Polymer-Modified Gadolinium Metal-Organic Framework

Biomacromolecules 2009, 10, 983–993

983

Polymer-Modified Gadolinium Metal-Organic Framework Nanoparticles Used as Multifunctional Nanomedicines for the Targeted Imaging and Treatment of Cancer Misty D. Rowe,† Douglas H. Thamm,‡ Susan L. Kraft,‡ and Stephen G. Boyes*,† Department of Chemistry and Geochemistry, Colorado School of Mines, Golden, Colorado 80401, and College of Veterinary Medicine and Biological Sciences, Animal Cancer Center, Colorado State University, Fort Collins, Colorado 80526 Received January 9, 2009; Revised Manuscript Received January 28, 2009

Novel nanoscale theragnostic devices were successfully prepared through attachment of well defined, multifunctional polymer chains to gadolinium (Gd) metal-organic framework (MOF) nanoparticles. Copolymers of poly(Nisopropylacrylamide)-co-poly(N-acryloxysuccinimide)-co-poly(fluorescein O-methacrylate) were prepared via reversible addition-fragmentation chain transfer (RAFT) polymerization. The succinimide functionality was utilized as a scaffold for attachment of both a therapeutic agent, such as methotrexate, and a targeting ligand, such as H-glycine-arginine-glycine-aspartate-serine-NH2 peptide. Employment of a trithiocarbonate RAFT agent allowed for reduction of the polymer end groups to thiolates providing a means of copolymer attachment through vacant orbitals on the Gd3+ ions at the surface of the Gd MOF nanoparticles. These versatile, nanoscale scaffolds were shown to be biocompatible and have cancer cell targeting, bimodal imaging, and disease treatment capabilities. This unique method provided a simple yet versatile route of producing polymer-nanoparticle theragnostic materials with an unprecedented degree of flexibility in the construct, potentially allowing for tunable loading capacities and spatial loading of targeting/treatment agents, while incorporating bimodal imaging capabilities through both magnetic resonance and fluorescence microscopy.

Introduction As of late, there has been an explosion in the development of nanomedicine platforms for application in molecular imaging and drug delivery.1-5 Nanoscale theragnostic systems that incorporate molecular targeting, therapeutic agents, and diagnostic imaging capabilities are emerging as the next generation of personalized medicines and have the potential to dramatically improve the therapeutic outcome of drug therapy.2,3,5,6 While there is almost unanimous agreement that these next generation, personalized nanomedicines will provide clinically important theragnostic devices, they have yet to reach clinical realization. Arguably, the primary reasons limiting application of these devices are poor design and manufacturing techniques.1,2,7,8 Thus, new nanomedicine platforms must be developed for the successful preparation of nanoscale theragnostic devices. One of the most promising platforms for the formation of new nanoscale theragnostic devices are nanoparticles, where the particle acts as the imaging component of the multifunctional theragnostic device. This is particularly beneficial due to the fact that a large concentration of imaging agent can be delivered to the desired location per targeted biorecognition event. However, in order to produce a nanoparticle theragnostic device that can be easily translated to clinical application, it is important that the imaging component of the nanoparticle be useful for application with common diagnostic imaging instrumentation, such as magnetic resonance imaging (MRI). Currently, MRI techniques employ high spin paramagnetic metals, such as iron (Fe) or gadolinium (Gd), as contrast agents. Fe-based contrast * To whom correspondence should be addressed. E-mail: sboyes@ mines.edu. † Colorado School of Mines. ‡ Colorado State University.

agents typically induce a large shortening of the transverse relaxation time (T2) and a high transverse relaxivity (r2), leading to a darkening effect and are thus called negative contrast agents, while Gd-based contrast agents produce a large shortening of the longitudinal relaxation time (T1) and high longitudinal relaxivity (r1) and are called positive contrast agents.9,10 Relaxivity values, r, are simply defined as the inverse of the relaxation time with respect to the contrast agent concentration. The ratio of r2/r1 is used to provide information about the contrast agent. Typically, r2/r1 values below 2 show brightening in T1-weighted images, providing a positive contrast agent.10 As there is a preference for the use of positive contrast agents at the clinical level, due to their wider dynamic range, contrast agents based on Gd3+ are the most widely used. However, Gd3+ is highly toxic to cells; as such for MRI application, the toxicity has been overcome by utilizing chelates to increase the stability and compatibility of the metal ion. Over the past few years, there has been an increasing focus on the use of nanoparticle based contrast agents. This attention has been driven by limitations with conventional contrast agents based on chelates of metal ions, such as a low concentration of metal ion per molecule, short retention times in vivo, and difficulty in functionalization to enable use in nanovectors.9,11-13 Most of the focus on nanoparticle contrast agents has been on the use of superparamagnetic iron oxides (SPIOs), which act as negative contrast agents. However, negative contrast agents, based on SPIOs, suffer from a series of drawbacks, including that MRI cannot distinguish the void between the contrast agent and the other signal voids, negative contrast agents are limited by partial volume effects, and tracking cells in vivo can be difficult.14,15 Recently, researchers have reported procedures for the preparation of nanoparticles containing Gd.10,13,16-23 These nanoparticles have been synthesized with both inorganic and

10.1021/bm900043e CCC: $40.75  2009 American Chemical Society Published on Web 03/16/2009

984

Biomacromolecules, Vol. 10, No. 4, 2009

Rowe et al.

Figure 1. Polymer-modified Gd metal-organic framework nanoparticles as a nanomedicine construct for targeted imaging and treatment of cancer.

organo-metallic compounds of Gd such as gadolinium oxide, gadolinium phosphate, gadolinium fluoride, gadolinium hexanedione, and acetylacetenoate mixed with emulsifying wax, and most recently, metal-organic frameworks (MOFs).10,13,16-24 Research into Gd nanoparticles has been focused on overcoming issues related to conventional contrast agents and on taking advantage of the favorable properties of the nanoparticles, such as improved relaxation times. While Gd nanoparticles exhibit relaxivities significantly higher than typical Gd chelates and also provide a contrast agent with higher molecular weights for improved retention times and a high concentration of Gd3+ ions per contrast agent particle, their application has been limited due to the difficulty in producing nanoparticles that are biocompatible, stable, and have specific surface functionality.10,17 Due to the strong potential of Gd nanoparticles as theragnostic nanodevices, researchers have attempted to overcome these limitations by developing methods to surface modify the nanoparticles.20,22,23 To date, these surface modification methods have demonstrated limited success, as they have resulted in poorly defined surfaces with lack of control over the functionality of the surface of the nanoparticle, instabilities in the coatings due to their noncovalent nature, or reduced imaging capabilities due to masking of the underlying Gd nanoparticle. As such, the search for a surface modification technique that provides control over surface functionality and architecture, the ability to incorporate a wide range of both targeting ligands and therapeutics, and produce a stable structure without diminishing the inherent imaging properties represents a significant challenge for researchers in the search for next generation nanomedicines. In this research, Gd MOF nanoparticles were surface modified by the covalent attachment of well-defined polymers containing both a targeting ligand and antineoplastic agent to produce a novel theragnostic nanodevice (Figure 1). Initially, biocompatible random copolymers composed of N-isopropylacrylamide (NIPAM), N-acryloxysuccinimide (NAOS), and fluorescein O-methacrylate (FMA) were synthesized via reversible additionfragmentation chain transfer (RAFT) polymerization techniques employing a trithiocarbonate RAFT agent (Scheme 1). RAFT polymerization shows great promise in the synthesis of multifunctional polymers due to the versatility of monomer selection and polymerization conditions, along with the ability to produce well-defined, narrow polydispersity index (PDI) polymers with both simple and complex architectures and a high degree of end-group control.25,26 Addition of FMA into the backbone of the copolymer allowed for fluorescent tagging of the copolymer and subsequent use as a cellular-scale imaging moiety via fluorescence microscopy. Incorporation of NAOS into the

Scheme 1. Modification of PNIPAM-co-PNAOS-co-PFMA Copolymer with Methotrexate and GRGDS-NH2 Residues

copolymer provided a means of tailoring our copolymer backbone with both targeting ligands, such as H-glycinearginine-glycine-aspartate-serine-NH2 (GRGDS-NH2) peptide motifs, and antineoplastic drugs, such as methotrexate (MTX). The enhanced reactivity of the succinimide group gives these copolymer structures the potential to be tailored with a wide

