Polymer Nanoparticles Covered with Phosphorylcholine Groups and

Feb 2, 2008 - All Publications/Website .... Polymer Nanoparticles Covered with Phosphorylcholine Groups and Immobilized with Antibody for ... photoele...
0 downloads 0 Views 642KB Size
Download PDF
828

Biomacromolecules 2008, 9, 828–833

Polymer Nanoparticles Covered with Phosphorylcholine Groups and Immobilized with Antibody for High-Affinity Separation of Proteins Yusuke Goto,† Ryosuke Matsuno,†,§ Tomohiro Konno,†,§ Madoka Takai,†,§ and Kazuhiko Ishihara*,†,‡,§ Department of Materials Engineering, Department of Bioengineering, School of Engineering, and Center for NanoBio Integration, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-8586, Japan Received October 19, 2007; Revised Manuscript Received December 10, 2007

Novel polymer nanoparticles were prepared for the selective capture of a specific protein from a mixture with high effectiveness. The nanoparticle surface was covered with hydrophilic phosphorylcholine groups and active ester groups for easy immobilization of antibodies. Phospholipid polymers (PMBN) composed of 2-methacryloyloxyethyl phosphorylcholine, n-butyl methacrylate, and p-nitrophenyloxycarbonyl polyethyleneglycol methacrylate, were synthesized for the surface modification of poly(L-lactic acid) nanoparticles. Surface analysis of the nanoparticles using laser-Doppler electrophoresis and X-ray photoelectron spectroscopy revealed that the surface of nanoparticles was covered with PMBN. Protein adsorption was evaluated with regard to the nonspecific adsorption on the nanoparticles that was effectively suppressed by the phosphorylcholine groups. The immobilization of antibodies on nanoparticles was carried out under physiological conditions to ensure specific binding of antigens. The antibody immobilized on the nanoparticles exhibited high activity and strong affinity for the antigen similar to that exhibited by an antibody in a solution. The selective binding of a specific protein as an antigen from a protein mixture was relatively high compared to that observed with conventional antibody-immobilized polymer nanoparticles. In conclusion, nanoparticles having both phosphorylcholine and active ester groups for antibody immobilization have strong potential for use in highly selective separation based on the biological affinities between biomolecules.

Introduction The purification of native or recombinant proteins is important for conducting research in the life sciences, particularly proteomics, and also for the applications of these molecules in these fields. Adsorptive and chromatographic separation processes have played a major role in protein separation.1 Affinity separation is based on the natural biological affinity displayed between biological macromolecules and complementary ligands such as enzymes and coenzymes, antibodies and antigens, and receptors and hormones. The dominant affinity purification technique is affinity chromatography.2,3 This technique enables a high degree of purification while providing a good recovery yield. However, its widespread use has been limited by the inefficiency and instability of conventional solid phases. A novel protein purification technique involves the use of affinity nanoparticles, an affinity purification method in which ligandimmobilized nanoparticles are used in a batchwise manner. Nanoparticles have a very high ratio of surface area to volume. This provides a tremendous driving force for diffusion, resulting in a rapid procedure. However, the nonspecific binding of proteins onto nanoparticles has complicated the identification of the target protein. Often, low purification efficiency necessitates pretreatment and partial purification of the source protein mixtures before attempting affinity purification. Thus, the most important property to be considered while developing affinity * Corresponding author. E-mail: [email protected]. Telephone: +81-3-5841-7124. Fax: +81-3-5841-8647. † Department of Materials Engineering, School of Engineering. § Center for NanoBio Integration. ‡ Department of Bioengineering, School of Engineering.

