Preparation and Characterization of Chemically Functionalized Silica

Dec 19, 2008 - Kwanyong Seo , Sunghun Lee , Younghun Jo , Myung-Hwa Jung , Jinhee Kim , David G. Churchill and Bongsoo Kim. The Journal of Physical ...
0 downloads 0 Views 435KB Size
Download PDF
536

J. Phys. Chem. B 2009, 113, 536–543

Preparation and Characterization of Chemically Functionalized Silica-Coated Magnetic Nanoparticles as a DNA Separator Kiho Kang,† Jinsub Choi,† Joong Hee Nam,† Sang Cheon Lee,† Kyung Ja Kim,† Sang-Won Lee,‡ and Jeong Ho Chang*,† Korea Institute of Ceramic Engineering and Technology, Seoul 153-801, Korea, and Department of Chemistry, Korea UniVersity, Seoul 136-701 ReceiVed: August 8, 2008; ReVised Manuscript ReceiVed: October 13, 2008

The work describes a simple and convenient process for highly efficient and direct DNA separation with functionalized silica-coated magnetic nanoparticles. Iron oxide magnetic nanoparticles and silica-coated magnetic nanoparticles were prepared uniformly, and the silica coating thickness could be easily controlled in a range from 10 to 50 nm by changing the concentration of silica precursor (TEOS) including controlled magnetic strength and particle size. A change in the surface modification on the nanoparticles was introduced by aminosilanization to enhance the selective DNA separation resulting from electrostatic interaction. The efficiency of the DNA separation was explored via the function of the amino-group numbers, particle size, the amount of the nanoparticles used, and the concentration of NaCl salt. The DNA adsorption yields were high in terms of the amount of triamino-functionalized nanoparticles used, and the average particle size was 25 nm. The adsorption efficiency of aminofunctionalized nanoparticles was the 4-5 times (80-100%) higher compared to silica-coated nanoparticles only (10-20%). DNA desorption efficiency showed an optimum level of over 0.7 M of the NaCl concentration. To elucidate the agglomeration of nanoparticles after electrostatic DNA binding, the Guinier plots were calculated from small-angle X-ray diffractions in a comparison of the results of energy diffraction TEM and confocal laser scanning microscopy. Additionally, the direct separation of human genomic DNA was achieved from human saliva and whole blood with high efficiency. Introduction Effective and creative nanostructured probes for sensing biomolecules such as deoxyribonucleic acid (DNA) and proteins with binding and cleaving interactions have been interesting in bioengineering and medicinal chemistry.1-6 Current DNA separation methods have several drawbacks that make them unsuitable for the manufacture of pharmaceutical grade materials. These methods often involve the use of solvents; toxic chemicals such as cesium chloride, ethidium bromide, phenol, and chloroform; or animal-derived enzymes such as ribonuclease A and lysozyme that are either not approved or not recommended by regulatory agencies. Moreover, the density-gradient ultracentrifugation of saccharose or cesium chloride or ethidium bromide is time consuming and difficult to scale up and uses toxic and mutagenic reagents. A common limitation of a commercial matrix is furthermore the low capacity for DNA due to the inability of large molecules to penetrate porous beads; it has not been reported that the matrix can be utilized to separate the DNA and protein simultaneously. Consequently, there is a need for large-scale processes to manufacture DNA of a high level of purity for use; thus, the requirements of regulatory agencies regarding purity, potency, safety, and efficacy must be achieved. Magnetic nanoparticles (MNPs) of iron oxides have been extensively utilized as the materials of choice for magnetic resonance imaging (MRI),7 tissue-specific release of therapeutic agents,8,9 the labeling and sorting of cells,10 and the separation * Corresponding author. E-mail: [email protected]. Fax: +82-2-32827811. † Korea Institute of Ceramic Engineering and Technology. ‡ Korea University.

