- Email: [email protected]
Adsorption of bovine serum albumin on nanosized magnetic particles
Journal of Colloid and Interface Science 271 (2004) 277–283 www.elsevier.com/locate/jcis
Adsorption of bovine serum albumin on nanosized magnetic particles Z.G. Peng, K. Hidajat, and M.S. Uddin ∗ Department of Chemical and Environmental Engineering, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260 Received 6 December 2002; accepted 10 December 2003
Abstract Adsorption of bovine serum albumin (BSA) on nanosized magnetic particles (Fe3 O4 ) was carried out in the presence of carbodiimide. The equilibrium and kinetics of the adsorption process were studied. Nanosized magnetic particles (Fe3 O4 ) were prepared by the chemical precipitation method using Fe2+ , Fe3+ salts, and ammonium hydroxide under a nitrogen atmosphere. Characterizations of magnetic particles were carried out using transmission electron microscopy (TEM) and a vibrating sample magnetometer (VSM). Fourier-transform infrared spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS) were used to confirm the attachment of BSA on magnetic particles. Effects of pH and salt concentrations were investigated on the adsorption process. The experimental results show that the adsorption of BSA on magnetic particles was affected greatly by the pH, while the effect of salt concentrations was insignificant at a low concentration range. The adsorption equilibrium isotherm was fitted well by the Langmuir model. The maximum adsorption of BSA on magnetic particles occurred at the isoelectric point of BSA. Adsorption kinetics was analyzed by a linear driving force mass-transfer model. BSA was desorbed from magnetic particles under alkaline conditions, which was confirmed by SDS–PAGE electrophoresis and FTIR results. 2004 Published by Elsevier Inc. Keywords: Nanosized magnetic particles; Bovine serum albumin; Zeta potential; Adsorption; Desorption
1. Introduction Magnetic separation is a recent developing technology and mostly applied in the field of bioseparation. The principle of this method is to utilize magnetic particles to bind the target molecules via ligand to form a complex that can be separated from the bulk solution by magnetic field gradient. Its application includes enzyme immobilization [1,2], cell sorting [3–5], protein adsorption and purification [6,7], nucleic acid detachment [8,9], and drug delivery [10]. Compared to conventional separation, the advantages of magnetic separation are attributed to its speed, accuracy, and simplicity. Many published works focused on the synthesis of micrometer-sized polymer matrix containing magnetic particles and its application in the separation of protein with the aid of a specific ligand coating the surface of the particles [11,12]. Only limited work has been published on the application of nanosized magnetic particles in the separation of proteins. Nanosized magnetic particles can produce larger specific surface areas and, therefore, may result in high ad* Corresponding author.
E-mail address: [email protected] (M.S. Uddin). 0021-9797/$ – see front matter 2004 Published by Elsevier Inc. doi:10.1016/j.jcis.2003.12.022
sorption capacity for proteins. Therefore, it may be useful to synthesize nanosized magnetic particles with large surface areas and utilize them as suitable carriers for the adsorption/desorption of protein. Water-soluble carbodiimide is used to immobilize protein on carboxyl-terminated polymer. Chen and Chen [13] used carbodiimide for the immobilization of lysozyme on polymer, while Dilgimen et al. [14] utilized carbodiimide to conjugate BSA on polymer. Recently, carbodiimide was also used to activate the direct adsorption of BSA and alkaline phosphatase [15,16] on magnetic particles. Yoon et al. [17] studied the separation of protein using submicrometersized magnetic particles without ligand and suggested that hydrogen bonding, hydrophobic interaction, and electrostatic repulsion may also promote the adsorption. Bergemann et al. [18] used nanosized (50 or 100 nm) magnetic particles coated with ionically susceptible ligand in bioseparation. Desorption of protein from magnetic particles had been studied either by acid or alkaline washing. Safarik and Safarikova [19] used a 0.01 M HCl solution to desorp lysozyme from magnetic chitin. Khng et al. [11] found that an acetic acid solution containing salt was more efficient than HCl or glycine–HCl for the desorption of trypsin from mag-
278
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
netic particles. Honda et al. [20] used alkaline condition (pH 10–13) for the desorption of recombinant Escherichia coli from chitosan-conjugated magnetite. Yoshida and Kataoka [21] noted that the desorption of BSA from cross-linked chitosan could be done by an alkaline buffer solution. Although there was some previous work carried out on protein adsorption/desorption on large sized magnetic particles, not much work has been published on protein adsorption/desorption on nanosized magnetic particles. In this work, the equilibrium and kinetics of BSA adsorption on nanosized magnetic particles were studied. The effect of pH and salt concentrations on adsorption was investigated in detail. Desorption of BSA was also studied.
2. Materials and methods 2.1. Materials Iron(II) chloride tetrahydrate (99%) was obtained from Fisher (USA). Iron(III) chloride hexahydrate (98%) was obtained from Nacalai Tesque (Japan). Ammonium hydroxide (25%) was purchased from Merck (USA). 1-Ethyl-3-(3dimethylaminopropyl)carbodiimide hydrochloride, bovine serum albumin (BSA), and tris(hydroxymethyl)aminomethane were purchased from Sigma-Aldrich (USA). All the chemicals were used as received without further treatment. The water used in this work was Milli-Q ultrapure water. 2.2. Methods 2.2.1. Preparation of Fe3 O4 magnetic particles Magnetic particles were prepared by the chemical precipitation method [22,23]. A complete precipitation of Fe3 O4 was achieved under alkaline conditions, while maintaining a molar ratio of Fe2+ :Fe3+ = 1:2 under a nonoxidizing environment. To obtain 1 g of Fe3 O4 precipitate, 0.86 g of FeCl2 ·4H2 O and 2.36 g FeCl3 ·6H2 O were dissolved under a N2 atmosphere in 40 ml of deaerated Milli-Q water with vigorous stirring (1000 rpm). As the solution was being heated to 80 ◦ C, 5 ml NH4 OH was added. To ensure the complete growth of the nanoparticle crystals, the reaction was carried out for 30 min at 80 ◦ C under constant stirring. The resulting suspension was cooled down to room temperature and then washed with Milli-Q water to remove unreacted chemicals. Finally, the wet magnetic particles were obtained by draining the water and used for the adsorption experiments. The solid content of the wet particles was measured by freezedrying. 2.2.2. Characterization of magnetic particles Characterizations of the magnetic particles were carried out using TEM for the size measurement and VSM for the magnetization curve. A bright-field TEM (Model JEM2010) was used for the size measurement of the magnetic particles. To prepare the sample for TEM measurement, a