nanoparticles is low nonspecific adsorption of proteins and other biomolecules for achieving a high signal-to-noise (S/N) ratio (signal: the target protein, noise: the other proteins). Various affinity particles have been developed based on polystyrene nanoparticles4 and magnetic nanoparticles.5,6 However, the problem of nonspecific adsorption has not yet been discussed sufficiently. Additionally, although the activity of the biomolecules immobilized on the particles decides the performance of the affinity particles, it has not ever been examined. Indeed, the surface simultaneously achieving reduction of nonspecific adsorption and retention of the molecule’s activity should be necessary for affinity nanoparticles to enhance its sensitivity. As new biocompatible polymers, phospholipid polymers have been synthesized from a methacrylate with a phosphorylcholine group, 2-methacryloyloxyethyl phosphorylcholine (MPC), and other vinyl compounds by radical copolymerization.7–12 Further, many research groups have made considerable contributions toward enhancing research in phospholipid polymers developed using MPC and its derivatives.13–19 Phospholipid polymers can form an artificial cell membrane structure by coating, blending with other polymers, and grafting to the polymer chains.9–11,20–24 Phospholipid polymers thus provide biologically inert properties. By controlling the composition of the MPC units in the polymer and the molecular weight of the polymer, water-soluble MPC polymers, including poly(MPC-co-n-butyl methacrylate (BMA)co-p-nitrophenyloxycarbonyl polyethyleneglycol methacrylate (MEONP)) (PMBN), could be prepared.25–28 PMBN could suspend poorly water-soluble organic compounds in an aqueous medium due to its amphiphilic nature, and the MEONP unit could bind biomolecules covalently under very mild conditions. Phospholipid polymer nanoparticles (PMBN/PLA-NP) were

10.1021/bm701161d CCC: $40.75  2008 American Chemical Society Published on Web 02/02/2008

Polymer Nanoparticles with Phosphorylcholine and Antibody

Biomacromolecules, Vol. 9, No. 3, 2008 829

Figure 2. Chemical structure of PMBN.

Figure 1. Scheme of the PMBN/PLA-NP’s function for an affinity purification.

prepared in an aqueous solution containing PMBN to cover the surfaces of hydrophobic poly(L-lactic acid) (PLA) by the solvent evaporation technique. We have reported the applications of various PMBN/PLA-NP-immobilized biomolecules, enzymes, and antibodies.26,29–32 From the previous studies, the PMBN/ PLA-NP could be a prospective candidate tool for affinity purification as shown in Figure 1, however, the potential property of PMBN-PLA-NP, especially the maintaining the activity of biomolecule immobilized on it, had not been investigated sufficiently. In this study, we evaluated the effects of phosphorylcholine groups on the suppression of protein adsorption and the retention of the antibody’s activity in the nanoparticles system. The affinity binding of the bovine serum albumin (BSA) and antiBSA antibody was examined as a model antigen–antibody interaction. An anti-BSA antibody was covalently immobilized onto PMBN/PLA-NP, and the performance of these immobilized nanoparticles in affinity separation was evaluated by using a plasma protein mixture solution.

Experimental Section Materials. MPC was synthesized using a previously reported method.7 BMA and 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC) were purchased from Nacalai Tesque, Inc. (Kyoto, Japan). BMA was purified under reduced pressure, and the fraction at 63 °C/24 mmHg was used. N-hydroxysuccinimide (NHS) was purchased from MERCK & Co., Inc. (Frankfurt, Germany), and 2,2′azobisisobutyronitrile (AIBN) was purchased from Kanto Chemical Co., Ltd. (Tokyo, Japan). PLA (average molecular weight: 2.0 × 104) was purchased from Wako Pure Chemical Industries Co., Ltd. (Osaka, Japan). Glycine and BSA labeled with fluorescein isothiocyanate (BSA/ FITC) were purchased from Sigma-Aldrich, Corp. (St. Louis, MO). Polybead microspheres (PS-NP, diameter ) 0.20 µm) and polybead carboxylate microspheres (PSc-NP, diameter ) 0.20 µm) were purchased from Polysciences Inc. (Warrington, PA). The anti-BSA antibody was purchased from Rockland, Inc. (Gilbertsville, PA). One of the γ-globulins (IgG, antichicken IgG) labeled with aminomethylcoumarin (anti-IgG/AMCA) was purchased from Chemicon (Temecula, CA). Human plasma fibrinogen (HPF) labeled with Alexa Fluor 594 (HPF/Alexa) was purchased from Invitrogen (Carlesbad, CA). All other reagents were of extra-pure reagent grade.

