Poly(2-methyl-2-oxazoline)-b-poly(tetrahydrofuran ... - ACS Publications

This material is available free of charge via the Internet at http://pubs.acs.org. ...... Jean-Pierre Gomez , William Même , Bazoly Rasolonjatovo , D...
0 downloads 0 Views 1MB Size
Subscriber access provided by Stony Brook University | University Libraries

Article

Poly (2-methyl-2-oxazoline)-b-poly(tetrahydrofuran)-b-poly(2methyl-2-oxazoline) amphiphilic triblock copolymers: synthesis, physicochemical characterizations and hydrosolubilizing properties Bazoly Rasolonjatovo, Jean Pierre Gomez, William meme, Cristine Goncalves, Cecile HUIN, Veronique Bennevault-Celton, Tony Le Gall, Tristan Montier, Pierre Lehn, herve cheradame, Patrick Midoux, and Philippe Guegan Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/bm5016656 • Publication Date (Web): 17 Dec 2014 Downloaded from http://pubs.acs.org on December 18, 2014

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biomacromolecules is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Poly (2-methyl-2-oxazoline)-b-poly(tetrahydrofuran)-b-poly(2-methyl2-oxazoline)

amphiphilic

triblock

copolymers:

synthesis,

physicochemical characterizations and hydrosolubilizing properties

Bazoly Rasolonjatovo, Jean-Pierre Gomez, William Même, Cristine Gonçalves, Cécile Huin, Véronique Bennevault-Celton, Tony Le Gall, Tristan Montier, Pierre Lehn, Hervé Cheradame, Patrick Midoux* and Philippe Guégan* *Corresponding authors: [email protected] and [email protected]

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 36

For Table of Content Graphic Use Only

+ +

Poly (2-methyl-2-oxazoline)-b-poly(tetrahydrofuran)-b-poly(2-methyl2-oxazoline)

amphiphilic

triblock

copolymers:

synthesis,

physicochemical characterizations and hydrosolubilizing properties

Bazoly Rasolonjatovo, Jean-Pierre Gomez, William Même, Cristine Gonçalves, Cécile Huin, Véronique Bennevault-Celton, Tony Le Gall, Tristan Montier, Pierre Lehn, Hervé Cheradame, Patrick Midoux* and Philippe Guégan*

ACS Paragon Plus Environment

Page 3 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Abstract Block copolymers assembled into micelles drain a large attention to improve drug delivery. The recent drawbacks of the poly(ethylene oxide) blocks (PEO) contained in amphiphilic pluronics derivatives made of a central poly(propylene oxide) block surrounded by two PEO blocks were recently revealed, opening the way to the design of new amphiphilic block copolymers able of self assembly in water, and to entrap molecules of interest. Here, a family of p(methyloxazoline)-b-p(tetrahydrofurane)-bp(methyloxazoline) triblock copolymers (called TBCP) is synthesized using cationic ring opening polymerization. Studies of micelles formation by using Dynamic Light Scattering, ITC, NMR DOSY and Fluorescence experiments, lead to draw a relationship between copolymer structure and the physicochemical properties of the block copolymers (CMC, Nagg, core diameter, shell thickness …). The packing parameter of the block copolymers indicates the formation of a core-corona structure. Hydrosolubilizing properties of TBCPs exemplified with curcumin selected as a highly insoluble drug model. Curcumin, a natural polyphenolic compound, has shown a large spectrum of biological and pharmacological activity, including anti-inflammatory, antimicrobial, anti-oxidant and anti-carcinogenic. An optimized formulation process reveals that the aggregation number is the parameter affecting the drug encapsulation. Patch clamp experiments carried out to study interaction of TBCP with cell membrane demonstrate their permeation property suitable to promote the cellular internalization of curcumin.

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Introduction Chemotherapeutic molecules are generally poorly soluble in the delivery aqueous medium, limiting their clinical application. Polymer micelles offer real opportunities to solubilize hydrophobic drugs.1-4 For this purpose, several commercial ABA copolymers have received attentions.1,2 For instance, pluronic polymers have been studied quite extensively thanks to the formation of core-shell micelles comprising the poly(ethylene oxide) (PEO) block as the hydrophilic shell of the corona and the poly(propylene oxide) (PPO) block as the core. Some triblock pluronic copolymers have shown capacity to pass the blood-brain barrier.5 Some poloxamers have even demonstrated their ability to increase the biodistribution of molecules poorly soluble in water such as genestein after oral delivery.3 In preclinical studies, doxorubicin bound to pluronic polymer has demonstrated increased efficacy compared to free doxorubicin.6 In the vaccine field, neutral amphiphilic copolymers mixed with antigenic proteins have increased significantly the humoral and cellular responses after intravenous injection.7 Recent works demonstrated that micelle-based vectors could accumulate in solid tumors, through the enhanced permeability and retention (EPR) effect, and play an important role in reducing the uptake by the reticuloendothelial system. Polymeric micelles are formed by a spontaneous self-assembly of the block copolymers in water thanks to the high water solubility difference of the individual blocks. The micellization process favors solubilization of the hydrophobic drugs present in the solution. The resulting nano-objects exhibit specific characteristics including their form (spherical, rod-like …), their hydrodynamic parameter, and their aggregation number, i.e; the number of individual block copolymer molecules forming the nano-objects that can strongly affect the behaviors of these nano-objects. Recent formulations for drug delivery focused on the use of poly(ethylene glycol) (PEG) because of its remarkable stealth properties, not shared by most of the neutral hydrophilic polymers.8 Furthermore, polymerization of ethylene oxide allows for the

