Research Article Cite This: ACS Appl. Mater. Interfaces 2019, 11, 23858−23869
www.acsami.org
Self-Assembled Au@Fe Core/Satellite Magnetic Nanoparticles for Versatile Biomolecule Functionalization Hongwei Chen,*,† Hongxiang Hu,† Chun Tao,† Ryan M. Clauson,† Ila Moncion,† Xin Luan,† Sangyeul Hwang,‡ Ashley Sough,‡ Kanokwan Sansanaphongpricha,† Jinhui Liao,† Hayley J. Paholak,† Nicholas O. Stevers,† Guoping Wang,‡ Bing Liu,‡ and Duxin Sun*,† †
Department of Pharmaceutical Sciences, College of Pharmacy, University of Michigan, Ann Arbor, Michigan 48109, United States IMRA America, Inc., 1044 Woodridge Avenue, Ann Arbor, Michigan 48105, United States
Downloaded via UNIV OF SOUTHERN INDIANA on July 17, 2019 at 11:12:16 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
‡
S Supporting Information *
ABSTRACT: Although the functionalization of magnetic nanoparticles (MNPs) with biomolecules has been widely explored for various biological applications, achieving efficient bioconjugations with a wide range of biomolecules through a single, universal, and versatile platform remains a challenge, which may significantly impact their applications’ outcomes. Here, we report a novel MNP platform composed of Au@Fe core/satellite nanoparticles (CSNPs) for versatile and efficient bioconjugations. The engineering of the CSNPs is facilely formed through the self-assembly of ultrasmall gold nanoparticles (AuNPs, 2−3 nm in diameter) around MNPs with a polysiloxane-containing polymer coating. The formation of the hybrid magnetic nanostructure is revealed by absorption spectroscopy, dynamic light scattering (DLS), transmission electron microscopy (TEM), element analysis using atomic absorption spectroscopy, and vibrating sample magnetometer. The versatility of biomolecule loading to the CSNP is revealed through the bioconjugation of a wide range of relevant biomolecules, including streptavidin, antibodies, peptides, and oligonucleotides. Characterizations including DLS, TEM, lateral flow strip assay, fluorescence assay, giant magnetoresistive nanosensor array, high-performance liquid chromatography, and absorption spectrum are performed to further confirm the efficiency of various bioconjugations to the CSNP. In conclusion, this study demonstrates that the CSNP is a novel MNP-based platform that offers versatile and efficient surface functionalization with various biomolecules. KEYWORDS: magnetic iron oxide nanoparticles, gold nanoparticles, bioconjugation, streptavidin, surface functionalization
1. INTRODUCTION Magnetic nanoparticles (MNPs) owing to their unique properties, such as superparamagnetism, larger surface-tovolume ratio, and easy separation by external magnetic fields, have gained tremendous attention in the past decade due to their potential role in various biomedical applications. Magnetic beads, for instance, have been widely utilized for the isolation of cells of interest and the pull-down purification of target proteins.1 In addition to a direct magnetic field response, MNPs have been used as biosensors for the detection of disease-related biomolecules in physiological liquids at very low detection limits by taking advantage of their nonlinear magnetization in a magnetic field with combinatorial frequencies.2−4 As a biodegradable and biocompatible nanoparticle, MNPs have also been widely explored for various clinical applications such as MRI contrast agents,5 inducers of hyperthermia under an alternative magnetic field,6 NIR light irradiation,7 and imaging-guidable drug carriers.8 MNPs have been traditionally synthesized by either coprecipitation using ferrous (Fe2+) and ferric (Fe3+) © 2019 American Chemical Society
