Article pubs.acs.org/Langmuir
Rapid Surface−Biostructure Interaction Analysis Using Strong MetalBased Nanomagnets Aline C. C. Rotzetter,†,‡ Christoph M. Schumacher,†,‡ Tamotsu Zako,*,‡ Wendelin J. Stark,*,† and Mizuo Maeda‡ †
ETH Zurich, Institute for Chemical and Bioengineering, Wolfgang-Pauli-Strasse 10, CH-8093 Zurich, Switzerland Bioengineering Laboratory, RIKEN Institute, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan
‡
S Supporting Information *
ABSTRACT: Nanomaterials are increasingly suggested for the selective adsorption and extraction of complex compounds in biomedicine. Binding of the latter requires specific surface modifications of the nanostructures. However, even complicated macromolecules such as proteins can afford affinities toward basic surface characteristics such as hydrophobicity, topology, and electrostatic charge. In this study, we address these more basic physical interactions. In a model system, the interaction of bovine serum albumin and amyloid β 42 fibrillar aggregates with carbon-coated cobalt nanoparticles, functionalized with various polymers differing in character, was studied. The possibility of rapid magnetic separation upon binding to the surface represents a valuable tool for studying surface interactions and selectivities. We find that the surface interaction of Aβ 42 fibrillar aggregates is mostly hydrophobic in nature. Because bovine serum albumin (BSA) is conformationally adaptive, it is known to bind surfaces with widely differing properties (charge, topology, and hydrophobicity). However, the rate of tight binding (no desorption upon washing) can vary largely depending on the extent of necessary conformational changes for a specific surface. We found that BSA can only bind slowly to polyethylenimine-coated nanomagnets. Under competitive conditions (high excess BSA compared to that for β 42 fibrillar aggregates), this effect is beneficial for targeting the fibrillar species. These findings highlight the possibility of selective extractions from complex media when advantageous basic physical surface properties are chosen.
■
the β-sheet structure of a large variety of amyloid fibrils and are therefore widely used in their analysis.24−27 The binding of amyloid fibrils to ultrastrong metal-based nanomagnets offers a convenient base for examining the interaction of a family of complex biological structures with different basic physical surface characteristics (i.e., electrostatic properties and hydrophobic interactions). Here, we modify the surface properties of carbon-coated cobalt nanomagnets with widely used polymers. The surface coating of nanomagnets with polymers can be achieved using different approaches, including physisorption,28 reactive linking,29 and direct polymerization on the particle surface.30 Different surface affinities of amyloid β 42 (Aβ) fibrils were systematically evaluated in solutions containing high concentrations of bovine serum albumin (BSA, 1 wt %, 100-fold excess compared to amyloid fibrils) to simulate a competitive environment similar to blood for surface attachment.
INTRODUCTION
Magnetic nanomaterials offer high specific surface areas, ease of separation, and the opportunity for application-tailored surface modifications. They have been proposed for use in catalysis1−3 and drug delivery4,5 and as contrast agents.6,7 Beside high chemical stability and advantageous properties of ferromagnetic cores, carbon-coated cobalt nanomagnets offer a simple platform for chemical functionalization and rapid physical movement.8 Previously, the selective removal of large biomolecules (e.g., inflammatory mediators) using nonoxidic nanomagnets in protein-rich media such as human whole blood by immobilized FAB fragments has been shown.9 This highlights the importance of specific surface-structure interactions under competitive adsorption conditions. However, the complex binding characteristics of a specific antibody usually restrict its use to broader target ranges. Beside structural selectivity, affinities of large biomolecules can also arise from basic physical surface properties including hydrophobicity,10 topology,11 and electric charge.12,13 This concept is successfully applied in phosphopeptide enrichment.14,15 Amyloid fibrils are complex strain-shaped protein structures that are formed from smaller peptide monomer units.16 Their physiological presence and tissue deposition are associated with neurodegenerative diseases such as morbus Alzheimer’s17 and different types of amyloidosis.18−23 Thioflavin T (ThT) or Congo red can access © 2013 American Chemical Society
■
EXPERIMENTAL SECTION
Carbon-Coated Cobalt Metal Nanoparticles. Carbon-coated nanoparticles were prepared according to Grass et al.8 or purchased from Turbobeads LLC (Turbobeads LLC, Switzerland, see Supporting Received: July 15, 2013 Revised: September 25, 2013 Published: October 23, 2013 14117
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
pH-Dependent Amyloid Fibril Adsorption Experiments. Solutions with different pH values were prepared by varying the amounts of dipotassium hydrogen phosphate solution and potassium dihydrogen phosphate solution, resulting in final pH values of 4.4, 6.0, 7.4, 8.0, and 9.2 (constant final salt concentration of 0.1 M).33 Sample tubes with 80 μL of amyloid-fibril-containing stock solution were centrifuged at 6200 rpm (Waken, model 8864) for 15 min, and the supernatants were carefully replaced by the previously prepared solutions differing in pH. Then, particle dispersions in deionized water were prepared at a concentration of 2.5 mg/mL for poly(acrylic acid) and 1 mg/L for polyethylenimine-coated nanomagnets, respectively. After brief ultrasonification (Hielscher Ultrasound Technology, UP50H Ultrasonic Processor), 20 μL of each dispersion was added to the previously prepared amyloid-fibril-containing solutions with diverging pH values. The extent of fibrillar extraction was analyzed as above.
Information for characterization). As-received nanoparticles were stirred in hydrochloric acid (10%) for 4 h to remove free metal ions or poorly coated magnetic particles.31 Thereafter, the particles were separated with a strong Nd2Fe14B magnet (51 × 51 × 25 mm3, B ≈ 0.5 T) and carefully washed with deionized water (three times) and ethanol (three times). This washing procedure was applied after all functionalization steps to remove remaining reagents. Finally, the particles were dried for 24 h at 60 °C before further functionalization. Surface Functionalization with Polymers. Polyethylenimine Functionalization. Following pre-established protocols,28 200 mg particles were stirred in 10 mL of dimethylacetamide (DMAc, Wako Chemicals), and then 200 mg of an aqueous polyethylenimine (PEI) solution (branched, Sigma-Aldrich, 50 wt %) was added. The dispersion was stirred for 2 h at room temperature. Hereafter, the functionalized particles were washed with DMAc (three times) before the general washing procedure described above. In these washing steps, only loosely physisorbed polymer is desorbed again, whereas the detachment of the inner polymer layer is unlikely to happen because of the loss of entropy.32 This irreversible polymer attachment was confirmed as described below. Poly(acrylic acid) and Poly(diallyldimethylammonium chloride) Functionalization. Two hundred milligram particles were dispersed in 10 mL of deionized water and 20 mg of poly(acrylic acid) (Wako Chemicals, Mw = 5000), and 100 mg of an aqueous poly(diallyldimethylammonium chloride) (polyDADMAC) solution (Sigma-Aldrich, Mw = 400 000−500 000, 20 wt %) was added. After 2 h of stirring, the functionalized particles were separated and then washed as described above. Additional Polymer Functionalization. See the Supporting Information. Irreversible Attachement of Polymer Coatings to the Nanoparticle Surfaces. To ensure irreversible surface attachment, polyDADMAC-functionalized nanoparticles were washed thoroughly with deionized water several times. After the 1st, 3rd, 5th and 10th washing steps, 10 mg of nanoparticles was removed and the carbon content was determined quantitatively by element microanalysis (Yanaco, MT-6, CHN Corder). Bovine Serum Albumin (BSA) Adsorption Experiments. Ten milligrams of bare or functionalized nanoparticles was added to 10 mL of a BSA solution (Jackson Immuno Research Laboratories Inc., IgGfree and protease-free, 1 wt %) and stirred overnight. Thereafter, the particles were washed by using the general washing procedure (see above). To determine the extent of BSA adsorption, quantitative element microanalysis was used. To reveal differences in the adsorption rate on differently coated particles, the same experiment was conducted for 5 min only. Amyloid β 42 Fibrillar Aggregate Preparation. Ten microliter aliquots containing 1 mM Aβ protein (human, 1−42, Peptide Institute Inc., Japan) in dimethyl sulfoxide (Wako Chemicals, dehydrated) were prepared and stored at −20 °C. The Aβ fibrillar aggregates were freshly prepared each time before the adsorption experiments. Therefore, 10 μL of defrosted Aβ-protein solution was added to 390 μL of phosphate-buffered saline solution (PBS, Takara Bio Inc., Japan, Dulbecco’s Formula (modified)) and gently mixed. The resulting solution was incubated for at least 12 h at 37 °C and directly used for the adsorption experiments described below (denoted as amyloidfibril-containing stock solution). Amyloid β 42 Fibrillar Aggregate Adsorption Experiments. Nanoparticle dispersions with different particle concentrations (0.1, 0.25, 1.0, 2.5, and 5.0 mg/mL) in PBS were prepared. For competing adsorption experiments, BSA (Jackson Immuno Research Laboratories Inc., IgG-free and protease-free, 5 wt %) was directly added to the initial nanoparticle dispersions. All dispersions were briefly ultrasonicated (Hielscher Ultrasound Technology, UP50H Ultrasonic Processor). Immediately afterwards, 20 μL was transferred into 80 μL of an amyloid-fibril-containing stock solution in sample tubes. The resulting dispersions were carefully mixed by pipetting 10 times up and down. After 10 min, the nanoparticles were collected with a magnet on the tube wall, and the supernatant was transferred to a new sample tube for analysis.
