- Email: [email protected]
Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins Wei Wang a,b , Ying Jing a , Shihai He a , Jian-Ping Wang a,∗ , Jian-Ping Zhai b,∗∗ a b
Department of Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN 55455, USA State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing 210046, PR China
a r t i c l e
i n f o
Article history: Received 6 June 2013 Received in revised form 27 November 2013 Accepted 29 November 2013 Available online xxx Keywords: FeCo magnetic nanoparticles Protein Surface modification Immobilization
a b s t r a c t Magnetic Fe70 Co30 nanoparticles with a cubic shape and a mean size of 15 ± 1.5 nm were fabricated using a magnetron–sputtering-based gas phase condensation deposition method. The particles had a high saturation magnetization of 220 emu/g, which is much higher than that of commercially available iron oxide nanoparticles. The FeCo nanoparticles were modified by 3-aminopropyltriethoxy silane and subsequently activated by glutaraldehyde, leading to successful attachment of aldehyde groups onto nanoparticle surfaces. Three proteins, namely streptavidin, PAPP-A antibody and Nectin-4 antibody, were immobilized on glutaraldehyde activated FeCo nanoparticles, and their loading levels were quantitatively evaluated. Our results show that loading capabilities are 95 g of streptavidin, 128 g of PAPP-A, and 125 g of Nectin-4 antibody per milligram of FeCo nanoparticles, and that the three immobilized proteins retain their binding bioactivity. The protein–FeCo conjugates may find valuable applications involving magnetic separation and purification of proteins and cells, and the magnetic detection of biomolecules. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Nanomaterials, having attracted much attention for decades as a result of their special and superior properties, are increasingly being applied in multidiscipline areas, especially in the fields of biology and biomedicine. Among the nanomaterials, many kinds of nanoparticles have been studied depending on their intrinsic properties, such as optical absorption (Au, Ag) [1,2], photoluminescence in the form of fluorescence (quantum dots, e.g. CdSe, CdS) [3,4], and magnetic moment (e.g. iron oxides) [5,6]. These moststudied nanoparticles can be potentially and multi-functionally used as diagnostic and therapeutic agents, delivery vehicles and imaging agents. In 1996, iron oxide magnetic nanoparticle (MNPs) based ferumoxides (Feridex I.V.) was approved by the US Food and Drug Administration (FDA) as an imaging agent for the detection of liver lesions, and now iron oxide MNPs have also been widely used in MRI contrast enhancement, hyperthermia, separation and purification of biomolecules, drug and gene delivery, and bimolecular detection [7–10]. MNPs with different compositions and sizes have been commercialized by a number of companies [11]. Most of these commercially available MNPs are iron oxides, which are easy to synthesize, low cost, and biocompatible. These proven qualities have further driven researchers to develop other kinds of
∗ Corresponding author. Tel.: +1 612 625 9509; fax: +1 612 625 4583. ∗∗ Corresponding author. Tel.: +86 025 83592903; fax: +86 025 83592903. E-mail addresses: [email protected] (J.-P. Wang), [email protected] (J.-P. Zhai).
MNPs. In addition to iron oxides, much attention has recently been paid to high-magnetic-moment materials [12–19]. Among them, bimetallic FeCo MNPs have the highest saturation magnetization (Ms), which makes it an attractive material choice for application to writing heads of hard disk drives [20]. Some methods have been reported for the fabrication of FeCo MNPs including reductive decomposition of Fe(III) acetylacetonate and Co(II) acetylacetonate [21], chemical coprecipitation [22], mechanical ball milling [23], chemical vapor deposition [24], pulsed laser ablation deposition [25], and interfacial diffusion [26]. Our FeCo MNPs were synthesized in gas phase using a magnetron–sputtering-based system, which can control the shape (cubic or spherical), sizes and composition of particles [27,28]. In order to fulfill full potential of these high-magnetic-moment nanoparticles for various biomedical applications, it is crucial to find and develop reliable and repeatable processes for their surface functionalization and protein conjugation. Basically, two different approaches can be employed to conjugate proteins to nanoparticles. The first method is through non-covalent interactions, such as electrostatic attraction [29], protein A interaction [30,31], and nickel-mediated interactions with His-tagged proteins [32,33]. The other method uses the direct covalent linkage between proteins and nanoparticle surfaces to form a conjugate that is stable regarding possible dissociation in complex biological media with other interfering biomolecules. The nanoparticle surfaces need prior modification with active groups, such as cyanogen bromide, aldehyde, epoxy, N-hydroxysuccinimide ester, and primary amino groups [34–37], that can be covalently bound to the
0927-7765/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
2
Scheme 1. Schematic illustration of an experimental overview.
carboxyl or amino groups of proteins. Some other techniques have also been developed for this purpose. Jung [38] used a site-specific photo cross-linker (benzophenone), and Wang [39] and Cho [40] developed Au coating on particle surfaces. In this work, the immobilization of proteins on FeCo nanoparticles was carried out via anchored aldehyde groups. 3-Aminopropyltriethoxysilane (APTES) was first used to modify particle surfaces, and then glutaraldehyde (Glu), an amine-reactive homobifunctional cross-linker used in biochemistry applications, could be bound to APTES leaving another aldehyde group at the end for capturing proteins (Scheme 1a). Conjugates of protein–MNPs have wide applications in biomedicine areas and have been extensively used in magnetic separation and purification of proteins and cells [41–46]. Additionally, Ma [47] used antibody–MNPs in a magnetic force induced microcantilever biosensor. These conjugates were also used in diagnostic magnetic resonance (DMR) and giant magnetoresistive (GMR) sensors [48,49]. Herein, three model proteins were selected for loading on the surfaces of FeCo MNPs with enhanced properties for the potential applications listed above. One is streptavidin, which has an extraordinarily high non-covalent affinity for biotin, and the biotin–streptavidin system is one of the most widely used affinity pairs in molecular, immunological, and cellular assays. The other two selected proteins were PAPP-A (Pregnancy-associated plasma protein A) and Nectin-4 monoclonal IgG antibodies. The loading amounts of these three proteins on FeCo MNPs were evaluated while binding bioactivity was confirmed for each protein.
2. Materials and methods 2.1. Materials 3-Aminopropyltriethoxy silane, 99% (APTES) was purchased from Acros organics. Glutaraldehyde solution (Grade II, 25% in H2 O), ethanol (≥99.5%, 200 proof), and tween 20 was purchased from Sigma–Aldrich. Phosphate buffered saline (PBS, 1×, pH 7.4) was purchased from Invitrogen. mPEG-amine (5 kDa) was purchased from Creative PEGWorks. Streptavidin-AF555 (2 mg/mL in PBS, pH 7.2) was purchased from Invitrogen. Recombinant human Nectin-4 [catalog number 2659-N4], monoclonal anti-human Nectin4 antibody [catalog number MAB2659], recombinant human
PAPP-A [catalog number 2487-ZN], monoclonal anti-human PAPPA antibody [catalog number MAB2487], and polyclonal anti-human PAPP-A biotinylated antibody [catalog number BAF2487] were purchased from R&D Systems, Inc., Minneapolis, USA. DyLight550 Microscale Antibody Labeling Kit [catalog number 84531] was purchased from Pierce, Thermo Scientific. The NHS ester functionalized microarray slide (No. DHD1-0023) was provided by SurModics, Inc. 2.2. Fabrication of FeCo MNPs FeCo MNPs were fabricated by using a magnetron–sputteringbased gas phase condensation deposition system [27,28]. The sputtering source used a planer target of FeCo alloy with atomic composition 70:30. Argon gas was used as sputtering gas and carrier gas. The FeCo nanoparticles were generated in the source chamber, and carried into the collection chamber by argon gas flow. Polyethylene glycol (PEG) pre-coated glass slides were placed in the collection chamber for collecting nanoparticles [50]. After deposition, the glass slides were taken out and washed by deionized (DI) water to transfer FeCo MNPs into solution. The water solution of nanoparticles was stored at 4–8 ◦ C for further modification and application. The concentration and composition of prepared samples were determined by ICP-OES (Optima 3000 DV, Perkin–Elmer). FeCo MNPs were dropped on a silicon wafer and dried using nitrogen gas. The particles on the wafer were further analyzed by an energy dispersive spectrometer (EDS) that was attached on a JEOL 6500 scanning electron microscope. Morphology and crystal structure of FeCo MNPs were characterized by a transmission electron microscope (TEM). Magnetic behavior measurement of FeCo MNPs was carried out at 300 K by a vibrating sample magnetometer (VSM, model 3900, Princeton Measurements Corporation). 2.3. Modification by 3-aminopropyltriethoxysilane (APTES) and glutaraldehyde FeCo MNPs (100 g Fe) were washed with ethanol three times, and were dispersed in 500 L ethanol/water (95/5 in volume ratio) solution with sonication. In the washing steps, the particles were harvested and separated by applying an external magnetic field. 10 L APTES was dissolved in 500 L ethanol/water (95/5) solution and stirred for 1 min, and then the APTES solution was added to
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