Methods Preparation of MPC polymers. MEONP was synthesized by a previously reported method.27 PMBN (Figure 2) was synthesized from the corresponding monomers such as MPC, BMA, and MEONP by a conventional radical polymerization technique using AIBN as an initiator.27,28 Polymerization was carried out at 60 °C. The reaction mixture was then poured into an excess amount of diethyl ether to precipitate the polymer, and the polymers were then collected by filtration and dried in vacuo. The composition of each monomer unit in PMBN was

determined by 1H NMR measurements (JEOL R-500 highresonance spectroscopy). The number-averaged molecular weight (Mn) was evaluated using gel permeation chromatography (GPC; JASCO, Tokyo, Japan) with poly(ethylene oxide) standards in a solvent system (water/methanol ) 3/7). Preparation of Phospholipid Polymer Nanoparticles. PLA nanoparticles coated with PMBN (PMBN/PLA-NP) were prepared using a solvent evaporation technique. 26,29–31 A brief explanation of the procedure is as follows. An aqueous solution (40 mL) containing 400 mg of PMBN (10 mg/mL) was placed in a glass bottle, and the solution was stirred at 400 rpm with cooling in an ice bath. PLA (20 mg) was dissolved in 2.0 mL of dichloromethane. The PLA solution was then added to the aqueous PMBN solution. The mixture was sonicated using a probe-type sonicator (Sonifier 250; Branson, Danbury, CT) for 30 min under a cool condition and was maintained under reduced pressure for 1 h to evaporate the dichrolomethane. The formed nanoparticles were collected by centrifugation at 10300 g at 4 °C for 30 min (Allegra 21R centrifuge, Beckman Coulter, Palo Alto, CA). To remove the excess PMBN in the solution, the nanoparticles were repeatedly washed with water by centrifugation and resuspended in water. The average diameter of the PMBN/PLA-NP was determined by dynamic light scattering (DLS-7000; Otsuka Electronics Co., Ltd., Tokyo, Japan). The morphology of the PMBN/PLA-NP was observed using atomic force microscopy (AFM; Nanoscope IIIa; Veeco, Tokyo, Japan). The surface potential of the PMBN/ PLA-NP was measured using laser-Doppler electrophoresis (ELS-8000; Otsuka Electronics Co., Ltd., Tokyo, Japan). The density of the MPC units on the PMBN/PLA-NP surface was evaluated using X-ray photoelectron spectroscopy (XPS; AXISHsi; Shimadzu/Kratos, Kyoto, Japan). The takeoff angle of the photoelectron was 90°. To determine the number of p-nitrophenyl active ester groups on the surface of the PMBN/PLANP, 0.01 N NaOH aqueous solution was added to the PMBN/ PLA-NP aqueous suspension to facilitate hydrolysis of the MEONP unit, resulting in the formation of p-nitrophenoxy anions. After centrifugation, the UV absorption of the supernatant at 400 nm based on the p-nitrophenoxy anions was measured using an UV/visible spectrometer (V-650; Jasco, Tokyo, Japan). Purified PMBN/PLA-NP were maintained at 4 °C at a concentration of 20 mg/mL before use. Evaluating Protein Adsorption onto Various Nanoparticles. The active ester groups on the PMBN/PLA-NP were reacted with glycine (3 mg/mL) beforehand to avoid reaction with proteins. BSA/FITC was dissolved in PBS (pH 8.2) to prepare a 100 µg/mL solution. After mixing 500 µL of the PMBN/PLA-NP solution and 500 µL of the BSA/FITC solution, the resulting solution was incubated for 24 h at 37 °C. Washes (repeating centrifugation at 10300g for 30 min and resuspension with 1 mL PBS) were done three times, and the nanoparticles were then isolated by centrifugation. The amount of BSA/FITC adsorbed on the nanoparticles was calculated by the measurement of fluorescence intensity in the solution based on free BSA/ FITC using a fluorescence spectrometer (FP-6500; JASCO, Tokyo, Japan) (excitation wavelength (λex) ) 490 nm, emission

830 Biomacromolecules, Vol. 9, No. 3, 2008

Goto et al.

Table 1. Characterization of PMBN mole fraction in polymerc

mole fraction in feed a

PMBN451 PMBN532b PMBN631b

MPCl

BMAm

MEONPn

MPCl

BMAm

MEONPn

yield (%)

time (h)

Mnd

0.40 0.50 0.60

0.50 0.30 0.30

0.10 0.20 0.10

0.38 0.59 0.73

0.58 0.36 0.25

0.04 0.05 0.02

69 53 62

5 3.5 3

3.7 × 104 5.8 × 104 8.0 × 104

[Monomer] ) a 1.0, b 0.5 mol L-1, [AIBN] ) a10, b5 mmol L-1, temperature at 60 °C in ethanol. Precipitated by diethyl ether/chloroform ) 8/2. c Determined by 1H NMR spectrum. d Number-averaged molecular weight (Mn) was determined by GPC in water/methanol ) 3/7, PEO standard.

