Poly(styrene-alt-maleic anhydride)-Based Diblock

Jun 23, 2014 - Center for Musculoskeletal Research, University of Rochester Medical Center, Rochester, New York 14642, United States. •S Supporting ...
0 downloads 0 Views 8MB Size
Download PDF
Article pubs.acs.org/Biomac

Poly(styrene-alt-maleic anhydride)-Based Diblock Copolymer Micelles Exhibit Versatile Hydrophobic Drug Loading, DrugDependent Release, and Internalization by Multidrug Resistant Ovarian Cancer Cells Michael P. Baranello,† Louisa Bauer,‡ and Danielle S.W. Benoit*,†,‡,§ †

Department of Chemical Engineering and ‡Department of Biomedical Engineering, University of Rochester, Rochester, New York 14627, United States § Center for Musculoskeletal Research, University of Rochester Medical Center, Rochester, New York 14642, United States S Supporting Information *

ABSTRACT: Amphiphilic diblock copolymers of poly(styrene-alt-maleic anhydride)-b-poly(styrene) (PSMA-b-PS) and poly(styrene-alt-maleic anhydride)-b-poly(butyl acrylate) (PSMA-b-PBA) were synthesized via reversible addition− fragmentation chain transfer (RAFT) polymerizations. Polymers were well-controlled with respect to molecular weight evolution and polydispersity indices (PDI < 1.2). Additionally, RAFT allowed for control of diblock compositions (i.e., ratio of hydrophilic PSMA blocks to hydrophobic PS/PBA blocks) and overall molecular weight, which resulted in reproducible self-assembly of diblocks into micelle nanoparticles with diameters of 20−100 nm. Parthenolide (PTL), a hydrophobic anticancer drug, was loaded and released from the micelles. The highest loading and prolonged release of PTL was observed from predominantly hydrophobic PSMA-b-PS micelles (e.g., PSMA100-bPS258), which exhibited the most ordered hydrophobic environment for more favorable core−drug interactions. PSMA100-b-PS258 micelles were further loaded with doxorubicin (DOX), as well as two hydrophobic fluorescent probes, nile red and IR-780. While PTL released quantitatively within 24 h, DOX, IR-780, and nile red showed release over 1 week, suggesting stronger drug−core interactions and/or hindrance due to less favorable drug−solvent interactions. Finally, uptake and intracellular localization of PSMA100-b-PS258 micelles by multidrug resistant (MDR) ovarian cancer cells was observed by transmission electron microscopy (TEM). Additionally, in vitro analyses showed DOX-loaded PSMA-b-PS micelles exhibited greater cytotoxicity to NCI/ADR RES cells than equivalent free DOX doses (75% reduction in cell viability by DOX-loaded micelles compared to 40% reduction in viability by free DOX at 10 μM DOX), likely due to avoidance of MDR mechanisms that limit free hydrophobic drug accumulation. The ability of micelles to achieve intracellular delivery via avoidance of MDR mechanisms, along with the versatility of chemical constituents and drug loading and release rates, offer many advantages for a variety of drug delivery applications.

1. INTRODUCTION Nanoparticles (NPs) have shown great promise for delivery of cancer therapeutics by overcoming several drug delivery barriers.1,2 Systemic delivery of chemotherapeutics often leads to rapid blood clearance, chemical and enzymatic degradation, and off-target side effects including toxicity.3−5 Additionally, drugs that do reach target tissue may have reduced efficacy due to poor tumor penetration and multidrug resistance (MDR) mechanisms that limit intracellular drug concentrations.6,7 NPmediated delivery, however, prevents premature drug clearance, and NPs can be designed to avoid renal filtration and removal by the reticuloendothelial system (RES).8,9 In addition to favorable pharmacokinetics, NP drug carriers can also passively accumulate in tumor tissue by a phenomenon known as the enhanced permeability and retention (EPR) effect, or actively bind to specific cells when decorated with surface targeting ligands.10,11 Finally, NPs may also evade MDR mechanisms by © XXXX American Chemical Society