copper grid (200 mesh and covered with formvar/carbon) was coated with a thin layer of diluted magnetic particle suspension. The copper film was then dried at room temperature for 24 h before the measurement. Wet magnetic particles were freeze-dried (Edwards freeze dryer, ESM 1342) for 24 h and then used for VSM measurement (Model 1600, DMS). The zeta potentials of Fe3 O4 magnetic particles at different pH were measured using a Brookhaven Zeta Plus 90 analyzer. Samples were prepared by diluting 20 mg wet magnetic particles in 10−3 M NaNO3 solution at different pH adjusted with diluted HNO3 and NaOH solution. Magnetic particles, BSA, and BSA attached magnetic particles were characterized by X-ray photoelectron spectroscopy (XPS) and Fourier-transform infrared spectroscopy (FTIR). XPS measurements were made on a VG ESCALAB MkII spectrometer with a MgKα X-ray source (1253.6 eV photons) at a constant retard ratio of 40. The samples were mounted on the standard sample studs by means of doublesided adhesive tape. The core-level signals were obtained at a photoelectron take-off angle of 75◦ (with respect to the sample surface). The X-ray source was run at a reduced power of 120 W. The pressure in the analysis chamber was maintained at 7.5 × 10−9 Torr or lower during each measurement. All binding energies (BEs) were referenced to the C1s neutral carbon peak at 284.6 eV. FTIR measurements were performed by Bio-Rad Model 400 using KBr as background. 2.2.3. Adsorption/desorption of protein by magnetic particles Adsorption of BSA on magnetic particles was carried out by mixing 4 ml of BSA solution of certain concentration, 110 mg wet solid magnetic particles, and 1 ml of 2.0 mg/ml freshly prepared carbodiimide solution. The mixture was left in a shaker operating at 150 times/min for 24 h to reach equilibrium. The solid content of the wet magnetic particles was measured to be 13.3%. Effect of pH on adsorption was evaluated in the range of pH 3.36–9.07. Three buffer systems, i.e., 0.01 M sodium acetic acid buffer (pH 3.36, 4.64), 0.01 M mono/disodium phosphate buffer (pH 5.19, 6.15), and 0.01 M tris–HCl buffer (7.25, 8.11, 9.07), were used for the experiments. The unadjusted pH of the mixture of magnetic particles and BSA was about 7.11. Effect of salt concentration was evaluated at different NaCl concentrations (0, 0.05, 0.1, 0.5, 1.0, 1.5 M). BSA concentration in supernatant was measured by UVspectrometer analysis (Shimadzu UV 1601 PC) at 280 nm [17,24]. Therefore, the adsorbed mass of protein can be calculated by mass balance. Adsorption kinetics of BSA on magnetic particles was carried out at pH 4.64 for two different feed concentrations (0.413 and 1.202 mg/ml). Desorption of BSA from magnetic particles was carried out using 0.5 M Na2 HPO4 (pH 9.35). For the desorption study, feed with a BSA concentration of 0.413 mg/g was used. Once equilibrium was reached for the adsorption, the
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
279
Fig. 1. Transmission electron micrographs of Fe3 O4 magnetic particles.
supernatant was separated from the magnetic particles by the help of a magnet. Then the particles were washed by Milli-Q water. Magnetic particles containing BSA were then mixed with 5 ml 0.5 M Na2 HPO4 solution. After 2 h, the supernatant was collected and analyzed using a UV-spectrometer and SDS–PAGE electrophoresis. The SDS–PAGE gel was stained with Coomassie blue. A sample of the solid particles was collected after the desorption experiment and freezedried for FTIR analysis.
3. Results and discussion 3.1. Characterization of magnetic particles Size and morphology of magnetic particles were characterized by TEM. A typical TEM micrograph of magnetic particles is shown in Fig. 1. It shows that the size of magnetic particles is about 10 nm, which is comparable to the reported value 8.5 nm [23]. It is known that magnetic particles less than about 30 nm will exhibit superparamagnetism [7]. Therefore, the prepared magnetic particles (Fe3 O4 ) have superparamagnetic properties and are expected to respond well to magnetic fields without any permanent magnetization. The superparamagnetic properties of the magnetic particles were also verified by the magnetization curve measured by VSM. A typical plot of magnetization versus applied magnetic field (M–H loop) at 393 K is shown in Fig. 2. The magnetization curve exhibits zero remanence and coercivity, and follows the Langevin function [25], which proves that magnetic particles have superparamagnetic properties. The saturation magnetization of the obtained Fe3 O4 magnetic particles is 76 emu/g Fe3 O4 , which is comparable to the reported magnetization 92 emu/g of bulk Fe3 O4 [26]. This large saturation magnetization of magnetic particles makes them very susceptible to magnetic fields, and therefore makes the solid and liquid phases separate easily.
Fig. 2. Magnetization curve of Fe3 O4 magnetic particles at 393 K.
3.2. Adsorption of BSA Adsorption of BSA on nanosized magnetic particles at different pH and salt concentrations in the presence of carbodiimide was carried out. As a catalyst, carbodiimide is used to activate a carboxyl group in one molecule and a free amino group in another molecule to form peptide bonds. Carbodiimide is used to bind protein directly on magnetic particles. It is proposed that either the OH− [15] or the NH2 group [27] on the surface of the magnetic particles is responsible for protein adsorption via carbodiimide. In the following sections, adsorption equilibrium, adsorption kinetics, FTIR, and XPS spectrum analyses of BSA adsorption were studied. 3.2.1. Effect of pH The effect of pH on BSA adsorption on magnetic particles is shown in Fig. 3. The pH was changed from 3.36 (below the pI of BSA) to 4.64 (very close to the pI of BSA, 4.7), and finally to 9.07 (above the pI of BSA). It shows that pH has a significant effect on the adsorption of BSA on magnetic particles. With the increase of pH from 4.64 to 9.07, the amount of adsorbed BSA on magnetic particles decreased significantly. The maximum adsorption of BSA occurred at pH 4.64, which is close to isoelectric point of BSA, pI = 4.7 [28,29]. When the pH decreased from 4.64 to 3.36, the adsorbed amount of BSA on magnetic particles also decreased. A possible explanation for pH effect on adsorption may be related to the surface charge of magnetic particles and BSA. The measured zeta potentials of magnetic particles in suspension at different pH are shown in Fig. 4. Results show that the isoelectric point of magnetic particles is about 6.8. This value agrees well with the literature value
280
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
Fig. 3. BSA adsorption equilibrium isotherms at different pH.
Fig. 5. Schematic illustration of zeta potential and electrostatic interaction between magnetic particle and BSA at different pH.