■
MATERIAL CHARACTERIZATION Nanoparticle Characterization. The surface functionalization of the nanomagnets was verified by diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) using KBr powder containing 5 wt % particles (Tensor 27 Spectrometer, Bruker, DiffuseIR, Pike Technologies). Bare Co/C nanoparticles were taken as a background to improve the spectrum quality. Furthermore, transmission electron micrographs (TEM) of the bare particles were recorded (FEI, Tecnai F30 ST, operated at 300 kV), and element microanalysis (Yanaco, MT-6, CHN Corder) was used for the quantification of the surface functionalization. Amyloid β 42 Fibrillar Aggregate Characterization. To confirm the formation of amyloid β fibrillar aggregates and the interaction with nanomagnets, scanning transmission electron microscopy (STEM) micrographs were recorded (FEI, Nova NanoSEM 450, operated at 30 kV). Amyloid β 42 Fibrillar Aggregate Adsorption Experiments. The adsorption of Aβ fibrillar aggregates on functionalized nanomagnets was determined by the number of aggregates remaining in the supernatant after collection by nanomagnets. Concentrations of Aβ fibrillar aggregates were determined by staining the fibrils with amyloid-specific dye ThT (Sigma-Aldrich) and subsequent fluorescence measurements. Therefore, 75 μL of the ThT solution (25 μM in PBS) was added to 25 μL of a sample probe and gently mixed by pipetting up and down in a 96-well plate (Nuncolon). Each sample was recorded three times. After 2 h, the fluorescence signal was recorded (TECAN, excitation wavelength 450 nm, detection wavelength 485 nm, bandwidth 10 nm) in order to quantify the remaining number of Aβ fibrillar aggregates. BSA appeared to influence the fluorescence signal slightly, making separate calibration necessary. (For more detailed information, see the Supporting Information.)
■
RESULTS AND DISCUSSION Characterization of the Nanomaterials and Amyloid β 42 Fibrillar Aggregates. According to electron micrographs (Figure 1a) the carbon-coated cobalt particles used here reveal a particle size distribution of between 5 and 50 nm in diameter. The particles appear to be almost spherical in shape and are typically coated with two to three well-structured graphite layers.8 To determine the number of functional groups (i.e., the amount of physisorbed polymer on the particles), element microanalysis was used. Corresponding polymer shell thicknesses were calculated according to the previously described 14118
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
Figure 1. (a) Transmission electron micrograph of carbon-coated cobalt nanomagnets. (b) Amyloid-catching procedure using nanomagnets. (c) Scanning transmission electron micrographs of amyloid β 42 fibrillar aggregates. (d) Scanning transmission electron micrographs of amyloid β 42 fibrillar aggregates after binding to PEI-coated nanoparticles.
Figure 2. Diffuse reflectance Fourier transform infrared spectra of polymer-coated Co/C nanoparticles.
contaminants from various media. In noncomplex liquids, the capturing, particularly of larger species (proteins, dyes, and toxins), can in some cases occur even without any further specific surface modification. This effect is due to entropic reasons. When a large entity is physisorbing onto a surface, small molecules in close proximity are displaced. The large entity as a unit holds three directional degrees of motional freedom in solution but only two on the surface. In contrast, many small molecules (i.e., solvent) change from two to three motional degrees of freedom upon release, resulting in an overall gain in entropy.34 The synthesis of the materials used here with high-molecular-weight polymers is based on this effect. Even water-soluble polymers can be irreversibly physisorbed as long as other effects such as electrostatic repulsion are not predominant. This behavior was verified by quantitative element microanalysis of polyDADMAC-functionalized nanomagnets after different numbers of washing steps.
method.31 (For details, see the Supporting Information.) All results are summarized in Table 1. To verify further the presence of the polymers and functional groups on the particles, diffuse reflectance Fourier transform infrared spectroscopy (DRIFTS) was performed and compared to the spectra of the pure substances (Figure 2). Characteristic absorption bands of the polymers confirm their physisorbed presence. Shifts and changes in the characteristic absorption bands can result from surface-related effects.31 After at least 12 h of incubation at 37 °C in PBS, Aβ fibrillar aggregates show lengths in the range of a few micrometers (Figure 1c). In a magnetic separation process, the fibrillar aggregates bind to the nanomagnet surface and can thereafter be magnetically separated (Figure 1b,d). Irreversible Polymer Attachment. Because of their high specific surface areas, nanoparticles are suitable for capturing
Table 1. Carbon Contents Obtained from Element Microanalysis, Enabling the Calculation of the Polymer Coatings in Weight Percent and Resulting Layer Thicknesses by Knowing the Polymer Carbon Content
14119
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
Because the carbon content of the functionalized material remains constant and differs significantly from the starting material (Table 2), a stable attachment of the highly watersoluble polymer can be assumed.
Table 3. Bound Amounts of BSA on Differently Functionalized Nanoparticles after Distinct Times nanoparticle Co/C Co/C−PEI Co/C−poly(acrylic acid) Co/C− polyDADMAC
Table 2. Confirmation of Stable PolyDADMAC Functionalization by Carbon Content Analysis after Different Numbers of Washing Stepsa
a
number of washing steps
carbon content (wt %)
1 3 5 10
8.36 8.33 8.28 8.31
BSA bound after 5 min (wt %)
BSA bound overnight (wt %)
3.51 0.36 1.47
9.51 11.13 6.98
2.89
9.03
adsorption can be strongly time- and surface-type-dependent. For competing interactions between different species, structural changes for tight adsorption can have a significant influence. This is particularly important for magnetic extractions that are conducted within short time frames of typically minutes. Surface Adsorption of Amyloid β 42 Fibrillar Aggregates. In contrast to BSA, amyloid β 42 fibrils are much larger and more rigid. Thus, conformational adaptions upon surface contact are more unlikely to occur because strong enthalpy barriers would have to be overcome.39 Figure 3a,b
Starting material (unfunctionalized nanomagnets): 7.70 wt % carbon.
Surface Coatings for Selective Extraction. When target capturing structures are attached to surfaces for applications in medicine, diagnostics, and biotechnology, the validation of the binding selectivity is preferably quantified in media containing other large biomolecules such as proteins. Still, it must be stated that surface groups that are selective for a target in their free state are often constricted or even blocked for target interaction upon surface attachment. Congo red is a well-known staining agent with an affinity for amyloid fibrils. Its interaction is based on the penetration of its flat aromatic units into the characteristic fibrillar β-sheet structure (gap width ∼5 Å).24,25 Earlier publications claimed an arising selectivity for amyloid fibrils when Congo red or similar dye molecules were directly physisorbed onto iron oxide nanoparticles.35 However, because flat-attached Congo red molecules are geometrically hindered to access the narrow β-sheet structure of amyloid fibrils, the reason for the observed fibril selectivity must be addressed as either surface entropy effects or electrostatic interactions. Here, we systematically investigate the surface adsorption of Aβ fibrillar aggregates on carbon-coated cobalt nanomagnets that were functionalized with various polymers, revealing different physical characteristics (charge and hydrophobicity). To afford a better understanding of biologically more relevant competitive adsorption, the surface affinities were also examined in solutions containing excess bovine serum albumin (100-fold excess in weight regarding fibrillar aggregates). This simulates an environment similar to that of blood plasma.36 Surface Adsorption of BSA. Because of its flexibility, BSA is known to undergo thermodynamically favorable conformational changes upon surface contact, regardless of the hydrophilic or hydrophobic nature of the surface.37,38 However, this event can take several hours to complete. To understand to what extent this happens on the materials used here, differently coated nanoparticles were stirred in a 1 wt % BSA solution for 5 min and overnight. After extensive washing, element microanalysis was conducted to quantify the bound amount of BSA from the resulting change in the carbon content (Table 3). All particles show considerable adsorption after prolonged exposition (overnight) to dissolved BSA. However, after short exposure to the same conditions, strong differences in adsorption occur. Only proteins that developed strong surface interactions in a short time were bound. Because PEI-coated nanomagnets developed the highest adsorption after a long exposure but revealed the lowest adsorption amounts after 5 min, it seems likely that time-consuming conformational changes are needed to bind tightly to this surface. Furthermore, previous reports38 have shown that the extent of BSA
Figure 3. Adsorption isotherms of amyloid β 42 fibrillar aggregates without (squares) and with BSA (circles) on (a) blank carbon-coated nanomagnets, (b) polyethylenimine-functionalized nanomagnets, (c) poly(acrylic acid)-functionalized nanomagnets, and (d) polyDADMAC-functionalized nanomagnets (broken lines, 60% extraction).
depicts adsorption isotherms of Aβ fibrillar aggregates on blank as well as PEI-coated nanomagnets at different particle concentrations. Both particle types exhibit similar adsorption behavior in solutions with PBS only. At the highest tested particle concentrations, we found almost complete fibril extraction (>95%). Comparable results were obtained for chemically similar chitosan-coated particles (Supporting Information). All adsorption isotherms reveal a curved shape, which is likely to be a dilution effect of the fibrillar concentration in the extracted medium. Adsorption isotherms on the charged polymer surfaces (negative charge under neutral conditions, poly(acrylic acid); positive charge, polydiallyl14120
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
Supporting Information for polyvinylsulfate- and poly(ethylene glycol)-coated particles.) As found earlier, BSA is not able to bind rapidly to PEI-functionalized particles. This enables a favorable adsorption environment for fibrils within the conducted extraction time frame, despite a much higher BSA concentration compared to that of fibrils (100-fold).