of biochemical products.11,12 Most of these applications require the nanoparticles to be chemically stable, uniform in size, and well dispersed in the media. The principle of the separation process is to utilize MNPs coated with intermediates such a surfactant, a polymer, a ligand, or silica in order to adsorb biomolecules, which thereafter can be separated according to the magnetic field gradient. Although there are many reports regarding the preparation of MNPs-silica composite particles, most showed an ill-defined structure and morphology as a result of a lack of systematic investigation of the formation of welldefined silica-coated MNPs (Si-MNPs). A small number of studies showed good preparation of Si-MNPs, but their major objective was not biomolecular sensing.13,14 Furthermore, biomolecular sensing with the molecularly functionalized MNPs is uncommon and scarcely available with a small size distribution in the nanoscale range. Chemical surface modifications of MNPs with suitable intermediates can be used to extract the desired target. The driving forces for the adsorptions are hydrophobic, electrostatic, and ligand binding interactions.15,16 Recently, we reported the feasibility of chemically functionalized Si-MNPs for high-throughput magnetic DNA separation process using electrostatic interaction with a negative phosphate backbone of DNA.17 The MNPs were synthesized with average particle size of 9 nm, and Si-MNPs were obtained in the range of 10-40 nm by a sol-gel method. However, the results had a lack of systematic investigation to elucidate the magnetic separation process due to the ill-defined morphology and dispersion. In this work, we prepared well-defined MNPs and demonstrated magnetic DNA separation as a function of the Si-MNPs’ particle sizes as well as control of the electrostatic strength

10.1021/jp807081b CCC: $40.75  2009 American Chemical Society Published on Web 12/19/2008

Magnetic Nanoparticles as DNA Separator

J. Phys. Chem. B, Vol. 113, No. 2, 2009 537

SCHEME 1: Schematic Presentation of the Preparation of Amino-Functionalized Silica-Coated Magnetic Nanoparticles for High-Throughput Direct DNA Separationa

a (1) magnetic nanoparticle (MNP, magnetite), (2) silica-coated MNPs (Si-MNPs), (3) amino-functionalized Si-MNPs, and (4) DNA-bound aminofunctionalized Si-MNPs.

according to various amino groups as shown in Scheme 1. The efficiency of magnetic DNA separation was calculated,and the detailed characterization was achieved to elucidate the interaction between DNA and Si-MNPs. It is expected that the aminofunctionalized silica-coated MNPs described here will be useful for magnetic DNA separation methods using a simple treatment and a high-throughput process involving selective interaction, replacing conventional silica-based processes. Experimental Section Synthesis of Amino-Functionalized Si-MNPs. The Si-MNPs were prepared by using a modification of a literature method.18 To prepare the Fe3O4 MNPs, a fresh mixture of FeCl2 · 4H2O (2 M) and FeCl3 · 6H2O (1 M) was added to ammonia solution (0.7 M) with vigorous stirring at room temperature. The obtained MNPs were separated by magnet and the supernatant was decanted. The collected MNPs were washed 3-4 times with deionized water and ethanol and then air-dried. Before the silicacoating process, the MNPs (90 mg) were dipersed in cyclohexane (120 mL) under sonication for 1 h at 50 °C in the presence of oleic acid (2 mL). The Igepal CO-520(24 g, Aldrich) disolved solution in cyclohexane (300 mL) was added to MNPs solution and then further stirred for 20 min. After stirring, a freshly prepared solution of tetraethyl orthosilicate (TEOS, 98%) was added dropwise. The molar concentration of TEOS was varied in the range of 10-50 mM to control the coating thickness. Aqueous ammonia solution was added until the pH of the mixture was raised to 12. The mixed solution was further stirred for 20 h at room temperature. Methanol was added to the solution to form dark precipitates, which were collected through centrifugation after removal of the supernatants. The dark precipitate was washed repeatedly through sonication in hexane. The removal of Igepal CO-520 was achieved by centrifugation after addtion of glacial acetic acid and chloroform, which process was repeated three times. The dark precipitate was further purified by sonication in hexane. To represent the functionality, amino-functionalized silanes such as aminopropyltrimethoxysilane (APTMS), N-[3-(trimethoxysilyl)propyl]ethylenediamine (TMPEA), N-[3-(trimethoxysilyl)propyl]diethylenetriamine (TMPDT), and silica-coated MNPs were reacted in toluene at 130 °C for 7 h. The final product was collected and washed with toluene three times. DNA Adsorption and Desorption Experiments. Fresh human blood and saliva were collected in a microtube and homogenized by vortex for 10 min before use. For DNA adsorption and desorption work with chemically functionalized MNPs, the blood chemically lysed to release from the cells. Proteinase K (10 µL of a 10 mL aqueous solution) was added to the whole blood, and the solution was then diluted to 300 µL in cell lysis buffer. After centrifugation for 10 min, the

