DNA Surface Modified Gadolinium Phosphate Nanoparticles as MRI

Article pubs.acs.org/bc

DNA Surface Modified Gadolinium Phosphate Nanoparticles as MRI Contrast Agents Matthieu F. Dumont,†,§ Celine Baligand,‡ Yichen Li,† Elisabeth S. Knowles,§ Mark W. Meisel,§ Glenn A. Walter,‡ and Daniel R. Talham*,† †

Department of Chemistry, University of Florida, Gainesville, Florida 32611-7200, United States Department of Physiology and Functional Genomic, University of Florida and the National High Magnetic Field Laboratory, Gainesville, Florida 32610-0274, United States § Department of Physics, University of Florida and the National High Magnetic Field Laboratory, Gainesville, Florida 32611, United States ‡

S Supporting Information *

ABSTRACT: Oligonucleotide modified gadolinium phosphate nanoparticles have been prepared and their magnetic resonance relaxivity properties measured. Nanoparticles of GdPO4·H2O were synthesized in a water/oil microemulsion using IGEPAL CO-520 as surfactant, resulting in 50 to 100 nm particles that are highly dispersible and stable in water. Using surface modification chemistry previously established for zirconium phosphonate surfaces, the particles are directly modified with 5′-phosphate terminated oligonucleotides, and the specific interaction of the divalent phosphate with Gd3+ sites at the surface is demonstrated. The ability of the modified nanoparticles to act as MRI contrast agents was determined by performing MR relaxivity measurements at 14.1 T. Solutions of nanopure water, Feridex, and Omniscan (FDA approved contrast agents) in 0.25% agarose were used for comparison and control purposes. MRI data confirm that GdPO4·H2O nanoparticles have relaxivities (r1, r2) comparable to those of commercially available contrast agents. In addition, the data suggest that biofunctionalization of the surface of the nanoparticles does not prevent their function as MRI contrast agents.

■

1 1 = + ri(s−1·mM−1) × [CA](mM) Tiapp Ti

INTRODUCTION Magnetic resonance imaging (MRI) is a noninvasive medical imaging technique that allows both morphological and functional evaluation of tissues, making it a powerful diagnosis and research tool. MRI contrast depends on a number of factors including proton density, longitudinal relaxation time (T1), and transverse relaxation time (T2). This contrast varies depending on the tissue type and can be altered with pathology, making it especially useful in neurological and oncological imaging.1−3 Despite the ability to image, in some cases, an increase in the endogenous contrast is needed to distinguish between two tissues that naturally produce very similar MR signal. This fact is particularly relevant to in vivo cell therapy protocols or gene expression tracking, where the therapeutic material does not offer any natural contrast with the tissue.4 Improved MR contrast is made possible by the use of contrast agents (CAs) that modify the rates of relaxation of the water protons in their vicinity. The efficiency of CAs is usually estimated as the magnitude of their effect on the water proton relaxation rates and is related to the concentration and relaxivity of the agents by the following equation: © 2012 American Chemical Society

i = 1, 2 (1)

where Tiapp is the apparent T1 or T2 of the tissue, ri are the relaxivities r1 or r2, and [CA] is the concentration of contrast agent.1,5,6 Magnetic resonance CAs are conventionally divided into two main classes depending on their superparamagnetic or paramagnetic properties. The first class is exemplified by superparamagnetic iron oxide nanoparticles (SPIOs). Superparamagnetic particles affect the transverse relaxation rate, R2, of the protons in their vicinity through spin−lattice interactions and are known as T2 or negative contrast agents.6,7 The second class of agent is mainly based on chelates of gadolinium(III). These molecules are strongly paramagnetic as a result of unpaired electrons giving a spin state of S = 7/2. Chelation is required to reduce their toxicity and make their use in vivo possible. These complexes operate through an inner-sphere spin−spin relaxation mechanism, show a high longitudinal Received: October 19, 2011 Revised: February 1, 2012 Published: April 2, 2012 951

dx.doi.org/10.1021/bc200553h | Bioconjugate Chem. 2012, 23, 951−957

Bioconjugate Chemistry

Article

Figure 1. Illustration of the surface of the gadolinium phosphate nanoparticles. Hydroxide and oxide ions present on the surface can be displaced by divalent phosphate terminated oligonucleotide, creating a coordinate covalent linkage.