FeCo solution immediately. The reaction mixture was sonicated for 1 h. The particles were washed three times with acetone, followed by being washed three times with DI water. 500 L glutaraldehyde (Glu) solution (10%) was added to 500 L APTES modified FeCo MNPs solution (in PBS), and the mixture was sonicated for 5 h. The Glu activated nanoparticles were washed three times with DI water, and dispersed in PBST (PBS with 0.02% tween 20) and stored at 4–8 ◦ C. The samples were characterized by an SSX-100 spectrometer with an Al Ka X-ray source. Shirley background subtraction and properly curve-fitted peaks were applied in the XPS spectra peak fitting procedure. The quantification was based on peak areas calculated from the high-resolution spectra. All binding energies were referenced to the C 1 s neutral carbon peak at 284.6 eV. 2.4. Protein immobilization on Glu activated FeCo MNPs Monoclonal PAPP-A and Nectin-4 antibodies were both fluorescently labeled using Dylight 550 microscale antibody labeling kit [Pierce Inc., catalogue number 84531] according to the procedure provided by the vendor. 200 L of streptavidin-AF555 (25 g/mL in PBS) was added to 200 L of Glu activated nanoparticles solution (in PBST). After being sonicated for 30 min, the mixture was shaken and incubated for 12 h at 4 ◦ C using a mechanical rotor. After that, the particles were collected by a magnet and washed three times with 400 L PBS. The supernatant and washings (unbound streptavidin-AF555) were combined and made up to 5 mL using PBS for quantification. The experiments were repeated three times using fresh samples to obtain reliable error bars. To determine the concentration of unbound streptavidin-AF555, 100 L of combined solution (5 mL) was placed in a 96-well plate and the fluorescence intensity was compared to a standard curve. Fluorescence measurements were performed on a Cytofluor II Fluorescent Microplate Reader. During the fluorescence measurement, the fluorophore was excited at 530/25 nm and the emission was collected at 590/35 nm. For inactivating aldehyde groups and reducing non-specific binding, 400 L of 0.5 mg/mL mPEG-amine solution was added to the streptavidin-AF555 labeled FeCo particles, and the mixture was shaken for 1 h at room temperature. After being washing with PBS, the modified particles were dispersed in 1.0% bovine serum albumin (BSA) solution (in PBST), and stored at 4–8 ◦ C. In a similar way, Dylight 550 labeled PAPP-A antibody and Nectin-4 antibody were respectively conjugated to Glu activated FeCo MNPs. All immobilizations were quantitively analyzed, and the final conjugates were stored in 1.0% BSA solution containing 0.02% tween 20. 2.5. Binding to the surface of functionalized slide The process of streptavidin–FeCo conjugate binding to a functionalized slide is briefly illustrated in Scheme 1b, and other two antibody–FeCo conjugates also have similar binding processes. PAPP-A biotinylated antibody (20 g/mL), PAPP-A antigen (20 g/mL), and Nectin-4 antigen (20 g/mL) were spotted on NHS ester modified slides using the sciFLEXARRAYER S5 (Scienion, Germany), respectively. The printed slides were incubated for 24 h at 4 ◦ C under a relative humidity of ∼90%. After that, these modified slides were washed with immunoassay stabilizer (SurModics, Cat. No. SC01), and were blocked with 1.0 mg/mL BSA for 30 min. After being washed with PBST, the three functionalized slides were respectively incubated with streptavidin-AF555 (20 g/mL), Dylight 550-PAPP-A antibody (20 g/mL), and Dylight 550-Nectin4 antibody (20 g/mL) to test the binding activity of the three proteins on the slide surfaces. After incubating for 1 h at room temperature, the unbound fluorescence dye labeled proteins were washed off using PBST and DI water. The protein bound slides were dried by nitrogen gas, and were further imaged using Olympus IX70 Invert Fluorescence Microscope under identical camera conditions.
3
Table 1 Saturation magnetization of selected magnetic nanoparticles. Materials
Size (nm)
Ms (emu/g)
Ref.
Iron oxides (Fex Oy ) MnFe2 O4 Fe FeCo Fe50 Co50 Fe70 Co30
5–11 5–12 7 8–10 68 15
43–60 37–52 212 144.5 211.9 220
[51] [52] [19] [12] [16] This work
According to the procedure for fluorescence dye labeled protein binding, the three kinds of streptavidin-AF555, Dylight 550-PAPPA antibody and Dylight 550-Nectin-4 antibody labeled FeCo MNPs (with highest protein loading on particles) were also incubated on PAPP-A biotinylated antibody, PAPP-A antigen, and Nectin-4 antigen modified slides, respectively. The particles bound slides were also imaged using a Fluorescence Microscope. 3. Results and discussion 3.1. Properties of FeCo MNPs The TEM bright-field image of FeCo MNPs is shown in Fig. 1a. The nanoparticles are primarily cubic in shape. The selected area electron diffraction (SAED) pattern (Fig. 1b) matches well with the diffraction pattern of body centered cubic (bcc) FeCo crystal structure, and no other crystal structure was detected. The cubic morphology and sharp diffraction rings indicate that the FeCo MNPs are well crystallized. As shown in Fig. 1c, the shell (FeCo oxides) and the core region have different contrast. It tells that several nanometers shell of oxides was formed after FeCo MNPs exposed to air. The oxide shell can facilitate further surface modification and bioconjugation considering the fast-developing techniques used in ferrite particles. The EDS analysis of FeCo MNPs is presented in Fig. 1d. The strongest Si peak was originated from the wafer substrate, and the peaks of Fe and Co were also detected. Quantitative analysis shows that the atomic ratio of Fe to Co for the nanoparticles is 70.06 to 29.94, which is very close to the composition of Fe70 Co30 target. The composition of FeCo MNPs was also confirmed by ICP measurement (70.67/29.33). In addition, Fe70 Co30 exhibited the highest magnetic moment among Fex Coy alloys according to the Slater–Pauling curve [20]. Fig. 2a shows that the FeCo MNPs have an average size of 15 nm with narrow size distribution (10.1% standard deviation), and a total number of 208 particles was measured for this sample. Superparamagnetic behavior was found in the prepared FeCo MNPs. Its saturation magnetization is up to 220 emu/g (Fig. 2b), which is close to the value for bulk material (Ms, 245 emu/g) [21]. As listed in Table 1, a comparison is made between FeCo nanoparticles and other common MNPs in terms of saturation magnetization. The FeCo MNPs made in this study have a high saturation magnetization even through their size is small. The responses to external static magnetic fields for FeCo and iron oxide nanoparticles (15 nm, 44.6 emu/g) were compared and presented in Fig. 2c. It is apparent that most of FeCo MNPs were attracted to the bottom of the tube within 60 s and supernatant liquid was clear. In contrast, the solution of iron oxide nanoparticles showed no obvious change after 60 s magnetic separation. The time of iron oxides separation needed up to 12 h in this condition. Therefore, FeCo MNPs will be more effective in biomedical areas such as hyperthermia, and separation of proteins and cells. 3.2. Surface modification of FeCo MNPs The surface of FeCo MNPs was modified via APTES–Glu method for conjugating proteins, and the process was illustrated in Scheme 1a. APTES first undergoes a condensation reaction on the
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8 4
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
Fig. 1. TEM bright-field images of FeCo nanoparticles at (a) low and (c) high magnifications. (b) Selected area electron diffraction pattern and (d) EDS spectrum of FeCo nanoparticles.
Fig. 2. (a) Size distribution of FeCo nanoparticles with mean size of 15.0 nm and a 10.1% distribution. Estimated numbers = 208. Line is fitting by using a normal distribution. (b) Hysteresis loop of FeCo nanoparticles at 300 K. (c) Photographs of FeCo nanoparticles and iron oxide nanoparticles solutions on a bar magnet under different times. The iron oxide nanoparticles with a diameter of 15 nm were provided by Ocean NanoTech, LLC.
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
5
Table 2 XPS binding energy (B.E.) and the share of different components. Samplesa
C1s Sample 1 Sample 2 Sample 3
C C/C H
C–N/C N
a
O C O
Share (%)
B.E. (eV)
Share (%)
B.E. (eV)
Share (%)
B.E. (eV)
Share (%)
284.6 284.6 284.6
76.4 74.9 61.7
– 285.7 285.6
– 15.0 24.4
286.3 286.5 286.8
14.9 6.1 8.2
288.8 288.6 288.6
8.7 4.0 5.7
–N
N1s Sample 1 Sample 2 Sample 3
C O/C O
B.E. (eV)
–NH3 +
–NH2
B.E. (eV)
Share (%) B.E. (eV)
B.E. (eV)
Share (%) B.E. (eV)
B.E. (eV)
Share (%) B.E. (eV)
— — 398.6
— — 18.0
— 399.6 399.9
— 75.0 67.7
— 401.6 401.4
— 25.0 14.5
Samples: Sample 1, bare FeCo; Sample 2, APTES modified FeCo; Sample 3, Glu activated FeCo.
nanoparticle surfaces, causing the linking of APTES with the particles. Then, Glu solution was added, and an aldehyde group reacted with the amino group of APTES via Schiff base formation. Another remained aldehyde group could be used for reacting with the active amino group on proteins. As a crosslinking agent with wide applications, Glu can react rapidly with NH2 rather than NH3 + . Therefore, Glu has its greatest potency in the neutral to alkaline pH range. On the other hand, Glu activity could be reduced in alkaline solution due to self-polymerization such as aldol condensation [53]. Thus, Glu treatment is carried out in PBS buffer to maintain solution pH around neutrality. Herein XPS technique is employed to verify the chemical states during modification processes. The C1s and N1s high-resolution spectra for different samples are presented in Fig. 3. The binding energy and the share of individual components are also summarized in Table 2. For bare FeCo MNPs, the C1s peak (Fig. 3a) has been detected, and the C1s spectra can be deconvoluted into three different components. These components are associated with the attached PEG, and may also be related to
surface contamination introduced by explosure to air and possible carbon contamination from the vacuum system [54–56]. The peak located at 284.6 eV is attributed to hydrocarbons (C C/C H), and the other two peaks centered at 283.6 and 288.8 eV correspond to C O/C O and O C O, respectively [57,58]. In Fig. 3b, no obvious peak for N1s is detected for bare FeCo nanoparticles. After APTES modification, the new peak at 285.7 eV in Fig. 3c can be assigned to C N/C N. Two peaks centered at 399.6 eV and 401.6 eV are observed for N1s (Fig. 3d), which proves the presence of the nitrogen-containing APTES on surfaces of FeCo particle. The peak at 399.6 eV corresponds to the free amino groups ( NH2 ) and another peak at 401.6 eV is assigned to hydrogen-bonded amine or positively charged amino form ( NH3 + ) [59]. When the Glu molecules were further attached to APTES modified FeCo MNPs, the surface C NH2 bond could change to C N C . The C1s peak for C N/C N is strongly increased, and the N1s peak at 398.6 eV associated with N is also observed (Fig. 3f). Meanwhile, the percentages of NH2 and NH3 + are reduced. From the share information of N , roughly 18.0 percent of amino groups on surfaces of APTES modified FeCo reacted with the Glu in this treatment step. 3.3. Proteins loading on FeCo MNPs
Fig. 3. XPS C1s and N1s core-level spectra of (a) and (b) bare FeCo nanoparticles, (c) and (d) APTES modified FeCo nanoparticles, (e) and (f) Glu activated FeCo nanoparticles.