(dimethylammonium chloride) (i.e., polyDADMAC)) depicted in Figure 3c,d show a significantly lower binding affinity. Following these results, it is likely that hydrophobic interactions are present. To verify this assumption further, the pH value of the extraction solution was varied in order to change surface charges (here, on poly(acrylic acid)-coated particles). Poly(acrylic acids) have pKa values of between 4.7 and 7, depending on the molecular weight, branching, and salt concentrations.40 This leads to various ratios of protonated carboxylic acid groups on the surface and should therefore influence the extraction performance if the above-stated assumptions are valid. A strong dependence of the fibril extraction extent reaching 85% at pH 4.4 and no extraction at pH 9.2 (following an almost linear trend) becomes apparent at constant particle and salt concentrations (Figure 4). The more carboxylic groups that
■
CONCLUSIONS In this study, we investigated the interaction of Aβ fibrillar aggregates as a model biostructure with physical surface properties of polymer-coated cobalt-based nanomagnets. Because of the possibility of magnetic separation within seconds, our setup represents a valuable tool for biostructure−surface interaction analysis, particularly for in vivo situations. Thus, it represents an alternative to current in vitro stopped-flow techniques, especially for larger biologically relevant sample volumes. In protein-rich model solutions under neutral conditions (1 wt % BSA, 100-fold excess in mass compared to amyloid fibrils), a competitive adsorption environment (similar to blood) between the fibrils and BSA was established. BSA is known for its high structural adaptability when binding to surfaces. We found that all investigated particle types revealed considerable BSA binding after prolonged exposure to BSA-rich solutions. However, binding extents after minutes largely differed. This effect can be used to target a lower-concentration biological species such as Aβ fibrillar aggregates. We observed that the latter have weaker affinities for charged polymer surfaces. This reveals the importance of hydrophobic interactions. By varying the pH level of solutions containing poly(acrylic acid)-coated nanomagnets and thus the surface charge on the particles, we furthermore confirmed this effect. Following these insights, we conducted extractions with good selectivity toward Aβ fibrils using PEI-functionalized nanoparticles. Because our findings are valid not only for the system described here but also more generally for a vast range of other biomolecules and structures, we want to highlight the practicability of this simple and rapid surface interaction analysis tool based on highly magnetic nanoparticles.
Figure 4. pH dependence of the surface adsorption of amyloid β 42 fibrillar aggregates on poly(acrylic acid)- and polyethylenimine (PEI)functionalized carbon-coated cobalt nanomagnets. Because the poly(acrylic acid)-coated surface is increasingly deprotonated at increasing pH, hydrophobic interactions decrease. At pH 9.2, surface adsorption is no longer possible. In the case of polyethylenimine, more constant behavior occurs because it provides constant surface properties under all examined conditions.
are deprotonated, the fewer hydrophobic interactions that may occur. Additionally, Hetenyi et al.41 found that Aβ fibrillar aggregates present a slightly negative charge at a pH of 7, meaning that protonation at higher acidity leads to a lesscharged fibril surface, further enhancing the entropy-driven interaction. Equivalent adsorption experiments with polyethylenimine-coated particles as shown in Figure 4 resulted in rather constant adsorption behavior when the pH is altered (presence of primary, secondary, and tertiary amines with different pKa’s42). This situation allows more constant binding over the observed pH range. Wang et al.43 found similar effects on differently charged self-assembled monolayer surfaces in an atomic force microscopy study. Competitive Surface Adsorption of Amyloid β 42 Fibrillar Aggregates (BSA Present). The presence of large concentrations of BSA compared to the target species establishes a strongly competitive environment for selective extractions. Such situations are frequent in natural systems such as blood. Because it often remains unclear what kind of interaction mechanisms takes place, magnetic nanoparticles can act as a convenient rapid analysis tool. As expected, all particles adsorbed amyloid fibrils to a lesser extent when BSA was present (Figure 3a−d). Despite this, the PEI-coated nanomagnets still revealed a considerable ability to bind amyloid fibrillar aggregates (75% extraction extent at highest particle concentration). For all of the other tested particle types, the drop in extraction ability was significantly higher. (See also the
■
ASSOCIATED CONTENT
S Supporting Information *
TEM micrographs, adsorption isotherms of other polymer surface coatings, IR spectra, and elemental microanalysis data. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Authors
*Phone: +81 48 467 93 12. Fax: +81 48 462 46 58. E-mail: [email protected]. *Phone: +41 44 632 09 80. Fax: +41 44 633 10 83. E-mail: [email protected]. Author Contributions
A.C.C.R. and C.M.S. contributed equally to this work. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We are grateful to the RIKEN Materials Characterization Support Unit for element analysis and TEM analysis. A.C.C.R. and C.M.S. are RIKEN short-term international program 14121
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
X.; Greengard, P.; Relkin, N. R. Intraneuronal A beta 42 accumulation in human brain. Am. J. Pathol. 2000, 156, 15−20. (18) Platt, G. W.; McParland, V. J.; Kalverda, A. P.; Homans, S. W.; Radford, S. E. Dynamics in the unfolded state of beta(2)-microglobulin studied by NMR. J. Mol. Biol. 2005, 346, 279−294. (19) Gertz, M. A.; Kyle, R. A. Myopathy in primary systemic amyloidosis. J. Neurol., Neurosurg., Psychiatry 1996, 60, 655−660. (20) Gharibyan, A. L.; Zamotin, V.; Yanamandra, K.; Moskaleva, O. S.; Margulis, B. A.; Kostanyan, I. A.; Morozova-Roche, L. A. Lysozyme amyloid oligomers and fibrils induce cellular death via different apoptotic/necrotic pathways. J. Mol. Biol. 2007, 365, 1337−1349. (21) Novitskaya, V.; Bocharova, O. V.; Bronstein, I.; Baskakov, I. V. Amyloid fibrils of mammalian prion protein are highly toxic to cultured cells and primary neurons. J. Biol. Chem. 2006, 281, 13828− 13836. (22) Engel, M. F. M.; Khemtemourian, L.; Kleijer, C. C.; Meeldijk, H. J. D.; Jacobs, J.; Verkleij, A. J.; de Kruijff, B.; Killian, J. A.; Hoppener, J. W. M. Membrane damage by human islet amyloid polypeptide through fibril growth at the membrane. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 6033−6038. (23) Zako, T.; Sakono, M.; Kobayashi, T.; Sörgjerd, K.; Nilsson, K. P. R.; Hammarström, P.; Lindgren, M.; Maeda, M. Cell interaction study of amyloid by using luminescent conjugated polythiophene: implication that amyloid cytotoxicity is correlated with prolonged cellular binding. ChemBioChem 2012, 13, 358−363. (24) Biancalana, M.; Koide, S. Molecular mechanism of Thioflavin-T binding to amyloid fibrils. Biochim. Biophys. Acta, Proteins Proteomics 2010, 1804, 1405−1412. (25) Schutz, A. K.; Soragni, A.; Hornemann, S.; Aguzzi, A.; Ernst, M.; Bockmann, A.; Meier, B. H. The amyloid-Congo red interface at atomic resolution. Angew. Chem., Int. Ed. 2011, 50, 5956−5960. (26) Puchtler, H.; Sweat, F.; Levine, M. On binding of Congo red by amyloid. J. Histochem. Cytochem. 1962, 10, 355−364. (27) Hobbs, J. R.; Morgan, A. D. Fluorescence microscopy with thioflavine-T in diagnosis of amyloid. J. Pathol. Bacteriol. 1963, 86, 437−442. (28) Fuhrer, R.; Herrmann, I. K.; Athanassiou, E. K.; Grass, R. N.; Stark, W. J. Immobilized beta-cyclodextrin on surface-modified carboncoated cobalt nanomagnets: reversible organic contaminant adsorption and enrichment from water. Langmuir 2011, 27, 1924−1929. (29) Schumacher, C. M.; Herrmann, I. K.; Bubenhofer, S. B.; Gschwind, S.; Hirt, A.-M.; Beck-Schimmer, B.; Günther, D.; Stark, W. J. Quantitative recovery of magnetic nanoparticles from flowing blood: trace analysis and the role of magnetization. Adv. Funct. Mater. 2013, 39, 4888−4896. (30) Zeltner, M.; Grass, R. N.; Schaetz, A.; Bubenhofer, S. B.; Luechinger, N. A.; Stark, W. J. Stable dispersions of ferromagnetic carbon-coated metal nanoparticles: preparation via surface initiated atom transfer radical polymerization. J. Mater. Chem. 2012, 22, 12064− 12071. (31) Schumacher, C. M.; Grass, R. N.; Rossier, M.; Athanassiou, E. K.; Stark, W. J. Physical defect formation in few layer graphene-like carbon on metals: influence of temperature, acidity, and chemical functionalization. Langmuir 2012, 28, 4565−4572. (32) Koehler, F. M.; Rossier, M.; Waelle, M.; Athanassiou, E. K.; Limbach, L. K.; Grass, R. N.; Gunther, D.; Stark, W. J. Magnetic EDTA: coupling heavy metal chelators to metal nanomagnets for rapid removal of cadmium, lead and copper from contaminated water. Chem. Commun. 2009, 4862−4864. (33) Gomori, G. Preparation of buffers for use in enzyme studies. Methods Enzymol. 1955, 1, 138−146. (34) Melzak, K. A.; Sherwood, C. S.; Turner, R. F. B.; Haynes, C. A. Driving forces for DNA adsorption to silica in perchlorate solutions. J. Colloid Interface Sci. 1996, 181, 635−644. (35) Skaat, H.; Margel, S. Synthesis of fluorescent-maghemite nanoparticles as multimodal imaging agents for amyloid-beta fibrils detection and removal by a magnetic field. Biochem. Biophys. Res. Commun. 2009, 386, 645−649.
