Multiple and Co-Nanoprecipitation Studies of Branched Hydrophobic

Mar 12, 2015 - Andrew Owen,. ‡ and Steve P. Rannard*. ,†. †. Department of Chemistry, University of Liverpool, Crown Street, Liverpool L69 7ZD, ...
0 downloads 0 Views 6MB Size
Download PDF
Article pubs.acs.org/Macromolecules

Multiple and Co-Nanoprecipitation Studies of Branched Hydrophobic Copolymers and A−B Amphiphilic Block Copolymers, Allowing Rapid Formation of Sterically Stabilized Nanoparticles in Aqueous Media Jane Ford,† Pierre Chambon,† Jocelyn North,† Fiona L. Hatton,† Marco Giardiello,† Andrew Owen,‡ and Steve P. Rannard*,† †

Department of Chemistry, University of Liverpool, Crown Street, Liverpool L69 7ZD, U.K. Department of Molecular and Clinical Pharmacology, University of Liverpool, Block H, 70 Pembroke Place, Liverpool L69 3GF, U.K.

‡

S Supporting Information *

ABSTRACT: Nanopreciptation of hydrophobic polymers into aqueous media is still a relatively poorly understood phenomenon. Here we present the first model studies of branched vinyl polymer nanoprecipitation and the co-nanoprecipitation of the branched polymers with linear amphiphilic A−B block copolymers. The instantaneous formation of nanoprecipitates is shown after addition of polymers in a good solvent to the aqueous antisolvent, and the limits of the solvent mixtures are determined. Multiple nanoprecipitations are also shown to be possible through direct addition of polymer solutions to aqueous nanoparticle dispersions after solvent removal, leading to particle concentration increase without modification of the initial nanoprecipitate size. When nanoprecipitation of branched polymer/A−B block copolymer mixtures were studied, variation of the ratio of the two polymer architectures led to varying z-average diameters, narrow particle size distributions, tunable stability to salt addition and storage within aqueous salt conditions. A mechanistic rationale was investigated using a simplified classical DLVO approach, and the impact of branched copolymer:A−B block copolymer ratio is discussed.

■

INTRODUCTION During recent years, many research groups have focused on the formation and utilization of organic nanoparticles due to their unique properties and potential biological applications.1 Nanoparticles may be generated via several approaches including the so-called “top-down” attrition of powders and slurries into the submicron range to form nanosuspensions.2 In the pharmaceutical industry, this direct processing of poorly water-soluble drugs produces solid drug nanoparticles (SDNs) which are stabilized by the addition of polymers and/or surfactants.3 Recently, the manipulation of oil-in-water emulsions, containing water-soluble polymers and surfactants within the continuous phase and water-immiscible organic solvent solutions of drugs as the dispersed phase, has generated SDNs without the use of attrition strategies.4−6 Alternatively, nanoparticles containing either covalently attached7 or encapsulated drugs8 have been generated using many polymer classes. Nanoparticle and colloidal suspensions are also used in many environments other than medicine such as home and personal care products, agrochemicals, coatings, and inks. Structures such as micelles and vesicles may be produced by the self-assembly of well-designed amphiphilic linear block copolymers,9−12 which are often synthesized using a controlled radical polymerizations13 such as atom transfer radical © XXXX American Chemical Society

polymerization (ATRP) or reversible addition−fragmentation transfer.14−16 Such techniques utilize copolymers with a welldefined number-average degree of polymerization (DPn) and narrow dispersity (Đ) within each hydrophobic and hydrophilic block segment which play an important role in self-assembly processes and outcomes. As an alternative to self-assembly, nanoprecipitation is receiving increasing interest due to its versatility and compatibility with a range of polymeric materials and the rapid formation of well-defined nanoparticles and nanocapsules in water.17−19 The technique typically employs a solution of hydrophobic polymer within a good, water-miscible organic solvent. Addition to water, as a solvent-miscible antisolvent, leads to nanoparticle formation via a proposed nucleation/ growth mechanism;20 solvated and expanded chains collapse on rapid dilution of the good solvent with subsequent association to form colloidally stable nanoparticles. This approach has been very successful in the large-scale generation of polymer nanoparticles under clinically relevant conditions, as highlighted in recent reports of positive phase II human clinical trial results Received: January 16, 2015 Revised: March 5, 2015

A

DOI: 10.1021/acs.macromol.5b00099 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

report here a relatively facile technique to introduce steric stability to branched polymer nanoprecipitates through conanoprecipitation in the presence of an A−B linear amphiphilic block copolymer (Figure 1D,E). Co-nanoprecipitation differs from conventional nanoprecipitation as two polymers are dissolved in the organic solvent prior to precipitation into water and combine to form uniform mixtures during the nanoparticle formation process.

from nanoprecipitates derived from linear A−B block copolymers of poly(ethylene glycol) (PEG) and either poly(lactic acid), the Accurin technology,21−23 or poly(alkyl cyanoacrylate)s. The majority of nanoprecipitation reports are limited to biodegradable polyesters, polystyrene,24 and methacrylic acid/ poly(alkyl cyanoacrylate) based polymers,25 and the influence of polymer architecture variation on nanoprecipitation has not been widely studied.26 Indeed, we recently reported the first nanoprecipitation of hydrophobic branched copolymers comprised predominantly of poly(2-hydroxypropyl methacrylate) (pHPMA) and containing a low molar concentration of ethylene glycol dimethacrylate (EGDMA) to yield high molecular weight complex soluble architectures27−29 (Figure 1A−C). These reports utilized ATRP to synthesize linear and

■

RESULTS AND DISCUSSION Synthesis of Polymers. Three p(HPMA)-derived polymer samples with varying architectures were synthesized for the study using previously reported copper-catalyzed (Cu(I)Cl/ 2,2′-bipyridine), methanolic ATRP chemistry at 30 °C; a linear polymer with a DPn = 50 monomer units (p(HPMA50)), a particle-forming branched copolymer comprising p(HPMA50) primary chains connected through the copolymerization of EGDMA (p(HPMA50-co-EGDMA0.9)), and an amphiphilic A− B block copolymer stabilizer comprising poly(ethylene glycol) (PEG) and p(HPMA120) block segments (p(PEG45-bHPMA120)) (Scheme 1). Scheme 1. Cu(I)Cl/2,2′-Bipyridine Catalyzed Methanolic ATRP of HPMA Forming (A) Linear Homopolymer, (B) Amphiphilic A−B Polymer Using a PEG45-Derived Macroinitiator, or (C) Branched Copolymer Incorporating EGDMAa

