- Email: [email protected]
Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling
Polymer xxx (2016) 1e11
Contents lists available at ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling Duc Nguyen a, Binh T.T. Pham a, Vien Huynh a, Byung J. Kim a, Nguyen T.H. Pham a, Stephanie A. Bickley c, Stephen K. Jones c, Algirdas Serelis b, Tim Davey b, Chris Such b, Brian S. Hawkett a, * a b c
Key Centre for Polymers and Colloids, School of Chemistry, University of Sydney, NSW 2006, Australia DuluxGroup (Australia), 1970 Princess Highway, Clayton, Victoria 3168, Australia Sirtex Medical Ltd., 101 Miller Street, North Sydney, NSW 2060, Australia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 May 2016 Received in revised form 18 August 2016 Accepted 19 August 2016 Available online xxx
Polymer encapsulation of raw ferrofluids was carried out by free radical emulsion polymerization using macro-RAFT copolymers as stabilizers. The iron oxide cores with positively charged surface at the pH of encapsulation were dispersed in a solution of negatively charged random macro-RAFT copolymer by ultrasonication. Uniform polymer encapsulated super paramagnetic iron oxide nanoparticles (SPIONs) were produced by feeding the mixture of methyl methacrylate (MMA) and n-butyl acrylate (BA) monomers into the dispersion containing water soluble initiator 4,4’-azobis (4-cyanopentanoic acid) (V501). Shell thickness could be controlled by adjusting the amount of coating monomer added. Polymer encapsulation of SPIONs aligned in a magnetic field produced polymer encapsulated SPION rods. Monodispersed sterically stabilized crosslinked polymer encapsulated SPIONs were synthesized using diblock and random macro-RAFT copolymers as stabilizers. Rhodamine labeled SPIONs were found to be noncytotoxic and suitable for cell labeling. © 2016 Elsevier Ltd. All rights reserved.
Keywords: SPIONs Iron oxide Magnetic particles Encapsulation Polymer coating RAFT Control radical Amphiphilic Cells labeling Monodispersed
1. Introduction Superparamagnetic iron oxide nanoparticles (SPIONs) are the most studied nanoparticles due to their relatively low cost and simple synthesis process [1]. They are normally in the form of magnetite [1], maghemite [2] or iron oxides doped with other metal oxides [1e3] which can enhance the particles overall magnetic properties. SPIONs are superparamagnetic nanoparticles which consist of a single magnetic domain and display magnetic properties only in the presence of an external magnetic field [1,2]. These attributes have led to SPIONs finding extensive applications in biomedical science from cell labeling [1,2,4,5], hyperthermia [1e3] to drug delivery [1,2,6]. However, iron oxide nanoparticles in their original form have a weakly charged surface which makes
* Corresponding author. E-mail address: [email protected] (B.S. Hawkett).
them susceptible to aggregation [1,7]. Surface modification involving attachment of stabilizers such as surfactant-like organic substances [1,2] or hydrophilic polymers [1,2] can increase their stability. The provision of steric stabilizers makes the SPIONs more stable, especially in high salt biological media [1,2]. KCPC has extensive experience in modifying SPIONs with macro-RAFT copolymers for cell labeling and drug delivery towards cancer treatment [8e12]. The modification process simply involves the dispersion of SPIONs in a macro-RAFT stabilizing solution. The macro-RAFT stabilizers are diblock in nature and consist of a short block that contains acid groups such as carboxylic [11] or phosphonic [9,13] which anchor the macro-RAFT copolymer to the SPION surface while the second block comprises relatively long blocks of polyethylene oxide or polyacrylamide, which serve as steric stabilizers for the SPIONs [9e11]. The modified SPIONs have been found to be extremely stable in highly concentrated salt solutions and biological media. However, this dispersion process still has an inherent drawback which is the exposure of the SPION
http://dx.doi.org/10.1016/j.polymer.2016.08.064 0032-3861/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
2
D. Nguyen et al. / Polymer xxx (2016) 1e11
surface to the dispersed media [9e14]. Furthermore, Jain et al. reported desorption of the polymer stabilizers when macro-RAFT diblocks of poly (acrylic acid) and polyacrylamide at pH greater than 12 and lower than 4 [11]. He attributed this desorption to full protonation of carboxylic groups at pH lower than 4. For pH greater than 12, the desorption was thought to be due to electrostatic repulsion between negatively charged surface and negatively charged diblock copolymers. This lead to reduced particle stability, affecting performance of the modified SPIONs when macro-RAFT diblocks of poly (acrylic acid) and poly (acrylamide) are used as stabilizers at such pH ranges [11]. One possible solution for desorption of this particular type of diblocks is to chain extend the living ends of macro-RAFT stabilizers, making them less labile on the particle surface. The chain extension process can be carried out using polymer encapsulation [15] which may also form a uniform thin polymer shell insulating the iron oxide surface from dispersing media. Polymer encapsulation of iron oxide nanoparticles has been carried out by non-RAFT free radical emulsion [16,17], miniemulsion [18] and microemulsion [19] polymerization but suffered from complexity, low efficiency, uncontrolled polymer shell thickness and only applicable on a small scale. Therefore, it is in our interests to develop a ready to scale-up method which can efficiently encapsulate SPIONs. In this work, we will explore the application of our previously developed polymer encapsulation techniques to encapsulate SPIONs within polymer shells [6,15]. The encapsulation process is based on RAFT mediated emulsion polymerization using macro-RAFT copolymers as stabilizers. In this process, nanoparticles are first dispersed in the macro-RAFT copolymer solution by sonication or milling followed by free radical emulsion polymerization in which controlled chain extension of adsorbed macro-RAFT stabilizers leads to formation of uniform polymer shells encapsulating nanoparticles in the centers of polymer particles. Throughout the process, particle stability is maintained by anchored charges or hydrophilic groups from the macro-RAFT stabilizers. The process is simple, versatile and has been successfully adapted to encapsulate a range of particulate materials [20e26]. In this work, polymer encapsulation of SPIONs will be carried out using macro-RAFT copolymers as stabilizers. The superparamagnetic properties will be taken advantage of to synthesize polymer encapsulated SPION nanorods. Incorporation of steric stabilizers to produce encapsulated SPIONs which are stable in high salt and biological media will be demonstrated. Such particles will be further modified by fluorescent dye attachment for cell labeling. 2. Experimental 2.1. Reagents Milli RO water was used in the synthesis of latexes and acrylic acid-containing RAFT agents. Acrylic acid (AA) (Aldrich) was purified by distillation under reduced pressure. Butyl acrylate (BA) (Aldrich) and methyl methacrylate (MMA) (Aldrich), divinyl benzene (80%, DVB) (Aldrich) and styrene (Sty) (Aldrich) had inhibitor removed by passing them through an inhibitor-removal column (Aldrich). Rhodamine B isothiocyanate (Aldrich), PBS buffer solution (Aldrich), N-(3-dimethylaminopropyl)-N0 -ethylcarbodiimide hydrochloride (EDC) (Aldrich), n-hydroxy succinimide (NHS) (Aldrich), 2,2'-(ethylenedioxy)bis-(ethylamine) (Diamine) (Aldrich), ammonium hydroxide (NH4OH) (Aldrich), carbon disulfide (Aldrich), acetone (Aldrich), tetrabutylammonium bromide (Aldrich), 2-bromopropanoic acid (Aldrich), dimethyl sulfoxide (DMSO) (Aldrich), diethyl ether (Aldrich, Univar), hydrochloric acid (Aldrich, Univar), sodium sulfate (Aldrich), 2,2’-
azobisisobutyronitrile (AIBN) (Wako), acrylamide (AAm) (Aldrich) and 4,4’-azobis (4-cyanopentanoic acid) (V-501) (Wako) were used as received. Dioxane (Aldrich) was distilled under reduced pressure before use. Super paramagnetic iron oxide nanoparticles (SPIONs) dispersion (5.25%) were obtained from by Sirtex Medical Ltd and used as supplied. The RAFT agent, 2-([(butylsulfanyl)carbonothioyl]sulfanyl) propanoic acid (PABTC), was synthesized as previously described [27]. Synthesis of short chain poly (BA-co-AA) macro-RAFT PABTC(BA7-co-AA10) copolymers using PABTC. The amphiphilic macro-RAFT copolymers were prepared as reported in our previous work [15]. PABTC (1.86 g, 7.8 mmol), AIBN (0.064 g, 0.39 mmol), AA (5.63 g, 78.1 mmol), BA (7.51 g, 58.6 mmol) in dioxane (15.00 g) was prepared in a 50 mL round-bottomed flask. 1,3,5-trioxane (0.09 g, 0.9 mmol) was added as an internal standard for proton NMR. This was stirred magnetically and sparged with nitrogen for 10 min. The flask was then heated at 70 C for 7 h under constant stirring. The final copolymer solution was 50.7% solids. The copolymers were characterized by 1H NMR (300 MHz, Acetoned6) and size exclusion chromatography (SEC). Based on 1H NMR, the conversion of BA and AA was around 97% and therefore the theoretical molecular weight is 1869.8 g mol1. The copolymers were characterized by MALDI (Bruker Daltonics Apex Ultra 7 T Fourier Transform Ion Cyclotron Resonance (FTICR) Mass Spectrometer and found to be narrowly distributed in a range from 1000 to 3000 (g/mole), peaked at 2000 (g/mole). 2.2. Synthesis of poly((acrylic acid)-block-poly(acrylamide)) macro-RAFT PABTC-(AA5-block-AAm60) using PABTC A solution of PABTC (0.74 g, 3.1 mmol), V501 (0.05 g, 0.2 mmol), acrylamide (13.40 g, 188.6 mmol) in dioxane (17.2 g), 1,3,5-trioxane (0.01 g) and water (20.5 g) was prepared in a 100 mL round bottom flask. This was stirred magnetically and sparged with nitrogen for 15 min. The flask was then heated at 70 C for 4 h. By 1H NMR, conversion was determined to be at 99.5% after 4 h. At the end of this period, AA (1.12 g, 15.6 mmol) and V501 (0.05 g, 0.2 mmol) was added to the flask. The mixture was deoxygenated and heating was continued at 70 C for a further 3 h s. The copolymer solution had 28.8% solids and conversion was found to be 96% by 1H NMR. 2.3. Encapsulation of SPIONs with poly(MMA-co-BA) using PABTC(BA7-co-AA10) copolymers as stabilizers Poly (MMA-co-BA) encapsulated SPIONs were synthesized using the following procedure. In a 100 ml beaker, macro-RAFT PABTC(BA7-co-AA10) solution (1.20 g, 0.3 mmol) was dispersed in water (60.1 g) to yield a cloudy yellow solution of pH 4. Ammonium hydroxide (2.8% solution in water) was added to the macro-RAFT solution to raise the pH to 9.0. This was followed by 1 min ultrasonication using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% to obtain a clear yellow solution with pH 7.0. To this solution, SPIONs (20.11 g) was added drop wise and thoroughly dispersed by ultrasonication using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% for 30 min. After sonication, the dispersion pH was measured at 6.5 and it was then transferred to a 100 mL round bottom flask containing V501 (0.04 g, 0.15 mmol) which was subsequently sealed and purged with nitrogen for 10 min. The whole flask was then immersed in an oil bath with a temperature setting of 70 C and was magnetically stirred. A deoxygenated 10:1 (weight ratio) solution (1 mL, 0.94 g) of methyl methacrylate (MMA), butyl acrylate (BA) was injected into the flask, while in the 70 C oil bath, at a rate of 2.0 mL/hour. This feed rate
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11
