Well-De Fined Cholesterol Polymers With pH-Controlled Membrane Switching Activity
Well-De Fined Cholesterol Polymers With pH-Controlled Membrane Switching Activity
Well-De Fined Cholesterol Polymers With pH-Controlled Membrane Switching Activity
pubs.acs.org/Biomac
■ INTRODUCTION
A number of potent therapeutic strategies rely on effective
at acidic pHs, enhancing interactions with cell membrane.
Poly(methacrylic acid), a pH-responsive polymer, has been
transport of macromolecular therapeutics, such as proteins/ widely utilized as an essential constituent for generating pH-
peptides and DNA/RNA, into cells. However, most macro- sensitive therapeutic delivery systems.31−34
molecular therapeutics cannot diffuse passively through cellular When compared with peptides and synthetic polymers, lipid-
membranes because of their large hydrodynamic size and based systems have been more effective in interacting with
hydrophilic nature.1−6 Various systems such as lipids,7−9 cellular membranes and enhancing the cytoplasmic delivery of
endosome-fusogenic peptides,10−13 and endosome-lytic poly- therapeutics. The naturally occurring hydrophobic building
mers14−26 have been developed to increase the intracellular blocks35 are essential for enhancing stability,36 stimulating self-
transport of macromolecular therapeutics. assembly,37,38 increasing cellular uptake via endocytosis
pH-responsive membrane-destabilizing polymers have im- mechanisms,39 disrupting endosomal membranes along with
proved the intracellular delivery of therapeutics both in facilitating effective transfection of macromolecular therapeu-
vitro27−29 and in vivo.30 These polymers that mimic the pH-
controllable membrane activity of endosome-fusogenic peptides Received: May 31, 2012
combine hydrophobic and acidic units in their structure and Revised: August 23, 2012
display a transition from a hydrophilic state to a lipophilic state Published: August 24, 2012
© 2012 American Chemical Society 3064 dx.doi.org/10.1021/bm300846e | Biomacromolecules 2012, 13, 3064−3075
Biomacromolecules Article
tics,8,4041 and generating carriers which are well tolerated in according to a modified procedure described previously in the
biological systems.42 For example, cholesterol, which accounts literature (Figure S1, Supporting Information).49,50 The RAFT agent,
for 20−25% of the lipid molecules in the cell membrane,43 4-(cyanopentanoic acid)-4-dithiobenzoate (CPADB), was synthesized
interacts easily with the cell membrane. It is easily transported according to the procedure reported in the literature.51,52 Membranes
for dialysis (MWCO 3500) were purchased from Fisher Biotech
into the cells by lipoproteins and albumin, usually via
(Cellu SepT4, regenerated cellulose-Tubular membrane).
endocytosis mechanism. In a number of studies,8,41 cholesterol Analytical Techniques. Nuclear Magnetic Resonance (NMR)
conjugation has proved to facilitate the entry of macro- Spectroscopy. All NMR spectra were taken using Bruker 300 MHz
molecular therapeutics into the cells. NMR spectrometer. According to their solubility, samples were
Considering the ability of cholesterol to cross cellular analyzed in CDCl3, DMSO-d6, or D2O as the NMR solvents.
membranes, we prepared well-defined copolymers of meth- Gel Permeation Chromatography (GPC). Gel permeation
acrylic acid and cholesteryl methacrylate, poly(methacrylic acid- chromatography was performed using HPLC grade tetrahydrofuran
co-cholesteryl methacrylate) P(MAA-co-CMA), as potential (THF) as the mobile phase. Polymer solutions (3−5 mg/mL in THF)
components of delivery systems for pH-controlled delivery of were injected to GPC at 40 °C (flow rate = 1 mL/min). A Shimadzu
therapeutics into cells. Hence, the ability of cholesterol to modular system comprising of a SIL-10AD autoinjector, a PL 5.0 mm
bead-size guard column (50 × 7.8 mm) followed by four linear PL
interact with lipid membranes was potentially synergized with
(Styragel) columns (105, 104, 103, and 500 Å) and a RID-10A
multiplicity and pH-sensitivity in a copolymer structure. The differential refractive-index detector was used. Calibration was
reversible addition−fragmentation chain transfer (RAFT) performed with commercial polystyrene standards ranging from 500
polymerization44−48 was used to synthesize well-defined to 106 g/mol.
copolymers with controlled molecular weight and narrow UV−Visible (UV−vis) Spectrophotometer. UV−visible spectra were
molecular weight distribution. Aqueous solution properties of obtained by a double beam Hitachi−UV spectrometer (Model No.: U-
the copolymers were investigated by varying techniques 2800) using UV solutions 2.1 software. The absorbance of polymer
revealing information on hydrodynamic size, conformation, solutions (0.125 mM) in buffer solutions with varying pH values was
and phase transition as a function of pH. The pH-dependent measured at 400 nm using quartz cuvettes.
membrane-destabilizing activities of copolymers were examined Differential Scanning Calorimeter (DSC). DSC analysis was
by the liposome leakage assay and further lipid layer performed using a Perkin-Elmer Pyris 1 under N2 atmosphere, from
interactions were explored by surface plasmon resonance −10 to 250 °C at a heating rate of 15 °C/min. High purity indium was
used to calibrate the calorimeter. Thermal history difference was
(SPR) spectroscopy. Furthermore, the cytotoxicity of copoly- erased by reheating the sample and recording a second DSC scan; 10−
mers with varying cholesterol content was investigated using a 15 mg polymer was used for each running.
human neuroblastoma cell line. Overall, the results support the Dynamic Light Scattering (DLS). Dynamic light scattering studies
value of these new copolymers for further investigations as were performed using a Malvern Zetasizer NaNo ZS Instrument
components of delivery systems for transporting therapeutics (Malvern, U.S.A.) equipped with a 4 mV He−Ne laser operating at λ =
through cellular membranes. 633 nm, an avalanche photodiode detector with high quantum
■
efficiency, and an ALV/LSE-5003 multiple tau digital correlator
EXPERIMENTAL SECTION electronics system. The polymer sample solutions were prepared at
0.125 mM concentration.