Figure 1. Schematic representation of nanoprecipitated and conanoprecipitated branched copolymer particles in water: (A) collapse of branched copolymer when entering a bad solvent environment; (B + C) association and aggregation to form nanoparticles; (D) presence of A−B block copolymer leading to (E) sterically stabilized branched polymer nanoparticles.

branched statistical copolymers containing either ethyl functionality at the chain ends or ideally branched dendrons of varying generation. When linear−dendritic hybrid polymers are branched, the resulting materials have been termed “hyperbranched polydendrons”.28,29 In all cases, uniform polymer nanoparticles were generated by nanoprecipitation into water and were shown to result from rapidly assembling collapsed branched polymers and association to produce structures of diameters ranging from 60 to 800 nm (Figure 1A−C). A range of parameters such as polymer concentration, dilution factor, primary polymer chain length, composition, and architecture were used to control the process.27 Within these studies, nanoprecipitates generated from linear polymers were observed to aggregate and undergo macrophase separation on standing in water, while nanoprecipitates from branched copolymers were stable in water for extended periods (years). The stability of such materials in water was overshadowed by the observation that small amounts of NaCl(aq) led to large aggregates, suggesting a predominant charge stabilization mechanism, supported by highly negatively charged zetapotential (ζ) measurements. Stability to aqueous electrolyte is a critical factor when considering various applications of nanoparticles; therefore, we

a

Reaction A: [initiator]1:[monomer]50:[Cu(I)Cl]1:[bpy]2. Reaction B: [macroinitiator]1:[monomer]120:[Cu(I)Cl]1:[bpy]2. Reaction C: [initiator]1:[EGDMA]0.9:[monomer]50[Cu(I)Cl]1:[bpy].

Either an ethyl-functional initiator (linear homopolymer and branched copolymer synthesis) or a PEG-derived macroinitiator (PEG45-Br; DPn = 45 monomer units; see Supporting Information Figures S1 and 2) was employed in the ATRP reactions. HPMA was polymerized to target chain lengths of 120 monomer units, for the linear A−B block copolymer and DPn = 50 monomer units as a homopolymer or within the primary chains of the branched vinyl polymer, which also B

DOI: 10.1021/acs.macromol.5b00099 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

phases including microfluidics,31−33 simple dropping/evaporation, and combinations of sample addition and dialysis.34 In previous studies we have employed a rapid addition of polymer solutions to volumes of stirred water and chose to continue using this approach to allow direct comparison.27−29 Prior to our studies of co-nanoprecipitation, investigations of good solvent/antisolvent phase ratios were conducted as it was apparent that addition of large volumes of good solvent to the antisolvent phase could result in a liquid environment capable of preventing nanoprecipitation. If the nanoprecipitation mechanism is valid, nanoparticles are formed very quickly on mixing and will be present within the mixed solvent/antisolvent environment during solvent evaporation stages, rather than being formed slowly from a dissolved state as the good solvent is removed. In this model, any nanoparticles formed on mixing would be expected to be somewhat solvent swollen during the evaporation stages, and therefore nanoparticle sizes would be measurably larger immediately after mixing than those observed after solvent removal. Additionally, the mechanism of nucleation and growth is reported to continue until colloidally stable nanoparticles are formed, and achievement of colloidal stability removes the driving force for further aggregation and macrophase separation. If this is the case, it is plausible to expect a multiple nanoprecipitation process to be achievable, involving the steps of (a) addition of polymer dissolved in a good solvent to the aqueous antisolvent, (b) removal of good solvent to generate stable nanoparticles in water, and (c) a second and identical addition of polymer solution in a good solvent to the aqueous nanoparticle dispersion, with (d) removal of the second volume of good solvent to create a greater number of nanoparticles whose size matches the original distribution. To test these concepts, a series of experiments were conducted where a consistent mass of polymer was precipitated into a set volume of water with increasing amounts of good solvent and double nanopreciptations were studied with respect to the volume of good solvent added. Nanoprecipitation of a Fixed Mass of Polymer from Varying Solvent Volumes. Two good solvents for p(HPMA) were studied, as recent reports have shown that the nature of the good solvent can affect nanoprecipitation outcomes.21 Tetrahydrofuran (THF) and acetone were chosen and solutions of the linear p(HPMA50) and branched p(HPMA50co-EGDMA0.9) polymers were generated at a range of concentrations from 25 to 1.5625 mg/mL. Varying volumes of each solution ranging from 0.5 mL of the 25 mg/mL solution to 8 mL of the 1.5625 mg/mL solution were added to 5 mL of water, thereby ensuring that a consistent fixed mass of 12.5 mg of each polymer was precipitated into a set volume of water under varying solvent conditions (Supporting Information Table S1). Maintaining a constant mass of polymer during the different nanoprecipitations was important to evaluate the quality of the mixed solvent environments. When the linear p(HPMA50) was studied as described above, no clear trends were observed when measuring the samples immediately after addition and after full evaporation of the good solvent phase, using dynamic light scattering (DLS). A very high intensity of light scattering was only observed when less than 1 mL of either THF or acetone was present within the aqueous antisolvent (Supporting Information Figures S5 and S6). Nanoprecipitation of the branched polymer showed considerably different behavior. Highly significant light scattering intensities were seen when the branched polymer

incorporated a 0.9:1 molar ratio of EGDMA to ethyl initiator to prevent gelation, but ensure intermolecular branching between primary chains. 1 H nuclear magnetic resonance (NMR) spectroscopic analysis of the three polymers showed high monomer conversion (>98%) and was used to determine the degree of polymerization of the A−B block copolymer (DPn(NMR) = approximately 90 monomer units; Supporting Information Figure S3). Triple detection size exclusion chromatography (SEC with DMF eluent) was used to determine molecular weights of the synthesized polymers; the linear p(HPMA50) homopolymer was determined as having a higher than targeted number-average molecular weight (Mn) of 9900 g/mol and a weight-average molecular weight (Mw) of 12 400 g/mol (dispersity (Đ) = 1.25); the p(PEG45-b-HPMA120) A−B block copolymer (Supporting Information Figure S4) was characterized as Mn = 20 300 g/mol, Mw = 25 000 g/mol (Đ = 1.23); and the branched particle-forming p(HPMA50-coEGDMA0.9) exhibited a much higher molecular weight (Mn = 24 200 g/mol; Mw = 251 400 g/mol) and dispersity (Đ = 10.4) as a result of the presence of the EGDMA during the polymerization (Figure 2).

