Engineered Polymer Nanoparticles Containing HydrophobicDipeptide for Inhibition of Amyloid‑β FibrillationHadas Skaat,† Ravit Chen,†,‡ Igor Grinberg, and Shlomo Margel*

Department of Chemistry, Bar-Ilan Institute of Nanotechnology and Advanced Materials, Ramat-Gan 52900, Israel

*S Supporting Information

ABSTRACT: Protein aggregation into amyloid fibrils isimplicated in the pathogenesis of many neurodegenerativediseases. Engineered nanoparticles have emerged as a potentialapproach to alter the kinetics of protein fibrillation process.Yet, there are only a few reports describing the use ofnanoparticles for inhibition of amyloid-β 40 (Aβ40) peptideaggregation, involved in Alzheimer’s disease (AD). In thepresent study, we designed new uniform biocompatible amino-acid-based polymer nanoparticles containing hydrophobicdipeptides in the polymer side chains. The dipeptide residueswere designed similarly to the hydrophobic core sequence of Aβ. Poly(N-acryloyl-L-phenylalanyl-L-phenylalanine methyl ester)(polyA-FF-ME) nanoparticles of 57 ± 6 nm were synthesized by dispersion polymerization of the monomer A-FF-ME in 2-methoxy ethanol, followed by precipitation of the obtained polymer in aqueous solution. Cell viability assay confirmed that nosignificant cytotoxic effect of the polyA-FF-ME nanoparticles on different human cell lines, e.g., PC-12 and SH-SY5Y, wasobserved. A significantly slow secondary structure transition from random coil to β-sheets during Aβ40 fibril formation wasobserved in the presence of these nanoparticles, resulting in significant inhibition of Aβ40 fibrillation kinetics. However, thepolyA-FF-ME analogous nanoparticles containing the L-alanyl-L-alanine (AA) dipeptide in the polymer side groups, polyA-AA-ME nanoparticles, accelerate the Aβ40 fibrillation kinetics. The polyA-FF-ME nanoparticles and the polyA-AA-ME nanoparticlesmay therefore contribute to a mechanistic understanding of the fibrillation process, leading to the development of therapeuticstrategies against amyloid-related diseases.

■ INTRODUCTION

The formation of amyloid aggregates is a pathological hallmarkof many diseases, including Parkinson’s, Huntington’s, madcow, prion, and Alzheimer’s disease (AD).1−3 The commonfeature of amyloid diseases is the transition of proteins from thenormally soluble form into amyloid fibrils, organized mainlyinto cross β-sheets, which accumulate in the extracellular spaceof various tissues.4−6 The extracellular deposits that character-ize AD are composed of fibrils of a small peptide with 39−43amino acids, the amyloid-β (Aβ) peptide.7 Aβ is continuouslysecreted by normal cells in culture and is detected as a solublepeptide in the plasma and cerebrospinal fluid of healthyindividuals.8,9 In AD patients, Aβ exists as insoluble fibrillaraggregates.10 It is generally accepted that the Aβ peptides canself-assemble to form neurological toxic aggregates with variousmorphologies, such as dimers, oligomers, protofibrils, andfibrils, which eventually deposit as insoluble plaques and causethe death of brain cells.11 Currently, there are no effective drugsto cure these diseases and treatment options are extremelylimited.12 A coherent pharmacological approach for preventingamyloid aggregation encompasses agents able to specificallystabilize the initial conformations through interfering with theintermediate species and/or destabilizing the β-sheet con-formations.13

Several studies on the aggregation process of Aβ haveidentified the critical peptidic sequence involved in amyloidaggregates formation.14−18 The hydrophobic core from residues17−20 of Aβ, LVFF sequence, is crucial for the formation of β-sheet structures.14−18 It was also demonstrated that thispeptidic region binds to its homologous sequence in Aβ andprevents its aggregation into amyloid fibrils.19−21 Suchinhibition is mainly based on recognition between the pairsof phenylalanine residues (FF) through hydrophobic inter-actions.14−21 This FF motif has served as a key platform for thedevelopment of peptide and peptidomimetic inhibitors of Aβfibrillation.17 Although peptide or protein analogues withspecific binding sites may be of primary importance for studiesof neurodegenerative disorders, therapeutic agents have notbeen developed within this strategy. The main problems are theblood-brain barrier (BBB) permeability, the complexity of theirsynthesis, and their low in vivo stability and efficacy.22

Recent literature concerning novel biomaterials indicatesincreasing interest in developing nanoparticles to detect,prevent, and treat protein-misfolding diseases.23−25 Theirpotential influence on protein fibrillation is a function of both

Received: April 8, 2012Revised: August 17, 2012Published: August 17, 2012

Article

pubs.acs.org/Biomac

© 2012 American Chemical Society 2662 dx.doi.org/10.1021/bm3011177 | Biomacromolecules 2012, 13, 2662−2670

the surface interfacial properties, including charge, of thenanoparticle and its enormous surface area to volume ratio.Increased local protein concentration on the nanoparticlesurface and/or changes in protein conformation upon bindingcould promote aggregation while trapping of early intermedi-ates may inhibit further aggregation.26 Engineered biocompat-ible nanoparticles offer advantages as therapeutic agents foramyloid-related diseases since they allow modification ofsurface properties so as to control the adsorption andinteraction processes. Moreover, in vitro and in vivo studieshave shown that nanoparticles are capable of overcoming thedifficulty of crossing the BBB and have greater in vivostability.27,28

Various nanoparticles such as copolymer particles of N-isopropylacrylamide/N-tert-butylacrylamide, cerium oxide,quantum dots (QDs), carbon nanotubes, and titanium oxidehave been reported to promote protein assembly into amyloidfibrils in vitro, by assisting the nucleation process.29,30 Only afew studies have been reported on the inhibitory effect ofnanoparticles on the Aβ fibrillation process. Very recently,Cabaleiro-Lago et al. reported the inhibition of the Aβ40 fibrilformation by copolymer nanoparticles of variable hydro-phobicity,31 and also demonstrated the dual effect ofcommercial polystyrene (PS) nanoparticles with aminomodification toward the Aβ40 and Aβ42 fibril formation.32 Yooet al. have shown an inhibition effect of CdTe QDs on Aβ40fibrillation.33 Fluorinated nanoparticles34 and sulfonated andsulfated PS nanoparticles35 have been also reported as potentialcandidates for the inhibition of Aβ fibril formation. Ourprevious work showed that the Leu-Pro-Phe-Phe-Asp (LPFFD)peptide conjugated iron oxide nanoparticles slightly inhibit theAβ40 fibrillation.