Fig. 4. Zeta potential of Fe3 O4 (20 mg/100 ml) in 10−3 M NaNO3 at different pH with/without carbodiimide (carbodiimide concentration 0.2 mg/ml).
pI = 6.5 [30]. Magnetic particles have positive charge below pI and negative charge above pI . The values of zeta potential of magnetic particles are 50 mV at pH 3 and −46 mV at pH 11, respectively. The large value of zeta potential may greatly affect the adsorption of BSA. Fig. 4 also shows that the presence of carbodiimide does not have a significant effect on the zeta potential of magnetic particles. BSA is an amphiphilic protein due to the presence of a NH2 and a COOH group in its molecular structure. It shows a different net charge at different pH media. The isoelectric point of bovine serum albumin is pI = 4.7. It indicates that BSA has a positive charge below pI 4.7 and negative charge above pI 4.7. Peters [31] reported that the net charge of BSA at pH 7 was −18 mV. Fig. 5 shows a schematic illustration of electrostatic interaction between magnetic particles and BSA at different pH media. It can be divided into several regions: (I) less than
pH 4.7, (II) pH 4.7, (III) pH 4.7–6.8, and (IV) greater than pH 6.8. In region I, the electrostatic effect is large due to both magnetic particles and BSA having a positive charge, and therefore, the electrostatic repulsion does not favor the adsorption of BSA on magnetic particles. Similarly, in region IV, both magnetic particles and BSA have a negative charge, also resulting in the decrease of adsorption of BSA on magnetic particles. Yoon et al. [17] reported the electrostatic repulsion between microsphere (negative charge) and BSA (negative charge) at pH 7. In region III, it shows that electrostatic interaction is one of the driving forces for the adsorption of BSA on the surface of magnetic particles. In this region, BSA has a negative charge while magnetic particles have a positive charge, which can promote the adsorption of BSA on magnetic particles. But in this region, the net difference of surface charges of BSA and magnetic particles is comparatively small and, therefore, less adsorption is expected. In region II at pH 4.7, the isoelectric point of BSA, magnetic particles have a positive charge while BSA shows a zero net surface charge. Peters [31] pointed out that BSA has several isomeric forms at different pH media and correspondingly has different α-helix contents, with a maximum α-helix content at its isoelectric point. This means that BSA molecules are in most compact states and result in a minimum intermolecular repulsion, which refers to the higher adsorption amount. Bajpai [32] also found that at isoelectric point pI 4.7, BSA undergoes the minimum conformational change and therefore the adsorption of BSA on glass pow-
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
Fig. 6. Effect of salt (NaCl) concentrations on BSA adsorption at pH 7.11 for two feed concentrations (0.413 and 1.202 mg/ml).
der coated with PVA reached a maximum. Chun and Stroeve [29] also suggested that at the isoelectric point of protein, electrostatic repulsion between the protein and the membrane surface was minimized and subsequently resulted in a larger flux than that at other pH values. In our cases, the experimental results also show that the maximum adsorption of BSA on magnetic particles occurred at pI 4.7. These results confirmed the previous suggestion. 3.2.2. Effect of salt concentration The effect of NaCl concentration on the adsorption of BSA for two different feed concentrations (0.413 and 1.202 mg/ml) was studied at pH 7.11. The results are shown in Fig. 6. Results show that when NaCl concentration was increased from 0 to 0.5 M, there was no significant change on BSA adsorption for 0.413 mg/ml feed solution and in case of 1.202 mg/ml feed solution, there is a slight increase, about 3%, in adsorption. However, when NaCl concentration was increased to 1.5 M, a significant change in adsorption was observed, 20% decrease in the case of 0.413 mg/ml feed and 34% for 1.202 mg/ml feed. Liu et al. [33] proposed that high concentrations of NaCl ions could cover the particle surface and form an ion shield, which can decrease the diffusivity of proteins and enlarge the absorbed proteins molecules, and therefore reduce the protein adsorption. 3.2.3. Adsorption equilibrium of BSA The adsorption equilibrium of BSA was studied at different pH and the results are shown in Fig. 3. The Langmuir
281
Fig. 7. FTIR spectra of (a) BSA, (b) BSA-attached magnetic particles, (c) magnetic particles, and (d) magnetic particles after desorption of BSA.
model is used to fit the experimental data. The Langmuir equation is expressed as Q=
Qm Ka C ∗ , 1 + Ka C ∗
(1)
where C ∗ (mg/ml) and Q (mg/g solid) are BSA concentration in the aqueous solution and the absorbed BSA on the solid at equilibrium, respectively. Qm is the maximum adsorption amount, and Ka (ml/mg) is the adsorption constant. The experimental data are fitted to the Langmuir model through nonlinear regression analysis. The fitted parameters of the model are summarized in Table 1. High R 2 values indicate that the Langmuir model predicts well the adsorption behavior over a wide range of pH. 3.2.4. FTIR and XPS spectra for BSA and magnetic particles FTIR and XPS spectra of BSA, Fe3 O4 , and BSA-adsorbed Fe3 O4 are analyzed to show the BSA adsorption on magnetic particles. The FTIR results are shown in Fig. 7. Spectrum a indicates that the characteristic bands of BSA occur at 1648 and 1540 cm−1 . Similar values were reported by previous work [15,27]. Spectrum c shows the characteristic band of Fe3 O4 at 600 cm−1 . After adsorption of BSA on magnetic particles, spectrum b shows both characteristic bands of BSA at 1648 and 1540 cm−1 and characteristic bands of Fe3 O4 at 600 cm−1 . It indicates that BSA was successfully attached to the surface of magnetic particles. Comparing spectra a,
Table 1 Langmuir model parameters for BSA adsorption isotherms on magnetic particles at different pH pH Qm (mg/g solid) Ka (ml/mg) R2
3.36
4.64
5.19
6.15
7.25
8.11
9.07
245.8 16.2 0.98
418.9 53.1 0.95
322.1 41.4 0.99
307.8 28.2 0.97
296.0 14.6 0.98
251.9 11.4 0.97
113.2 4.69 0.99
282
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
Fig. 9. Adsorption kinetics of BSA on magnetic particles at pH 4.64 for two feed concentrations (0.413 and 1.202 mg/ml). Fig. 8. XPS wide scan spectra of (a) Fe3 O4 , (b) BSA-attached magnetic particles, and (c) BSA.
Table 2 Desorption results of BSA from magnetic particles Initial BSA Adsorbed BSA Desorbed BSA Desorption concentration (mg/ml) (mg/g solid) (mg/g solid) percentage (%)
b, and c, it is found that the chemical interaction between Fe3 O4 and BSA is not significant, since little shift of the IR band due to the binding is observed. Fig. 8 shows the XPS wide scan spectrum of BSA, magnetic particles, and BSA attached on magnetic particles. Spectrum a shows the Fe2p peak of magnetic particles at about 710.3 eV. BSA contains N element, which show the N1s peak at about 399.6 eV in spectrum c. After BSA adsorption on magnetic particles, spectrum b shows both the N1s peak at about 399.6 eV and the Fe2p peak at about 710.3 eV. Therefore, XPS analysis of the surface composition confirmed the adsorption of BSA on magnetic particles. The XPS results agree well with that of FTIR.
Substituting Eq. (4) into Eq. (3), the following differential equation is obtained: (C0 − C)V /S dC . = KL a C − − (5) dt Qm Ka − Ka (C0 − C)V /S
3.2.5. Adsorption kinetics Study on the adsorption kinetics of BSA on magnetic particles was carried out at pH 4.64 at two different feed concentrations (0.413 and 1.202 mg/ml), and the results are shown in Fig. 9. A linear driving force mass-transfer model is used to describe the adsorption kinetics,
3.3. Desorption of BSA from magnetic particles
−
dC = KL a(C − C ∗ ), dt
(2)
where KL a (min−1 ) is the overall mass-transfer coefficient, C (mg/ml) is the concentration of BSA in solution at time t, and C ∗ (mg/ml) is the equilibrium concentration. C ∗ is obtained from Langmuir isotherm Eq. (1) and the resulting equation becomes Q dC = KL a C − . − (3) dt Qm Ka − QKa The mass balance of BSA can be expressed as C0 V = CV + SQ,
(4)
where C0 (mg/ml) is the initial concentration of BSA, V (ml) is the volume of the feed mixture, and S (g) is the total quantity of the solid.