supernatant of blood waste was removed and washed. The supernatant was again transferred into a new tube and 600 µL of binding buffer added. The solution was stirred for 10 min and centrifuged for 2 min at 12 000 rpm. Finally, TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 8) was added and then centrifuged for 1 min. DNA adsorption and desorption experiments were achieved that a total of 50 µL of crude cell lysates, 10 µL of aminofunctionalized Si-MNPs (10 mg/mL), and 50 µL of TAE buffer (40 mM Tris, 20 mM acetic acid, 1 mM EDTA, pH 8.0) were mixed and incubated for 10 min at laboratory temperature. The amino-functionalized Si-MNPs with adsorbed DNA were then separated using a magnetic separator for 1 min, the supernatant was collected, and the aminofunctionalized Si-MNPs were washed with 500 µL of TAE buffer and dried shortly. DNA captured to the amino-functionalized Si-MNPs was eluted into 50 µL of TAE buffer. DNA in the eluate (1 µL) was used as DNA matrix in PCR amplification and for agarose gel electrophoresis. The concentration of nucleic acids in solution can be readily calculated from absorbance at 260 nm. Agarose gel electrophoresis was achieved from a horizontal gel electrophoresis unit in TAE buffer including a 1% agarose gel and 0.1 µg/mL ethidium bromide (EtBr). Electrophoresis was carried out for each different sample at 100 V for 20 min involving the supernatants before/after adsorption, wash, and desorption work. Instrumental Analysis. The nitrogen adsorption and desorption isotherms were measured using a Quantachrome Autosorb-6 system. The sample was pretreated at 100 °C overnight in the vacuum line. The pore size distributions were calculated from the analysis of the adsorption branch of the isotherm using the Barrett-Joyner-Halenda (BJH) method. The pore volume was taken at the five points of P/P0. The XPS measurements were obtained by SSI 2803-S spectrometer with Al KR. X-ray source was 12 kV and 20 mA. All core-level spectra were obtained at a photoelectron takeoff angle of 35° with respect to the sample surface. To compensate for the surface charging effect, all binding energies were referenced to a C 1s neutral carbon peak at 284.6 eV. The N 1s peaks were deconvoluted into the components consisting of a Gaussian line shape Lorentzian function. The SAXD patterns were taken with 40 kV, 160 mA Cu KR radiation using a Rigaku Denki instrument. The scattered intensity was measured over the scattering vectors q ) (4π/ λ)sin θ, where 2θ is the total scattering angle and λ the wavelength (Cu KR radiation, λ ) 1.54 Å) generated from a rotating anode source which was monochromatized by crystal monochromator. The scattering curves were measured by using a point focusing scintillation detector. Confocal laser scanning microscopy was performed on a Leica-TCS SP2 AOBS model. Cy5-labeled DNA bound amino-functionalized Si-MNPs were excited at 650 nm and their emission spectrum was captured.