orchestrating drug uptake via specific endocytic pathways rather than passive diffusion.5 Several NP drug delivery platforms including liposomes,12 polymer−drug conjugates,13 nanocrystals,14 and polymer micelles15 have been investigated clinically to improve the efficacy of a variety of therapeutics. The Food and Drug Administration (FDA) has already approved a number of NP formulations as cancer therapeutics, with many more in various stages of clinical development.2,10 Among the several types of NP drug carriers, synthetic polymer micelles are of particular interest because they readily self-assemble into highly stable core−shell NPs, are capable of loading therapeutic drugs at high concentrations, and can be chemically modified for better Received: March 29, 2014 Revised: June 10, 2014

A

dx.doi.org/10.1021/bm500468d | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

Scheme 1. One-Step Synthesis of Poly(styrene-alt-maleic anhydride)-b-poly(styrene) via RAFT Polymerizationsa

a

Amphiphilic diblock copolymers can be conveniently synthesized in one step by providing an excess of styrene monomer in the above reaction. 4Cyano-4-dodecylsulfanyltrithiocarbonyl sulfanyl pentanoic acid (DCT) was used as the RAFT chain transfer agent.

Scheme 2. Two-Step Synthesis of Poly(styrene-alt-maleic anhydride)-b-poly(styrene) and Poly(styrene-alt-maleic anhydride)-bpoly(butyl acrylate) via RAFT Polymerizationsa

a A living PSMA macro-CTA is made in the first step by providing a 1:1 molar ratio of Sty/MA monomers in the presence of RAFT CTA. In the second step, Sty or BA monomers are added to a reaction vessel containing the living macro-CTA and thermal initiator to produce amphiphilic diblock copolymers of PSMA-b-PS or PSMA-b-PBA, respectively.

However, this study is the first to systematically characterize the self-assembly (size, morphology, stability), drug loading, and release of PSMA-b-PS diblocks that span a wide range of overall molecular weights and hydrophilic PSMA/hydrophobic PS ratios. In addition, PSMA-b-PBA micelles were similarly characterized to assess the effect of alternative core compositions with more amorphous physical properties. Importantly, the micelles characterized in this report were synthesized under identical conditions, which allows for a more direct comparison of the polymer NPs. Further investigation of drug loading and release behavior of one micelle formulation with a panel of hydrophobic drugs (or drug-like molecules) including parthenolide (PTL), doxorubicin (DOX), nile red, and IR-780 was used to study the effect of drug properties and drug−polymer interactions on loading and release from polymer micelles. Finally, successful micelle uptake by multidrug resistant ovarian cancer cells was demonstrated via transmission electron microscopy (TEM) and flow cytometry, and in vitro cytotoxicity profiles were analyzed to assess therapeutic advantages of DOX-loaded micelles over free drugs toward multidrug resistant cancer cells.

drug−polymer interactions, surface targeting, and/or responsive functionalities (e.g., pH, redox, temperature).14,16 With many key micelle characteristics (e.g., NP size, morphology, aqueous stability, drug loading/release, cellular interaction, etc.) depending on diblock copolymer molecular weight and composition, use of well-controlled polymerization techniques, such as atom transfer radical polymerization (ATRP),17 nitroxide-mediated polymerization (NMP),18 and reversible addition−fragmentation chain transfer (RAFT)19−22 polymerization, is essential to investigate the structure−function relationship of synthetic polymer micelles.23 RAFT polymerizations allow for great breadth of monomer incorporation and polymer end functionalities,22,24 and polymer growth is linear as a result of a RAFT chain transfer agent (CTA), which allows for controlled, block-by-block addition of monomers.25 In this study, RAFT polymerizations were used to synthesize well-controlled amphiphilic diblock copolymers of poly(styrene-alt-maleic anhydride)-b-poly(styrene) (PSMA-b-PS) and poly(styrene-alt-maleic anhydride)-b-poly(butyl acrylate) (PSMA-b-PBA) that readily self-assembled into water-soluble micelle nanoparticles. The size, morphology, and stability of a variety of similar PSMA-b-PS micelles have been investigated,26−30 including a few studies into hydrophobic drug loading and release behavior of these NP formulations.31,32 B