0.413
135.7 ± 1.6
126.2 ± 1.1
92.9 ± 2.0
Equation (5) is integrated numerically using Polymath 4, a standard routine for solving nonlinear ordinary differential equations. The fitted curve and the experimental data are compared in Fig. 9. It is found that the experimental results are fitted well with KL a = 0.025 (min−1 ) for 0.413 mg/ml and KL a = 0.022 (min−1 ) for 1.202 mg/ml, respectively.
The desorption of BSA from magnetic particles was carried out under alkaline conditions of pH 9.35. The adsorbed and desorbed quantities of BSA at equilibrium for feed concentration 0.413 mg/ml are shown in Table 2. Desorption of more than 90% was achieved. SDS–PAGE electrophoresis experiments were carried out for BSA in three different samples, feed solution, and supernatants after adsorption and desorption experiments. The results are shown in Fig. 10. It is found that feed and supernatant from desorption experiments contain BSA bands in SDS–PAGE gel. Whereas the raffinate from the adsorption experiments does not contain any detectable BSA. This confirms the adsorption and subsequent desorption of BSA on magnetic particles. Some extrabands were observed for the desorbed supernatant sample compared to the BSA feed solutions. This could be the protein complexes formed by BSA molecules, which are activated by carbodiimide used in the adsorption process. But compared to the bulk band of BSA, it could be neglected.
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
283
Acknowledgment This work was financially supported by the National University of Singapore Research Fund (R-279-000-085-112).
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] Fig. 10. SDS–PAGE electrophoresis gel stained by Coomassie blue (lane 1, 0.413 mg/ml BSA feed solution; lane 2, supernatant after adsorption experiment; lane 3, supernatant after desorption experiment). Each lane was loaded with 20 µl of sample.
[12]
The FTIR spectrum of magnetic particles after desorption is shown in Fig. 7. It shows the characteristic band of Fe3 O4 at 600 cm−1 . But the absence of a BSA characteristic band confirms the desorption of BSA from magnetic particles.
[15]
[13] [14]
[16]
[17]
4. Conclusions Nanosized magnetic particles Fe3 O4 were prepared by the chemical precipitation method and characterized by TEM and VSM for size and supermagnetic properties. Adsorption of BSA on magnetic particles was carried out in the presence of carbodiimide under different pH and salt concentrations. It has been found that pH has a great effect on BSA adsorption while the effect of salt at a lower concentration range was comparatively insignificant. The maximum adsorption of BSA on magnetic particles occurred at the isoelectric point of BSA. The adsorption equilibrium results are fitted well by the Langmuir model while a linear driving force mass-transfer model is used for fitting kinetic data. FTIR and XPS spectra are used to confirm the attachment of BSA on the magnetic particles. Desorption of BSA was carried out under alkaline conditions, which was confirmed by SDS–PAGE electrophoresis and FTIR results. More than 90% desorption efficiency was achieved.
[18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]
R.F.H. Dekker, Appl. Biochem. Biotechnol. 22 (1989) 289. A. Kondo, H. Fukuda, J. Ferment. Bioeng. 84 (1997) 337. Y. Haik, V. Pai, C.J. Chen, J. Magn. Magn. Mater. 194 (1999) 254. J.P. Hancock, J.T. Kemshead, J. Immunol. Methods 164 (1993) 51. R.S. Molday, L.L. Molday, FEBS Lett. 170 (1984) 232. T. Abudiab, R.R. Beitle, J. Chromatogr. A 795 (1998) 211. S.M. O’Brien, O.R.T. Thomas, P. Dunnill, J. Biotechnol. 50 (1996) 13. P.R. Levison, S.E. Badger, J. Dennis, P. Hathi, M.J. Davies, I.J. Bruce, D. Schimkat, J. Chromatogr. A 816 (1998) 107. M. Uhlen, Nature 340 (1989) 733. A.N. Rusetski, E.K. Ruuge, J. Magn. Magn. Mater. 85 (1990) 299. H.P. Khng, D. Cunliffe, S. Davies, N.A. Turner, E.N. Vulfson, Biotechnol. Bioeng. 60 (1998) 419. M. Suzuki, M. Shinkai, M. Kamihira, T. Kobayashi, Biotechnol. Appl. Biosci. 21 (1995) 335. J.P. Chen, Y.C. Chen, Bioresource Technol. 60 (1997) 231. A.S. Dilgimen, Z. Mustafaeva, M. Demchenko, T. Kaneko, Y. Osada, M. Mustafaev, Biomaterials 22 (2001) 2383. R.V. Mehta, R.V. Upadhyay, S.W. Charles, C.N. Ramchand, Biotechnol. Tech. 11 (1997) 493. M. Koneracka, P. Kopcansky, M. Antalik, M. Timko, C.N. Ramchand, D. Lobo, R.V. Mehta, R.V. Upadhyay, J. Magn. Magn. Mater. 201 (1999) 427. J.Y. Yoon, J.H. Lee, J.H. Kim, W.S. Kim, Colloids Surf. B 10 (1998) 365. C. Bergemann, D. Muller-Schulte, J. Oster, L. a Brassard, A.S. Lubbe, J. Magn. Magn. Mater. 194 (1999) 45. I. Safarik, M. Safarikova, J. Biochem. Biophys. Methods 27 (1993) 327. H. Honda, A. Kawabe, M. Skinkai, T. Kobayashi, Biochem. Eng. J. 3 (1999) 157. H. Yoshida, T. Kataoka, Chem. Eng. J. Biochem. Eng. 41 (1989) B11. A. Wooding, M. Kilner, D.B. Lambrick, J. Colloid Interface Sci. 144 (1991) 236. Y.S. Kang, S. Risbud, J.F. Rabolt, P. Stroeve, Chem. Mater. 8 (1996) 2209. D. Tanyolac, A.R. Ozdural, J. Appl. Polym. Sci. 80 (2001) 707. R.W. Chantrell, J. Popplewell, S.W. Charles, IEEE Trans. Magn. MAG-14 (1978) 975. V.S. Zaitsev, D.S. Filimonov, I.A. Presnyakov, R.J. Gambino, B. Chu, J. Colloid Interface Sci. 212 (1999) 49. D.H. Chen, M.H. Liao, J. Mol. Catal. B 16 (2002) 283. W.K. Lee, J.S. Ko, H.M. Kim, J. Colloid Interface Sci. 246 (2002) 70. K.Y. Chun, P. Stroeve, Langmuir 18 (2002) 4653. I. Iwasaki, S.R.B. Cooke, Y.S. Kim, SME Trans. 223 (1962) 113. T.J. Peters, Adv. Protein Chem. 37 (1985) 161. A.K. Bajpai, J. Appl. Polym. Sci. 78 (2000) 933. H.S. Liu, Y.C. Wang, W.Y. Chen, Colloids Surf. B 5 (1995) 25.