36

Poly(amino acid) nanoparticles have recently attracted greatattention due to their potential nontoxicity, biocompatibility,nonimmunogenicity, and biofunctionality. These nanoparticlescontaining amino acid moieties are potentially useful in manydifferent biomedical applications, such as drug or gene deliveryagents, tissue engineering scaffolds, and chiral recognition.37−40

The synthesis and radical polymerization of various acrylmonomers having amino acid moieties in the side chain havebeen reported.41 Their corresponding polymers, poly(N-acryloyl amino acids), are expected to be functional materials,and much attention has been paid to their unique polymer-

ization behavior, structures, and properties.42−44 Yet, to the bestof our knowledge, there is no report on the preparation ofpolymeric nanoparticles from N-acryloyl amino acid mono-mers.The present article describes a novel designed synthesis of

uniform biocompatible amino-acid-based polymer nanopar-ticles containing hydrophobic dipeptides in the polymer sidechains. The dipeptide residues were designed similarly to thehydrophobic core sequence of Aβ. Poly(N-acryloyl-L-phenyl-alanyl-L-phenylalanine methyl ester) (polyA-FF-ME) nano-particles of 57 ± 6 nm were synthesized by dispersionpolymerization of the monomer A-FF-ME in 2-methoxyethanol, followed by precipitation of the obtained polymer inaqueous solution. These nanoparticles were well-characterizedincluding their cytotoxic effect on different human cell lines, i.e.,PC-12 (pheochromocytoma cell line) and SH-SY5Y (neuro-blastoma cell line). Kinetics of the Aβ40 fibrillation process inthe absence and the presence of varying concentrations of thepolyA-FF-ME nanoparticles, and their analogues polyA-AA-MEnanoparticles containing the L-alanyl-L-alanine (AA) dipeptideinstead of FF in the polymer side chains, have been elucidated.A significantly slow direct secondary structure transition fromrandom coil to β-sheets during Aβ40 fibril formation wasobserved in the presence of the polyA-FF-ME nanoparticles.This observation indicates that these nanoparticles significantlyincrease the lag time and hence the kinetics of Aβ40 fibrillation.However, the opposite behavior was observed in the presenceof the polyA-AA-ME nanoparticles, indicating acceleration ofAβ40 fibrillation kinetics.

■ MATERIALS AND METHODSMaterials. The following analytical-grade chemicals were pur-

chased from commercial sources and used without further purification:L-phenylalanyl-L-phenylalanine (FF), acetyl chloride, methanol anhy-drous, acryloyl chloride, triethylamine (TEA), dimethylaminopyridine(DMAP), dry dichloromethane (DCM), hydrochloric acid (1 M),hexane, magnesium sulfate, sodium bicarbonate, sodium hydrogensulfate, calcium chloride, 2-methoxyethanol, polyvinylpyrrolidone(PVP, MW 360 000), benzoyl peroxide (BP), amyloid β proteinfragment 1−40 (Aβ40), trifluoroacetic acid (TFA), 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), dimethyl sulfoxide (DMSO), andThioflavin T (ThT) from Sigma (Israel); L-alanyl-L-alanine methylester (AA-ME) hydrochloride salt from D-Chem (Israel); Dulbecco’smodified eagle medium (DMEM), fetal calf serum (FCS), L-glutamine,penicillin, streptomycin, and 2,3-bis-(2-methoxy-4-nitro-5-sulfophen-

Figure 1. Synthetic route of the polyA-FF-ME nanoparticles.

Biomacromolecules Article

dx.doi.org/10.1021/bm3011177 | Biomacromolecules 2012, 13, 2662−26702663

yl)-2H-tetrazolium-5-carboxanilide (XTT) cell viability assay kit fromBiological-Industries (Israel). Water was purified by passing deionizedwater through an Elgastat Spectrum reverse osmosis system (Elga,High Wycombe, UK). The cell lines PC-12 and SH-SY5Y werepurchased from ATCC (USA). The culture medium used in this studywas composed of 90% DMEM, 10% FCS, 2 mM L-glutamine, 50units/mL penicillin, and 50 μg/mL streptomycin.Synthesis of A-FF-ME. The monomer A-FF-ME was synthesized

by esterification of FF followed by its reaction with acryloyl chloride,according to a procedure reported previously,45,46 as shown in Figure 1(steps I−II). Briefly, FF (1 g, 3.2 mmol) was dissolved in a flaskcontaining anhydrous methanol (50 mL). The flask was immersed inan ice bath and acetyl chloride (5 mL) was then added dropwise. Thereaction mixture was stirred for 2 h at 0 °C and then stirred at roomtemperature overnight. The obtained FF methyl ester (FF-ME)hydrochloride salt was then isolated by removal under reducedpressure of methyl acetate and methanol. Pure white brilliant crystalsof FF-ME hydrochloride salt were obtained in quantitative yield(98%). The obtained FF-ME hydrochloride salt (1.08 g, 3.0 mmol)was dissolved in dried DCM (45 mL). TEA (890 μL, 6.4 mmol) andDMAP (40 mg, 0.32 mmol) were then added to the solution. Thereaction mixture was immersed in an ice bath. Acryloyl chloride (410μL, 5 mmol) dissolved in dry DCM (5 mL) was then added dropwiseto the reaction mixture. The solution was stirred at room temperatureovernight and then diluted with further amount of DCM (40 mL).The DCM solution was then washed with HCl (1 M) and brine. Theorganic phase was then dried over MgSO4, filtered, and evaporated toproduce a white solid. Pure white crystals of N-acryloyl-FF-ME (A-FF-ME) were obtained by recrystallization of the white solid twice fromDCM/hexane (yield 44%).

1H NMR (CDCl3, 200 MHz): δ 7.26−7.17 (m, 8H, aromatichydrogen), 6.95−7.00 (m, aromatic hydrogen, 2H), 6.3 (dd, J2 = 1.6Hz, J3 = 16.8 Hz, 1H, CH2CH), 6.14 (m, NH, 2H), 6.1 (dd, J3cis =10.0 Hz J3trans = 16.8 Hz, 1H, CH2CH), 5.65−5.70 (dd, J2 = 1.6Hz, J3 = 10.0 Hz, 1H, CH2CH), 4.75 (m, 2H, C(O)NHCH), 3.68(s, 3H, OCH3), 2.93−3.13 (m, 4H, PhCH2).