associates. Financial support by ETH Zurich, RIKEN Institute, and the Swiss National Science Foundation (no. 406440131268) is greatly appreciated.
■
REFERENCES
(1) Zhang, S. X.; Zhao, X. L.; Niu, H. Y.; Shi, Y. L.; Cai, Y. Q.; Jiang, G. B. Superparamagnetic Fe3O4 nanoparticles as catalysts for the catalytic oxidation of phenolic and aniline compounds. J. Hazard. Mater. 2009, 167, 560−566. (2) Schatz, A.; Hager, M.; Reiser, O. Cu(II)-Azabis(oxazoline)complexes immobilized on superparamagnetic magnetite@silica-nanoparticles: a highly selective and recyclable catalyst for the kinetic resolution of 1,2-diols. Adv. Funct. Mater. 2009, 19, 2109−2115. (3) Megia-Fernandez, A.; Ortega-Munoz, M.; Lopez-Jaramillo, J.; Hernandez-Mateo, F.; Santoyo-Gonzalez, F. Non-magnetic and magnetic supported copper(I) chelating adsorbents as efficient heterogeneous catalysts and copper scavengers for click chemistry. Adv. Synth. Catal. 2010, 352, 3306−3320. (4) LaVan, D. A.; McGuire, T.; Langer, R. Small-scale systems for in vivo drug delivery. Nat. Biotechnol. 2003, 21, 1184−1191. (5) Weissleder, R.; Kelly, K.; Sun, E. Y.; Shtatland, T.; Josephson, L. Cell-specific targeting of nanoparticles by multivalent attachment of small molecules. Nat. Biotechnol. 2005, 23, 1418−1423. (6) Hong, R. Y.; Feng, B.; Chen, L. L.; Liu, G. H.; Li, H. Z.; Zheng, Y.; Wei, D. G. Synthesis, characterization and MRI application of dextran-coated Fe3O4 magnetic nanoparticles. Biochem. Eng. J. 2008, 42, 290−300. (7) Zhao, M.; Beauregard, D. A.; Loizou, L.; Davletov, B.; Brindle, K. M. Non-invasive detection of apoptosis using magnetic resonance imaging and a targeted contrast agent. Nat. Med. 2001, 7, 1241−1244. (8) Grass, R. N.; Athanassiou, E. K.; Stark, W. J. Covalently functionalized cobalt nanoparticles as a platform for magnetic separations in organic synthesis. Angew. Chem., Int. Ed. 2007, 46, 4909−4912. (9) Herrmann, I. K.; Bernabei, R. E.; Urner, M.; Grass, R. N.; BeckSchimmer, B.; Stark, W. J. Device for continuous extracorporeal blood purification using target-specific metal nanomagnets. Nephrol., Dial., Transplant. 2011, 26, 2948−U1516. (10) Wei, J. H.; Igarashi, T.; Okumori, N.; Maetani, T.; Liu, B. L.; Yoshinari, M. Influence of surface wettability on competitive protein adsorption and initial attachment of osteoblasts. Biomed. Mater. 2009, 4, 045002. (11) Tsougeni, K.; Petrou, P. S.; Papageorgiou, D. P.; Kakabakos, S. E.; Tserepi, A.; Gogolides, E. Controlled protein adsorption on microfluidic channels with engineered roughness and wettability. Sens. Actuators, B 2012, 161, 216−222. (12) Pasche, S.; Voros, J.; Griesser, H. J.; Spencer, N. D.; Textor, M. Effects of ionic strength and surface charge on protein adsorption at PEGylated surfaces. J. Phys. Chem. B 2005, 109, 17545−17552. (13) Xie, H. G.; Li, X. X.; Lv, G. J.; Xie, W. Y.; Zhu, J.; Luxbacher, T.; Ma, R.; Ma, X. J. Effect of surface wettability and charge on protein adsorption onto implantable alginate-chitosan-alginate microcapsule surfaces. J. Biomed. Mater. Res. 2010, 92A, 1357−1365. (14) Tan, F.; Zhang, Y.; Mi, W.; Wang, J.; Wei, J.; Cai, Y.; Qian, X. Enrichment of phosphopeptides by Fe3+-immobilized magnetic nanoparticles for phosphoproteome analysis of the plasma membrane of mouse liver. J. Proteome Res. 2008, 7, 1078−1087. (15) Zhou, H. J.; Tian, R. J.; Ye, M. L.; Xu, S. Y.; Feng, S.; Pan, C. S.; Jiang, X. G.; Li, X.; Zou, H. F. Highly specific enrichment of phosphopeptides by zirconium dioxide nanoparticles for phosphoproteome analysis. Electrophoresis 2007, 28, 2201−2215. (16) Zako, T.; Sakono, M.; Hashimoto, N.; Hara, M.; Maeda, M. Bovine insulin filaments induced by reducing disulfide bonds show a different morphology, secondary structure, and cell toxicity from intact insulin amyloid fibrils. Biophys. J. 2009, 96, 3331−3340. (17) Gouras, G. K.; Tsai, J.; Naslund, J.; Vincent, B.; Edgar, M.; Checler, F.; Greenfield, J. P.; Haroutunian, V.; Buxbaum, J. D.; Xu, H. 14122
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
(36) Wang, H.; Shi, Z. M. In vitro biodegradation behavior of magnesium and magnesium alloy. J. Biomed. Mater. Res. 2011, 98B, 203−209. (37) Norde, W.; Giacomelli, C. E. BSA structural changes during homomolecular exchange between the adsorbed and the dissolved states. J. Biotechnol. 2000, 79, 259−268. (38) Lee, S. H.; Ruckenstein, E. Adsorption of proteins onto polymeric surfaces of different hydrophilicitiesa case study with bovine serum albumin. J. Colloid Interface Sci. 1988, 125, 365−379. (39) Andrade, J. D.; Hlady, V. Protein Adsorption and Materials Biocompatibility: A Tutorial Review and Suggested Hypotheses. Biopolymers/Non-Exclusion HPLC; Springer: Berlin, 1986; Vol. 79, pp 1−63. (40) Shibayama, M.; Ikkai, F.; Inamoto, S.; Nomura, S.; Han, C. C. pH and salt concentration dependence of the microstructure of poly(N-isopropylacrylamide-co-acrylic acid) gels. J. Chem. Phys. 1996, 105, 4358−4366. (41) Hetenyi, A.; Fulop, L.; Martinek, T. A.; Weber, E.; Soos, K.; Penke, B. Ligand-induced flocculation of neurotoxic fibrillar A beta(1− 42) by noncovalent crosslinking. ChemBioChem 2008, 9, 748−757. (42) Demadis, K. D.; Paspalaki, M.; Theodorou, J. Controlled release of bis(phosphonate) pharmaceuticals from cationic biodegradable polymeric matrices. Ind. Eng. Chem. Res. 2011, 50, 5873−5876. (43) Wang, Q. M.; Shah, N.; Zhao, J.; Wang, C. S.; Zhao, C.; Liu, L. Y.; Li, L. Y.; Zhou, F. M.; Zheng, J. Structural, morphological, and kinetic studies of beta-amyloid peptide aggregation on self-assembled monolayers. Phys. Chem. Chem. Phys. 2011, 13, 15200−15210.