Adsorption of bovine serum albumin on nanosized magnetic particles Z.G. Peng, K. Hidajat, and M.S. Uddin ∗ Department of Chemical and Environmental Engineering, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260 Received 6 December 2002; accepted 10 December 2003
Abstract Adsorption of bovine serum albumin (BSA) on nanosized magnetic particles (Fe3 O4 ) was carried out in the presence of carbodiimide. The equilibrium and kinetics of the adsorption process were studied. Nanosized magnetic particles (Fe3 O4 ) were prepared by the chemical precipitation method using Fe2+ , Fe3+ salts, and ammonium hydroxide under a nitrogen atmosphere. Characterizations of magnetic particles were carried out using transmission electron microscopy (TEM) and a vibrating sample magnetometer (VSM). Fourier-transform infrared spectroscopy (FTIR) and X-ray photoelectron spectroscopy (XPS) were used to confirm the attachment of BSA on magnetic particles. Effects of pH and salt concentrations were investigated on the adsorption process. The experimental results show that the adsorption of BSA on magnetic particles was affected greatly by the pH, while the effect of salt concentrations was insignificant at a low concentration range. The adsorption equilibrium isotherm was fitted well by the Langmuir model. The maximum adsorption of BSA on magnetic particles occurred at the isoelectric point of BSA. Adsorption kinetics was analyzed by a linear driving force mass-transfer model. BSA was desorbed from magnetic particles under alkaline conditions, which was confirmed by SDS–PAGE electrophoresis and FTIR results. 2004 Published by Elsevier Inc. Keywords: Nanosized magnetic particles; Bovine serum albumin; Zeta potential; Adsorption; Desorption
1. Introduction Magnetic separation is a recent developing technology and mostly applied in the field of bioseparation. The principle of this method is to utilize magnetic particles to bind the target molecules via ligand to form a complex that can be separated from the bulk solution by magnetic field gradient. Its application includes enzyme immobilization [1,2], cell sorting [3–5], protein adsorption and purification [6,7], nucleic acid detachment [8,9], and drug delivery [10]. Compared to conventional separation, the advantages of magnetic separation are attributed to its speed, accuracy, and simplicity. Many published works focused on the synthesis of micrometer-sized polymer matrix containing magnetic particles and its application in the separation of protein with the aid of a specific ligand coating the surface of the particles [11,12]. Only limited work has been published on the application of nanosized magnetic particles in the separation of proteins. Nanosized magnetic particles can produce larger specific surface areas and, therefore, may result in high ad* Corresponding author.
E-mail address: [email protected] (M.S. Uddin). 0021-9797/$ – see front matter 2004 Published by Elsevier Inc. doi:10.1016/j.jcis.2003.12.022
sorption capacity for proteins. Therefore, it may be useful to synthesize nanosized magnetic particles with large surface areas and utilize them as suitable carriers for the adsorption/desorption of protein. Water-soluble carbodiimide is used to immobilize protein on carboxyl-terminated polymer. Chen and Chen [13] used carbodiimide for the immobilization of lysozyme on polymer, while Dilgimen et al. [14] utilized carbodiimide to conjugate BSA on polymer. Recently, carbodiimide was also used to activate the direct adsorption of BSA and alkaline phosphatase [15,16] on magnetic particles. Yoon et al. [17] studied the separation of protein using submicrometersized magnetic particles without ligand and suggested that hydrogen bonding, hydrophobic interaction, and electrostatic repulsion may also promote the adsorption. Bergemann et al. [18] used nanosized (50 or 100 nm) magnetic particles coated with ionically susceptible ligand in bioseparation. Desorption of protein from magnetic particles had been studied either by acid or alkaline washing. Safarik and Safarikova [19] used a 0.01 M HCl solution to desorp lysozyme from magnetic chitin. Khng et al. [11] found that an acetic acid solution containing salt was more efficient than HCl or glycine–HCl for the desorption of trypsin from mag-
278
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
netic particles. Honda et al. [20] used alkaline condition (pH 10–13) for the desorption of recombinant Escherichia coli from chitosan-conjugated magnetite. Yoshida and Kataoka [21] noted that the desorption of BSA from cross-linked chitosan could be done by an alkaline buffer solution. Although there was some previous work carried out on protein adsorption/desorption on large sized magnetic particles, not much work has been published on protein adsorption/desorption on nanosized magnetic particles. In this work, the equilibrium and kinetics of BSA adsorption on nanosized magnetic particles were studied. The effect of pH and salt concentrations on adsorption was investigated in detail. Desorption of BSA was also studied.
2. Materials and methods 2.1. Materials Iron(II) chloride tetrahydrate (99%) was obtained from Fisher (USA). Iron(III) chloride hexahydrate (98%) was obtained from Nacalai Tesque (Japan). Ammonium hydroxide (25%) was purchased from Merck (USA). 1-Ethyl-3-(3dimethylaminopropyl)carbodiimide hydrochloride, bovine serum albumin (BSA), and tris(hydroxymethyl)aminomethane were purchased from Sigma-Aldrich (USA). All the chemicals were used as received without further treatment. The water used in this work was Milli-Q ultrapure water. 2.2. Methods 2.2.1. Preparation of Fe3 O4 magnetic particles Magnetic particles were prepared by the chemical precipitation method [22,23]. A complete precipitation of Fe3 O4 was achieved under alkaline conditions, while maintaining a molar ratio of Fe2+ :Fe3+ = 1:2 under a nonoxidizing environment. To obtain 1 g of Fe3 O4 precipitate, 0.86 g of FeCl2 ·4H2 O and 2.36 g FeCl3 ·6H2 O were dissolved under a N2 atmosphere in 40 ml of deaerated Milli-Q water with vigorous stirring (1000 rpm). As the solution was being heated to 80 ◦ C, 5 ml NH4 OH was added. To ensure the complete growth of the nanoparticle crystals, the reaction was carried out for 30 min at 80 ◦ C under constant stirring. The resulting suspension was cooled down to room temperature and then washed with Milli-Q water to remove unreacted chemicals. Finally, the wet magnetic particles were obtained by draining the water and used for the adsorption experiments. The solid content of the wet particles was measured by freezedrying. 2.2.2. Characterization of magnetic particles Characterizations of the magnetic particles were carried out using TEM for the size measurement and VSM for the magnetization curve. A bright-field TEM (Model JEM2010) was used for the size measurement of the magnetic particles. To prepare the sample for TEM measurement, a