After APTES-Glu modification, the free and outward aldehyde groups on particle surfaces would be active for protein conjugation. Streptavidin and IgG antibody contain active amino groups, which can react with aldehyde for forming a Schiff base as the proteins are covalently immobilized. To quantitatively evaluate loading of these three proteins, fluorescent labeling was introduced during protein conjugation. Herein streptavidin-AF555, Dylight 550 labeled PAPPA antibody and Nectin-4 antibody were used. To determine the maximum level of these three proteins that could be loaded on the FeCo MNPs, varying ratio of protein to particle was attempted. The detailed loading information of these three proteins on FeCo MNPs is presented in Fig. 4. In all the three cases, the amount of bound proteins increased as the ratio of protein to particle was increased at the beginning, and the proteins loading appeared to saturate as the added ratio reached a value. The maximum loading amount for the three proteins is 95 g of streptavidin, 128 g of PAPP-A, and 125 g of Nectin-4 antibody per milligram of FeCo MNPs. As a known homo-tetramer, streptavidin has a size of around 5 nm [60,61]. If a compact monolayer of streptavidin is formed on surfaces of the cubic FeCo MNPs (15 nm), one FeCo nanoparticle can roughly bind with 54 streptavidin molecules, and this is approximate 300 g of streptavidin per milligram of FeCo nanoparticles. Actually, the true case is more complicated, and it is hard to achieve this theoretical coverage. In our condition for the maximum loading, it appears that 31.6 percent of available loading surface
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8 6
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
Fig. 4. Amount of the three proteins loaded on Glu activated FeCo nanoparticles. Points, mean; error bars, standard deviation; n = 3.
has been occupied by streptavidin. The structure of IgG antibody is briefly illustrated in Scheme 2. It is a Y-shaped molecule that contains two Fab fragments (the arm) and an Fc fragment (the stalk), and each Fab has one binding site for specific antigen. Based on the three-dimensional size of IgG antibody (approximately, width = 15.0 nm, height = 10.0 nm, thickness = 5.0 nm) [62,63], the loaded number of antibody can vary from 6 to 24 depending on the binding orientation in the theoretical coverage. Theoretical minimum and maximum loadings are approximately equal to 90.4 and 361.5 g of IgG per milligram of FeCo nanoparticles. From the
Scheme 2. The structure of IgG antibody and the orientations of IgG immobilization on particle surfaces.
loading results of PAPP-A and Nectin-4 antibodies, multiple binding orientations should exist for the two samples. The orientation of antibodies on particle surfaces is crucial for its further applications due to the position of antigen-binding sites. Three possible orientations are presented in Scheme 2, and other orientations may also exist in real samples. As an antibody is immobilized on a particle surface via Fc fragment, the two Fab fragments are away from the binding surface (Scheme 2a), so the two antigen-binding sites
Fig. 5. Fluorescence microscopy images of (a) streptavidin-AF555 binding to PAPP-A biotinylated antibody immobilized surface, (b) Dylight550-PAPP-A antibody binding to PAPP-A antigen immobilized surface, and (c) Dylight550-Nectin-4 antibody binding to Nectin-4 antigen immobilized surface. All the scale bars are 200 m.
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
7
Fig. 6. Fluorescence microscopy images. (a) The solution of streptavidin-AF555 labeled FeCo nanoparticles was placed on a strong magnet for thoroughly pulling down particles to the bottom, then the supernatant was binding to PAPP-A biotinylated antibody immobilized surface. (b) Streptavidin-AF555, (c) Dylight 550-PAPP-A antibody and (d) Dylight 550-Nectin-4 antibody labeled FeCo nanoparticles bind to PAPP-A biotinylated antibody, PAPP-A antigen, and Nectin-4 antigen immobilized surfaces, respectively. All the scale bars are 100 m.
would be both active. This orientation can make the antibody be fully used for binding two antigens. In Scheme 2b, one Fab fragment is closely linked with particle surface, thus, antigen cannot bind to this blocked binding site even though the site may be still active. Another outward Fab fragment can be used for antigen binding. The most undesirable orientation is that the two Fab fragments are both bound on the surface (Scheme 2c). In this case, both two binding sites are blocked and the antibody has no chance to bind with an antigen. Since the active amino groups can be found in both Fc and Fab fragments, multiple orientations should always occur on amine-reactive surfaces. Increasing the amount of immobilized antibodies is a way to improve the antigen-binding capability of antibody–particle conjugates. 3.4. Binding to the surface of functionalized slide A direct two-layer binding method was used (see experimental section) to test the bioactivity of the synthesized three protein–FeCo conjugates. First, immobilized proteins on slide surfaces were tested. In Fig. 5a, the printed microarray spots were clearly observed under fluorescence microscope after the incubation of streptavidin-AF555, indicating that PAPP-A biotinylated antibody was successfully immobilized on the NHS ester activated slide, and the immobilized biotinylated antibody was bound to streptavidin-AF555. From the microarray image, the diameter of one spot is about 200 m. The spots quality is good and the background signal of non-specific binding is low. Fig. 5b and 5c also suggest that PAPP-A and Nectin-4 antigens have both been captured on the slide surfaces, and their subsequent binding to fluorescence dye labeled antibodies has also been achieved. To ensure that the fluorescence signals are generated by the three conjugates, the solutions were magnetically separated, and the supernatants without conjugates were respectively incubated on the three functionalized slides. Free proteins in the conjugate solutions would compete with the conjugates for binding, leading to fewer bound protein–FeCo conjugates. The result for the supernatant of streptavidin-AF555 labeled FeCo solution is shown in Fig. 6a, and no obvious fluorescence signal is observed. Fig. 6b, in which the fluorescence signal of spots is detected, shows a significant contrast as comparing with Fig. 6a. This implies that the fluorescence signal is originated from streptavidin-AF555 coated
on FeCo nanoparticles, and this conjugate is successfully bound to PAPP-A biotinylated antibody immobilized surface. In our experiments, the supernatants of Dylight 550-PAPP-A antibody and Dylight 550-Nectin-4 antibody labeled FeCo solutions after magnetic separation have been tested, with similar results to that shown in Fig. 6a. The fluorescence images of the two kinds of antibody labeled FeCo nanoparticles (Fig. 6c and d) show that the Dylight 550-PAPP-A and Dylight 550-Nectin-4 antibody have been immobilized on FeCo particle surfaces, and the two antibody–FeCo conjugates can respectively bind to the PAPP-A and Nectin-4 antigens. FeCo MNPs could be more efficient in biomedical applications than iron oxide MNPs since they have much higher magnetic moment. Protein–FeCo complexes could provide an easy way for efficient, automated purification of proteins and cell sorting. Magnetic activated cell sorting (MACS) is much gentler than centrifugation. It is compatible with cells, and isolated cells could be directly used for further culturing and flow cytometry analysis. Protein–FeCo conjugates may also have a potential application in magnetic hyperthermia, which has been investigated for cancer treatment. Magnetic guidance could improve the delivery of MNPs to a localized tumor region, which is then specifically and exclusively heated to death by means of the magnetic losses of MNPs in an alternating magnetic field. Additionally, in GMR bio-sensing for molecular diagnostics, FeCo MNPs could have an advantage over iron oxide MNPs in improving the sensitivity due to their much higher magnetic moment. 4. Conclusion Bimetallic FeCo nanoparticles have superior magnetic properties, and could be potentially applied in a variety of biomedical areas. Fe70 Co30 MNPs with cubic shape were successfully prepared using a magnetron–sputtering-based gas phase condensation method. APTES and glutaraldehyde (Glu) were successfully modified on the nanoparticle surfaces, and active aldehyde groups were able to capture biomolecules that contain amine groups. This APTES–Glu surface modification could be applied to other kinds of nanoparticles for conjugating biomolecules. Streptavidin, PAPP-A antibody and Nectin-4 antibody were successfully immobilized on APTES-Glu modified FeCo nanoparticles, and their loading levels