Materials. The initiator, 2,2-azobisisobutyronitrile (AIBN), was Atomic Force Microscopy (AFM). Atomic force microscopy (AFM)
recrystallized twice from methanol prior to use. High purity nitrogen studies were performed using a Nanomagnetic AFM apparatus in
(Linde gases, 99.99%) was used for removing oxygen from the tapping mode. A total of 20 μL of the copolymer solution (0.125 mM)
polymerization solutions. tert-Butyl methacrylate (t-BMA) monomer at pH 5.0 was placed on silicon substrate, and the sample was dried
(Aldrich, 99%) was purified via basic alumina gel column under atmospheric conditions at room temperature. The scans were
chromatography before use. Triethylamine (Sigma-Aldrich, 99%)
typically done at rates between 1 and 4 Hz. The images were obtained
was stored with sodium hydroxide pellets (Univar) for 2 days prior
using a silicon nitride cantilever with a nominal force constant of 0.38
to use. Cholesterol (Sigma, 98%), methacryloyl chloride (Fluka,
N/m.
>97%), dichloromethane (Univar, analytical grade reagent), toluene
fMax Fluorescence Spectroscopy. A fMax Fluorescence Plate
(Univar, analytical grade reagent), methanol (Univar, analytical grade
reagent), acetone (Univar, analytical grade), trifluoro acetic acid Reader from fMax Fluorescence Microplate Molecular Devices
(Sigma), Triton X100 (Sigma), phosphate buffer saline (PBS) pellets Corporation was used for measurements in Liposome Lysis Assay
(Sigma), potassium dihydrogen phosphate (Univar), sulforhodamine and CellTiter-Blue Assay with fluorescence readings measured at
B (Sigma), egg yolk phosphatidylcholine (Sigma), chloroform 488λex/585λem nm and 540λex/590λem nm, respectively. Measurements
(Sigma), hexadecanethiol (Sigma), Dulbecco’s modified Eagle medium were performed by using the fMax Fluorescence user software
(DMEM; Lonza), 10% Australian fetal bovine serum (FBS; Lonza), program.
and CellTiter-Blue cell viability assay (Promega) were used as Methods. The procedure for the RAFT homopolymerizations of
received. Cholesteryl methacrylate (CMA) monomer was synthesized cholesteryl methacrylate (CMA) and tert-butyl methacrylate (t-BMA)
was described in Supporting Information (Schemes S1 and S2 and Healthcare Biacore) to monitor polymer binding onto lipid bilayers in
Figures S2 and S3). real-time. SPR gold slides, used for lipid bilayer immobilization, were
RAFT Copolymerization of tert-Butyl Methacrylate (t-BMA) and first washed with ethanol, treated with 0.1 mM hexadecanethiol/DCM
Cholesteryl Methacrylate (CMA). The copolymerizations were solution overnight (to produce self-assembled monolayer of
conducted at varying comonomer concentrations as shown in Table hexadecanethiol), then rinsed thoroughly with DCM, and finally
1. t-BMA and CMA monomers, initiator (2,2-azobisisobutyronitrile dried under a gentle stream of nitrogen.
(AIBN)), and RAFT agent (4-(cyanopentanoic acid)-4-dithiobenzoate Liposomes of different compositions were prepared to mimic both
(CPADB)) were dissolved in toluene. The solution was sealed with a the cell plasma membrane (composed of cholesterol/phosphatidylcho-
rubber septum and then degassed using nitrogen for 30 min in an ice line constituents at a cholesterol/phosphatidylcholine molar ratio of
bath. The polymerization was carried out by immersing the vial into a 0.42) and endosome membrane (molar ratio of 0.80) lipid
preheated oil bath of 68 °C. At predetermined time intervals arrangements.54 Small, unilamellar vesicles (SUV) were prepared by
polymerization solution was drawn out from the vial using a syringe sonication, according to the protocol described by Morrissey
for further analysis. The monomer conversion was determined by 1H Laboratories.55 In brief, egg yolk phosphatidylcholine (8.4 mg, 0.01
NMR analysis of the crude polymerization mixtures. The polymer mmol for plasma membrane; 7 mg, 0.08 mmol for endosome
sample was purified by precipitation twice in methanol and dried membrane) and cholesterol (1.6 mg, 0.04 mmol for plasma
under vacuo overnight. 1H NMR spectrum of the purified product is membrane; 3 mg, 0.07 mmol for endosome) were solubilized in 1
shown in Supporting Information (Figure S7A).1H NMR (CDCl3, 300 mL of chloroform (in a round-bottom flask) and dried under nitrogen,
MHz) ppm: 5.3 (d, 1H, -CCH-, olefin group in cholesterol), 4.5 (m, generating a lipid film that was stored overnight at +4 °C. The thin
1H, -COO-CH-), 2.3 (d, 2H, -CH-CH2-), 1.81 (s, 2H, -C-CH2-), 1.41 lipid film was resuspended in 5 mL phosphate buffered (pH 7.4) or
(s, 9H, -C-(CH3)3), 1.02 (s, 3H, -C-CH3), 0.93 (d, 3H, -CH-CH3), citrate-phosphate buffer (pH 5.0) by vortex mixing, hydrated for 1 h at
0.88−0.84 (q, 6H, -CH-(CH3)2), 0.68 (s, 3H, -C-CH3). The molecular room temperature and then sonicated forming SUVs.
weight of the purified polymers was obtained by gel permeation Liposomes were injected to the SPR at 20 μL/min for 6.5 min until
chromatography (GPC) using THF as the mobile phase. the surfaces were saturated. The solution was allowed to sit on the
Hydrolysis of Poly(tert-butyl methacrylate-co-cholesteryl meth- surface for 20 min. During this period, liposomes were attached to self-
acrylate) to Poly(methacrylic acid-co-cholesteryl methacrylate). assembled hexadecane monolayer surface and then disrupted to form
The copolymer P(t-BMA-co-CMA) (325 mg, 2.3 mmol tert-butyl lipid bilayer on the surface. Surfaces covered with lipid bilayer were
ester) was dissolved in dichloromethane (DCM; 5 mL). The mixture then washed with several pulses (ranging from 30 s to 5 min) of
was allowed to stir for 10 min to dissolve the polymer. Trifluoro acetic related buffer (citrate-phosphate buffer at pH 5.0 or phosphate buffer
acid (TFA; 0.894 mL, 114 mmol) was slowly added while vigorously at pH 7.4) to remove loosely bound lipids and stabilize the surfaces.
stirring. The reaction continued to stir for 32 h at room temperature. Polymer solutions (2 mol % CMA, with (Mn)GPC of 16500 g/mol; 4
Excess TFA along with DCM was removed at room temperature with mol % CMA with (Mn)GPC 15800 g/mol; 8 mol % CMA with (Mn)GPC
air flowing through the vial and dried under vacuum overnight. 18000 g/mol) prepared at 0.5 mM concentration in related buffer
Removal of tert-butyl group from the copolymer was verified by 1H solutions (citrate-phosphate buffer at pH 5.0 or phosphate buffer at
NMR pH 7.4) were injected over the surfaces at 20 μL/min flow rate,
1
H NMR (DMSO, 300 MHz) ppm: 12.3 (s, 1H, -CO-OH), 5.3 (d, allowing real-time monitoring of the interactions between polymers
1H, -CCH-, olefin group in cholesterol), 4.5(m, 1H, -COO-CH-), and lipid bilayers of different composition in acidic and neutral
2.3 (d, 2H, -CH-CH2-), 1.82 (s, 2H, -C-CH2-), 1.02 (s, 3H, -C-CH3), medium. Lipid bilayer surfaces that reached saturation point were
0.93 (d, 3H, -CH-CH3), 0.88−0.84 (q, 6H, -CH-(CH3)2), 0.68 (s, 3H, rinsed with 6.5 min injections of their related buffer solutions.
