 2.0 mg/ml. 3.3. Affinity precipitation Fig. 3. FT-IR spectra of (a) NVCL; (b) MAA; (c) PNVCL-co-MAA.
Fig. 4. Influence of temperature on the transmittance at 540 nm of PNVCL-co-MAA with different contents of NVCL and MAA. Polymers with NVCL: MAA ratios of 1:0.04 (), 1:0.08 (䊉), 1:0.12 (), 1:0.16 (), 1:0.20 () were explored.
observed (see Fig. 4). Different LCSTs were obtained owing to the varying contents of MAA in the PNVCL-co-MAA macromolecule. The LCST generally rises with increasing MAA content as the balance between hydrophilic and hydrophobic groups altered [20]. It should be noted that even polymers with a high content of the hydrophilic monomer (MAA) exhibit clear transmittance changes. PNVCL-co-MAA (4) was used for further studies as it was felt to be the most effective polymer for affinity precipitation owing to its relatively high LCST. 3.2. Metal ion loading Cu2+ ions can be easily linked to the polymer by the addition of copper sulfate to PNVCL-co-MAA. The reaction scheme for copper cation-loading is given in Fig. 1b. A range of various conditions for metal ion loading were explored, and are shown in Fig. 5. The pH is a critical parameter for ion loading, since it affects the specificity of copper binding, surface charge, concentration of counter ions on the copolymer, and the degree of ionization of the adsorbate during the reaction. The influence of pH on the coupling of Cu2+ to the copolymer is depicted in Fig. 5a. The coupling efficiency initially increases with pH because of
For affinity precipitation of BSA, various factors (pH, temperature, concentrations of NaCl and BSA) were investigated. Fig. 6a illustrates the effect of pH on the precipitation of BSA. The maximum amount of protein was precipitated at pH 6.0. Changes in the charges at the surface of the protein occur with varying pH. Under acidic conditions the specific binding between metal ions and protein molecules increases with rising pH because of the deprotonation of histidine [21,22]. However, in alkaline conditions the copper ion on the copolymer will precipitate leading to the decrease in the amount of BSA absorbed in polymer. The optimal value of pH for affinity precipitation was pH 6.0. As depicted in Fig. 4, PNVCL-co-MAA displays a LCST in aqueous solution, and hence its solubility will be affected by temperature. This parameter was investigated, and the results shown in Fig. 6b. These experiments demonstrated that the amount of precipitated BSA increased as the temperature rose from 0 to 10 ◦ C, and then declined with further increases in temperature. Therefore, the optimum temperature for affinity precipitation appears to be 10 ◦ C. Although the LCST is greater than 10 ◦ C, the balance between hydrophilic and hydrophobic groups in the copolymer changes prior to the onset of LCST. The effect of NaCl concentration on affinity precipitation of BSA was also studied, and the results are presented in Fig. 6c. A decline in BSA precipitation occurs in response to an increasing NaCl concentration. This reduction in affinity precipitation arises as a result of a number of facts. First, at higher NaCl concentrations, ions shield the surface charge of the protein, resulting in reduced electrostatic interactions between charged protein groups and the copper loaded onto the polymer. Second, the rise in salt concentration will affect the hydrophobic interaction between proteins [23]. The initial BSA concentration was also studied, this is an important consideration because of the fixed number of binding sites on the copolymer. As shown in Fig. 6d, the maximum BSA precipitation level reached at 1.0 mg/ml initial BSA concentration. The binding of protein was at 37.5 mg BSA per gram of PNVCL-co-MAA-Cu. Increasing the BSA concentration further did not result in increased binding owing to saturation of the binding sites on the copolymer. Therefore, the optimized conditions for BSA precipitation were observed when the pH, temperature and BSA concentration were 6.0, 10 ◦ C and 1.0 mg/ml. These optimized conditions were subsequently adopted in BSA elution studies. 3.4. Elution The influence of three factors (pH, NaCl concentration and EDTA concentration) on the elution of BSA from protein/copolymer composite was investigated. EDTA competitively binds to copper ions. Since the interaction between EDTA and Cu2+ is much stronger
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
285
Fig. 5. The effect of processing parameters on Cu2+ loading (mean ± SD, n = 3): (a) pH (2.0 mg/ml Cu2+ at room temperature); and (b) the initial concentration of Cu2+ (pH 5.0 at room temperature). The ligand density is the amount of Cu2+ (mg) loaded per gram of PNVCL-co-MAA.