Figure 2. Triple detection SEC chromatography (DMF) of linear (green traces) and branched (red traces) poly(HPMA) samples. Responses are shown for the refractive index (solid lines) and right angle light scattering (dotted lines) detectors.

The broad dispersity of the molecular weight distribution is typical of branched copolymers formed via this mechanism; however, the complexity of the mixed architectures within the polymer sample has been shown to allow highly uniform nanoprecipitates.27−29 The high Mw value also implies that at least 50 wt % of the sample comprises branched polymer structures containing >25 conjoined primary chains, assuming attainment of the same DPn within the branched polymerization as seen for the linear polymerization in the absence of EGDMA (Mn = 9900 g/mol), as shown in previous reports of HPMA polymerization under these conditions27,30 and the overlaid SEC analysis from our materials (Figure 2). Nanoprecipitation Model Studies. Although the nanoprecipitation of polymers has been studied for several decades, aspects of the mechanism are not fully understood or studied in detail. Different approaches have been employed to lead to efficient and rapid mixing of the miscible solvent/antisolvent C

DOI: 10.1021/acs.macromol.5b00099 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules solutions were added to water up to 3 mL of good solvent (Figure 3Ai), and similarly, nanoparticles with z-average diameters (Dz) less than 250 nm were readily observed (Figure 3Aii). After evaporation of the good solvent, to leave nanoprecipitated particles dispersed in water, the measured Dz values were lower than seen in the presence of the good solvent. The nanoprecipitated particles also exhibited a uniform size distribution (PdI values 3 mL of each acetone solution were used in the nanoprecipitation, a sudden decrease in the light scattering intensity was seen, with a concomitant loss of quality within the correlation functions, within the samples measured immediately after acetone solution addition (Figure 3Ai). A corresponding decrease was observed in the measured Dz values; however, after solvent evaporation, particles with Dz values >1 μm were observed (Figure 3Aii) with broad polydispersities (PdI > 0.3); the slow formation of particles during evaporation of acetone is clearly not controlled. An identical trend was seen when using THF as the good solvent phase, but up to 4 mL volumes of polymer solution were able to generate nanoprecipitates immediately after addition and the sudden decrease in light scattering of samples containing good solvent and antisolvent was seen at volumes >4 mL (Supporting Information Figure S7). The solvent swollen nanoprecipitates formed immediately after THF solution addition were significantly larger than those derived from acetone solutions, and this increased size was also seen within the nanoparticles after solvent evaporation. It is unclear whether these differences indicate considerable or slight differences in the solubility of branched p(HPMA) in these aqueous solvent environments but the larger swollen nanopreciptates, and resulting solvent-free aqueous dispersions, do suggest different factors controlling the nucleation and growth mechanism when using these two solvents. Multiple Nanoprecipitations. As mentioned above, the formation of colloidally stable nanoparticles through nanoprecipitation has the potential to allow additional subsequent nanoprecipitations into the same antisolvent mixture, after removal of good solvent and assuming the mixed solvent/ antisolvent environment does not pass the nanoprecipitation boundary (Figure 3B) or perturb the already-formed nanoparticle distribution. THF and acetone solutions of p(HPMA50-co-EGDMA0.9) were generated at a concentration of 5 mg/mL, and varying volumes of the solutions from 0.5 to 4 mL were rapidly added to 5 mL of water with vigorous stirring; samples were studied by DLS after solvent evaporation. As can be seen from Table 1, for repeated additions of the branched polymer solution up to 2 mL of acetone addition, the recorded Dz values of the nanopreciptated particles are almost identical and only slight changes in PdI are observed (Supporting Information Figure S8) as would be expected if no perturbation of the initial nanodispersion occurs (Figure 3Biv) and the existing nanoparticles do not act as nuclei for further growth (Figure 3Biv). When repeating additions of volumes >2 mL, the nanoparticles formed vary considerably after the second addition and subsequent evaporation of good solvent; repeated addition of 4 mL of the polymer solution in acetone led to considerable phase separation. When using THF as the good solvent phase, similar behavior was observed but repeated nanoprecipitations only tolerated volumes of 50%) of the dissolved branched copolymer comprising >25 primary chains per molecule, the rapid transition from solvated polymer chains to solvent swollen collapsed nuclei will lead to an appreciable number of large nuclei that can further assemble to form the larger, colloidally stable nanoprecipitates. This process may be greatly facilitated by the branched architecture and the proximity of the primary chains. Co-Nanoprecipitation Studies. In general, nanoprecipitations of p(HPMA50-co-EGDMA0.9) from acetone gave lower PdI and smaller Dz values prior to, and after, good solvent removal; therefore, THF was not selected for further study. Acetone also has the practical advantage of a lower boiling point to aid removal. In order to avoid the boundary between the nanoprecipitation region and the phase ratios that were clearly seen to act as a good solvent (Figure 3), the addition of just 1 mL of polymer solution to 5 mL of water was chosen for co-nanoprecipitation studies. The concentration of polymer within the good solvent has also been shown to impact Dz values,27 and a study of nanoprecipitation using various volumes of solutions with polymer concentrations 25, 10, and 5 mg/mL confirmed that, irrespective of good solvent choice, solutions containing the branched polymer at 5 mg/mL consistently produced smaller nanoparticles (Supporting Information Figure S9). All co-nanoprecipitation studies were therefore conducted using 1 mL addition of a 5 mg/mL polymer solution in acetone to 5 mL of water and removal of good solvent over 24 h. The inclusion of a stabilizing amphiphilic A−B block copolymer into the nanoprecipitation process was evaluated by dissolving varying combinations of p(PEG45-b-HPMA120) and p(HPMA50-co-EGDMA0.9) in acetone, at a total polymer concentration of 5 mg/mL, for 12 h to ensure complete polymer dissolution, prior to rapid addition to water. The broad molecular weight distributions of branched polymers synthesized by branched vinyl polymerization are known to include linear polymers corresponding to the primary chains that have not been joined into the complex architectures; these linear chains are incorporated into the branched polymer nanoparticles during nanoprecipitation. As such, the incorporation of the linear p(PEG45-b-HPMA120) block copolymer, comprising a long p(HPMA) segment, during the nucleation and growth process would be reasonable to be expected, resulting in a co-nanoprecipitate and introducing sterically stabilizing PEG chains. It is important to note that the p(PEG45-b-HPMA120) is not readily soluble in water; a solubility of 70 wt %, resulted in nanoparticles with diameters >150 nm. As previously reported, the zeta-potential of the nanoprecipitated p(HPMA50-co-EGDMA0.9) was highly negative (ζ = −40.5 mV),27 but nanoparticles formed solely of A−B block copolymer showed lower values (ζ = −10.5 mV), consistent with PEG-stabilized particles or micelles.35,36 Incorporation of just 10 wt % of the stabilizing A−B block copolymer into the acetone solution with the branched p(HPMA50-co-EGDMA0.9) led to a considerable decrease in ζ values, indicating incorporation of the block copolymer; Dz and PdI values were also considerably reduced from the values obtained from nanoprecipitation of the linear A−B block copolymer alone. Dn values determined by DLS were consistent with the same assessment by SEM (220 particles per reported DSEM value, Supporting Information Figure S11); although evidence of nanoparticle formation was seen, SEM analysis became increasingly difficult with increasing amounts of A−B block copolymer, some film formation and observation of objects resembling flattened vesicles at higher ratios (Figure 4 and Supporting Information Figures S12−S15). The influence of the A−B block copolymer on the composition of the nanoparticles was studied using fluorometric analysis of pyrene encapsulated within the nanoprecipitates, produced via a ter-nanoprecipitation of the two copolymers and a hydrophobic organic small molecule. Variation in the fine structure of the fluorescence spectrum, specifically the ratio of the intensities of the first and third vibrational bands (the I1/I3 ratio) from the pyrene guest molecule, allowed an indication of the polarity within the nanoparticle (Supporting Information Figure S16). Reports of I1/I3 ratios of pyrene in water vary from 1.8 to 2.0 at 20 °C,37,38 and I1/I3 was recorded as being 1.81 within our study. When pyrene was nanoprecipitated with p(HPMA50-co-EGDMA0.9), an I1/I3 ratio = 1.58 was measured; addition of pyrene to the co-nanoprecipitation solution containing both polymers led to