was chosen based on our previous works [15,23] on pigment encapsulation in consideration of the amount of initiator used. After monomer addition, the heating continued for another five and half hours to produce stable brown latex of the encapsulated SPIONs with 2.1% solids. 2.4. Further shell growth of the polymer encapsulated SPIONs with polystyrene The poly (MMA-co-BA) shells encapsulating SPIONs were further grown with polystyrene using the following procedure. A solution containing styrene (1.00 g) and AIBN (0.04 g) was prepared in a 100 mL round bottom flask. To this solution, the polymer encapsulated SPIONs latex (10.10 g) and water (40.11 g) was added and thoroughly mixed for 10 min by magnetic stirring. The flask was subsequently sealed and purged with nitrogen for another 10 min before being heated in an oil bath at 80 C for 3 h under constant stirring. The final brown latex was found to be stable and contained 2.2% solids. 2.5. Synthesis of poly(MMA-co-BA) encapsulated SPION nanorods using PABTC-(BA7-co-AA10) copolymers as stabilizers Poly (MMA-co-BA) encapsulated SPION nanorods were synthesized using the following procedure. In a 100 ml beaker, PABTC(BA7-co-AA10) macro-RAFT solution (0.6 g, 0.15 mmol) was dispersed in water (31.2 g) to yield a cloudy yellow solution of pH 4. Ammonium hydroxide (2.8% solution in water) was added to the macro-RAFT solution to raise the pH to 9.0. This was followed by 1 min ultrasonication using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% to obtain a yellow clear solution with pH 7.5. To this solution, SPIONs (10.10 g) were added drop wise and thoroughly dispersed by ultrasonication using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% for 30 min. A magnetic stirrer was used to mix the dispersion throughout the ultrasonication. After sonication, the dispersion pH was measured at 7.6 and it was then transferred to a 100 mL round bottom flask containing V501 (0.03 g, 0.15 mmol) which was subsequently sealed and purged with nitrogen for 10 min. The whole flask was then immersed in an oil bath with a temperature setting of 70 C and was magnetically stirred. A deoxygenated 10:1 (weight ratio) solution (5 mL, 4.69 g) of methyl methacrylate (MMA), butyl acrylate (BA) was injected into the flask, while in the 70 C oil bath, at a rate of 1.0 mL/h. Since SPION nanorods were larger, this slower monomer feed rate (compared to the previous SPION encapsulation experiment) was chosen to minimize the amount of unreacted monomer present in the system at any given time. This helped to avoid aggregation of the nanorod dispersion by the presence of excessive numbers of monomer droplets. After monomer addition, the heating continued for another five and half hours to produce stable brown latex of the encapsulated SPIONs. Encapsulation and modification of SPIONs with polymers using PABTC-(BA7-co-AA10) and PABTC-(AA5-block-AAm60) macro-RAFT copolymers as electrostatic and steric stabilizers, respectively. 2.6. Poly(MMA-co-BA) encapsulation of SPIONs Poly (MMA-co-BA) encapsulated SPIONs were synthesized using the following procedure. In a 100 ml beaker, PABTC-(BA7-co-AA10) solution (0.32 g, 0.06 mmol) and PABTC-(AA5-block-AAm60) macroRAFT solution (0.40 g, 0.02 mmol) was dispersed in water (50.6 g) to yield a cloudy yellow solution of pH 3.5. Ammonium hydroxide (2.8% solution in water) was added to the macro-RAFT solution to raise the pH to 8.5. This was followed by 1 min of ultrasonication
3
using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% to obtain a yellow clear solution with pH 7.2. To this solution, SPIONs (10.0 g) was added drop wise and thoroughly dispersed by ultrasonication using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% for 15 min. After sonication, the dispersion, which had a pH of 7.5, was transferred to a 100 mL round bottom flask containing V501 (0.03 g, 0.11 mmol) which was subsequently sealed and purged with nitrogen for 10 min. The whole flask was then immersed in an oil bath with a temperature setting of 70 C and was magnetically stirred. A deoxygenated 10:1 (weight ratio) solution (5 mL, 4.69 g) of methyl methacrylate (MMA), butyl acrylate (BA) was injected into the flask, while in the 70 C oil bath, at a rate of 2.0 mL/h. After monomer addition, the heating continued for another five and half hours to produce stable brown latex of the encapsulated SPIONs. The latex was characterized by DLS and TEM. 2.7. Crosslinking sterically stabilized polymer encapsulated SPIONs with DVB The Poly (MMA-co-BA) shells of the encapsulated SPIONs were crosslinked with DVB (5.6% by weight on polymer) using the procedure described in Reference 23 [23]. A solution containing DVB (0.26 g) and V501 (0.04 g) was prepared in a 100 mL round bottom flask. To this solution, the polymer encapsulated SPIONs latex (48.3 g) was added and thoroughly mixed for 10 min by magnetic stirring. The flask was subsequently sealed and purged with nitrogen for another 10 min before heated in an oil bath at 80 C for 2 h under constant stirring. The final brown latex was found to be stable and was characterized by DLS and TEM. 2.8. Conjugation of sterically stabilized crosslinked polymer encapsulated SPIONs with Rhodamine B isothiocyanate The sterically stabilized crosslinked polymer encapsulated SPIONs were conjugated with Rhodamine B isothiocyanate using the following procedure. A latex dispersion containing the above crosslinked polymer encapsulated SPIONs (10.1 g) was prepared in a 25 mL beaker. To this dispersion, 0.03 g of 1% NHS solution was added, followed by 0.03 g of 1% EDC solution. The dispersion was thoroughly mixed by magnetic stirring. After 60 min, 3 g of 0.25% diamine was added and the mixing was continued for another 60 min. To this dispersion, a Rhodamine B isothiocyanate dye (0.0017 g) in 1 g of DMSO was then added and the mixture was stirred for 2 h. After reaction, free dye was removed by repeated centrifugations (3 15 min at 14500 rpm) and washing (3 times) with mili-Q water. The washed SPION deposit was re-dispersed in 25 g of water by ultrasonication for 10 min at 30% amplitude followed by purification by dialysis against water to obtain a stable Rhodamine labeled SPIONs dispersion with 1.2% solids. PBS was used to prepare a stock dispersion for cell imaging. 2.9. In vitro evaluation for the bioactivity of polymer encapsulated iron oxide nanoparticles 2.9.1. Cell culture DLD-1 (ATCC, CCL-221) cells were maintained as confluent monolayers in Advanced DMEM (Gibco, 12491e015), supplemented with 2% (v/v) FCS. The cells were incubated under standard culturing conditions (at 37 C with 5% (v/v) CO2 under humidified conditions). Cells were routinely sub-cultured by removal of the medium, washing with PBS (In vitro technologies, IVT3001302) and detached from the culture-ware by 0.25% (w/v) trypsin (SAFC Biosciences, 59430C). The cell suspension was collected in a centrifuge
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11
tube with the addition of complete growth medium and spun down at approximately 1000 g for 3 min. The supernatant was discarded and the cells re-suspended in fresh growth medium. Cells were then counted on a haemocytometer and an appropriate number of cells were seeded for further experiments and subculturing. 2.9.2. Cytotoxicity screening in monolayer cells A 1 105 cells/well suspension was seeded into each well of a 96-well plate (Corning, 3599) and incubated for 24 h under standard culturing conditions. Prior to the assay, polymer encapsulated iron oxide nanoparticles were sterilized by filtration through a 0.45 mm filter. The dose volume of the nanoparticles was kept below 24% of the total volume and diluted in complete culture medium to be added to quintuplet wells spanning a 2-log range of final concentrations (500, 250, 125, 62.5, 31.25, 15.625 and 7.8125 ppm). The cells were then incubated for 72 h before 1.0 mM MTT (thiazolyl blue tetrazolium bromide, Alfa Aesar, L11939) was added and further incubated for 4 h. The media was removed and 150 mL of DMSO was added to each well. The absorbance at 600 nm was then measured for the plate in a Victor3V microplate reader (Perkin Elmer, 1420). IC50 values were determined as the drug concentration that reduced the absorbance to 50% of that in untreated control wells. 2.9.3. Confocal imaging of cells treated with the Rhodamine labeled SPIONs 1 105 cells/well suspension was seeded into each well of a 12well glass bottom plate (Mattek, P12G-1.5e10-F.s) and incubated for 24 h under standard culturing conditions. 20 ppm filter sterilized nanoparticles were dosed onto the DLD-1 cells for 24 h 20 min prior to imaging, cells were treated with 1 mg/mL Hoechst 33258 (Sigma, 94403-1 ML). Excess nanoparticles and Hoechst were washed away with PBS and fresh culturing medium was added. Confocal images were generated on the Olympus FV-1000 microscope using UPLAPO 40 oil objective (N.A.: 1.00) with the sampling speed of 2 ms/pixel. Rhodamine B signals were acquired with 559 nm laser excitation and 570e670 nm emission while Hoechst was excited with the 405 nm laser and emissions collected from 425 to 475 nm. 2.9.4. Particle characterizations Particles were characterized by TEM (Philips CM120 Biofilter), dynamic light scattering (DLS) (Zetasizer Nano ZS, Malvern) or by TGA (TA Instruments TGA Discovery). Zeta potential measurements were also performed on the Malvern Zetasizer Nano ZS using the Smoluchowski equation as a basis for calculation. For TGA dry sample preparation, the polymer encapsulated SPION particles were separated from the homo-polymer particles by centrifugation, redispersed and washed three times with water prior to drying in an oven at 100 C for 12 h. Fluorescence of the Rhodamine labeled SPIONs was characterized using a Shimadzu (Kyoto, Japan) RF5301 PC spectrofluorometer at 550 and 559 nm excitation wavelengths. Magnetic measurements were conducted using a Quantum Design 9 T PPMS (Physical Property Measurement System) equipped with a VSM (Vibrating Sample Magnetometer). Powdered samples were housed in a polypropylene sample holder held in place with a brass half tube holder. All measurements were carried out at 300 K. The magnetic moment was measured as a function of magnetic field strength and then converted to mass magnetisation (emu/g) by dividing the moment (emu) by the mass of SPIONs in each sample (g). Raw SPIONs (12.36 mg) and polymer encapsulated SPIONs (10.51 mg, 73.1% iron oxide, analyzed by TGA) were dried in an oven overnight at 100 C prior to magnetic measurements using
a 0.4 T electromagnet. 3. Results and discussions 3.1. SPIONs The aqueous SPION dispersion was supplied as a dark brown dispersion and observed to be stable in their supplied form at pH 4. The SPIONs were found to have strong surface positive charge of 44 mV at pH 4.2 (Fig. 1). TGA shows that the SPIONs contained no organic stabilizers, with negligible weight loss at 700 C (Fig. 2, black curve). Therefore, the positive surface charge was attributed to the presence of Metal-OHþ 2 groups on the surface of the nanoparticles as discussed by Lucas et al. [7] The positive surface charge together with negative counter ions helped to maintain SPION stability at low pH. However, as the pH increased, the positive charge decreased due to de-protonation and formation of more OH groups [7]. The reduction in surface charge destabilized the SPIONs, leading to visual aggregate formation. As the pH was raised to 7.2, the surface approached the point of zero charge at which 1.4 mm average aggregates were measured by DLS. By TEM (Fig. 3), it was found that SPIONs are not spherical in shape and quite polydispersed, with particle diameters ranging from 10 to 50 nm. Broad size distribution was also consistent with DLS results in Table 1 in which average particle diameters were measured to be 89 nm with a high PDI of 0.410. Since the DLS measurement gave a much larger size than what observed by TEM, it was likely that a large number of SPIONs exist as aggregates despite the high positive surface charge. This observation is common with large SPIONs due to the magnetic attraction between iron oxide particles even in zero magnetic field which can be overcome by surface modifications [28]. 3.2. Dispersion of SPIONs in macro-RAFT solution Dispersion of SPIONs in macro-RAFT solution was carried out in the same manner as previously described [15] and is shown in Fig. 4. Short chain macro-RAFT copolymers were solubilized at pH 6.5 prior to the dispersion. Once mixed with the iron oxide nanoparticles, negatively charged carboxylic functional groups facilitate the copolymer adsorption onto the positively charged particle surface via opposite charge interactions [29]. In the case of iron oxide, polymer adsorption was also possible via complex formation between carboxyl groups and iron atoms on particle surface [11]. After sonication, the SPION surface was covered with the macroRAFT copolymers, imparting strong negative charge as measured by Zeta potential of 54 mV at pH 6.5 (Table 1). This adsorption was further confirmed by TGA (Fig. 2, blue curve) where weight loss due to the copolymer was found to be about 10%. As a result, the dispersed SPIONs were more stable in the aqueous phase and less
50 40 30 20 10 0
8
7
6
5
4
Zeta PotenƟal (mV)
4
3
pH Fig. 1. Zeta potentials of SPIONs versus pH of dispersions.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11
5
Fig. 2. TGA curves of raw (black), dispersed (blue), poly (MMA-co-BA) encapsulated (green) and poly (Sty/MMA/BA) encapsulated (red) SPIONs. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
RAFT solution. This was probably triggered by the reduction in surface charge when the pH was suddenly increased from 4 to 6.5, making SPIONs less stable before they had a chance to interact with macro-RAFT stabilizers. The subsequent ultrasonication helped to break up some aggregates but some remained, resulting in an overall increase in particle size as reported in Table 1. 3.3. Polymer encapsulation of SPIONs
Fig. 3. TEM image of raw, not encapsulated SPIONs.