-C-CH3). Liposome Leakage Assay. Liposomes composed of cholesterol/
To obtain the final product, the copolymer, poly(methacrylic acid- phosphatidylcholine constituents at a cholesterol/phosphatidylcholine
co-cholesteryl methacrylate) P(MAA-co-CMA), was dissolved in basic molar ratio of 0.42 were prepared intending to mimic the cell plasma
aqueous solution (10 mM NaOH solution prepared with deionized membrane lipid composition.54 The liposomes were prepared by
water) and transferred to dialysis tubing (MWCO: 3500). The sonication according to a protocol described by Morrissey
solution was dialyzed against deionized water for 3 days followed by Laboratories.55 Briefly, egg yolk phophatidylcholine (42 mg, 0.05
freeze-drying giving a white powder. The average yield was 90% for mmol) and cholesterol (8 mg, 0.02 mmol) were dissolved in 1 mL
overall hydrolysis process. The final polymer was analyzed by 1H chloroform and placed in a round-bottom flask. A thin lipid film was
NMR (Figure S7B, Supporting Information). generated on the bottom of the flask by evaporation of the chlorofom
1
H NMR (D2O, 300 MHz) ppm: 5.3 (d, 1H, -CCH-, olefin via nitrogen gas and stored overnight at +4 °C. The lipid film was
group in cholesterol), 4.5 (m, 1H, -COO-CH-), 1.81 (s, 2H, -C-CH2-), hydrated with sulforhodamine B (SRB; 2.5 mM, 2.5 mL) solution in
1.02 (s, 3H, -C-CH3), 0.68 (s, 3H, -C-CH3). phosphate buffer saline (pH 7.4) and then vortexed vigorously. The
Turbidity Assay. The pH-responsive phase behaviors of polymers lipid solution was incubated for 1 h at room temperature followed by
were studied by measuring the turbidity change of polymer solutions sonication generating small, unilamellar vesicles (SUV) with diameters
at varying pH values via UV−visible spectroscopy.53 For spectropho- in the range of 15−50 nm. Untrapped SRB molecules were eliminated
tometric analysis, citrate buffer solutions in the pH range of 3.0−7.0 by size exclusion chromatography on a Sephadex G-25 column (bed
were prepared by mixing citric acid (0.1 M) and dibasic sodium volume 8.3 mL, bed height 5 cm; PD-10) using phosphate buffer saline
phosphate (0.2 M) aqueous solutions. A phosphate buffer solution as the eluting solution. 750 μL of the SRB loaded liposomes were
(0.1 M) at pH 7.4 was prepared by mixing sodium phosphate added to the isotonic buffer solution (3.75 mL, citrate-phosphate
monobasic (0.1 M) and sodium phosphate dibasic (0.1 M) aqueous buffer at pH 5.0 or phosphate buffer at pH 7.4). The solution was
solutions. The ionic strength of the buffer solutions were adjusted to equally divided into 6 vials (750 μL) and the final volume was made
0.1 M by the addition of NaCl to yield isotonic solutions. Three up to 900 μL by addition of isotonic buffer solution (in the absence of
different P(MAA-co-CMA) samples with varying cholesterol content, polymer), Triton X-100 solution (10% in Milli-Q water) or copolymer
(2 mol % CMA, with number average molecular weight (Mn)GPC of solution prepared in Milli-Q water at 1.5 μM or 3.0 μM concentration
16500 g/mol and polydispersity index (PDI) of 1.19; 4 mol % CMA, (2 mol % CMA, with (Mn)GPC of 16500 g/mol; 4 mol % CMA with
with (Mn)GPC 15800 g/mol and PDI 1.10; 8 mol % CMA, with (Mn)GPC 15800 g/mol; 8 mol % CMA with (Mn)GPC 18000 g/mol or
(Mn)GPC 18000 g/mol and PDI 1.11) were dissolved in buffer poly(methacrylic acid), PMAA, cholesterol mole content 0% with
solutions. The final copolymer concentration was 0.125 mM. The (Mn)GPC 36000 g/mol). The reason for using dilute copolymer
absorbance of each polymer solution from acidic pH to neutral pH was solutions were to determine the effectiveness of the copolymers in
detected by a UV−visible spectrophotometer at 400 nm. selective (pH-dependent) membrane destabilization, as detailed in
Surface Plasmon Resonance. Surface plasmon resonance (SPR) Results and Discussion. After 1 h incubation, the solutions were
experiments were performed using a Biacore 2000 system (GE transferred to a 96-well plate and the fluorescent intensity (λex = 488
Scheme 1. Synthesis of P(t-BMA-co-CMA) via RAFT Polymerization and Subsequent Acid Hydrolysis, Yielding
Poly(methacrylic acid-co-cholesteryl methacrylate) (P(MAA-co-CMA))
nm, λem = 585 nm) of the resulting solutions was measured by a fMax untreated cells (cells only in media, positive control for 100%
fluorescence plate reader (fMax Fluorescence Microplate Molecular viability), respectively. FB, is the fluorescence intensity of blank wells
Devices Corporation). In a control experiment, a series of SRB-loaded (wells with media and polymer treatment, no cells) where the values
liposomes were analyzed by the same protocol without using polymer. are similar to 10% Triton X-100 treatment, which corresponds to 0%
This experiment suggested that SPR-loaded liposomes, without viability. Samples were tested in triplicate using cells at different
polymer treatment, were stable in isotonic buffer solutions (pH 5.0 passage numbers.
■
and 7.4) as they showed similar fluorescence trends in both
environments. The calibration curves of SRB fluorescence were RESULTS AND DISCUSSION
obtained at both pH values prior to the liposome leakage experiment.