13C NMR (CDCl3, 200MHz): δ 171.1 (C(O)NH, 1C), 170.1 (C(O)OCH3, 1C), 165.1(C(O)CHCH, 1C), 130.2, 135.4 (2C, aromatic quaternary carbon),130.2, 129.3, 129.1, 128.67, 128.60, 127.29 (aromatic carbon, 10C),127.14, 127.08 (2C, double bond), 54.0, 53.2 (2C, C(O)NHCH), 52.2(1C, OCH3), 38.1, 37.8 (2C, PhCH2). TOF MS ES+: 381 (MH+,100%), 403 (MNa+, 50%). HRMS MALDI: 403.1628 (MNa+).Elemental analysis (calculated): C, 69.46; N, 7.36; H, 6.36; O,16.82. Found: C, 68.22; N, 7.21; H, 6.47; O, 16.45.Synthesis of PolyA-FF-ME Nanoparticles. In a typical experi-

ment, 30 mg of the vinylic monomer A-FF-ME, 10 mg PVP, and 2 mgof BP were added to 1 mL of nitrogen-bubbled 2-methoxy ethanol.The mixture was shaken at room temperature to dissolve the solids,giving concentrations of 3%, 1%, and 0.2% (w/v) A-FF-ME, PVP, andBP, respectively. The mixture was then shaken at 75 οC for 18 h(Figure 1, step III). The formed soluble polyA-FF-ME wasprecipitated as uniform nanoparticles by adding 100 μL of its solutionto 1 mL distilled water (Figure 1, step IV). The formed polyA-FF-MEnanoparticles were then washed by extensive dialysis against water.Dried polyA-FF-ME nanoparticles were then obtained by lyophiliza-tion.

1H NMR (CDCl3, 200 MHz): δ 7.3−6.8 (broad, aromatichydrogens and amide hydrogens, 12H), 4.8−4.3 (broad, C(O)NHCH,2H), 3.7−3.3 (broad, 3H, OCH3), 3.2−2.7 (broad, 4H, PhCH2), 2.4−2.1 (broad, 1H, CHCH2), 2.1−1.8 (broad, 2H, CHCH2).Synthesis of A-AA-ME. The monomer A-AA-ME was synthesized

by the reaction of commercial AA-ME hydrochloride salt with acryloylchloride in cold water, according to a procedure that was reportedpreviously.42 Briefly, the AA-ME hydrochloride salt (1 g, 4.74 mmol)was dissolved in water (10 mL). NaHCO3 (880 mg, 9.49 mmol) wasthen added to the solution. The reaction mixture was immersed in anice bath. Acryloyl chloride (470 μL, 5.7 mmol) was then addeddropwise to the solution. The chilled solution was vigorously stirredfor 30 min. The solution was acidified with NaHSO4 (1 M) to pH 3and was then washed three times with ethyl acetate. The organic phase

was then dried over MgSO4, filtered, and evaporated to produce whitesolid. Pure white crystals of N-acryloyl-AA-ME (A-AA-ME) wereobtained by recrystallization of the white solid twice from DCM/hexane (yield 44%).

1H NMR (CDCl3, 300 MHz): δ 6.6 (broad doublet, J3 = 6.8 Hz,NH near the double bond, 1H), 6.3 (dd, J2 = 0.6 Hz, J3 = 16.8 Hz, 1H,CH2CH), 6.28 (broad, NH, 1H), 6.16−6.05 (dd, J3cis = 10.5 Hz,J3trans = 16.8 Hz, 1H, CH2CH), 5.69 (dd, J2 = 0.6 Hz, J3 = 10.5 Hz,1H, CH2CH), 4.59 (double-quartet, J3 = 6.8 H, J3 = 7.2 Hz, 1H,C(O)NHCH), 4.54 (double-quartet, J3 = 6.8 H, J3 = 7.2 Hz, 1H,C(O)NHCH), 3.76 (s, 3H, OCH3), 1.43 (d, J3 = 6.8 3H, CHCH3),1.41 (d, J3 = 7.2, 3H, CHCH3).

13C NMR (CDCl3, 200 MHz): δ 173.4(C(O)NH, 1C), 172.0 (C(O)OCH3, 1C), 165.4 (C(O)CHCH,1C), 130.6, 127.6 (2C, double bond), 52.9 (1C, OCH3), 49.0, 48.5(2C, C(O)NHCHCH3), 18.8, 18.5 (2C, CHCH3). TOF MS ES+: 229(MH+, 12%), 251 (MNa+, 100%). HRMS MALDI: 229.1215 (MH+).Elemental analysis (calculated): C, 52.62; N, 12.27; H, 7.07. Found: C,51.96; N, 12.30; H, 7.02.

Synthesis of PolyA-AA-ME Nanoparticles. The polyA-AA-MEwas prepared by a similar procedure to that described above for thepolyA-FF-ME, substituting the monomer A-FF-ME for the A-AA-ME.The formed soluble polyA-FF-ME was precipitated as uniformnanoparticles by adding 100 μL of its solution to 0.5 mL distilledwater. The formed polyA-AA-ME nanoparticles were then washed byextensive dialysis against water. Dried polyA-AA-ME nanoparticleswere then obtained by lyophilization.

1H NMR (D2O, 300 MHz): δ 4.6−4.2 (broad, C(O)NHCH, 2H),3.7−3.6 (broad, 3H, OCH3), 2.3−2.1 (broad, 1H, CHCH2), 2.1−1.8(broad, 2H, CHCH2), 1.5−1.3 (broad 6H, CHCH3).

Cytotoxicity Assay. XTT assay was performed to determine thecytotoxicity of the polyA-FF-ME nanoparticles on PC-12 and SH-SY5Y cell lines.47 Cells were seeded in a 96-well plate at a density of 1× 104 cells/well in 100 μL culture medium and grown in a humidified5% CO2 atmosphere at 37

οC. After 18 h at 37 °C, different volumesof polyA-FF-ME nanoparticles dispersed in water were added to thecells, giving final concentrations of 0.02 and 0.2 mg/mL per well. Afterincubation for 24 or 48 h at 37 °C, 50 μL XTT solution was added toeach well according to the kit manufacturer’s instructions. Absorbancewas read at 480 nm, and the absorbance of correspondingconcentrations of the nanoparticles was subtracted from the reading.Cell viability was determined as a percentage of the negative control(cultured cells in medium without nanoparticles).