14123
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Rapid Surface−Biostructure Interaction Analysis Using Strong MetalBased Nanomagnets Aline C. C. Rotzetter,†,‡ Christoph M. Schumacher,†,‡ Tamotsu Zako,*,‡ Wendelin J. Stark,*,† and Mizuo Maeda‡ †
ETH Zurich, Institute for Chemical and Bioengineering, Wolfgang-Pauli-Strasse 10, CH-8093 Zurich, Switzerland Bioengineering Laboratory, RIKEN Institute, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan
‡
S Supporting Information *
ABSTRACT: Nanomaterials are increasingly suggested for the selective adsorption and extraction of complex compounds in biomedicine. Binding of the latter requires specific surface modifications of the nanostructures. However, even complicated macromolecules such as proteins can afford affinities toward basic surface characteristics such as hydrophobicity, topology, and electrostatic charge. In this study, we address these more basic physical interactions. In a model system, the interaction of bovine serum albumin and amyloid β 42 fibrillar aggregates with carbon-coated cobalt nanoparticles, functionalized with various polymers differing in character, was studied. The possibility of rapid magnetic separation upon binding to the surface represents a valuable tool for studying surface interactions and selectivities. We find that the surface interaction of Aβ 42 fibrillar aggregates is mostly hydrophobic in nature. Because bovine serum albumin (BSA) is conformationally adaptive, it is known to bind surfaces with widely differing properties (charge, topology, and hydrophobicity). However, the rate of tight binding (no desorption upon washing) can vary largely depending on the extent of necessary conformational changes for a specific surface. We found that BSA can only bind slowly to polyethylenimine-coated nanomagnets. Under competitive conditions (high excess BSA compared to that for β 42 fibrillar aggregates), this effect is beneficial for targeting the fibrillar species. These findings highlight the possibility of selective extractions from complex media when advantageous basic physical surface properties are chosen.
■
the β-sheet structure of a large variety of amyloid fibrils and are therefore widely used in their analysis.24−27 The binding of amyloid fibrils to ultrastrong metal-based nanomagnets offers a convenient base for examining the interaction of a family of complex biological structures with different basic physical surface characteristics (i.e., electrostatic properties and hydrophobic interactions). Here, we modify the surface properties of carbon-coated cobalt nanomagnets with widely used polymers. The surface coating of nanomagnets with polymers can be achieved using different approaches, including physisorption,28 reactive linking,29 and direct polymerization on the particle surface.30 Different surface affinities of amyloid β 42 (Aβ) fibrils were systematically evaluated in solutions containing high concentrations of bovine serum albumin (BSA, 1 wt %, 100-fold excess compared to amyloid fibrils) to simulate a competitive environment similar to blood for surface attachment.
INTRODUCTION
Magnetic nanomaterials offer high specific surface areas, ease of separation, and the opportunity for application-tailored surface modifications. They have been proposed for use in catalysis1−3 and drug delivery4,5 and as contrast agents.6,7 Beside high chemical stability and advantageous properties of ferromagnetic cores, carbon-coated cobalt nanomagnets offer a simple platform for chemical functionalization and rapid physical movement.8 Previously, the selective removal of large biomolecules (e.g., inflammatory mediators) using nonoxidic nanomagnets in protein-rich media such as human whole blood by immobilized FAB fragments has been shown.9 This highlights the importance of specific surface-structure interactions under competitive adsorption conditions. However, the complex binding characteristics of a specific antibody usually restrict its use to broader target ranges. Beside structural selectivity, affinities of large biomolecules can also arise from basic physical surface properties including hydrophobicity,10 topology,11 and electric charge.12,13 This concept is successfully applied in phosphopeptide enrichment.14,15 Amyloid fibrils are complex strain-shaped protein structures that are formed from smaller peptide monomer units.16 Their physiological presence and tissue deposition are associated with neurodegenerative diseases such as morbus Alzheimer’s17 and different types of amyloidosis.18−23 Thioflavin T (ThT) or Congo red can access © 2013 American Chemical Society
■
EXPERIMENTAL SECTION
Carbon-Coated Cobalt Metal Nanoparticles. Carbon-coated nanoparticles were prepared according to Grass et al.8 or purchased from Turbobeads LLC (Turbobeads LLC, Switzerland, see Supporting Received: July 15, 2013 Revised: September 25, 2013 Published: October 23, 2013 14117
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
pH-Dependent Amyloid Fibril Adsorption Experiments. Solutions with different pH values were prepared by varying the amounts of dipotassium hydrogen phosphate solution and potassium dihydrogen phosphate solution, resulting in final pH values of 4.4, 6.0, 7.4, 8.0, and 9.2 (constant final salt concentration of 0.1 M).33 Sample tubes with 80 μL of amyloid-fibril-containing stock solution were centrifuged at 6200 rpm (Waken, model 8864) for 15 min, and the supernatants were carefully replaced by the previously prepared solutions differing in pH. Then, particle dispersions in deionized water were prepared at a concentration of 2.5 mg/mL for poly(acrylic acid) and 1 mg/L for polyethylenimine-coated nanomagnets, respectively. After brief ultrasonification (Hielscher Ultrasound Technology, UP50H Ultrasonic Processor), 20 μL of each dispersion was added to the previously prepared amyloid-fibril-containing solutions with diverging pH values. The extent of fibrillar extraction was analyzed as above.
Information for characterization). As-received nanoparticles were stirred in hydrochloric acid (10%) for 4 h to remove free metal ions or poorly coated magnetic particles.31 Thereafter, the particles were separated with a strong Nd2Fe14B magnet (51 × 51 × 25 mm3, B ≈ 0.5 T) and carefully washed with deionized water (three times) and ethanol (three times). This washing procedure was applied after all functionalization steps to remove remaining reagents. Finally, the particles were dried for 24 h at 60 °C before further functionalization. Surface Functionalization with Polymers. Polyethylenimine Functionalization. Following pre-established protocols,28 200 mg particles were stirred in 10 mL of dimethylacetamide (DMAc, Wako Chemicals), and then 200 mg of an aqueous polyethylenimine (PEI) solution (branched, Sigma-Aldrich, 50 wt %) was added. The dispersion was stirred for 2 h at room temperature. Hereafter, the functionalized particles were washed with DMAc (three times) before the general washing procedure described above. In these washing steps, only loosely physisorbed polymer is desorbed again, whereas the detachment of the inner polymer layer is unlikely to happen because of the loss of entropy.32 This irreversible polymer attachment was confirmed as described below. Poly(acrylic acid) and Poly(diallyldimethylammonium chloride) Functionalization. Two hundred milligram particles were dispersed in 10 mL of deionized water and 20 mg of poly(acrylic acid) (Wako Chemicals, Mw = 5000), and 100 mg of an aqueous poly(diallyldimethylammonium chloride) (polyDADMAC) solution (Sigma-Aldrich, Mw = 400 000−500 000, 20 wt %) was added. After 2 h of stirring, the functionalized particles were separated and then washed as described above. Additional Polymer Functionalization. See the Supporting Information. Irreversible Attachement of Polymer Coatings to the Nanoparticle Surfaces. To ensure irreversible surface attachment, polyDADMAC-functionalized nanoparticles were washed thoroughly with deionized water several times. After the 1st, 3rd, 5th and 10th washing steps, 10 mg of nanoparticles was removed and the carbon content was determined quantitatively by element microanalysis (Yanaco, MT-6, CHN Corder). Bovine Serum Albumin (BSA) Adsorption Experiments. Ten milligrams of bare or functionalized nanoparticles was added to 10 mL of a BSA solution (Jackson Immuno Research Laboratories Inc., IgGfree and protease-free, 1 wt %) and stirred overnight. Thereafter, the particles were washed by using the general washing procedure (see above). To determine the extent of BSA adsorption, quantitative element microanalysis was used. To reveal differences in the adsorption rate on differently coated particles, the same experiment was conducted for 5 min only. Amyloid β 42 Fibrillar Aggregate Preparation. Ten microliter aliquots containing 1 mM Aβ protein (human, 1−42, Peptide Institute Inc., Japan) in dimethyl sulfoxide (Wako Chemicals, dehydrated) were prepared and stored at −20 °C. The Aβ fibrillar aggregates were freshly prepared each time before the adsorption experiments. Therefore, 10 μL of defrosted Aβ-protein solution was added to 390 μL of phosphate-buffered saline solution (PBS, Takara Bio Inc., Japan, Dulbecco’s Formula (modified)) and gently mixed. The resulting solution was incubated for at least 12 h at 37 °C and directly used for the adsorption experiments described below (denoted as amyloidfibril-containing stock solution). Amyloid β 42 Fibrillar Aggregate Adsorption Experiments. Nanoparticle dispersions with different particle concentrations (0.1, 0.25, 1.0, 2.5, and 5.0 mg/mL) in PBS were prepared. For competing adsorption experiments, BSA (Jackson Immuno Research Laboratories Inc., IgG-free and protease-free, 5 wt %) was directly added to the initial nanoparticle dispersions. All dispersions were briefly ultrasonicated (Hielscher Ultrasound Technology, UP50H Ultrasonic Processor). Immediately afterwards, 20 μL was transferred into 80 μL of an amyloid-fibril-containing stock solution in sample tubes. The resulting dispersions were carefully mixed by pipetting 10 times up and down. After 10 min, the nanoparticles were collected with a magnet on the tube wall, and the supernatant was transferred to a new sample tube for analysis.