relaxivity r1 and generate a positive contrast by brightening of the image. They are known as T1 or positive contrast agents.5 Recent studies have raised some concerns about the release of Gd3+ in the body, leading to pathology such as nephrogenic systemic fibrosis.8−11 The need to keep gadolinium in a more stable form has led to the investigation of different types of gadolinium-based nanomaterials. Some recent examples are small particulate gadolinium oxides,12−14 GdF3 nanoparticles,15 Gd-DOTA attached to SiO2-coated semiconductor quantum dots (DOTA is tetraazacyclododecanetetraacetic acid),16 and Gd(BDC)1.5(H2O)2 nanorods (BDC is 1,4-benzendicarboxylate).17 Similarly, Hifumi et al.18,19 reported the use of dextrancoated gadolinium phosphate nanoparticles that act as a positive contrast agent. Gadolinium phosphate is potentially highly desirable because of its extremely low solubility and its predisposition to form paramagnetic nanoparticles.20,21 For many nanoparticle CAs, targeting of the agents relies on the effects of particle size, localizing where they are not excluded by their size, such as tumor sites with high vascularity. Driven by the goal to improve efficiency and to lower dosage, techniques have been developed to modify the surface of nanoparticles with bioactive targeting molecules such as oligonucleotides, antibodies, or proteins. Surface modification has the added benefit of facilitating biocompatibility, reducing toxicity, and avoiding immediate elimination from the body. Methods to modify nanoparticles with bioactive molecules include approaches such as using the avidin/biotin interaction to link groups to the surface and passivating particles with functionalized polymers.2,22−24 These techniques have been successful, but often require significant effort through specialized synthesis and multiple steps. Recent studies in our laboratories have shown that oligonucleotides bearing a 5′-terminus phosphate covalently bind directly to zirconium phosphonate-modified surfaces through the 5′-phosphate and that this interaction can be used to immobilize oligonucleotide probes for applications such as DNA arrays.25−27 In the present article, we report that the same type of chemistry can be applied to the surface of gadolinium phosphate nanoparticles (GdPO4NP) (Figure 1). Since the Ka values of Gd3+ and Zr4+ are close, the metal ions show similar affinity for divalent phosphate. A phosphorylated oligonucleotide was selected as the surface modifier because of its potential for cancer-cell marker recognition and its high biocompatibility, mandatory for in vivo applications. Specific binding of the terminal phosphate of an oligonucleotide on the GdPO4NPs has been established by confocal laser scanning microscopy. In order to demonstrate that the oligonucleotides

covalently attached to the nanoparticle surface remain bioactive, hybridization experiments were performed with a fluorescently labeled complement. Subsequent relaxivity measurements established that the biofunctionalization of the surface of the GdPO4NPs does not alter their role as MRI contrast agents. The DNA-modified GdPO4 nanoparticles (GdPO4NP-DNA) exhibit relaxivities comparable to those of FDA approved contrast agents.

■

EXPERIMENTAL PROCEDURES Materials. All oligonucleotides were custom synthesized and purchased from Sigma-Aldrich (St. Louis, MO) and used as received. The oligonucleotide strands were HPLC-purified by Sigma-Aldrich before use. The appropriate amount of sodium saline citrate (SSC) buffer (pH 8) was added to the oligonucleotide to make 100 μM solutions, and the DNA solutions were subsequently aliquoted into one-time-use volumes (10 μL). The aliquots were stored at −20 °C. All other reagents were of analytical grade and used as received from commercial sources. High Resolution Transmission Electron Microscopy (HRTEM). Microscopy was performed on a JEOL 2010F HRTEM at 200 kV. TEM grids (carbon film on holey carbon support film, 400 mesh, copper from Ted-Pella, Inc.) were prepared by dropping onto the grid 20 μL of a solution containing 5 mg of sample in 2 mL of H2O dispersed by sonication for 30 min. Energy Dispersive X-ray Spectroscopy (EDS). Analysis was performed with an Oxford Instruments EDS X-ray Microanalysis System coupled to the HRTEM microscope. A total of 4 scans were completed on different parts of the sample and averaged to give relative atomic percentages for gadolinium and phosphorus. X-ray Powder Diffraction (XRD). Powder data were obtained using a Philips APD 3720 powder diffractometer. One hundred milligrams of nanoparticles were mounted on doublesided tape backed by a glass slide. Confocal Laser Scanning Microscopy (CLSM). DNAmodified gadolinium phosphate nanoparticles were imaged using a confocal microscope setup consisting of an Olympus IX-81 inverted microscope with an Olympus FluoView 500 confocal scanning system.