Preparation of the Aβ40 Fibrils in the Absence and Presenceof the PolyA-FF-ME Nanoparticles or PolyA-AA-ME Nano-particles. Lyophilized Aβ40, synthesized by Sigma Israel, was stored at−20 °C immediately upon arrival. To obtain a homogeneous solutionfree of aggregates, a variant of Zagorski’s protocol48 was followed, asdiscussed in our previous publication.36 Briefly, 0.5 mg of the Aβ40 wasfirst dissolved with TFA, followed by TFA evaporation with N2. Thisprocess was then repeated two more times. To remove TFAthoroughly, HFIP was added and then evaporated with N2. ThisHFIP treatment was also repeated another two times. The dry aliquotwas completely resuspended in a solution containing 0.5 M DMSOand 0.1 M PBS (pH 7.4) to a final volume of 2.88 mL, in order toreach a final Aβ40 concentration of 40 μM. Before use, all solutionswere filtered through 0.20 μm pore size filters. For initiating the Aβ40fibrillation process, samples of the Aβ40 solutions (40 μM) wereincubated in 1.5 mL eppendorf tubes in a bath heater at 37 °C withgentle shaking. To monitor the appearance and growth of fibrils,aliquots from the tubes were taken at different times and added to ablack 96-well plate, and the ThT fluorescence (20 μM ThT added toeach well) was measured at 485 nm with excitation at 435 nm in aplate reader.49

A similar process to that described above was also performed in thepresence of different concentrations of the polyA-FF-ME nanoparticlesor polyA-AA-ME nanoparticles. Briefly, different volumes, 2.6−26 μL,(10−100% (w/wAβ40)) of an aqueous dispersion of nanoparticles (3.4mg/mL) were added to 0.5 mL of 40 μM Aβ40 PBS solution, asdescribed above. The formation of the fibrils was then initiated by

Biomacromolecules Article

dx.doi.org/10.1021/bm3011177 | Biomacromolecules 2012, 13, 2662−26702664

quickly raising the temperature of the aqueous mixture from roomtemperature to 37 °C at different time intervals.Characterization. 1H and 13C NMR spectra were obtained with a

Bruker AC 200 or 300 MHz spectrometer. Chloroform-d chemicalshifts are expressed in ppm downfield from tetramethylsilane used asinternal standard. High-resolution mass spectra were obtained on anAutoFlex III Tof/Tof (Bruker, Germany) in MALDI mode. Elementalanalysis was performed using an elemental analyzer, modelFlashEA1112 (Thermo Electron Corporation). Fourier transforminfrared (FTIR) analysis was performed with a ALPHA FTIRspectrophotometer equipped with attenuated total reflection (ATR)model A220/D-01, Bruker. Low-resolution transmission electronmicroscopy (TEM) pictures were obtained with a FEI TECNAI C2BIOTWIN electron microscope with 120 kV accelerating voltage.Samples for TEM were prepared by placing a drop of diluted sampleon a 400-mesh carbon-coated copper grid. High-resolution scanningelectron microscopy (HRSEM) images were obtained with a FEIMagellan 400 L electron microscope. Samples for HRSEM wereprepared by spreading a drop of diluted sample on a glass surface andthen drying it at room temperature. The dried sample was coated withgold in a vacuum for 4 min before viewing under HRSEM. Theaverage size and size distribution of the dry nanoparticles weredetermined by measuring the diameter of more than 100 particles withthe image analysis software AnalySIS Auto (Soft Imaging SystemGmbH, Germany). Hydrodynamic diameter and size distribution ofthe nanoparticles dispersed in an aqueous phase were measured usinga particle analyzer, model NANOPHOX (Sympatec GmbH,Germany). The thermal behavior of the nanoparticles was measuredby ThermoGravimetric Analysis (TGA) and differential scanningcalorimetric (DSC), model STAR-1 System, Mettler Toledo. Theanalysis was performed with approximately 10 mg of dried samples in adynamic nitrogen atmosphere (20 mL/min) at heating rate of 10 °C/min. Fluorescence intensity and absorbance at room temperature weremeasured using a multiplate reader Synergy 4, using Gen 5 software.Electrokinetic properties (ζ-potential) as a function of pH weredetermined by Zetasizer (Zetasizer 2000, Malvern Instruments, UK).Circular dichroism (CD) spectra were measured on a Jasco J710spectropolarimeter at 22 °C over the wavelength range 180−260 nmwith 0.2 nm resolution. All measurements in solution were recorded ina 0.1 cm path length cell. Each spectrum is an average of fourmeasurements. The spectra were corrected by subtracting the buffersolution or nanoparticles’ dispersion baseline. All samples of Aβ40solutions were diluted with phosphate buffer (PB) solution to a finalconcentration of 10 μM. Aβ40 solutions sampled at long incubationtimes (120 and 336 h) when the fibrils were already formed werecentrifuged prior to scanning.

■ RESULTS AND DISCUSSION

Designed Synthesis of the polyA-FF-ME Nanopar-ticles. Recent progress in elucidating the structural propertiesof Aβ fibrils has enabled the design of inhibitors of Aβ fibrilformation.14−21 In the present study, we designed new uniformbiocompatible amino-acid-based polymer nanoparticles forinhibition of Aβ40 fibrillation process. The FF residues, thehydrophobic core of residues 19−20 of the Aβ protein, arewell-known as crucial sequences for the formation of β-sheetstructures that trigger the fibrillation process.14−18 It has beensuggested that the Aβ assembly is partially driven byhydrophobic interactions through recognition between the FFpairs.14−18 Short peptides homologous to the hydrophobic coreof Aβ, for example, LPFFD, were designed to bind specificallyto full-length Aβ and used as inhibitors of its fibrillation processvia the FF recognition motif.19−21 The rational of our design isto engineer polymer nanoparticles containing extremely highconcentrations of the FF motif. This motif in the nanoparticlestherefore should act as a recognition motif within Aβ andinterferes in its fibril formation. Figure 1 describes the syntheticroute of the monomer A-FF-ME, followed by its polymer-ization and precipitation to form the polyA-FF-ME nano-particles.