■
MATERIAL CHARACTERIZATION Nanoparticle Characterization. The surface functionalization of the nanomagnets was verified by diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) using KBr powder containing 5 wt % particles (Tensor 27 Spectrometer, Bruker, DiffuseIR, Pike Technologies). Bare Co/C nanoparticles were taken as a background to improve the spectrum quality. Furthermore, transmission electron micrographs (TEM) of the bare particles were recorded (FEI, Tecnai F30 ST, operated at 300 kV), and element microanalysis (Yanaco, MT-6, CHN Corder) was used for the quantification of the surface functionalization. Amyloid β 42 Fibrillar Aggregate Characterization. To confirm the formation of amyloid β fibrillar aggregates and the interaction with nanomagnets, scanning transmission electron microscopy (STEM) micrographs were recorded (FEI, Nova NanoSEM 450, operated at 30 kV). Amyloid β 42 Fibrillar Aggregate Adsorption Experiments. The adsorption of Aβ fibrillar aggregates on functionalized nanomagnets was determined by the number of aggregates remaining in the supernatant after collection by nanomagnets. Concentrations of Aβ fibrillar aggregates were determined by staining the fibrils with amyloid-specific dye ThT (Sigma-Aldrich) and subsequent fluorescence measurements. Therefore, 75 μL of the ThT solution (25 μM in PBS) was added to 25 μL of a sample probe and gently mixed by pipetting up and down in a 96-well plate (Nuncolon). Each sample was recorded three times. After 2 h, the fluorescence signal was recorded (TECAN, excitation wavelength 450 nm, detection wavelength 485 nm, bandwidth 10 nm) in order to quantify the remaining number of Aβ fibrillar aggregates. BSA appeared to influence the fluorescence signal slightly, making separate calibration necessary. (For more detailed information, see the Supporting Information.)
■
RESULTS AND DISCUSSION Characterization of the Nanomaterials and Amyloid β 42 Fibrillar Aggregates. According to electron micrographs (Figure 1a) the carbon-coated cobalt particles used here reveal a particle size distribution of between 5 and 50 nm in diameter. The particles appear to be almost spherical in shape and are typically coated with two to three well-structured graphite layers.8 To determine the number of functional groups (i.e., the amount of physisorbed polymer on the particles), element microanalysis was used. Corresponding polymer shell thicknesses were calculated according to the previously described 14118
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
Figure 1. (a) Transmission electron micrograph of carbon-coated cobalt nanomagnets. (b) Amyloid-catching procedure using nanomagnets. (c) Scanning transmission electron micrographs of amyloid β 42 fibrillar aggregates. (d) Scanning transmission electron micrographs of amyloid β 42 fibrillar aggregates after binding to PEI-coated nanoparticles.
Figure 2. Diffuse reflectance Fourier transform infrared spectra of polymer-coated Co/C nanoparticles.
contaminants from various media. In noncomplex liquids, the capturing, particularly of larger species (proteins, dyes, and toxins), can in some cases occur even without any further specific surface modification. This effect is due to entropic reasons. When a large entity is physisorbing onto a surface, small molecules in close proximity are displaced. The large entity as a unit holds three directional degrees of motional freedom in solution but only two on the surface. In contrast, many small molecules (i.e., solvent) change from two to three motional degrees of freedom upon release, resulting in an overall gain in entropy.34 The synthesis of the materials used here with high-molecular-weight polymers is based on this effect. Even water-soluble polymers can be irreversibly physisorbed as long as other effects such as electrostatic repulsion are not predominant. This behavior was verified by quantitative element microanalysis of polyDADMAC-functionalized nanomagnets after different numbers of washing steps.
method.31 (For details, see the Supporting Information.) All results are summarized in Table 1. To verify further the presence of the polymers and functional groups on the particles, diffuse reflectance Fourier transform infrared spectroscopy (DRIFTS) was performed and compared to the spectra of the pure substances (Figure 2). Characteristic absorption bands of the polymers confirm their physisorbed presence. Shifts and changes in the characteristic absorption bands can result from surface-related effects.31 After at least 12 h of incubation at 37 °C in PBS, Aβ fibrillar aggregates show lengths in the range of a few micrometers (Figure 1c). In a magnetic separation process, the fibrillar aggregates bind to the nanomagnet surface and can thereafter be magnetically separated (Figure 1b,d). Irreversible Polymer Attachment. Because of their high specific surface areas, nanoparticles are suitable for capturing
Table 1. Carbon Contents Obtained from Element Microanalysis, Enabling the Calculation of the Polymer Coatings in Weight Percent and Resulting Layer Thicknesses by Knowing the Polymer Carbon Content
14119
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
Because the carbon content of the functionalized material remains constant and differs significantly from the starting material (Table 2), a stable attachment of the highly watersoluble polymer can be assumed.
Table 3. Bound Amounts of BSA on Differently Functionalized Nanoparticles after Distinct Times nanoparticle Co/C Co/C−PEI Co/C−poly(acrylic acid) Co/C− polyDADMAC
Table 2. Confirmation of Stable PolyDADMAC Functionalization by Carbon Content Analysis after Different Numbers of Washing Stepsa
a
number of washing steps
carbon content (wt %)
1 3 5 10
8.36 8.33 8.28 8.31
BSA bound after 5 min (wt %)
BSA bound overnight (wt %)
3.51 0.36 1.47
9.51 11.13 6.98
2.89
9.03
adsorption can be strongly time- and surface-type-dependent. For competing interactions between different species, structural changes for tight adsorption can have a significant influence. This is particularly important for magnetic extractions that are conducted within short time frames of typically minutes. Surface Adsorption of Amyloid β 42 Fibrillar Aggregates. In contrast to BSA, amyloid β 42 fibrils are much larger and more rigid. Thus, conformational adaptions upon surface contact are more unlikely to occur because strong enthalpy barriers would have to be overcome.39 Figure 3a,b
Starting material (unfunctionalized nanomagnets): 7.70 wt % carbon.
Surface Coatings for Selective Extraction. When target capturing structures are attached to surfaces for applications in medicine, diagnostics, and biotechnology, the validation of the binding selectivity is preferably quantified in media containing other large biomolecules such as proteins. Still, it must be stated that surface groups that are selective for a target in their free state are often constricted or even blocked for target interaction upon surface attachment. Congo red is a well-known staining agent with an affinity for amyloid fibrils. Its interaction is based on the penetration of its flat aromatic units into the characteristic fibrillar β-sheet structure (gap width ∼5 Å).24,25 Earlier publications claimed an arising selectivity for amyloid fibrils when Congo red or similar dye molecules were directly physisorbed onto iron oxide nanoparticles.35 However, because flat-attached Congo red molecules are geometrically hindered to access the narrow β-sheet structure of amyloid fibrils, the reason for the observed fibril selectivity must be addressed as either surface entropy effects or electrostatic interactions. Here, we systematically investigate the surface adsorption of Aβ fibrillar aggregates on carbon-coated cobalt nanomagnets that were functionalized with various polymers, revealing different physical characteristics (charge and hydrophobicity). To afford a better understanding of biologically more relevant competitive adsorption, the surface affinities were also examined in solutions containing excess bovine serum albumin (100-fold excess in weight regarding fibrillar aggregates). This simulates an environment similar to that of blood plasma.36 Surface Adsorption of BSA. Because of its flexibility, BSA is known to undergo thermodynamically favorable conformational changes upon surface contact, regardless of the hydrophilic or hydrophobic nature of the surface.37,38 However, this event can take several hours to complete. To understand to what extent this happens on the materials used here, differently coated nanoparticles were stirred in a 1 wt % BSA solution for 5 min and overnight. After extensive washing, element microanalysis was conducted to quantify the bound amount of BSA from the resulting change in the carbon content (Table 3). All particles show considerable adsorption after prolonged exposition (overnight) to dissolved BSA. However, after short exposure to the same conditions, strong differences in adsorption occur. Only proteins that developed strong surface interactions in a short time were bound. Because PEI-coated nanomagnets developed the highest adsorption after a long exposure but revealed the lowest adsorption amounts after 5 min, it seems likely that time-consuming conformational changes are needed to bind tightly to this surface. Furthermore, previous reports38 have shown that the extent of BSA
Figure 3. Adsorption isotherms of amyloid β 42 fibrillar aggregates without (squares) and with BSA (circles) on (a) blank carbon-coated nanomagnets, (b) polyethylenimine-functionalized nanomagnets, (c) poly(acrylic acid)-functionalized nanomagnets, and (d) polyDADMAC-functionalized nanomagnets (broken lines, 60% extraction).
depicts adsorption isotherms of Aβ fibrillar aggregates on blank as well as PEI-coated nanomagnets at different particle concentrations. Both particle types exhibit similar adsorption behavior in solutions with PBS only. At the highest tested particle concentrations, we found almost complete fibril extraction (>95%). Comparable results were obtained for chemically similar chitosan-coated particles (Supporting Information). All adsorption isotherms reveal a curved shape, which is likely to be a dilution effect of the fibrillar concentration in the extracted medium. Adsorption isotherms on the charged polymer surfaces (negative charge under neutral conditions, poly(acrylic acid); positive charge, polydiallyl14120
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
Supporting Information for polyvinylsulfate- and poly(ethylene glycol)-coated particles.) As found earlier, BSA is not able to bind rapidly to PEI-functionalized particles. This enables a favorable adsorption environment for fibrils within the conducted extraction time frame, despite a much higher BSA concentration compared to that of fibrils (100-fold).