Figure 4. Scanning electron microscopy images of nanoprecipitates and co-nanoprecipitates produced using varying p(HPMA50-coEGDMA0.9):p(PEG45-b-HPMA120) weight ratios: (A) 100:00, (B) 80:20, (C) 60:40, (D) 40:60, (E) 20:80, and (F) 0:100.

a nonsystematic variation in I1/I3 from 1.56 to 1.60 (Supporting Information Table S3) across the 100:0 to 50:50 weight ratios of p(HPMA50-co-EGDMA0.9):p(PEG45-b-HPMA120). This indicates no appreciable differences in the polarity and composition of the internal nanoparticle environment and the substantial absence of PEG chains within the nanoparticle core. The combination of reduced ζ values and the lack of variance in the pyrene I1/I3 ratio suggests that PEG45 chains are at the surface of the co-nanoprecipitates, and these should offer a level of resistance to salt addition due to steric stabilization. This was evaluated across the range of nanoparticle compositions outlined in Table 2. Aliquots (20 μL) of an aqueous 0.5 M NaCl solution were added to 1 mL samples of the conanoprecipitated aqueous dispersions, and Dz and PdI values were measured during a period of 7 days at ambient temperature (Figure 5A). As shown previously, nanoparticles formed without the A−B block copolymer exhibited immediate aggregation/precipitation (Figure 5A,B and Supporting Information Figure S17). The addition of 10 wt % of the A−B block copolymer had no meaningful effect on the salt stability of the co-nanoprecipitates; however, 20 wt % addition led to stability over the full 7 day test period; an immediate approximate doubling in Dz was seen on addition of salt (110 to 190 nm), and little variation was observed from this point (Supporting Information Table S4, Figures S18 and S19). Further increases in A−B block copolymer led to increasing stability as demonstrated by a F

DOI: 10.1021/acs.macromol.5b00099 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

branched copolymer, and very little change was noted over the 7 day storage in the presence of NaCl (Dz(7)/Dz(blank) = 1.25). The ability of the co-nanoprecipitates to tolerate salt was in marked contrast to the behavior of the branched copolymer p(HPMA50-co-EGDMA0.9) nanoprecipitated in the absence of the A−B block copolymer. Although stability to a set concentration of salt is of interest, it is also important to establish the impact of varying salt concentrations over a relatively short period of time, for example after injection or during formulation with other ingredients. As such, an evaluation of the behavior of fresh samples of the varying conanoprecipitates was studied as successive aliquots of the 0.5 M NaCl solution were added to 1 mL samples (Figure 6); again, DLS measurement was used to study the Dz and PdI of the nanoparticles.

Figure 5. Analysis of co-nanoprecipitated particles with varying p(HPMA50-co-EGDMA0.9):p(PEG45-b-HPMA120) weight ratios after addition of aqueous 0.5 M NaCl (20 μL into 1 mL of sample at 1 mg/ mL). (A) Z-average diameter variation of samples. Open squares: red = 100:0, blue = 90:10, green = 0:100. Open circles: red = 80:20, blue = 70:30, green = 60:40, gray = 50:50, purple = 40:60, black = 30:70, orange = 20:80, dark pink = 10:90. (B) Photograph of p(HPMA50-coEGDMA0.9):p(PEG45-b-HPMA120) co-nanoprecipitates in water at different weight ratios: 100:0 (i) before salt addition and (ii) immediately after salt addition; 60:40 (iii) before salt addition and (iv) immediately after salt addition. (C) DLS size distributions of aqueous co-nanoprecipitates of p(HPMA50-co-EGDMA0.9):p(PEG45-bHPMA120) with a 60:40 wt ratio before and after NaCl addition.