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
8
were also explored. Multiple immobilizing orientations of PAPP-A and Nectin-4 antibodies on FeCo surfaces were found. Fluorescence microscopy showed that the resulting three kinds of protein–FeCo conjugates retained their bioactivity and could bind to functionalized surfaces. Acknowledgments This work was supported by the GMR Innovation Grant through the Office for Technology Commercialization, University of Minnesota. Parts of this work were carried out in the Characterization Facility, University of Minnesota, which receives partial support from NSF through the MRSEC program. Wei Wang also acknowledges the financial support from China Scholarship Council. References [1] X. Huang, I.H. El-Sayed, W. Qian, M.A. El-Sayed, J. Am. Chem. Soc. 128 (2006) 2115–2120. [2] W. Wu, T. Zhou, A. Berliner, P. Banerjee, S. Zhou, Chem. Mater. 22 (2010) 1966–1976. [3] X. Fang, M. Han, G. Lu, W. Tua, Z. Dai, Sens. Actuators B 168 (2012) 271–276. [4] H.S. Mansura, J.C. Gonzalez, A.A.P. Mansur, Colloids Surf. B 84 (2011) 360–368. [5] N. Nasongkla, E. Bey, J. Ren, H. Ai, C. Khemtong, J.S. Guthi, S.F. Chin, A.D. Sherry, D.A. Boothman, J. Gao, Nano Lett. 6 (2006) 2427–2430. [6] J.H. Lee, K. Lee, S.H. Moon, Y. Lee, T.G. Park, J. Cheon, Angew. Chem. Int. Ed. 48 (2009) 4174–4179. [7] J. Xie, J. Huang, X. Li, S. Sun, X. Chen, Curr. Med. Chem. 16 (2009) 1278–1294. [8] S. Yu, G. Wu, X. Gu, J. Wang, Y. Wang, H. Gao, J. Ma, Colloids Surf. B 103 (2013) 15–22. [9] J. Jabera, E. Mohsen, Colloids Surf. B 102 (2013) 265–272. [10] A. Kaushik, R. Khan, P.R. Solanki, P. Pandey, J. Alam, S. Ahmad, B.D. Malhotra, Biosens. Bioelectron. 24 (2008) 676–683. [11] M. Arruebo, M. Valladares, A. Gonzalez-Fernandez, J. Nanomater. (2009) 439389. [12] V. Tzitzios, G. Basina, D. Niarchos, W. Li, G. Hadjipanayis, J. Appl. Phys. 109 (2011) 07A313. [13] M. Zamanpour, Y. Chen, B. Hu, K. Carroll, Z.J. Huba, J. Appl. Phys. 111 (2012) 07B528. [14] L.M. Lacroix, S. Lachaize, A. Falqui, M. Respaud, B. Chaudret, J. Am. Chem. Soc. 131 (2009) 549–557. [15] J. Bai, J.P. Wang, Appl. Phys. Lett. 87 (2005) 152502. [16] X.W. Wei, G.X. Zhu, Y.J. Liu, Y.H. Ni, Y. Song, Z. Xu, Chem. Mater. 20 (2008) 6248–6253. [17] C. Wang, D.R. Baer, J.E. Amonette, M.H. Engelhard, J. Antony, Y. Qiang, J. Am. Chem. Soc. 131 (2009) 8824–8832. [18] K.N. Collier, N.J. Jones, K.J. Miller, Y.L. Qin, D.E. Laughlin, M.E. McHenry, J. Appl. Phys. 105 (2009) 07A328. [19] F. Dumestre, B. Chaudret, C. Amiens, P. Renaud, F. Peter, Science 303 (2004) 821–823. [20] N.A. Spaldin, Magnetic Materials: Fundamentals and Applications, Cambridge University Press, Cambridge, UK, 2010, pp. 76–77. [21] G.S. Chaubey, G. Barcena, N. Poudyal, C. Rong, J. Gao, S. Sun, J.P. Liu, J. Am. Chem. Soc. 129 (2007) 7214–7215. [22] M. Hesani, A. Yazdani, B. Abedi Ravan, M. Ghazanfari, Solid State Commun. 150 (2010) 594–597. [23] N.E. Feninechea, R. Hamzaouib, O. El Kedim, Mater. Lett. 57 (2003) 4165–4169.
[24] W.S. Seo, J.H. Lee, X. Sun, Y. Suzuki, D. Mann, Z. Liu, M. Terashima, P.C. Yang, M.V. McConnell, D.G. Nishimura, H. Dai, Nat. Mater. 5 (2006) 971–976. [25] H. Happy, S.R. Mohanty, P. Lee, T.L. Tan, S.V. Springham, A. Patran, R.V. Ramanujan, R.S. Rawat, Appl. Surf. Sci. 252 (2006) 2806–2816. [26] C. Wang, S. Peng, L.M. Lacroix, S. Sun, Nano Res. 2 (2009) 380–385. [27] J. Bai, Y.H. Xu, J.P. Wang, IEEE Trans. Magn. 43 (2007) 3340–3342. [28] Y.H. Xu, J. Bai, J.P. Wang, J. Magn. Magn. Mater. 311 (2007) 131–134. [29] A. Al-Saadi, C.H. Yu, V.V. Khutoryanskiy, S.J. Shih, A. Crossley, S.C. Tsang, J. Phys. Chem. C 113 (2009) 15260–15265. [30] R. Danczyk, B. Krieder, A. North, T. Webster, H. HogenEsch, A. Rundell, Biotechnol. Bioeng. 84 (2003) 215–223. [31] D. Qin, X. He, K. Wang, X.J. Zhao, W. Tan, J. Chen, J. Biomed. Biotechnol. (2007) 89364. [32] J.S. Kim, C. Alexander Valencia, R. Liu, W. Lin, Bioconjugate Chem. 18 (2007) 333–341. [33] M. De, S. Rana, V.M. Rotello, Macromol. Biosci. 9 (2009) 174–178. [34] M. Fuentes, C. Mateo, J.M. Guisan, R. Fernandez-Lafuente, Biosens. Bioelectron. 20 (2005) 1380–1387. [35] Y. Jung, J.Y. Jeonga, B.H. Chung, Analyst 133 (2008) 697–701. [36] S. Ju, W.S. Yeo, Nanotechnology 23 (2012) 135701. [37] E.R. Balmayor, I. Pashkuleva, A.M. Frias, H.S. Azevedo, R.L. Reis, J. R. Soc. Interface 8 (2011) 896–908. [38] Y. Jung, J.M. Lee, J.W. Kim, J. Yoon, H. Cho, B.H. Chung, Anal. Chem. 81 (2009) 936–942. [39] L. Wang, H.Y. Park, S.I. Lim, M.J. Schadt, D. Mott, J. Luo, X. Wang, C.J. Zhong, J. Mater. Chem. 18 (2009) 2629–2635. [40] S.J. Cho, B.R. Jarrett, A.Y. Louie, S.M. Kauzlarich, Nanotechnology 17 (2006) 640–644. [41] J.M. Nam, C.S. Thaxton, C.A. Mirkin, Science 301 (2003) 1884–1886. [42] H. Gu, K. Xu, C. Xu, B. Xu, Chem. Commun. 9 (2006) 941–949. [43] M. Takahashi, T. Yoshino, H. Takeyama, T. Matsunaga, Biotechnol. Prog. 25 (2009) 219–226. [44] D.W. Inglis, R. Riehn, R.H. Austin, J.C. Sturm, Appl. Phys. Lett. 85 (2004) 5093–5095. [45] H.Y. Park, M.J. Schadt, L. Wang, I.S. Lim, P.N. Njoki, S.H. Kim, M.Y. Jang, J. Luo, C.J. Zhong, Langmuir 23 (2007) 9050–9056. [46] P. Zhu, D.R. Shelton, S. Li, D.L. Adams, J.S. Karns, P. Amstutz, C.M. Tang, Biosens. Bioelectron. 30 (2011) 337–341. [47] L. Ma, C. Wang, M. Zhang, Sens. Actuators B 160 (2011) 650–655. [48] J.B. Haun, T.J. Yoon, H. Lee, R. Weissleder, WIREs Nanomed. Nanobiotechnol. 2 (2010) 291–304. [49] R.S. Gaster, L. Xu, S.J. Han, R.J. Wilson, D.A. Hall, S.J. Osterfeld, H. Yu, S.X. Wang, Nat. Nanotechnol. 6 (2011) 314–320. [50] J.P. Wang, J.M. Qiu, T.A. Taton, B.S. Kim, IEEE Trans. Magn. 42 (2006) 3042–3047. [51] A. Demortiere, P. Panissod, B.P. Pichon, G. Pourroy, D. Guillon, B. Donnio, S. Begin-Colin, Nanoscale 3 (2011) 225–232. [52] D.H. Kim, Y.T. Thai, D.E. Nikles, C.S. Brazel, IEEE Trans. Magn. 45 (2009) 64–70. [53] M. Isabelle, D. Catherine, J.B. Michel, C.W. Karen, BioTechniques 37 (2004) 790–802. [54] Y. Yamamoto, Y. Matsumoto, H. Koinuma, Appl. Surf. Sci. 238 (2004) 189–192. [55] Y. Mizuno, F.K. King, Y. Yamauchi, T. Homma, A. Tanaka, Y. Takakuwa, T. Momose, J. Vac. Sci. Technol. A 20 (2002) 1716–1721. [56] L. Zhao, Q. Jiang, J. Lian, Appl. Surf. Sci. 254 (2008) 4620–4625. [57] T. Ramanathan, F.T. Fisher, R.S. Ruoff, L.S. Brinson, Chem. Mater. 17 (2005) 1290–1295. [58] M. Dhayal, D.M. Ratner, Langmuir 25 (2009) 2181–2187. [59] M. Driffield, D.M. Goodall, A.S. Klute, D.K. Smith, K. Wilson, Langmuir 18 (2002) 8660–8665. [60] A. Kuzuya, K. Numajiri, M. Kimura, M. Komiyama, Nucleic Acids Symp. Ser. 52 (2008) 681–682. [61] T. Ooya, T. Kawashima, N. Yui, Biotechnol. Bioprocess Eng. 6 (2001) 293–300. [62] R.J. Leatherbarrow, M. Stedman, T.N.C. Wells, J. Mol. Biol. 221 (1991) 361–365. [63] K. Wadu-Mesthrige, N.A. Amro, G.Y. Liu, Scanning 22 (2000) 380–388.