■

METHODS Gd(PO4)·H 2O Nanoparticle (GdPO 4NP) synthesis. GdPO4NPs were obtained by combining two precursor mixtures. The first was prepared by adding 500 mg of 952



Article

measurements. Control phantoms consisting of 0.25% agarose alone were concurrently imaged. All samples were secured together inside a 10 mm NMR glass tube (Wilmad-LabGlass, Vineland, NJ) to prevent movement. The NMR tube was filled with perfluorotri-n-butylamine (FC-43 Fluorinert, 3M, Maplewood, MN) in order to avoid magnetic susceptibility artifacts from the air−glass interfaces. MR Relaxation Time Measurements. All MR measurements were performed at a constant temperature of 22 °C in a vertical 14.1 T magnet (Avance 600, Bruker Instruments, Billerica, MA) equipped with microimaging gradients (micro-5, Bruker Instruments). Samples were placed into a 10 mmdiameter Helmholtz probe. The spectrometer was interfaced with ParaVision 3.0.2 software (Bruker). For each of the tested contrast agents, T1 and T2 relaxation times were measured in the same axial 0.5 mm-thick slice, positioned across the middle of the different capillaries. A variable repetition time (TR) spin echo sequence was used for T1 measurements (variable TR = (15; 7.5; 4; 3; 2; 1.5; 1; 0.75; 0.5; 0.25; 0.15; 0.1; 0.075; 0.05; 0.015 s); echo time (TE) = 4 ms; matrix size = 128 × 128, FOV = 1 × 1 cm2). T2 was measured with a multiple spin echo sequence (TR/TE = 4.5 ms/15 s, interecho spacing ΔTE = 5 ms (32 echoes), matrix size = 128 × 128, FOV = 1 × 1 cm2). MR Data Analysis. All MR data were analyzed using ParaVision 3.0.2 software (Bruker) and Excel (Microsoft Inc.). For each sample, the signal intensity on magnitude images was averaged within regions of interest (ROIs) and plotted against TR for T1 recovery curves or TE for T2 decay curves. Data were then fitted to the following monoexponential functions:

Gd(NO3)3 in 5 mL of H2O to a solution of IGEPAL CO-520 (20 mL) dissolved in 100 mL of cyclohexane under vigorous stirring. The second was a mixture of NaH2PO4 (700 mg) in 5 mL of H2O combined with a solution of IGEPAL CO-520 (20 mL) dissolved in 100 mL of cyclohexane. The two precursor mixtures were combined by dropwise addition using a peristaltic pump (10 mL/hour). Once the addition was complete, the mixture was left under vigorous stirring. After one hour, the microemulsion was broken with 200 mL of acetone. The nanoparticles were collected by centrifugation, washed with 50 mL of water and 2 × 50 mL of acetone, and left to dry under N2. The composition was confirmed by comparison to published X-ray data. Gadolinium Surface Preparation. GdPO4NP (20 mg) were redispersed by sonication in 20 mL of nanopure water. A solution of 2 mg of Gd(NO3)3 in 5 mL of H2O was then added to the nanoparticle suspension and sonicated for 20 min. The nanoparticles were subsequently washed by 3 successive cycles consisting of isolation by centrifugation and redispersion in nanopure water. The Gd-modified GdPO4NPs were kept redispersed in 10 mL of water. DNA Functionalization of GdPO4NP. A solution of phosphorylated-single-stranded DNA 5′-[phos]-ATCTAACTGCTGCGCCCGCCG-3′ (250 nM,100 μL) was added to 20 mg of Gd-modified GdPO4NPs in 10 mL of water and left to incubate for 24 h. The nanoparticles were then washed with nanopure water to eliminate unbound DNA and kept in 10 mL of nanopure water. Control experiments were carried out using the nonphosphorylated probe (5′-ATCTAACTGCTGCGCCCGCCG-3′). Probe Hybridization. A solution (100 μL, 250 nM) of Cy3-labeled complement (5′-[Cy3]-CGGCGGCGCAGCAGTTAGAT-3′) was added to 5 mL of the DNA modified GdPO4NP (GdPO4NP-DNA) and left to incubate for 24 h. The nanoparticles were then washed with nanopure water to eliminate unbound Cy3-DNA and kept in 10 mL of nanopure water. The changing surface properties of the nanoparticles through the synthesis and modification steps were monitored using zeta potential. The measured zeta potential value of GdPO4NP immediately after synthesis is −17 mV. Following treatment with Gd(NO3)3, the zeta potential increased to 25 mV, confirming the formation of a cationic layer on the surface of the particles. This Gd3+ layer proved to be robust to rinsing. After incubation with the phosphorylated oligonucleotide, the measured surface potential was again negative; −23 mV corresponded to surface binding of the negatively charged DNA. Values of the zeta potential measurements confirm the successive changes at every stage of the surface modification. Phantom Preparation. The GdPO4NP-DNA (2 mM) MR contrast agent was compared to commercial iron oxide (Feridex, 11.2 mg·mL−1, Berlex Laboratories, Montville, NJ), and gadolinium chelate (Omniscan, Gd-DTPA-BMA, 500 mM, Berlex Laboratories, Montville, NJ). GdPO4NP-DNA standards were prepared in concentrations of 2, 1.5, 1, 0.5, 0.25, 0.12, and 0.06 mM in a 0.25% agarose solution prepared with agar gelose (Acros Organics, Pittsburgh, PA). Omniscan and Feridex standards were made using the same procedure, in concentrations of 25, 12.5, 6.25, 3.13, 1.56, and 0.78 mM and 2.5, 1.87, 1.25, 0.63, 0.31, 0.16, and 0.08 mM, respectively. Borosilicate capillary tubes (Curtin-Matheson Scientific, Broomall, PA) were then filled with the solutions and sealed with clay (Critoseal, McCormick Scientific, Richmond, IL). The use of agarose averts possible sedimentation during relaxation