susceptible to pH change. As reported in Table 1, there was a slight increase in average particle size to 104 nm. However, TEM evidence in Fig. 5 showed that the adsorption of short chain macro-RAFT copolymers did not alter the size of individual iron oxide nanoparticles since they were similar to ones in Fig. 3. Therefore, the average size increase was probably due to aggregate formation during the mixing of the iron oxide dispersion with the macro-
To encapsulate the dispersed SPIONs, BA/MMA monomer mixture was slowly fed into the dispersion in the presence of the water soluble initiator at 70 C. The polymerization mechanism was thoroughly discussed in our previous work [15] and depicted here in Fig. 4. The process followed RAFT controlled free radical emulsion polymerization, involving a continuous and even growth of the macro-RAFT copolymers which was adsorbed onto the nanoparticle surface. Such controlled polymer growth produced uniform polymer layers encapsulating the SPIONs. As seen in Fig. 6A, thin and even polymer shells of 5e6 nm were observed on all iron oxide particles and aggregates. This was further confirmed by TGA evidence presented in Fig. 2 (green curve) in which weight loss due to the polymer on the SPION surface was found to be about 26.9%. Throughout the encapsulation process, SPIONs were stabilized by the negatively charged carboxyl groups from the adsorbed macroRAFT copolymers. As polymer shell thickness increased, the negative charge was maintained on the outer surface of the encapsulated iron oxide nanoparticles. As shown in Table 1, the zeta potential measured for the encapsulated particles was 55 mV, the same as for the dispersed SPIONs. Particle stability provided by anchored carboxyl groups allowed further shell growth without the
Table 1 Evolution of Zeta potentials and sizes of SPIONs particles throughout the encapsulation process.
Sirtex iron oxide nanoparticle Dispersed iron oxide nanoparticle in macro-RAFT Poly(MMA-co-BA) encapsulated iron oxide nanoparticle Poly(Styrene/MMA/BA) encapsulated iron oxide nanoparticle
Z potential (mV)
Electrophoretic mobility (mmcm/Vs)
pH of dispersion
Size (nm)
PDI
44 ¡54 ¡55 NA
3.489 ¡4.14 ¡4.303 NA
4.2 7.3 6.7 NA
89 104 94 90
0.41 0.26 0.25 0.12
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
6
D. Nguyen et al. / Polymer xxx (2016) 1e11
Fig. 4. Dispersion and encapsulation of SPIONs using PABTC-(BA7-co-AA10) macro-RAFT copolymer.
Fig. 5. SPIONs after dispersion in macro-RAFT solution.
increased to more than 60%. The uniform polymer layers in Fig. 6B indicated an even distribution of the monomer and initiator across particles. From 6A to 6B, it was observed that encapsulated SPIONs became more spherical as more monomer was added. This was similar to the case of Titanium dioxide pigment encapsulation where more spherical shapes were obtained with increasing shell thickness [15]. This was attributed to the minimization of particle surface area as particle the size grew. It was also found that particle sizes became more monodispersed as shown in Fig. 6B. This was in agreement with DLS measurements (Table 1) in here particle size was shown to be largely similar while size distribution consistently decreased as monomer was added. This was probably due to removal of unstable large particles as aggregates during the emulsion polymerization. Particle transformation from irregular to spherical shapes reduced the discrepancy between measured dimensions per particles and therefore was important in reducing polydispersity. In Fig. 6B, homogenous polymer particles containing no SPIONs with sizes from 28 to 33 nm were clearly present. They were formed during the emulsion polymerization process probably due to the presence of labile macro-RAFT copolymers and other water soluble species from decomposition of V501 initiator.
Fig. 6. Polymer coated SPIONs: A) with MMA/BA (10/1); B) further growth with styrene.
need of additional surfactants. Using poly (MMA-co-BA) coated SPIONs as seeds, free radical emulsion polymerization of styrene was carried out. Relatively hydrophobic AIBN initiator was chosen due to its high solubility in styrene. The reaction was completed in a very simple manner in which the monomer swollen polymer shells acted as the locus of polymerization. As shown in Fig. 6B, polymer shells grew to 16e17 nm in thickness during this process. This was reconfirmed by TGA (Fig. 2, red curve) in which the polymer content of poly (Sty/MMA/BA) encapsulated SPIONs significantly
Growth of the polymer particles competed with the encapsulation process for monomer and free radicals, reducing the efficiency of the later. Therefore, it is desirable in the future to minimize polymer particle formation by optimizing the amount of macro-RAFT and V501 required for the encapsulation process. 3.4. Synthesis of polymer encapsulated SPION nanorods One key feature of SPIONs particles is their super paramagnetic
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11
Fig. 7. Spikes formation from poly (MMA-co-BA) encapsulated SPIONs under influence of a magnet.
7
the ultrasonic dispersion of SPIONs in the macro-RAFT copolymer solution. Due to the close proximity between aligned individual iron oxide nanoparticles, inter-particle adsorption of the macroRAFT copolymers probably occurred. They acted as glue holding individual nanoparticles together in alignment as SPION rods even when the magnetic field was removed. Polymer encapsulation of such aggregates encased SPION rods in centers of polymer shells as shown by TEMs in Fig. 10A and B. In these images, polymer encapsulated SPION rods with lengths from 100 to 500 nm were observed. However, there were a large number of spherical polymer encapsulated SPIONs. This indicated a non-uniform magnetic field during dispersion especially for regions distanced from the spin bar where the field was too weak to align SPIONs. Possible solutions to exert a uniform magnetic field during dispersion to raise the yield of SPION nanorods will be investigated in future work. 3.5. Synthesis of sterically stabilized polymer encapsulated SPIONs
properties in which they interact with the magnetic field once applied [8,11]. It is therefore critical that encapsulated SPIONs continue to behave in this manner after encapsulation. Fig. 7 clearly demonstrated that poly (MMA-co-BA) encapsulated SPIONs still maintained their magnetic properties by forming sharp spikes once the concentrated dispersion was placed on top of a permanent magnet. Spike formation is typical ferrofluid behavior and was due to alignment of encapsulated SPIONs under the influence of the magnetic field generated by the magnet [8,10,11,13]. Magnetic measurements (Fig. 8) found that mass magnetization of polymer encapsulated SPIONs saturated at a lower value than that of the raw SPIONs. This reduction in magnetic property was attributed to the presence of non-magnetic polymer layers which made the magnetic diameter smaller than physical diameter of particles [11]. SPION alignments under the magnetic field were further taken advantage of to synthesize polymer encapsulated SPION nanorods in the process shown in Fig. 9 [8]. In this process, a magnetic field was generated by a magnetic spin bar and magnetic stirrer during
For polymer encapsulated SPIONs to be used for biological applications, it is essential for the final particles to be stable in high salt concentrations [9,12]. Unfortunately, electrostatically stabilized polymer encapsulated iron oxide nanoparticles are not suitable for this purpose. They become unstable once dispersed in high salt environment due to the suppression of the electrical double layer which in turn reduces the repulsive force between particles. In previous work [8,11], macro-RAFT diblock copolymer RAFT PABTC(AA10-block-AAm14) was successfully used to stabilize SPIONs in 2 M NaCl solution. In our case, a similar macro-RAFT diblock copolymer RAFT PABTC-(AA5-block-AAm60) was utilized in conjunction with macro-RAFT copolymer PABTC-(BA7-co-AA10) for the dispersion and encapsulation of SPIONs as depicted in Fig. 11. As in the case of the macro-RAFT random copolymers, adsorption of the diblock copolymers onto the SPION surface was likely via opposite charge interaction or complex formation. On the other hand, the long poly (acrylamide) blocks of the adsorbed macroRAFT copolymers would provide steric stability by forming
Fig. 8. Mass magnetization of raw and poly (MMA-co-BA) encapsulated SPIONs as a function of magnetic field using a 0.4 T electromagnet.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
8
D. Nguyen et al. / Polymer xxx (2016) 1e11
Fig. 9. Polymer encapsulation of SPION rods using macro-RAFT copolymers as the stabilizers.
Fig. 10. Poly (MMA-co-BA) encapsulated SPION nanorods.
hydrophilic hairy layers which prevented surface contact between particles. Polymer encapsulation was carried out in the same manner as previously discussed. A monomer mixture of MMA/BA (10/1 by weight) was slowly fed into the dispersion at 70 C in the presence of V501 initiator. Chain extension of macro-RAFT copolymers by free radical emulsion polymerization produced uniform polymer shell encapsulated SPIONs as shown in Fig. 12A. Particles were observed to be similar in size with diameters varying from 70 to 78 nm. By DLS, it is reported in Table 2 that final latex particles were monodispersed (0.07 PDI), having an average hydrodynamic diameter of 125 nm. The larger diameter measured by DLS was expected due to the extension of the poly (AAm) blocks in the aqueous medium. In the dry state, they collapsed in volume and therefore were not observable by TEM. The polymer encapsulated SPIONs were further strengthened by crosslinking with DVB. As shown in Fig. 12B, the crosslinked polymer encapsulated SPIONs slightly increased in size to a range from 78 to 94 nm. DLS measurement in Table 2 reported monodispersed particles (0.04 PDI) with diameters almost unchanged from before the crosslinking reaction. Compared to poly (MMA-co-BA), crosslinked polymer networks should make encapsulating shells harder and less
susceptible to solvents. 3.6. Dye conjugation with sterically stabilized polymer encapsulated SPIONs and cell labeling As previously discussed, the key target for using steric stabilizer macro-RAFT PABTC-(AA5-block-AAm60) was to have encapsulated SPIONs stable in high salt environments. As reported in Table 2, both uncrosslinked and crosslinked SPIONs were found to be relatively stable in PBS solution with measured average diameters slightly lower than those in water. The small reductions in size were probably due to suppression of electrical double layer by the high ionic strength. This may lead to some aggregation between particles as shown by increases in PDI. However, particle stability was markedly improved compared to the charge stabilized encapsulated SPIONs. This was readily observed (Fig. 13B) a sediment of large aggregates at the bottom of vial B once charge stabilized encapsulated SPIONs were mixed with PBS. The sterically stabilized polymer encapsulated SPIONs contain an abundance of carboxylic groups on the surface which can be readily conjugated with a fluorescent dye to produce a
Fig. 11. Poly (MMA-co-BA) encapsulated SPIONs using PABTC-(BA7-co-AA10) and PABTC-(AA5-block-AAm60) macro-RAFT copolymers as stabilizers.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11
9
Fig. 12. Polymer encapsulated SPIONs using macro-RAFT PABTC-(AA5-block-AAm60) and PABTC-(BA7-co-AA10) as stabilizers: A) poly (MMA-co-BA) encapsulated SPIONs; B) crosslinked poly (MMA-co-BA) encapsulated SPIONs.