than that between BSA and the cation, the protein molecules can be effectively eluted by adding EDTA [24]. Fig. 7a confirms this: the amount of eluted BSA increases with EDTA concentration. The maximum BSA elution occurs with 60 mM EDTA. The influence of pH on BSA elution is quite different to that in affinity precipitation (see Fig. 7b). The amount of eluted protein decreases as pH increases. The pH profoundly affects the surface charge of the protein. At low pH conditions, the protein surface
carries a positive charge, leading to electrostatic repulsion between the proteins and the Cu2+ ion, which helps the elution of protein. An increase in pH decreases the electrostatic repulsion, and therefore the BSA was harder to elute. The pH yielding optimal elution of BSA was determined to be 4.0. A popular approach to achieve effective elution is to change the ionic strength of the medium. The data in Fig. 7c show that the amount of eluted BSA rises in response to increasing NaCl
Fig. 6. The influence of different parameters on BSA precipitation (mean ± SD, n = 3): (a) pH (10 ◦ C, 1.0 mg/ml BSA); (b) temperature (pH 6.0, 1.0 mg/ml BSA); (c) NaCl (pH 6.0, 10 ◦ C, 1.0 mg/ml BSA); and (d) initial BSA concentration (pH 6.0, 10 ◦ C).
286
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
Fig. 7. The influence of environmental conditions on BSA elution (mean ± SD, n = 3): (a) EDTA concentration (pH 4.0, 0.2 M NaCl); (b) pH (0.06 M EDTA, 0.2 M NaCl); and (c) NaCl concentration (0.06 M EDTA, pH 4.0).
concentration. Solvated Na+ and Cl− ions reduce the electrostatic interactions between protein molecules and the Cu2+ ions on the copolymer, and hence the amount of eluted BSA increases. However, when the amount of NaCl was greater than 0.2 M, the hydrophobic interactions between copolymer molecules increased, leading to condensation and reducing the amount of eluted BSA.
The optimal NaCl concentration for elution was hence determined to be 0.20 mol/l. 3.5. Recycling of PNVCL-co-MAA-Cu In order to demonstrate the ability to reuse PNVCL-co-MAA and to investigate its stability, the precipitation and elution cycle was repeated five times using the optimal conditions for both precipitation (pH 6.0, 10 ◦ C, 1.0 mg/ml BSA) and elution (pH 4.0, 0.2 M NaCl, 0.06 M EDTA). Fig. 8 outlines the results of performing the precipitation/elution cycle multiple times. There was a slight reduction in the BSA adsorption capacity after 5 cycles. This may be caused by degradation of the polymer. However, the degradation is slight and the polymer remains relatively stable for at least 3 precipitation/elution cycles. This suggests that binding to the active sites in PNVCL-co-MAA-Cu is reversible, and the polymer can be recycled, thereby providing a highly effective system for BSA precipitation. 4. Conclusions
Fig. 8. The outcome of multiple BSA precipitation/elution cycles using PNVCL-coMAA-Cu (mean ± SD, n = 3). Precipitation conditions: 1.0 mg/ml BSA solution, pH 6.0, 10 ◦ C; elution conditions: pH 4.0, 0.2 M NaCl and 0.06 M EDTA.