T1: (TR) = A(1 − e(−TR/T)1 ) + y0

(2)

T2: (TE) = A(1 − e(−TE/T2)) + y0

(3)

In both cases, the constant y0 was set to zero, and the low signal intensity values were excluded from the fit when inferior to 3 times the standard deviation of the noise in the images.28 The obtained T1 and T2 values were converted into R1 and R2 relaxation rates (1/T1 (s−1), 1/T2 (s−1)). Finally, R1 and R2 values were plotted against the concentration of the corresponding contrast agent, and r1 and r2 (mM−1 s−1) relaxivities were obtained as the slope of the resulting linear plots using the following equations:

■

R1 = r1 × [CA](mM) + R 1agarose

(4)

R 2 = r2 × [CA](mM) + R 2agarose

(5)

RESULTS Nanoparticle Synthesis. The nanoparticles were synthesized in a water-in-oil microemulsion containing Gd(NO3)3 and NaH2PO4. The assembly was formed by adding the precursors in an aqueous phase to a cyclohexane solution containing a surfactant (IGEPAL CO-520) to form reverse micelles containing microdroplets of the precursors dispersed in the reaction media. When the two precursor solutions were mixed, the reverse micelles coalesced leading to formation of gadolinium phosphate. The use of a peristaltic pump to mix the reactants gave particles with well-defined shapes and surfaces. High resolution transmission electron microscopy images show rod-shaped nanoparticles averaging 50 nm in length and 953



Article

that were not pretreated with Gd3+ and incubated in phosphorylated DNA, and particles that were not pretreated with Gd3+ and incubated in nonphosphorylated DNA. The four conditions are shown schematically in Figure 4.

10 nm in width as seen in Figure 2. Lattice fringes visible in the images are consistent with a highly crystalline product. To

Figure 2. TEM images of GdPO4·H2O nanoparticles modified with DNA. The lattice planes are visible on the close up (inset).

further prove the composition and the crystallinity of the synthesized materials, a powder X-ray diffraction (XRD) analysis was performed (Figure 3). The XRD patterns are

Figure 4. Schematic of different modification procedures (top) and confocal laser scanning microscopy images of drops of dispersions of the GdPO4·H2O nanoparticles in water following each procedure (bottom). (a) Nanoparticles treated with Gd3+ and followed by phosphorylated oligonucleotide; (b) treated with Gd3+ and followed by nonphosphorylated oligonucleotide; (c) without postsynthesis treatment with Gd3+ but with phosphorylated oligonucleotide; and (d) without postsynthesis treatment with Gd3+ and with nonphosphorylated oligonucleotide. In each case, the nanoparticles are shown after incubation with Cy3-labeled complement and methodical rinsing. Because the fluorophore is attached to the complementary strand, robust fluorescence indicates successful modification and hybridization steps. The fluorescence intensities are color coded from blue (low) to green, yellow, red, and white (saturation).

Figure 3. X-ray diffraction pattern of the gadolinium phosphate nanoparticles indexed to GdPO4·H2O. (JCPDS no. 39-232; space groupn P3121, #152).

consistent with the known hexagonal GdPO4·H2O with a = 6.90 Å and c = 6.35 Å (JCPDS no. 39-232; space group, P3121, #152).21,29,30 Magnetic susceptibility measurement of the GdPO4·H2O nanoparticles show paramagnetic behavior at room and physiological temperatures (Supporting Information). Surface Modification of the GdPO4NPs. The phosphateterminated oligonucleotides are expected to specifically bind the particle surface through the divalent phosphate, in line with observations made for zirconium phosphate and zirconium phosphonate surfaces.25−27 The specific Gd3+-divalent phosphate linkage requires that Gd3+ sites are accessible at the particle surface (Figure 1). The as-synthesized GdPO4·H2O are phosphate-rich at the surface, as confirmed by zeta potential measurements (Supporting Information). Therefore, a postsynthesis treatment of the particles with excess Gd3+ is employed to ensure a Gd3+-rich surface for oligonucleotide binding. The specific interaction depicted in Figure 1 was verified by monitoring phosphate-terminated oligonucleotide binding to Gd3+ treated particles and comparing it to three control experiments. The controls include particles that were pretreated with Gd3+ and incubated in nonphosphorylated DNA, particles

