A Modular Monolayer Coating Enables Cell Targeting by Luminescent

Feb 18, 2009 - R. Hari Krishna , B. M. Nagabhushana , H. Nagabhushana , N. Suriya Murthy , S. C. Sharma , C. Shivakumara , and R. P. S. Chakradhar...
0 downloads 0 Views 590KB Size
Bioconjugate Chem. 2009, 20, 437–439

437

A Modular Monolayer Coating Enables Cell Targeting by Luminescent Yttria Nanoparticles Christopher A. Traina, T. Joseph Dennes, and Jeffrey Schwartz* Department of Chemistry, Princeton University, Princeton, New Jersey 08544-1009. Received December 18, 2008; Revised Manuscript Received February 7, 2009

Luminescent Eu3+-doped Y2O3 nanoparticles are functionalized for cell targeting using a modular, multisegmented approach based on a phosphonate monolayer platform. The first segment provides hydrolytic stability for the particle-organic interface; the second enables aqueous suspendability; the third is used to bond cell attachment molecules. In vitro imaging experiments showed enhanced cell attachment of activated nanoparticles conjugated with cell attachment peptides compared to control nanoparticles. Peptide-derivatized nanoparticles are not displaced from the cells by their soluble peptide analogue, which suggests strong, polyvalent cell-particle interactions.

Effective biotargeting luminescent nanoparticles must be nontoxic, small relative to the system they are imaging, stable over the course of the experiment, and able to efficiently bond to the desired entity. Rare earth ion-doped yttria (Y2O3) nanoparticles are an appealing class of such targeting agents: they are benign in vitro (1) and in vivo (2) and can be made with diameters of 12 days; thus, 1 is used as our first module. Treating 1 with p-nitrophenyl chloroformate readily gives activated p-nitrophenyl carbonate-derivatized 2, and the second module is introduced by reaction with tri(ethylene glycol) monoamine to give 3; 3 has good aqueous suspendability and is also stable to hydrolysis at physiological pH. Treating 3 with p-nitrophenyl chloroformate gives 4, which can then bond the desired biotargeting molecule via reaction with a primary amino group (see Scheme 1). In a typical procedure, Y2O3 powder (ca. 1 g; Aldrich) was dispersed by sonication in 60 mL THF. A solution of * E-mail: [email protected].

11-hydroxyundecylphosphonic acid (10) (5; 12 mg, 4.76 × 10-5 mol, in 25 mL THF) was added to the suspension, and the reaction mixture was stirred for 3 h at room temperature. The powder was collected by centrifugation, washed to remove excess phosphonic acid, and then dried in vacuo to give 1, which was then resuspended in 60 mL THF. p-Nitrophenyl chloroformate (80 mg) and diisopropylethylamine (0.75 mL) were added, and the mixture was stirred for 15 h. The powder was then recovered by centrifugation and washed to give 2 (νCdO ) 1768 cm-1). It was then treated with tri(ethylene glycol) monoamine (35 mg) and diisopropylethylamine (0.75 mL) in THF with stirring for 20 h. The powder was collected by centrifugation, and excess reagents were removed to give 3, which was then treated with p-nitrophenyl chloroformate (80 mg) as described above. The product was then washed to give 4 (νCdO ) 1768; 1700 cm-1). To determine the surface loading of the two modules on Y2O3 particles, a fluorescent dye was coupled to 4, cleaved, and quantified in solution (11); in this way, the surface loading was measured to be 0.25 ( 0.01 mol/cm2. Full synthesis and characterization details are given as Supporting Information. As a first step to determining their biocompatibility, cell culture studies using NIH 3T3 fibroblasts were performed to probe cell adhesion on untreated and treated Y2O3 particles. Using a standard IR pellet press, aliquots of 4 were pressed into disks (d ) 12 mm, h ) 1.5 mm) that were then kept in a solution of Gly-Arg-Gly-Asp-Ser-Pro-Lys (GRGDSPK; 20 mg in 40 mL deionized water adjusted to pH 8-8.5 with aqueous NaOH) for 24 h; peptides containing the RGDS sequence are well-known models for the cell recognition site in fibronectin (12). The disks were then removed and rinsed with aqueous NaOH (pH 8.5), deionized water, then methanol to give peptideterminated 6a (see Scheme 1). Three surfaces were examined for cell adhesion: unmodified Y2O3, 3, and 6a. Samples of these particles were pressed into disks and put into cell culture (Supporting Information Figure 2). Quantification of cell adhesion for each surface was assessed by counting the number of cells in randomly generated microscope fields (Figure 1; 10× magnification, minimum of six fields for each type of sample; for cell counting data, see Supporting Information Figure 3). Unmodified Y2O3 supported some cell adhesion; importantly, the number of adherent cells was reduced by almost half on 3 and was doubled on 6a. A two-way analysis of variance (ANOVA) showed a statistically significant difference between

10.1021/bc800551x CCC: $40.75  2009 American Chemical Society Published on Web 02/18/2009

438 Bioconjugate Chem., Vol. 20, No. 3, 2009

Communications

Figure 1. Cell adhesion images for pelletized Y2O3: untreated (left), 3 (middle), 6a (right); cell counting data are given in Supporting Information.