Gd Nanoparticles as Multifunctional Nanomedicines

range of targeting-treatment combinations. Additionally, the inherent flexibility of RAFT polymerizations makes it an ideal candidate to produce well-defined polymer structures, with controllable molecular weights, capable of providing increased therapeutic/targeting agent loading and loading efficiency.27,28

Experimental Section Materials. All chemicals were purchased from Sigma-Aldrich, with the exception of the H-glycine-arginine-glycine-aspartate-serine-NH2 (GRGDS-NH2) peptide motif, which was purchased from AnaSpec. Tissue culture media and reagents were purchased from Fisher Scientific. NIPAM was doubly recrystallized in hexanes before use. Azobisisobutyronitrile (AIBN) was doubly recrystallized from methanol prior to use. Triethylamine was distilled under pressure and stored in the freezer prior to use. All other chemicals were used as received. Synthesis of S-1-Dodecyl S′-(r,r-Dimethylacetic acid) Trithiocarbonate (DATC). DATC was prepared via Lai et al.’s literature procedure.29 1H NMR (60 MHz, CDCl3, δ, ppm): 0.89 (t, 3H), 1.25-1.53 (m, 20H), 1.73 (s, 6H), 3.35 (t, 2H). FTIR (cm-1): 1702 (CdO), 1065 (CdS). PNIPAM-co-PNAOS-co-PFMA Copolymer Synthesis via RAFT Polymerization. All copolymers were synthesized with 33 wt % monomer in dioxane at 60 °C via RAFT polymerization techniques utilizing a 18.75:1 molar ratio of DATC, as the RAFT agent, to AIBN, as the initiator. NIPAM comprised the majority of the copolymer backbone, with NAOS being incorporated into the backbone at 10, 17, 25, 33, and 50 wt % loading to provide a reactive sight for attachment of targeting ligands and therapeutic agents. FMA, at approximately 0.5 wt % loading, was added near the end of the polymerization, allowing for fluorescent tagging of the biocopolymer for subsequent imaging capabilities in fluorescent microscopy. For example, to produce PNIPAM-co-PNAOS-co-PFMA with a 10 wt % loading of NAOS, anhydrous dioxane (21.9 mL), NIPAM (7.00 g, 61.9 mmol), NAOS (0.700 g, 4.14 mmol), and DATC (0.340 g, 0.931mmol) were added to a 150 mL Schlenk flask equipped with a stir bar. The flask was sealed and the solution was gently degassed for 45 min and then left under a high-purity nitrogen atmosphere. The flask was allowed to stir at room temperature until the monomer and DATC were completely dissolved. To a second 150 mL Schlenk flask equipped with a stir bar was added AIBN (0.0082 g, 0.050 mmol). This flask was sealed with a rubber septum, subjected to three evacuationnitrogen purge cycles and left under a nitrogen atmosphere. The monomer solution was then transferred via cannula to the initiator containing flask. The reaction was then heated for 24 h at 60 °C, at which time a 2 mL sample was taken by syringe for analysis. To a 25 mL Schlenk tube equipped with a stir bar, anhydrous dioxane (10 mL) and fluorescein O-methacrylate (0.035 g, 0.087 mmol) were added. The tube was sealed and the solution was gently degassed for 45 min and then left under a high-purity nitrogen atmosphere. The monomer solution was then transferred via cannula to the PNIPAM-co-PNAOS copolymer containing flask. The reaction was then heated for an additional 6 h at 60 °C to allow the addition of PFMA to the PNIPAMco-PNAOS copolymer. Polymer was isolated from the solution by evaporating residual solvent under vacuum at 40 °C overnight. Polymer was then purified via aqueous extraction to remove residual monomer. Attachment of GRGDS-NH2 and MTX to PNIPAM-co-PNAOSco-PFMA Copolymer. The PNIPAM-co-PNAOS-co-PFMA copolymers were subsequently reacted through a condensation reaction of the succinimide groups with primary amine groups of the targeting ligand and/or therapeutic. Reactions were carried out at room temperature in deuterated N,N-dimethylsulfoxide (DMSO) with 0.01 M triethylamine and stirring at room temperature for 24 h. Unreacted targeting and/or therapeutic agents were then removed via silica column chromatography and aqueous/solvent extraction. MTX and GRGDS-NH2 were incorporated into the biocopolymer backbone at 10, 25, and 33 wt % loading capacity.


985

Formation of Gd MOF Nanoparticles. Gd MOF nanoparticles were synthesized via a reverse microemulsion system developed by Reiter and co-workers employing 0.05 M gadolinium(III) chloride, 0.05 M cetyltrimethylammonium bromide (CTAB), as the surfactant, and 0.075 M 1,4-benzenedicarboxylic acid methyl ammonium salt (1,4-BDC), as the chelating agent.16 A water to surfactant ratio of 10:1 was utilized to provide nanoplatelet shapes with an average of 20-25 nm in width and 100-150 nm in length. After 24 h, the microemulsion was centrifuged at 5000 rpm for a total of 30min. After removal of the supernatant the particles were washed with ethanol (15 mL), sonicated, and then recentrifuged for 30min. The supernatant was discarded. The particles underwent an additional cycle of dispersement in ethanol (15 mL), sonication, and centrifugation to remove any excess reactants. The nanoparticle solution was resuspended in ethanol and then allowed to dry. Surface Modification of Gd MOF Nanoparticles with RAFT Copolymers. In a typical experiment, one of the following RAFT copolymers: (1) unmodified PNIPAM-co-PNAOS-co-PFMA, (2) PNIPAM-co-PNAOS-co-PFMA tailored with MTX, or (3) PNIPAMco-PNAOS-co-PFMA tailored with MTX and GRGDS-NH2 (0.1 g) was added to 25 mL of anhydrous N,N-dimethylformamide (DMF) in a 150 mL Schlenk flask equipped with a stir bar and then sealed with a rubber septum. The RAFT copolymer solution was purged with high purity nitrogen and then subsequently left under a nitrogen atmosphere. The RAFT agent terminated copolymer was then converted to a thiolate terminated copolymer, through aminolysis, by the addition of 0.075 M hexylamine (0.45 mL) and stirring for 1 h at room temperature. Gd MOF nanoparticles (0.01 g) were suspended in an additional 25 mL of DMF in a second 150 mL Schlenk flask equipped with a stir bar, then sealed with a rubber septum. The nanoparticle solution was then purged with high purity nitrogen for 30min and was subsequently left under a nitrogen atmosphere. The polymer solution was then transferred via cannula to the Gd MOF nanoparticle solution. The resulting solution was allowed to stir for 24 h at room temperature under a nitrogen atmosphere. Untethered polymer was removed from the polymermodified Gd MOF nanoparticles through repeated centrifugation and resuspension in DMF (2×) and ethanol (2×), followed by drying. Cell Growth Inhibition Studies. Initial cell growth inhibition studies were performed using FITZ-HSA, an Rvβ3 integrin expressing canine endothelial sarcoma cell line, developed in the laboratory of D. Thamm. The cells were maintained in tissue culture using Minimum Essential Media (MEM) supplemented with nonessential amino acids, Lglutamine, MEM vitamins, penicillin/streptomycin, sodium pyruvate, and 10% fetal bovine serum (C/10). Cells were maintained under standard conditions (37 °C, 5% CO2 in a humidified atmosphere) and passaged by trypsinization. Two × 103 cells were seeded in 96-well plate in C/10 medium and allowed to adhere overnight. The following day, the plates were washed and varying concentrations of drug were added. The cells were then incubated for 72 h. Relative viable cell number was measured using a bioreductive fluorescent assay according to manufacturer directions using a multidetection microplate reader. Fluorescence Imaging. Initial cell imaging studies were performed employing FITZ-HSA tumor cells at the Animal Cancer Center at Colorado State University. Copolymer-modified Gd MOF nanoparticles tailored with or without MTX and GRGDS-NH2 were incubated with FITZ-HSA for 1, 4, and 24 h at 37 °C in standard culture medium containing 10% PBS in a 5% CO2 atmosphere. Cells were then extensively washed, fixed, and stained with propidium iodide for fluorescence imaging. Images were acquired using a microscope fitted with FITC and Texas Red filter sets. In Vitro Relaxivity through Measurements by MRI. Preliminary characterization of the relaxation properties of the unmodified Gd MOF nanoparticles and tailored Gd MOF nanoparticles was accomplished through T1 and T2 relaxation measurements made in vitro with a direct comparison to the clinically utilized contrast agents, Magnevist and Multihance. Measurements were made on a 1.5 T scanner. Scanning for T1 calculations were done by spin echo imaging using repetition