ions by a base in an aqueous solution, or thermal decomposition using an iron precursor such as FeO(OH) in an organic solution.9 Following organic synthesis, MNPs must be further coated with an additional hydrophilic coating, such as heterobifunctional poly(ethylene glycol) (PEG) molecules,10,11 amphiphilic polymers,12,13 amphiphilic lipids,14,15 or a silica shell16,17 to achieve aqueous stabilization. Regardless of the synthetic strategy, the ultimate challenge to their successful translation in biologically relevant applications lies in the mechanism by which they facilitate biomolecule surface functionalization.18,19 Despite the fact that the functionalization of MNPs with various biomolecules has been widely explored previously, achieving efficient conjugations of a wide range of biomolecules through a single versatile platform still remains a considerable challenge to this day. What is more, the Received: March 28, 2019 Accepted: June 18, 2019 Published: June 18, 2019 23858
DOI: 10.1021/acsami.9b05544 ACS Appl. Mater. Interfaces 2019, 11, 23858−23869
Research Article
ACS Applied Materials & Interfaces
Orthopyridyldisulfide−PEG−succinimidyl ester (OPSS−PEG−NHS, 5 kD) was obtained from Creative PEGWorks (Winston-Salem, NC). Streptavidin was purchased from Jackson ImmunoResearch (West Grove, PA). Lipoic acid−PEG−NHS (LA−PEG−NHS, 3.4 kD) was purchased from NanoCS. E7 peptides with a sequence of CSKKKQAEPDRAHYNIVTFCCKCD from Elim Biopharm (Hayward, CA) were used as received. Single-stranded DNA (ssDNA) oligonucleotides were purchased from Bio Basic (Toronto, Canada). ssDNA1: 5′-SH-AAAAAAAAAATGCATGGTTGATAGG-3′. ssDNA2: 5′-SH-AAAAAAAAAACCTATCAACCATGCA-3′. Milli-Q water (18.2 MΩ cm) was prepared using a Milli-Q Academic water purification system (Billerica, MA). UV−visible spectra were recorded in a BioTek microplate reader (Synergy 2). The nanoparticle hydrodynamic size and ζ-potential were measured using a dynamic light scattering (DLS) instrument (Malvern Zeta Sizer Nano S-90) equipped with a 22 mW He−Ne laser operating at λ = 632.8 nm. The magnetic nanocrystals and core−satellite structure were viewed by transmission electron microscopy (TEM) (JEOLJEM-1400Plus, 80 kV). 2.2. Synthesis of IONPs Coated with Polysiloxane-Containing Diblock Copolymer. Spherical IONPs (15 nm in diameter) were synthesized in an organic solvent by thermal decomposition, as reported previously.7 Cubic IONPs (25 nm in edge length) were synthesized following previously reported method with slight modification.43 Diblock copolymer (PEO-b-PγMPS) was synthesized by the reversible addition of fragmentation chain transfer polymerization, as previously reported.13 The preparation of the polymercoated MNPs with either single or clustered core was previously reported.42,44 The IONP iron concentration was determined using ophenanthroline (ACS reagent, 99%) after digestion with hydrochloric acid (ACS reagent, 37%), as previously described.42 2.3. Synthesis of AuNPs. AuNPs were synthesized using sodium sulfide (Na2S) as the reducing reagent as reported before with a slight modification.45 Gold in the form of chloroauric acid (HAuCl4) was prepared to a concentration of 100 mM as a stock solution and was diluted to 2.0 mM before use. Na2S (50 mM) was prepared and aged in the dark for 40−48 h prior to use and was diluted to 1.0 mM before use. The volume ratio of Na2S to HAuCl4 was varied from 2.5/1.0 to 3.0/1.0. UV/vis spectra were recorded to monitor the reaction. Without specification, the reaction with a volume ratio of 3.0/1.0 was chosen to be used in the following steps. 2.4. Synthesis of Core/Satellite Nanoparticles (CSNPs). CSNPs were made by incubating AuNPs with polymer-coated IONPs at 4 °C. In a typical experiment, 2 mg Fe of IONPs (0.5 mL) was mixed with 6 mL of AuNP solution for spherical IONPs and 4 mL for cubic IONPs if without specification. The formed CSNPs were purified by a magnet to remove unbound AuNPs after overnight incubation. Upon removing the supernatant, the concentrated solution was collected, and its iron and gold concentrations were measured by atomic absorption spectroscopy after digestion using aqua regia followed by proper dilutions. 