Figure 6. Analysis of co-nanoprecipitated particles with varying p(HPMA50-co-EGDMA0.9):p(PEG45-b-HPMA120) weight ratios after sequential addition of aliquots of aqueous 0.5 M NaCl to a 1 mL sample of nanoparticles at 1 mg/mL. Dz variation of samples is shown. Open squares: red = 100:0, blue = 90:10, green = 0:100. Open circles: red = 80:20, blue = 70:30, green = 60:40, gray = 50:50, purple = 40:60, black = 30:70, orange = 20:80, dark pink = 10:90.

decreasing impact on Dz immediately after salt addition, low Dz values on storage, and maintenance of low PdI values (100% of the available PEG45 chains (i.e., too few PEG45 chains were present) within the nanoprecipitation solution to generate complete coverage of the nanoparticle surfaces (Figure 7B); this ratio was the only nanoprecipitate containing the A−B block copolymer that aggregated instantly on addition of NaCl. In addition, when the PEG45 required to cover the surfaces of the nanoparticles fell below 20% of the available PEG45 within the polymer mixture (i.e., a large excess of A−B block copolymer is present), a loss of polymer size control and an indication of salt instability were seen. This may be another limit to successful co-nanoprecipitation as competitive nanoprecipitation of the poorly soluble A−B block copolymer may lead to inhomogeneity within the composition of the nanoparticles or the potential for vesicle and micelle formation, as discussed earlier. The generality of these findings will require further study.

■

■

CONCLUSIONS The nanoprecipitation studies that have been conducted support the previous views that there is a clear advantage to using branched p(HPMA-co-EGDMA) copolymers rather than linear homopolymers when seeking to achieve narrow, monomodal polymer nanoparticle distributions. The direct observations that (a) nanoparticles are formed immediately after addition of the polymer within a good solvent to the antisolvent, (b) the addition of large volumes of good solvent can prevent nanoprecipitation, and (c) repeated nanoprecipitations into the same antisolvent sample led to increased numbers of particles that match previous nanoprecipitations and also support the nucleation and growth mechanism that has been proposed by us and other groups.20,28 It appears that branched chains can effectively collapse to form relatively large nuclei that assemble to form the larger, colloidally stable nanoparticles. Under conditions where these initial nanoparticles are not perturbed by further addition of good solvent, new nuclei are formed that assemble into a second, but virtually indistinguishable, population of nanoparticles. Previous studies have investigated the interaction energies of particles and unimers from single polymer chains and shown an overall attractive force leading to growth of particles.24 If the relatively large nuclei from collapsed branched polymers present a more repulsive interaction with existing nanoparticles, or two nuclei present a higher attractive force than nuclei and exisiting nanoparticles, the growth of a new particle distribution would be expected.

EXPERIMENTAL SECTION

Materials. Poly(ethylene glycol) monomethyl ether (Mn ∼ 2000 g/mol), triethylamine (TEA, 99%), 4-(dimethylamino)pyridine (DMAP, 99%), 2-bromo-2-methylpropionyl bromide (BIBB, 98%), 2-hydroxypropyl methacrylate (HPMA, 97% mixture of isomers), ethyl 2-bromoisobutyrate (EBIB, 98%), Cu(I)Cl (99%), 2,2′-bipyridyl (bpy, 99%), ethylene glycol dimethacrylate (EGDMA, 98%), basic aluminum oxide, and Dowex marathon exchange beads were purchased from Sigma-Aldrich and used without any further purification. All solvents used were analytical grade. Instrumentation. 1H nuclear magnetic resonance (NMR) spectra were recorded in either DMSO-d6 or D2O using a 400 MHz Bruker Avance spectrometer. Chemical shifts are reported in parts per million (ppm) with respect to an internal reference of tetramethylsilane. Average molecular weights and dispersities (Đ) of the PEG45 macroinitiator and p(HPMA)s were estimated using a triple detection size exclusion chromatography instrument (SEC; Malvern/Viscotek) equipped with a GPCmax VE2001 autosampler, two Viscotek D6000 columns (and a guard column), and a triple detector array TDA305 (refractive index, light scattering, and viscometer) with a mobile phase of DMF containing 0.01 M lithium bromide at 60 °C and a flow rate of 1 mL min−1. Dynamic light scattering (DLS) measurements were performed using a Malvern Zetasizer Nano ZS (4 mW He−Ne 633 nm laser) on nanoparticle dispersions at 1 mg mL−1 (unless otherwise stated) and 25 °C. Size measurements were obtained as an average of three individual measurements. Nanoparticle dispersions were measured directly without additional filtration or centrifugation. Zeta-potentials were determined using the same apparatus described above. The measurements were obtained for aqueous dispersions at 1 mg mL−1 I