986


time (TR) values of 10000, 5000, 2500, 1000, 500, 250, 150, 100, 50, 30, 25 ms, and minimum echo time (TE), 5 nm slice thickness, 128 × 128 matrix. Scanning for T2 calculations were done by spin echo imaging using TR of 1500, and four different TE values of 15, 30, 45, and 60 ms, 5 nm slice thickness, 128 × 128 matrix. All samples, including standard contrast agents, were serially diluted in deionized ultrafiltered (DIUF) water, degassed with high purity nitrogen, and sealed in polypropylene vials. Analysis was performed by acquiring signal intensity (I) measurements via region-of-interest analysis of the samples for all pulse sequences with T1 and T2 values being calculated using Ii ) Io,I (1 - exp(-t/Ti)). Dialysis Purification and Stability Studies of Unmodified and Polymer-Modified Gd MOF Nanoparticles. In a typical procedure, the unmodified and polymer-modified Gd MOF nanoparticles were first purified as discussed earlier. The unmodified and polymer-modified Gd MOF nanoparticles were then suspended in a hydrochloric acid solution (pH ) 3), sonicated for 10 min, and then centrifuged for 30 min, after which the supernatant was discarded. The unmodified and polymer-modified Gd MOF nanoparticles were resuspended in the acidic solution, sonicated for an additional 10 min, and then centrifuged once more. The supernatant was then discarded and replaced with DIUF water. The unmodified and polymer-modified Gd MOF nanoparticles were then sonicated for 10 min and centrifuged for 30 min before the supernatant was discarded. Next, in six congruent experiments, the unmodified and polymer-modified Gd MOF nanoparticles were dialyzed (molecular weight cut off of 3500 Dalton) against DIUF water, 0.09 wt % sodium chloride isotonic solution, and phosphate-buffered saline (PBS) solution (pH ) 7.4) at 37 °C for 1 week, with the aqueous dialysate solution being changed at several time increments along the experiment. The aqueous dialysate was then acidified in a 1% nitric acid solution before inductively coupled plasma-atomic emission spectroscopy (ICP-AES) was utilized to determine the concentration of Gd3+ leached to the dialysate from the unmodified and polymermodified Gd MOF nanoparticles. In each experiment the contribution of the carrier solution was subtracted to baseline the data. Characterization. Gel permeation chromatography (GPC) was used to determine absolute molecular weights and molecular weight distributions of the PNIPAM copolymer samples using a dn/dc value of 0.077 mL/g in DMF at 55 °C with a flow rate of 1.0 mL/min (Viscotek GPC pump; columns: ViscoGel I-series G-3000 and G-4000 mixed bed columns: molecular weight range 0-60 × 103 and 0-400 × 103 g/mol, respectively). Detection consisted of a Viscotek refractive index detector operating at λ ) 660 nm, and a Viscotek model 270 series platform, consisting of a laser light scattering detector (operating at 3 mW, λ ) 670 nm with detector angles of 7° and 90°) and a four capillary viscometer. PNIPAM copolymer molecular weights were calculated with respect to polystyrene and poly(methyl methacrylate) standards. 1 H NMR spectra of the polymers were obtained on a Chemagnetics CMX Infinity 400 solids/liquids NMR spectrometer and data obtained was manipulated in Galactic GRAMS AI software. Sample concentrations were 5% (w/v) in deuterated DMSO containing 1% TMS as an internal reference. Transmission electron microscopy (TEM) was performed on a Philips CM200 with an accelerating voltage of 120 kV. A Keen View Soft Imaging System coupled to iTEM Universal TEM Imaging Platform Software was utilized to acquire digital TEM images. Thermal gravimetric analysis (TGA) was achieved on a Seiko TG/DTA 220 calibrated against an indium standard. Alumina pans were used and samples were run with a helium gas sweep. The heating program was as follows: 30 to 110 °C at 20 °C/min, hold 30min at 110 °C, heat from 110-850 at 7 °C/min, hold at 850 °C for 30 min, then cool to 30 °C. Attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectra were collected utilizing a Smart SAGA attachment coupled with a Thermo-Electron Nicolet 4700 spectrometer, collecting 16 background scans and 64 sample scans, and utilizing Nicolet’s OMNIC software. The relative viable cell number was measured using a bioreductive fluorescent assay, Alamar Blue (Prorrega, Madison, WI), according to manufacturer directions using a Synergy

Rowe et al. HT multidetection microplate reader with KC4 software from Bio-Tek Instruments. Fluorescence images were acquired using a Zeiss Axioplan microscope fitted with FITC and Texas Red filter sets. Magnetic resonance imaging and measurements were made on a 1.5 T General Electric Signa LX scanner from GE Healthcare. ICP-AES data was acquired on a Perkin-Elmer Optima 3000 ICP-AES instrument following the EPA 200.7 standardized method. The instrument was calibrated with an internal scandium standard and recalibrated if there was greater than 20% drift from the 50 ppm concentration. Samples were diluted in a 1% nitric solution to give a total volume of 10 mL and run against an internal quality control gadolinium standard from High Purity Standards using a two point calibration.

Results and Discussion Nanoparticles demonstrate tremendous potential for application in the general area of nanomedicines. In particular, the use of nanoparticles for therapeutic imaging, biosensing, biolabeling, and drug delivery may provide dramatic improvements in the clinical application of nanomedicines.1-4,6,30 However, the application of nanoparticles in such areas critically depends on the ability to modify the surface properties on the particles to introduce advanced functionality, biocompatibility, and stability in a well defined manner. Recently there has been a great deal of attention on the development of Gd nanoparticles as MRI contrast agents, biosensors, and for drug delivery. While a variety of different types of Gd nanoparticles have been synthesized,10,13,16-24 the work of Rieter and co-workers developing nanoscale MOFs consisting of Gd is particularly interesting.16 To date, there have been limited reports regarding the functionalization or surface modification of these Gd MOF nanoparticles.17,31 This was achieved by silica coating the nanoscale metal-organic frameworks of Gd and the ligand 1,4BDC followed by functionalization with dipicolinic acid for luminescence sensing. However, to the best of our knowledge there are no reports on the functionalization of Gd nanoparticles with a molecular targeting agent, a chemotherapeutic, and a fluorescent component for multimodal imaging. This work describes for the first time a general method to covalently modify the surface of the Gd MOF nanoparticles with highly functional biocompatible polymers synthesized via RAFT polymerization containing a targeting ligand, an antineoplastic drug, and a fluorescent monomer using a novel surface modification technique. Synthesis of Highly Functional Biocompatible Polymers via RAFT Polymerization. The first step in the preparation of the polymer-modified Gd MOF nanoparticles involved the synthesis of PNIPAM-co-PNAOS-co-PFMA random copolymers via RAFT polymerization employing the RAFT agent, DATC. Trithiocarbonates, such as DATC, have been shown to be effective RAFT agents for the polymerization of a wide range of monomers including acrylates, methacrylates, and acrylamides.32,29,33-39 The NIPAM monomer was chosen because of its extensive use in polymers for biobased applications and the overall biocompatibility of PNIPAM,40 which comprises the major component of the prepared random copolymers in most cases. The NAOS monomer was incorporated into the copolymer as a site for the attachment of different targeting and/or therapeutic agents. It has been widely reported that primary amines and alcohols will react readily with the succinimide functionality present on the NAOS monomer.27,28 As each of the targeting and therapeutic agents chosen to be used in this work contain either an available primary amine or alcohol, by introducing the NAOS monomer at different weight percentages into the copolymer, the targeting and chemotherapeutic loading can be easily modified. The content of the NAOS monomer



987

Table 1. Molecular Weight Properties of RAFT Copolymers Determined by Gel Permeation Chromatography polymer structure PNIPAM-co PNIPAM-co PNIPAM-co PNIPAM-co PNIPAM-co a

a

-PNAOS-co -PFMA at 10 wt % NAOS -PNAOS-co -PFMA at 17 wt % NAOS -PNAOS-co -PFMA at 25 wt % NAOS -PNAOS at 33 wt % NAOS -PNAOS at 50 wt % NAOS

theor Mn (g/mol)

exptl Mn (g/mol)