dx.doi.org/10.1021/bm500468d | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

system equipped with a solvent pump (Shimadzu LC-20AD), a differential refractometer (Shimadzu RID-10A), and a light scattering detector (Wyatt Technology DAWN TREOS). A 3 μm linear gel column (Tosoh TSK-Gel Super HM-N, 6.0 mm ID × 15 cm) was used in series with a 3 μm guard column (Tosoh Biosciences) in an oven chamber operating at 60 °C. The mobile phase consisted of spectroscopic grade DMF containing 0.5 M lithium bromide. A flow rate of 0.35 mL/min was used to analyze polymers dissolved in the mobile phase. Refractive index increments (dn/dc) of PSMA macroCTAs and PSMA-b-PS polymers were determined experimentally to be 0.121 and 0.142 mL/g, respectively. PSMA-b-PS polymer molecular weights and all polydispersity indices (PDIs) were calculated using ASTRA 6 software (Wyatt Technology). For PSMA-b-PBA polymers, LC Solutions (Shimadzu) software was used to determine molecular weight from peak areas using a refractive index detector and poly(methyl methacrylate) (PMMA) standards. Representative GPC traces for synthesized polymers are included in the Supporting Information. 2.3.2. 1H NMR Analysis. 1H NMR (Bruker 300 MHz) was performed on polymers (dissolved in deuterated DMSO) to confirm the conversion of monomers. NMR spectra of PSMA-b-PS prepolymerization solutions were compared to the spectra of the same solutions after a 72 h RAFT polymerization (without precipitation) to confirm complete conversion of maleic anhydride monomer. 2.3.3. Elemental Analysis. PSMA-b-PS diblock copolymers were analyzed at Columbia Analytical Services (Tucson, AZ) for C, H, and O content. C and H content were measured by high temperature combustion/(IR−ASTM), and O content was measured using oxygen pyrolysis/(IR−ASTM). Using GPC, 1H NMR, and elemental analysis, experimentally determined PSMA-b-PS diblock compositions were compared to theoretical calculations as seen previously27,29 (data presented in the Supporting Information). 2.3.4. Differential Scanning Calorimetry. Glass transition temperatures (Tg) of dry PSMA, PSMA-b-PS, and PSMA-b-PBA copolymers were compared using a TA Instruments Q2000 differential scanning calorimeter (DSC). Polymers were dried in a vacuum oven at 80 °C for 24 h to ensure all residual solvent was removed. Dry polymer (15− 20 mg) was placed in a hermetically sealed aluminum DSC pan and subjected to a heat/cool/heat cycle at a linear heating rate of 5 °C/ min. PSMA and PSMA-b-PS copolymers were examined over a temperature range of 40−170 °C, and PSMA-b-PBA copolymers were examined over a temperature range of −70−170 °C. Tg for each component was determined from the second heating curve and analyzed using Universal Analysis software (TA Instruments). Representative DSC thermograms are included in the Supporting Information. 2.4. Micelle Nanoparticle Formation and Characterization. 2.4.1. Self-Assembly of PSMA-b-PS and PSMA-b-PBA Diblock Copolymers into Micelle NPs. Dried diblock copolymers (200 mg) were dissolved in 30 mL of DMF. An equivalent volume of water was added to stirring polymer solutions at a rate of 24 μL/min via syringe pump.32 Upon completion of the syringe pump, polymer micelle solutions were dialyzed (MWCO 6000−8000 kDa) against water for 3 days, with replacement of dialysis media twice daily. Following dialysis, micelle solutions were passed through 0.45 μm hydrophilic PTFE syringe filters. Micelle concentrations were determined by lyophilization. Micelles were stored at 4 °C in distilled water. 2.4.2. Particle Size and Zeta-Potential Measurements by Dynamic Light Scattering (DLS). PSMA-b-PS and PSMA-b-PBA micelles were dissolved in PBS (0.1 mg/mL) and sized by dynamic light scattering (DLS, Malvern Instruments, Worcestershire, U.K.) using a 3.0 mm quartz cuvette with 633 nm incident laser source. Hydrodynamic diameters presented in Table 2 are number-based averages. DLS size distributions of synthesized micelles are included in the Supporting Information. 2.4.3. Particle Size and Morphology by Electron Microscopy. Transmission electron microscopy (TEM) was used to characterize the size and morphology of micelle NPs. Polymer micelles were solvated in PBS at 0.1 mg/mL and incubated 1:1 with 2% (v/v in