Table 2 Sizes of steric stabilized polymer encapsulated SPIONs before and after conjugation with Rhodamine B isothiocyanate. Size (nm) PDI Poly(MMA-co-BA) encapsulated SPIONs in water Poly(MMA-co-BA) encapsulated SPIONs in PBS Crosslinked poly(MMA-co-BA) encapsulated SPIONs in water Crosslinked poly(MMA-co-BA) encapsulated SPIONs in PBS Dye conjugated polymer encapsulated SPIONs in water Dye conjugated polymer encapsulated SPIONs in PBS
125 123 124 121 141 145
0.07 0.11 0.04 0.09 0.03 0.08
monodispersed superparamagnetic imaging agent. As depicted in Fig. 14, the conjugation process followed a well-used method which
involved reacting carboxylic functional groups with EDC/NHS/ Diamine to produce active primary amines on the particle surface [9]. Once formed, such amines groups were reacted with isothiocyanate groups of Rhodamine B isothiocyanate [30], producing dye conjugated polymer encapsulated SPIONs. As shown in Table 2, dye conjugation slightly increased the hydrodynamic diameter to 141 nm while particles maintained their monodispersity with a PDI of 0.03. More importantly, the particles were still stable in PBS solution with particle size remaining unchanged at 145 nm and almost no aggregates were observed (Fig. 13A). The size increase was probably due to attachment of bulky dye molecules especially to the carboxyl group at poly (AAm) polymer chain ends. The modified particles still maintained their magnetic properties. As shown in Fig. 13C, the dye conjugated particles were readily
Fig. 13. Polymer encapsulated SPIONs in PBS: A) with macro-RAFT PABTC-(AA5-block-AAm60) as the steric stabilizer; B) without steric stabilizer; C) sterically stabilized dye conjugated polymer encapsulated SPIONs were extracted by a magnet.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
10
D. Nguyen et al. / Polymer xxx (2016) 1e11
Fig. 14. Dye conjugation of steric stabilized polymer encapsulated SPIONs.
Fig. 15. Fluorescent emission spectrum of Rhodamine B labeled SPIONs at 100 mg/mL.
extracted by a permanent magnet, suggesting their suitability for separation applications. Dye conjugated polymer encapsulated SPIONs were found to be fluorescent under excitation with UV light (Fig. 15). Similar emission spectra at different excitation wavelength showed that the fluorescent dye (fluorophore) was chemically attached to the surface of particles and not due to adsorbed dye molecules. Incorporation of a fluorophore onto the encapsulated nanoparticles allows for visualization and tracking in biological systems. Unconjugated free Rhodamine B isothiocyanate tends to localize primarily in the mitochondria and partly in the cytoplasm due to its positive charge and lipophilicity [31]. When the fluorophore is part of the encapsulated SPIONs, it showed a vesicular localization (Fig. 16), which could be due to its mode of entry into the cells. The encapsulated SPIONs would enter the cells through macropinocytosis more so than the receptor-mediated endocytosis due to a size restriction of approximately 120 nm via the latter route of entry [32]. As the local pH becomes acidic within these vesicles, many pH sensitive nanoparticles degrade and cause cytotoxic effects within the cells. Since the encapsulating shells composed of crosslinked polymer
Fig. 16. Confocal images of DLD-1 cells treated with 20 mg/mL Rhodamine B labeled SPIONs. Scale bar represents 20 mm.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11 Table 3 IC50 values (mg/mL) of the Rhodamine B labeled SPIONs in DLD-1 cells at 72 h. Cell line
Cell type
IC50 (mg/mL)
Highest concentration tested (mg/mL)
DLD-1
Human colorectal adenocarcinoma
>1000
1000
networks should help protect the SPIONs from the external environment and, the stability of the SPIONs should be further enhanced. As such, when DLD-1 cells were treated with the encapsulated SPIONs, they were non-cytotoxic up to a concentration of 1000 mg/mL (Table 3), which is about hundred fold higher than the concentration relevant for in vitro biological studies. Increasing the stability of nanomaterials by encapsulation is an attractive strategy that could strengthen the transition of the nanomaterials onto an in vivo testing and beyond.
4. Conclusions In this work, we successfully dispersed and encapsulated SPIONs particles with uniform polymer shells using macro-RAFT copolymers as stabilizers. The encapsulation process was carried out by free radical emulsion polymerization by chain extensions of macro-RAFT stabilizers. Polymer encapsulated SPIONs were found to retain their magnetic properties, displaying ferrofluid spikes under the influence of a magnet. Polymer encapsulated SPION nanorods were also successfully synthesized based on alignment of SPIONs in a magnetic field. For the cell labeling application, monodispersed sterically stabilized polymer encapsulated SPIONs were produced by incorporating diblock steric stabilizers during the dispersion and encapsulation stages. Such particles were found to be stable in PBS solution and were further modified by conjugation with Rhodamine B isothiocyanate. The dye conjugated polymer encapsulated SPIONs displayed no cytotoxicity and successfully used for cell labeling.
Acknowledgement The authors thank the ARC, Sirtex Medical Ltd, Syngenta Crop Protection and DuluxGroup (Australia) for financial support. We B. thank A/Prof. Ron Clarke for an access to the Fluorometer; Rene Macquart for magnetic measurements of samples; ACMM unit at the University of Sydney for providing the facility and instruments for particle imaging and cell study.
11
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.polymer.2016.08.064. References [1] A.K. Gupta, M. Gupta, Biomaterials 26 (2005) 3995. [2] L. Li, W. Jiang, K. Luo, H.M. Song, F. Lan, Y. Wu, Z.W. Gu, Theranostics 3 (2013) 595. [3] S. Patra, E. Roy, P. Karfa, S. Kumar, R. Madhuri, P.K. Sharma, Acs Appl. Mater Inter 7 (2015) 9235. [4] C. Huang, K.G. Neoh, L. Wang, E.T. Kang, B. Shuter, Contrast Media Mol. I 6 (2011) 298. [5] L. Zhang, Y. Wang, Y.H. Tang, Z. Jiao, C.Y. Xie, H.J. Zhang, P. Gu, X.B. Wei, G.Y. Yang, H.C. Gu, C.F. Zhang, Nanoscale 5 (2013) 4506. [6] D. Maitland, S.B. Campbell, J. Chen, T. Hoare, Rsc Adv. 6 (2016) 15770. [7] I.T. Lucas, S. Durand-Vidal, E. Dubois, J. Chevalet, P.J. Turq, Phys. Chem. C 111 (2007) 18568. [8] N. Jain, C.K. Liu, B.S. Hawkett, G.G. Warr, W.A. Hamilton, J. Appl. Crystallogr. 47 (2014) 41. [9] N.S. Bryce, B.T.T. Pham, N.W.S. Fong, N. Jain, E.H. Pan, R.M. Whan, T.W. Hambley, B.S. Hawkett, Biomater. Sci-Uk 1 (2013) 1260. [10] N. Jain, X.L. Zhang, B.S. Hawkett, G.G. Warr, Acs Appl. Mater Inter 3 (2011) 662. [11] N. Jain, Y.J. Wang, S.K. Jones, B.S. Hawkett, G.G. Warr, Langmuir 26 (2010) 4465. [12] B.T.T. Pham, N. Jain, P.W. Kuchel, B.E. Chapman, S.A. Bickley, S.K. Jones, B.S. Hawkett, Int. J. Nanomed 10 (2015) 6645. [13] L.B. King, E. Meyer, M.A. Hopkins, B.S. Hawkett, N. Jain, Langmuir 30 (2014) 14143. [14] S.S. Eamegdool, M.W. Weible, B.T.T. Pham, B.S. Hawkett, S.M. Grieve, T. ChanLing, Biomaterials 35 (2014) 5549. [15] D. Nguyen, H.S. Zondanos, J.M. Farrugia, A.K. Serelis, C.H. Such, B.S. Hawkett, Langmuir 24 (2008) 2140. [16] K.M. Ho, P. Li, Langmuir 24 (2008) 1801. [17] L. Chen, D. Niu, C.H. Lee, Y. Yao, K. Lui, K.M. Ho, P. Li, Part. Part. Syst. Charact. (2016), http://dx.doi.org/10.1002/ppsc.201600095. [18] L.P. Ramirez, K. Landfester, Macromol. Chem. Physic 204 (2003) 22. [19] H. Xu, L. Cheng, C. Wang, X.X. Ma, Y.G. Li, Z. Liu, Biomaterials 32 (2011) 9364. [20] S.I. Ali, J.P.A. Heuts, B.S. Hawkett, A.M. van Herk, Langmuir 25 (2009) 10523. [21] V.T. Huynh, D. Nguyen, C.H. Such, B.S. Hawkett, J. Polym. Sci. Pol. Chem. 53 (2015) 1413. [22] D. Nguyen, C.H. Such, B.S. Hawkett, J. Polym. Sci. Pol. Chem. 51 (2013) 250. [23] D. Nguyen, C. Such, B.J. Hawkett, Polym. Sci. Part A Polym. Chem. 50 (2012) 346. [24] N. Zgheib, J.L. Putaux, A. Thill, E. Bourgeat-Lami, F. D'Agosto, M. Lansalot, Polym. Chem-Uk 4 (2013) 607. [25] M.A.M. Mballa, S.I. Ali, J.P.A. Heuts, A.M. van Herk, Polym. Int. 61 (2012) 861. [26] S.I. Ali, J.P.A. Heuts, A.M. van Herk, Soft Matter 7 (2011) 5382. [27] C.J. Ferguson, R.J. Hughes, D. Nguyen, B.T.T. Pham, R.G. Gilbert, A.K. Serelis, C.H. Such, B.S. Hawkett, Macromolecules 38 (2005) 2191. [28] A.P. Philipse, M.P.B. Vanbruggen, C. Pathmamanoharan, Langmuir 10 (1994) 92. [29] M. Baalousha, Sci. Total Environ. 407 (2009) 2093. [30] A. Diac, M. Focsan, C. Socaci, A.M. Gabudean, C. Farcau, D. Maniu, E. Vasile, A. Terec, L.M. Veca, S. Astilean, Rsc Adv. 5 (2015) 77662. [31] P. Reungpatthanaphong, S. Dechsupa, J. Meesungnoen, C. Loetchutinat, S. Mankhetkorn, J. Biochem. Bioph Meth 57 (2003) 1. [32] G. Sahay, D.Y. Alakhova, A.V. Kabanov, J. Control Release 145 (2010) 182.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
Contents lists available at ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling Duc Nguyen a, Binh T.T. Pham a, Vien Huynh a, Byung J. Kim a, Nguyen T.H. Pham a, Stephanie A. Bickley c, Stephen K. Jones c, Algirdas Serelis b, Tim Davey b, Chris Such b, Brian S. Hawkett a, * a b c
Key Centre for Polymers and Colloids, School of Chemistry, University of Sydney, NSW 2006, Australia DuluxGroup (Australia), 1970 Princess Highway, Clayton, Victoria 3168, Australia Sirtex Medical Ltd., 101 Miller Street, North Sydney, NSW 2060, Australia
a r t i c l e i n f o
a b s t r a c t
Article history: Received 31 May 2016 Received in revised form 18 August 2016 Accepted 19 August 2016 Available online xxx
Polymer encapsulation of raw ferrofluids was carried out by free radical emulsion polymerization using macro-RAFT copolymers as stabilizers. The iron oxide cores with positively charged surface at the pH of encapsulation were dispersed in a solution of negatively charged random macro-RAFT copolymer by ultrasonication. Uniform polymer encapsulated super paramagnetic iron oxide nanoparticles (SPIONs) were produced by feeding the mixture of methyl methacrylate (MMA) and n-butyl acrylate (BA) monomers into the dispersion containing water soluble initiator 4,4’-azobis (4-cyanopentanoic acid) (V501). Shell thickness could be controlled by adjusting the amount of coating monomer added. Polymer encapsulation of SPIONs aligned in a magnetic field produced polymer encapsulated SPION rods. Monodispersed sterically stabilized crosslinked polymer encapsulated SPIONs were synthesized using diblock and random macro-RAFT copolymers as stabilizers. Rhodamine labeled SPIONs were found to be noncytotoxic and suitable for cell labeling. © 2016 Elsevier Ltd. All rights reserved.