538 J. Phys. Chem. B, Vol. 113, No. 2, 2009 Results and Discussion Characterization of Amino-Functionalized Si-MNPs. The MNPs were prepared by the chemical precipitation of Fe2+ and Fe3+ salts with a molar ratio of 1:2 in a basic solution.19 The Si-MNPs were directly produced by the sol-gel reaction of a TEOS precursor, in which the coating layer served as a biocompatible and versatile group for further biomolecular functionalization. The thickness of the silica layer was controlled by the TEOS concentration for optimal binding with DNA. To effectively capture the DNA, Si-MNPs must be functionalized reproducibly on the silica surface. Three types of amino functional groups for DNA separation were introduced on the surface of Si-MNPs as a function of the number of amine group. These groups were mono-, di-, and triaminopropylalkoxysilanes. The morphology of the synthesized Si-MNPs including magnetic nanoparticles (magnetite) was confirmed by transmission electron microscopy (TEM) and high-resolution TEM (HR-TEM), which showed that the MNPs were highly dispersive and that their coating thickness was controlled. The prepared magnetite showed a single domain including a single-crystalline cubic spinel structure, in addition to a uniform size as calculated from XRD data (data is not shown here) that used the Scherrer equation showing 8 ( 2 nm, which was in good agreement with those obtained from the TEM images (Figure 1a).20 As the molar concentration of the added TEOS increased, the silica coating layer encapsulating the MNPs became thicker, which can be observed in the TEM images (Figure 1b-d). This process involved the base-catalyzed hydrolysis of TEOS and the subsequent condensation of silica onto the surfaces of the cores of the MNPs. The ratio between the concentration of MNPs and TEOS was optimized to avoid the homogeneous nucleation of silica and subsequent formation of core-free silica spheres. Although several factors, including the reaction time and the concentration of ammonia solution, could be adjusted to control the thickness of the silica shell, the control of the TEOS concentration is most convenient and reproducible when adjusting the shell thickness with strong magnetic strength. Moreover, the silica layer served as a biocompatible and versatile group for further molecular functionalization. The estimated particle sizes, surface areas, pore volumes, and saturation magnetization values as a function of the added TEOS molar concentration on the MNPs were plotted using TEM images and a particle size analyzer, as well as Brauner-Emmet-Teller (BET) measurements with nitrogen adsorption and a vibrating sample magnetometer (VSM) (Figure 1e). As the molar concentration of added TEOS increased, the particle sizes linearly increased with each additional 10 nm within the concentration range of 10-50 mM of the added TEOS: consequently, particle sizes were regularly distributed in a range of 15-60 nm. According to the results of the BET measurement, the surface areas and pore volumes of the silica-coated MNPs showed a linear decrease due to the increase of the amorphous silica layers caused by the sol-gel coating. Moreover, the MNPs and all of the Si-MNPs showed superparamagnetic characteristics, and the saturation magnetization (Ms) of the synthesized silica-coated MNPs decreased dramatically upon the increase of the silica coating layers from 61.42 to 33.87 emu/g, which is consistent with the increased amount of added TEOS from 10 to 50 mM. To confirm the formation of the silica coating layers and the assembly of three types of amino groups on MNP, cyclic voltammetry (CV), and X-ray photoelectron spectroscopy (XPS) were utilized. The cyclic voltammograms were achieved in 0.5 M sulfuric acid electrolyte solution with continuous cycling at a scan rate of 100 mV s-1 of magnetite (MNP), and Si-MNPs