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
ARTICLE IN PRESS Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb
Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins Wei Wang a,b , Ying Jing a , Shihai He a , Jian-Ping Wang a,∗ , Jian-Ping Zhai b,∗∗ a b
Department of Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN 55455, USA State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing 210046, PR China
a r t i c l e
i n f o
Article history: Received 6 June 2013 Received in revised form 27 November 2013 Accepted 29 November 2013 Available online xxx Keywords: FeCo magnetic nanoparticles Protein Surface modification Immobilization
a b s t r a c t Magnetic Fe70 Co30 nanoparticles with a cubic shape and a mean size of 15 ± 1.5 nm were fabricated using a magnetron–sputtering-based gas phase condensation deposition method. The particles had a high saturation magnetization of 220 emu/g, which is much higher than that of commercially available iron oxide nanoparticles. The FeCo nanoparticles were modified by 3-aminopropyltriethoxy silane and subsequently activated by glutaraldehyde, leading to successful attachment of aldehyde groups onto nanoparticle surfaces. Three proteins, namely streptavidin, PAPP-A antibody and Nectin-4 antibody, were immobilized on glutaraldehyde activated FeCo nanoparticles, and their loading levels were quantitatively evaluated. Our results show that loading capabilities are 95 g of streptavidin, 128 g of PAPP-A, and 125 g of Nectin-4 antibody per milligram of FeCo nanoparticles, and that the three immobilized proteins retain their binding bioactivity. The protein–FeCo conjugates may find valuable applications involving magnetic separation and purification of proteins and cells, and the magnetic detection of biomolecules. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Nanomaterials, having attracted much attention for decades as a result of their special and superior properties, are increasingly being applied in multidiscipline areas, especially in the fields of biology and biomedicine. Among the nanomaterials, many kinds of nanoparticles have been studied depending on their intrinsic properties, such as optical absorption (Au, Ag) [1,2], photoluminescence in the form of fluorescence (quantum dots, e.g. CdSe, CdS) [3,4], and magnetic moment (e.g. iron oxides) [5,6]. These moststudied nanoparticles can be potentially and multi-functionally used as diagnostic and therapeutic agents, delivery vehicles and imaging agents. In 1996, iron oxide magnetic nanoparticle (MNPs) based ferumoxides (Feridex I.V.) was approved by the US Food and Drug Administration (FDA) as an imaging agent for the detection of liver lesions, and now iron oxide MNPs have also been widely used in MRI contrast enhancement, hyperthermia, separation and purification of biomolecules, drug and gene delivery, and bimolecular detection [7–10]. MNPs with different compositions and sizes have been commercialized by a number of companies [11]. Most of these commercially available MNPs are iron oxides, which are easy to synthesize, low cost, and biocompatible. These proven qualities have further driven researchers to develop other kinds of
∗ Corresponding author. Tel.: +1 612 625 9509; fax: +1 612 625 4583. ∗∗ Corresponding author. Tel.: +86 025 83592903; fax: +86 025 83592903. E-mail addresses: [email protected] (J.-P. Wang), [email protected] (J.-P. Zhai).
MNPs. In addition to iron oxides, much attention has recently been paid to high-magnetic-moment materials [12–19]. Among them, bimetallic FeCo MNPs have the highest saturation magnetization (Ms), which makes it an attractive material choice for application to writing heads of hard disk drives [20]. Some methods have been reported for the fabrication of FeCo MNPs including reductive decomposition of Fe(III) acetylacetonate and Co(II) acetylacetonate [21], chemical coprecipitation [22], mechanical ball milling [23], chemical vapor deposition [24], pulsed laser ablation deposition [25], and interfacial diffusion [26]. Our FeCo MNPs were synthesized in gas phase using a magnetron–sputtering-based system, which can control the shape (cubic or spherical), sizes and composition of particles [27,28]. In order to fulfill full potential of these high-magnetic-moment nanoparticles for various biomedical applications, it is crucial to find and develop reliable and repeatable processes for their surface functionalization and protein conjugation. Basically, two different approaches can be employed to conjugate proteins to nanoparticles. The first method is through non-covalent interactions, such as electrostatic attraction [29], protein A interaction [30,31], and nickel-mediated interactions with His-tagged proteins [32,33]. The other method uses the direct covalent linkage between proteins and nanoparticle surfaces to form a conjugate that is stable regarding possible dissociation in complex biological media with other interfering biomolecules. The nanoparticle surfaces need prior modification with active groups, such as cyanogen bromide, aldehyde, epoxy, N-hydroxysuccinimide ester, and primary amino groups [34–37], that can be covalently bound to the
0927-7765/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
2
Scheme 1. Schematic illustration of an experimental overview.
carboxyl or amino groups of proteins. Some other techniques have also been developed for this purpose. Jung [38] used a site-specific photo cross-linker (benzophenone), and Wang [39] and Cho [40] developed Au coating on particle surfaces. In this work, the immobilization of proteins on FeCo nanoparticles was carried out via anchored aldehyde groups. 3-Aminopropyltriethoxysilane (APTES) was first used to modify particle surfaces, and then glutaraldehyde (Glu), an amine-reactive homobifunctional cross-linker used in biochemistry applications, could be bound to APTES leaving another aldehyde group at the end for capturing proteins (Scheme 1a). Conjugates of protein–MNPs have wide applications in biomedicine areas and have been extensively used in magnetic separation and purification of proteins and cells [41–46]. Additionally, Ma [47] used antibody–MNPs in a magnetic force induced microcantilever biosensor. These conjugates were also used in diagnostic magnetic resonance (DMR) and giant magnetoresistive (GMR) sensors [48,49]. Herein, three model proteins were selected for loading on the surfaces of FeCo MNPs with enhanced properties for the potential applications listed above. One is streptavidin, which has an extraordinarily high non-covalent affinity for biotin, and the biotin–streptavidin system is one of the most widely used affinity pairs in molecular, immunological, and cellular assays. The other two selected proteins were PAPP-A (Pregnancy-associated plasma protein A) and Nectin-4 monoclonal IgG antibodies. The loading amounts of these three proteins on FeCo MNPs were evaluated while binding bioactivity was confirmed for each protein.
2. Materials and methods 2.1. Materials 3-Aminopropyltriethoxy silane, 99% (APTES) was purchased from Acros organics. Glutaraldehyde solution (Grade II, 25% in H2 O), ethanol (≥99.5%, 200 proof), and tween 20 was purchased from Sigma–Aldrich. Phosphate buffered saline (PBS, 1×, pH 7.4) was purchased from Invitrogen. mPEG-amine (5 kDa) was purchased from Creative PEGWorks. Streptavidin-AF555 (2 mg/mL in PBS, pH 7.2) was purchased from Invitrogen. Recombinant human Nectin-4 [catalog number 2659-N4], monoclonal anti-human Nectin4 antibody [catalog number MAB2659], recombinant human
PAPP-A [catalog number 2487-ZN], monoclonal anti-human PAPPA antibody [catalog number MAB2487], and polyclonal anti-human PAPP-A biotinylated antibody [catalog number BAF2487] were purchased from R&D Systems, Inc., Minneapolis, USA. DyLight550 Microscale Antibody Labeling Kit [catalog number 84531] was purchased from Pierce, Thermo Scientific. The NHS ester functionalized microarray slide (No. DHD1-0023) was provided by SurModics, Inc. 2.2. Fabrication of FeCo MNPs FeCo MNPs were fabricated by using a magnetron–sputteringbased gas phase condensation deposition system [27,28]. The sputtering source used a planer target of FeCo alloy with atomic composition 70:30. Argon gas was used as sputtering gas and carrier gas. The FeCo nanoparticles were generated in the source chamber, and carried into the collection chamber by argon gas flow. Polyethylene glycol (PEG) pre-coated glass slides were placed in the collection chamber for collecting nanoparticles [50]. After deposition, the glass slides were taken out and washed by deionized (DI) water to transfer FeCo MNPs into solution. The water solution of nanoparticles was stored at 4–8 ◦ C for further modification and application. The concentration and composition of prepared samples were determined by ICP-OES (Optima 3000 DV, Perkin–Elmer). FeCo MNPs were dropped on a silicon wafer and dried using nitrogen gas. The particles on the wafer were further analyzed by an energy dispersive spectrometer (EDS) that was attached on a JEOL 6500 scanning electron microscope. Morphology and crystal structure of FeCo MNPs were characterized by a transmission electron microscope (TEM). Magnetic behavior measurement of FeCo MNPs was carried out at 300 K by a vibrating sample magnetometer (VSM, model 3900, Princeton Measurements Corporation). 2.3. Modification by 3-aminopropyltriethoxysilane (APTES) and glutaraldehyde FeCo MNPs (100 g Fe) were washed with ethanol three times, and were dispersed in 500 L ethanol/water (95/5 in volume ratio) solution with sonication. In the washing steps, the particles were harvested and separated by applying an external magnetic field. 10 L APTES was dissolved in 500 L ethanol/water (95/5) solution and stirred for 1 min, and then the APTES solution was added to
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
FeCo solution immediately. The reaction mixture was sonicated for 1 h. The particles were washed three times with acetone, followed by being washed three times with DI water. 500 L glutaraldehyde (Glu) solution (10%) was added to 500 L APTES modified FeCo MNPs solution (in PBS), and the mixture was sonicated for 5 h. The Glu activated nanoparticles were washed three times with DI water, and dispersed in PBST (PBS with 0.02% tween 20) and stored at 4–8 ◦ C. The samples were characterized by an SSX-100 spectrometer with an Al Ka X-ray source. Shirley background subtraction and properly curve-fitted peaks were applied in the XPS spectra peak fitting procedure. The quantification was based on peak areas calculated from the high-resolution spectra. All binding energies were referenced to the C 1 s neutral carbon peak at 284.6 eV. 2.4. Protein immobilization on Glu activated FeCo MNPs Monoclonal PAPP-A and Nectin-4 antibodies were both fluorescently labeled using Dylight 550 microscale antibody labeling kit [Pierce Inc., catalogue number 84531] according to the procedure provided by the vendor. 200 L of streptavidin-AF555 (25 g/mL in PBS) was added to 200 L of Glu activated nanoparticles solution (in PBST). After being sonicated for 30 min, the mixture was shaken and incubated for 12 h at 4 ◦ C using a mechanical rotor. After that, the particles were collected by a magnet and washed three times with 400 L PBS. The supernatant and washings (unbound streptavidin-AF555) were combined and made up to 5 mL using PBS for quantification. The experiments were repeated three times using fresh samples to obtain reliable error bars. To determine the concentration of unbound streptavidin-AF555, 100 L of combined solution (5 mL) was placed in a 96-well plate and the fluorescence intensity was compared to a standard curve. Fluorescence measurements were performed on a Cytofluor II Fluorescent Microplate Reader. During the fluorescence measurement, the fluorophore was excited at 530/25 nm and the emission was collected at 590/35 nm. For inactivating aldehyde groups and reducing non-specific binding, 400 L of 0.5 mg/mL mPEG-amine solution was added to the streptavidin-AF555 labeled FeCo particles, and the mixture was shaken for 1 h at room temperature. After being washing with PBS, the modified particles were dispersed in 1.0% bovine serum albumin (BSA) solution (in PBST), and stored at 4–8 ◦ C. In a similar way, Dylight 550 labeled PAPP-A antibody and Nectin-4 antibody were respectively conjugated to Glu activated FeCo MNPs. All immobilizations were quantitively analyzed, and the final conjugates were stored in 1.0% BSA solution containing 0.02% tween 20. 2.5. Binding to the surface of functionalized slide The process of streptavidin–FeCo conjugate binding to a functionalized slide is briefly illustrated in Scheme 1b, and other two antibody–FeCo conjugates also have similar binding processes. PAPP-A biotinylated antibody (20 g/mL), PAPP-A antigen (20 g/mL), and Nectin-4 antigen (20 g/mL) were spotted on NHS ester modified slides using the sciFLEXARRAYER S5 (Scienion, Germany), respectively. The printed slides were incubated for 24 h at 4 ◦ C under a relative humidity of ∼90%. After that, these modified slides were washed with immunoassay stabilizer (SurModics, Cat. No. SC01), and were blocked with 1.0 mg/mL BSA for 30 min. After being washed with PBST, the three functionalized slides were respectively incubated with streptavidin-AF555 (20 g/mL), Dylight 550-PAPP-A antibody (20 g/mL), and Dylight 550-Nectin4 antibody (20 g/mL) to test the binding activity of the three proteins on the slide surfaces. After incubating for 1 h at room temperature, the unbound fluorescence dye labeled proteins were washed off using PBST and DI water. The protein bound slides were dried by nitrogen gas, and were further imaged using Olympus IX70 Invert Fluorescence Microscope under identical camera conditions.