(dimethylammonium chloride) (i.e., polyDADMAC)) depicted in Figure 3c,d show a significantly lower binding affinity. Following these results, it is likely that hydrophobic interactions are present. To verify this assumption further, the pH value of the extraction solution was varied in order to change surface charges (here, on poly(acrylic acid)-coated particles). Poly(acrylic acids) have pKa values of between 4.7 and 7, depending on the molecular weight, branching, and salt concentrations.40 This leads to various ratios of protonated carboxylic acid groups on the surface and should therefore influence the extraction performance if the above-stated assumptions are valid. A strong dependence of the fibril extraction extent reaching 85% at pH 4.4 and no extraction at pH 9.2 (following an almost linear trend) becomes apparent at constant particle and salt concentrations (Figure 4). The more carboxylic groups that
■
CONCLUSIONS In this study, we investigated the interaction of Aβ fibrillar aggregates as a model biostructure with physical surface properties of polymer-coated cobalt-based nanomagnets. Because of the possibility of magnetic separation within seconds, our setup represents a valuable tool for biostructure−surface interaction analysis, particularly for in vivo situations. Thus, it represents an alternative to current in vitro stopped-flow techniques, especially for larger biologically relevant sample volumes. In protein-rich model solutions under neutral conditions (1 wt % BSA, 100-fold excess in mass compared to amyloid fibrils), a competitive adsorption environment (similar to blood) between the fibrils and BSA was established. BSA is known for its high structural adaptability when binding to surfaces. We found that all investigated particle types revealed considerable BSA binding after prolonged exposure to BSA-rich solutions. However, binding extents after minutes largely differed. This effect can be used to target a lower-concentration biological species such as Aβ fibrillar aggregates. We observed that the latter have weaker affinities for charged polymer surfaces. This reveals the importance of hydrophobic interactions. By varying the pH level of solutions containing poly(acrylic acid)-coated nanomagnets and thus the surface charge on the particles, we furthermore confirmed this effect. Following these insights, we conducted extractions with good selectivity toward Aβ fibrils using PEI-functionalized nanoparticles. Because our findings are valid not only for the system described here but also more generally for a vast range of other biomolecules and structures, we want to highlight the practicability of this simple and rapid surface interaction analysis tool based on highly magnetic nanoparticles.
Figure 4. pH dependence of the surface adsorption of amyloid β 42 fibrillar aggregates on poly(acrylic acid)- and polyethylenimine (PEI)functionalized carbon-coated cobalt nanomagnets. Because the poly(acrylic acid)-coated surface is increasingly deprotonated at increasing pH, hydrophobic interactions decrease. At pH 9.2, surface adsorption is no longer possible. In the case of polyethylenimine, more constant behavior occurs because it provides constant surface properties under all examined conditions.
are deprotonated, the fewer hydrophobic interactions that may occur. Additionally, Hetenyi et al.41 found that Aβ fibrillar aggregates present a slightly negative charge at a pH of 7, meaning that protonation at higher acidity leads to a lesscharged fibril surface, further enhancing the entropy-driven interaction. Equivalent adsorption experiments with polyethylenimine-coated particles as shown in Figure 4 resulted in rather constant adsorption behavior when the pH is altered (presence of primary, secondary, and tertiary amines with different pKa’s42). This situation allows more constant binding over the observed pH range. Wang et al.43 found similar effects on differently charged self-assembled monolayer surfaces in an atomic force microscopy study. Competitive Surface Adsorption of Amyloid β 42 Fibrillar Aggregates (BSA Present). The presence of large concentrations of BSA compared to the target species establishes a strongly competitive environment for selective extractions. Such situations are frequent in natural systems such as blood. Because it often remains unclear what kind of interaction mechanisms takes place, magnetic nanoparticles can act as a convenient rapid analysis tool. As expected, all particles adsorbed amyloid fibrils to a lesser extent when BSA was present (Figure 3a−d). Despite this, the PEI-coated nanomagnets still revealed a considerable ability to bind amyloid fibrillar aggregates (75% extraction extent at highest particle concentration). For all of the other tested particle types, the drop in extraction ability was significantly higher. (See also the
■
ASSOCIATED CONTENT
S Supporting Information *
TEM micrographs, adsorption isotherms of other polymer surface coatings, IR spectra, and elemental microanalysis data. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Authors
*Phone: +81 48 467 93 12. Fax: +81 48 462 46 58. E-mail: [email protected]. *Phone: +41 44 632 09 80. Fax: +41 44 633 10 83. E-mail: [email protected]. Author Contributions
A.C.C.R. and C.M.S. contributed equally to this work. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We are grateful to the RIKEN Materials Characterization Support Unit for element analysis and TEM analysis. A.C.C.R. and C.M.S. are RIKEN short-term international program 14121
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
X.; Greengard, P.; Relkin, N. R. Intraneuronal A beta 42 accumulation in human brain. Am. J. Pathol. 2000, 156, 15−20. (18) Platt, G. W.; McParland, V. J.; Kalverda, A. P.; Homans, S. W.; Radford, S. E. Dynamics in the unfolded state of beta(2)-microglobulin studied by NMR. J. Mol. Biol. 2005, 346, 279−294. (19) Gertz, M. A.; Kyle, R. A. Myopathy in primary systemic amyloidosis. J. Neurol., Neurosurg., Psychiatry 1996, 60, 655−660. (20) Gharibyan, A. L.; Zamotin, V.; Yanamandra, K.; Moskaleva, O. S.; Margulis, B. A.; Kostanyan, I. A.; Morozova-Roche, L. A. Lysozyme amyloid oligomers and fibrils induce cellular death via different apoptotic/necrotic pathways. J. Mol. Biol. 2007, 365, 1337−1349. (21) Novitskaya, V.; Bocharova, O. V.; Bronstein, I.; Baskakov, I. V. Amyloid fibrils of mammalian prion protein are highly toxic to cultured cells and primary neurons. J. Biol. Chem. 2006, 281, 13828− 13836. (22) Engel, M. F. M.; Khemtemourian, L.; Kleijer, C. C.; Meeldijk, H. J. D.; Jacobs, J.; Verkleij, A. J.; de Kruijff, B.; Killian, J. A.; Hoppener, J. W. M. Membrane damage by human islet amyloid polypeptide through fibril growth at the membrane. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 6033−6038. (23) Zako, T.; Sakono, M.; Kobayashi, T.; Sörgjerd, K.; Nilsson, K. P. R.; Hammarström, P.; Lindgren, M.; Maeda, M. Cell interaction study of amyloid by using luminescent conjugated polythiophene: implication that amyloid cytotoxicity is correlated with prolonged cellular binding. ChemBioChem 2012, 13, 358−363. (24) Biancalana, M.; Koide, S. Molecular mechanism of Thioflavin-T binding to amyloid fibrils. Biochim. Biophys. Acta, Proteins Proteomics 2010, 1804, 1405−1412. (25) Schutz, A. K.; Soragni, A.; Hornemann, S.; Aguzzi, A.; Ernst, M.; Bockmann, A.; Meier, B. H. The amyloid-Congo red interface at atomic resolution. Angew. Chem., Int. Ed. 2011, 50, 5956−5960. (26) Puchtler, H.; Sweat, F.; Levine, M. On binding of Congo red by amyloid. J. Histochem. Cytochem. 1962, 10, 355−364. (27) Hobbs, J. R.; Morgan, A. D. Fluorescence microscopy with thioflavine-T in diagnosis of amyloid. J. Pathol. Bacteriol. 1963, 86, 437−442. (28) Fuhrer, R.; Herrmann, I. K.; Athanassiou, E. K.; Grass, R. N.; Stark, W. J. Immobilized beta-cyclodextrin on surface-modified carboncoated cobalt nanomagnets: reversible organic contaminant adsorption and enrichment from water. Langmuir 2011, 27, 1924−1929. (29) Schumacher, C. M.; Herrmann, I. K.; Bubenhofer, S. B.; Gschwind, S.; Hirt, A.-M.; Beck-Schimmer, B.; Günther, D.; Stark, W. J. Quantitative recovery of magnetic nanoparticles from flowing blood: trace analysis and the role of magnetization. Adv. Funct. Mater. 2013, 39, 4888−4896. (30) Zeltner, M.; Grass, R. N.; Schaetz, A.; Bubenhofer, S. B.; Luechinger, N. A.; Stark, W. J. Stable dispersions of ferromagnetic carbon-coated metal nanoparticles: preparation via surface initiated atom transfer radical polymerization. J. Mater. Chem. 2012, 22, 12064− 12071. (31) Schumacher, C. M.; Grass, R. N.; Rossier, M.; Athanassiou, E. K.; Stark, W. J. Physical defect formation in few layer graphene-like carbon on metals: influence of temperature, acidity, and chemical functionalization. Langmuir 2012, 28, 4565−4572. (32) Koehler, F. M.; Rossier, M.; Waelle, M.; Athanassiou, E. K.; Limbach, L. K.; Grass, R. N.; Gunther, D.; Stark, W. J. Magnetic EDTA: coupling heavy metal chelators to metal nanomagnets for rapid removal of cadmium, lead and copper from contaminated water. Chem. Commun. 2009, 4862−4864. (33) Gomori, G. Preparation of buffers for use in enzyme studies. Methods Enzymol. 1955, 1, 138−146. (34) Melzak, K. A.; Sherwood, C. S.; Turner, R. F. B.; Haynes, C. A. Driving forces for DNA adsorption to silica in perchlorate solutions. J. Colloid Interface Sci. 1996, 181, 635−644. (35) Skaat, H.; Margel, S. Synthesis of fluorescent-maghemite nanoparticles as multimodal imaging agents for amyloid-beta fibrils detection and removal by a magnetic field. Biochem. Biophys. Res. Commun. 2009, 386, 645−649.