Characterization of the polyA-FF-ME Nanoparticles.Figure 2 presents the dry (A) and hydrodynamic (B) diametersand size distributions of the polyA-FF-ME nanoparticles asmeasured by HRSEM images and light scattering, respectively.The dry diameter is 57 ± 6 nm, while the hydrodynamicdiameter is 200 ± 27 nm. The difference in the diametermeasured by HRSEM and light scattering is due to the fact thatthe second method also takes into account the surface-adsorbedsolvent (water) molecules.FTIR spectra of the monomer A-FF-ME (A) and the polyA-

FF-ME nanoparticles (B) are shown in Figure 3. The IRspectrum of the monomer A-FF-ME (Figure 3A) indicatesstretching vibrations at 1757 and 1206 cm−1 corresponding tothe methyl ester (ME) groups, and stretching vibrations at1672, 1647, and 1552 cm−1 corresponding to the amide groups.The absorbance peak at 1624 cm−1 corresponds to thestretching vibration of CC bonds. The absorbance peaksfrom 2980 to 3070 and from 1456 to 1496 cm−1 arecharacteristic of the C−H stretching (aromatic and aliphatic)and CH2 groups, and peaks at 700 and 3269 cm−1 correspondto the stretching of C−C and N−H bonds, respectively. The IR

Figure 2. HRSEM image (A) and size histogram (B) of the polyA-FF-ME nanoparticles.

Biomacromolecules Article

dx.doi.org/10.1021/bm3011177 | Biomacromolecules 2012, 13, 2662−26702665

spectrum of the polyA-FF-ME nanoparticles (Figure 3B) issimilar to that of the monomer except for the typical stretchingvibration of CC bonds that is not present in the polyA-FF-ME nanoparticles. The chemical structures of the monomer A-FF-ME and its corresponding polymer were also confirmed by1H NMR analysis (in CDCl3), as shown in Figure 4. The 1HNMR spectrum of the monomer (Figure 4A) reveals the typicalpeaks of the acrylic protons at 6.3−6.1 ppm and 5.70−5.65ppm, the aromatic protons at 7.2 and 7.0 ppm, the amideprotons at 6.1 ppm, the proton attached to chiral carbon atomat 4.75 ppm, the methylene protons at 3.1 ppm, and the MEprotons at 3.68 ppm. The 1H NMR spectrum of the polyA-FF-ME nanoparticles (Figure 4B) reveals the complete absence ofthe acrylic protons at 5.5−6.5 ppm and the presence of broadpeaks of the aromatic and amide protons at 7.3−6.8 ppm, theproton attached to chiral carbon atom at 4.8−4.3 ppm, the MEprotons at 3.7−3.3 ppm, the methylene protons at 3.2−2.7ppm, and the aliphatic chain protons of the polymer backboneat 2.4−1.8 ppm.The thermal behavior of the polyA-FF-ME nanoparticles is

shown in Figure 5. The TGA curve exhibits a steep slopebetween 280 and 450 °C, indicating a 68% weight loss due tothe polymer decomposition, leaving residual carbon. The Tm ofthe polyA-FF-ME nanoparticles measured by DSC is ca. 165 °C(data not shown).To learn about the stability of the polyA-FF-ME nano-

particles in aqueous medium, ζ-potential measurements can beemployed as an indirect indicator of their surface charge. Figure6 illustrates the ζ-potential curve of the polyA-FF-MEnanoparticles as a function of pH. The titration curve of thepolyA-FF-ME nanoparticles, obtained by adding HCl to anaqueous dispersion of the polyA-FF-ME nanoparticles, showsan increase of the ζ-potential from −13 to 2 mV while changingthe pH from 11 to 2. This change in the surface potential canbe attributed to the surface ME groups of the polyA-FF-MEnanoparticles. In both basic and acidic environments, the MEgroups undergo partial hydrolysis. In a basic environment, thesurface charge potential is negative due to hydrolysis of MEgroups to CO−O−, whereas in an acidic environment, thesurface charge potential is slightly positive due to theprotonation of the carboxyl groups.50 The measured isoelectricpoint of the polyA-FF-ME nanoparticles is at pH 4.2 as shownin Figure 6.

In Vitro Cytotoxicity Study of polyA-FF-ME Nano-particles. The cytotoxic effect of the polyA-FF-ME nano-particles on different human cell lines, e.g., PC-12 and SH-SY5Y, was examined using XTT cell viability assay, as shown in

Figure 3. FTIR spectra of the monomer A-FF-ME (A) and the polyA-FF-ME nanoparticles (B).

Figure 4. 1H NMR spectra (in CDCl3) of the monomer A-FF-ME (A)and the polyA-FF-ME nanoparticles (B).

Figure 5. TGA thermogram of the polyA-FF-ME nanoparticles.

Biomacromolecules Article

dx.doi.org/10.1021/bm3011177 | Biomacromolecules 2012, 13, 2662−26702666

Figure 7. This figure shows the cell viability % in the presenceof different concentrations of the nanoparticles after incubation

for 24 or 48 h. After 24 h of exposure of the PC-12 and SH-SY5Y cells to the nanoparticles, no significant toxicity wasobserved for nanoparticle concentration of 0.2 mg/mL(concentration used for our fibrillation studies) or even tentimes higher (ca. 2.0 mg/mL). Minor toxicity was observedafter 48 h of exposure to nanoparticle concentration of 0.2 mg/mL for PC-12 only, and to the higher concentration of 2.0 mg/mL for both cell lines, with approximately 80% cell viability.Effect of polyA-FF-ME Nanoparticles on the Kinetics

of Aβ40 Fibrillation Process. Kinetics of the Aβ40 fibrilformation in PBS at 37 °C, monitored by the temporaldevelopment of ThT binding,49 in the absence (A) andpresence (B−D) of increasing concentrations of the polyA-FF-ME nanoparticles are shown in Figure 8. It should be noted

that the stability against agglomeration of these nanoparticleswas found to be dependent at the ionic strength of the PBS,e.g., in 0.1 M PBS, they are stable, while in 0.5 M PBS, theyaggregate. Due to this instability issue, the kinetics wasaccomplished in the presence of 0.1 M PBS as a continuousphase. In the absence of the nanoparticles (Figure 8A), themain growth of the Aβ40 fibrils, corresponding to the end of thelag time, occurred approximately 60 h after initiating thefibrillation process, and was completed after approximately 120h. On the other hand, the kinetics of the Aβ40 fibril growth inthe presence of increasing concentrations of the polyA-FF-MEnanoparticles was significantly delayed (Figure 8B−D). Forexample, Figure 8 exhibits that the presence of 10 (B), 20 (C),and 100 (D) % (w/wAβ40) of the polyA-FF-ME nanoparticlesinhibits the initiation of the fibril formation to be only after 85,99, and 233 h, respectively. These observations indicate that thepolyA-FF-ME nanoparticles substantially increase the fibrilla-tion lag time and thereby inhibit the kinetics of the nucleationand oligomer formation prior to the formation of the Aβ40fibrils. This inhibitory effect might be explained by the presenceof the pairs of FF residues of these nanoparticles, which areknown for their high affinity to the corresponding residues ofAβ40 prefibril aggregates, e.g., monomers and oligomers,through hydrophobic interactions. This high binding affinitydisturbs the monomer-critical nuclei equilibrium by trappingthe monomers and/or blocking the growing oligomer ends onthe surface of the nanoparticles, thereby decreasing theirsolution concentration and interfering with their elongation toform fibrils. However, the ability to distinguish whether thesenanoparticles adsorbed the monomers, or oligomers, or both,remains to be investigated.Our results imply that the nucleation process of Aβ40 is

strongly disturbed by the presence of the polyA-FF-MEnanoparticles assuming a tight interaction between thenanoparticles and Aβ40 prefibril intermediates. A similar effectwas reported for the fibrillation process of Aβ and islet amyloidpolypeptide (IAPP) proteins in the presence of copolymeric N-isopropylacrylamide/N-tert-butylacrylamide nanoparticles withdifferent ratios between these monomers having variablehydrophobicity.25,26,31 In these studies, the retardation of the