2. MATERIALS AND METHODS 2.1. Materials. Unless otherwise specified, all chemicals were purchased from Sigma-Aldrich (St. Louis, MO). Styrene (99%, ACS grade) and butyl acrylate (BA, 99% pure ACS grade) were purified by distillation. Maleic anhydride (MA) was recrystallized from chloroform. 2,2′-Azo-bis(isobutylnitrile) (AIBN) was recrystallized from methanol. All solvents used were spectroscopic grade. Unless otherwise specified, all water used was distilled with resistivity > 18 MΩ (ultrapure). 2.2. RAFT Polymer Synthesis. 4-Cyano-4-dodecylsulfanyltrithiocarbonyl sulfanyl pentanoic acid (DCT) was synthesized as described in ref 33 and used as the RAFT chain transfer agent (CTA) in all polymerizations (see the Supporting Information for details of DCT synthesis). 2.2.1. Synthesis of Poly(styrene-alt-maleic anhydride)-b-poly(styrene) (PSMA-b-PS) Diblock Copolymers. Amphiphilic PSMA-b-PS copolymers were made in either a one- or two-step RAFT polymerization process (refer to Schemes 1 and 2, respectively). For the one-step process, styrene (Sty) was added in excess of maleic anhydride (MA) (4:1 [Sty]/[MA]) in the presence of DCT (100:1 [monomer]/[CTA]) and AIBN thermal initiator (10:1 [CTA]/ [initiator]) in dioxane (50% w/w). The dissolved mixture was kept on ice and purged with nitrogen for 45 min. After purging, the solution was placed in a 60 °C oil bath for polymerization. The polymer molecular weight was monitored throughout the polymerization using gel permeation chromatography (GPC). After the desired molecular weight was attained, the polymer sample was exposed to air and diluted with acetone prior to isolating the PSMA-b-PS product by precipitation in petroleum ether. The final product was dried under vacuum at room temperature. One-step RAFT polymerizations were typically designed to produce 2 g of polymer diblock and achieved ∼70% yield and up to 90% conversion. PSMA-b-PS polymers were also synthesized in a two-step RAFT polymerization process by generating a “living” PSMA macro-chain transfer agent (macro-CTA). PSMA macro-CTA was synthesized by reacting a 1:1 molar ratio of Sty and MA in the presence of DCT (300:1 [monomer]/[CTA]) and AIBN thermal initiator (20:1 [CTA]/[initiator]) in dioxane (50% w/w). The dissolved mixture was kept on ice and purged with nitrogen for 45 min. Following the purge, the solution was placed in a 60 °C oil bath and allowed to polymerize. After the desired molecular weight was achieved (monitored by GPC), the polymer was exposed to air and diluted with acetone prior to isolating the PSMA macro-CTA by precipitation in ethyl ether. The PSMA polymer intermediate was dried under vacuum at room temperature (typically ∼6 g of dry mass, ∼70% yield). A hydrophobic PS block was added to the hydrophilic PSMA block by reinitiating the living macro-CTA with AIBN in the presence of Sty monomer. PSMA macro-CTA, Sty monomer (300:1 [macro-CTA]/ [Sty]), and AIBN (2:1 [macro-CTA]/[initiator]) were dissolved in dimethylformamide (DMF, 50% w/w) and purged on ice for 45 min. The solution was subsequently placed in a 60 °C oil bath and allowed to polymerize for an additional 72 h. The desired PSMA-b-PS diblock was exposed to air, diluted in acetone, purified by precipitation in petroleum ether, and dried under vacuum at room temperature. 2.2.2. Synthesis of Poly(styrene-alt-maleic anhydride)-b-poly(butyl acrylate) (PSMA-b-PBA) Diblock Copolymers. A two-step RAFT polymerization was also employed to produce PSMA-b-PBA amphiphilic diblock copolymers. PSMA macro-CTA was synthesized as described in section 2.2.1. In the second step of the polymerization, a hydrophobic butyl acrylate (BA) block was added to the PSMA macro-CTA. PSMA macro-CTA, BA (215:1 [macro-CTA]/[BA monomer]), and AIBN (2:1 [macro-CTA]/[initiator]) were dissolved in DMF (50% w/w) and purged on ice with nitrogen for 45 min. Following the purge, the solution was placed in a 60 °C oil bath and allowed to polymerize for an additional 72 h. PSMA-b-PBA diblock was exposed to air, diluted in acetone, purified by precipitation in petroleum ether, and dried under vacuum at room temperature. 2.3. Polymer Characterization. 2.3.1. Determination of Polymer Molecular Weight by Gel Permeation Chromatography (GPC). Gel permeation chromatography was preformed on a Shimadzu C