DOI: 10.1021/acs.macromol.5b00099 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules (unless otherwise stated) and 25 °C using disposable capillary zeta cells. Scanning electron microscopy SEM images were obtained using a Hitachi S-4800 FE-SEM. Aqueous nanoparticle samples (0.1 mg mL−1 in water) were dropped onto a silicon wafer mounted on an aluminum stub with a carbon tab, dried overnight at ambient temperature, and then gold sputter-coated (EMITECH K550X) at 20 mA for 2 min. Fluorescence spectra were obtained using a Shimadzu RF-5301PC spectrofluorophotometer. Emission spectra for pyrene were recorded between 300 and 500 nm. An excitation wavelength of λex = 335 nm was used for all studies as well as an excitational slit width of 1.5 nm and an emission slit width of 1.5 nm with a scan rate of 100 nm/min. Syntheses and Procedures. Synthesis of Poly(ethylene glycol) Monofunctional ATRP Macroinitiator (PEG45-Br). In a typical synthesis, MeO-PEG45-OH (30 g, 15 mmol, 1 equiv) was dissolved in toluene (100 mL) in the presence of triethylamine (2.275 g, 22.5 mmol, 1.5 equiv) and 4-(dimethylamino)pyridine (0.092 g, 0.75 mmol 0.05 equiv) in a two-necked round-bottomed flask fitted with an addition funnel, a nitrogen inlet/outlet, and a stirrer bar. 2-Bromo-2methylpropionyl bromide (5.175 g, 22.5 mmol, 1.5 equiv), diluted with toluene (15 mL), was placed into an addition funnel. The reactor was put under stirring at ambient temperature, the 2-bromo-2methylpropionyl bromide solution was added slowly over a period of 20−30 min, and the reaction was left to stir for 24 h. Caution: Reaction is exothermic. The rapid formation of a white precipitate (triethylamine salt Et3NH+Br−) indicated the progress of the reaction. The reaction medium was filtered and concentrated on the rotary evaporator. The resulting product was diluted in acetone and purified by precipitation into cold petroleum ether (40−60). This last step was repeated, and the product was finally dried under vacuum at 40 °C for 24 h. The resulting macroinitiator was confirmed by 1H NMR in D2O, triple detection SEC with a mobile phase of DMF, and MALDI-TOF mass spectrometry. Yield = 75%. 1H NMR (400 MHz, D2O): δ ppm = 4.31 (m, 2H), 3.76 (m, 2H), 3.56−3.4 (m, 181H), 3.55 (m, 2H), 3.31 (s, 3H), and 1.89 (s, 6H). MALDI-TOF: MNa+ = distribution ∼2054 Da. SEC: Mn = 2000 g/mol, Mw = 2100 g/mol, and Đ = 1.05. Atom Transfer Radical Polymerization (ATRP): Synthesis of Linear p(PEG45-b-HPMA120). In a typical reaction, targeting a number-average degree of polymerization (DPn) of 120 monomer units, PEG45-Br macroinitiator (0.62 g, 0.29 mmol, 1 equiv) and HPMA (5 g, 34.68 mmol, 120 equiv) were added to a round-bottomed flask equipped with a nitrogen inlet/outlet and a stirrer bar. Methanol was added (33.5 wt %, based on HPMA), and the solution was degassed via nitrogen sparge for 10−15 min. The copper salt Cu(I)Cl (0.029 g, 0.29 mmol, 1 equiv) and bpy ligand (0.09 g, 0.58 mmol, 2 equiv) were added to the flask, and the reaction medium was further degassed for 5 min. The reaction was carried out at 30 °C, and the monomer conversion was monitored by 1H NMR spectroscopy. The polymerization was terminated by exposure to air, and addition of THF when the HPMA monomer had reached >98% conversion. The copper catalytic system was removed by passing the polymer solution through an alumina column. The solution was concentrated by rotary evaporation, and the sample was purified by precipitation into 40−60 petroleum ether. The resulting polymer was characterized by 1H NMR in DMSO-d6 and triple detection SEC with a mobile phase of DMF. 1 H NMR (400 MHz DMSO-d6): δ ppm = 4.94−4.50 (br, OH), 3.94− 3.56 (br, 3H, CH2 and CH from pendant group), 3.56−3.40 (br, 184H), 3.24 (s, 3H), 2.10−1.50 (br, 2H, CH2 from polymer backbone), 1.30−0.50 (br, 6H, CH3 from polymer backbone and CH3 from pendant group). SEC: Mn = 20 300 g/mol, Mw = 25 000 g/ mol, and Đ = 1.23. Atom Transfer Radical Polymerization (ATRP): Synthesis of Linear p(HPMA50). In a typical ATRP synthesis, EBIB (0.14 g, 0.69 mmol, 1 equiv) and HPMA (DPn = 50 monomer units, 5 g, 34.68 mmol, 50 equiv) were added to a round-bottomed flask equipped with a nitrogen inlet/outlet and a stirrer bar. Methanol was added (50 wt %, based on HPMA), and the solution was degassed via nitrogen sparge for 10−15 min. The copper salt Cu(I)Cl (0.069 g, 0.69 mmol, 1 equiv) and bpy ligand (0.22 g, 1.39 mmol, 2 equiv) were added to the flask, and the reaction medium was further degassed for 5 min. The