Kang et al. were treated with 10 mM TEOS by applying a cyclic potential (between -0.5 and 1.3 V versus Ag/AgCl electrode) to the electrode involving electrochemical oxidation and reduction changes (Figure 2a). The shape of the cyclic voltammograms between MNP and Si-MNP showed a remarkable difference, in which the MNPs indicate strong electrochemical reactions including oxidation to ferric oxide and further reduction to ferrous ion upon the applied voltage (inset graph). However, Si-MNP indicated no electrochemical response. This result showed that the silica precursor TEOS would be perfectly coated on MNPs, as the silica layer is highly inert during the electrochemical reaction. Moreover, the cyclic voltammograms were more reactive with the increase in the amine number on the Si-MNPs, which is attributed to the increase of the positive charge from mono- to triamino functionality. X-ray photoelectron spectroscopy (XPS) with a wide of the scan spectra of the amino-functionalized Si-MNPs was conducted for the characterization of molecularly assembled surface (Figure 2b). The spectrum showed that the Fe electronic configuration peaks of magnetic nanoparticles were 846, 723, 710, 95, and 56 eV for 2s, 2p1/2, 2p3/2, 3s, and 3p, respectively. After the silica coating process, most of the Fe electronic configuration peaks were not detected, but Si peaks were detected at 149 and 100 eV for 2s and 2p, respectively. Finally, amino-functionalization on the surface of the MNPs was initially observed on the two peaks of N 1s and C 1s at 399 and 285 eV, respectively. For a quantitative analysis of the new C and N peaks, the atomic concentration profiles in XPS were calculated, and the result showed that the concentration of N 1s and C 1s increased to 5.5% and 21.5% from 3.3% and 17.6%, respectively, with the number of amino groups (Figure S3). These results are in good agreement with those of thermal gravity (TG) and FT-IR (ATR method) measurements. Magnetic DNA Separation with Amino-Functionalized SiMNPs. The surface modification by electrostatic strength changes and the particle size dependency of the Si-MNPs after amino-functionalization were explored through the adsorption and desorption of human DNA (45 000 bp) using electrophoresis and spectroscopic analyses of ultraviolet absorbance and fluorescence emission (Figure 3). The effective DNA adsorption work was achieved corresponding to the average size of the amino-functionalized silica-coated MNPs, including the SiMNPs, in which they showed 8-10 times enhanced adsorption ability compared to unmodified MNPs from a resident (supernatant) solution after DNA binding. Furthermore, a higher amino functionality on the Si-MNPs surface showed greater DNA adsorption in the order of triamino (TA) > diamino (DA) > monoamino (MA) functionality due to the increase of the level of electrostatic interaction between positive amino groups and the negative phosphate backbone of DNA. DNA adsorption was perfectly conducted with the use of an average particle size of 25 nm with DA and TA functionality. A spectroscopic analysis to explore the DNA separation work was also conducted using the ultraviolet absorbance and fluorescence emission measurements with the resident (supernatant) solution after DNA binding at wavelengths of 260 and 302 nm, respectively. Strong adsorption and emission bands at 260 and 302 nm derived from the base part of the DNA proved that DNA was actually bound to the amino-functionalized Si-MNPs as a function of the average particle size. This result revealed that the absorbance and fluorescence intensity gradually decreased with the decrease in the particle sizes. The fluorescence intensity was drastically reduced in the 15 nm particles that were used, and almost disappeared in 25 nm particles of TA-Si-MNPs used due to the

Magnetic Nanoparticles as DNA Separator

J. Phys. Chem. B, Vol. 113, No. 2, 2009 539

Figure 1. TEM images of MNPs and Si-MNPs: (a) MNPs (HR-TEM image inserted), (b) mixture of Si-MNPs and silica colloids (before washing), (c) 20 mM TEOS added Si-MNPs, (d) 50 mM TEOS added Si-MNPs. (e) Relationship among the particle size, magnetic strength, BET surface areas, and pore volume as a function of the additional TEOS concentration.

perfect DNA binding of the amino-functionalized MNPs. Another DNA separation experiment was also conducted as a function of the NaCl salt concentration with an average size of 25 nm for the adsorption and desorption measurements. The conventional DNA desorption process is also associated with temperature rises and NaCl salt additions. The process adapted in this work sought to change the ionic strength for the

dissociation using various molar concentrations of NaCl salt from 0.1 to 0.7 M, in which the sodium salt interacted with the phosphate backbone of DNA and decreased the bond strength of DNA-amino-functionalized silica-coated MNPs. The solution of the lower NaCl molar concentration did not dissociate the bounded DNA much, but the MA-Si-MNPs sample showed good dissociation (it was easily broken down) due to the weak

540 J. Phys. Chem. B, Vol. 113, No. 2, 2009

Kang et al.