PDI

7500 8900 10100 12000 16000

7720 9950 12900 12190 17850

1.09 1.11 1.06 1.08 1.12

PNIPAM-co-PNAOS-co-PFMA is defined as poly(N-isopropylacrylamide)-co-poly(N-acryloxysuccinimide)-co-poly(fluorescein O-methacrylate).

was varied between 10 and 50 wt %, allowing for control over the concentration of targeting and/or therapeutic attached to the copolymer. The addition of the FMA monomer into the copolymer, at 0.5 wt %, allowed for the introduction of a fluorescence moiety providing in vitro cellular-level imaging and making the final polymer-modified Gd MOF nanoparticle bimodal with respect to imaging. The use of RAFT polymerization techniques to produce the polymers for attachment to the Gd MOF nanoparticles offers a number of advantages. First, due to the living nature of RAFT polymerizations, the molecular weight of the copolymer can be controlled by simply varying the ratio of monomer to RAFT agent and the extent of conversion of the polymerization. As can be seen in Table 1, a range of copolymer molecular weights, from 7500-16000 g/mol, were prepared, and in each case there is very good agreement between the theoretical and the experimental molecular weights. Second, RAFT polymerization is arguably the best living radical polymerization technique for the polymerization of functional monomers, such as NIPAM, NAOS, and FMA.25,41,42 As such, the incorporation of a higher weight percentage of NAOS monomer into the initial polymerization system led to copolymers with increased molecular weights while still maintaining excellent control. This is evidenced by the fact that, for every copolymer synthesized, the experimental number average molecular weights (Mn) are comparable to the theoretical Mn values (Table 1). For example, the experimental Mn of PNIPAM-co-PNAOS-co-PFMA at 10 wt % NAOS of 7720 g/mol was within 3% of the theoretical Mn of 7500 g/mol. Furthermore, 1H NMR confirmed the experimental weight percentages of PNAOS in each of the RAFT copolymers synthesized to be within 1.5 wt % of their corresponding theoretical values. Finally, due to the controlled nature of RAFT polymerizations, the PDI for each of the copolymers synthesized was less than 1.12, indicating very narrow molecular weight distributions, and each GPC curve indicated monomodal molecular weight distributions. Additionally, because the copolymerization of NIPAM and NAOS monomers was allowed to proceed to moderate conversions before the addition of the FMA monomer, 1H NMR confirmed the copolymer backbones to have approximately random structures with blocky PFMA characteristics near the chain ends. Control of the molecular weight is imperative as it will provide control over the concentration of targeting/therapeutic agent that can be incorporated into the final polymer, minimize heterogeneity in the polymer, and provide a high degree of chain-end functionality, making these constructs especially effective for incorporation into a nanoscale theragnostic device. As mentioned previously, a theragnostic nanodevice must contain an imaging component, molecular targeting ligand, and therapeutic agent. Thus, once the random copolymers were synthesized, the ability to incorporate a targeting ligand and/or therapeutic agent onto the polymer backbone via reaction with the NAOS segments was investigated. The peptide GRGDSNH2 was chosen as the targeting ligand and MTX was chosen as the antineoplastic chemotherapeutic agent. The GRGDS-NH2

ligand was chosen due to its ability to target the RVβ3 integrin, which is expressed in angiogenic vasculature in a variety of tumors.43 The chemotherapeutic, MTX, was chosen as it is known to prevent cell proliferation and to induce apoptosis in multiple types of cancer cells through a variety of mechanisms.44 Attachment of the MTX and/or GRGDS-NH2 peptide motifs to the copolymer was achieved via a condensation reaction between the succinimide group on the NAOS monomer and the primary amine present on each motif in the presence of 0.01 M triethylamine (Scheme 1). Successful attachment of the GRGDSNH2 motif and MTX was qualitatively confirmed by 1H NMR spectroscopy, shown in Supporting Information. For example, in the case of the attachment of the MTX chemotherapeutic, after reaction of the PNIPAM-co-PNAOS-co-PFMA copolymer at 25 wt % NAOS with MTX, the copolymer was purified by both column chromatography and extraction to remove any unreacted drug and subsequently 1H NMR confirmed the appearance of carboxylic acid protons at 10.5 ppm and the decrease in intensity of the succinimide CH2 peak at 2.8 ppm due to the attachment of MTX to the polymer backbone. Similar results were observed for the GRGDS-NH2 attachment, where a decrease in intensity of the succinimide CH2 peak at 2.8 ppm was observed with a corresponding appearance of methylene peaks attributed to the backbone of the pentapeptide motif at 4.2, 4.4, 4.6, and 4.8 ppm. 1H NMR spectra for PNIPAM-coPNAOS-co-PFMA at 25 wt % NAOS, along with the RAFT copolymer modified with either MTX, or MTX and GRGDSNH2 are included in Supporting Information. The 1H NMR studies also provided information regarding critical characteristics of the multifunctional copolymers including the number of drug molecules per polymer chain and the number of targeting ligands per polymer chain. For example, the RAFT copolymer containing 25 wt % NAOS, with a molecular weight of 12900 g/mol and PDI of 1.06, contained 15 molecules of MTX and 5 molecules of GRGDS-NH2 per polymer chain. By varying the amount of NAOS or the concentration of MTX or GRGDSNH2 used in the modification steps, the number of molecules of each can easily be modified and controlled. Although our data confirmed that copolymers containing a range of weight percentages of NAOS could be controllably synthesized, to limit the length of this article, we will only focus on the RAFT copolymer containing 25 wt % NAOS. It should be noted that as the RAFT polymerization technique allows for the preparation of well-defined copolymers containing the NAOS monomer, the reactivity of the succinimide group can be used to produce copolymer structures tailored with a wide range of targetingtreatment combinations, via postpolymerization modification, for diagnosis and treatment of not only cancer, but a wide range of different diseases. Gd MOF Nanoparticle Synthesis. As mentioned above, this research focused on the development of a novel surface modification technique for the functionalization of Gd MOF nanoparticles to produce a nanoscale theragnostic device. The use of Gd MOF nanoparticles as positive contrast agents for MRI should provide several advantages over the clinically

988


Rowe et al.

Figure 2. ATR-FTIR spectra of the (a) PNIPAM-co-PNAOS-co-PFMA RAFT copolymer, along with the (c) unmodified and (b) copolymermodified gadolinium (Gd) metal-organic framework (MOF) nanoparticles. (d) Transmission electron microscopy (TEM) image of unmodified Gd MOF nanoparticles synthesized via a reverse microemulsion system providing an average width of 20-25 nm and length of 100-150 nm. (e) TEM image of copolymer-modified Gd MOF nanoparticles with an average film thickness of approximately 9 nm.

Figure 3. Concentration of Gd3+ leached from unmodified and polymer-modified gadolinium (Gd) metal-organic framework (MOF) nanoparticles during dialysis purification graphed as a function of time. Gd3+ concentration determined by inductively coupled plasmaabsorption emission spectroscopy.

employed Gd chelates, such as enhanced imaging through magnetic resonance and increased biostability.10,12,13,16,18-22,24 The Gd nanoparticles used in this research were prepared via a reverse microemulsion procedure developed by Reiter and coworkers.16 This method is reported to produce nanoscale MOFs with the general composition Gd(1,4-BDC)1.5(H2O)2. The chelating ligand 1,4-BDC and a water to surfactant ratio of 10 were employed to provide Gd MOF nanoparticles with an average width of 20-25 nm and length of 100-150 nm, as determined from TEM images (Figure 2d). The synthesized Gd MOF nanoparticles were also characterized by ATR-FTIR. The ATR-FTIR spectrum, Figure 3c, showed a characteristic outof-plane dCsH aromatic stretch at 725 cm-1, symmetric carboxylate stretch at 1400 cm-1, an asymmetric carboxylate stretch at 1540 cm-1, along with 2855 cm-1, 2925 cm-1, and 3065 cm-1, which are attributed to the sCsH stretching