2.5. Magnetization Hysteresis Loop Measurement. Magnetic hysteresis loops were measured using a vibrating sample magnetometer (VSM) Lake Shore VSM 7400 (Lake Shore Cryotronics Inc., Westerville, OH). An aqueous solution (91.85 μL) with a Fe concentration of 0.2 mg/mL was loaded into a VSM holder. Experiments were carried out in a magnetic field of up to 1000 Oe at room temperature (295 K). 2.6. CSNP Streptavidin Conjugation. To prepare a streptavidin conjugation, streptavidin (1 mg in 100 μL of 10 mM K-phosphate buffer, pH 7.2) was first modified with OPSS−PEG−NHS (30 μL of 10 mg/mL in K-buffer, freshly made) with a protection from light and a gentle magnetic stirring for 2 h at room temperature. OPSS−PEG− streptavidin conjugates were separated from an unbound linker using a 10 k Nanosep spin filter followed by resuspension in 200 μL of Kbuffer. After the addition of 10 μL of 10 mM ethylenediaminetetraacetic acid (EDTA) and 5 μL of 100 mM dithiothreitol, the mixture was stirred for 30 min at room temperature followed by two rounds of purification using Nanosep and a final resuspension in 200 μL of Kbuffer. One milligram Fe of CSNP solution was then mixed with
bioconjugation mechanism and resulting physical and chemical biomaterial properties may significantly impact their applications’ outcomes.20 For instance, a common method of surface functionalization of MNPs is to conjugate various antibodies or peptides to MNPs through the 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) coupling reactions.21−24 The challenge associated with this method is that MNPs could easily form aggregates during the conjugation reaction due to the surface charge change caused by the different buffer conditions required for EDC activation and the subsequent coupling reaction.22 In addition to the limited conjugation efficiency from traditional strategies, the integrity of the proteinfunctionalized MNPs also remains a critical concern.25,26 For instance, although biomolecules can be conjugated onto MNPs with functionalized phospholipid coatings, these coatings may not be sufficiently stable to withstand such modifications due to the potential detachment of phospholipids that are only connected to the core of the MNP through weak hydrophobic interactions of intercalated hydrocarbon chains.15,25−27 As an alternative mechanism for MNP surface functionalization, bioconjugations to silica-coated MNPs have also been explored previously. However, the tedious sol−gel process and low concentration of the pretreated MNPs required to avoid aggregations may limit this type of MNPs for large-scale production.28 Therefore, the novel design of a magnetic platform with an efficient biomolecule conjugation is still highly desired.29,30 Gold nanoparticle (AuNP) due to its versatile capability for efficient biomolecule conjugation through Au−S interaction has been widely explored for various biomedical applications.31,32 Hybrid AuNPs with MNPs may provide a promising strategy to enhance MNP bioconjugation efficiency and versatility.33−36 However, most previous strategies to build Au@Fe hybrids were focused on forming a continuous Au shell around the MNP core to achieve a unique plasmatic absorption in the near-infrared range for photothermal therapeutic applications,37−40 which may compromise the core magnetic property. Although core/satellite nanostructures composed of small-sized satellite AuNPs around the MNP core were recently reported,41 the large size of MNP cores (∼300 nm) together with the coating layers around MNPs makes the reported system very difficult for various bioapplications that require individual suspension of the functional MNPs in aqueous solutions. MNPs (15−20 nm) coated with a polysiloxane-containing diblock copolymer for various bioapplications have been previously reported by our laboratory.7,13,42 In this work, we report a facile and novel strategy to decorate the polymer-coated MNPs with multiple ultrasmall AuNPs (2−3 nm) to form Au@Fe core/satellite nanoparticles (CSNPs) as a versatile platform for efficient conjugations with various biomolecules.