A fluorescently labeled complementary oligonucleotide strand (target) was used to confirm the modification of the particles. Confocal microscopy detects fluorescence on nanoparticles dispersed in water. When the fluorescence is maintained after successive rinsing procedures, the target is successfully hybridized to the probe strands covalently attached to the surface of the particle (Figure 4). Confocal microscopy shows that the greatest binding affinity is for the phosphorylated DNA with GdPO4NP treated with Gd3+. This result points out that nonspecifically bound DNA is removed from the particles following the wash procedures. Particles pretreated with Gd(NO3)3 and incubated with nonphosphorylated oligonucleotides showed some areas of saturated fluorescence, but it is not uniform, which can be explained by a few aggregates of particles trapping nonspecifically bound DNA. By far, the greatest extent of hybridization occurred on Gd3+treated GdPO4NPs that were subsequently modified with phosphorylated oligonucleotides. 954



Article

Figure 5. MR relaxivity data and image of serial dilutions of DNA modified GdPO4·H2O nanoparticles (GdPO4NP-DNA). (I) and (II) Plots of the values of longitudinal (T1) and transverse (T2), relaxation respectively, measured against the concentration of Gd3+. (III) T2 weighted images of successive dilutions of GdPO4NP-DNA (in sequence, from a to h: 2, 1.5, 1, 0.5, 0.25, 0.12, 0.06 mM, and pure agarose).

Table 1. Comparison of the Chemical and Magnetic Characteristics of Functionalized GdPO4·H2O Nanoparticles, Omniscan, and Feridex relaxivity (mM−1s−1)c

a

a

contrast agent

structure

magnetism

GdPO4NP-DNA Omniscan Feridex

NP C NP

P P SP

b

chemical composition

r1

r2

r2 / r1

GdPO4 Gd3+-DTPA-BMA iron oxide

0.2 4.6 1.9

12.8 5.4 313

60.9 1.1 164.7

NP, nanoparticles; C, chelate. bP, paramagnetic; SP, superparamagnetic. cMR measurements were performed at 600 MHz (14.1 T field).

MR Measurements. Dispersions of the DNA-modified nanoparticles in agarose were used to evaluate the r2/r1 values. Longitudinal and transversal relaxation rates, R1 and R2, were measured in the same phantom consisting of solutions containing increased concentrations of GdPO4NP-DNA in agarose. As expected, a linear relationship was established between the relaxation rates and the agent concentration (Figure 5). Longitudinal and transversal relaxivities were derived from the linear fits of R1 and R2 versus gadolinium concentration. Relaxivities were found to be r1 = 0.2 mM−1·s−1 and r2 = 12.8 mM−1·s−1 from which we calculated the ratio r2/r1 = 60.9 (Table 1). A T2 weighted image of successive dilutions of GdPO4NP-DNA showed an obvious negative contrast, increasing with the agent concentration (Figure 5). All together, these data confirmed the efficacy of GdPO4NPDNA as an MR contrast agent. Identical measurements were carried out with commercial gadolinium chelate and SPIO, using the same sequences. The values of r1 and r2 are reported in Table 1. Omniscan, a Gd-chelate positive contrast agent was found to have an r2/r1 value of 1.1 at 14.1 T, whereas Feridex, an SPIO negative contrast agent, had an r2/r1 of 164.7. These results were consistent with the specifications of these agents. The value of r2/r1 = 60.9 obtained for GdPO4NP-DNA and the loss of signal observed with increasing concentration confirmed that GdPO4NP-DNA has the characteristics of a T2 or negative contrast agent.