wavelength (λem) ) 520 nm). Adsorption of BSA/FITC on the other nanoparticles was also investigated by the same procedure. Antigen–Antibody Reaction on Nanoparticles. The antiBSA antibody was dissolved in PBS (pH 8.2) to prepare a 1 mg/mL solution. After mixing 1 mL of the PMBN/PLA-NP solution (4 mg/mL) and 1 mL of the anti-BSA antibody solution, the resulting solution was stored for 24 h at 4 °C. After washes, the amount of immobilized anti-BSA antibody on the PMBN/ PLA-NP was determined from the UV absorbance of the supernatant at 400 nm using UV/visible spectroscopy; the absorbance was based on the p-nitrophenoxy anion released during the immobilizing reaction. The remaining active ester groups on the nanoparticles were also reacted with glycine (3 mg/mL) for 24 h at 4 °C. To observe the antigen–antibody reaction, 500 µL of the BSA/FITC solution (100 µg/mL) was added to 500 µL of the anti-BSA antibody-immobilized PMBN/ PLA-NP suspension, and the resulting solution was stored for 24 h at 37 °C in Tris-HCl buffer (pH 7.4). After washes, the amount of reacted BSA on the nanoparticles was determined from the fluorescence intensity of the resuspension based on free BSA/FITC. Evaluation of Dissociation Constant of Antibody– Antigen Complexes on Nanoparticles. Assuming polyvalent binding, the equilibrium dissociation constant of antibody–antigen complexes, Kd, can be expressed by the following eq 1:

θ )

[Ag]n Kd + [Ag]n

(1)

where θ is the fractional occupancy, that is, the ratio of bound antibody–antigen complexes to total antibody concentration, [Ag] is the antigen concentration, and n is the Hill coefficient. The Hill plot, the linear transformation that is commonly used, is derived by rearranging eq 1 and using the following eq 2:

[ 1 -θ θ ] ) nlog[Ag] - logK

log

d

(2)

To evaluate the dissociation constant of BSA from the antiBSA antibody immobilized on the nanoparticles, 500 µL of the suspension of anti-BSA-immobilized PMBNPLA-NP (2 mg/ mL) was added to 500 µL of the BSA/FITC solutions at various concentrations (2.25-36 µg/mL). The anti-BSA antibody was immobilized on the PSc-NP as a control. The procedure is as follows. EDC (95.8 mg) and NHS (57.5 mg) were dissolved in 16.0 mL of distilled water. Four mL of PSc-NP suspension was added to the solution and reacted for 24 h at room temperature to form active ester groups on the nanoparticles. After centrifugation, the purified nanoparticles (PSc-suc-NP) were obtained as a precipitate. One mL of the anti-BSA antibody solution (1 mg/mL) and 1 mL of the PScsuc-NP resuspension in Tris-HCl buffer (4 mg/mL) were mixed and stored for 24 h at 4 °C. To compare Kd, 500 µL of the suspension of anti-BSA-immobilized PSc-suc-NP was added to 500 µL of the BSA/FITC solution at varying concentrations.

Affinity Selection Test of Target Protein. BSA/FITC, antiIgG/AMCA, and HPF/Alexa were dissolved in Tris-HCl buffer (pH 7.4) to prepare the protein mixture solution (100, 60, and 14 µg/mL, respectively). Five hundred mL of this mixture was added to 500 µL of the anti-BSA antibody-immobilized PMBN/ PLA-NP suspension and stored for 24 h at 37 °C. After washes and centrifugation at 10000g for 30 min to precipitate the nanoparticles, the fluorescence intensity of the resuspension in 1 mL Tris-HCl buffer was measured. The λex of the fluorescence dyes FITC, AMCA, and Alexa were 350, 490, and 592 nm, respectively. The fluorescence intensity at 450, 520, and 612 nm, which corresponded to that of each fluorescence dye, was recorded to determine the concentration of each protein in the resuspension. Using the anti-BSA antibody-immobilized PScsuc-NP, the same procedure was carried out to evaluate the effects of PMBN on the adsorption of proteins.