Figure 2. TEM image of Y2O3/Eu3+nanoparticles shows diameters of ca. 100 nm.

the mean number of cells attached to RGD-coated and control pellets (for untreated Y2O3 vs 3 p ) 0.02; for untreated Y2O3 vs 6a, p ) 0.001; for 3 vs 6a, p ) 0.0001). To demonstrate the utility of this modular approach for bioimaging purposes, bioconjugation of luminescent nanoparticles was investigated. Eu3+-doped (5%) Y2O3 nanoparticles were synthesized by precipitation of Y(OH)CO3 particles from a refluxing, aqueous solution of Y3+ and Eu3+ salts and urea (13); calcining at 785 °C for 1 h gave cubic Y2O3:Eu3+ (by powder X-ray diffraction (13)). Particles with diameters ranging from ∼40 to 180 nm (by TEM) could be obtained by adjustment of reagent concentrations; those with diameters between 80 and 120 nm (Figure 2) were used for subsequent experimentation. A strong band at 612 nm and less intense peaks at 591 and 630 nm were recorded in the particle emission spectrum and are characteristic of Eu3+ luminescence (14). Importantly, these

nanoparticles can be easily visualized using a standard fluorescence microscope under UV (∼365 nm) or blue (∼488 nm) excitation. The luminescent yttria nanoparticles were sonicated in THF to break up any agglomeration prior to and during monolayer deposition of 5 to give 1, which was derivatized with tetra(ethylene glycol) monoamine and then GRGDSPK to give 6a as described above. Similarly prepared control 6b used cell adhesion-inactive GRADSPK (12) as a control. Suspensions of 6a and 6b were introduced into separate fibroblast cell culture dishes (105 cells/mL; cells were prestained with cell-tracker dye) which had a confluent cell layer on the plating substrate and the particles were allowed to attach for 3 h; the cells were then washed to remove unbound particles, fixed, and visualized by fluorescence microscopy. There was no change in the emission spectra for 6a- or 6b-treated vs untreated nanoparticles; only the doped nanoparticles were visible under UV (365 nm) illumination, but blue (488 nm) irradiation excited both dyestained cells and nanoparticles. Images of cells treated with control 6b showed only some large particle aggregates but no specific particle binding to the cells. In contrast, 6a showed cell attachment with little aggregation (see Supporting Information Figure 8). Since 6a and 6b have similar wetting properties and charges, it is unlikely that they would have different aggregative properties in the absence of cell binding opportunities. Nanoparticle-cell binding was further distinguished from simple precipitation using a nonconfluent cell layer (∼50% substrate coverage) that was treated with 6a or 6b. Most of the

Figure 3. Nonspecific adsorption/aggregation of RAD-modified nanoparticles 6b (left) versus adherence to cells (green) by RGD-modified nanoparticles 6a (right) in nonconfluent cell layers. The nanoparticles luminesce red. Scheme 1. The Modular Approach to Surface Functionalization of Undoped and Doped Yttria Particles (Red Spheres)

Communications

cells treated with 6a were covered with nanoparticles but very little 6a deposited on the substrate between the cells; few cells exposed to 6b were bound to particles (Figure 3). Cell binding by 6a was not inhibited by soluble GRGDSPK. RGD-integrin binding is reversible (15), but due to their size, particles of 6a cannot be resuspended after cell interaction, which inhibits displacement by the soluble peptide. Importantly for cellular targeting, nanoparticles of 6a may act as polyvalent binding species (16), which would greatly enhance cell binding of the bioconjugated particles vs the soluble peptide: On the basis of calculated surface loadings of the two modules on the Y2O3 particles, and assuming an average particle diameter of ∼100 nm, each nanoparticle can have >104 surface-attached peptide molecules on it, thus making it possible for each particle to bind simultaneously to several receptor sites on the cell surface. We have shown that the Y2O3 surface can be easily and reproducibly functionalized in high yield via a modular scheme based on organophosphonate monolayers that enables control of nanoparticle surface properties. The nanoparticles can be made more cell adhesive or more cell resistant than the native ones by bonding an RGD-containing peptide or an oligo(ethylene glycol) moiety, respectively. This modular scheme was used to derivatize luminescent, Eu3+-doped nanoparticles with an RGD peptide; in vitro experiments showed that increased cell surface-nanoparticle binding interactions could be visualized, and this binding was not suppressed by added soluble peptide. The modular concept may be expanded to examine peptide-receptor specificities; a small library of differently “colored” nanoparticles, each terminated with a different peptide sequence, might in principle be used to identify the presence of peptide-ligand specific cell surface receptor sites.