associates. Financial support by ETH Zurich, RIKEN Institute, and the Swiss National Science Foundation (no. 406440131268) is greatly appreciated.
■
REFERENCES
(1) Zhang, S. X.; Zhao, X. L.; Niu, H. Y.; Shi, Y. L.; Cai, Y. Q.; Jiang, G. B. Superparamagnetic Fe3O4 nanoparticles as catalysts for the catalytic oxidation of phenolic and aniline compounds. J. Hazard. Mater. 2009, 167, 560−566. (2) Schatz, A.; Hager, M.; Reiser, O. Cu(II)-Azabis(oxazoline)complexes immobilized on superparamagnetic magnetite@silica-nanoparticles: a highly selective and recyclable catalyst for the kinetic resolution of 1,2-diols. Adv. Funct. Mater. 2009, 19, 2109−2115. (3) Megia-Fernandez, A.; Ortega-Munoz, M.; Lopez-Jaramillo, J.; Hernandez-Mateo, F.; Santoyo-Gonzalez, F. Non-magnetic and magnetic supported copper(I) chelating adsorbents as efficient heterogeneous catalysts and copper scavengers for click chemistry. Adv. Synth. Catal. 2010, 352, 3306−3320. (4) LaVan, D. A.; McGuire, T.; Langer, R. Small-scale systems for in vivo drug delivery. Nat. Biotechnol. 2003, 21, 1184−1191. (5) Weissleder, R.; Kelly, K.; Sun, E. Y.; Shtatland, T.; Josephson, L. Cell-specific targeting of nanoparticles by multivalent attachment of small molecules. Nat. Biotechnol. 2005, 23, 1418−1423. (6) Hong, R. Y.; Feng, B.; Chen, L. L.; Liu, G. H.; Li, H. Z.; Zheng, Y.; Wei, D. G. Synthesis, characterization and MRI application of dextran-coated Fe3O4 magnetic nanoparticles. Biochem. Eng. J. 2008, 42, 290−300. (7) Zhao, M.; Beauregard, D. A.; Loizou, L.; Davletov, B.; Brindle, K. M. Non-invasive detection of apoptosis using magnetic resonance imaging and a targeted contrast agent. Nat. Med. 2001, 7, 1241−1244. (8) Grass, R. N.; Athanassiou, E. K.; Stark, W. J. Covalently functionalized cobalt nanoparticles as a platform for magnetic separations in organic synthesis. Angew. Chem., Int. Ed. 2007, 46, 4909−4912. (9) Herrmann, I. K.; Bernabei, R. E.; Urner, M.; Grass, R. N.; BeckSchimmer, B.; Stark, W. J. Device for continuous extracorporeal blood purification using target-specific metal nanomagnets. Nephrol., Dial., Transplant. 2011, 26, 2948−U1516. (10) Wei, J. H.; Igarashi, T.; Okumori, N.; Maetani, T.; Liu, B. L.; Yoshinari, M. Influence of surface wettability on competitive protein adsorption and initial attachment of osteoblasts. Biomed. Mater. 2009, 4, 045002. (11) Tsougeni, K.; Petrou, P. S.; Papageorgiou, D. P.; Kakabakos, S. E.; Tserepi, A.; Gogolides, E. Controlled protein adsorption on microfluidic channels with engineered roughness and wettability. Sens. Actuators, B 2012, 161, 216−222. (12) Pasche, S.; Voros, J.; Griesser, H. J.; Spencer, N. D.; Textor, M. Effects of ionic strength and surface charge on protein adsorption at PEGylated surfaces. J. Phys. Chem. B 2005, 109, 17545−17552. (13) Xie, H. G.; Li, X. X.; Lv, G. J.; Xie, W. Y.; Zhu, J.; Luxbacher, T.; Ma, R.; Ma, X. J. Effect of surface wettability and charge on protein adsorption onto implantable alginate-chitosan-alginate microcapsule surfaces. J. Biomed. Mater. Res. 2010, 92A, 1357−1365. (14) Tan, F.; Zhang, Y.; Mi, W.; Wang, J.; Wei, J.; Cai, Y.; Qian, X. Enrichment of phosphopeptides by Fe3+-immobilized magnetic nanoparticles for phosphoproteome analysis of the plasma membrane of mouse liver. J. Proteome Res. 2008, 7, 1078−1087. (15) Zhou, H. J.; Tian, R. J.; Ye, M. L.; Xu, S. Y.; Feng, S.; Pan, C. S.; Jiang, X. G.; Li, X.; Zou, H. F. Highly specific enrichment of phosphopeptides by zirconium dioxide nanoparticles for phosphoproteome analysis. Electrophoresis 2007, 28, 2201−2215. (16) Zako, T.; Sakono, M.; Hashimoto, N.; Hara, M.; Maeda, M. Bovine insulin filaments induced by reducing disulfide bonds show a different morphology, secondary structure, and cell toxicity from intact insulin amyloid fibrils. Biophys. J. 2009, 96, 3331−3340. (17) Gouras, G. K.; Tsai, J.; Naslund, J.; Vincent, B.; Edgar, M.; Checler, F.; Greenfield, J. P.; Haroutunian, V.; Buxbaum, J. D.; Xu, H. 14122
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
Langmuir
Article
(36) Wang, H.; Shi, Z. M. In vitro biodegradation behavior of magnesium and magnesium alloy. J. Biomed. Mater. Res. 2011, 98B, 203−209. (37) Norde, W.; Giacomelli, C. E. BSA structural changes during homomolecular exchange between the adsorbed and the dissolved states. J. Biotechnol. 2000, 79, 259−268. (38) Lee, S. H.; Ruckenstein, E. Adsorption of proteins onto polymeric surfaces of different hydrophilicitiesa case study with bovine serum albumin. J. Colloid Interface Sci. 1988, 125, 365−379. (39) Andrade, J. D.; Hlady, V. Protein Adsorption and Materials Biocompatibility: A Tutorial Review and Suggested Hypotheses. Biopolymers/Non-Exclusion HPLC; Springer: Berlin, 1986; Vol. 79, pp 1−63. (40) Shibayama, M.; Ikkai, F.; Inamoto, S.; Nomura, S.; Han, C. C. pH and salt concentration dependence of the microstructure of poly(N-isopropylacrylamide-co-acrylic acid) gels. J. Chem. Phys. 1996, 105, 4358−4366. (41) Hetenyi, A.; Fulop, L.; Martinek, T. A.; Weber, E.; Soos, K.; Penke, B. Ligand-induced flocculation of neurotoxic fibrillar A beta(1− 42) by noncovalent crosslinking. ChemBioChem 2008, 9, 748−757. (42) Demadis, K. D.; Paspalaki, M.; Theodorou, J. Controlled release of bis(phosphonate) pharmaceuticals from cationic biodegradable polymeric matrices. Ind. Eng. Chem. Res. 2011, 50, 5873−5876. (43) Wang, Q. M.; Shah, N.; Zhao, J.; Wang, C. S.; Zhao, C.; Liu, L. Y.; Li, L. Y.; Zhou, F. M.; Zheng, J. Structural, morphological, and kinetic studies of beta-amyloid peptide aggregation on self-assembled monolayers. Phys. Chem. Chem. Phys. 2011, 13, 15200−15210.
14123
dx.doi.org/10.1021/la4026427 | Langmuir 2013, 29, 14117−14123