reaction was carried out at 30 °C, and the monomer conversion was monitored by 1H NMR spectroscopy. The polymerization was terminated by exposure to air and addition of THF when the HPMA monomer had reached >98% conversion. The polymer was purified using Dowex Marathon exchange beads (∼10 g) in order to remove excess copper catalyst followed by passing the sample through a basic alumina column. Excess THF was removed under vacuum to concentrate the sample before precipitation into cold hexane. The resulting polymer was characterized by 1H NMR in DMSO-d6 and triple detection SEC with a mobile phase of DMF. 1H NMR (400 MHz DMSO-d6): δ ppm = 4.85−4.5 (br, OH), 4.12−3.150 (br, 3H, CH2 and CH from pendant group), 2.11−1.52 (br, 2H, CH2 from polymer backbone), 1.32−0.50 (br, 6H, CH3 from polymer backbone and CH3 from pendant group). SEC: Mn = 9900 g/mol, Mw = 12 400 g/mol, and Đ = 1.25. Synthesis of Branched p(HPMA50-co-EGDMA0.9). Branched p(HPMA50-co-EGDMA0.9) is synthesized and purified using the procedure described above for linear p(HPMA50) with the addition of the branching agent/comonomer EGDMA (0.12 g, 0.62 mmol, 0.9 equiv to EBIB initiator). The resulting branched material was characterized by 1H NMR in DMSO-d6 and triple detection SEC with a mobile phase of DMF. 1H NMR (400 MHz DMSO-d6): δ ppm = 4.85−4.50 (br, OH), 4.12−3.15 (br, 3H, CH2 and CH from pendant group), 2.11−1.52 (br, 2H, CH2 from polymer backbone), 1.32−0.50 (br, 6H, CH3 from the polymer backbone and CH3 from pendant group). SEC: Mn = 24 200 g/mol, Mw = 251 400 g/mol, and Đ = 10.4. Preparation of Polymer Nanoparticles. In a typical co-nanoprecipitation, targeting a weight fraction of 50% A−B diblock copolymer and 50% branched copolymer, a total mass of 50 mg of material is weighed out (25 mg of p(HPMA50-co-EGDMA0.9) and 25 mg of p(PEG45-b-HPMA120)) into a vial. The polymers were dissolved in 10 mL of analytical grade acetone over a period of 12 h to ensure complete solubilization. 1 mL of the resulting 5 mg mL−1 solution of polymers was added rapidly to 5 mL of distilled water (under vigorous stirring). The mixture was left for 24 h at ambient temperature to ensure complete acetone evaporation, leading to a final polymer concentration in water of 1 mg mL−1. Encapsulation of Pyrene. Pyrene was dissolved in acetone to give a stock solution (0.1 mg mL−1). The stock solution (300 μL, 0.1 mg mL−1) was added to an empty vial, and the acetone was allowed to evaporate to leave 30 μg of dye. For a typical 50:50 wt % conanoprecipitation, p(HPMA50-co-EGDMA0.9) (15 mg), p(PEG45-bHPMA120) (15 mg), and 6 mL of acetone were added to the vial to give a final concentration of polymer 5 mg mL−1 and pyrene 5 μg mL−1. The solution was left to stir for period of 12 h to ensure complete solubilization. The nanoparticles were prepared by rapid addition of 1 mL of the solution into vigorously stirring distilled water (5 mL). The mixture was left for 24 h at ambient temperature to ensure complete acetone evaporation, leading to a final polymer concentration in water of 1 mg mL−1 and 1 μg mL−1 pyrene in water. Solubility Evaluation of p(PEG45-b-HPMA120). The solubility of p(PEG45-b-HPMA120) in H2O was determined by rolling a 0.5 mg mL−1 solution of p(PEG45-b-HPMA120) in distilled H2O for 24 h. The sample was then allowed to settle, and 3 mL of the supernatant was added to a preweighed dish. After water evaporation (vacuum oven) the weight was recorded, and the amount of p(PEG45-b-HPMA120) was calculated. This procedure was repeated three times to generate an average value and standard deviation. NaCl Stability Studies. Storage over a 7 Day Period. Prior to study, 1 mL of the 1 mg mL−1 nanodispersion was added to a vial, and 20 μL of an aqueous NaCl (0.5M) solution was added. For the instant addition measurement, the solution was agitated before addition to a disposable sizing cuvette for DLS analysis (average of three measurements). The sample was returned to the vial for the 1 and 7 day measurements, and studies were carried out as described above. NaCl Stability Studies. Repeated Aqueous NaCl Addition. Prior to study, 1 mL of the 1 mg mL−1 nanodispersion was added to a disposable sizing cuvette, and multiple additions of aqueous NaCl (0.5 M, 2−2000 μL) were added, agitated quickly, and instantly measured using DLS (average of two measurements). J

DOI: 10.1021/acs.macromol.5b00099 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

■

(20) Lepeltier, E.; Bourgaux, C.; Couvreur, P. Adv. Drug Delivery Rev. 2014, 71, 86−97. (21) Cheng, J.; Teply, B. A.; Sherifi, I.; Sung, J.; Luther, G.; Gu, F. X.; Levy-Nissenbaum, E.; Radovic-Moreno, A. F.; Langer, R.; Farokhzad, O. C. Biomaterials 2007, 28, 869−876. (22) Farokhzad, O. C.; Jon, S.; Khademhosseini, A.; Tran, T-N. T.; LaVan, D. A.; Langer, R. Cancer Res. 2004, 64, 7668−7672. (23) Kamaly, N.; Fredman, G.; Subramanian, M.; Gadde, S.; Pesic, A.; Cheung, L.; Fayad, Z. A.; Langer, R.; Tabas, I.; Farokhzad, O. C. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 6506−6511. (24) Zhang, C.; Pansare, V. J.; Prud’homme, R. K.; Priestley, R. D. Soft Matter 2012, 8, 86−93. (25) (a) Khoee, S.; Yaghoobian, M. Eur. J. Med. Chem. 2009, 44, 2392−2399. (b) Le Droumaguet, B.; Nicolas, J.; Brambilla, D.; Mura, S.; Maksimenko, A.; De Kimpe, L.; Salvati, E.; Zona, C.; Airoldi, C.; Canovi, M.; Gobbi, M.; Magali, N.; La Ferla, B.; Nicotra, F.; Scheper, W.; Flores, O.; Masserini, M.; Andrieux, K.; Couvreur, P. ACS Nano 2012, 6, 5866−5879. (26) Hornig, S.; Heinze, T.; Becer, C. R.; Schubert, U. S. J. Mater. Chem. 2009, 19, 3838−3840. (27) Slater, R. A.; McDonald, T. O.; Adams, D. J.; Draper, E. R.; Weaver, J. V. M; Rannard, S. P. Soft Matter 2012, 8, 9816−9827. (28) Hatton, F. L.; Chambon, P.; McDonald, T. O.; Owen, A.; Rannard, S. P. Chem. Sci. 2014, 5, 1844−1853. (29) Hatton, F. L.; Tatham, L. M.; Tidbury, L. R.; Chambon, P.; He, T.; Owen, A.; Rannard, S. Chem. Sci. 2014, DOI: 10.1039/ C4SC02889A. (30) Li, Y.; Armes, S. P. Macromolecules 2005, 38, 8155−8162. (31) Valencia, P. M.; Farokhzad, O. C.; Karnik, R.; Langer, R. Nat. Nanotechnol. 2012, 7, 623−629. (32) Valencia, P. M.; Pridgen, E. M.; Rhee, M.; Langer, R.; Farokhzad, O. C.; Karnik, R. ACS Nano 2013, 7, 10671−10680. (33) Capretto, L.; Cheng, W.; Carugo, D.; Katsamenis, O. L.; Hill, M.; Zhang, X. Nanotechnology 2012, 23, 375602. (34) Zhang, C.; Chung, J. W.; Priestley, R. D. Macromol. Rapid Commun. 2012, 33, 1798−1803. (35) Ji, R.; Cheng, J.; Yang, T.; Song, C.-C.; Li, L.; Du, F.-S.; Li, Z.-C. Biomacromolecules 2014, 15, 3531−3539. (36) Deng, H.; Liu, J.; Zhao, X.; Zhang, Y.; Liu, J.; Xu, S.; Deng, L.; Dong, A.; Zhang, J. Biomacromolecules 2014, 15, 4281−4292. (37) Dong, D. C.; Winnik, M. A. Can. J. Chem. 1984, 62, 2560−2565. (38) Motokawa, R.; Morishita, K.; Koizumi, S.; Nakahira, T.; Annaka, M. Macromolecules 2005, 38, 5748−5760. (39) Lebovka, N. K. Adv. Polym. Sci. 2014, 255, 57−96. (40) Schneidera, C.; Hanisch, M.; Wedel, B.; Jusufi, A.; Ballauff, M. J. Colloid Interface Sci. 2011, 358, 62−67. (41) Derjaguin, B. V.; Landau, L. D. Acta Physicochim. URSS 1941, 14, 633−662. (42) Verwey, E. J. W.; Overbreek, J. T. G. Theory of the Stability of Lyophobic Colloids, the Interaction of Sol Particles Having an Electrical Double Layer; Elsevier: Amsterdam, The Netherlands, 1948. (43) Grasso, D.; Subramaniam, K.; Butkus, M.; Strevett, K.; Bergendahl, J. Rev. Environ. Sci. Bio/Technol. 2002, 1, 17−38. (44) Doane, T. L.; Chuang, C.-H.; Hill, R. J.; Burda, C. Acc. Chem. Res. 2012, 45, 317−326. (45) Ishikawa, Y.; Katoh, Y.; Ohshima, H. Colloids Surf., B 2005, 42, 53−58. (46) Cruje, C.; Chithrani, D. B. J. Nanomed. Res. 2014, 1, 00006. (47) Tadros, T. Adv. Colloid Interface Sci. 2011, 168, 263−277. (48) Wu, W.; Majkrzak, C. F.; Satija, S. K.; Ankner, J. F.; Orts, W. J.; Satkowski, M.; Smith, S. D. Polymer 1992, 33, 5081−5084.