Keywords: SPIONs Iron oxide Magnetic particles Encapsulation Polymer coating RAFT Control radical Amphiphilic Cells labeling Monodispersed
1. Introduction Superparamagnetic iron oxide nanoparticles (SPIONs) are the most studied nanoparticles due to their relatively low cost and simple synthesis process [1]. They are normally in the form of magnetite [1], maghemite [2] or iron oxides doped with other metal oxides [1e3] which can enhance the particles overall magnetic properties. SPIONs are superparamagnetic nanoparticles which consist of a single magnetic domain and display magnetic properties only in the presence of an external magnetic field [1,2]. These attributes have led to SPIONs finding extensive applications in biomedical science from cell labeling [1,2,4,5], hyperthermia [1e3] to drug delivery [1,2,6]. However, iron oxide nanoparticles in their original form have a weakly charged surface which makes
* Corresponding author. E-mail address: [email protected] (B.S. Hawkett).
them susceptible to aggregation [1,7]. Surface modification involving attachment of stabilizers such as surfactant-like organic substances [1,2] or hydrophilic polymers [1,2] can increase their stability. The provision of steric stabilizers makes the SPIONs more stable, especially in high salt biological media [1,2]. KCPC has extensive experience in modifying SPIONs with macro-RAFT copolymers for cell labeling and drug delivery towards cancer treatment [8e12]. The modification process simply involves the dispersion of SPIONs in a macro-RAFT stabilizing solution. The macro-RAFT stabilizers are diblock in nature and consist of a short block that contains acid groups such as carboxylic [11] or phosphonic [9,13] which anchor the macro-RAFT copolymer to the SPION surface while the second block comprises relatively long blocks of polyethylene oxide or polyacrylamide, which serve as steric stabilizers for the SPIONs [9e11]. The modified SPIONs have been found to be extremely stable in highly concentrated salt solutions and biological media. However, this dispersion process still has an inherent drawback which is the exposure of the SPION
http://dx.doi.org/10.1016/j.polymer.2016.08.064 0032-3861/© 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
2
D. Nguyen et al. / Polymer xxx (2016) 1e11
surface to the dispersed media [9e14]. Furthermore, Jain et al. reported desorption of the polymer stabilizers when macro-RAFT diblocks of poly (acrylic acid) and polyacrylamide at pH greater than 12 and lower than 4 [11]. He attributed this desorption to full protonation of carboxylic groups at pH lower than 4. For pH greater than 12, the desorption was thought to be due to electrostatic repulsion between negatively charged surface and negatively charged diblock copolymers. This lead to reduced particle stability, affecting performance of the modified SPIONs when macro-RAFT diblocks of poly (acrylic acid) and poly (acrylamide) are used as stabilizers at such pH ranges [11]. One possible solution for desorption of this particular type of diblocks is to chain extend the living ends of macro-RAFT stabilizers, making them less labile on the particle surface. The chain extension process can be carried out using polymer encapsulation [15] which may also form a uniform thin polymer shell insulating the iron oxide surface from dispersing media. Polymer encapsulation of iron oxide nanoparticles has been carried out by non-RAFT free radical emulsion [16,17], miniemulsion [18] and microemulsion [19] polymerization but suffered from complexity, low efficiency, uncontrolled polymer shell thickness and only applicable on a small scale. Therefore, it is in our interests to develop a ready to scale-up method which can efficiently encapsulate SPIONs. In this work, we will explore the application of our previously developed polymer encapsulation techniques to encapsulate SPIONs within polymer shells [6,15]. The encapsulation process is based on RAFT mediated emulsion polymerization using macro-RAFT copolymers as stabilizers. In this process, nanoparticles are first dispersed in the macro-RAFT copolymer solution by sonication or milling followed by free radical emulsion polymerization in which controlled chain extension of adsorbed macro-RAFT stabilizers leads to formation of uniform polymer shells encapsulating nanoparticles in the centers of polymer particles. Throughout the process, particle stability is maintained by anchored charges or hydrophilic groups from the macro-RAFT stabilizers. The process is simple, versatile and has been successfully adapted to encapsulate a range of particulate materials [20e26]. In this work, polymer encapsulation of SPIONs will be carried out using macro-RAFT copolymers as stabilizers. The superparamagnetic properties will be taken advantage of to synthesize polymer encapsulated SPION nanorods. Incorporation of steric stabilizers to produce encapsulated SPIONs which are stable in high salt and biological media will be demonstrated. Such particles will be further modified by fluorescent dye attachment for cell labeling. 2. Experimental 2.1. Reagents Milli RO water was used in the synthesis of latexes and acrylic acid-containing RAFT agents. Acrylic acid (AA) (Aldrich) was purified by distillation under reduced pressure. Butyl acrylate (BA) (Aldrich) and methyl methacrylate (MMA) (Aldrich), divinyl benzene (80%, DVB) (Aldrich) and styrene (Sty) (Aldrich) had inhibitor removed by passing them through an inhibitor-removal column (Aldrich). Rhodamine B isothiocyanate (Aldrich), PBS buffer solution (Aldrich), N-(3-dimethylaminopropyl)-N0 -ethylcarbodiimide hydrochloride (EDC) (Aldrich), n-hydroxy succinimide (NHS) (Aldrich), 2,2'-(ethylenedioxy)bis-(ethylamine) (Diamine) (Aldrich), ammonium hydroxide (NH4OH) (Aldrich), carbon disulfide (Aldrich), acetone (Aldrich), tetrabutylammonium bromide (Aldrich), 2-bromopropanoic acid (Aldrich), dimethyl sulfoxide (DMSO) (Aldrich), diethyl ether (Aldrich, Univar), hydrochloric acid (Aldrich, Univar), sodium sulfate (Aldrich), 2,2’-
azobisisobutyronitrile (AIBN) (Wako), acrylamide (AAm) (Aldrich) and 4,4’-azobis (4-cyanopentanoic acid) (V-501) (Wako) were used as received. Dioxane (Aldrich) was distilled under reduced pressure before use. Super paramagnetic iron oxide nanoparticles (SPIONs) dispersion (5.25%) were obtained from by Sirtex Medical Ltd and used as supplied. The RAFT agent, 2-([(butylsulfanyl)carbonothioyl]sulfanyl) propanoic acid (PABTC), was synthesized as previously described [27]. Synthesis of short chain poly (BA-co-AA) macro-RAFT PABTC(BA7-co-AA10) copolymers using PABTC. The amphiphilic macro-RAFT copolymers were prepared as reported in our previous work [15]. PABTC (1.86 g, 7.8 mmol), AIBN (0.064 g, 0.39 mmol), AA (5.63 g, 78.1 mmol), BA (7.51 g, 58.6 mmol) in dioxane (15.00 g) was prepared in a 50 mL round-bottomed flask. 1,3,5-trioxane (0.09 g, 0.9 mmol) was added as an internal standard for proton NMR. This was stirred magnetically and sparged with nitrogen for 10 min. The flask was then heated at 70 C for 7 h under constant stirring. The final copolymer solution was 50.7% solids. The copolymers were characterized by 1H NMR (300 MHz, Acetoned6) and size exclusion chromatography (SEC). Based on 1H NMR, the conversion of BA and AA was around 97% and therefore the theoretical molecular weight is 1869.8 g mol1. The copolymers were characterized by MALDI (Bruker Daltonics Apex Ultra 7 T Fourier Transform Ion Cyclotron Resonance (FTICR) Mass Spectrometer and found to be narrowly distributed in a range from 1000 to 3000 (g/mole), peaked at 2000 (g/mole). 2.2. Synthesis of poly((acrylic acid)-block-poly(acrylamide)) macro-RAFT PABTC-(AA5-block-AAm60) using PABTC A solution of PABTC (0.74 g, 3.1 mmol), V501 (0.05 g, 0.2 mmol), acrylamide (13.40 g, 188.6 mmol) in dioxane (17.2 g), 1,3,5-trioxane (0.01 g) and water (20.5 g) was prepared in a 100 mL round bottom flask. This was stirred magnetically and sparged with nitrogen for 15 min. The flask was then heated at 70 C for 4 h. By 1H NMR, conversion was determined to be at 99.5% after 4 h. At the end of this period, AA (1.12 g, 15.6 mmol) and V501 (0.05 g, 0.2 mmol) was added to the flask. The mixture was deoxygenated and heating was continued at 70 C for a further 3 h s. The copolymer solution had 28.8% solids and conversion was found to be 96% by 1H NMR. 2.3. Encapsulation of SPIONs with poly(MMA-co-BA) using PABTC(BA7-co-AA10) copolymers as stabilizers Poly (MMA-co-BA) encapsulated SPIONs were synthesized using the following procedure. In a 100 ml beaker, macro-RAFT PABTC(BA7-co-AA10) solution (1.20 g, 0.3 mmol) was dispersed in water (60.1 g) to yield a cloudy yellow solution of pH 4. Ammonium hydroxide (2.8% solution in water) was added to the macro-RAFT solution to raise the pH to 9.0. This was followed by 1 min ultrasonication using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% to obtain a clear yellow solution with pH 7.0. To this solution, SPIONs (20.11 g) was added drop wise and thoroughly dispersed by ultrasonication using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% for 30 min. After sonication, the dispersion pH was measured at 6.5 and it was then transferred to a 100 mL round bottom flask containing V501 (0.04 g, 0.15 mmol) which was subsequently sealed and purged with nitrogen for 10 min. The whole flask was then immersed in an oil bath with a temperature setting of 70 C and was magnetically stirred. A deoxygenated 10:1 (weight ratio) solution (1 mL, 0.94 g) of methyl methacrylate (MMA), butyl acrylate (BA) was injected into the flask, while in the 70 C oil bath, at a rate of 2.0 mL/hour. This feed rate
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11
was chosen based on our previous works [15,23] on pigment encapsulation in consideration of the amount of initiator used. After monomer addition, the heating continued for another five and half hours to produce stable brown latex of the encapsulated SPIONs with 2.1% solids. 2.4. Further shell growth of the polymer encapsulated SPIONs with polystyrene The poly (MMA-co-BA) shells encapsulating SPIONs were further grown with polystyrene using the following procedure. A solution containing styrene (1.00 g) and AIBN (0.04 g) was prepared in a 100 mL round bottom flask. To this solution, the polymer encapsulated SPIONs latex (10.10 g) and water (40.11 g) was added and thoroughly mixed for 10 min by magnetic stirring. The flask was subsequently sealed and purged with nitrogen for another 10 min before being heated in an oil bath at 80 C for 3 h under constant stirring. The final brown latex was found to be stable and contained 2.2% solids. 2.5. Synthesis of poly(MMA-co-BA) encapsulated SPION nanorods using PABTC-(BA7-co-AA10) copolymers as stabilizers Poly (MMA-co-BA) encapsulated SPION nanorods were synthesized using the following procedure. In a 100 ml beaker, PABTC(BA7-co-AA10) macro-RAFT solution (0.6 g, 0.15 mmol) was dispersed in water (31.2 g) to yield a cloudy yellow solution of pH 4. Ammonium hydroxide (2.8% solution in water) was added to the macro-RAFT solution to raise the pH to 9.0. This was followed by 1 min ultrasonication using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% to obtain a yellow clear solution with pH 7.5. To this solution, SPIONs (10.10 g) were added drop wise and thoroughly dispersed by ultrasonication using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% for 30 min. A magnetic stirrer was used to mix the dispersion throughout the ultrasonication. After sonication, the dispersion pH was measured at 7.6 and it was then transferred to a 100 mL round bottom flask containing V501 (0.03 g, 0.15 mmol) which was subsequently sealed and purged with nitrogen for 10 min. The whole flask was then immersed in an oil bath with a temperature setting of 70 C and was magnetically stirred. A deoxygenated 10:1 (weight ratio) solution (5 mL, 4.69 g) of methyl methacrylate (MMA), butyl acrylate (BA) was injected into the flask, while in the 70 C oil bath, at a rate of 1.0 mL/h. Since SPION nanorods were larger, this slower monomer feed rate (compared to the previous SPION encapsulation experiment) was chosen to minimize the amount of unreacted monomer present in the system at any given time. This helped to avoid aggregation of the nanorod dispersion by the presence of excessive numbers of monomer droplets. After monomer addition, the heating continued for another five and half hours to produce stable brown latex of the encapsulated SPIONs. Encapsulation and modification of SPIONs with polymers using PABTC-(BA7-co-AA10) and PABTC-(AA5-block-AAm60) macro-RAFT copolymers as electrostatic and steric stabilizers, respectively. 2.6. Poly(MMA-co-BA) encapsulation of SPIONs Poly (MMA-co-BA) encapsulated SPIONs were synthesized using the following procedure. In a 100 ml beaker, PABTC-(BA7-co-AA10) solution (0.32 g, 0.06 mmol) and PABTC-(AA5-block-AAm60) macroRAFT solution (0.40 g, 0.02 mmol) was dispersed in water (50.6 g) to yield a cloudy yellow solution of pH 3.5. Ammonium hydroxide (2.8% solution in water) was added to the macro-RAFT solution to raise the pH to 8.5. This was followed by 1 min of ultrasonication