groups, and the specificity comes from the fact that the divalent phosphate can displace the oxides and hydroxides on the surface, whereas the phosphodiester backbone [(RO)2PO2−] of a DNA molecule is not sufficiently basic, only forming electrostatic or hydrogen bonding interactions.25−27 In the absence of the terminal phosphate, the DNA probe is easily washed away. Interaction between phosphate terminating the DNA and the surfaces of acidic metal phosphates and phosphonates has been shown to be stable under a range of conditions.25−27 The same principle applies for DNA binding to GdPO4NP. Oligonucleotide binding to these particles occurs through a coordinate covalent linkage with the terminal 5′-phosphate of the oligonucleotides. Therefore, binding to these particles does not require any modification of the DNA molecule to obtain specific immobilization or any polymer or organic coating of the particles. Furthermore, the phosphate-terminated oligonucleotides specifically bound to the GdPO4NP were able to hybridize with their complementary sequence, demonstrating sustained biological activity. The procedures should enable efficient surface modification with targeting markers associated with pathological conditions while simultaneously providing a biocompatible outer-shell to the nanoparticles. MR Measurements. Despite their content in Gd3+, which is usually utilized as a positive contrast agent, the GdPO4NPDNA nanoparticles developed in this study proved to generate negative MR contrast. This observation was confirmed by the high r2/r1 value of 60.9 and the generation of a negative contrast on T2-weighted images at 14.1 T. In general, the development of positive MR contrast agents at high magnetic fields is challenging. Indeed, the sensitivity of such compounds depends on the complex interplay between different structural and dynamics properties, such as rotational mobility or the residence lifetime of the water protons at the metal site.31 High r1 relaxivities have been achieved in a few cases. For example, using similar gadolinium phosphate nanoparticles to those studied here but coated with dextran, Hifumi et al.18,19 reported

■

DISCUSSION Surface Modification. The results demonstrate specific binding of phosphate terminated DNA oligonucleotides to the surface of GdPO4·H2O nanoparticles. The result is consistent with previous work on zirconium phosphate and zirconium phosphonate surfaces that showed that coordinate covalent bonds can be formed between the divalent phosphate group at the terminus of oligonucleotides and Zr4+ ions at the surface. The acidic Zr4+ surface sites are bound by oxide and hydroxide 955



Article

small r2/r1 values of 1.1, although at very low magnetic field (0.47 T). Relaxivity measurements of the oligonucleotide modified GdPO4NPs at 1.47 T gave r2/r1 of 1.7 ± 0.2, confirming the positive contrast agent properties observed by these authors at low field (Supporting Information). At higher fields (> 3 T), the relaxivity of gadolinium-based agents decreases rapidly due to their frequency dependence, making the achievement of high r1 a bigger challenge.31 The DNA-modified GdPO4·H2O particles induced an important change in the relaxivity of surrounding water, indicating high sensitivity as a negative MRI contrast agent. The ability to easily functionalize the particles with phosphateterminated DNA, such as that enzymatically produced by polymerase chain reaction (PCR), suggests this system could potentially enable cancer-cell recognition in order to specifically enhance the contrast of the tumors with surrounding tissue.32,33 In addition, surface modification combined with the low solubility of GdPO4·H2O means that Gd3+ release is slow. Measurements of the supernatant of suspensions kept for 18 months showed no significant discharge of Gd3+ (Supporting Information). Additionally, data suggesting in vivo stability and low cytotoxicity have been reported for similar gadolinium phosphate nanoparticles by Hifumi et al.18

in the Advanced Magnetic Resonance Imaging and Spectroscopy facility (AMRIS) in the McKnight Brain Institute (MBI) at the University of Florida, which is supported by the National High Magnet Field Laboratory and the State of Florida. We gratefully acknowledge Weihong Tan and Suwussa Bamrungsap from the Chemistry Department at University of Florida for MR relaxivity studies, Kerry Siebein at the UF Major Analytical Instrument Center for HRTEM imaging and EDS analysis work, and Dan Plant at AMRIS/MBI for MRI work.