Percent leakage was defined as Copolymerization of tert-Butyl Methacrylate and
Cholesteryl Methacrylate. Owing to the solubility mismatch
%leakage = (Fp − F0)/(F100 − F0) × 100 between CMA monomer (water insoluble) and methacrylic
acid monomer (water soluble), a two-step reaction route was
where F0 and Fp are the fluorescence intensity of blank solution (SRB- adopted for polymerization (Scheme 1, step 1). Cholesterol
loaded liposomes in buffer solutions without polymer) and polymer
solutions (SRB-loaded liposomes in buffer solutions with polymer), was first esterified forming a methacrylate and then used for
respectively. A 100% leakage, F100, was taken as the fluorescence copolymerization with t-BMA in the presence of CPABD, a
intensity of the SRB-loaded liposome solutions after the addition of dithioester RAFT agent, which has been widely used for the
10% Triton X-100, which corresponded to a complete liposome lysis. polymerization of methacrylates.56 pH-responsive copolymers
Determination of Cell Viability via CellTiter-Blue Assay. Human of CMA were then formed by the hydrolysis of t-BMA units of
neuroblastoma (SH-EP) cell line was used to determine the toxicity of the copolymers to methacrylic acid units. tert-Butyl ester has
the polymers. The cells were grown in Dulbecco’s modified Eagle received significant attention as an effective protecting group
medium (DMEM) containing 4.5 g/L glucose and L-glutamine, due to the labile ester-alkyl bond.57 The tert-butyl protecting
supplemented with 10% Australian Fetal Bovine Serum (FBS). Cells group can be easily removed via acid-catalyzed processes such
were grown at 37 °C in 5% CO2 and 95% humidity.
The cytotoxicity of the P(MAA-co-CMA) on SH-EP cells was as toluene/toluenesulfonic acid,57,58 dioxane/HCl,59 or di-
evaluated using the CellTiter-Blue cell viability assay. The assay was chloromethane (DCM)/trifluoroacetic acid (TFA) treat-
performed according to the manufacturer’s protocol. In brief, SH-EP ments.60,61
cells were seeded in to 96-well plates at 3 × 103 cells/mL and grown GPC traces of P(t-BMA-co-CMA) (obtained at [t-BMA]/
for 24 h at 37 °C, 5% CO2 in DMEM containing 10% FBS. Cells were [CMA]/[RAFT]/[AIBN] molar ratio of 200.0/4.0/1.0/0.2,
then exposed to varying concentrations of P(MAA-co-CMA) and total monomer concentration of 4.08 M), shown in Figure 1A,
incubated for another 72 h. After the incubation period, 20 μL of reveals the formation of monomodal molecular weight
CellTiter-Blue was added to each well, and the plate was further distribution during polymerization. The monomodal GPC
incubated at 37 °C, 5% CO2 for 2−3 h. The fluorescence was recorded chromatograms indicate that the polymerization occurs in the
at λex = 540 nm and λem = 590 nm. The effect of the presence of
polymer on the viability of the cells was calculated as follows: absence of side reactions, which can cause branched
polydisperse polymers.62 It is evident from GPC traces that
⎧ F − FB ⎫ the molecular weight of P(t-BMA-co-CMA) increases with
%cell viability = ⎨ Tr ⎬ × 100 monomer conversion. The semilogarithmic monomer con-
⎩ FUntr − FB ⎭
version (determined by 1H NMR) shown in Figure 1B
FTr and FUntr are the fluorescence intensity of treated cells indicates that monomer conversion increases concomitantly
(treatmented with polymer solutions at varying concentrations) and with time. The evolution of the Mn with monomer conversion
3067 dx.doi.org/10.1021/bm300846e | Biomacromolecules 2012, 13, 3064−3075
Biomacromolecules Article
(Figure 1C) reveals a linear pseudo-first-order kinetic plot molar ratio or the polymerization time at a fixed monomer/
along with a narrow polydispersity profile of the produced RAFT agent molar ratio result in larger molecular weight
polymers. As a result, both narrow molecular weight polymers, proving that polymerization of (t-BMA-co-CMA)
distributions and linear increase of molecular weights with could be controlled via the RAFT mechanism.
monomer conversion indicate that the polymerization of P(t- Purified polymers were analyzed by 1H NMR using CDCl3 as
BMA-co-CMA) performed at [t-BMA]/[CMA]/[RAFT]/ the solvent (Figure S7A, Supporting Information). Signals at
[AIBN] ratio of 200.0/4.0/1.0/0.2 (4.0M/0.08M/0.02M/4 5.3 and 4.5 ppm, indicative of one olefinic proton in cholesterol
mM) was a well-controlled polymerization consistent with the moiety and methine, -COO-CH-, the main link between
known traits of the RAFT mechanism. cholesterol moiety and ester group, were observed in equal
Table 1, which contains polymerization conditions for P(t- intensities, suggesting that the cholesterol units stay intact
BMA-co-CMA) at varying monomer/RAFT agent ratios, also during polymerization process.
confirms that the synthesized polymers hold RAFT-controlled To obtain polymers with varying cholesterol contents,
character. In summary, increasing the monomer/RAFT agent polymerizations were performed with increasing CMA content
3068 dx.doi.org/10.1021/bm300846e | Biomacromolecules 2012, 13, 3064−3075
Biomacromolecules Article
in the feed (Table 2). The composition of the copolymers Conversion of tert-butyl ester to carboxylic acids was
could be calculated from the integrations of relevant proton achieved by treating P(t-BMA-co-CMA) with TFA at room
temperature for 32 h (Scheme 1, step 2). TFA treatment was
Table 2. Copolymers of t-BMA and CMA Prepared at chosen due to its selectivity in cleaving tert-butyl ester groups
Varying Compositions without affecting other types of esters.63 Initially the reaction
was conducted on P(t-BMA) and the specific deprotection of
CMA in the CMA in the
CMA in the copolymer copolymer tert-butyl groups was monitored by 1H NMR at different time
feed (mol (mol % of (wt % of intervals (Scheme S3 and Figures S4 and S5, Supporting
% of total resultant resultant conversionb Mnc Information). Indication of hydrolysis being complete was
monomer)a copolymer) copolymer) (%) (g/mol) PDId
determined by the disappearance of the tert-butyl protons peak
2 2 6 73 16500 1.19
at 1.41 ppm along with the formation of an acidic proton at
4 4 12 76 15800 1.10
12.04 ppm.64 The signals intensity decreased proportional to
8 8 23 69 18000 1.11
time; after 16 h, the tert-butyl group had nearly disappeared,
10 10 28 64 17900 1.14
although the reaction was continued to remove side products
12 12 32 60 18300 1.08
such as CF3C(CH3)3 (1.01 ppm) and OHC(CH3)3 (1.19 ppm,
13 13 34 62 18600 1.27
4.64 ppm). Extended reaction time resulted in a clearer
15 16 40 67 21700 1.23
a
hydrolysis with minimum side products (Figure S4, Supporting
[Total monomer]/[RAFT]/[AIBN] molar ratio in the feed = 204.0/ Information).65 In a control experiment, the effects of acid
1.0/0.2. Total monomer concentration in the feed, 4.08 M; hydrolysis on PCMA were investigated. PCMA was treated
polymerization temperature, 68 °C; solvent, toluene. bMonomer
conversion determined by 1H NMR. cThe number of average with TFA for 32 h (Figure S6, Supporting Information).