Figure 6. ζ-potential of the polyA-FF-ME nanoparticles as a functionof pH. The reported values are the average of five measurements. Theerror bars indicate the standard deviation.

Figure 7. Viability of PC-12 and SH-SY5Y cells exposed to increasingconcentrations of polyA-FF-ME nanoparticles after 24 or 48 h at 37°C, as determined by the XTT assay described in the Materials andMethods. The reported values are the average of measurementsperformed on at least eight wells for each tested sample of twoindependent experiments, and expressed as a percentage of thenegative control (cultured cells in medium without nanoparticles).The error bars indicate the standard error of the mean.

Figure 8. Kinetics of the Aβ40 fibrils formation in PBS at 37 °C in theabsence (A) and presence of 10 (B), 20 (C), and 100 (D) % (w/wAβ40) of the polyA-FF-ME nanoparticles. The kinetics of the Aβ40fibril formation were measured using ThT binding assay, as describedin the Materials and Methods. The reported values are the averages offive measurements of each triplicate tested sample. The error barsindicate the standard deviation.

Biomacromolecules Article

dx.doi.org/10.1021/bm3011177 | Biomacromolecules 2012, 13, 2662−26702667

fibrillation process was also explained by the depletion of freemonomers and oligomers (prefibrils) from solution due to theiradsorption on the nanoparticle surface. This observation led toan increase in the lag time of the fibrillation process but did notprevent the fibril formation. It should be noted that, untilrecently, it was assumed that Aβ had to be assembled intoamyloid fibrils to exert its cytotoxic effect.14,16,21 However,recent studies have revealed that the soluble oligomers aresignificantly more toxic to neuronal cells than the monomersand the formed fibrils.17,20,21,25,35 Therefore, there is greatinterest in developing nanoparticles that are able to disturb themonomer-critical nuclei equilibrium by trapping activemonomers and prefibril intermediates on the nanoparticlesurface, thereby delaying further growth, thereby reducing theirtoxicity in solution.31,33−35

Our assumption that the observed inhibitory effect in thepresence of the polyA-FF-ME nanoparticles is due to specifichydrophobic interactions between the nanoparticles and the Aβprefibril aggregates, via the FF recognition motif, was examinedby replacing the polyA-FF-ME nanoparticles by polyA-AA-MEnanoparticles. The chemical structures of the monomer A-AA-ME and its corresponding polymeric nanoparticles wereconfirmed by 1H NMR analysis, as shown in the SupportingInformation, Figure S1. The dry and hydrodynamic diametersof the polyA-AA-ME nanoparticles are 52 ± 8 and 214 ± 23nm, respectively, which are similar to those obtained for thepolyA-FF-ME nanoparticles, as shown in the SupportingInformation, Figure S2. Figure 9 presents typical sigmoidal

curves showing the kinetics of the Aβ40 fibrils formation in PBSat 37 °C in the absence (A) and the presence of 10 (B), 20 (C),and 100 (D) % (w/wAβ40) of the polyA-AA-ME nanoparticles.It is rather interesting to note that the observed behavior is theopposite of the behavior obtained in the presence of similarconcentrations of the polyA-FF-ME nanoparticles. The kineticsof the Aβ40 fibrils growth in the presence of 10, 20, and 100%(w/wAβ40) were accelerated, and initiated after 53, 38, and 18 h,respectively (see Figure 9). This promoting effect probablyresults from the absence of the FF interfering moieties in these

nanoparticles. Therefore, the increased local concentration ofthe Aβ40 prefibrils adsorbed on the polyA-AA-ME nanoparticlesurface promoted their growth into fibrils. It is important tonote that the promoting effect achieved in the presence of thepolyA-AA-ME nanoparticles may also be considered a desiredprocess. The polyA-AA-ME nanoparticles decrease thefibrillation lag time, thereby shorting the half-life time of theAβ40 prefibrils, thereby decreasing their toxicity in solution.The secondary conformational change of Aβ40 from α-helix

and/or random coil to β-sheet is probably the key step in theprocess to form oligomeric aggregates.11 Hence, CD measur-ments were performed to determine the secondary structurechanges during the Aβ40 fibril formation in the absence and thepresence of the polyA-FF-ME nanoparticles. Figure 10 presents

CD spectra of the Aβ40 in the absence (A) and the presence(B) of 100% (w/wAβ40) of the polyA-FF-ME nanoparticles atdifferent time periods during the fibrillation process. In theabsence of the nanoparticles (Figure 10A) before the initiationof the fibrillation process, the CD spectrum of the freshlyprepared Aβ40 aqueous solution at pH 7.4 shows one strongminimum at 198 nm indicative of a typical spectrum of a nativeprotein dominated by random coil conformation, as describedin the literature.36,51,52 By heating the Aβ40 solution to 37 °Cfor 120 h, the CD spectrum of the Aβ40 displays a loss ofintensity of the 198 nm peak accompanied by a maximumellipticity at 200 nm and a broad minimum ellipticity at 217nm, which is typical of the presence of extensive β-sheetstructures. Almost the same CD spectra of Aβ40 were observedin the presence of 100% (w/wAβ40) of the polyA-FF-MEnanoparticles. It is rather interesting to note that the CD

Figure 9. Kinetics of the Aβ40 fibril formation in PBS at 37 °C in theabsence (A) and presence of 10 (B), 20 (C), and 100 (D) % (w/wAβ40) of the polyA-AA-ME nanoparticles. The kinetics of the Aβ40fibril formation was measured using ThT binding assay, as described inthe Materials and Methods. The reported values are the average of fivemeasurements of each triplicate tested sample. The error bars indicatethe standard deviation.

