Article pubs.acs.org/Biomac
Blood Clearance and Biodistribution of Polymer Brush-Afforded Silica Particles Prepared by Surface-Initiated Living Radical Polymerization Kohji Ohno,*,† Tatsuki Akashi,† Yoshinobu Tsujii,†,‡ Masaya Yamamoto,§ and Yasuhiko Tabata§ †
Institute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan JST, CREST, Tokyo 102-0075, Japan § Institute for Frontier Medical Sciences, Kyoto University, Kyoto 606-8507, Japan ‡
S Supporting Information *
ABSTRACT: The physiological properties of polymer brushafforded silica particles prepared by surface-initiated living radical polymerization were investigated in terms of the circulation lifetime in the blood and distribution in tissues. Hydrophilic polymers consisting mainly of poly(poly(ethylene glycol) methyl ether methacrylate) were grafted onto silica particles by surface-initiated atom transfer radical polymerization that was mediated by a copper complex to produce hairy hybrid particles. A series of hybrid particles was synthesized by varying the diameter of the silica core and the chain length of the polymer brush to examine the relationship between their physicochemical and physiological properties. The hybrid particles were injected intravenously into mice to investigate systematically their blood clearance and body distribution. It was revealed that the structural features of the hybrid particles significantly affected their in vivo pharmacokinetics. Some hybrid particles exhibited an excellently prolonged circulation lifetime in the blood with a half life of ∼20 h. When such hybrid particles were injected intravenously into a tumor-bearing mouse, they preferentially accumulated in tumor tissue. The tumor-targeted delivery was optically visualized using hybrid particles grafted with fluorescence-labeled polymer brushes.
■
INTRODUCTION Colloidal particles have been intensively researched for the past several decades with respect to biorelated applications.1−10 Many types of particulate architectures including organic polymeric particles, 11−24 inorganic particles, 25−34 liposomes,35−39 micelles,40−44 star polymers,45−47 and dendrimers48−51 have been used as key materials in the production of functional tools for biomedical therapy and diagnosis and for the bioanalysis of sensors and separations. To provide excellent performance in such applications, it is important to control the surface properties of particulate architectures because particle surfaces are in direct contact with external media and occasionally interact with biomacromolecules such as proteins and polysaccharides. A common method for the surface modification of particulate architectures is the grafting of hydrophilic polymers to the particle surfaces. The introduction of poly(ethylene glycol) (PEG), that is, PEGylation, is a common strategy to provide biocompatibility and good solubility in aqueous media.52−57 The major advantage to be PEGylation reduces nonspecific protein adsorption to surfaces of particulate materials in biomedical appliances. This imparts stealth characteristics to particulate architectures when they are injected into the body system; that is, such particles can minimize the nonspecific interactions with opsonin proteins and the uptake by the reticulo-endothelial system (RES); consequently, the © 2012 American Chemical Society
stealth particles have an increased circulation lifetime in the blood.58−61 The development of long-circulating particles in the blood has been continuously investigated in many fields of biomedical applications. Besides surface modification, many factors of polymermodified particulate architectures affect their biodistribution and blood clearance. For instance, particle size is of great importance in determining the fate of colloidal particles in the body system regardless of whether they are surface-modified or not.62 The regime of surface modification using polymers that can be controlled by chain lengths and surface density on particle surfaces is also critical to controlling the way in which particulate architectures interact with biomolecules.63−66 Whereas it is generally well-accepted that the aforementioned factors impact the physiological properties of particles, it should be realized that the range of structural features of particulate architectures investigated so far is very limited. For example, one study varied the chain lengths of the PEG grafted onto particles to investigate the effect of the chain length on the circulation lifetime in the blood; however, the molecular weights of PEG that were applied ranged only from about 1k to Received: December 28, 2011 Revised: February 8, 2012 Published: February 10, 2012 927
dx.doi.org/10.1021/bm201855m | Biomacromolecules 2012, 13, 927−936
Biomacromolecules
Article