2. EXPERIMENTAL SECTION 2.1. Materials and Instruments. Iron oxide(III) (FeO(OH), hydrated, catalyst grade, 30−50 mesh), oleic acid (technical grade, 90%), 1-octadecene (technical grade, 90%), anhydrous tetrahydrofuran (THF, 99.8%), sodium sulfide, chloroauric acid, ammonium iron(II), sulfate hexahydrate (Fe(NH4)2(SO4)2·6H2O, ACS reagent, 99%), o-phenanthroline monohydrate (ACS reagent, 99%), hydroquinone (ACS reagent, 99%), nitric acid (ACS reagent, 70%), hydrochloric acid (ACS reagent, 37%), dithiothreitol, and antihuman IgG Ab were purchased from Sigma-Aldrich. FluoSpheres biotinlabeled microspheres were purchased from ThermoFisher Scientific. 23859
DOI: 10.1021/acsami.9b05544 ACS Appl. Mater. Interfaces 2019, 11, 23858−23869
Research Article
ACS Applied Materials & Interfaces modified streptavidin with 100 μL (0.5 mg) for spherical core and 50 μL (0.25 mg) for the cubic core. After incubation overnight at 4 °C, the streptavidin-conjugated CSNPs were purified with a magnet and stored at 4 °C. 2.7. Antibody Conjugation Efficacy Study. Antihuman IgG was chosen to demonstrate the CSNP antibody conjugation and its conjugation efficacy. Antihuman IgG (0.4 mg) was first conjugated to 0.176 mg LA-PEG3400−NHS by stirring in K-buffer, protected from light, for 2 h. PEG−antibody conjugates were separated from unbound PEG using a Nanosep filter. Next, fluorescein was conjugated to the antibody bound to PEG3400 by stirring 12.6 μg of NHS−fluorescein with the PEG−antihuman IgG solution in Kbuffer, protected from light, for 2 h. PEG−antibody−fluorescein conjugates were separated from unbound fluorescein using a Nanosep filter. To determine the number of antibodies conjugated per nanoparticle, a series of concentrations of PEG−antibody−fluorescein conjugates were incubated overnight with 25 μL of 1 mg/mL (25 μg) CSNPs. These concentrations correspond to molar ratios of 50, 30, 20, 5, 1, and 0 antibodies per nanoparticle. After overnight incubation, the CSNPs were purified using a magnet and the supernatant collected for fluorescent measurement. Fluorescent measurement of the supernatant was done using a Biotek Cy5 plate reader, and fluorescent intensity of the supernatant was compared to that of controls (PEG−antibody−fluorescein conjugates at the same concentrations in a buffer, but without incubation with nanoparticles). 2.8. Specific Binding Using Biotin-Labeled Fluorescent Beads. To know the successful streptavidin conjugation to CSNPs and the specific binding to biotin, we first used FluoSpheres biotinlabeled microspheres to confirm. Ten microliters of FluoSpheres was mixed with 0.2 mg Fe of either streptavidin−CSNPs or CSNPs as a control (spherical core) with a final solution volume of up to 1 mL with Milli-Q water and incubated at 4 °C for 2 h. The mixtures were applied in a magnet for the separation of fluorescent beads from CSNPs for 6 h. The supernatants were collected for fluorescent analysis using a Biotek Cy5 plate reader. 2.9. Lateral Flow Strip Test. Biotinylated bovine serum albumin (BSA) was printed at a concentration of either 0.1 mg/mL (first line/ test line) or 0.5 mg/mL (second line/control line). In a 96-well tray, 50 μL of 1× phosphate-buffered saline (PBS) + 0.1% Tween-20 (running buffer) was combined with 1 μL of streptavidin-conjugated CSNPs or control nanoparticles (1 mg/mL Fe) and mixed well. A strip was dropped into the well, and the particles were allowed to flow for 3 min. For Prussian blue staining enhanced strip test, nanoparticle solutions were further diluted with 10× and 100×. After a regular run, the strips were washed with running buffer twice with 3 min for each round. The trips were then dipped into a freshly made Prussian blue staining solution (10% potassium ferrocyanide and 10% hydrochloric acid) for 15 min. 2.10. Giant Magnetoresistive (GMR) Nanosensor Array. The GMR sensor surface was modified with biotinylated BSA, as reported before.2 The GMR sensor array was placed in the test station. Streptavidin-conjugated CSNPs or commercial magnetic nanoparticles from Miltenyi Biotec., referred to as “MACS”, which were used as a control, were added to the reaction well with the same final concentrations. The GMR sensor array was monitored in real time as the streptavidin−CSNPs bound to the sensor surface through streptavidin−biotin interaction and the binding curves were then plotted. 