vibrations of the 1,4-BDC bridging ligand, and 3460 cm-1, which was attributed to the sOH stretch of the water ligand. Surface Modification of Gd MOF Nanoparticles with RAFT Copolymers. The surface modification of nanoparticles to incorporate advanced functionality and biocompatibility is a critical step in the development of the next generation of nanoscale theragnostic devices. Surface modification of Gd MOF nanoparticles with well-defined polymers synthesized via RAFT polymerization potentially offers one of the most versatile methods to incorporate all of the desired components of a theragnostic device into multifunctional platforms for nanomedicines. As discussed above, the use of RAFT polymerization to prepare the highly functional biocompatible copolymers offers numerous advantages. In addition to the previously discussed advantages, RAFT polymers have a high degree of chain end functionality, where one end of the chain will contain a thiocarbonylthio moiety due to limited termination events during the polymerization.25 In this case, the employment of the RAFT agent, DATC, provides trithiocarbonate terminated chains which can be reduced in the presence of a nucleophile, such as hexylamine, to provide a thiol moiety.32,45,46 Due to the fact thiols have been shown to effectively react with a variety of surfaces including metallic and semiconducting nanoparticles, RAFT polymerization has become one of the premier polymerization techniques to prepare polymer functionalized surfaces.45 However, despite this fact, RAFT polymers have, to the best of our knowledge, never been used for the modification of either MOFs or Gd MOF nanoparticles. In this research, modification of the Gd MOF nanoparticles was achieved via initial aminolysis, using hexylamine, of the trithiocarbonate end group of the RAFT copolymers to a thiolate functionality under inert and basic conditions. In each case, both ATR-FTIR and 1H NMR spectroscopy confirmed near quantitative reduction of the trithiocarbonate endgroups of the RAFT copolymers to thiols with the addition of hexylamine. Subsequently, the thiolate terminated copolymer was covalently attached to the nanoparticle surface through a coordination reaction between the polymer chain thiolate end-group moiety and vacant orbitals on the Gd3+ ions at the surface of the Gd MOF nanoparticles. Thiols are known to form stable metal


thiolate compounds with many metal ions and thiolates are also known to coordinate strongly to many organometallic compounds including iron porphyrins.47,48 It should be noted that, after polymer deposition and prior to characterization or use, the nanoparticles were washed several times with a good solvent for the polymer to remove any untethered polymer from the system. Successful modification of the Gd MOF nanoparticles with RAFT copolymers was characterized through both TEM (Figure 2d,e) and ATR-FTIR spectroscopy (Figure 2a-c). TEM images (Figure 2e) indicate a relatively uniform coating of polymer around the Gd MOF nanoparticles after deposition, with an average thickness of approximately 9 nm. ATR-FTIR was also utilized to confirm the addition of the polymer onto the nanoparticles without the loss of the copolymer functionality. Several of the characteristic stretches of the free copolymer, including the carbonyl stretch at 1735 cm-1, succinimide stretch at 1650 cm-1, and methylene stretches from 2810-3000 cm-1, display good transference to the polymer-modified Gd MOF nanoparticles when compared to the unmodified Gd MOF nanoparticles (Figure 2a-c). In an attempt to further confirm the need for reduction of the RAFT copolymer endgroups for successful attachment, modification of the Gd MOF nanoparticles was attempted without the use of a reducing agent. Both TEM and ATR-FTIR showed unsuccessful surface modification of the Gd MOF nanoparticles confirming the need for the thiolate polymer end group to coordinate with empty Gd3+ orbitals of the Gd MOF nanoparticle for successful modification. Furthermore, a small molecule thiol, dodecanethiol, was used to modify the Gd MOF nanoparticles to confirm the route of our hypothesized surface modification technique. ATR-FTIR confirmed the addition of the small molecule with several of the characteristic stretches appearing in the spectrum of the dodecanethiolate modified Gd MOF nanoparticles. The combination of all of these experiments, confirmed the thiolate end groups of the RAFT polymer, formed after aminolysis, were providing a route to attachment of the copolymers and suggested that the thiolate end groups were coordinating to empty orbitals of the Gd3+ of the Gd MOF nanoparticle surface. One of the advantages of using preformed polymers to covalently modify the Gd MOF nanoparticles is that the grafting density, or number of polymer chains per surface area, can be easily determined using a combination of TGA, to determine the mass of polymer per mass of nanoparticles, and TEM to determine the surface area of the nanoparticles. Using this information, combined with a density of 2.529 g/cm3 for the Gd MOF nanoparticles, the grafting density was determined to be approximately 0.97 chains/nm2, which corresponds to approximately 25000 polymer chains per Gd MOF nanoparticle. When the previously determined concentration of MTX and GRGDS-NH2 for the 25 wt % NAOS RAFT copolymer was used, this corresponds to approximately 37.5 × 104 molecules of MTX and 12.5 × 104 molecules of GRGDS-NH2 per Gd MOF nanoparticle. Due to the narrow molecular weight distributions produced in RAFT polymerization, approximately the same concentration of therapeutic/targeting agents will be incorporated to each polymer chain, making each of the polymer-modified Gd MOF nanoparticles very similar in composition. To the best of our knowledge, this is the first time Gd MOF nanoparticles have been successfully surfaced modified with well-defined, highly functional RAFT copolymers for the production of multifunctional nanodevices for image guided cancer intervention. The developed method has the added advantage that the copolymers can be well characterized through common spectroscopic and chromatographic techniques and


989

modified with a wide variety of targeting ligands or therapeutics before surface functionalization. This provides tremendous flexibility in controlling the loading capacity of the molecular targeting agent and the chemotherapeutic, while allowing for a wide range of different therapeutic and target ligand combinations. In Vitro Stability of Unmodified and Polymer-Modified Gd MOF Nanoparticles. Recently there has been an increasing focus on nephrogenic fibrosing dermopathy (NFD), which is a disease that affects a subset of patients with renal insufficiency and has been linked to exposure to Gd3+ released from conventional MRI contrast agents.49-51 As such, the stability of the Gd MOF nanoparticles used in this work is of critical importance. Recently, literature has shown a simplified method of determining the concentration of free Gd3+ leached from the Gd MOF nanoparticles.11,31 The unmodified and polymermodified Gd MOF nanoparticles were purified by dialysis using the literature procedure to provide information about their stability.11,31 In this procedure, the unmodified and polymermodified Gd MOF nanoparticles underwent three washings at a pH ) 3, as the mild acidity should be sufficient to overcome the van der Waals forces or electrostatic attraction of any physisorbed Gd3+ on the Gd MOF nanoparticle surface, without degrading the nanoscale Gd-based MOF or polymer coating. In an attempt to investigate the in vitro stability of the unmodified and polymer-modified Gd MOF nanoparticles, the nanoparticles were then dialyzed against DIUF water, a 0.09 wt % sodium chloride isotonic solution, and a PBS solution for up to one week with the aqueous dialysate solution being analyzed at several time increments along the experiment. In each case, the aqueous solution was then characterized by ICPAES to determine the concentration of Gd3+ leached from each of the Gd MOF nanoparticle constructs (Figure 3). These results demonstrate that in each case a very small quantity of Gd3+ is released over a period of one week and the vast majority of this release occurred very early in the experiment with no further release of Gd3+ seen after the 12 h dialysis period. In the case of the water and PBS dialysis experiments, each of the unmodified Gd MOF nanoparticle samples lost 0.0139 µmol Gd3+ and 0.0307 µmol Gd3+, respectively. In parallel experiments with water and PBS dialysate, the polymer-modified Gd MOF nanoparticle only leached 0.000328 µmol Gd3+ and 0.00364 µmol Gd3+, respectively. The maximum loss of Gd3+ was determined to be near 1.12 µmol Gd3+ for the unmodified Gd MOF nanoparticle in the saline solution. The results also show that the polymer-modified Gd MOF nanoparticles release significantly less Gd3+, 0.0681 µmol Gd3+, in the saline solution, which suggests that the polymer modification process improves the stability of the Gd MOF nanoparticles. This trend of less Gd3+ leaching to the dialysate with polymer modification was also seen in both the water and PBS experiments. These results demonstrate that the Gd MOF nanoparticles demonstrate excellent stability over long periods of time and that the polymer modification of the nanoparticles improves the stability in all cases. MRI of Polymer-Modified Gd MOF Nanoparticles. One of the primary requirements of any nanoscale theragnostic device is the ability to image the construct using standard clinical techniques. Arguably, MRI has become one of the most important techniques in medical diagnosis, since the discovery of the X-ray. Effectively, MRI measures the characteristics of the hydrogen nuclei of water and the signal intensity depends on the amount of water in the image area.9,52 In many instances, contrast agents are used in MRI to accelerate the rate of relaxation of water molecules in close proximity to the agent.

990


Rowe et al.