3
using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% to obtain a yellow clear solution with pH 7.2. To this solution, SPIONs (10.0 g) was added drop wise and thoroughly dispersed by ultrasonication using a Vibra-Cell Ultrasonic Processor (Sonics and Materials, Inc.) standard probe at amplitude of 30% for 15 min. After sonication, the dispersion, which had a pH of 7.5, was transferred to a 100 mL round bottom flask containing V501 (0.03 g, 0.11 mmol) which was subsequently sealed and purged with nitrogen for 10 min. The whole flask was then immersed in an oil bath with a temperature setting of 70 C and was magnetically stirred. A deoxygenated 10:1 (weight ratio) solution (5 mL, 4.69 g) of methyl methacrylate (MMA), butyl acrylate (BA) was injected into the flask, while in the 70 C oil bath, at a rate of 2.0 mL/h. After monomer addition, the heating continued for another five and half hours to produce stable brown latex of the encapsulated SPIONs. The latex was characterized by DLS and TEM. 2.7. Crosslinking sterically stabilized polymer encapsulated SPIONs with DVB The Poly (MMA-co-BA) shells of the encapsulated SPIONs were crosslinked with DVB (5.6% by weight on polymer) using the procedure described in Reference 23 [23]. A solution containing DVB (0.26 g) and V501 (0.04 g) was prepared in a 100 mL round bottom flask. To this solution, the polymer encapsulated SPIONs latex (48.3 g) was added and thoroughly mixed for 10 min by magnetic stirring. The flask was subsequently sealed and purged with nitrogen for another 10 min before heated in an oil bath at 80 C for 2 h under constant stirring. The final brown latex was found to be stable and was characterized by DLS and TEM. 2.8. Conjugation of sterically stabilized crosslinked polymer encapsulated SPIONs with Rhodamine B isothiocyanate The sterically stabilized crosslinked polymer encapsulated SPIONs were conjugated with Rhodamine B isothiocyanate using the following procedure. A latex dispersion containing the above crosslinked polymer encapsulated SPIONs (10.1 g) was prepared in a 25 mL beaker. To this dispersion, 0.03 g of 1% NHS solution was added, followed by 0.03 g of 1% EDC solution. The dispersion was thoroughly mixed by magnetic stirring. After 60 min, 3 g of 0.25% diamine was added and the mixing was continued for another 60 min. To this dispersion, a Rhodamine B isothiocyanate dye (0.0017 g) in 1 g of DMSO was then added and the mixture was stirred for 2 h. After reaction, free dye was removed by repeated centrifugations (3 15 min at 14500 rpm) and washing (3 times) with mili-Q water. The washed SPION deposit was re-dispersed in 25 g of water by ultrasonication for 10 min at 30% amplitude followed by purification by dialysis against water to obtain a stable Rhodamine labeled SPIONs dispersion with 1.2% solids. PBS was used to prepare a stock dispersion for cell imaging. 2.9. In vitro evaluation for the bioactivity of polymer encapsulated iron oxide nanoparticles 2.9.1. Cell culture DLD-1 (ATCC, CCL-221) cells were maintained as confluent monolayers in Advanced DMEM (Gibco, 12491e015), supplemented with 2% (v/v) FCS. The cells were incubated under standard culturing conditions (at 37 C with 5% (v/v) CO2 under humidified conditions). Cells were routinely sub-cultured by removal of the medium, washing with PBS (In vitro technologies, IVT3001302) and detached from the culture-ware by 0.25% (w/v) trypsin (SAFC Biosciences, 59430C). The cell suspension was collected in a centrifuge
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11
tube with the addition of complete growth medium and spun down at approximately 1000 g for 3 min. The supernatant was discarded and the cells re-suspended in fresh growth medium. Cells were then counted on a haemocytometer and an appropriate number of cells were seeded for further experiments and subculturing. 2.9.2. Cytotoxicity screening in monolayer cells A 1 105 cells/well suspension was seeded into each well of a 96-well plate (Corning, 3599) and incubated for 24 h under standard culturing conditions. Prior to the assay, polymer encapsulated iron oxide nanoparticles were sterilized by filtration through a 0.45 mm filter. The dose volume of the nanoparticles was kept below 24% of the total volume and diluted in complete culture medium to be added to quintuplet wells spanning a 2-log range of final concentrations (500, 250, 125, 62.5, 31.25, 15.625 and 7.8125 ppm). The cells were then incubated for 72 h before 1.0 mM MTT (thiazolyl blue tetrazolium bromide, Alfa Aesar, L11939) was added and further incubated for 4 h. The media was removed and 150 mL of DMSO was added to each well. The absorbance at 600 nm was then measured for the plate in a Victor3V microplate reader (Perkin Elmer, 1420). IC50 values were determined as the drug concentration that reduced the absorbance to 50% of that in untreated control wells. 2.9.3. Confocal imaging of cells treated with the Rhodamine labeled SPIONs 1 105 cells/well suspension was seeded into each well of a 12well glass bottom plate (Mattek, P12G-1.5e10-F.s) and incubated for 24 h under standard culturing conditions. 20 ppm filter sterilized nanoparticles were dosed onto the DLD-1 cells for 24 h 20 min prior to imaging, cells were treated with 1 mg/mL Hoechst 33258 (Sigma, 94403-1 ML). Excess nanoparticles and Hoechst were washed away with PBS and fresh culturing medium was added. Confocal images were generated on the Olympus FV-1000 microscope using UPLAPO 40 oil objective (N.A.: 1.00) with the sampling speed of 2 ms/pixel. Rhodamine B signals were acquired with 559 nm laser excitation and 570e670 nm emission while Hoechst was excited with the 405 nm laser and emissions collected from 425 to 475 nm. 2.9.4. Particle characterizations Particles were characterized by TEM (Philips CM120 Biofilter), dynamic light scattering (DLS) (Zetasizer Nano ZS, Malvern) or by TGA (TA Instruments TGA Discovery). Zeta potential measurements were also performed on the Malvern Zetasizer Nano ZS using the Smoluchowski equation as a basis for calculation. For TGA dry sample preparation, the polymer encapsulated SPION particles were separated from the homo-polymer particles by centrifugation, redispersed and washed three times with water prior to drying in an oven at 100 C for 12 h. Fluorescence of the Rhodamine labeled SPIONs was characterized using a Shimadzu (Kyoto, Japan) RF5301 PC spectrofluorometer at 550 and 559 nm excitation wavelengths. Magnetic measurements were conducted using a Quantum Design 9 T PPMS (Physical Property Measurement System) equipped with a VSM (Vibrating Sample Magnetometer). Powdered samples were housed in a polypropylene sample holder held in place with a brass half tube holder. All measurements were carried out at 300 K. The magnetic moment was measured as a function of magnetic field strength and then converted to mass magnetisation (emu/g) by dividing the moment (emu) by the mass of SPIONs in each sample (g). Raw SPIONs (12.36 mg) and polymer encapsulated SPIONs (10.51 mg, 73.1% iron oxide, analyzed by TGA) were dried in an oven overnight at 100 C prior to magnetic measurements using
a 0.4 T electromagnet. 3. Results and discussions 3.1. SPIONs The aqueous SPION dispersion was supplied as a dark brown dispersion and observed to be stable in their supplied form at pH 4. The SPIONs were found to have strong surface positive charge of 44 mV at pH 4.2 (Fig. 1). TGA shows that the SPIONs contained no organic stabilizers, with negligible weight loss at 700 C (Fig. 2, black curve). Therefore, the positive surface charge was attributed to the presence of Metal-OHþ 2 groups on the surface of the nanoparticles as discussed by Lucas et al. [7] The positive surface charge together with negative counter ions helped to maintain SPION stability at low pH. However, as the pH increased, the positive charge decreased due to de-protonation and formation of more OH groups [7]. The reduction in surface charge destabilized the SPIONs, leading to visual aggregate formation. As the pH was raised to 7.2, the surface approached the point of zero charge at which 1.4 mm average aggregates were measured by DLS. By TEM (Fig. 3), it was found that SPIONs are not spherical in shape and quite polydispersed, with particle diameters ranging from 10 to 50 nm. Broad size distribution was also consistent with DLS results in Table 1 in which average particle diameters were measured to be 89 nm with a high PDI of 0.410. Since the DLS measurement gave a much larger size than what observed by TEM, it was likely that a large number of SPIONs exist as aggregates despite the high positive surface charge. This observation is common with large SPIONs due to the magnetic attraction between iron oxide particles even in zero magnetic field which can be overcome by surface modifications [28]. 3.2. Dispersion of SPIONs in macro-RAFT solution Dispersion of SPIONs in macro-RAFT solution was carried out in the same manner as previously described [15] and is shown in Fig. 4. Short chain macro-RAFT copolymers were solubilized at pH 6.5 prior to the dispersion. Once mixed with the iron oxide nanoparticles, negatively charged carboxylic functional groups facilitate the copolymer adsorption onto the positively charged particle surface via opposite charge interactions [29]. In the case of iron oxide, polymer adsorption was also possible via complex formation between carboxyl groups and iron atoms on particle surface [11]. After sonication, the SPION surface was covered with the macroRAFT copolymers, imparting strong negative charge as measured by Zeta potential of 54 mV at pH 6.5 (Table 1). This adsorption was further confirmed by TGA (Fig. 2, blue curve) where weight loss due to the copolymer was found to be about 10%. As a result, the dispersed SPIONs were more stable in the aqueous phase and less
50 40 30 20 10 0
8
7
6
5
4
Zeta PotenƟal (mV)
4
3
pH Fig. 1. Zeta potentials of SPIONs versus pH of dispersions.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11
5
Fig. 2. TGA curves of raw (black), dispersed (blue), poly (MMA-co-BA) encapsulated (green) and poly (Sty/MMA/BA) encapsulated (red) SPIONs. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
RAFT solution. This was probably triggered by the reduction in surface charge when the pH was suddenly increased from 4 to 6.5, making SPIONs less stable before they had a chance to interact with macro-RAFT stabilizers. The subsequent ultrasonication helped to break up some aggregates but some remained, resulting in an overall increase in particle size as reported in Table 1. 3.3. Polymer encapsulation of SPIONs
Fig. 3. TEM image of raw, not encapsulated SPIONs.