copper grid (200 mesh and covered with formvar/carbon) was coated with a thin layer of diluted magnetic particle suspension. The copper film was then dried at room temperature for 24 h before the measurement. Wet magnetic particles were freeze-dried (Edwards freeze dryer, ESM 1342) for 24 h and then used for VSM measurement (Model 1600, DMS). The zeta potentials of Fe3 O4 magnetic particles at different pH were measured using a Brookhaven Zeta Plus 90 analyzer. Samples were prepared by diluting 20 mg wet magnetic particles in 10−3 M NaNO3 solution at different pH adjusted with diluted HNO3 and NaOH solution. Magnetic particles, BSA, and BSA attached magnetic particles were characterized by X-ray photoelectron spectroscopy (XPS) and Fourier-transform infrared spectroscopy (FTIR). XPS measurements were made on a VG ESCALAB MkII spectrometer with a MgKα X-ray source (1253.6 eV photons) at a constant retard ratio of 40. The samples were mounted on the standard sample studs by means of doublesided adhesive tape. The core-level signals were obtained at a photoelectron take-off angle of 75◦ (with respect to the sample surface). The X-ray source was run at a reduced power of 120 W. The pressure in the analysis chamber was maintained at 7.5 × 10−9 Torr or lower during each measurement. All binding energies (BEs) were referenced to the C1s neutral carbon peak at 284.6 eV. FTIR measurements were performed by Bio-Rad Model 400 using KBr as background. 2.2.3. Adsorption/desorption of protein by magnetic particles Adsorption of BSA on magnetic particles was carried out by mixing 4 ml of BSA solution of certain concentration, 110 mg wet solid magnetic particles, and 1 ml of 2.0 mg/ml freshly prepared carbodiimide solution. The mixture was left in a shaker operating at 150 times/min for 24 h to reach equilibrium. The solid content of the wet magnetic particles was measured to be 13.3%. Effect of pH on adsorption was evaluated in the range of pH 3.36–9.07. Three buffer systems, i.e., 0.01 M sodium acetic acid buffer (pH 3.36, 4.64), 0.01 M mono/disodium phosphate buffer (pH 5.19, 6.15), and 0.01 M tris–HCl buffer (7.25, 8.11, 9.07), were used for the experiments. The unadjusted pH of the mixture of magnetic particles and BSA was about 7.11. Effect of salt concentration was evaluated at different NaCl concentrations (0, 0.05, 0.1, 0.5, 1.0, 1.5 M). BSA concentration in supernatant was measured by UVspectrometer analysis (Shimadzu UV 1601 PC) at 280 nm [17,24]. Therefore, the adsorbed mass of protein can be calculated by mass balance. Adsorption kinetics of BSA on magnetic particles was carried out at pH 4.64 for two different feed concentrations (0.413 and 1.202 mg/ml). Desorption of BSA from magnetic particles was carried out using 0.5 M Na2 HPO4 (pH 9.35). For the desorption study, feed with a BSA concentration of 0.413 mg/g was used. Once equilibrium was reached for the adsorption, the
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
279
Fig. 1. Transmission electron micrographs of Fe3 O4 magnetic particles.
supernatant was separated from the magnetic particles by the help of a magnet. Then the particles were washed by Milli-Q water. Magnetic particles containing BSA were then mixed with 5 ml 0.5 M Na2 HPO4 solution. After 2 h, the supernatant was collected and analyzed using a UV-spectrometer and SDS–PAGE electrophoresis. The SDS–PAGE gel was stained with Coomassie blue. A sample of the solid particles was collected after the desorption experiment and freezedried for FTIR analysis.
3. Results and discussion 3.1. Characterization of magnetic particles Size and morphology of magnetic particles were characterized by TEM. A typical TEM micrograph of magnetic particles is shown in Fig. 1. It shows that the size of magnetic particles is about 10 nm, which is comparable to the reported value 8.5 nm [23]. It is known that magnetic particles less than about 30 nm will exhibit superparamagnetism [7]. Therefore, the prepared magnetic particles (Fe3 O4 ) have superparamagnetic properties and are expected to respond well to magnetic fields without any permanent magnetization. The superparamagnetic properties of the magnetic particles were also verified by the magnetization curve measured by VSM. A typical plot of magnetization versus applied magnetic field (M–H loop) at 393 K is shown in Fig. 2. The magnetization curve exhibits zero remanence and coercivity, and follows the Langevin function [25], which proves that magnetic particles have superparamagnetic properties. The saturation magnetization of the obtained Fe3 O4 magnetic particles is 76 emu/g Fe3 O4 , which is comparable to the reported magnetization 92 emu/g of bulk Fe3 O4 [26]. This large saturation magnetization of magnetic particles makes them very susceptible to magnetic fields, and therefore makes the solid and liquid phases separate easily.
Fig. 2. Magnetization curve of Fe3 O4 magnetic particles at 393 K.
3.2. Adsorption of BSA Adsorption of BSA on nanosized magnetic particles at different pH and salt concentrations in the presence of carbodiimide was carried out. As a catalyst, carbodiimide is used to activate a carboxyl group in one molecule and a free amino group in another molecule to form peptide bonds. Carbodiimide is used to bind protein directly on magnetic particles. It is proposed that either the OH− [15] or the NH2 group [27] on the surface of the magnetic particles is responsible for protein adsorption via carbodiimide. In the following sections, adsorption equilibrium, adsorption kinetics, FTIR, and XPS spectrum analyses of BSA adsorption were studied. 3.2.1. Effect of pH The effect of pH on BSA adsorption on magnetic particles is shown in Fig. 3. The pH was changed from 3.36 (below the pI of BSA) to 4.64 (very close to the pI of BSA, 4.7), and finally to 9.07 (above the pI of BSA). It shows that pH has a significant effect on the adsorption of BSA on magnetic particles. With the increase of pH from 4.64 to 9.07, the amount of adsorbed BSA on magnetic particles decreased significantly. The maximum adsorption of BSA occurred at pH 4.64, which is close to isoelectric point of BSA, pI = 4.7 [28,29]. When the pH decreased from 4.64 to 3.36, the adsorbed amount of BSA on magnetic particles also decreased. A possible explanation for pH effect on adsorption may be related to the surface charge of magnetic particles and BSA. The measured zeta potentials of magnetic particles in suspension at different pH are shown in Fig. 4. Results show that the isoelectric point of magnetic particles is about 6.8. This value agrees well with the literature value
280
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
Fig. 3. BSA adsorption equilibrium isotherms at different pH.
Fig. 5. Schematic illustration of zeta potential and electrostatic interaction between magnetic particle and BSA at different pH.
Fig. 4. Zeta potential of Fe3 O4 (20 mg/100 ml) in 10−3 M NaNO3 at different pH with/without carbodiimide (carbodiimide concentration 0.2 mg/ml).