Table 2. Experimental Relaxivity Data for Clinical Magnetic Resonance Imaging Contrast Agents, Multihance and Magnevist, along with the Synthesized Gadolinium Nanotheragnostic Devices contrast agent

r1 (mM-1s-1)

r2 (mM-1s-1)

r2/r1

Magnevist Multihance unmodified gadolinium (Gd) metal-organic framework (MOF) nanoparticles PNIPAM-co -PNAOS-co -PFMA modified Gd MOF nanoparticles MTX-copolymer-modified Gd MOF nanoparticles GRGDS-NH2+MTX-copolymer-modified Gd MOF nanoparticles

13.44 19.45 9.86

21.40 30.44 17.94

1.59 1.56 1.82

33.43 38.52 14.45

47.25 53.92 25.29

1.41 1.40 1.75

This results in a large increase in the contrast between the specific organ or tissue of interest and the surrounding tissue. Gd-chelate compounds are widely used as positive contrast agents for MRI, as they produce brightening in T1-weighted images; however, recently researchers have investigated the use of Gd nanoparticles as positive contrast agents for MRI.10,13,17,24 The polymer-modified Gd MOF nanoparticles prepared in this work provide the specific advantage of multimodal imaging capability. The incorporation of the FMA monomer into the RAFT copolymer allows for cellular level imaging via fluorescence microscopy, while the Gd MOF nanoparticle acts as a positive contrast agent for MRI, providing diagnostic imaging at the clinical level. The combination of fluorescence imaging and MRI has received a great deal of attention because they ally the sensitivity of the fluorescence component with the high degree of spatial resolution of MRI.13,31,53 To provide information about the clinical imaging viability of the copolymermodified Gd MOF nanoparticles as a positive contrast agent, in vitro MR imaging using a 1.5 T scanner was employed to determine relaxivity properties of the nanoparticle scaffolds. The results shown in Table 2 compares the calculated relaxivity values, r1 and r2, of several polymer-modified Gd MOF nanoparticles to unmodified Gd MOF nanoparticles and to the clinically employed positive MRI contrasts agents, gadopentetate dimeglumine (Magnevist) and gadobenate dimeglumine (Multihance). The r1 and r2 values are simply the inverse of the measured T1 and T2 relaxation values, respectively, in relation to the concentration of Gd3+. The associated relaxivity curves (Supporting Information) depict the calculated relaxivity rate for the various compounds with respect to their Gd3+ concentration calculated by ICP-AES. The relaxivity values demonstrate that both the unmodified and polymer-modified Gd MOF nanoparticles result in a large shortening of the T1 relaxation time and, thus, behave as positive contrast agents. Of particular note is the fact that the polymer modification of the Gd MOF nanoparticles increased the relaxivity to approximately triple (r1 ) 33.4 mM-1 s-1) the relaxation value when compared to the unmodified Gd MOF nanoparticles (r1 ) 9.86 mM-1 s-1). This phenomenon is attributed to increased water retention by the hydrophilic RAFT copolymer matrix attached to the surface of the Gd MOF nanoparticles, thus enhancing T1 relaxation shortening effects. To effectively compare the relaxation times of the unmodified and polymer-modified Gd MOF nanoparticles to clinically used Gd chelate contrast agents it is important to compare the concentration of Gd3+ in each sample. As such, ICP-AES was employed to approximate Gd3+ concentrations in the Gd MOF nanoparticles and polymer-modified Gd MOF nanoparticles in comparison to that of Magnevist and Multihance. The relaxivity data shows that the r1 value of the polymer-modified Gd MOF nanoparticles (r1 ) 33.4 mM-1 s-1) is nearly double that of the conventional clinically used Gd contrast agents (r1Magnevist ) 13.4 mM-1 s-1 and r1Multihance ) 19.5 mM-1 s-1). However, the relaxivity values for the Gd MOF nanoparticles, which were modified with the MTX/GRGDS-

NH2 containing polymer, were calculated using Gd3+ concentrations in the range of 0.16 to 0.013 mmol/L, while the conventional contrast agent, Magnevist, was determined to be in the range of 0.51 to 0.046 mmol/L. These concentrations of Gd3+ for the Gd MOF nanoparticles systems are approximately three times less than the clinically employed contrast agents, Magnevist and Multihance. As the relaxivity should be approximately proportional to the concentration of Gd3+, these results suggest that equal concentrations of Gd3+ from polymermodified Gd MOF nanoparticles should produce comparable, or potentially better, relaxation rates when compared to the clinically used contrast agents. This data demonstrates the feasibility of achieving clinically useful T1 shortening effects at lower Gd3+ doses with these novel polymer-modified Gd MOF nanoparticles. Molecular Targeting of Polymer-Modified Gd MOF Nanoparticles. In addition to an imaging component, one of the other essential requirements of a successful theragnostic nanodevice is the presence of a molecular targeting component or ligand to increase the targeted selectivity of the system and, in the case of nanoparticles, to take advantage of the large amount of imaging agent that can be delivered to the desired location per targeting biorecognition event, thus reducing the amount of Gd3+ required for effective MR imaging.7 The synthesized polymer-modified Gd MOF nanoparticles have been designed to incorporate dual targeting components. The first targeting component is incorporated by tailoring the PNIPAMco-PNAOS-co-PFMA copolymer with an active targeting ligand, GRGDS-NH2. As discussed earlier, GRGDS-NH2 has been widely used as a molecular targeting agent, as it can specifically bind to overexpressed Rvβ3 integrins on tumor cells and induce receptor-mediated endocytosis for cellular uptake.43 The second potential route of targeting is passive targeting through the enhanced permeability and retention (EPR) effect. This tumor targeting mechanism results from the increased vasculature present in most tumors resulting in preferential extravasation and protracted lodging of particulates of a particular size.54 The EPR effect can be taken advantage of due to the controllable size of the Gd MOF nanoparticles used to produce the polymermodified constructs. To investigate the ability of the polymermodified Gd MOF nanoparticles to be selectively targeted to cancer cells using the active GRGDS-NH2 ligand, fluorescence microscopy was used. As mentioned previously, the incorporation of the FMA monomer into the RAFT copolymer provides a means of cellular level fluorescence imaging. To evaluate the ability to selectively target the polymer-modified Gd MOF nanoparticles, FITZ-HSA, an Rvβ3-expressing canine endothelial sarcoma cell line, was incubated with polymer-modified Gd MOF nanoparticles, where the polymer had either been tailored with the GRGDS-NH2 ligand or contained no GRGDS-NH2, for 1, 4, or 24 h followed by extensive washing, fixation, and fluorescence microscopy. As can be seen in Figure 4a, in the case where the polymer attached to the Gd MOF nanoparticles had been tailored with GRGDS-NH2 significant cellular surface



991

Figure 4. (a) Fluorescence microscopy image of FITZ-HSA cells incubated with MTX/GRGDS-NH2 tailored copolymer-modified gadolinium (Gd) metal-organic framework (MOF) nanoparticles at 1, 4, and 24 h and (b) corresponding microscopy image of FITZ-HSA cells stained with propidium iodide for the 24 h study. (c) Fluorescence microscopy image of FITZ-HSA cells incubated with the control system of MTX tailored copolymer-modified Gd MOF nanoparticles at 1, 4, and 24 h and (d) corresponding microscopy image of FITZ-HSA cells stained with propidium iodide for the 24 h study. (e) Fluorescence microscopy image of FITZ-HSA cells pretreated with free GRGDS-NH2 peptide then incubated with the MTX/GRGD tailored copolymer-modified Gd MOF nanoparticles at 24 h and (e) corresponding microscopy image of FITZ-HSA cells stained with propidium iodide for the 24 h study.