susceptible to pH change. As reported in Table 1, there was a slight increase in average particle size to 104 nm. However, TEM evidence in Fig. 5 showed that the adsorption of short chain macro-RAFT copolymers did not alter the size of individual iron oxide nanoparticles since they were similar to ones in Fig. 3. Therefore, the average size increase was probably due to aggregate formation during the mixing of the iron oxide dispersion with the macro-
To encapsulate the dispersed SPIONs, BA/MMA monomer mixture was slowly fed into the dispersion in the presence of the water soluble initiator at 70 C. The polymerization mechanism was thoroughly discussed in our previous work [15] and depicted here in Fig. 4. The process followed RAFT controlled free radical emulsion polymerization, involving a continuous and even growth of the macro-RAFT copolymers which was adsorbed onto the nanoparticle surface. Such controlled polymer growth produced uniform polymer layers encapsulating the SPIONs. As seen in Fig. 6A, thin and even polymer shells of 5e6 nm were observed on all iron oxide particles and aggregates. This was further confirmed by TGA evidence presented in Fig. 2 (green curve) in which weight loss due to the polymer on the SPION surface was found to be about 26.9%. Throughout the encapsulation process, SPIONs were stabilized by the negatively charged carboxyl groups from the adsorbed macroRAFT copolymers. As polymer shell thickness increased, the negative charge was maintained on the outer surface of the encapsulated iron oxide nanoparticles. As shown in Table 1, the zeta potential measured for the encapsulated particles was 55 mV, the same as for the dispersed SPIONs. Particle stability provided by anchored carboxyl groups allowed further shell growth without the
Table 1 Evolution of Zeta potentials and sizes of SPIONs particles throughout the encapsulation process.
Sirtex iron oxide nanoparticle Dispersed iron oxide nanoparticle in macro-RAFT Poly(MMA-co-BA) encapsulated iron oxide nanoparticle Poly(Styrene/MMA/BA) encapsulated iron oxide nanoparticle
Z potential (mV)
Electrophoretic mobility (mmcm/Vs)
pH of dispersion
Size (nm)
PDI
44 ¡54 ¡55 NA
3.489 ¡4.14 ¡4.303 NA
4.2 7.3 6.7 NA
89 104 94 90
0.41 0.26 0.25 0.12
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
6
D. Nguyen et al. / Polymer xxx (2016) 1e11
Fig. 4. Dispersion and encapsulation of SPIONs using PABTC-(BA7-co-AA10) macro-RAFT copolymer.
Fig. 5. SPIONs after dispersion in macro-RAFT solution.
increased to more than 60%. The uniform polymer layers in Fig. 6B indicated an even distribution of the monomer and initiator across particles. From 6A to 6B, it was observed that encapsulated SPIONs became more spherical as more monomer was added. This was similar to the case of Titanium dioxide pigment encapsulation where more spherical shapes were obtained with increasing shell thickness [15]. This was attributed to the minimization of particle surface area as particle the size grew. It was also found that particle sizes became more monodispersed as shown in Fig. 6B. This was in agreement with DLS measurements (Table 1) in here particle size was shown to be largely similar while size distribution consistently decreased as monomer was added. This was probably due to removal of unstable large particles as aggregates during the emulsion polymerization. Particle transformation from irregular to spherical shapes reduced the discrepancy between measured dimensions per particles and therefore was important in reducing polydispersity. In Fig. 6B, homogenous polymer particles containing no SPIONs with sizes from 28 to 33 nm were clearly present. They were formed during the emulsion polymerization process probably due to the presence of labile macro-RAFT copolymers and other water soluble species from decomposition of V501 initiator.
Fig. 6. Polymer coated SPIONs: A) with MMA/BA (10/1); B) further growth with styrene.
need of additional surfactants. Using poly (MMA-co-BA) coated SPIONs as seeds, free radical emulsion polymerization of styrene was carried out. Relatively hydrophobic AIBN initiator was chosen due to its high solubility in styrene. The reaction was completed in a very simple manner in which the monomer swollen polymer shells acted as the locus of polymerization. As shown in Fig. 6B, polymer shells grew to 16e17 nm in thickness during this process. This was reconfirmed by TGA (Fig. 2, red curve) in which the polymer content of poly (Sty/MMA/BA) encapsulated SPIONs significantly
Growth of the polymer particles competed with the encapsulation process for monomer and free radicals, reducing the efficiency of the later. Therefore, it is desirable in the future to minimize polymer particle formation by optimizing the amount of macro-RAFT and V501 required for the encapsulation process. 3.4. Synthesis of polymer encapsulated SPION nanorods One key feature of SPIONs particles is their super paramagnetic
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11
Fig. 7. Spikes formation from poly (MMA-co-BA) encapsulated SPIONs under influence of a magnet.
7
the ultrasonic dispersion of SPIONs in the macro-RAFT copolymer solution. Due to the close proximity between aligned individual iron oxide nanoparticles, inter-particle adsorption of the macroRAFT copolymers probably occurred. They acted as glue holding individual nanoparticles together in alignment as SPION rods even when the magnetic field was removed. Polymer encapsulation of such aggregates encased SPION rods in centers of polymer shells as shown by TEMs in Fig. 10A and B. In these images, polymer encapsulated SPION rods with lengths from 100 to 500 nm were observed. However, there were a large number of spherical polymer encapsulated SPIONs. This indicated a non-uniform magnetic field during dispersion especially for regions distanced from the spin bar where the field was too weak to align SPIONs. Possible solutions to exert a uniform magnetic field during dispersion to raise the yield of SPION nanorods will be investigated in future work. 3.5. Synthesis of sterically stabilized polymer encapsulated SPIONs
properties in which they interact with the magnetic field once applied [8,11]. It is therefore critical that encapsulated SPIONs continue to behave in this manner after encapsulation. Fig. 7 clearly demonstrated that poly (MMA-co-BA) encapsulated SPIONs still maintained their magnetic properties by forming sharp spikes once the concentrated dispersion was placed on top of a permanent magnet. Spike formation is typical ferrofluid behavior and was due to alignment of encapsulated SPIONs under the influence of the magnetic field generated by the magnet [8,10,11,13]. Magnetic measurements (Fig. 8) found that mass magnetization of polymer encapsulated SPIONs saturated at a lower value than that of the raw SPIONs. This reduction in magnetic property was attributed to the presence of non-magnetic polymer layers which made the magnetic diameter smaller than physical diameter of particles [11]. SPION alignments under the magnetic field were further taken advantage of to synthesize polymer encapsulated SPION nanorods in the process shown in Fig. 9 [8]. In this process, a magnetic field was generated by a magnetic spin bar and magnetic stirrer during
For polymer encapsulated SPIONs to be used for biological applications, it is essential for the final particles to be stable in high salt concentrations [9,12]. Unfortunately, electrostatically stabilized polymer encapsulated iron oxide nanoparticles are not suitable for this purpose. They become unstable once dispersed in high salt environment due to the suppression of the electrical double layer which in turn reduces the repulsive force between particles. In previous work [8,11], macro-RAFT diblock copolymer RAFT PABTC(AA10-block-AAm14) was successfully used to stabilize SPIONs in 2 M NaCl solution. In our case, a similar macro-RAFT diblock copolymer RAFT PABTC-(AA5-block-AAm60) was utilized in conjunction with macro-RAFT copolymer PABTC-(BA7-co-AA10) for the dispersion and encapsulation of SPIONs as depicted in Fig. 11. As in the case of the macro-RAFT random copolymers, adsorption of the diblock copolymers onto the SPION surface was likely via opposite charge interaction or complex formation. On the other hand, the long poly (acrylamide) blocks of the adsorbed macroRAFT copolymers would provide steric stability by forming
Fig. 8. Mass magnetization of raw and poly (MMA-co-BA) encapsulated SPIONs as a function of magnetic field using a 0.4 T electromagnet.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
8
D. Nguyen et al. / Polymer xxx (2016) 1e11
Fig. 9. Polymer encapsulation of SPION rods using macro-RAFT copolymers as the stabilizers.
Fig. 10. Poly (MMA-co-BA) encapsulated SPION nanorods.
hydrophilic hairy layers which prevented surface contact between particles. Polymer encapsulation was carried out in the same manner as previously discussed. A monomer mixture of MMA/BA (10/1 by weight) was slowly fed into the dispersion at 70 C in the presence of V501 initiator. Chain extension of macro-RAFT copolymers by free radical emulsion polymerization produced uniform polymer shell encapsulated SPIONs as shown in Fig. 12A. Particles were observed to be similar in size with diameters varying from 70 to 78 nm. By DLS, it is reported in Table 2 that final latex particles were monodispersed (0.07 PDI), having an average hydrodynamic diameter of 125 nm. The larger diameter measured by DLS was expected due to the extension of the poly (AAm) blocks in the aqueous medium. In the dry state, they collapsed in volume and therefore were not observable by TEM. The polymer encapsulated SPIONs were further strengthened by crosslinking with DVB. As shown in Fig. 12B, the crosslinked polymer encapsulated SPIONs slightly increased in size to a range from 78 to 94 nm. DLS measurement in Table 2 reported monodispersed particles (0.04 PDI) with diameters almost unchanged from before the crosslinking reaction. Compared to poly (MMA-co-BA), crosslinked polymer networks should make encapsulating shells harder and less
susceptible to solvents. 3.6. Dye conjugation with sterically stabilized polymer encapsulated SPIONs and cell labeling As previously discussed, the key target for using steric stabilizer macro-RAFT PABTC-(AA5-block-AAm60) was to have encapsulated SPIONs stable in high salt environments. As reported in Table 2, both uncrosslinked and crosslinked SPIONs were found to be relatively stable in PBS solution with measured average diameters slightly lower than those in water. The small reductions in size were probably due to suppression of electrical double layer by the high ionic strength. This may lead to some aggregation between particles as shown by increases in PDI. However, particle stability was markedly improved compared to the charge stabilized encapsulated SPIONs. This was readily observed (Fig. 13B) a sediment of large aggregates at the bottom of vial B once charge stabilized encapsulated SPIONs were mixed with PBS. The sterically stabilized polymer encapsulated SPIONs contain an abundance of carboxylic groups on the surface which can be readily conjugated with a fluorescent dye to produce a
Fig. 11. Poly (MMA-co-BA) encapsulated SPIONs using PABTC-(BA7-co-AA10) and PABTC-(AA5-block-AAm60) macro-RAFT copolymers as stabilizers.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11
9
Fig. 12. Polymer encapsulated SPIONs using macro-RAFT PABTC-(AA5-block-AAm60) and PABTC-(BA7-co-AA10) as stabilizers: A) poly (MMA-co-BA) encapsulated SPIONs; B) crosslinked poly (MMA-co-BA) encapsulated SPIONs.