2.11. CD4+ T-Cell Isolation with Streptavidin-Conjugated ̈ BALB/c mice suspended CSNPs. Splenocytes (107 cells) from naive in Hanks’ balanced salt solution and 2% fetal bovine serum (FBS) were incubated with 10 μL of the biotin−antibody cocktail (Biotin anti-CD8a, CD11b, CD11c, CD19, CD24, CD45R/B220, CD49b, CD105, I-A/I-E, TER-119/erythroid, TCR-γδ from Biolegend mouse CD4 T-cell negative selection kit) for 15 min followed by incubation with 10 μL of streptavidin-conjugated CSNPs (clustered MNP core) (1.0 mg Fe/mL) for 15 min. Cell suspension was then applied with STEMCELL magnet for 5 min. The liquid was collected for fluorescent dye-labeled CD3 and CD4 antibody staining. Untreated splenocytes were also stained with CD3 and CD4 antibodies for
control. Flow cytometry data were acquired on an FACSCanto II (BD Biosciences) and analyzed with flow software Summit. 2.12. Protein Pull-Down with Streptavidin-Conjugated CSNPs. Streptavidin-conjugated CSNPs (clustered MNP core) were incubated 30 min at room temperature with mIgG (30 μg) or biomIgG (30 μg) in 100 μL of PBS containing 1.0 mg of bovine serum albumin and 30 μg of actin and myosin. After washing with PBS, the pull-downs were loaded to 4−20% sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE). The gel was stained with a silver staining kit. 2.13. Peptide Conjugation. Different amount of E7 peptide (10 mg/mL) was added into the solution of CSNPs (50 μL, 2 mg/mL), and the molar ratio of E7 to CSNPs was fixed at 0, 100, 500, 1 K, 3 K, 5 K, 8 K, or 10 K. The mixture was then reacted at room temperature for 2 h and diluted with water to 200 μL before centrifugation (14 000 rpm, 20 min). The supernatant was extracted, and the unbound peptide concentration was then measured by high-performance liquid chromatography (HPLC). The mobile phase for HPLC was combined with acetonitrile (0.1% trifluoroacetyl (TFA)) and water (0.1% TFA), with acetonitrile concentration changes from 0 to 10% among 0−5 min and from 10 to 35% among 5−30 min. Absorption wavelength was set at 220 nm. The peptide-conjugated CSNP pellet after centrifugation was resuspended with Milli-Q water. 2.14. Serum Stability of Peptide-Conjugated CSNPs. AuNP surface retention kinetics on CSNPs post different incubation time points in both PBS and 50% FBS/PBS at 37 °C was evaluated by measuring the CSNPs’ Au/Fe mass ratio via inductively coupled plasma mass spectrometry. The Z-average hydrodynamic size of the CSNPs in both PBS and 50% FBS/PBS at 37 °C post different incubation times was measured by DLS. 2.15. CSNP Oligonucleotide Conjugation. ssDNA was conjugated to either CSNPs or AuNPs through the Au−S binding. The modification was carried out based on a reported procedure.46 The nanoparticles were incubated with ssDNA (2 nmol) for 1 min. Then, citrate buffer was added to a final concentration of 10 mM with a pH of 3.0 for additional 5 min. CSNP-ssDNA1 was purified by a magnet. AuNP-ssDNA2 was purified through centrifugation at 20 000 rpm for 50 min. Nanoparticles were collected and washed with water 3 times. The CSNP-ssDNA1 and AuNP-ssDNA2 were then incubated in 0.4 M NaCl, 1× PBS, and 0.01% SDS. After incubating overnight, the nanoparticles were purified by a magnet. The DNA-hybridized CSNPs were heated by a dry bath incubator at 55 °C for 10 min. Electrophoresis was applied to separate the released AuNPs. Agarose gels (1%, w/v) were used, and tris−acetate−EDTA (1×, TAE) was used as a running buffer. Nanoparticles (7 μL) were mixed with glycerol (3 μL) and added into the gels. Electrophoresis was performed at 60 V for 20 min. The reddish color of AuNPs allows band identification. The samples after heating were mixed with ssDNA1 to block the AuNP-ssDNA2. To observe the UV absorbance of the released AuNPs, the samples were heated with the same method, and the supernatant was collected after applying the magnet to isolate MNPs. 2.16. Statistical Analysis. Data are presented as mean ± standard deviation. A direct comparison between two groups was conducted by Student’s nonpaired t test. A p value of