Figure 10. CD spectra of the Aβ40 in the absence (A) and the presence(B) of 100% (w/wAβ40) of the polyA-FF-ME nanoparticles dispersedin PB solution at different time periods of the fibrillation process.

Biomacromolecules Article

dx.doi.org/10.1021/bm3011177 | Biomacromolecules 2012, 13, 2662−26702668

spectrum of the freshly prepared Aβ40 aqueous solution in thepresence of 100% (w/wAβ40) of the polyA-FF-ME nanoparticles(Figure 10B) exhibits one strong minimum at 198 nm and aweak maximum at 224 nm, which are stable for more than 233h of heating. Only beyond that time does the transformation toβ-sheet structure start to occur.To further confirm the achieved inhibitory effect on the Aβ40

fibril formation in the presence of the polyA-FF-ME nano-particles, TEM images were obtained to detect morphologicalchanges in Aβ40 solution, as shown in Figure 11. In the case ofcontrol Aβ40 solution, in the absence of nanoparticles (Figure11A), massive networks of entangled fibrils with microsizedlengths were observed after 120 h of the fibrillation process. Onthe other hand, when Aβ40 solution was incubated in thepresence of 100% (w/wAβ40) of the polyA-FF-ME nano-particles, only free nanoparticles in the background wereobserved even after 200 h of the fibrillation process (Figure11B), indicating that the formation of the fibrils had not yetoccurred. After 233 h (Figure 11C), absorption of amorphousAβ40 aggregates on the surface of the polyA-FF-ME nano-particles was initiated. The internal dark and bright areasaround the nanoparticles (as indicated by white arrows)illustrate the adsorbed dense and less dense multiple layers ofthe amorphous prefibril aggregates. It is further interesting tonote that after 330 h (Figure 11D) the formed fibrillarnetworks were much less massive than those observed in the

absence of the nanoparticles. These observations indicate thetrapping of amorphous Aβ40 aggregates by their adsorption onthe surface of the polyA-FF-ME nanoparticles through specificinteractions, leading to their partial depletion from solutionwhich interferes with their elongation into massive fibrils.

■ CONCLUSIONS

The present manuscript describes the design, synthesis, andcharacterization of novel polyA-FF-ME nanoparticles, bypolymerization of the synthesized monomer A-FF-ME,followed by precipitation of the soluble polymer in aqueoussolution. These studies demonstrate the significant inhibition ofthe Aβ40 fibrillation process in the presence of thesenanoparticles. This inhibition is probably due to theintermolecular attractive hydrophobic interactions betweenthe pairs of FF residues of the nanoparticles with thecorresponding residues of the Aβ40 prefibrillar aggregates,which disrupt the self-assembly of Aβ40 into fibrils. Thisobservation was confirmed by the promoting effect achieved inthe presence of the polyA-AA-ME nanoparticles.The present studies were performed using pure Aβ40 without

competition from other proteins for binding to the nanoparticlesurface. These conditions are quite unlike those in any realisticclinical situation. Still, our studies intend to provide a newmechanistic insight into amyloidogenic peptide interaction with

Figure 11. TEM images of the Aβ40 in the absence of the polyA-FF-ME nanoparticles, 120 h after initiation of the fibrillation process (A), and in thepresence of 100% (w/wAβ40) of the polyA-FF-ME nanoparticles dispersed in PBS, 200 h (B), 233 h (C), and 330 h (D) after initiation of thefibrillation process.

Biomacromolecules Article

dx.doi.org/10.1021/bm3011177 | Biomacromolecules 2012, 13, 2662−26702669

nanoparticles. Further work would be necessary to assess thesuitable mechanistic interference of the polyA-FF-ME nano-particles and the polyA-AA-ME nanoparticles with the Aβ40self-assembly. In future work, we wish to extend this study toother amyloidogenic proteins under competitive conditions.

■ ASSOCIATED CONTENT*S Supporting Information1H NMR spectra of the monomer A-AA-ME and the polyA-AA-ME nanoparticles. TEM image and size histogram of thepolyA-AA-ME nanoparticles. This material is available free ofcharge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: Shlomo.margel@mail.biu.ac.il. Tel: 972-3-5318861.Fax: 972-3-6355208.Present Address‡Currently on sabbatical from Israel Institute for BiologicalResearch, Ness-Ziona 74100, Israel.Author Contributions†These authors contributed equally to this study.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThese studies were partially supported by a Minerva Grant(Microscale & Nanoscale Particles and Films).

■ REFERENCES(1) Chiti, F.; Dobson, C. M. Annu. Rev. Biochem. 2006, 75, 333−366.(2) Sipe, J. D.; Benson, M. D.; Buxbaum, J. N.; Ikeda, S.; Merlini, G.;Saraiva, M. J.; Westermark, P. Amyloid 2010, 17, 101−104.(3) Selkoe, D. J. Nature 2003, 426, 900−904.(4) Sunde, M.; Blake, C. C. Q. Rev. Biophys. 1998, 31, 1−39.(5) Ohnishi, S.; Takano, K. Cell. Mol. Life Sci. 2004, 61, 511−524.(6) Kagan, B. L.; Thundimadathil, J. Adv. Exp. Med. Biol. 2010, 677,150−167.(7) Price, D. L.; Sisodia, S. S.; Gandy, S. E. Curr. Opin. Neurol. 1995,8, 268−274.(8) Haass, C.; Schlossmacher, M. G.; Hung, A. Y.; Vigo-Pelfrey, C.;Mellon, A.; Ostaszewski, B. L.; Lieberburg, I.; Koo, E. H.; Schenk, D.;Teplow, D. B.; Selkoe, D. J. Nature 1992, 359, 322−325.(9) Seubert, P.; Vigo-Pelfrey, C.; Esch, F.; Lee, M.; Dovey, H.; Davis,D.; Sinha, S.; Schiossmacher, M.; Whaley, J.; Swindlehurst, C.;McCormack, R.; Wolfert, R.; Selkoe, D. J.; Lieberburg, I.; Schenk, D.Nature 1992, 359, 325−327.(10) Selkoe, D. J. Physiol. Rev. 2001, 81, 741−766.(11) Kumar, S.; Udgaonkar, J. B. Curr. Sci. 2010, 98, 639−656.(12) Karran, E.; Mercken, M.; Strooper, B. D. Nat. Rev. DrugDiscovery 2011, 10, 698−712.(13) Hard, T.; Lendel, C. J. Mol. Biol. 2012, 421, 441−465.(14) Hilbich, C.; Kisters-Woike, B.; Reed, J.; Masters, C. L.;Beyreuther, K. J. Mol. Biol. 1992, 228, 460−473.(15) Gazit, E. FEBS J. 2005, 272, 5971−5978.(16) Soto, C.; Sigurdsson, E. M.; Morelli, L.; Kumar, R. A.; Castano,E. M.; Frangione, B. Nat. Med. 1998, 4, 822−826.(17) Takahashi, T.; Mihara, H. Acc. Chem. Res. 2008, 41, 1309−1318.(18) Marshall, K. E.; Morris, K. L.; Charlton, D.; O’Reilly, N.; Lewis,L.; Walden, H.; Serpell, L. C. Biochemistry 2011, 50, 2061−2071.(19) Viet, M. H.; Ngo, S. T.; Lam, N. S.; Li, M. S. J. Phys. Chem. B2011, 115, 7433−7446.(20) Rocha, S.; Cardoso, I.; Borner, H.; Pereira, M. C.; Saraiva, M. J.;Coelho, M. Biochem. Biophys. Res. Commun. 2009, 380, 397−401.