■

(1) Fukumori, Y., and Ichikawa, H. (2006) Nanoparticles for cancer therapy and diagnosis. Adv. Powder Technol. 17, 28−33. (2) Louie, A. (2010) Multimodality imaging probes: design and challenges. Chem. Rev. 110, 3146−3171. (3) Oghabian, M. A., and Farahbakhsh, N. M. (2010) Potential use of nanoparticle based contrast agents in MRI: a molecular imaging perspective. J. Biomed. Nanotechnol. 6, 203−207. (4) Cahill, K. S., Germain, S., Byrne, B. J., and Walter, G. A. (2004) Non-invasive analysis of myoblast transplants in rodent cardiac muscle. Int. J. Cardiovasc. Imaging 20, 593−597. (5) Caravan, P., Ellison, J. J., McMurry, T. J., and Lauffer, R. B. (1999) Gadolinium(III) chelates as MRI contrast agents: structure, dynamics, and applications. Chem. Rev. 99, 2293−2321. (6) Sharma, P., Brown, S., Walter, G., Santra, S., and Moudgil, B. (2006) Nanoparticles for bioimaging. Adv. Colloid Interface Sci. 123, 471−476. (7) Laurent, S., Forge, D., Port, M., Roch, A., Robic, C., Vander Elst, L., and Muller, R. N. (2008) Magnetic iron oxide nanoparticles: synthesis, stabilization, vectorization, physicochemical characterizations, and biological applications. Chem. Rev. 108, 2064−2069. (8) Boyd, A. S., Zic, J. A., and Abraham, J. L. (2007) Gadolinium deposition in nephrogenic fibrosing dermopathy. J. Am. Acad. Dermatol. 56, 27−31. (9) DaSilva, M., Deming, M. O., Fligiel, S. E. G., Dame, M. K., Johnson, K. J., Swartz, R. D., and Varani, J. (2010) Responses of human skin in organ culture and human skin fibroblasts to a gadolinium-based MRI contrast agent comparison of skin from patients with end-stage renal disease and skin from healthy subjects. Invest. Radiol. 45, 733−737. (10) Ersoy, H., and Rybicki, F. J. (2007) Biochemical safety profiles of gadolinium-based extracellular contrast agents and nephrogenic systemic fibrosis. J. Magn. Reson. Im. 26, 1190−1194. (11) Thakral, C., and Abraham, J. L. (2009) Gadolinium-induced nephrogenic systemic fibrosis is associated with insoluble Gd deposits in tissues: in vivo transmetallation confirmed by microanalysis. J. Cutananeous Pathol. 36, 1244−1248. (12) Bridot, J.-L., Faure, A.-C., Laurent, S., Riviere, C., Billotey, C., Hiba, B., Janier, M., Josserand, V., Coll, J.-L., Vander Elst, L., Muller, R., Roux, S., Perriat, P., and Tillement, O. (2007) Hybrid gadolinium oxide nanoparticles: a multimodal contrast agents for in vivo imaging. J. Am. Chem. Soc. 129, 5076−5081. (13) McDonald, M. A., and Watkin, K. L. (2006) Investigations into the physicochemical properties of dextran small particulate gadolinium oxide nanoparticles. Acad. Radiol. 13, 421−426. (14) Park, J. Y., Baek, M. J., Choi, E. S., Woo, S., Kim, J. H., Kim, T. J., Jung, J. C., Chae, K. S., Chang, Y., and Lee, G. H. (2009) Paramagnetic ultrasmall gadolinium oxide nanoparticles as advanced T1MRI contrast agent: account for large longitudinal relaxivity, optimal particle diameter, and in vivo T1MR images. ACS Nano 3, 3663−3668. (15) Cheung, E. N., Alvares, R. D. A., Oakden, W., Chaudhary, R., Hill, M. L., Pichaandi, J., Mo, G. C. H., Yip, C., Macdonald, P. M., Stanisz, G. J., van Veggel, F. C., and Prosser, R. S. (2010) Polymerstabilized lanthanide fluoride nanoparticle aggregates as contrast agents for magnetic resonance imaging and computed tomography. Chem. Mater. 22, 4728−4733.

■

CONCLUSIONS Using surface modification chemistry previously demonstrated on zirconium phosphonate surfaces, it is shown that GdPO4·H2O nanoparticles can be easily functionalized with DNA probes. The key interaction is coordinate covalent binding of the oligonucleotide terminal phosphate to active Gd3+ sites on the surface of the particles. The single-step binding of phosphate-terminated oligonucleotides avoids complicated surface modification schemes that involve multiple synthetic steps. The binding is achieved in one step, in biological conditions, and in the case of a DNA oligonucleotide, the bioactivity is retained. The MR relaxivity measurements confirmed that the DNA-modified GdPO4·H2O nanoparticles have relaxivities comparable to those of commercially available contrast agents and that the biofunctionalization of the surface of the nanoparticles does not prevent their function as MRI contrast agents.

■

ASSOCIATED CONTENT

S Supporting Information *

Details of the magnetic measurement and the magnetic susceptibility, χ, as a function of temperature, T; titration of free Gd3+ with xylenol orange; and low field relaxivity measurements on GdPO4NP-DNA. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*Phone: (352) 392-9016. Fax (352) 392-3255. E-mail: [email protected]fl.edu. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported, in part, by the U.S. National Science Foundation and the National Institutes of Health through grants CHE-0957155 (to D.R.T.), DMR-0701400 (to M.W.M.) and P01HL059412 (to G.A.W.). The MRI work was performed 956