ACKNOWLEDGMENT The authors thank the National Science Foundation, CHE0612572, for support of this research. They also thank Prof. Jean E. Schwarzbauer, Department of Molecular Biology, Princeton University, for critical assistance. Supporting Information Available: Full experimental details for synthesis of 1, 2, 3, 4, and 6; a calibration curve for measuring surface loading of 4; calculation of surface loading; cell adhesion cartoon and counting data; IR spectra for 1, 2, 3, and 4; images of nanoparticles of 6a and 6b on a confluent layer of cells. This material is available free of charge via the Internet at http://pubs.acs.org.

LITERATURE CITED (1) Schubert, D., Dargusch, R., Raitano, J., and Chan, S.-W. (2006) Cerium and yttrium oxide nanoparticles are neuroprotective. Biochem. Biophys. Res. Commun. 342, 86–91.

Bioconjugate Chem., Vol. 20, No. 3, 2009 439 (2) Lim, S. F., Riehn, R., Ryu, W. S., Khanarian, N., Tung, C.-k., Tank, D., and Austin, R. H. (2006) In Vivo and Scanning Electron Microscopy Imaging of Upconverting Nanophosphors in Caenorhabditis elegans. Nano Lett. 6, 169–174. (3) Qin, X., Yokomori, T., and Ju, Y. (2007) Flame synthesis and characterization of rare-earth (Er3+, Ho3+, and Tm3+) doped upconversion nanophosphors. Appl. Phys. Lett. 90, 073104-1-3. (4) Maas, H., Currao, A., and Calzaferri, C. (2002) Encapsulated lanthanides as luminescent materials. Angew. Chem., Int. Ed. 41, 2495–2497. (5) Michalet, X., Pinaud, F., Lacoste, T. D., Dahan, M., Bruchez, M. P., Alivisatos, A. P., and Weiss, S. (2001) Properties of fluorescent semiconductor nanocrystals and their application to biological labeling. Single Mol. 2, 261–276. (6) He, C., Guan, Y., Yao, L., Cai, W., Li, X., and Yao, Z. (2003) Synthesis and photoluminescence of nano-Y2O3:Eu3+ phosphors. Mater. Res. Bull. 38, 973–979. (7) Matsuura, D. (2002) Red, green, and blue upconversion luminescence of trivalent-rare-earth ion-doped Y2O3 nanocrystals. Appl. Phys. Lett. 81, 4526–4528. (8) Texier, I., Heinrich, E., Berger, M., Tillement, O., Louis, C., and Peltie, P. (2007) Luminescent up-converting nano-crystals for in vivo imaging. Proc. SPIE 6449, 64490D-1–64490D-11. (9) Traina, C. A., and Schwartz, J. (2007) Surface modification of Y2O3 nanoparticles. Langmuir 23, 9158–9161. (10) Putvinski, T. M., Schilling, M. L., Katz, H. E., Chidsey, C. E. D., Mujsce, A. M., and Emerson, A. B. (1990) Selfassembly of organic multilayers with polar order using zirconium phosphate bonding between layers. Langmuir 6, 1567–1571. (11) Silverman, B. M., Wieghaus, K. A., and Schwartz, J. (2005) Comparative stabilities of siloxane vs. phosphonate monolayers on a key titanium alloy. Langmuir 21, 225–228. (12) Pierschbacher, M. D., and Ruoslahti, E. (1984) Variants of the cell recognition site of fibronectin that retain attachmentpromoting activity. Proc. Natl. Acad. Sci. U.S.A. 81, 5985–5988. (13) Kaczmarek, W. A., and Riesen, H. (2006) Structural and morphological properties of RE3+ doped sesquioxide Y2O3 spherical nanoparticles. J. Mater. Sci. 41, 8320–8328. (14) Pang, Q., Shi, J., Liu, Y., Xing, D., Gong, M., and Xu, N. (2003) A novel approach for preparation of Y2O3:Eu3+ nanoparticles by microemulsion-microwave heating. Mater. Sci. Eng., B 103, 57–61. (15) Ruoslahti, E., and Pierschbacher, M. D. (1987) New perspectives in cell adhesion: RGD and integrins. Science 238, 491– 497. (16) Mammen, M., Choi, S.-K., and Whitesides, G. M. (1998) Polyvalent interactions in biological systems: implications for design and use of multivalent ligands and inhibitors. Angew. Chem., Int. Ed. 37, 2754–2794. BC800551X