binding can be visualized after 1 h of incubation. This is evidenced by a fluorescent halo forming around the surface of the FITZ-HSA cells after washing of the cells before fluorescence microscopy. After 4 h of incubation, the formation of the fluorescent halo becomes even more pronounced. Furthermore, fluorescence microscopy images (Figure 4a) show internalization of the GRGDS-NH2 tailored polymer-modified Gd MOF nanoparticles by the cells after an incubation period of 24 h. This was suggested by staining of the nuclei of FITZHSA cells with propidium iodide (PI) allowing for visualization of spatial location of the cells. As can be seen in Figure 4b, cells stained with the PI showed the same spatial location as the accumulated fluorescent-tagged nanoparticles, confirming selective uptake of the targeted polymer-modified Gd MOF nanoparticles by the tumor cells. To demonstrate the importance of the active molecular targeting agent to promote cellular uptake, a control system was analyzed, where the Gd MOF nanoparticles were modified with the PNIPAM-co-PNAOS-coPFMA copolymer that had not been tailored with GRGDS-NH2. As can be seen in Figure 4c, no significant amount of fluorescence was seen in FITZ-HSA cells that were incubated with Gd MOF nanoparticles that did not contain the GRGDSNH2 at 1, 4, or 24 h. This was expected as extensive washing of the cells before fluorescence microscopy should remove all polymer-modified Gd MOF nanoparticles and the nontargeted nature of the nanoparticles should provide no attachment mechanism to the cancer cells by means of active targeting. It

should be noted that after 24 h, the nuclei of the FITZ-HSA cells that were stained with PI were still present when viewed with the Texas Red filter (Figure 4d). These results confirm preferential uptake of the targeted polymer-modified Gd MOF nanoparticles was due to active targeting of the Rvβ3-integrins by the GRGDS-NH2 ligand. To further confirm specificity of targeting, FITZ-HSA cells were pretreated with free GRGDSNH2 peptide to block the Rvβ3-integrins before introduction of the targeted polymer-modified Gd MOF nanoparticles. The GRGDS-NH2 targeted polymer-modified Gd MOF nanoparticles were then introduced, and fluorescent microscopy demonstrated that there was no specific binding for the nanoparticles, further exhibiting the targeting specificity of our nanoparticle constructs (Figures 4e,f). These results demonstrate that an active targeting functionality can be added to these Gd MOF nanoparticles, while maintaining the properties required for clinical diagnostic imaging. In addition, these constructs demonstrated fluorescence capability, which makes the constructs bimodal in regard to imaging, unlike other clinical imaging agents, such as Multihance or Magnevist. Cell Growth Inhibition of Polymer-Modified Gd MOF Nanoparticles. To demonstrate that the polymer-modified Gd MOF nanoparticle based platforms contain all the basic requirements of a theragnostic nanodevice, growth inhibition studies were performed using FITZ-HSA tumor cells. Samples were incubated with FITZ-HSA at 37 °C in standard culture medium

992


Rowe et al.

therapeutic but with the added benefit that the device has targeting ability and bimodal imaging through fluorescence and MR.

Conclusions

Figure 5. Cell growth inhibition studies for reagents involved in multifunctional nanoparticle formation, including: Methotrexate (MTX), PNIPAM-co-PNAOS-co-PFMA tailored with MTX, gadolinium (Gd) metal-organic framework (MOF) nanoparticles modified with PNIPAMco-PNAOS-co-PFMA tailored with MTX. Finally, the control, Gd MOF nanoparticles modified with PNIPAM-co-PNAOS-co-PFMA, without the MTX therapeutic, is shown for comparison. Dilutions (10-fold) of each sample were carried out, with Dilution 1 having the highest concentration of the therapeutic, MTX, and Dilution 4 having the lowest concentration of MTX. The concentration of MTX for each dilution of each therapeutic sample is as follows: Dilution 1 ) 1.15 mM, Dilution 2 ) 0.115 mM, Dilution 3 ) 0.0115 mM, and Dilution 4 ) 0.00115 M. Again, there is no MTX in Dilutions 1′, 2′, 3′, and 4′.

containing 10% PBS for 72 h in a 5% CO2 atmosphere. The reagents for nanodevice formation, that is, Gd(III) chloride salt, 1,4-benzenedicarboxylic acid methylammonium salt, unmodified Gd MOF nanoparticles, and Gd MOF nanoparticles modified with RAFT copolymers were studied. The Gd(III) chloride salt, 1,4-benzenedicarboxylic acid methylammonium salt, and the unmodified Gd MOF nanoparticles resulted in significant cell growth inhibition at high concentrations (Supporting Information). However, it is important to note that modification of the Gd MOF nanoparticles with the RAFT copolymer not containing MTX increased cell viability (Supporting Information). The increased cell viability is attributed to the coating of the Gd MOF nanoparticles with copolymers consisting of the biocompatible polymer PNIPAM. This infers that the presence of the RAFT copolymer on the surface of the Gd MOF nanoparticles increases the biocompatible nature of the nanodevice. To demonstrate the ability of polymer-modified nanoparticles to act as chemotherapeutics, Gd MOF nanoparticles were modified with RAFT copolymers containing MTX and both MTX and GRGDS-NH2, after which additional cell growth inhibition studies were performed (Figure 5). As a control, both MTX and the RAFT copolymer, unattached to the nanoparticles, were also used in cell growth inhibition studies (Figure 5). As stated previously, MTX is a widely used chemotherapy agent that has been employed in many drug delivery systems because of its ability to inhibit the folate synthesis metabolic pathway.44 Figure 5 demonstrates that the MTX-containing polymer-modified Gd MOF nanoparticles show a dose-dependent inhibition of growth of the FITZ-HSA tumor cells that was comparable to that of the free MTX drug, on an equal concentration of MTX basis. These results demonstrate that attachment of the MTX drug to the copolymer and subsequent modification of Gd MOF nanoparticles with the MTX copolymer does not affect the overall function of the drug. Addition of the MTX to the theragnostic construct provides similar performance to the free

The development of nanoscale theragnostic devices for the diagnosis and treatment of cancer represents one of the primary targets of the general field of nanomedicine. Despite the fact that the incredible potential of these devices is widely recognized, their clinical application is yet to be realized due to poor design and manufacturing techniques. This research has demonstrated incorporation of an MRI contrast agent (Gd MOF nanoparticle), a cellular level imaging agent (PFMA), a targeting moiety (GRGDS-NH2), and an antineoplastic drug (MTX) into a single nanoscale theragnostic device, with preservation of all the functions of the individual components. Specifically, versatile highly functional copolymers of PNIPAM-co-PNAOSco-PFMA have been synthesized via RAFT polymerization. Due to the well-defined nature of the copolymers formed they are ideal as a scaffold for attachment of therapeutic and/or targeting moieties. PNIPAM-co-PNAOS-co-PFMA copolymers were successfully tailored with both a chemotherapeutic, MTX, and with a targeting ligand, GRGDS-NH2, to form the multifunctional polymeric structure. Following preparation of the RAFT copolymers, quantitative aminolysis of the trithiocarbonate end groups in basic conditions provided thiolate end groups which allowed for the direct attachment of the copolymers to the surface of the Gd nanoparticle metal-organic frameworks. Successful modification was attributed to thiolate attachment through vacant orbitals on the Gd3+ ions at the surface of the Gd MOF nanoparticles and long-term stability in a range of aqueous medium was confirmed. To evaluate the potential of the RAFT copolymer-modified nanoparticles as a theragnostic nanodevice, MRI, fluorescence imaging, and cell growth inhibition studies were performed. In the MRI studies, the RAFT copolymer-modified Gd MOF nanoparticles showed T1 relaxation times that were one-third the value of the unmodified Gd MOF nanoparticles. Furthermore, T1 relaxation times of the polymer-modified Gd MOF nanoparticles are of the same order of magnitude as those of conventional clinically used Gd contrast agents. The incorporation of FMA into the copolymer constructs provided polymer-modified Gd MOF nanoparticles that were successfully used as a bimodal diagnostic imaging device for both MR and fluorescence imaging. Additionally, using fluorescence imaging, it was demonstrated that the RAFT copolymermodified Gd MOF nanoparticles containing the targeting moiety, GRGDS-NH2, showed active targeting toward FITZ-HSA tumor cells, which overexpress Rvβ3-integrins. Gd MOF nanoparticles modified with the RAFT copolymer containing MTX showed dose-dependent treatment of FITZ-HSA cancer cells established by cell viability measurements. By taking advantage of advancements in both nanotechnology and polymer science, this research has effectively prepared new nanoscale multifunctional devices with tumor targeting, treatment, and diagnostic imaging capability. The combination of these technologies has the potential to advance the therapeutic outcome of drug treatment for cancer sufferers and improve the quality of life for patients. Acknowledgment. The authors would like to thank the Colorado State University Cancer Supercluster for funding, along with Professor Brent Sumerlin and Andrew Vogt of Southern Methodist University for GPC characterization of the PNIPAM-co-PNAOS-co-PFMA RAFT copolymers.