dx.doi.org/10.1021/bm500468d | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

loading efficiency = 100 × (mg drug loaded/mg initial drug mass), and drug loading capacity = 100 × (mg drug loaded/mg micelles). 2.4.6. Drug Release from Polymer Micelles. Drug-loaded micelles (5 mL, 4 mg/mL) were dialyzed (MWCO 6000−8000 kDa) against 2 L of phosphate buffered saline (PBS) at 37 °C. Release media was replaced once per day, and 100 μL samples were taken from the dialysis tubing at various times. For PTL release, samples were taken at 1, 4, 8, and 24 h. Slower releasing drugs were dialyzed against PBS at 37 °C and sampled once daily. Drug release was quantified using the detection methods described in section 2.4.5. In control experiments, free PTL and DOX released quantitatively from dialysis membranes within 24 h. Free nile red and IR-780 became insoluble during dialysis and precipitated to the bottom of membranes. To ensure that precipitated nile red and IR-780 did not artificially impact release results, samples of nile red, IR-780, nile-red loaded micelles, and IR780-loaded micelles were centrifuged at 5000 rpm for 5 min and the supernatant was used for determination of drug concentration in dialysis tubing. Using these methods, free nile red and IR-780 was undetectable within 1 h, while drug from loaded micelles released as described in Figure 3C. 2.5. Transmission Electron Microscopy of NP Uptake by NCI/ ADR RES Cancer Cells. Circular (12 mm diameter) glass coverslips (VWR) were placed in 12-well tissue culture plates (1 coverslip per well). A total of 40 000 NCI/ADR-RES ovarian cancer cells dispersed in complete growth media (MEM + 10% fetal bovine serum + 1% penicillin−streptomycin) were added to each well (1 mL) and allowed to adhere to glass slide overnight. The following day, cells were replenished with fresh media containing micelles at 0.04 mg/mL (untreated cells replenished with fresh media not containing micelles). Cells were incubated at 37 °C (5% CO2) for 2 h, washed three times with PBS, and fixed with 2.5% glutaraldehyde and 4% paraformaldehyde in 0.1 M Millonig’s buffer. Fixed coverslips were processed into epoxy resin and thin sectioned at 70 nm onto grids using a “pop-off” technique before TEM imaging.36 2.6. Intracellular Uptake and Accumulation of DOX via Micelle Delivery to NCI/ADR RES Cells. Cells resuspended in complete growth media (MEM + 10% fetal bovine serum + 1% penicillin−streptomycin) were added to TC-treated 24-well plates (Greiner Bio-One) at 10 000 cells/cm2 and allowed to adhere and spread for 24 h. Cells were subsequently washed with PBS and incubated in 1 mL of fresh media containing either free DOX or DOXloaded PSMA-b-PS micelles at 10 μg/mL DOX. Cells were incubated at 37 °C (5% CO2) for 1, 2, or 4 h, subsequently washed thrice with PBS, trypsinized, and suspended in 0.5% BSA in PBS. Flow cytometry data was collected using an Accuri C6 flow cytometer (BD Biosciences). The median fluorescence intensity of FL2 detector (blue 488 nm laser, 585 nm detector with 40 nm band pass filter) was analyzed for 5000 gated cell events using FlowJo software (version 10.0.6). In separate experiments, trypan blue dye was used to quench extracellular fluorescence (as described previously37), and no significant change in median fluorescence was observed for free DOX or DOX-loaded micelles. Therefore, all fluorescence data presented Figure 5A is due to intracellular DOX molecules. 2.7. Cytotoxicity of NCI/ADR RES Cells Treated with DOXLoaded PSMA-b-PS Micelles. Cells resuspended in complete growth media (MEM + 10% fetal bovine serum + 1% penicillin− streptomycin) were added to TC-treated 24-well plates (Greiner BioOne) at 10 000 cells/cm2 and allowed to adhere and spread for 24 h. Cells were subsequently washed with PBS and incubated in 1 mL of fresh media containing either free DOX (dissolved in DMSO, diluted in PBS) or DOX-loaded PSMA-b-PS micelles at 0.1, 0.5, 1, 5, or 10 μg/mL DOX. Untreated control groups refer to cells incubated with PBS (at highest equivalent dose) in fresh growth media. Cytotoxicity of unloaded PSMA-b-PS micelles (at highest equivalent doses) was also tested. Cells were incubated at 37 °C (5% CO2) for 24 h and cell viability was quantified using Quant-iT PicoGreen DNA quantificaiton kit (Invitrogen). 2.8. Statistical Analysis. Statistical significance was determined using Prism software (GraphPad Version 6.0). Drug loading and release data represent the average of two independently synthesized