20k.16,52,62−64 With respect to particle size, studies have rarely systematically investigated the effect of particle sizes ranging from nano- to micrometers using a similar regime of surface modification.20,35,65 One of the reasons for the lack of a comprehensive study on these factors is the difficulty of the synthesis of particles with a large variety of structural features. The goal of the present study is to overcome the abovementioned problem and to investigate the physiological properties, such as blood circulation and biodistribution of particulate architectures, of a broad range of particles with a variety of structural features. A more systematic understanding of the effect of structural features on the performance of particles will increase the development and applicability of this rapidly growing area of particle-based biofunctional materials. For this, we will synthesize polymer brush-afforded particles with a variety of structural features based on a surface-initiated living radical polymerization (SI-LRP) technique, which is used to graft polymers onto solid surfaces functionalized with LRP initiators.67−70 SI-LRP allows us to control precisely the molecular weight and polydispersity of graft polymers. Interestingly, SI-LRP provides polymer brushes with an extremely high grafting density of nearly 50% surface occupation by polymer chains.67 Because of a unique structural feature of polymer brushes synthesized via SI-LRP, which is not evident in low density grafting surface, the brush surfaces exhibited extraordinary functions and properties. For example, hydrophilic polymer brushes prepared by SI-LRP showed excellent antifouling characteristics against proteins or cells.71−76 This property encouraged us to adopt SI-LRP for the synthesis of hybrid particles grafted with a hydrophilic polymer brush to achieve the above-mentioned purpose of this study. We previously succeeded for the first time in preparing perfectly dispersed, monodisperse silica particles (SiPs) grafted with a hydrophobic polymer brush of high grafting density by surface-initiated atom transfer radical polymerization (SIATRP).77 Using this system, we could readily control the diameter of the SiP core and the chain length of graft polymers by simply changing the polymerization conditions. This makes it possible to prepare hybrid particles with a broad range of structural features. In addition, because of the high uniformity and high dispersibility of such hybrid particles, we were able to fabricate 2- and 3-D ordered particle arrays.78−84 This product homogeneity is also believed to be of great importance in
achieving the goal of our present work because a well-defined structure provides particles with uniform characteristics, which is crucial for a systematic investigation of the effect of structural features on their functions and properties. In this work, we used poly(poly(ethylene glycol) methyl ether methacrylate) (PPEGMA) brushes to decorate the surfaces of SiPs. PPEGMA is hydrophilic and is therefore readily soluble in aqueous media even under physiological conditions and has an excellent biocompatibility.70,72,73,76 We produced the PPEGMA brushes with a broad range of molecular weights up to ∼1 million (based on poly(methyl methacrylate), PMMA, calibration) using SI-ATRP at normal and high pressures to investigate the dependence of the particle properties on the graft chain length. We used SiPs with sizes ranging from 15 nm to 1.5 μm in diameter to investigate the dependence of the particle properties on the size of the SiP core. The various combinations of the lengths of PPEGMA brushes and sizes of SiP cores enabled the production of hybrid particles with a variety of structural features. A systematic investigation using such hybrid particles were carried out to understand how the structural features of particles impact their physiological properties such as circulation lifetime in the blood and biodistribution. Furthermore, we presented the tumor-targeting ability of hybrid particles that have a long blood-circulation lifetime and tried to perform optical bioimaging of the tumor-targeting by using chemically functionalized polymer brushes.
■
EXPERIMENTAL SECTION
Materials. Poly(ethylene glycol) methyl ether methacrylate (PEGMA, number-average molecular weight Mn = 475, Scheme 1) and 4,4′-dinonyl-2,2′-bipyridine (dNbipy, 97%) were purchased from Aldrich and used without further purification. N,N-Dimethylformamide (DMF, 99.5%) and anisole (99.5%) were purchased from Wako Pure Chemicals, Osaka, Japan. Ethyl 2-bromoisobutyrate (2-(EiB)Br, 98%) was obtained from Tokyo Chemical Industry, Tokyo, Japan. 2,2′Bipyridine (Bipy, 99%) and chloramine T (99%) were used as received from Nacalai Tesque, Osaka, Japan. Copper(I) chloride (Cu(I)Cl, 99.9%), copper(I) bromide (Cu(I)Br, 99.9%), and copper(II) bromide (Cu(II)Br2, 99.9%) were purchased from Wako Pure Chemicals. Methacryloyl chloride (80%) and 4-hydroxyphenethyl alcohol (98%) were obtained from Tokyo Chemical Industry. Acetonitrile (anhydrous, 99.5%) and N-methyl-2-pyrrolidone (99.5%) were procured from Wako Pure Chemicals and Aldrich, respectively. 125I-labeled
Scheme 1. Schematic Representation of Synthesis of Hydrophilic Polymer Brush-Afforded Silica Particles and Chemical Structures of Monomers Used for Surface-Initiated Atom Transfer Radical Polymerizations
928
dx.doi.org/10.1021/bm201855m | Biomacromolecules 2012, 13, 927−936
Biomacromolecules
Article