Results and Discussion Because under biological conditions, polymer nanoparticles exhibit good dispersion in an aqueous medium and a large surface area, the separation and analysis of a specific biomolecule from a mixture is possible. During nanoparticles preparation, surface properties particularly, hydrophilicity that may be important for nanoparticles dispersion in an aqueous medium should be regulated. Moreover, to enhance specific separation, biological affinity is induced by the immobilization antibodies on nanoparticles. Further, it is essential to prevent nonspecific adsorption on nanoparticles. Therefore, we decided to develop a protein adsorption-resistant surface on nanoparticles followed by antibody immobilization under mild conditions. To achieve this, phosphorylcholine groups were assembled on the nanoparticles surface. Our previous articles reported the preparation of nanoparticles covered with phospholipid polymers that could suppress protein adsorption effectively.29,30,32 On these nanoparticles, proteins were immobilized covalently by the reaction between the active ester groups located on the nanoparticles surface and the amino group of the proteins. In this research, affinity separation of a protein mixture using nanoparticles covered with PMBN was investigated. Characterization of MPC Polymers. The copolymerization of MPC, BMA, and MEONP proceeded homogeneously, and PMBN was obtained. The characteristics of PMBN are summarized in Table 1. The monomer composition of the polymer was in good agreement with the monomer compositions in the feed. Because PMBN was water soluble with amphiphilic properties, it could form polymer aggregates in an aqueous medium at concentrations greater than 0.1 mg/mL. Characterization of Phospholipid Polymer Nanoparticles. PMBN/PLA-NP could be prepared using a solvent evaporation technique. The procedure used for nanoparticles preparation was as follows. The core polymer PLA was dissolved in dichloromethane but was insoluble in water. Droplets of the PLA solution were suspended in an aqueous

Polymer Nanoparticles with Phosphorylcholine and Antibody

Biomacromolecules, Vol. 9, No. 3, 2008 831

Table 2. Surface Property of PMBN/PLA-NP averaged diametera (nm)

ζ potentialb (mV)

P/C ratioc

amount of active ester groups (10-9 mol/mg-nanoparticle)d

290 240 280

-2.0 -6.7 -2.8

0.019 0.021 0.016

4.7 3.1 1.9

PMBN451/PLA-NP PMBN532/PLA-NP PMBN631/PLA-NP Determined by p-nitrophenol

a

dynamic light scattering measurement

b

laser-doppler electrophoretic measurement

Figure 3. AFM picture of PMBN/PLA-NP.

solution with the PMBN aggregate. Dichloromethane has a low boiling temperature and was thus evaporated easily. At this point, PLA was precipitated at the interface of the aqueous medium containing PMBN. At this interface, the PMBN chains and PLA chains form entanglements. Thus, a stable PMBN coating layer was formed on the surface of the PLA core. The surface properties of the PMBN/PLA-NP are summarized in Table 2. The average diameters of the PMBN/PLA-NP ranged from 250 to 300 nm, and the polydispersity index was 7.3 × 10-2. This distribution range of the diameters indicated a monodispersion as calculated from the cumulant method. Figure 3 shows an AFM image of the PMBN/PLA-NP, indicating the spherical shape of these particles. The ζ potential of the PMBN/ PLA-NP ranged from -6.8 to -2.0 mV. The ζ potential of the PMBN/PLA-NP was nearly zero, whereas that of the PLA nanoparticles was -60 mV.26,32 This is because of the electrically neutrality of the phosphorylcholine groups due to the formation of an inner salt between phosphate anions and trimethylammonium cations. XPS analysis and ζ potential measurements revealed that the surface of PMBN/PLA-NP was covered with phosphorylcholine groups. XPS analysis indicated that the PMBN/PLA-NP had specific peaks attributed to component atoms such as a phosphorus peak at 135 eV, a nitrogen peak at 403 eV, an oxygen peak at 530 eV, and a strong carbon peak at 285 eV that was attributed to methyl or methylene groups, as shown in Figure 4 (C 1s XPS and O 1s spectra were not shown). The P/C ratios were calculated from the amounts of phosphorus and carbon atoms. The nanoparticles prepared from various PMBNs had almost the same P/C ratio of approximately 0.02. The PMBN/PLA-NP had good dispersion ability in an aqueous medium due to the extremely hydrophilic nature of the phosphorylcholine group. To quantify the amount of p-nitrophenyl ester groups on the PMBN/PLA-NP, the dissociation of p-nitrophenoxy ions by the hydrolysis of active ester was observed using UV/visible spectroscopy at 400 nm. The results confirmed the presence of active ester groups on the surface of PMBN/PLA-NP. The values were approximately 10-9 mol/mg of nanoparticles.

c

X-ray photospectroscopy

d

UV adsorption of

Figure 4. N 1s and P 2p XPS spectra of PMBN/PLA-NP.