molecular weight determined by THF GPC analysis using PS Results indicated that acid hydrolysis conditions applied for the
standards. dPolydispersity index. cleavage of tert-butyl ester had no effect on P(CMA) ensuring
this treatment to be applicable to P(t-BMA-co-CMA).
Subsequently, acid hydrolysis was performed on P(t-BMA-co-
signals in the 1H NMR spectra. The total integration between CMA) and monitored by 1H NMR in DMSO revealing
2.4 and 0.6 ppm illustrates overlapping t-BMA, 14H, and successful deprotection of tert-butyl groups (Figure S7,
cholesterol moiety, 48H peaks. The mole ratio was calculated Supporting Information). The dramatic change in solubility
by the following equation: CMA mol % = [∫ a/(((∫ v − (48 × of the polymers (becoming soluble in aqueous solution) upon
∫ a))/14) + ∫ a)] × 100; t-BMA mol % = [((∫ v − (48 × acid treatment was also an indicator that the ester bonds had
∫ a))/14)/(((∫ v − (48 × ∫ a))/14) + ∫ a)] × 100 (Figure been cleaved. Subsequent to acid hydrolysis, copolymers having
S7A, Supporting Information). According to the NMR analysis 2, 4, and 8 mol % CMA became water-soluble. With the
(Table 2), the copolymer composition (between 60 and 76% increase in cholesterol content (more than 8 mol %), the
monomer conversions) was the same as the comonomer feed, deprotected copolymers, P(MAA-co-CMA), seemed to remain
suggesting a statistical placement of the two monomers along water insoluble. The water-soluble copolymers (having 2, 4,
the copolymer chain and almost equal reactivity of the and 8 mol % CMA) were used for further analysis.
monomers toward both propagating species. In this study, The thermal properties of the copolymers were briefly
statistical copolymers instead of block copolymers were examined via differential scanning calorimetry (DSC). Single
intended to be synthesized to minimize phase segregation of glass transition temperatures (Tg) were observed for homopol-
cholesterol units in aqueous solution (avoid from micellization) ymers and copolymers in the range of 82−228 °C (Figure S8,
and maximize their interaction with lipid membranes. Supporting Information). Poly(cholesteryl methacrylate)
Figure 2. DLS results of the copolymers (0.125 mM) having 2, 4, and 8 mol % CMA in buffer solutions at varying pHs (pH 4.5, 5.5, 6.5, 7.4) at 25
and 37 °C. Anova statistical error values point out the results to be within *P < 0.0005 and **P < 0.005 accuracy.
Figure 5. Surface plasmon resonance (SPR) real-time monitoring of binding of the copolymers with 2, 4, or 8 mol % CMA onto cell plasma
membrane-mimicking lipid bilayer (cholesterol/phosphatidylcholine molar ratio of 0.42). Polymers (0.5 mM) in buffer solutions at pH 5.0 or 7.4
were injected over the lipid bilayer surface at a flow rate of 20 μL/min for 6.5 min. After reaching saturation, surfaces were rinsed (shown by arrow)
with related buffer solutions for the same period of time.
Figure 6. SPR real-time monitoring the binding of the 2 mol % CMA copolymer onto lipid bilayer mimicking the plasma (PM) or endosome
membrane (EM) in acidic or neutral conditions. PM consisted of cholesterol/phosphatidylcholine constituents at a cholesterol/phosphatidylcholine
molar ratio of 0.42, and EM consisted of cholesterol and phosphatidylcholine at a mole ratio of 0.80. The related polymer solution (0.5 mM) was
injected over the lipid bilayer surface at a flow rate of 20 μL/min for 6.5 min; upon reaching saturation, it was then rinsed with related buffer solution
(indicated by the arrow).
In contrast, the copolymer with 2 mol % CMA appeared to surface coverage.79 Figure 5 displays the interaction of the
exist in extended chain conformation because of its higher copolymer containing 2 mol % CMA with lipid bilayer surface
hydrophilicity compared to the copolymer having 8 mol % at both pHs to be significantly higher than the interactions of 4
CMA. This result was in good accord with the results obtained and 8 mol % CMA containing copolymers. As supported by
from DLS and turbidity experiments. AFM analysis, the higher hydrophobicity of the copolymers
Surface Plasmon Resonance. The interactions between with 4 and 8 mol % CMA compared to 2 mol % CMA
polymer molecules and lipid bilayer immobilized surfaces were copolymer might be causing the self-organization of the
investigated by SPR. Lipid bilayer comprised of varying copolymer chains in aqueous solution to minimize the exposure
cholesterol/phosphatidylcholine constituents were generated of hydrophobic cholesterol units to water, which in total
as cellular membrane models for the plasma and endosome reduces the interactions of the cholesterol units along the
membranes.54 Polymers at 0.5 mM concentration, composed of chains with the lipid layer and, thus, demonstrating lower
varying cholesterol units (2, 4, and 8 mol % CMA), were binding profile. Furthermore, in acidic medium, the copolymer
interacted first with the plasma membrane-mimicking surfaces having 2 mol % CMA exhibits almost five times greater binding
in slightly acidic and neutral pH. Surfaces were treated with when compared with the binding signal intensity of the same
polymer solutions of 0.5 mM concentration to ensure copolymer at neutral pH. The increase in surface binding at
interaction of the polymer with lipid layer and obtain complete lower pH can be explained by the relatively higher hydro-
3071 dx.doi.org/10.1021/bm300846e | Biomacromolecules 2012, 13, 3064−3075
Biomacromolecules Article
phobicity (exposed cholesterol moieties) along with the lower membrane-lytic effect of the copolymer containing 2 mol %
ionic charge (protonated MAA units) of the copolymer, CMA, was found to be the strongest (Figure 7). This result, in
possibly increasing the interaction of the copolymer chains
with phosphatidylcholine/cholesterol lipid bilayer.74,80
Upon determining the copolymer composition displaying the
highest binding to the plasma membrane-mimicking surface,
the polymer−lipid interaction was further investigated in
relation to lipid bilayer composition (Figure 6). The interaction
of the 2 mol % CMA copolymer with plasma or endosome
membrane-mimicking lipid bilayer arrangements having differ-
ent cholesterol contents (molar ratio of 0.42 and 0.80,
respectively) were compared in a pH-dependent manner
using SPR. The binding of the copolymer with both lipid
bilayers at acidic pH was higher when compared with the
binding at neutral pH, as expected. Interestingly, the binding at
acidic condition with the plasma membrane-mimicking bilayer
was higher than the binding with endosome membrane-
mimicking bilayer which had almost 2-fold higher cholesterol
content. This was attributed to the possible interactions formed
among protonated MAA units of the copolymer and
phophatidylcholine components in the lipid bilayer surfaces.