water) phosphotungstic acid negative stain on 150 mesh Formvar/ carbon coated grids for 5 min. Excess aqueous solution was wicked away with filter paper.34 The grids were photographed at 80 000−200 000 magnification using a Hitachi 7650 transmission electron microscope operating at 80 kV with an attached Gatan 11 megapixel Erlangshen digital camera. 2.4.4. Determination of Critical Micelle Concentration (CMC) by PRODAN Assay. CMC measurements were made using a PRODAN assay as previously described. 3 5 Briefly, 6-propionyl-2(dimethylamino)naphthalene (PRODAN) was dissolved in methanol at 24 μM, and 10 μL of this solution was placed in each well of a 96well black plate. The plate was kept covered in a chemical fume hood overnight to allow the methanol to evaporate. Polymer micelle concentrations ranging from 0.1 ng/mL to 0.5 mg/mL were made in PBS, and 100 μL of each micelle solution was added to individual wells containing dry PRODAN dye. Micelles were allowed to incubate with PRODAN dye at 4 °C overnight. Samples were analyzed using a fluorescent plate reader (Tecan). An emission scan ranging from 400 to 600 nm was performed for PRODAN excitation at 360 nm for each micelle dilution. The ratio of the peak hydrophobic emission intensity (436 nm) to the peak hydrophilic emission intensity (518 nm) was plotted against the polymer concentration on a logarithmic scale (see Supporting Information Figure 6). The intersection of the hydrophilic and hydrophobic regimes of the dilution curve is designated the CMC.35 2.4.5. Hydrophobic Drug Loading into Polymer Micelles. Comparison of loading for one hydrophobic drug (parthenolide, PTL) was made across all micelle compositions. Additionally, the micelles that displayed the best drug loading (PSMA100-b-PS258) were tested for their ability to load a variety of hydrophobic molecules, including doxorubicin (DOX), nile red, and IR-780. For all loading comparisons, 10 mL of micelles dissolved in water at a concentration of 2 mg/mL was added to stirring solutions of hydrophobic drugs (3 mg) dissolved in 1 mL of chloroform. For DOX loading, triethylamine (TEA; 4:1 [TEA]/[DOX]) was added to the initial drug−chloroform solution to allow neutralization of DOX-HCl drug ion and increase DOX solubility in the organic phase. The drug−micelle solution was stirred vigorously overnight uncovered in a chemical fume hood shielded from light to allow evaporation of chloroform. Free drug was separated from drug-loaded micelles by centrifugation and centrifugal filtration. Drug−micelle solutions were centrifuged at 2000 rpm for 10 min to remove any insoluble drug aggregates that may have formed during loading, followed by three rounds of centrifugal filtration using a 100 000 MWCO Amicon Ultra-15 centrifugal filter device (Millipore) to remove any unloaded soluble drug. The final concentrate was reconstituted to 5 mL in water, passed through 0.45 μm PVDF aqueous syringe filters, and lyophilized. A mock PTL loading experiment, in which water (no micelles) was used as the aqueous phase, was performed to validate that PTL could be completely removed by the described purification method. The final concentrate of the mock sample showed no detectable PTL by HPLC. Additionally, after the third centrifugal filtration of PTL-loaded micelles, no PTL was detected in the filtrate, suggesting that three rounds of centrifugal filtration was sufficient to remove free, soluble PTL from solution. PTL loading was quantified using high performance liquid chromatography (HPLC) with a mobile phase consisting of HPLCgrade water and methanol. HPLC analysis was completed using a Kromasil C18 column (50 mm × 4.6 mm, 5 μm particle size, 100 Å pore size). The column effluent was monitored with a variable wavelength UV−vis detector at 210 nm (Shimadzu). Flow conditions were as follows: 0.5 mL/min flow rate; gradient elution 0−3 min 95% water in methanol, 3−10 min 30% water in methanol. DOX, nile red, and IR-780 loading was quantified by diluting aqueous drug-loaded micelles in methanol (1:10, aqueous/methanol) and comparing absorbance readings to the standard curves prepared for each molecule in methanol at the relevant wavelength (DOX, 485 nm; nile red, 550 nm; IR-780, 780 nm). Drug loading efficiency and capacity were calculated according to the following definitions: drug D