ASSOCIATED CONTENT

S Supporting Information *

Figures and tables showing NMR spectra, MALDI-MS spectra, SEC chromatograms, DLS distributions and analysis, SEM images, calculations of concentrations for nanoprecipitation solutions, fluorescence spectra, and raw data from sample behavioral studies. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*(S.R.) E-mail [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors thank the Engineering and Physical Sciences Research Council for PhD studentships (J.F., F.L.H.), a vacation bursary (J.N.), and grant funding (EP/I038721/1) that underpinned the research. The University of Liverpool and the Centre for Materials Discovery at Liverpool are also gratefully acknowledged for access to scanning electron microscopy.

■

REFERENCES

(1) Letchford, K.; Burt, H. Eur. J. Pharm. Biopharm. 2007, 65, 259− 269. (2) Rabinow, B. E. Nat. Rev. Drug Discovery 2004, 3, 785−796. (3) Pawar, V. K.; Singh, Y.; Meher, J. G.; Gupta, S.; Chourasia, M. K. J. Controlled Release 2014, 183, 51−66. (4) Zhang, H.; Wang, D.; Butler, R.; Campbell, N. L.; Long, J.; Tan, B.; Duncalf, D. J.; Foster, A. J.; Hopkinson, A.; Taylor, D.; Angus, D.; Cooper, A. I.; Rannard, S. P. Nat. Nanotechnol. 2008, 3, 506−511. (5) McDonald, T. O.; Giardiello, M.; Martin, P.; Siccardi, M.; Liptrott, N. J.; Smith, D.; Roberts, P.; Curley, P.; Schipani, A.; Khoo, S. H.; Long, J.; Foster, A. J.; Rannard, S. P.; Owen, A. Adv. Healthcare Mater. 2014, 3, 400−411. (6) McDonald, T. O.; Martin, P.; Patterson, J. P.; Smith, D.; Giardiello, M.; Marcello, M.; See, V.; O’Reilly, R. K.; Owen, A.; Rannard, S. Adv. Funct. Mater. 2012, 22, 2469−2478. (7) Guan, X.; Hu, X.; Liu, S.; Huang, Y.; Jing, X.; Xie, Z. RSC Adv. 2014, 4, 55187−55194. (8) Haladjova, E.; Toncheva-Moncheva, N.; Apostolova, M. D.; Trzebicka, B.; Dworak, A.; Petrov, P.; Dimitrov, I.; Rangelov, S.; Tsvetanov, C. B. Biomacromolecules 2014, 15, 4377−4395. (9) Blanazs, A.; Armes, S. P.; Ryan, A. J. Macromol. Rapid Commun. 2009, 30, 267−277. (10) Chambon, P.; Blanazs, A.; Battaglia, G.; Armes, S. P. Langmuir 2011, 28, 1196−1205. (11) Cotanda, P.; Lu, A.; Patterson, J. P.; Petzetakis, N.; O’Reilly, R. K. Macromolecules 2012, 45, 2377−2384. (12) van Nostrum, C. F. Soft Matter 2011, 7, 3246−3259. (13) Wang, J.-S.; Matyjaszewski, K. Macromolecules 1995, 28, 7901− 7910. (14) Monteiro, M. J.; Cunningham, M. F. Macromolecules 2012, 45, 4939−4957. (15) Holder, S. J.; Sommerdijk, N. A. J. M. Polym. Chem. 2011, 2, 1018−1028. (16) Matyjaszewski, K. ACS Symp. Ser. 2000, 768, 2−26. (17) Mora-Huertas, C. E.; Fessi, H.; Elaissari, A. Int. J. Pharm. 2010, 385, 113−142. (18) Yan, X.; Delgado, M.; Fu, A.; Alcouffe, P.; Gouin, S. G.; Fleury, E.; Katz, J. L.; Ganachaud, F.; Bernard, J. Angew. Chem., Int. Ed. 2014, 53, 6910−6913. (19) Schubert, S.; Delaney, J. T.; Schubert, U. S. Soft Matter 2011, 7, 1581−1588. K

DOI: 10.1021/acs.macromol.5b00099 Macromolecules XXXX, XXX, XXX−XXX