Figure 5. BSA adsorption onto various nanoparticles.

BSA Adsorption on Various Nanoparticles. The effects of the phosphorylcholine groups on the adsorption of BSA to the nanoparticles were evaluated. To avoid chemical reaction with BSA, the p-nitrophenyl ester groups on PMBN/PLA-NP were reacted with glycine for blocking the active ester groups in this experiment. Figure 5 shows the amount of BSA adsorbed on various nanoparticles. Numerous BSA molecules were adsorbed on the PS-NP, whereas these molecules were hardly adsorbed on the glycine-reacted PMBN/PLA-NP. Among the PMBN/ PLA-NPs, the PMBN532/PLA-NP showed the lowest BSA adsorption; the value was approximately 1/300 as compared with that on PS-NP. The BSA adsorption test indicated that the PMBN/PLA-NP showed excellent suppression of BSA adsorption. It is well known that phosphorylcholine groups exhibit antiadsorption properties even when the nanoparticles surface bearing phosphorylcholine groups is in contact with human plasma.10,11,23,24 On the other hand, the surface of PS-NP was hydrophobic and adsorbed proteins. PMBN/PLA-NP was suitable for developing affinity nanoparticles because of their considerable ability to prevent nonspecific adsorption. Thus, PMBN532/PLA-NP was employed in the following experiments. Antigen–Antibody Reaction on Nanoparticles. The activity and function of the antibody immobilized on the PMBN/PLA-

832 Biomacromolecules, Vol. 9, No. 3, 2008

Figure 6. Antigen–antibody reaction on the surface of PMBN532/ PLA-NP.

Figure 7. Hill plots of the reaction between BSA and various nanoparticles (a) PMBN532/PLA-NP, (b) PSc-suc-NP.

NP was evaluated by observing the antigen–antibody reaction. For the preparation of PLA-NP, we used PMBN532 as the MPC polymer due to its ability to suppress the nonspecific adsorption of BSA, as shown in Figure 5. Figure 6 shows the reactive binding of BSA to anti-BSA antibody-immobilized PMBN/ PLA-NP. The amount of BSA on the anti-BSA antibodyimmobilized PMBN/PLA-NP was higher than that on PMBN/ PLA-NP without antibody immobilization. This result indicates that the amount of BSA increased by reacting with the antibody immobilized on the surface of PMBN/PLA-NP. On the basis of this result, the antibody-immobilized PMBN/PLA-NP could function as affinity nanoparticles. Evaluation of the Dissociation Constant of Antigen– Antibody Complexes on Nanoparticles. The dissociation constant of the antigen-antibody complex substantially defines the performance of affinity-based separation, diagnosis, and detection systems. To evaluate the performance of PMBN/PLANP as affinity nanoparticles, the dissociation constant of the antigen-antibody complex on PMBN/PLA-NP was measured. Figure 7 shows the Hill plot of the reaction between BSA and various nanoparticles. The dissociation constant could be calculated from these plots, and it was observed to be 2.7 × 10-7 M for the anti-BSA antibody-immobilized PMBN/PLANP and 1.3 × 10-5 M for the anti-BSA antibody-immobilized PSc-suc-NP. Thus, the affinity of the anti-BSA antibody to BSA observed on the PMBN/PLA-NP was approximately 200-fold higher than that on the PSc-suc-NP. In addition, the Hill coefficients of PMBN/PLA-NP and PSc-suc-NP were 1.3 and 0.76, respectively.

Goto et al.

Figure 8. Affinity separation from protein mixture solution, BSA (gray), IgG (black), HPF (white).