Figure 2. (a) Cyclic voltammograms and (b) XPS spectrum: (i) MNPs, (ii) Si-MNPs, (iii) MA-Si-MNPs, (iv) DA-Si-MNPs, and (v) TA-SiMNPs.

electrostatic interactions. Consequently, additional amino-functionalized nanoparticles were required to dissociate the DNA with the addition of sodium salt. The DNA dissociation property with the amino-functionalized Si-MNPs inversely depended on the number of amino groups in the order of MA > DA > TA functionality. The effective DNA dissociation was unaffected by the functionality on the Si-MNPs over 0.7 M NaCl concentrations, as the all of the adsorbed DNA on the particles were perfectly dissociated: the ionic strength over the concentration was fully sufficient to dissociate for DNA desorption. Microscopic Characterization for DNA-Bound AminoFunctionalized Si-MNPs. Further studies of the electrostatic interaction between a target DNA labeled by cyanine (Cy5) and functionalized particles were achieved via confocal microscopy, a particle size analyzer, TEM, and small-angle XRD (SAXD) (Figure 4). The confocal images provided the DNA adsorption efficiency as a function of the level of amino functionality. With an increase in the level of amino functionality, the red color from the adsorbed Cy5 labeled DNA covers the entire detection area due to the strong electrostatic binding with DNA, with the exception of Si-MNP. Moreover, the amino-functionalized particles appear to be aggregated to each other in order to adsorb the target DNA effectively. This result was also confirmed by particle size analysis, in which the size distribution bands were determined to be located at 1.93, 4.03, and 6.53 µm, corresponding to MA-Si-MNPs, DA-Si-MNPs, and TA-Si-MNPs, respectively. The bandwidths of the particle size distribution

Figure 3. (a) DNA adsorption yields as a function of the average particle size and the number of amino groups on Si-MNPs (400 µg), (b) UV-visible and fluorescent spectrum corresponding to DNA binding with DA-Si-MNPs (0.5 mg) from the resident solution (1, 15 nm; 2, 25 nm; 3, 35 nm; 4, 45 nm; and 5, 55 nm of average particle size), and (c) DNA desorption yields from aminofunctionalized SiMNPs (25 nm) as a function of the NaCl concentration.

spectrum were broadened due to the increase in the hydrophilic aggregative force of the amino functionality. Guinier plots were also obtained to elucidate the shape and size distribution of the DNA-bound nanoparticles from small-angle X-ray diffraction (SAXD) patterns in the low-angle region.21 A SAXD analysis of the ln [q3I(q)] versus the [q2] plot showed nearly homogeneous matrices, but gave a thoroughly different appearance after TA-Si-MNPs, in which a distinct peak at 1.7 nm-2 can be attributed to the higher additive scattering compared to that of MA- or DA-Si-MNPs. The ln [I(q)] versus ln [q] plots were inserted, and all plots showed an exponential dependence on q, in which the SAXD intensity demonstrated an exponent between -2.1 and -1.2. Moreover, the average size of the amino-

Magnetic Nanoparticles as DNA Separator

J. Phys. Chem. B, Vol. 113, No. 2, 2009 541

Figure 4. Confocal laser scanning microscopy images of excited Cy5-labeled DNA-bound amino-functionalized Si-MNPs at 650 nm excited (a) MA-Si-MNPs, (b) DA-Si-MNPs, and (c) TA-Si-MNPs (the scale bar of the inserted image is 8 µm). (d) The particle-size distribution from particle size analyzer and a schematic diagram (I: MA-Si-MNPs, II: DA-Si-MNPs, III: TA-Si-MNPs), and (e) Guinier plots from SAXD.

functionalized nanoparticles could be roughly estimated, and the morphology of nanoparticles has spherical shape according to the TEM images. Thus, the size of the spherical nanoparticles (R) was calculated from the radius of gyration (Rg). The radius

of gyration was obtained from the plot as a slope (slope ) -(1/ 3)Rg2), and the average particle size (R) was calculated from the equation Rg2 ) (3/5)R2.22 The radius of gyration of aminofunctionalized Si-MNPs was 28.4, 29.7, and 30.8 nm corre-