Supporting Information Available. 1H NMR characterization of the modified and unmodified RAFT copolymer, R1 and R2 relaxation curves, and cell growth inhibition data for the components of the nanomedicine device. This material is available free of charge via the Internet at http://pubs.acs.org.

References and Notes (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)

(14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26)

Ferrari, M. Nat. ReV. Cancer 2005, 5, 162–171. Panchapakesan, B. Oncol. Issues 2005, NoVember/December, 20–23. Park, K. J. Controlled Release 2007, 120, 1–3. Service, R. F. Science 2005, 310, 1132–1134. Wickline, S. A.; Lanza, G. M. Circulation 2003, 107, 1092–1095. Nasongkla, N.; Bey, E.; Ren, J.; Ai, H.; Khemtong, C.; Guthi, J. S.; Chin, S.-F.; Sherry, A. D.; Boothman, D. A.; Gao, J. Nano Lett. 2006, 6, 2427–2430. Allen, T. M. Nat. ReV. Cancer 2002, 2, 750–763. Jain, R. K. Nat. Med. 1998, 4, 655–657. Caravan, P.; Ellison, J. J.; McMurry, T. J.; Lauffer, R. B. Chem. ReV. 1999, 99, 2293–2352. Hifumi, H.; Yamaoka, S.; Tanimoto, A.; Citterio, D.; Suzuki, K. J. Am. Chem. Soc. 2006, 128, 15090–15091. Taylor, K. M. L.; Kim, J. S.; Rieter, W. J.; An, H.; Lin, W.; Lin, W. J. Am. Chem. Soc. 2008, 130, 2154–2155. Reynolds, C. H.; Annan, N.; Beshah, K.; Huber, J. H.; Shaber, S. H.; Lenkinski, R. E.; Wortman, J. A. J. Am. Chem. Soc. 2000, 122, 8940– 8945. Bridot, J.-L.; Faure, A.-C.; Laurent, S.; Riviere, C.; Billotey, C.; Hiba, B.; Jainer, M.; Josserand, V.; Coll, J.-L.; Vander Elst, L.; Muller, R.; Roux, S.; Perriat, P.; Tillement, O. J. Am. Chem. Soc. 2007, 129, 5076– 5084. Rogers, W. J.; Meyer, C. H.; Kramer, C. M. Nat. Clin. Pract. CardioVasc. Med. 2006, 3, 554–562. Taboada, E.; Rodriquez, E.; Roig, A.; Oro, J.; Roch, A.; Muller, R. N. Langmuir 2007, 23, 4583–4588. Rieter, W. J.; Taylor, K. M. L.; An, H.; Lin, W.; Lin, W. J. Am. Chem. Soc. 2006, 128, 9024–9025. Rieter, W. J.; Taylor, K. M. L.; Lin, W. J. Am. Chem. Soc. 2007, 129, 9852–9853. Oyewumi, M., R.J. Int. J. Pharm. 2003, 251, 85–97. Oyewumi, M.; Yokel, R. A.; Jay, M.; Coakley, T.; Mumper, R. J. J. Controlled Release 2004, 95, 613–626. Oyewumi, M. O.; Mumper, R. J. Bioconjugate Chem. 2002, 13, 1328– 1335. Oyewumi, M. O.; Mumper, R. J. Drug DeV. Ind. Pharm. 2002, 28, 317–328. Oyewumi, M. O.; Liu, S.; Moscow, J. A.; Mumper, R. J. Bioconjugate Chem. 2004, 14, 404–411. Oyewumi, M. O.; Yokel, R. A.; Jay, M.; Coakley, T.; Mumper, R. J. J. Controlled Release 2004, 95, 613–626. Evanics, F.; Diamente, P. R.; Van Veggel, F. C. J. M.; Stanisz, G. J.; Prosser, R. S. Chem. Mater. 2006, 18, 2499–2505. Perrier, S.; Takolpuckdee, P. J. Polym. Sci., Part A: Polym. Chem. 2005, 43, 5347–5393. Le, T. P., Moad, G., Rizzardo, E., Thang, S. H. SH PCT Int. Appl. WO 98 01478 A1 980115, 1998.


993

(27) Yanjarappa, M. J.; Gujraty, K. V.; Joshi, A.; Saraph, A.; Kane, R. S. Biomacromolecules 2006, 7, 1665–1670. (28) Favier, A.; D’Angosto, F.; Charreyre, M.-T.; Pichot, C. Polymer 2004, 45, 7821–7830. (29) Lai, J. T.; Filla, D.; Shea, R. Macromolecules 2002, 35, 6754–6756. (30) Yezhelyev, M. V.; Gao, X.; Xing, Y.; Al-Hajj, A.; Nie, S.; O’Regan, R. M. Lancet Oncol. 2006, 7, 657–667. (31) Kim, J. S. R., W. J.; Taylor, K. M. L.; An, H.; Lin, W.; Lin, W. J. Am. Chem. Soc. 2007, 129, 8962–8963. (32) Xu, J.; He, J.; Fan, D.; Wang, X.; Yang, Y. Macromolecules 2006, 39, 8616–8624. (33) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Krstina, J.; Moad, G.; Postma, A.; Thang, S. H. Macromolecules 2000, 33, 243–245. (34) Rowe-Konopacki, M. D.; Boyes, S. G. Macromolecules 2007, 40, 879– 888. (35) Rowe, M. D.; Hammer, B. A. G.; Boyes, S. G. Macromolecules 2008, 41, 4147–4157. (36) Hotchkiss, J. W.; Lowe, A. B.; Boyes, S. G. Chem. Mater. 2007, 19, 6–13. (37) Hentschel, J.; Bleek, K.; Ernst, O.; Lutz, J.-F.; Borner, H. G. Macromolecules 2008, 41, 1073–1075. (38) An, Z.; Shi, Q.; Tang, W.; Tsung, C.-K.; Hawker, C. J.; Stucky, G. D. J. Am. Chem. Soc. 2007, 129, 14493–14499. (39) Bai, Z.; He, Y.; Young, N. P.; Lodge, T. P. Macromolecules 2008, 41, 6615–6617. (40) Vihola, H.; Laukkanen, A.; Valtola, L.; Tenhu, H.; Hirvonen, J. Biomaterials 2005, 26, 3055–3064. (41) Moad, G.; Chiefari, J.; Chong, Y. K.; Krstina, J.; Maydunne, R. T. A.; Postma, A.; Rizzardo, E.; Thang, S. H. Polym. Int. 2000, 49, 993– 1001. (42) Lowe, A. B.; McCormick, C. L. Prog. Polym. Sci. 2007, 32, 283– 351. (43) Ye, Y.; Bloch, S.; Xu, B.; Achilefu, S. J. Med. Chem. 2006, 49, 2268– 2275. (44) Kojima, C.; Kono, K.; Maruyama, K.; Takagishi, T. Bioconjugate Chem. 2000, 11, 910–917. (45) Sumerlin, B. S.; Lowe, A. B.; Stroud, P. A.; Zhang, P.; Urban, M. W.; McCormick, C. L. Langmuir 2003, 19, 5559–5562. (46) Moad, G.; Chong, Y. K.; Postma, A.; Rizzardo, E.; Thang, S. H. Polymer 2005, 46, 8458–8468. (47) Larsen, T. H.; Sigman, M.; Ghezelbash, A.; Doty, R. C.; Korgel, B. A. J. Am. Chem. Soc. 2003, 125, 5638–5639. (48) Urano, Y.; Higuchi, T.; Hirobe, M.; Nagano, T. J. Am. Chem. Soc. 1997, 119, 12008–12009. (49) Cowper, S. E. Am. J. Kidney Dis. 2005, 46, 763–765. (50) Grobner, T. Nephrol., Dial., Transplant. 2006, 21, 1104–1108. (51) High, W. A.; Ayers, R. A.; Chandler, J.; Zito, G.; Cowper, S. E. J. Am. Acad. Dermatol. 2007, 56, 21–26. (52) Artemov, D.; Bhujwalla, Z. M.; Bulte, J. W. M. Curr. Pharm. Biotechnol. 2004, 5, 485–494. (53) Taylor, K. M. L.; Rieter, W. J.; Lin, W. J. Am. Chem. Soc. 2008, 130, 14358–14359. (54) Brigger, I.; Dubernet, C.; Couvreur, P. AdV. Drug DeliVery ReV. 2002, 54, 631–651.

BM900043E

Polymer-Modified Gadolinium Metal-Organic Framework

Recommend Documents