dx.doi.org/10.1021/bm500468d | Biomacromolecules XXXX, XXX, XXX−XXX

Biomacromolecules

Article

Table 1. Characteristics of a PSMA Macro-CTA and PSMA-b-PS and PSMA-b-PBA Amphiphilic Diblock Copolymers Synthesized by One- and Two-Step RAFT Polymerizations polymer composition

RAFT polymerization process

diblock Mn (kDa)

diblock PDI (Mw/Mn)

PSMA repeat units/PS or PBA repeat units

hydrophilic Mn/ hydrophobic Mn

glass transition temperature (Tg, °C)

PSMA114 PSMA63-b-PS10 PSMA59-b-PS63 PSMA100-b-PS258 PSMA114-b-PS308 PSMA114-b-PBA118 PSMA186-b-PBA107

1-step 2-step 1-step 1-step 2-step 2-step 2-step

23.1a 13.8a 18.5a 46.2a 55.1a 38.1b 51.3b

1.02 1.1 1.1 1.05 1.03 1.02 1.07

6.3 0.9 0.4 0.4 1.0 1.7

12.2 1.8 0.7 0.7 1.5 3.3

140 NT NT NT 90, 160 −44, 166 NT

a

Number-average molecular weight (Mn) of PSMA and PSMA-b-PS diblocks determined by GPC with light scattering detection (PSMA dn/dc = 0.121, PSMA-b-PS dn/dc = 0.142). bMn of PSMA-b-PBA diblocks determined by GPC with refractive index detection calibrated with PMMA standards. All PDIs determined using combined light scattering and refractive index detection with ASTRA 6 software (Wyatt Technology). DSC was used to determine the glass transition temperature(s) (Tg) of representative polymers. DSC thermograms can be found in the Supporting Information (NT denotes polymers were not tested by DSC).