pI = 6.5 [30]. Magnetic particles have positive charge below pI and negative charge above pI . The values of zeta potential of magnetic particles are 50 mV at pH 3 and −46 mV at pH 11, respectively. The large value of zeta potential may greatly affect the adsorption of BSA. Fig. 4 also shows that the presence of carbodiimide does not have a significant effect on the zeta potential of magnetic particles. BSA is an amphiphilic protein due to the presence of a NH2 and a COOH group in its molecular structure. It shows a different net charge at different pH media. The isoelectric point of bovine serum albumin is pI = 4.7. It indicates that BSA has a positive charge below pI 4.7 and negative charge above pI 4.7. Peters [31] reported that the net charge of BSA at pH 7 was −18 mV. Fig. 5 shows a schematic illustration of electrostatic interaction between magnetic particles and BSA at different pH media. It can be divided into several regions: (I) less than
pH 4.7, (II) pH 4.7, (III) pH 4.7–6.8, and (IV) greater than pH 6.8. In region I, the electrostatic effect is large due to both magnetic particles and BSA having a positive charge, and therefore, the electrostatic repulsion does not favor the adsorption of BSA on magnetic particles. Similarly, in region IV, both magnetic particles and BSA have a negative charge, also resulting in the decrease of adsorption of BSA on magnetic particles. Yoon et al. [17] reported the electrostatic repulsion between microsphere (negative charge) and BSA (negative charge) at pH 7. In region III, it shows that electrostatic interaction is one of the driving forces for the adsorption of BSA on the surface of magnetic particles. In this region, BSA has a negative charge while magnetic particles have a positive charge, which can promote the adsorption of BSA on magnetic particles. But in this region, the net difference of surface charges of BSA and magnetic particles is comparatively small and, therefore, less adsorption is expected. In region II at pH 4.7, the isoelectric point of BSA, magnetic particles have a positive charge while BSA shows a zero net surface charge. Peters [31] pointed out that BSA has several isomeric forms at different pH media and correspondingly has different α-helix contents, with a maximum α-helix content at its isoelectric point. This means that BSA molecules are in most compact states and result in a minimum intermolecular repulsion, which refers to the higher adsorption amount. Bajpai [32] also found that at isoelectric point pI 4.7, BSA undergoes the minimum conformational change and therefore the adsorption of BSA on glass pow-
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
Fig. 6. Effect of salt (NaCl) concentrations on BSA adsorption at pH 7.11 for two feed concentrations (0.413 and 1.202 mg/ml).
der coated with PVA reached a maximum. Chun and Stroeve [29] also suggested that at the isoelectric point of protein, electrostatic repulsion between the protein and the membrane surface was minimized and subsequently resulted in a larger flux than that at other pH values. In our cases, the experimental results also show that the maximum adsorption of BSA on magnetic particles occurred at pI 4.7. These results confirmed the previous suggestion. 3.2.2. Effect of salt concentration The effect of NaCl concentration on the adsorption of BSA for two different feed concentrations (0.413 and 1.202 mg/ml) was studied at pH 7.11. The results are shown in Fig. 6. Results show that when NaCl concentration was increased from 0 to 0.5 M, there was no significant change on BSA adsorption for 0.413 mg/ml feed solution and in case of 1.202 mg/ml feed solution, there is a slight increase, about 3%, in adsorption. However, when NaCl concentration was increased to 1.5 M, a significant change in adsorption was observed, 20% decrease in the case of 0.413 mg/ml feed and 34% for 1.202 mg/ml feed. Liu et al. [33] proposed that high concentrations of NaCl ions could cover the particle surface and form an ion shield, which can decrease the diffusivity of proteins and enlarge the absorbed proteins molecules, and therefore reduce the protein adsorption. 3.2.3. Adsorption equilibrium of BSA The adsorption equilibrium of BSA was studied at different pH and the results are shown in Fig. 3. The Langmuir
281
Fig. 7. FTIR spectra of (a) BSA, (b) BSA-attached magnetic particles, (c) magnetic particles, and (d) magnetic particles after desorption of BSA.
model is used to fit the experimental data. The Langmuir equation is expressed as Q=
Qm Ka C ∗ , 1 + Ka C ∗
(1)
where C ∗ (mg/ml) and Q (mg/g solid) are BSA concentration in the aqueous solution and the absorbed BSA on the solid at equilibrium, respectively. Qm is the maximum adsorption amount, and Ka (ml/mg) is the adsorption constant. The experimental data are fitted to the Langmuir model through nonlinear regression analysis. The fitted parameters of the model are summarized in Table 1. High R 2 values indicate that the Langmuir model predicts well the adsorption behavior over a wide range of pH. 3.2.4. FTIR and XPS spectra for BSA and magnetic particles FTIR and XPS spectra of BSA, Fe3 O4 , and BSA-adsorbed Fe3 O4 are analyzed to show the BSA adsorption on magnetic particles. The FTIR results are shown in Fig. 7. Spectrum a indicates that the characteristic bands of BSA occur at 1648 and 1540 cm−1 . Similar values were reported by previous work [15,27]. Spectrum c shows the characteristic band of Fe3 O4 at 600 cm−1 . After adsorption of BSA on magnetic particles, spectrum b shows both characteristic bands of BSA at 1648 and 1540 cm−1 and characteristic bands of Fe3 O4 at 600 cm−1 . It indicates that BSA was successfully attached to the surface of magnetic particles. Comparing spectra a,
Table 1 Langmuir model parameters for BSA adsorption isotherms on magnetic particles at different pH pH Qm (mg/g solid) Ka (ml/mg) R2
3.36
4.64
5.19
6.15
7.25
8.11
9.07
245.8 16.2 0.98
418.9 53.1 0.95
322.1 41.4 0.99
307.8 28.2 0.97
296.0 14.6 0.98
251.9 11.4 0.97
113.2 4.69 0.99
282
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
Fig. 9. Adsorption kinetics of BSA on magnetic particles at pH 4.64 for two feed concentrations (0.413 and 1.202 mg/ml). Fig. 8. XPS wide scan spectra of (a) Fe3 O4 , (b) BSA-attached magnetic particles, and (c) BSA.
Table 2 Desorption results of BSA from magnetic particles Initial BSA Adsorbed BSA Desorbed BSA Desorption concentration (mg/ml) (mg/g solid) (mg/g solid) percentage (%)
b, and c, it is found that the chemical interaction between Fe3 O4 and BSA is not significant, since little shift of the IR band due to the binding is observed. Fig. 8 shows the XPS wide scan spectrum of BSA, magnetic particles, and BSA attached on magnetic particles. Spectrum a shows the Fe2p peak of magnetic particles at about 710.3 eV. BSA contains N element, which show the N1s peak at about 399.6 eV in spectrum c. After BSA adsorption on magnetic particles, spectrum b shows both the N1s peak at about 399.6 eV and the Fe2p peak at about 710.3 eV. Therefore, XPS analysis of the surface composition confirmed the adsorption of BSA on magnetic particles. The XPS results agree well with that of FTIR.
Substituting Eq. (4) into Eq. (3), the following differential equation is obtained: (C0 − C)V /S dC . = KL a C − − (5) dt Qm Ka − Ka (C0 − C)V /S
3.2.5. Adsorption kinetics Study on the adsorption kinetics of BSA on magnetic particles was carried out at pH 4.64 at two different feed concentrations (0.413 and 1.202 mg/ml), and the results are shown in Fig. 9. A linear driving force mass-transfer model is used to describe the adsorption kinetics,
3.3. Desorption of BSA from magnetic particles
−
dC = KL a(C − C ∗ ), dt
(2)
where KL a (min−1 ) is the overall mass-transfer coefficient, C (mg/ml) is the concentration of BSA in solution at time t, and C ∗ (mg/ml) is the equilibrium concentration. C ∗ is obtained from Langmuir isotherm Eq. (1) and the resulting equation becomes Q dC = KL a C − . − (3) dt Qm Ka − QKa The mass balance of BSA can be expressed as C0 V = CV + SQ,
(4)
where C0 (mg/ml) is the initial concentration of BSA, V (ml) is the volume of the feed mixture, and S (g) is the total quantity of the solid.