3
Table 1 Saturation magnetization of selected magnetic nanoparticles. Materials
Size (nm)
Ms (emu/g)
Ref.
Iron oxides (Fex Oy ) MnFe2 O4 Fe FeCo Fe50 Co50 Fe70 Co30
5–11 5–12 7 8–10 68 15
43–60 37–52 212 144.5 211.9 220
[51] [52] [19] [12] [16] This work
According to the procedure for fluorescence dye labeled protein binding, the three kinds of streptavidin-AF555, Dylight 550-PAPPA antibody and Dylight 550-Nectin-4 antibody labeled FeCo MNPs (with highest protein loading on particles) were also incubated on PAPP-A biotinylated antibody, PAPP-A antigen, and Nectin-4 antigen modified slides, respectively. The particles bound slides were also imaged using a Fluorescence Microscope. 3. Results and discussion 3.1. Properties of FeCo MNPs The TEM bright-field image of FeCo MNPs is shown in Fig. 1a. The nanoparticles are primarily cubic in shape. The selected area electron diffraction (SAED) pattern (Fig. 1b) matches well with the diffraction pattern of body centered cubic (bcc) FeCo crystal structure, and no other crystal structure was detected. The cubic morphology and sharp diffraction rings indicate that the FeCo MNPs are well crystallized. As shown in Fig. 1c, the shell (FeCo oxides) and the core region have different contrast. It tells that several nanometers shell of oxides was formed after FeCo MNPs exposed to air. The oxide shell can facilitate further surface modification and bioconjugation considering the fast-developing techniques used in ferrite particles. The EDS analysis of FeCo MNPs is presented in Fig. 1d. The strongest Si peak was originated from the wafer substrate, and the peaks of Fe and Co were also detected. Quantitative analysis shows that the atomic ratio of Fe to Co for the nanoparticles is 70.06 to 29.94, which is very close to the composition of Fe70 Co30 target. The composition of FeCo MNPs was also confirmed by ICP measurement (70.67/29.33). In addition, Fe70 Co30 exhibited the highest magnetic moment among Fex Coy alloys according to the Slater–Pauling curve [20]. Fig. 2a shows that the FeCo MNPs have an average size of 15 nm with narrow size distribution (10.1% standard deviation), and a total number of 208 particles was measured for this sample. Superparamagnetic behavior was found in the prepared FeCo MNPs. Its saturation magnetization is up to 220 emu/g (Fig. 2b), which is close to the value for bulk material (Ms, 245 emu/g) [21]. As listed in Table 1, a comparison is made between FeCo nanoparticles and other common MNPs in terms of saturation magnetization. The FeCo MNPs made in this study have a high saturation magnetization even through their size is small. The responses to external static magnetic fields for FeCo and iron oxide nanoparticles (15 nm, 44.6 emu/g) were compared and presented in Fig. 2c. It is apparent that most of FeCo MNPs were attracted to the bottom of the tube within 60 s and supernatant liquid was clear. In contrast, the solution of iron oxide nanoparticles showed no obvious change after 60 s magnetic separation. The time of iron oxides separation needed up to 12 h in this condition. Therefore, FeCo MNPs will be more effective in biomedical areas such as hyperthermia, and separation of proteins and cells. 3.2. Surface modification of FeCo MNPs The surface of FeCo MNPs was modified via APTES–Glu method for conjugating proteins, and the process was illustrated in Scheme 1a. APTES first undergoes a condensation reaction on the
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8 4
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
Fig. 1. TEM bright-field images of FeCo nanoparticles at (a) low and (c) high magnifications. (b) Selected area electron diffraction pattern and (d) EDS spectrum of FeCo nanoparticles.
Fig. 2. (a) Size distribution of FeCo nanoparticles with mean size of 15.0 nm and a 10.1% distribution. Estimated numbers = 208. Line is fitting by using a normal distribution. (b) Hysteresis loop of FeCo nanoparticles at 300 K. (c) Photographs of FeCo nanoparticles and iron oxide nanoparticles solutions on a bar magnet under different times. The iron oxide nanoparticles with a diameter of 15 nm were provided by Ocean NanoTech, LLC.
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
5
Table 2 XPS binding energy (B.E.) and the share of different components. Samplesa
C1s Sample 1 Sample 2 Sample 3
C C/C H
C–N/C N
a
O C O
Share (%)
B.E. (eV)
Share (%)
B.E. (eV)
Share (%)
B.E. (eV)
Share (%)
284.6 284.6 284.6
76.4 74.9 61.7
– 285.7 285.6
– 15.0 24.4
286.3 286.5 286.8
14.9 6.1 8.2
288.8 288.6 288.6
8.7 4.0 5.7
–N
N1s Sample 1 Sample 2 Sample 3
C O/C O
B.E. (eV)
–NH3 +
–NH2
B.E. (eV)
Share (%) B.E. (eV)
B.E. (eV)
Share (%) B.E. (eV)
B.E. (eV)
Share (%) B.E. (eV)
— — 398.6
— — 18.0
— 399.6 399.9
— 75.0 67.7
— 401.6 401.4
— 25.0 14.5
Samples: Sample 1, bare FeCo; Sample 2, APTES modified FeCo; Sample 3, Glu activated FeCo.
nanoparticle surfaces, causing the linking of APTES with the particles. Then, Glu solution was added, and an aldehyde group reacted with the amino group of APTES via Schiff base formation. Another remained aldehyde group could be used for reacting with the active amino group on proteins. As a crosslinking agent with wide applications, Glu can react rapidly with NH2 rather than NH3 + . Therefore, Glu has its greatest potency in the neutral to alkaline pH range. On the other hand, Glu activity could be reduced in alkaline solution due to self-polymerization such as aldol condensation [53]. Thus, Glu treatment is carried out in PBS buffer to maintain solution pH around neutrality. Herein XPS technique is employed to verify the chemical states during modification processes. The C1s and N1s high-resolution spectra for different samples are presented in Fig. 3. The binding energy and the share of individual components are also summarized in Table 2. For bare FeCo MNPs, the C1s peak (Fig. 3a) has been detected, and the C1s spectra can be deconvoluted into three different components. These components are associated with the attached PEG, and may also be related to
surface contamination introduced by explosure to air and possible carbon contamination from the vacuum system [54–56]. The peak located at 284.6 eV is attributed to hydrocarbons (C C/C H), and the other two peaks centered at 283.6 and 288.8 eV correspond to C O/C O and O C O, respectively [57,58]. In Fig. 3b, no obvious peak for N1s is detected for bare FeCo nanoparticles. After APTES modification, the new peak at 285.7 eV in Fig. 3c can be assigned to C N/C N. Two peaks centered at 399.6 eV and 401.6 eV are observed for N1s (Fig. 3d), which proves the presence of the nitrogen-containing APTES on surfaces of FeCo particle. The peak at 399.6 eV corresponds to the free amino groups ( NH2 ) and another peak at 401.6 eV is assigned to hydrogen-bonded amine or positively charged amino form ( NH3 + ) [59]. When the Glu molecules were further attached to APTES modified FeCo MNPs, the surface C NH2 bond could change to C N C . The C1s peak for C N/C N is strongly increased, and the N1s peak at 398.6 eV associated with N is also observed (Fig. 3f). Meanwhile, the percentages of NH2 and NH3 + are reduced. From the share information of N , roughly 18.0 percent of amino groups on surfaces of APTES modified FeCo reacted with the Glu in this treatment step. 3.3. Proteins loading on FeCo MNPs
Fig. 3. XPS C1s and N1s core-level spectra of (a) and (b) bare FeCo nanoparticles, (c) and (d) APTES modified FeCo nanoparticles, (e) and (f) Glu activated FeCo nanoparticles.