Intermolecular hydrogen bonding created at acidic pH between
phosphatidylcholine units (higher content in plasma mem-
brane-mimicking bilayer) and copolymer chains is likely to
dominate intra- and intermolecular hydrogen bonds between
MAA units along copolymer chains, consequently leading to
greater interactions with the lipid bilayer.81,82 It is also possible
that the higher rigidity of the endosome membrane-mimicking
bilayer because of the higher cholesterol content decreases the
interaction of the bilayer with the polymer chains. Consistent
with these results, at pH 7.4 there was almost no binding with Figure 7. Results of liposome lysis assay. The lysis obtained with the
both membrane models, possibly because of the electrostatic copolymers containing 2, 4, and 8 mol % CMA are shown in the
repulsion between negatively charged copolymers and anionic graph, PMAA represents lysis gained by poly(methacrylic acid) only
lipid bilayer. (with no cholesterol units), and 10% Triton X-100 (T) signifies
Liposome Lysis Assay. The extent of pH-dependent maximum lysis. Polymer concentrations are (A) 1.5 or (B) 3.0 μM.
hydrophobic association of the copolymers has been identified The results are the average of two different sets of experiments
performed in triplicate. Error bars represent standard deviation. Anova
as an important factor in determining the degree of membrane statistical analysis results reveal the values to be within *P < 0.05 and
destabilization activity which is required for intracellular drug **P < 0.006 accuracy.
delivery vehicles.83 Amphiphilic copolymers containing 2, 4,
and 8 mol % CMA, were analyzed for their pH-dependent
membrane-lytic properties using a liposome leakage assay, as accord with the SPR data reveals that the copolymer having 2
described in the literature.84 The assay was first performed by mol % CMA exhibits the greatest polymer−lipid interactions
treating the SRB-loaded liposomes with copolymer solutions at and displays the highest membrane destabilization activity at
0.125 mM concentrations at pH 5.0 and 7.4. Complete release pH 5.0. The lower membrane-lytic activity of the copolymers
of SRB from liposomes was observed with no pH-dependent with higher CMA content can be attributed to the self-
profile (data not shown). This suggests that the copolymers at organization of these copolymers in solution, minimizing the
0.125 mM cause nonspecific (pH-independent) destabilization hydrophobic interactions between copolymer CMA units and
of liposomes causing the leakage of SRB at both pH values liposomes. Amphiphilic polymers have been shown to interact
tested (pH 5.0 and 7.4). Here it should be noted that the with lipid vesicles by inserting their hydrophobic segments into
copolymers at a higher concentration (0.5 mM) show pH- the lipid layer as a result of the hydrophobic interactions.85
dependent interaction with the lipid layer anchored onto a The liposome lysis by the copolymers was observed to be
surface (as SPR results indicated). The discrepancy between pH-dependent.86,87 The membrane-lytic activity at pH 5.0 was
the results obtained from SPR and liposome leakage assay 3-fold (in the case of 2 mol % CMA), 2-fold (4 mol % CMA),
might be due to the fact that the lipid assembly on the surface is and 1.4-fold higher (8 mol % CMA) than the lysis at pH 7.4.
well-packed and, thus, less flexible than the lipid layer of The copolymers displayed an average of 34% SRB leakage at
liposomes, which minimizes the interactions with the neutral pH. Similar residual lytic activities at neutral pH have
copolymer chains at ionized (hydrophilic) state (at pH 7.4) also been reported previously for various pH-dependent
even at high concentrations such as 0.5 mM. membrane disruptive polymers.21,70,88 As the pH of the
The liposome leakage assay was repeated using copolymer solution is increased, the copolymer chains become negatively
solutions at much lower concentrations (1.5 and 3.0 μM) to charged and more hydrophilic because of the ionization of the
avoid from nonspecific (pH-independent) destabilization of carboxylic acid groups of MAA units.78 This decreases the
liposomes. Based on the release profiles of SRB from liposomes interactions between phophatidylcholine liposomes and the
after incubation with the copolymers, the pH-dependent copolymer chains. Hydrogen bonds that may form at acidic
3072 dx.doi.org/10.1021/bm300846e | Biomacromolecules 2012, 13, 3064−3075
Biomacromolecules
■
Article
■
of the cells with different concentrations of polymers, ranging
from 0.06 to 30 μM (per 3 × 103 cells), are shown in Figure 8. ASSOCIATED CONTENT
Cell viability (%) is expressed as a function against the
untreated cells (control for 100% viability) and 10% Triton X- * Supporting Information
S
100 treatment (control for 0% viability). Experimental methods for synthesis of cholesteryl methacrylate
(CMA) monomer, homopolymerizations of CMA and t-BMA,
and acid hydrolysis, differential scanning calorimetry traces, and
representative size distribution plots of copolymers, Schemes
S1−S3, Figures S1−S9, and Table S1. This material is available
free of charge via the Internet at http://pubs.acs.org.
■ AUTHOR INFORMATION
Corresponding Author
*Tel.: +90-232-750 6660. E-mail: volgabulmus@iyte.edu.tr.
Notes
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
The authors acknowledge The University of New South Wales
Gold Star Award and Dr. Ling-Jiun Wong for her help on
preliminary literature search. V.B. also acknowledges The
Figure 8. Viability of human neuroblastoma SH-EP cells after
incubation with P(MAA-co-CMA) copolymers that consist of 2, 4,
Scientific and Technological Research Council of Turkey
(TUBITAK) for research Grant No. 111T960.