Article

(16) Gerion, D., Herberg, J., Bok, R., Gjersing, E., Ramon, E., Maxwell, R., Kurhanewicz, J., Budinger, T. F., Gray, J. W., Shuman, M. A., and Chen, F. F. (2007) Paramagnetic silica-coated nanocrystals as an advanced MRI contrast agent. J. Phys. Chem. C 111, 12542−12546. (17) Rieter, W. J., Taylor, K. M. L., An, H., Lin, W., and Lin, W. (2006) Nanoscale metal-organic frameworks as potential multimodal contrast enhancing agents. J. Am. Chem. Soc. 128, 9024−9029. (18) Hifumi, H., Yamaoka, S., Tanimoto, A., Akatsu, T., Shindo, Y., Honda, A., Citterio, D., Oka, K., Kuribayashi, S., and Suzuki, K. (2009) Dextran coated gadolinium phosphate nanoparticles for magnetic resonance tumor imaging. J. Mater. Chem. 19, 6393−6397. (19) Hifumi, H., Yamaoka, S., Tanimoto, A., Citterio, D., and Suzuki, K. (2006) Gadolinium-based hybrid nanoparticles as a positive MR contrast agent. J. Am. Chem. Soc. 128, 15090−15094. (20) Fang, Y. P., Xu, A. W., Song, R. Q., Zhang, H. X., You, L. P., Yu, J. C., and Liu, H. Q. (2003) Systematic synthesis and characterization of single-crystal lanthanide orthophosphate nanowires. J. Am. Chem. Soc. 125, 16025−16029. (21) Lessing, P. A., and Erickson, A. W. (2003) Synthesis and characterization of gadolinium phosphate neutron absorber. J. Eur. Ceramic Soc. 23, 3049−3054. (22) Huang, Y.-F., Sefah, K., Bamrungsap, S., Chang, H.-T., and Tan, W. (2008) Selective photothermal therapy for mixed cancer cells using aptamer-conjugated nanorods. Langmuir 24, 11860−11863. (23) Medley, C. D., Smith, J. E., Tang, Z., Wu, Y., Bamrungsap, S., and Tan, W. (2008) Gold nanoparticle-based colorimetric assay for the direct detection of cancerous cells. Anal. Chem. 80, 1067−1070. (24) Villaraza, L. A. J., Bumb, A., and Brechbiel, M. W. (2010) Macromolecules, dendrimers, and nanomaterials in magnetic resonance imaging: the interplay between size, function, and pharmacokinetics. Chem. Rev. 110, 2921−2924. (25) Cinier, M., Petit, M., Williams, M. N., Fabre, R. M., Pecorari, F., Talham, D. R., Bujoli, B., and Tellier, C. (2009) Bisphosphonate adaptors for specific protein binding on zirconium phosphonate-based microarrays. Bioconjugate Chem. 20, 2270−2276. (26) Lane, S., Monot, J., Petit, M., Bujoli, B., and Talham, D. (2007) XPS investigation of DNA binding to zirconium-phosphonate surfaces. Colloid. Surf. B 58, 34−37. (27) Monot, J., Petit, M., Lane, S. M., Guisle, I., Lager, J., Tellier, C., Talham, D. R., and Bujoli, B. (2008) Towards zirconium phosphonatebased microarrays for probing DNA-protein interactions: critical influence of the location of the probe anchoring groups. J. Am. Chem. Soc. 130, 6243−6247. (28) Hutton, C., Bork, A., Josephs, O., Deichmann, R., Ashburner, J., and Turner, R. (2002) Image distortion correction in fMRI: a quantitative evaluation. NeuroImage 16, 217−240. (29) Huo, Z. Y., Chen, C., Chu, D., Li, H. H., and Li, Y. D. (2007) Systematic synthesis of lanthanide phosphate nanocrystals. Chem. Eur. J. 13, 7708. (30) Parreu, I., Sola, R., Gavalda, J., Massons, J., Diaz, F., and Aguila, M. (2005) Crystal growth, structural characterization, and linear thermal evolution of KGd(PO3)4. Chem. Mater. 17, 822−826. (31) Terreno, E., Castelli, D. D., Viale, A., and Aime, S. (2010) Challenges for molecular magnetic resonance imaging. Chem. Rev. 110, 3019−3046. (32) Wang, H., Yang, R., Yang, L., and Tan, W. (2009) Nucleic acid conjugated nanomaterials for enhanced molecular recognition. ACS Nano 3, 2451−2460. (33) Tang, Z., Parekh, P., Turner, P., Moyer, R. W., and Tan, W. (2009) Generating aptamers for recognition of virus-infected cells. Clin. Chem. 55, 813−822.

957


DNA Surface Modified Gadolinium Phosphate Nanoparticles as MRI

Recommend Documents