After APTES-Glu modification, the free and outward aldehyde groups on particle surfaces would be active for protein conjugation. Streptavidin and IgG antibody contain active amino groups, which can react with aldehyde for forming a Schiff base as the proteins are covalently immobilized. To quantitatively evaluate loading of these three proteins, fluorescent labeling was introduced during protein conjugation. Herein streptavidin-AF555, Dylight 550 labeled PAPPA antibody and Nectin-4 antibody were used. To determine the maximum level of these three proteins that could be loaded on the FeCo MNPs, varying ratio of protein to particle was attempted. The detailed loading information of these three proteins on FeCo MNPs is presented in Fig. 4. In all the three cases, the amount of bound proteins increased as the ratio of protein to particle was increased at the beginning, and the proteins loading appeared to saturate as the added ratio reached a value. The maximum loading amount for the three proteins is 95 g of streptavidin, 128 g of PAPP-A, and 125 g of Nectin-4 antibody per milligram of FeCo MNPs. As a known homo-tetramer, streptavidin has a size of around 5 nm [60,61]. If a compact monolayer of streptavidin is formed on surfaces of the cubic FeCo MNPs (15 nm), one FeCo nanoparticle can roughly bind with 54 streptavidin molecules, and this is approximate 300 g of streptavidin per milligram of FeCo nanoparticles. Actually, the true case is more complicated, and it is hard to achieve this theoretical coverage. In our condition for the maximum loading, it appears that 31.6 percent of available loading surface
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8 6
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
Fig. 4. Amount of the three proteins loaded on Glu activated FeCo nanoparticles. Points, mean; error bars, standard deviation; n = 3.
has been occupied by streptavidin. The structure of IgG antibody is briefly illustrated in Scheme 2. It is a Y-shaped molecule that contains two Fab fragments (the arm) and an Fc fragment (the stalk), and each Fab has one binding site for specific antigen. Based on the three-dimensional size of IgG antibody (approximately, width = 15.0 nm, height = 10.0 nm, thickness = 5.0 nm) [62,63], the loaded number of antibody can vary from 6 to 24 depending on the binding orientation in the theoretical coverage. Theoretical minimum and maximum loadings are approximately equal to 90.4 and 361.5 g of IgG per milligram of FeCo nanoparticles. From the
Scheme 2. The structure of IgG antibody and the orientations of IgG immobilization on particle surfaces.
loading results of PAPP-A and Nectin-4 antibodies, multiple binding orientations should exist for the two samples. The orientation of antibodies on particle surfaces is crucial for its further applications due to the position of antigen-binding sites. Three possible orientations are presented in Scheme 2, and other orientations may also exist in real samples. As an antibody is immobilized on a particle surface via Fc fragment, the two Fab fragments are away from the binding surface (Scheme 2a), so the two antigen-binding sites
Fig. 5. Fluorescence microscopy images of (a) streptavidin-AF555 binding to PAPP-A biotinylated antibody immobilized surface, (b) Dylight550-PAPP-A antibody binding to PAPP-A antigen immobilized surface, and (c) Dylight550-Nectin-4 antibody binding to Nectin-4 antigen immobilized surface. All the scale bars are 200 m.
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
7
Fig. 6. Fluorescence microscopy images. (a) The solution of streptavidin-AF555 labeled FeCo nanoparticles was placed on a strong magnet for thoroughly pulling down particles to the bottom, then the supernatant was binding to PAPP-A biotinylated antibody immobilized surface. (b) Streptavidin-AF555, (c) Dylight 550-PAPP-A antibody and (d) Dylight 550-Nectin-4 antibody labeled FeCo nanoparticles bind to PAPP-A biotinylated antibody, PAPP-A antigen, and Nectin-4 antigen immobilized surfaces, respectively. All the scale bars are 100 m.
would be both active. This orientation can make the antibody be fully used for binding two antigens. In Scheme 2b, one Fab fragment is closely linked with particle surface, thus, antigen cannot bind to this blocked binding site even though the site may be still active. Another outward Fab fragment can be used for antigen binding. The most undesirable orientation is that the two Fab fragments are both bound on the surface (Scheme 2c). In this case, both two binding sites are blocked and the antibody has no chance to bind with an antigen. Since the active amino groups can be found in both Fc and Fab fragments, multiple orientations should always occur on amine-reactive surfaces. Increasing the amount of immobilized antibodies is a way to improve the antigen-binding capability of antibody–particle conjugates. 3.4. Binding to the surface of functionalized slide A direct two-layer binding method was used (see experimental section) to test the bioactivity of the synthesized three protein–FeCo conjugates. First, immobilized proteins on slide surfaces were tested. In Fig. 5a, the printed microarray spots were clearly observed under fluorescence microscope after the incubation of streptavidin-AF555, indicating that PAPP-A biotinylated antibody was successfully immobilized on the NHS ester activated slide, and the immobilized biotinylated antibody was bound to streptavidin-AF555. From the microarray image, the diameter of one spot is about 200 m. The spots quality is good and the background signal of non-specific binding is low. Fig. 5b and 5c also suggest that PAPP-A and Nectin-4 antigens have both been captured on the slide surfaces, and their subsequent binding to fluorescence dye labeled antibodies has also been achieved. To ensure that the fluorescence signals are generated by the three conjugates, the solutions were magnetically separated, and the supernatants without conjugates were respectively incubated on the three functionalized slides. Free proteins in the conjugate solutions would compete with the conjugates for binding, leading to fewer bound protein–FeCo conjugates. The result for the supernatant of streptavidin-AF555 labeled FeCo solution is shown in Fig. 6a, and no obvious fluorescence signal is observed. Fig. 6b, in which the fluorescence signal of spots is detected, shows a significant contrast as comparing with Fig. 6a. This implies that the fluorescence signal is originated from streptavidin-AF555 coated
on FeCo nanoparticles, and this conjugate is successfully bound to PAPP-A biotinylated antibody immobilized surface. In our experiments, the supernatants of Dylight 550-PAPP-A antibody and Dylight 550-Nectin-4 antibody labeled FeCo solutions after magnetic separation have been tested, with similar results to that shown in Fig. 6a. The fluorescence images of the two kinds of antibody labeled FeCo nanoparticles (Fig. 6c and d) show that the Dylight 550-PAPP-A and Dylight 550-Nectin-4 antibody have been immobilized on FeCo particle surfaces, and the two antibody–FeCo conjugates can respectively bind to the PAPP-A and Nectin-4 antigens. FeCo MNPs could be more efficient in biomedical applications than iron oxide MNPs since they have much higher magnetic moment. Protein–FeCo complexes could provide an easy way for efficient, automated purification of proteins and cell sorting. Magnetic activated cell sorting (MACS) is much gentler than centrifugation. It is compatible with cells, and isolated cells could be directly used for further culturing and flow cytometry analysis. Protein–FeCo conjugates may also have a potential application in magnetic hyperthermia, which has been investigated for cancer treatment. Magnetic guidance could improve the delivery of MNPs to a localized tumor region, which is then specifically and exclusively heated to death by means of the magnetic losses of MNPs in an alternating magnetic field. Additionally, in GMR bio-sensing for molecular diagnostics, FeCo MNPs could have an advantage over iron oxide MNPs in improving the sensitivity due to their much higher magnetic moment. 4. Conclusion Bimetallic FeCo nanoparticles have superior magnetic properties, and could be potentially applied in a variety of biomedical areas. Fe70 Co30 MNPs with cubic shape were successfully prepared using a magnetron–sputtering-based gas phase condensation method. APTES and glutaraldehyde (Glu) were successfully modified on the nanoparticle surfaces, and active aldehyde groups were able to capture biomolecules that contain amine groups. This APTES–Glu surface modification could be applied to other kinds of nanoparticles for conjugating biomolecules. Streptavidin, PAPP-A antibody and Nectin-4 antibody were successfully immobilized on APTES-Glu modified FeCo nanoparticles, and their loading levels