■
and 8 mol % CMA along with P(MAA) for 72 h, measured with
CellTiter-Blue assay. The assay was repeated three times in triplicate
and the viability results were normalized according to the positive REFERENCES
control (untreated cells). Error bars represent standard deviation. (1) Jones, A. T.; Gumbleton, M.; Duncan, R. Adv. Drug Delivery Rev.
Anova statistical error values point out the results to be within *P < 2003, 55 (11), 1353−1357.
0.06 and **P < 0.001 accuracy. (2) Zhao, F.; Zhao, Y.; Liu, Y.; Chang, X.; Chen, C.; Zhao, Y. Small
2011, 7 (10), 1322−1337.
(3) Pack, D. W.; Hoffman, A. S.; Pun, S.; Stayton, P. S. Nat. Rev. Drug
Discovery 2005, 4 (7), 581−593.
Statistical copolymers composed of 2, 4, and 8 mol % CMA (4) Troiber, C.; Wagner, E. Bioconjugate Chem. 2011, 22 (9), 1737−
along with P(MAA) at the concentrations tested displayed 1752.
(5) Varkouhi, A. K.; Scholte, M.; Storm, G.; Haisma, H. J. J.
similar behavior on the cells. The viability of the cells for 30 μM Controlled Release 2011, 151 (3), 220−228.
to 0.06 μM treatments varied between 86 and 100%, indicating (6) Mukherjee, S.; Ghosh, R. N.; Maxfield, F. R. Physiol. Rev. 1997,
no significant toxicity of the copolymers. 77 (3), 759−803.
(7) Huang, Z. H.; Li, W. J.; MacKay, J. A.; Szoka, F. C. Mol. Ther. (37) Gu, Z. X.; Yuan, Y.; He, J. L.; Zhang, M. Z.; Ni, P. H. Langmuir
2005, 11 (3), 409−417. 2009, 25 (9), 5199−5208.
(8) Wolfrum, C.; Shi, S.; Jayaprakash, K. N.; Jayaraman, M.; Wang, (38) Zhou, Y.; Sharma, N.; Deshmukh, P.; Lakhman, R. K.; Jain, M.;
G.; Pandey, R. K.; Rajeev, K. G.; Nakayama, T.; Charrise, K.; Ndungo, Kasi, R. M. J. Am. Chem. Soc. 2011, 134 (3), 1630−1641.
E. M.; Zimmermann, T.; Koteliansky, V.; Manoharan, M.; Stoffel, M. (39) Bajaj, A.; Kondaiah, P.; Bhattacharya, S. Bioconjugate Chem.
Nat. Biotechnol. 2007, 25 (10), 1149−1157. 2008, 19, 1640−1651.
(9) Soutschek, J.; Akinc, A.; Bramlage, B.; Charisse, K.; Constien, R.; (40) Kim, W. J.; Chang, C. W.; Lee, M.; Kim, S. W. J. Controlled
Donoghue, M.; Elbashir, S.; Geick, A.; Hadwiger, P.; Harborth, J.; Release 2007, 118 (3), 357−363.
John, M.; Kesavan, V.; Lavine, G.; Pandey, R. K.; Racie, T.; Rajeev, K. (41) Incani, V.; Lavasanifar, A.; Uludag, H. Soft Matter 2010, 6 (10),
G.; Rohl, I.; Toudjarska, I.; Wang, G.; Wuschko, S.; Bumcrot, D.; 2124−2138.
Koteliansky, V.; Limmer, S.; Manoharan, M.; Vornlocher, H.-P. Nature (42) Jones, S. P.; Gabrielson, N. P.; Pack, D. W.; Smith, D. K. Chem.
2004, 432 (7014), 173−178. Commun. (Cambridge, U.K.) 2008, 39, 4700−4702.
(10) Meyer, M.; Philipp, A.; Oskuee, R.; Schmidt, C.; Wagner, E. J. (43) Ikonen, E. Nat. Rev. Mol. Cell Biol. 2008, 9 (2), 125−138.
Am. Chem. Soc. 2008, 130 (11), 3272−3273. (44) Boyer, C.; Bulmus, V.; Davis, T. P.; Ladmiral, V.; Liu, J. Q.;
(11) Torchilin, V. P. Biopolymers 2008, 90 (5), 604−610. Perrier, S. Chem. Rev. 2009, 109 (11), 5402−5436.
(12) Stewart, K. M.; Horton, K. L.; Kelley, S. O. Org. Biomol. Chem. (45) Smith, D.; Holley, A. C.; McCormick, C. L. Polym. Chem. 2011,
2008, 6 (13), 2242−2255. 2 (7), 1428−1441.
(13) Li, W. J.; Nicol, F.; Szoka, F. C. Adv. Drug Delivery Rev. 2004, 56 (46) Xu, X.; Flores, J. D.; McCormick, C. L. Macromolecules 2011, 44
(7), 967−985. (6), 1327−1334.
(14) Bulmus, V.; Woodward, M.; Lin, L.; Murthy, N.; Stayton, P.; (47) Huang, X.; Boyer, C.; Davis, T. P.; Bulmus, V. Polym. Chem.
Hoffman, A. J. Controlled Release 2003, 93 (2), 105−120. 2011, 2 (7), 1505−1512.
(15) Hoffman, A. S.; Stayton, P. S.; Ei-Sayed, M. E. H.; Murthy, N.; (48) Bulmus, V. Polym. Chem. 2011, 2 (7), 1463−1472.
Bulmus, V.; Lackey, C.; Cheung, C. J. Biomed. Nanotechnol. 2007, 3, (49) Sevimli, S.; Sagnella, S.; Kavallaris, M.; Bulmus, V.; Davis, T. P.
213−217. Polym. Chem. 2012, 3 (8), 2057−2069.
(16) Murthy, N.; Robichaud, J. R.; Tirrell, D. A.; Stayton, P. S.; (50) Zhou, Y. X.; Kasi, R. M. J. Polym. Sci., Part A: Polym. Chem.
Hoffman, A. S. J. Controlled Release 1999, 61 (1−2), 137−143. 2008, 46 (20), 6801−6809.
(17) Grube, S.; Wolfrurn, U.; Langguth, P. Biomacromolecules 2008, 9 (51) Mitsukami, Y.; Donovan, M. S.; Lowe, A. B.; McCormick, C. L.
(5), 1398−1405. Macromolecules 2001, 34 (7), 2248−2256.