542 J. Phys. Chem. B, Vol. 113, No. 2, 2009

Kang et al. in which a large amount of genomic DNA had collected (lane 3) and was successfully amplified via PCR process (lane 6). Genomic DNA was also obtained from whole blood, and the efficiency was proportionally dependent on the amounts of the DA-Si-MNPs used. To compare the use with identical amounts of particles, the separation efficiency showed 6-10 times enhancement in DA-Si-MNPs over commercialized 800 nm SiMNPs. Conclusions

Figure 5. Agarose gel electrophoresis images of direct DNA separation from human saliva and whole blood without biopsy. (a) (i) Elution of genomic DNA (5 µL) from saliva (100 µL) before PCR, and (ii) elution of genomic DNA (1 µL) after PCR amplification with commercial SiMNPs (800 nm), silica membrane, and with TA-Si-MNPs (25 nm) corresponding to lanes 1, 2, and 3 for before PCR, and lanes 4, 5, and 6 for after PCR, respectively, and (b) elution of genomic DNA (5 µL) from each blood sample of 10, 20, 100, and 200 µL with (i) commercial Si-MNPs (800 nm) and (ii) TA-Si-MNPs (25 nm).

sponding to MA-Si-MNPs, DA-Si-MNPs, and TA-Si-MNPs, respectively. The calculation of the amino-functionalized nanoparticle size resulted in 36.7, 38.3, and 40 nm for MA-Si-MNPs, DA-Si-MNPs, and TA-Si-MNPs, respectively. Consequently, the Guinier plot revealed that the DNA-bound region in the aggregation of amino-functionalized Si-MNPs increased to 1.7, 3.3, and 5.0 nm for MA-Si-MNPs, DA-Si-MNPs, and TA-SiMNPs, respectively. Direct Human DNA Separation with Comparison of Conventional Procedure and Am-Si-MNPs Procedure. To demonstrate clinical feasibility, high throughput and efficient DNA separation were directly achieved in human saliva and whole blood samples without the need for a biopsy (Figure 5). To determine the separation efficiency and optimize the experimental conditions, three types of materials were used: commercialized silica porous membrane, 800 nm Si-MNPs, and synthesized DA-Si-MNPs 30 nm in size before and after a polymerase chain reaction (PCR). An electrophoresis image of agarose gel showed that the DNA separation from saliva would be preferred to DA-Si-MNPs over commercialized materials,

A simple and convenient process was demonstrated for highly efficient and direct DNA separation with functionalized silicacoated magnetic nanoparticles. Iron oxide MNPs and Si-MNPs were uniformly obtained, and the silica coating thickness could be easily controlled in the range of 15-55 nm by changing the concentration of TEOS precursor including a controlled magnetic strength and particle size. The introduction of electrostatic strength changes with aminosilanization on the surfaces of the nanoparticles, which enhances the DNA separation efficiency due to the amount of electrostatic interaction. The DNA separation efficiency was explored via the function of the aminogroup number, particles size, amount of Si-MNPs used, and NaCl concentration. The DNA adsorption yields were high in terms of the use of triamino-functionalized nanoparticles with an average particle size of 25 nm. The adsorption efficiency of the amino-functionalized nanoparticles was 4-5 times (80-100%) higher compared to only silica-coated nanoparticles (10-20%). DNA desorption efficiency showed the optimum level at over 0.7 M of NaCl concentration. To elucidate the agglomeration of nanoparticles after electrostatic DNA binging, the Guinier plots were calculated from small-angle X-ray diffraction results, as compared with the results of energy diffraction TEM, and confocal laser scanning microcopy. In addition, the direct separation of human genomic DNA was achieved from human saliva and whole blood with high efficiency. This preliminary study could enable the design and construction of an automatic system with high-throughput biomolecular purification with functionalized magnetic nanoparticles and would be applied for clinical diagnoses and proteins/enzymes recognition processes using the appropriate surface modification technique. Acknowledgment. This work was supported by a grant from the Fundmental R&D Program for Core Technology of Materials funded by the Ministry of Knowledge Economy, Republic of Korea. We thank Dr. Yeonhwan Jeong (Kyoto Institute of Technology, Japan) for helpful discussion on small-angle XRD study. Supporting Information Available: Detailed characterization of the obtained materials: TEM images, wide-angle XRD data, magnetic property, BET isotherms, attenuated total reflection (ATR) FT-IR spectrum, and elemental analysis of XPS; electron diffraction (ED) measurement for DNA-bound aminofunctionalized Si-MNPs by TEM, and DNA adsorption yields as function of used amounts of nanoparticles, fluorescent images and cell transfection results. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Boal, A. K.; Synthesis and application of magnetic nanoparticles; Kluwer Academic/Plenum Publishers: New York, 2004. (2) Katz, E. A.; Shipway, N.; Willner, I. Biomaterial-nanoparticle hybrid systems; Wiley-VCH: Weinheim, Germany, 2004. (3) Kuhara, M.; Takeyama, H.; Tanaka, T.; Matsunaga, T. Anal. Chem. 2004, 76, 6207.