(21) Lowe, T. L.; Strzelec, A.; Kiessling, L. L.; Murphy, R. M.Biochemistry 2001, 40, 7882−7889.(22) Estrada, L. D.; Soto, C. Curr. Top. Med. Chem. 2007, 7, 115−126.(23) Spuch, C.; Saida, O.; Navarro, C. Recent Pat. Drug DeliveryFormulation 2012, 6, 2−18.(24) Sahni, J. K.; Doggui, S.; Ali, J.; Baboota, S.; Dao, L; Ramassamy,C. J. Controlled Release 2011, 152, 208−231.(25) Brambilla, D.; Le Droumaguet, B.; Nicolas, J.; Hashemi, S. H.;Wu, L. P.; Moghimi, S. M.; Couvreur, P.; Andrieux, K. Nanomedicine2011, 7, 521−540.(26) Cabaleiro-Lago, C.; Szczepankiewicz, O.; Linse, S. Langmuir2012, 28, 1852−1857.(27) Silva, G. A. Nat. Rev. Neurosci. 2006, 7, 65−74.(28) Kabanov, A. V.; Gendelman, H. E. Prog. Polym. Sci. 2007, 32,1054−1082.(29) Linse, S.; Cabaleiro-Lago, C.; Xue, W. F.; Lynch, I.; Lindman, S.;Thulin, E.; Radford, S. E.; Dawson, K. A. Proc. Natl. Acad. Sci. U. S. A.2007, 104, 8691−8696.(30) Wu, W. H.; Sun, X.; Yu, Y. P.; Hu, J.; Zhao, L.; Liu, Q.; Zhao, Y.F.; Li, Y. M. Biochem. Biophys. Res. Commun. 2008, 373, 315−318.(31) Cabaleiro-Lago, C.; Quinlan-Pluck, F.; Lynch, I.; Lindman, S.;Minogue, A. M.; Thulin, E.; Walsh, D. M.; Dawson, K. A.; Linse, S. J.Am. Chem. Soc. 2008, 130, 15437−15443.(32) Cabaleiro-Lago, C.; Quinlan-Pluck, F.; Lynch, I.; Dawson, K. A.;Linse, S. ACS Chem. Neurosci. 2010, 1, 279−287.(33) Yoo, S. I.; Yang, M.; Brender, J. R.; Subramanian, V.; Sun, K.;Joo, N. E.; Jeong, S. H.; Ramamoorthy, A.; Kotov, N. A. Angew. Chem.,Int. Ed. 2011, 50, 5110−5115.(34) Rocha, S.; Thunemann, A. F.; Pereira, M. C.; Coelho, M.;Mohwald, H.; Brezesinski, G. Biophys. Chem. 2008, 137, 35−42.(35) Saraiva, A. M.; Cardoso, I.; Saraiva, M. J.; Tauer, K.; Pereira, M.C.; Coelho, M. A.; Mohwald, H.; Brezesinski, G. Macromol. Biosci.2010, 10, 1152−1163.(36) Skaat, H; Shafir, G.; Margel, S. J. Nanopart. Res. 2011, 13, 3521−3534.(37) Dutta, P.; Dey, J. Int. J. Pharm. 2011, 421, 353−363.(38) Yang, H. M.; Lee, H. J.; Park, C. W.; Yoon, S. R.; Lim, S.; Jung,B. H.; Kim, J. D. Chem. Commun. 2011, 47, 5322−5324.(39) Akagi, T.; Shima, F.; Akashi, M. Biomaterials 2011, 32, 4959−4967.(40) Skey, J.; O’Reilly, R. K. J. Polym. Sci., Part A: Polym. Chem. 2008,46, 3690−3702.(41) O’Reilly, R. K. Polym. Int. 2010, 59, 568−573.(42) Bentolila, A.; Vlodavsky, I.; Ishai-Michaeli, R.; Kovalchuk, O.;Haloun, C.; Domb, A. J. J. Med. Chem. 2000, 43, 2591−2600.(43) Mori, H.; Matsuyama, M.; Sutoh, K.; Endo, T. Macromolecules2006, 39, 4351−4360.(44) Sanda, F.; Endo, T. Macromol. Chem. Phys. 1999, 200, 2651−2661.(45) Melamed, O.; Margel, S. Colloids Surf., A 2002, 208, 147−157.(46) Nudelman, A.; Bechor, Y.; Falb, E.; Fischer, B.; Wexler, B. A.;Nudelman, A. Synth. Commun. 1998, 28, 471−474.(47) Roehm, N. W.; Rodgers, G. H.; Hatfield, S. M.; Glasebrook, A.L. J. Immunol. Methods 1991, 142, 257−265.(48) Chen, S.; Wetzel, R. Protein Sci. 2001, 10, 887−891.(49) LeVine, H. Amyloid 1995, 2, 1−6.(50) Akagi, T.; Baba, M.; Akashi, M. Polymer 2007, 48, 6729−6747.(51) Terzi, E.; Holzemann, G.; Seelig, J. J. Mol. Biol. 1995, 252, 633−642.(52) Bao, Q.; Luo, Y.; Li, W.; Sun, X.; Zhu, C.; Li, P.; Huang, Z. X.;Tan, X. J. Biol. Inorg. Chem. 2011, 16, 809−816.

Biomacromolecules Article

dx.doi.org/10.1021/bm3011177 | Biomacromolecules 2012, 13, 2662−26702670