(18) Yessine, M. A.; Dufresne, M. H.; Meier, C.; Petereit, H. U.; (52) Thang, S. H.; Chong, Y. K.; Mayadunne, R. T. A.; Moad, G.;
Leroux, J. C. Bioconjugate Chem. 2007, 18 (3), 1010−1014. Rizzardo, E. Tetrahedron Lett. 1999, 40 (12), 2435−2438.
(19) Funhoff, A. M.; Van Nostrum, C. F.; Lok, M. C.; Kruijtzer, J. A. (53) Chen, G. H.; Hoffman, A. S. Macromol. Chem. Phys. 1995, 196
W.; Crommelin, D. J. A.; Hennink, W. E. J. Controlled Release 2005, (4), 1251−1259.
101 (1−3), 233−246. (54) Evans, W. H.; Hardison, W. G. M. Biochem. J. 1985, 232 (1),
(20) Convertine, A. J.; Benoit, D. S. W.; Duvall, C. L.; Hoffman, A. 33−36.
S.; Stayton, P. S. J. Controlled Release 2009, 133 (3), 221−229. (55) Morrissey, J. H. Morrissey Laboratory Protocol for Preparing
(21) Henry, S. M.; El-Sayed, M. E. H.; Pirie, C. M.; Hoffman, A. S.; Phospholipid Vesicles (SUV) by Sonication; Department of Biochemis-
Stayton, P. S. Biomacromolecules 2006, 7 (8), 2407−2414. try, University of Illinois: Urbana-Champaign, Urbana, IL 61801,
(22) Foster, S.; Duvall, C. L.; Crownover, E. F.; Hoffman, A. S.; 2001; http://tf7.org/suv.pdf.
Stayton, P. S. Bioconjugate Chem. 2010, 21 (12), 2205−2212. (56) Wong, L. J.; Boyer, C.; Jia, Z. F.; Zareie, H. M.; Davis, T. P.;
(23) Convertine, A. J.; Diab, C.; Prieve, M.; Paschal, A.; Hoffman, A. Bulmus, V. Biomacromolecules 2008, 9 (7), 1934−1944.
S.; Johnson, P. H.; Stayton, P. S. Biomacromolecules 2010, 11 (11), (57) Kilian, L.; Wang, Z. H.; Long, T. E. J. Polym. Sci., Part A: Polym.
2904−2911. Chem. 2003, 41 (19), 3083−3093.
(24) Duvall, C. L.; Convertine, A. J.; Benoit, D. S. W.; Hoffman, A. (58) Deporter, C. D.; Long, T. E.; McGrath, J. E. Polym. Int. 1994, 33
S.; Stayton, P. S. Mol. Pharm. 2010, 7 (2), 468−476. (2), 205−216.
(25) Benoit, D. S. W.; Srinivasan, S.; Shubin, A. D.; Stayton, P. S. (59) Mengel, C.; Meyer, W. H.; Wegner, G. Macromol. Chem. Phys.
Biomacromolecules 2011, 12 (7), 2708−2714. 2001, 202 (7), 1138−1149.
(26) Benoit, D. S. W.; Henry, S. M.; Shubin, A. D.; Hoffman, A. S.; (60) Lee, S. C.; Kim, K. J.; Jeong, Y. K.; Chang, J. H.; Choi, J.
Stayton, P. S. Mol. Pharm. 2010, 7 (2), 442−455. Macromolecules 2005, 38 (22), 9291−9297.
(27) Murthy, N.; Campbell, J.; Fausto, N.; Hoffman, A. S.; Stayton, (61) Nystrom, A. M.; Wooley, K. L. Tetrahedron 2008, 64 (36),
P. S. J. Controlled Release 2003, 89 (3), 365−374. 8543−8552.
(28) Cheung, C. Y.; Murthy, N.; Stayton, P. S.; Hoffman, A. S. (62) El-Sayed, M. E. H.; Hoffman, A. S.; Stayton, P. S. Expert Opin.
Bioconjugate Chem. 2001, 12 (6), 906−910. Biol. Ther. 2005, 5 (1), 23−32.
(29) Lackey, C. A.; Press, O. W.; Hoffman, A. S.; Stayton, P. S. (63) Jiang, X.; Zhao, B. Macromolecules 2008, 41 (23), 9366−9375.
Bioconjugate Chem. 2002, 13 (5), 996−1001. (64) Choi, W. J.; Kim, Y. B.; Kwon, S. K.; Lim, K. T.; Choi, S. K. J.
(30) Kyriakides, T. R.; Cheung, C. Y.; Murthy, N.; Bornstein, P.; Polym. Sci., Part A: Polym. Chem. 1992, 30 (10), 2143−2148.
Stayton, P. S.; Hoffman, A. S. J. Controlled Release 2002, 78 (1−3), (65) Ma, Q. G.; Wooley, K. L. J. Polym. Sci., Part A: Polym. Chem.
295−303. 2000, 38, 4805−4820.
(31) Sant, V. P.; Smith, D.; Leroux, J.-C. J. Controlled Release 2004, 97 (66) Lai, J. H. Macromolecules 1984, 17 (5), 1010−1012.
(2), 301−312. (67) Odian, G. G. Principles of Polymerization, 4th ed.; John Wiley &
(32) Yang, Y. Q.; Zheng, L. S.; Guo, X. D.; Qian, Y.; Zhang, L. J. Sons, Inc.: New York, 2004; Vol. 1, pp 140−141.
Biomacromolecules 2010, 12 (1), 116−122. (68) Ruiz-Perez, L.; Pryke, A.; Sommer, M.; Battaglia, G.; Soutar, I.;
(33) Schüwer, N.; Klok, H.-A. Langmuir 2011, 27 (8), 4789−4796. Swanson, L.; Geoghegan, M. Macromolecules 2008, 41 (6), 2203−
(34) Alarcon, C. d. l. H.; Pennadam, S.; Alexander, C. Chem. Soc. Rev. 2211.
2005, 34 (3), 276−285. (69) Heitz, C.; Rawiso, M.; François, J. Polymer 1999, 40 (7), 1637−
(35) Zhou, Y.; Briand, V.; Sharma, N.; Ahn, S.-k.; Kasi, R. Materials 1650.
2009, 2 (2), 636−660. (70) Kusonwiriyawong, C.; van de Wetering, P.; Hubbell, J. A.;
(36) Wen, J.; Mao, H.-Q.; Li, W.; Lin, K. Y.; Leong, K. W. J. Pharm. Merkle, H. P.; Walter, E. Eur. J. Pharm. Biopharm. 2003, 56 (2), 237−
Sci. 2004, 93 (8), 2142−2157. 246.