Successful diblock synthesis was confirmed by GPC and 1H NMR. Using DSC, PSMA114-b-PBA118 showed two Tg’s at −44 and 166 °C, displaying less order in the core-forming (hydrophobic) segments than PSMA-b-PS copolymers (Tgs at 90 and 160 °C). Final PSMA-b-PBA products displayed similarly low PDIs compared to PSMA-b-PS polymers, but addition of BA to the PSMA macroCTA was slower than Sty addition and usually resulted in a plateau in molecular weight over a 72 h reaction at 60 °C (∼60% conversion). For this reason, the most hydrophobic PSMA-b-PBA diblock was limited to a hydrophilic Mn/hydrophobic Mn ratio of 1.5. This plateau is not uncommon for RAFT polymerizations, due to decreased probability of monomer−CTA interaction over long reaction times.25 To achieve a lower ratio (more hydrophobic polymer diblock), a smaller PSMA macro-CTA or a higher [BA]/[macro-CTA] feed ratio could be employed.41 Overall, RAFT syntheses resulted in low PDI amphiphilic diblock copolymers of PSMA-b-PS and PSMA-bPBA with well controlled molecular weights and distinct hydrophilic (PSMA) Mn/hydrophobic (PS/PBA) Mn ratios. 3.2. Amphiphilic Diblock Copolymers Self-Assemble into Uniform Polymer Micelle NPs. All amphiphilic diblock copolymers listed in Table 1 were capable of self-assembling into micelle NPs using a DMF−water emulsion technique (see section 2.4.1). The dropwise addition of water protected against rapid polymer precipitation, and subsequent dialysis against water resulted in complete removal of organic solvent. Table 2 lists characteristics of the polymer micelles, including hydrodynamic diameter (number-based, measured by DLS), zeta potential, and CMC. Others have used a variety of nucleophiles (including sodium hydroxide and primary/ secondary amines) to explicitly ring-open maleic anhydride units and increase polymer hydrophilicity.39 Interestingly, all PSMA-b-PS and PSMA-b-PBA micelles presented in Table 2 were soluble in aqueous solutions, indicating the hydrolysis of MA within coronas without nucleophilic modification. Maleic anhydride groups in alternating copolymers have been shown to have rapid ring hydrolysis (on the order of seconds to minutes) at neutral pH, and similar hydrolysis is expected with alternating PSMA units herein.42 The presence of hydrolyzed MA is supported by negative zeta potentials observed for all micelles. Importantly, PSMA exhibits maximum chain stretching in aqueous media around neutral pH, and the degree of PSMA ionization has been shown to have little effect on PSMAb-PS micelle size.31

micelle batches prepared, loaded, and released under identical conditions as described in sections 2.4.1, 2.4.5, and 2.4.6, respectively. For one-variable (Figure 2) and two-variable (Figures 3, 5, and 6) comparisons, one-way or two-way ANOVAs with Tukey's posthoc analysis were used to determine statistical significance (p values specifically indicated in figure legends).

3. RESULTS AND DISCUSSION 3.1. RAFT Polymerizations Used to Controllably Synthesize PSMA-b-PS and PSMA-b-PBA Amphiphilic Diblock Copolymers. RAFT polymerizations were exploited to synthesize PSMA macroCTAs, as well as PSMA-b-PS and PSMA-b-PBA diblock copolymers. Previous studies have employed one-27,29,31,38,39 and two-step28,32 RAFT polymerizations to synthesize PSMA-b-PS block copolymers. As compatibility of monomers and RAFT agent is essential to polymerization kinetics, conversion, and polydispersity, guidelines for selection of a suitable RAFT CTA for desired polymers can be found in literature.25 4-Cyano-4-dodecylsulfanyltrithiocarbonyl sulfanyl pentanoic acid (DCT) is a trithiocarbonate RAFT CTA that allows for high addition/fragmentation rates and stable radical leaving groups when used with Sty, MA, and BA monomers, which promotes polymer monodispersity.25 For these features, DCT has been used for controlled growth of poly(styrene-co-methyl acrylic acid) copolymers.40 In the present study, DCT was used as the RAFT CTA to controllably synthesize all polymers. Successful DCT synthesis and monomer conversion was confirmed by 1H NMR (Supporting Information Figure 1). Additionally, GPC was used to determine the molecular weights and polydispersity indices of RAFT-synthesized polymers. Characteristics of the amphiphilic diblock copolymers investigated in this study are presented in Table 1 with all polymers exhibiting low PDIs (70% efficiency) for PTL to