Table 2 Sizes of steric stabilized polymer encapsulated SPIONs before and after conjugation with Rhodamine B isothiocyanate. Size (nm) PDI Poly(MMA-co-BA) encapsulated SPIONs in water Poly(MMA-co-BA) encapsulated SPIONs in PBS Crosslinked poly(MMA-co-BA) encapsulated SPIONs in water Crosslinked poly(MMA-co-BA) encapsulated SPIONs in PBS Dye conjugated polymer encapsulated SPIONs in water Dye conjugated polymer encapsulated SPIONs in PBS
125 123 124 121 141 145
0.07 0.11 0.04 0.09 0.03 0.08
monodispersed superparamagnetic imaging agent. As depicted in Fig. 14, the conjugation process followed a well-used method which
involved reacting carboxylic functional groups with EDC/NHS/ Diamine to produce active primary amines on the particle surface [9]. Once formed, such amines groups were reacted with isothiocyanate groups of Rhodamine B isothiocyanate [30], producing dye conjugated polymer encapsulated SPIONs. As shown in Table 2, dye conjugation slightly increased the hydrodynamic diameter to 141 nm while particles maintained their monodispersity with a PDI of 0.03. More importantly, the particles were still stable in PBS solution with particle size remaining unchanged at 145 nm and almost no aggregates were observed (Fig. 13A). The size increase was probably due to attachment of bulky dye molecules especially to the carboxyl group at poly (AAm) polymer chain ends. The modified particles still maintained their magnetic properties. As shown in Fig. 13C, the dye conjugated particles were readily
Fig. 13. Polymer encapsulated SPIONs in PBS: A) with macro-RAFT PABTC-(AA5-block-AAm60) as the steric stabilizer; B) without steric stabilizer; C) sterically stabilized dye conjugated polymer encapsulated SPIONs were extracted by a magnet.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
10
D. Nguyen et al. / Polymer xxx (2016) 1e11
Fig. 14. Dye conjugation of steric stabilized polymer encapsulated SPIONs.
Fig. 15. Fluorescent emission spectrum of Rhodamine B labeled SPIONs at 100 mg/mL.
extracted by a permanent magnet, suggesting their suitability for separation applications. Dye conjugated polymer encapsulated SPIONs were found to be fluorescent under excitation with UV light (Fig. 15). Similar emission spectra at different excitation wavelength showed that the fluorescent dye (fluorophore) was chemically attached to the surface of particles and not due to adsorbed dye molecules. Incorporation of a fluorophore onto the encapsulated nanoparticles allows for visualization and tracking in biological systems. Unconjugated free Rhodamine B isothiocyanate tends to localize primarily in the mitochondria and partly in the cytoplasm due to its positive charge and lipophilicity [31]. When the fluorophore is part of the encapsulated SPIONs, it showed a vesicular localization (Fig. 16), which could be due to its mode of entry into the cells. The encapsulated SPIONs would enter the cells through macropinocytosis more so than the receptor-mediated endocytosis due to a size restriction of approximately 120 nm via the latter route of entry [32]. As the local pH becomes acidic within these vesicles, many pH sensitive nanoparticles degrade and cause cytotoxic effects within the cells. Since the encapsulating shells composed of crosslinked polymer
Fig. 16. Confocal images of DLD-1 cells treated with 20 mg/mL Rhodamine B labeled SPIONs. Scale bar represents 20 mm.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064
D. Nguyen et al. / Polymer xxx (2016) 1e11 Table 3 IC50 values (mg/mL) of the Rhodamine B labeled SPIONs in DLD-1 cells at 72 h. Cell line
Cell type
IC50 (mg/mL)
Highest concentration tested (mg/mL)
DLD-1
Human colorectal adenocarcinoma
>1000
1000
networks should help protect the SPIONs from the external environment and, the stability of the SPIONs should be further enhanced. As such, when DLD-1 cells were treated with the encapsulated SPIONs, they were non-cytotoxic up to a concentration of 1000 mg/mL (Table 3), which is about hundred fold higher than the concentration relevant for in vitro biological studies. Increasing the stability of nanomaterials by encapsulation is an attractive strategy that could strengthen the transition of the nanomaterials onto an in vivo testing and beyond.
4. Conclusions In this work, we successfully dispersed and encapsulated SPIONs particles with uniform polymer shells using macro-RAFT copolymers as stabilizers. The encapsulation process was carried out by free radical emulsion polymerization by chain extensions of macro-RAFT stabilizers. Polymer encapsulated SPIONs were found to retain their magnetic properties, displaying ferrofluid spikes under the influence of a magnet. Polymer encapsulated SPION nanorods were also successfully synthesized based on alignment of SPIONs in a magnetic field. For the cell labeling application, monodispersed sterically stabilized polymer encapsulated SPIONs were produced by incorporating diblock steric stabilizers during the dispersion and encapsulation stages. Such particles were found to be stable in PBS solution and were further modified by conjugation with Rhodamine B isothiocyanate. The dye conjugated polymer encapsulated SPIONs displayed no cytotoxicity and successfully used for cell labeling.
Acknowledgement The authors thank the ARC, Sirtex Medical Ltd, Syngenta Crop Protection and DuluxGroup (Australia) for financial support. We B. thank A/Prof. Ron Clarke for an access to the Fluorometer; Rene Macquart for magnetic measurements of samples; ACMM unit at the University of Sydney for providing the facility and instruments for particle imaging and cell study.
11
Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.polymer.2016.08.064. References [1] A.K. Gupta, M. Gupta, Biomaterials 26 (2005) 3995. [2] L. Li, W. Jiang, K. Luo, H.M. Song, F. Lan, Y. Wu, Z.W. Gu, Theranostics 3 (2013) 595. [3] S. Patra, E. Roy, P. Karfa, S. Kumar, R. Madhuri, P.K. Sharma, Acs Appl. Mater Inter 7 (2015) 9235. [4] C. Huang, K.G. Neoh, L. Wang, E.T. Kang, B. Shuter, Contrast Media Mol. I 6 (2011) 298. [5] L. Zhang, Y. Wang, Y.H. Tang, Z. Jiao, C.Y. Xie, H.J. Zhang, P. Gu, X.B. Wei, G.Y. Yang, H.C. Gu, C.F. Zhang, Nanoscale 5 (2013) 4506. [6] D. Maitland, S.B. Campbell, J. Chen, T. Hoare, Rsc Adv. 6 (2016) 15770. [7] I.T. Lucas, S. Durand-Vidal, E. Dubois, J. Chevalet, P.J. Turq, Phys. Chem. C 111 (2007) 18568. [8] N. Jain, C.K. Liu, B.S. Hawkett, G.G. Warr, W.A. Hamilton, J. Appl. Crystallogr. 47 (2014) 41. [9] N.S. Bryce, B.T.T. Pham, N.W.S. Fong, N. Jain, E.H. Pan, R.M. Whan, T.W. Hambley, B.S. Hawkett, Biomater. Sci-Uk 1 (2013) 1260. [10] N. Jain, X.L. Zhang, B.S. Hawkett, G.G. Warr, Acs Appl. Mater Inter 3 (2011) 662. [11] N. Jain, Y.J. Wang, S.K. Jones, B.S. Hawkett, G.G. Warr, Langmuir 26 (2010) 4465. [12] B.T.T. Pham, N. Jain, P.W. Kuchel, B.E. Chapman, S.A. Bickley, S.K. Jones, B.S. Hawkett, Int. J. Nanomed 10 (2015) 6645. [13] L.B. King, E. Meyer, M.A. Hopkins, B.S. Hawkett, N. Jain, Langmuir 30 (2014) 14143. [14] S.S. Eamegdool, M.W. Weible, B.T.T. Pham, B.S. Hawkett, S.M. Grieve, T. ChanLing, Biomaterials 35 (2014) 5549. [15] D. Nguyen, H.S. Zondanos, J.M. Farrugia, A.K. Serelis, C.H. Such, B.S. Hawkett, Langmuir 24 (2008) 2140. [16] K.M. Ho, P. Li, Langmuir 24 (2008) 1801. [17] L. Chen, D. Niu, C.H. Lee, Y. Yao, K. Lui, K.M. Ho, P. Li, Part. Part. Syst. Charact. (2016), http://dx.doi.org/10.1002/ppsc.201600095. [18] L.P. Ramirez, K. Landfester, Macromol. Chem. Physic 204 (2003) 22. [19] H. Xu, L. Cheng, C. Wang, X.X. Ma, Y.G. Li, Z. Liu, Biomaterials 32 (2011) 9364. [20] S.I. Ali, J.P.A. Heuts, B.S. Hawkett, A.M. van Herk, Langmuir 25 (2009) 10523. [21] V.T. Huynh, D. Nguyen, C.H. Such, B.S. Hawkett, J. Polym. Sci. Pol. Chem. 53 (2015) 1413. [22] D. Nguyen, C.H. Such, B.S. Hawkett, J. Polym. Sci. Pol. Chem. 51 (2013) 250. [23] D. Nguyen, C. Such, B.J. Hawkett, Polym. Sci. Part A Polym. Chem. 50 (2012) 346. [24] N. Zgheib, J.L. Putaux, A. Thill, E. Bourgeat-Lami, F. D'Agosto, M. Lansalot, Polym. Chem-Uk 4 (2013) 607. [25] M.A.M. Mballa, S.I. Ali, J.P.A. Heuts, A.M. van Herk, Polym. Int. 61 (2012) 861. [26] S.I. Ali, J.P.A. Heuts, A.M. van Herk, Soft Matter 7 (2011) 5382. [27] C.J. Ferguson, R.J. Hughes, D. Nguyen, B.T.T. Pham, R.G. Gilbert, A.K. Serelis, C.H. Such, B.S. Hawkett, Macromolecules 38 (2005) 2191. [28] A.P. Philipse, M.P.B. Vanbruggen, C. Pathmamanoharan, Langmuir 10 (1994) 92. [29] M. Baalousha, Sci. Total Environ. 407 (2009) 2093. [30] A. Diac, M. Focsan, C. Socaci, A.M. Gabudean, C. Farcau, D. Maniu, E. Vasile, A. Terec, L.M. Veca, S. Astilean, Rsc Adv. 5 (2015) 77662. [31] P. Reungpatthanaphong, S. Dechsupa, J. Meesungnoen, C. Loetchutinat, S. Mankhetkorn, J. Biochem. Bioph Meth 57 (2003) 1. [32] G. Sahay, D.Y. Alakhova, A.V. Kabanov, J. Control Release 145 (2010) 182.
Please cite this article in press as: D. Nguyen, et al., Monodispersed polymer encapsulated superparamagnetic iron oxide nanoparticles for cell labeling, Polymer (2016), http://dx.doi.org/10.1016/j.polymer.2016.08.064