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050
G Model COLSUB-6152; No. of Pages 8
ARTICLE IN PRESS W. Wang et al. / Colloids and Surfaces B: Biointerfaces xxx (2013) xxx–xxx
8
were also explored. Multiple immobilizing orientations of PAPP-A and Nectin-4 antibodies on FeCo surfaces were found. Fluorescence microscopy showed that the resulting three kinds of protein–FeCo conjugates retained their bioactivity and could bind to functionalized surfaces. Acknowledgments This work was supported by the GMR Innovation Grant through the Office for Technology Commercialization, University of Minnesota. Parts of this work were carried out in the Characterization Facility, University of Minnesota, which receives partial support from NSF through the MRSEC program. Wei Wang also acknowledges the financial support from China Scholarship Council. References [1] X. Huang, I.H. El-Sayed, W. Qian, M.A. El-Sayed, J. Am. Chem. Soc. 128 (2006) 2115–2120. [2] W. Wu, T. Zhou, A. Berliner, P. Banerjee, S. Zhou, Chem. Mater. 22 (2010) 1966–1976. [3] X. Fang, M. Han, G. Lu, W. Tua, Z. Dai, Sens. Actuators B 168 (2012) 271–276. [4] H.S. Mansura, J.C. Gonzalez, A.A.P. Mansur, Colloids Surf. B 84 (2011) 360–368. [5] N. Nasongkla, E. Bey, J. Ren, H. Ai, C. Khemtong, J.S. Guthi, S.F. Chin, A.D. Sherry, D.A. Boothman, J. Gao, Nano Lett. 6 (2006) 2427–2430. [6] J.H. Lee, K. Lee, S.H. Moon, Y. Lee, T.G. Park, J. Cheon, Angew. Chem. Int. Ed. 48 (2009) 4174–4179. [7] J. Xie, J. Huang, X. Li, S. Sun, X. Chen, Curr. Med. Chem. 16 (2009) 1278–1294. [8] S. Yu, G. Wu, X. Gu, J. Wang, Y. Wang, H. Gao, J. Ma, Colloids Surf. B 103 (2013) 15–22. [9] J. Jabera, E. Mohsen, Colloids Surf. B 102 (2013) 265–272. [10] A. Kaushik, R. Khan, P.R. Solanki, P. Pandey, J. Alam, S. Ahmad, B.D. Malhotra, Biosens. Bioelectron. 24 (2008) 676–683. [11] M. Arruebo, M. Valladares, A. Gonzalez-Fernandez, J. Nanomater. (2009) 439389. [12] V. Tzitzios, G. Basina, D. Niarchos, W. Li, G. Hadjipanayis, J. Appl. Phys. 109 (2011) 07A313. [13] M. Zamanpour, Y. Chen, B. Hu, K. Carroll, Z.J. Huba, J. Appl. Phys. 111 (2012) 07B528. [14] L.M. Lacroix, S. Lachaize, A. Falqui, M. Respaud, B. Chaudret, J. Am. Chem. Soc. 131 (2009) 549–557. [15] J. Bai, J.P. Wang, Appl. Phys. Lett. 87 (2005) 152502. [16] X.W. Wei, G.X. Zhu, Y.J. Liu, Y.H. Ni, Y. Song, Z. Xu, Chem. Mater. 20 (2008) 6248–6253. [17] C. Wang, D.R. Baer, J.E. Amonette, M.H. Engelhard, J. Antony, Y. Qiang, J. Am. Chem. Soc. 131 (2009) 8824–8832. [18] K.N. Collier, N.J. Jones, K.J. Miller, Y.L. Qin, D.E. Laughlin, M.E. McHenry, J. Appl. Phys. 105 (2009) 07A328. [19] F. Dumestre, B. Chaudret, C. Amiens, P. Renaud, F. Peter, Science 303 (2004) 821–823. [20] N.A. Spaldin, Magnetic Materials: Fundamentals and Applications, Cambridge University Press, Cambridge, UK, 2010, pp. 76–77. [21] G.S. Chaubey, G. Barcena, N. Poudyal, C. Rong, J. Gao, S. Sun, J.P. Liu, J. Am. Chem. Soc. 129 (2007) 7214–7215. [22] M. Hesani, A. Yazdani, B. Abedi Ravan, M. Ghazanfari, Solid State Commun. 150 (2010) 594–597. [23] N.E. Feninechea, R. Hamzaouib, O. El Kedim, Mater. Lett. 57 (2003) 4165–4169.
[24] W.S. Seo, J.H. Lee, X. Sun, Y. Suzuki, D. Mann, Z. Liu, M. Terashima, P.C. Yang, M.V. McConnell, D.G. Nishimura, H. Dai, Nat. Mater. 5 (2006) 971–976. [25] H. Happy, S.R. Mohanty, P. Lee, T.L. Tan, S.V. Springham, A. Patran, R.V. Ramanujan, R.S. Rawat, Appl. Surf. Sci. 252 (2006) 2806–2816. [26] C. Wang, S. Peng, L.M. Lacroix, S. Sun, Nano Res. 2 (2009) 380–385. [27] J. Bai, Y.H. Xu, J.P. Wang, IEEE Trans. Magn. 43 (2007) 3340–3342. [28] Y.H. Xu, J. Bai, J.P. Wang, J. Magn. Magn. Mater. 311 (2007) 131–134. [29] A. Al-Saadi, C.H. Yu, V.V. Khutoryanskiy, S.J. Shih, A. Crossley, S.C. Tsang, J. Phys. Chem. C 113 (2009) 15260–15265. [30] R. Danczyk, B. Krieder, A. North, T. Webster, H. HogenEsch, A. Rundell, Biotechnol. Bioeng. 84 (2003) 215–223. [31] D. Qin, X. He, K. Wang, X.J. Zhao, W. Tan, J. Chen, J. Biomed. Biotechnol. (2007) 89364. [32] J.S. Kim, C. Alexander Valencia, R. Liu, W. Lin, Bioconjugate Chem. 18 (2007) 333–341. [33] M. De, S. Rana, V.M. Rotello, Macromol. Biosci. 9 (2009) 174–178. [34] M. Fuentes, C. Mateo, J.M. Guisan, R. Fernandez-Lafuente, Biosens. Bioelectron. 20 (2005) 1380–1387. [35] Y. Jung, J.Y. Jeonga, B.H. Chung, Analyst 133 (2008) 697–701. [36] S. Ju, W.S. Yeo, Nanotechnology 23 (2012) 135701. [37] E.R. Balmayor, I. Pashkuleva, A.M. Frias, H.S. Azevedo, R.L. Reis, J. R. Soc. Interface 8 (2011) 896–908. [38] Y. Jung, J.M. Lee, J.W. Kim, J. Yoon, H. Cho, B.H. Chung, Anal. Chem. 81 (2009) 936–942. [39] L. Wang, H.Y. Park, S.I. Lim, M.J. Schadt, D. Mott, J. Luo, X. Wang, C.J. Zhong, J. Mater. Chem. 18 (2009) 2629–2635. [40] S.J. Cho, B.R. Jarrett, A.Y. Louie, S.M. Kauzlarich, Nanotechnology 17 (2006) 640–644. [41] J.M. Nam, C.S. Thaxton, C.A. Mirkin, Science 301 (2003) 1884–1886. [42] H. Gu, K. Xu, C. Xu, B. Xu, Chem. Commun. 9 (2006) 941–949. [43] M. Takahashi, T. Yoshino, H. Takeyama, T. Matsunaga, Biotechnol. Prog. 25 (2009) 219–226. [44] D.W. Inglis, R. Riehn, R.H. Austin, J.C. Sturm, Appl. Phys. Lett. 85 (2004) 5093–5095. [45] H.Y. Park, M.J. Schadt, L. Wang, I.S. Lim, P.N. Njoki, S.H. Kim, M.Y. Jang, J. Luo, C.J. Zhong, Langmuir 23 (2007) 9050–9056. [46] P. Zhu, D.R. Shelton, S. Li, D.L. Adams, J.S. Karns, P. Amstutz, C.M. Tang, Biosens. Bioelectron. 30 (2011) 337–341. [47] L. Ma, C. Wang, M. Zhang, Sens. Actuators B 160 (2011) 650–655. [48] J.B. Haun, T.J. Yoon, H. Lee, R. Weissleder, WIREs Nanomed. Nanobiotechnol. 2 (2010) 291–304. [49] R.S. Gaster, L. Xu, S.J. Han, R.J. Wilson, D.A. Hall, S.J. Osterfeld, H. Yu, S.X. Wang, Nat. Nanotechnol. 6 (2011) 314–320. [50] J.P. Wang, J.M. Qiu, T.A. Taton, B.S. Kim, IEEE Trans. Magn. 42 (2006) 3042–3047. [51] A. Demortiere, P. Panissod, B.P. Pichon, G. Pourroy, D. Guillon, B. Donnio, S. Begin-Colin, Nanoscale 3 (2011) 225–232. [52] D.H. Kim, Y.T. Thai, D.E. Nikles, C.S. Brazel, IEEE Trans. Magn. 45 (2009) 64–70. [53] M. Isabelle, D. Catherine, J.B. Michel, C.W. Karen, BioTechniques 37 (2004) 790–802. [54] Y. Yamamoto, Y. Matsumoto, H. Koinuma, Appl. Surf. Sci. 238 (2004) 189–192. [55] Y. Mizuno, F.K. King, Y. Yamauchi, T. Homma, A. Tanaka, Y. Takakuwa, T. Momose, J. Vac. Sci. Technol. A 20 (2002) 1716–1721. [56] L. Zhao, Q. Jiang, J. Lian, Appl. Surf. Sci. 254 (2008) 4620–4625. [57] T. Ramanathan, F.T. Fisher, R.S. Ruoff, L.S. Brinson, Chem. Mater. 17 (2005) 1290–1295. [58] M. Dhayal, D.M. Ratner, Langmuir 25 (2009) 2181–2187. [59] M. Driffield, D.M. Goodall, A.S. Klute, D.K. Smith, K. Wilson, Langmuir 18 (2002) 8660–8665. [60] A. Kuzuya, K. Numajiri, M. Kimura, M. Komiyama, Nucleic Acids Symp. Ser. 52 (2008) 681–682. [61] T. Ooya, T. Kawashima, N. Yui, Biotechnol. Bioprocess Eng. 6 (2001) 293–300. [62] R.J. Leatherbarrow, M. Stedman, T.N.C. Wells, J. Mol. Biol. 221 (1991) 361–365. [63] K. Wadu-Mesthrige, N.A. Amro, G.Y. Liu, Scanning 22 (2000) 380–388.
Please cite this article in press as: W. Wang, et al., Surface modification and bioconjugation of FeCo magnetic nanoparticles with proteins, Colloids Surf. B: Biointerfaces (2013), http://dx.doi.org/10.1016/j.colsurfb.2013.11.050