(TAITEC, Saitama, Japan, Personal H-10) at 60 °C, and, after a prescribed time, t, quenched to room temperature. An aliquot of the solution was taken out for NMR analysis to estimate monomer conversion and for GPC analysis to determine the molecular weight and its distribution. The rest of the reaction mixture was diluted by acetone and centrifuged to collect the polymer-grafted SiPs. The cycle of centrifugation and redispersion in acetone was repeated five times to obtain polymer-grafted SiPs that were perfectly free of unbound (free) polymer. The polymer-grafted SiPs were redispersed in water and centrifuged. The collected particles were then redispersed in a 2 wt % aqueous EDTA solution and centrifuged; this cycle was repeated twice to remove a trace amount of copper. Finally, the particles were purified by repeated cycles of redispersion/centrifugation in water. In some sample preparations, the surface-initiated ATRP was carried out under a high pressure, as per our previous report:86 In brief, a deoxygenated polymerization solution was put in a screw vial made of polyethylene and sealed in a glovebox. The sample was further packed in a polyethylene-coated aluminum sheet to prevent oxygen contamination, put into the chamber of the high-pressure reaction system (HPS-700, Syn Corporation, Kyoto, Japan) preheated at 60 °C, and pressurized up to 200 MPa. In a typical run, the solution polymerization of PEGMA and PEMA in anisole was carried out at 60 °C for 5 h with the feed of PEGMA (9.98 g, 21 mmol), PEMA (22 mg, 0.107 mmol), 2-(EiB)Br (0.4 mg, 2 μmol), Cu(I)Cl (21 mg, 0.212 mmol), dNbipy (173 mg, 0.423 mmol), and the initiator-fixed SiPs with a 130 nm average diameter (100 mg), giving a monomer conversion of 6% and a free polymer with a Mn of 95 000 and a Mw/Mn of 1.12, where Mn and Mw are the number- and weight-average molecular weights, respectively, and Mw/Mn is the polydispersity index. The polymer-grafted SiPs were purified by repeated cycles of centrifugation (12 000 rpm for 30 min) and redispersion in acetone (5 × 50 mL) and in water (3 × 50 mL). Radioiodination of Polymer Brush-Afforded Particles. SiPs surface-modified with brushes of random (statistical) copolymer, poly(PEGMA-r-PEMA), were radioiodinated on the phenol residues of the PEMA units using a chloramine T method.87,88 To 100 μL of a 1 wt % hybrid particle suspension in water was added 5 μL of a Na125I solution (740 MBq/mL) and 100 μL of a solution of chloramine T (0.2 mg/mL) dissolved in a 0.5 M potassium phosphate-buffered solution (pH 7.5) containing 0.5 M NaCl. After agitation by a vortex mixer for 2 min, 100 μL of sodium metabisulfite solution (4 mg/mL) in water was added to terminate the radioiodination reaction. After agitation by the vortex mixer for a further 2 min, the resultant mixture was passed through a column of prepacked Sephadex G25 (GE Healthcare, PD-10) using phosphate-buffered saline (PBS, pH 7.4) as the eluent to remove unreacted, free 125I ions from the 125I-labeled hybrid particles. Evaluation of the Body Distribution of Polymer BrushAfforded Particles after Intravenous Injection. Female BALB/c mice aged 6 weeks (Shimizu Laboratory Supplies, Kyoto, Japan) with and without a tumor mass at the left thigh received an intravenous injection of 100 μL of 125I-labeled hybrid particles solution into the tail vein. Periodically, blood aliquots as low as 50 μL were taken from retro-orbital plexus. The blood radioactivity was measured by a gamma counter (Autowell Gamma System Aloka ARC-301B, Aloka, Tokyo, Japan) to assess the time course of the particle concentration in blood. At a prescribed time, the mice were sacrificed, and their tissues and organs were excised and rinsed quickly with cold water to remove the surface blood. The radioactivity of the excised body tissues was measured by the gamma counter. The urine and feces of mice were collected to measure their radioactivity for excrement evaluation. For mice carrying a tumor mass, the tumor mass was removed to measure its radioactivity. The animal experiments were carried out according to the Institutional Guidance of Kyoto University on Animal Experimentation. The experiment was carried out three times independently for each sample at each time point. Preparation of Fluorescence-Labeled Hybrid Particles. SiPs grafted with a polymer brush composed of poly(PEGMA-r-VBP) (50 mg), which were prepared by the polymerization conditions of sample 12 in Table 1, were dispersed in ethanol (35 mL). Hydrazine hydrate
sodium (Na125I) in a 0.1% NaOH aqueous solution (740 MBq/mL) was obtained from PerkinElmer Life & Analytical Sciences, Waltham, MA. Cy5.5 monofunctional dye (peak excitation: 675 nm, peak emission: 694 nm) was purchased from GE Healthcare U.K., Buckinghamshire, U.K. Ethylenediaminetetraacetic acid disodium salt dehydrate (EDTA, 99%) was purchased from Wako Pure Chemicals. The SiPs (SEAHOSTER, 20 wt % suspension of SiP in ethylene glycol) were kindly donated by Nippon Shokubai, Osaka, Japan. The average diameters of the SiPs were 130 (KE-E10), 290 (KE-E30), 740 (KE-E70), and 1550 nm (KE-E150) with relative standard deviations of