0.413
135.7 ± 1.6
126.2 ± 1.1
92.9 ± 2.0
Equation (5) is integrated numerically using Polymath 4, a standard routine for solving nonlinear ordinary differential equations. The fitted curve and the experimental data are compared in Fig. 9. It is found that the experimental results are fitted well with KL a = 0.025 (min−1 ) for 0.413 mg/ml and KL a = 0.022 (min−1 ) for 1.202 mg/ml, respectively.
The desorption of BSA from magnetic particles was carried out under alkaline conditions of pH 9.35. The adsorbed and desorbed quantities of BSA at equilibrium for feed concentration 0.413 mg/ml are shown in Table 2. Desorption of more than 90% was achieved. SDS–PAGE electrophoresis experiments were carried out for BSA in three different samples, feed solution, and supernatants after adsorption and desorption experiments. The results are shown in Fig. 10. It is found that feed and supernatant from desorption experiments contain BSA bands in SDS–PAGE gel. Whereas the raffinate from the adsorption experiments does not contain any detectable BSA. This confirms the adsorption and subsequent desorption of BSA on magnetic particles. Some extrabands were observed for the desorbed supernatant sample compared to the BSA feed solutions. This could be the protein complexes formed by BSA molecules, which are activated by carbodiimide used in the adsorption process. But compared to the bulk band of BSA, it could be neglected.
Z.G. Peng et al. / Journal of Colloid and Interface Science 271 (2004) 277–283
283
Acknowledgment This work was financially supported by the National University of Singapore Research Fund (R-279-000-085-112).
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] Fig. 10. SDS–PAGE electrophoresis gel stained by Coomassie blue (lane 1, 0.413 mg/ml BSA feed solution; lane 2, supernatant after adsorption experiment; lane 3, supernatant after desorption experiment). Each lane was loaded with 20 µl of sample.
[12]
The FTIR spectrum of magnetic particles after desorption is shown in Fig. 7. It shows the characteristic band of Fe3 O4 at 600 cm−1 . But the absence of a BSA characteristic band confirms the desorption of BSA from magnetic particles.
[15]
[13] [14]
[16]
[17]
4. Conclusions Nanosized magnetic particles Fe3 O4 were prepared by the chemical precipitation method and characterized by TEM and VSM for size and supermagnetic properties. Adsorption of BSA on magnetic particles was carried out in the presence of carbodiimide under different pH and salt concentrations. It has been found that pH has a great effect on BSA adsorption while the effect of salt at a lower concentration range was comparatively insignificant. The maximum adsorption of BSA on magnetic particles occurred at the isoelectric point of BSA. The adsorption equilibrium results are fitted well by the Langmuir model while a linear driving force mass-transfer model is used for fitting kinetic data. FTIR and XPS spectra are used to confirm the attachment of BSA on the magnetic particles. Desorption of BSA was carried out under alkaline conditions, which was confirmed by SDS–PAGE electrophoresis and FTIR results. More than 90% desorption efficiency was achieved.
[18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33]
R.F.H. Dekker, Appl. Biochem. Biotechnol. 22 (1989) 289. A. Kondo, H. Fukuda, J. Ferment. Bioeng. 84 (1997) 337. Y. Haik, V. Pai, C.J. Chen, J. Magn. Magn. Mater. 194 (1999) 254. J.P. Hancock, J.T. Kemshead, J. Immunol. Methods 164 (1993) 51. R.S. Molday, L.L. Molday, FEBS Lett. 170 (1984) 232. T. Abudiab, R.R. Beitle, J. Chromatogr. A 795 (1998) 211. S.M. O’Brien, O.R.T. Thomas, P. Dunnill, J. Biotechnol. 50 (1996) 13. P.R. Levison, S.E. Badger, J. Dennis, P. Hathi, M.J. Davies, I.J. Bruce, D. Schimkat, J. Chromatogr. A 816 (1998) 107. M. Uhlen, Nature 340 (1989) 733. A.N. Rusetski, E.K. Ruuge, J. Magn. Magn. Mater. 85 (1990) 299. H.P. Khng, D. Cunliffe, S. Davies, N.A. Turner, E.N. Vulfson, Biotechnol. Bioeng. 60 (1998) 419. M. Suzuki, M. Shinkai, M. Kamihira, T. Kobayashi, Biotechnol. Appl. Biosci. 21 (1995) 335. J.P. Chen, Y.C. Chen, Bioresource Technol. 60 (1997) 231. A.S. Dilgimen, Z. Mustafaeva, M. Demchenko, T. Kaneko, Y. Osada, M. Mustafaev, Biomaterials 22 (2001) 2383. R.V. Mehta, R.V. Upadhyay, S.W. Charles, C.N. Ramchand, Biotechnol. Tech. 11 (1997) 493. M. Koneracka, P. Kopcansky, M. Antalik, M. Timko, C.N. Ramchand, D. Lobo, R.V. Mehta, R.V. Upadhyay, J. Magn. Magn. Mater. 201 (1999) 427. J.Y. Yoon, J.H. Lee, J.H. Kim, W.S. Kim, Colloids Surf. B 10 (1998) 365. C. Bergemann, D. Muller-Schulte, J. Oster, L. a Brassard, A.S. Lubbe, J. Magn. Magn. Mater. 194 (1999) 45. I. Safarik, M. Safarikova, J. Biochem. Biophys. Methods 27 (1993) 327. H. Honda, A. Kawabe, M. Skinkai, T. Kobayashi, Biochem. Eng. J. 3 (1999) 157. H. Yoshida, T. Kataoka, Chem. Eng. J. Biochem. Eng. 41 (1989) B11. A. Wooding, M. Kilner, D.B. Lambrick, J. Colloid Interface Sci. 144 (1991) 236. Y.S. Kang, S. Risbud, J.F. Rabolt, P. Stroeve, Chem. Mater. 8 (1996) 2209. D. Tanyolac, A.R. Ozdural, J. Appl. Polym. Sci. 80 (2001) 707. R.W. Chantrell, J. Popplewell, S.W. Charles, IEEE Trans. Magn. MAG-14 (1978) 975. V.S. Zaitsev, D.S. Filimonov, I.A. Presnyakov, R.J. Gambino, B. Chu, J. Colloid Interface Sci. 212 (1999) 49. D.H. Chen, M.H. Liao, J. Mol. Catal. B 16 (2002) 283. W.K. Lee, J.S. Ko, H.M. Kim, J. Colloid Interface Sci. 246 (2002) 70. K.Y. Chun, P. Stroeve, Langmuir 18 (2002) 4653. I. Iwasaki, S.R.B. Cooke, Y.S. Kim, SME Trans. 223 (1962) 113. T.J. Peters, Adv. Protein Chem. 37 (1985) 161. A.K. Bajpai, J. Appl. Polym. Sci. 78 (2000) 933. H.S. Liu, Y.C. Wang, W.Y. Chen, Colloids Surf. B 5 (1995) 25.