In this investigation, a thermo-sensitive copolymer of Nvinylcaprolactam and methacrylic acid (PNVCL-co-MAA) was copolymerized and subsequently loaded with Cu2+ . After metal ion loading, the copolymer was used as an affinity macroligand for the precipitation of BSA. The parameters that affect the affinity precipitation and elution of BSA were evaluated. The results indicated that the addition of NaCl decreased the amount of precipitated protein, and the optimal conditions for BSA precipitation
Y.-Q. Ling et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 281–287
were with pH, temperature and BSA concentration at 6.0, 10 ◦ C and 1.0 mg/ml respectively. For the elution of the polymer-bound BSA, the optimum conditions are pH 4.0, 0.2 M NaCl, and 0.06 M EDTA. Under these conditions, maximum amounts of BSA precipitation and elution of 37.5 mg/g and 33.7 mg/g respectively were achieved. Moreover, the copper-loaded copolymer can be effectively recycled for multiple precipitation/elution uses. These results suggest that the thermo-sensitive copolymer synthesized here could be developed further for larger scale protein preparations. Acknowledgments This work was financially supported by the National Natural Science Foundation of China (21006010) and the Fundamental Research Funds for the Central Universities. The work was also supported in part by the UK–China Joint Laboratory for Therapeutic Textiles (based in Donghua University), the Biomedical Textile Materials “111 Project” of the Ministry of Education of P.R. China (no. B07024) and the Research Fund for the Doctoral Program of Higher Education (no. 2009007512001). References [1] F.H. Arnold, J.J. Chalmers, M.S. Saunders, M.S. Croughan, H.W. Blanch, C.R. Wilke, ACS Symp. Ser. 271 (1985) 113–122.
287
[2] V. Gaberc-Porekar, V. Menart, J. Biochem. Bioph. Methods 49 (2001) 335–360. [3] A.J. Fatiadi, Crit. Rev. Anal. Chem. 18 (1987) 1–44. [4] A. Kumar, P.O. Wahlund, C. Kepka, I.Y. Galaev, B. Mattiasson, Biotechnol. Bioeng. 84 (2003) 494–503. [5] B. Mattiasson, A. Kumar, A.E. Ivanov, I.Y. Galaev, Nat. Protoc. 2 (2007) 213–220. [6] H.X. Wang, J.F. Wan, X.J. Cao, Sep. Purif. Technol. 68 (2009) 172–177. [7] A. Kumar, A.A.M. Khalil, I.Y. Galaev, B. Mattiasson, Enzyme Microb. Technol. 33 (2003) 113–117. [8] A. Kumar, A. Srivastava, I.Y. Galaev, B. Mattiasson, Prog. Polym. Sci. 32 (2007) 1205–1237. [9] A.S. Hoffman, P.S. Stayton, Prog. Polym. Sci. 32 (2007) 922–932. [10] A.S. Hoffman, P.S. Stayton, T. Shimoboji, G.H. Chen, Z.L. Ding, A. Chilkoti, C. Long, M. Miura, J.P. Chen, T. Park, N. Monji, C.A. Cole, J.M. Harris, K. Nakamae, Macromol. Symp. 118 (1997) 553–563. [11] A.S. Hoffman, T. Inoue, G.H. Chen, K. Nakamae, Polym. Gels Networks 5 (1997) 561–575. [12] I.Y. Galaev, B. Mattiasson, Biotechnol. Bioeng. 41 (1993) 1101–1106. [13] I.Y. Galaev, B. Mattiasson, Biotechnol. Technol. 6 (1992) 353–358. [14] L.Z. Meng, L.F. Zhang, Y. Liang, Polym. Adv. Technol. 21 (2010) 720–725. [15] J. Porath, J. Carlsson, I. Olsson, G. Belfrage, Nature 258 (1975) 598–599. [16] F. Garret-Flaudy, R. Freitag, J. Polym. Sci. Polym. Chem. 38 (2000) 4218–4229. [17] I.Y. Galaev, A. Kumar, R. Agarwal, M.N. Gupta, B. Mattiasson, Appl. Biochem. Biotechnol. 68 (1997) 121–133. [18] W.H. Jiang, S.J. Han, J. Polym. Sci. Polym. Phys. 36 (1998) 1275–1281. [19] L.M. Zhu, Z.Y. He, B.W. Christopher, Y.T. Zhou, H.L. Nie, Sep. Sci. Technol. 45 (2010) 525–534. [20] E.J. Lee, Y.H. Kim, Fiber Polym. 11 (2010) 164–169. [21] H. Fitzpatrick, P.F. Luckham, S. Eriksen, K. Hammond, Colloids Surf. 65 (1992) 43–49. [22] D. Gao, D.Q. Lin, S.J. Yao, Chem. Eng. Sci. 61 (2006) 7260–7268. [23] A.K. Bajpai, Indian J. Chem. A 36 (1997) 705–707. [24] S.H. Li, J. Wang, M.P. Zhao, J. Sep. Sci. 32 (2009) 3359–3363.