Magnetic Nanoparticles as DNA Separator (4) Park, S. J.; Taton, T. A.; Mirkin, C. H. Science 2002, 295, 1503. (5) Houseman, B. T.; Huh, J. H.; Kron, S. J.; Mrksich, M. Nat. Biotechnol. 2002, 20, 270. (6) Warner, M. G.; Hutchison, J. E. Nat. Mater. 2003, 2, 272. (7) Babes, L.; Denizot, B.; Tanguy, G.; Jacques, J.; Jeune, L.; Jallet, P. J. Colloid Interface Sci. 1999, 212, 474. (8) Byren, S. J.; Corr, S. A.; Gunko, Y. K.; Kelly, J. M.; Brougham, D. F.; Ghosh, S. Chem. Commun. 2004, 2560. (9) See, for example: (a) Osaka, T.; Matsunaga, T.; Nakanishi, T.; Arakaki, A.; Niwa, D.; Iida, H. Anal. Bioanal. Chem 2006, 384, 593. (10) Chemla, Y. R.; Crossman, H. L.; Poon, Y.; McDermott, R.; Stevens, R.; Alper, M. D.; Clarke, J. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 14268. (11) Katz, E.; Weizmann, Y.; Willner, I. J. Am. Chem. Soc. 2005, 127, 9191. (12) Willner, I.; Katz, E. Langmuir 2006, 22, 1409. (13) Lu, Y.; Yin, Y.; Mayers, B. T.; Xia, Y. Nano Lett. 2002, 2, 183. (14) Landfester, K.; Ramirez, L. P. J. Phys.: Condens. Matter 2003, 15, 1345.

J. Phys. Chem. B, Vol. 113, No. 2, 2009 543 (15) Zhao, X.; Tapec, R.; Wang, K.; Tan, W Anal. Chem. 2003, 75, 3476. (16) Oktem, H. A.; Bayramoglu, G.; Ozalp, V. C.; Arica, M. Y. Biotechnol. Prog. 2007, 23, 146. (17) Park, M. E.; Chang, J. H. Mater. Sci. Eng., C 2007, 27, 1232. (18) Park, R. J.; An, K.; Hwang, Y.; Park, J.-C.; Noh, H.-J.; Kim, J.Y.; Park, J. H.; Hwang, N.-M.; Hyeon, T. Nat. Mater. 2004, 3, 891. (19) Santra, S.; Wang, K. M.; Tapec, R.; Theodoropoulou, N.; Hebard, A.; Tan, W. Langmuir 2001, 17, 2900. (20) Roe, R. J. Methods of X-ray and neutron scattering in polymer science; Oxford University Press: New York, 2000. (21) Jeong, Y.; Hanabusa, K.; Masunaga, H.; Akiba, I.; Miyoshi, K.; Sakurai, S.; Sakurai, K. Langmuir 2005, 21, 586. (22) Strander, A.; Cardinaux, F.; Schurtenberger, P. J. Phys. Chem. B 2006, 110, 21222.

JP807081B