The Kd value generally ranges from 10-7 to 10-10 for an antigen–antibody complex.33 The Kd value of the anti-BSA antibody immobilized on PMBN/PLA-NP for BSA is considered valid, while that for the anti-BSA antibody immobilized on PScsuc-NP is higher than the reported value. This indicates that the antibody immobilized on the PMBN/PLA-NP had a strong affinity toward the antigen, maintaining that the activity of the antibody even when immobilized on the nanoparticles. However, in the case of antibodies immobilized on the PSc-suc-NP, a large Kd value was observed; this was due to a weakening in the affinity for the antigen by denaturation of the antibody during immobilization. These results indicate the effects of the phosphorylcholine groups in preventing the denaturation of the antibody. The Hill coefficient n describes the cooperativity of antigen binding. If n > 1, cooperativity is positive, i.e., once one antigen molecule binds to the antibody, the affinity of the antibody for another antigen increases. If n < 1, cooperativity is negative, that is, once one antigen molecule binds to the antibody, the affinity of the antibody for other antigens decreases. In the case of the anti-BSA antigen-immobilized PMBN/PLA-NP, the reaction to BSA indicated slight positive cooperativity, whereas in the case of anti-BSA antigen-immobilized PSc-suc-NP, slightly negative cooperativity was observed with regard to antigen binding. These results may influence the amount of protein adsorption and separation efficiency. Affinity Selection of Target Protein. The capture of target proteins from the protein mixture with high efficiency was evaluated using antibody-immobilized PMBN/PLA-NP. Figure 8 shows the amount of proteins bound to the nanoparticles. Both the anti-BSA antibody-immobilized PMBN/PLA-NP and PScsuc-NP could bind BSA significantly compared with other plasma proteins such as IgG and HPF. The selectivity of the anti-BSA antibody-immobilized PMBN/PLA-NP for BSA against that for IgG and fibrinogen was 48 and 88, respectively. Comparison of PMBN/PLA-NP and PSc-suc-NP revealed a significant difference in BSA binding affinity, that is, the affinity of PMBN/PLA-NP for BSA was approximately two times greater than that of PSc-suc-NP. According to Figure 8, The anti-BSA antibody-immobilized PMBN/PLA-NP showed a high selectivity for BSA compared with other plasma proteins from the protein mixture. This selectivity that exceeded that for IgG and HPF, which was 48 and 88, respectively. These values were larger than those observed for anti-BSA antibody-immobilized PSc-suc-NP. This is due to the difference in the state of the antibody immobilized on the nanoparticles. The amount of bound IgG and HPF was almost the same in both nanoparticles. Nonspecific adsorption on the nanoparticle surface hardly occurred because of the immobilization of the antibody. However, the activity of

Polymer Nanoparticles with Phosphorylcholine and Antibody

Biomacromolecules, Vol. 9, No. 3, 2008 833

the antibody is different as discussed previously. Therefore, the selectivity for BSA was significantly different. We already reported that the proteins embedded in phosphorylcholine groups show high structural stability and maintain their activity.30 The same phenomenon was observed even on the nanoparticles. From these findings, PMBN/PLA-NP can be considered as a promising platform for the immobilization of proteins that possess high biological affinity toward target molecules.

(8) Ueda, T.; Oshida, H.; Kurita, K.; Ishihara, K.; Nakabayashi, N. Polym. J. 1992, 24, 1259–1269. (9) Ishihara, K.; Aragaki, R.; Ueda, T.; Watanabe, A.; Nakabayashi, N. J. Biomed. Mater. Res. 1990, 24, 1069–1077. (10) Ishihara, K.; Ziats, N. P.; Tierney, B. P.; Nakabayashi, N.; Anderson, J. M. J. Biomed. Mater. Res. 1991, 25, 1397–1407. (11) Ishihara, K.; Oshida, H.; Ueda, T.; Endo, Y.; Watanabe, A.; Nakabayashi, N. J. Biomed. Mater. Res. 1992, 26, 1543–1552. (12) Ishihara, K.; Nomura, H.; Mihara, T.; Kurita, K.; Iwasaki, Y.; Nakabayashi, N. J. Biomed. Mater. Res. 1998, 39, 323–330. (13) Sugiyama, K; Ohga, K.; Aoki, H.; Amaya, N. Macromol. Chem. Phys. 1995, 196, 1907–1916. (14) Oishi, T.; Uchiyama, H.; Onimura, K. Polym. J. 1998, 30, 17–20. (15) Campbell, E. J.; Wiles, M. C.; Yianni, P. ASAIO J. 1994, M858– M858. (16) Miyazawa, K.; Winnik, F. M. J. Phys. Chem. B 2003, 107, 10677– 10682. (17) Lam, J. K. W.; Ma, Y.; Ames, S. P.; Lewis, A. L.; Baidwin, T.; Stolnik, S. J. Controlled Release 2004, 100, 293–312. (18) Goreish, H. H; Lewis, A. L.; Rose, S.; Lloyd, A. W. J. Biomed. Mater. Res. B 2004, 68, 1–9. (19) Lewis, A. L. Colloid Surf., B 2000, 18, 261–275. (20) Moro, T.; Takatori, Y.; Ishihara, K.; Konno, T.; Takigawa, Y.; Matsushita, T.; Chung, U. I.; Nakamura, K.; Kawaguchi, H. Nat. Mater. 2004, 3, 829–836. (21) Hasegawa, T.; Iwasaki, Y.; Ishihara, K. J. Biomed. Mater. Res. B 2002, 63, 333–341. (22) Goda, T.; Konno, T.; Takai, M.; Moro, T.; Ishihara, K. Biomaterials 2006, 27, 5151–5160. (23) Sawada, S.; Iwasaki, Y.; Nakabayashi, N.; Ishihara, K. J. Biomed. Mater. Res. A 2006, 79, 476–484. (24) Ishihara, K. Sci. Technol. AdV. Mater. 2000, 1, 131–138. (25) Ishihara, K.; Iwasaki, Y.; Nakabayashi, N. Polym. J. 1999, 31, 1231– 1236. (26) Konno, T.; Watanabe, J.; Ishihara, K. Biomacromolecules 2004, 5, 342–347. (27) Takei, K.; Konno, T.; Watanabe, J.; Ishihara, K. Biomacromolecules 2004, 5, 858–862. (28) Konno, T.; Watanabe, J.; Ishihara, K. J. Biomed. Mater. Res. 2003, 65A, 209–214. (29) Park, J.; Kurosawa, S.; Watanabe, J.; Ishihara, K. Anal. Chem. 2004, 76, 2649–2655. (30) Watanabe, J.; Ishihara, K. Biomacromolecules 2006, 7, 171–175. (31) Ito, T.; Watanabe, J.; Takai, M.; Konno, T.; Iwasaki, Y.; Ishihara, K. Colloids Surf., B 2006, 50, 55–60. (32) Konno, T.; Kurita, K.; Iwasaki, Y.; Nakabayashi, N.; Ishihara, K. Biomaterials 2001, 22, 1883–1889. (33) Garcia, A. A., Eds.; Bioseparation Process Science; Blackwell Science Inc.: Cambridge, MA, 1999.

Conclusion To enhance antigen–antibody affinity, nanoparticles covered with both phosphorylcholine and active ester groups were prepared. PMBN/PLA-NP could immobilize an antibody effectively under mild conditions. The antibody immobilized on PMBN/PLA-NP maintained high activity. The highly selective binding of target proteins from a protein mixture could be realized using PMBN/PLA-NP. On the basis of the study results, PMBN/PLA-NP could be regarded as a successful novel candidate tool for the bioaffinity separation of proteins. Acknowledgment. We thank Dr. K. Fukumoto, Dr. T. Ito, Dr. T. Goda, and Dr. T. Hoshi at The University of Tokyo for their helpful discussions and support for preparation of nanoparticles. We appreciate the help received from Prof. K. Kataoka at the University of the Tokyo for the DLS measurement. A part of this research was supported from Asahi Glass Foundation.

References and Notes (1) Rousseau, R. W., Ed.; Handbook of Separation Process Technology; Wiley-Interscience: New York, 1987. (2) Epton, R., Ed.; Hydrophobic Ion Exchange, and Affinity Methods; Halsted Press: New York, 1979. (3) McGregor, W. C., Eds. Membrane Separation in Biotechnology; Marcel Dekker: New York, 1986. (4) Gebben, B.; van Houwelingen, G. D. B.; Zhang, W.; van de Boomgaard, Th.; Smolders, C. A. Colloids Surf., B 1994, 3, 75–84. (5) Gu, H.; Xu, K.; Xu, C.; Xu, B. Chem. Commun. 2006, 94, 1–949. (6) Park, H-Y.; Schadt, M. J.; Wang, L.; Lim, I-I. S.; Njaki, P. N.; Kim, S. H.; Jang, M-Y.; Luo, J.; Zhong, C-J. Langmuir 2007, 23, 9050– 9056. (7) Ishihara, K.; Ueda, T.; Nakabayashi, N. Polym. J. 1990, 22, 355– 360.

BM701161D