ACS Paragon Plus Environment

Page 4 of 36

Page 5 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

synthesis of polymers with Đ lower than 1.1 with precise and controlled molar masses,9 important parameters for drug delivery applications. Despite the large advantage of PEO as shielding counterparts from protein interactions, it was recently reported various unfavourable effects due either to the polymer itself or by side product formed during its synthesis.10 Various polymers are discussed as alternative to the PEO. Recently, the poly(2-ethyl-2-oxazoline) was reported to preserve excellent in vitro cell viability and good hemocompatibility, opening up the way to the use of the polyoxazoline family as an alternative to PEO.11 The use of poly(oxazoline)-based nano-objects for biologic application was suggested. 12 In this work, we have synthesized a series of poly(2-methyl-2-)-b-polytetrahydrofuran-

b-poly(2-methyl-2-oxazoline) amphiphilic triblock copolymers (called TBCP) using the poly(2-methyl-2-oxazoline) block. Moreover, the PPO block of pluronic and poloxamer triblock copolymers was replaced by a pTHF block in order to decrease the hydrophilichydrophobic balance (HLB) of the TBCPs, and to favor later on drug loading in the nanocarriers. Micelle formation in aqueous solution was investigated using four different techniques. We evaluated their capacity to interact with cell membranes and to solubilize curcumin, a very hydrophobic molecule that exhibits a large panel of biological activities. The correlation between the TBCP physicochemical parameters, their propensity to form micelles and their capacity to solubilize curcumin is reported.

Materials and methods All reagents were purchased from Sigma (St. Quentin Fallavier, France) unless otherwise stated. Tetrahydrofuran (THF) was purified by distillation from sodium, under reflux, in the presence of benzophenone, until a persistent blue color was obtained. 2-methyl-2-oxazoline (MeOx) was purified by refluxing over calcium

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

hydride,

under

nitrogen,

and

distilled

prior

use.

Page 6 of 36

Acetonitrile

(ACN)

and

dichloromethane (CH2Cl2) were dried by refluxing over CaH2 under N2 and subsequently distilled prior use. Deionized water was obtained with a Millipore Milli-Q system.

Synthesis of poly(2-methyl-2-oxazoline)-block-poly(tetrahydrofuran)-block-poly(2methyl-2-oxazoline) triblock copolymers(pMeOx-b-pTHF-b-pMeOx). The synthesis was performed as previously described.13-14

Briefly, to a 250 mL reaction flask

containing 129 mmol of dry THF, 2.109 mmol of trifluoromethanesulfonic anhydride (Tf2O) were added at -9°C. The reaction mixture was stirred during 15 minutes and the polymerization was quenched by adding 8.37 mmol of MeOx at -9°C. Evaporation of residual THF under reduced pressure yielded to a pTHF prepolymer as a white solid. This solid was dissolved in 40 mL of dry ACN and the temperature was increased to 80°C. 54 mmol of MeOx were added and the solution was stirred for 24 hours. The reaction was quenched by adding 4mL of 2M sodium carbonate solution (Na2CO3) and stirred for another 1 hour at room temperature. The triblock copolymer was obtained by chloroform extraction, evaporation of the organic phase and drying for 2 days under vacuum. The molar masses of the blocks were varied by adjusting the polymerization times.

Synthesis of homopolymer poly(2-methyl-2-oxazoline).15 To a solution of 8 mL acetonitrile and 409.7 mg of methyl p-toluenesulfonate (2.2mmol), 2g of 2-methyl-2-oxazoline (23.5 mmol) was added under inert atmosphere. Polymerization was carried out at 80°C for 24 hours. Two mL of Na2CO3 2M were used to end the reaction at reflux at 100°C and let stirred overnight. The reaction medium was precipitated into ethyl ether, filtrated and dried under vacuum. 1.82 g of polymer was obtained. The molar mass, determined by 1H NMR analysis in CDCl3 and SEC at room temperature, was 870 g.mol-1.

ACS Paragon Plus Environment

Page 7 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Structural characterization of copolymers Molar masses were determined by 1H NMR analysis at 300 MHz in CDCl3 at room temperature on a Bruker Avance AM300 spectrometer. Molar masses and Đ were determined by size exclusion chromatography (SEC) at 60°C with DMF as eluent and a copolymer concentration of 5 mg.mL-1. The columns were a PSS GRAM 103Ǻ and a PSS GRAM 30 Ǻ. SEC was calibrated using PMMA standards. DOSY NMR measurements were performed at 298 K with a 300 MHz Bruker Avance AM300 spectrometer equipped with a Bruker multinuclear z-gradients inverse probe head able to produce gradients in the z direction with strength 55 G.cm-1. DOSY spectra were acquired with the ledbpgp2s pulse program from Bruker topspin software. All spectra were recorded with 32 K time domain datapoints in the t2 dimension and 32 t1 increments. The gradient strength was logarithmically incremented in 32 steps from 2 % up to 95 % of the maximum gradient strength. All measurements were performed with a compromised diffusion delay (D) of 200 ms in order to keep the relaxation contribution to the signal attenuation constant for all samples. The gradient pulse length (d) was 5 ms in order to ensure full signal attenuation. The diffusion dimension of the 2D DOSY spectra was processed by means of the Bruker topspin software (version 2.1).

TBCP behaviors in solution - Micelle characterization. DOSY NMR. TBCPs were directly dissolved in D2O at different concentrations and the values for the diffusion coefficients (D) were collected at each concentration.

Fluorescence spectroscopy. The critical micellar concentration (CMC) was determined by fluorescence spectroscopy using Nile Red as a fluorescent probe. Increasing concentrations of TBCP were dissolved directly in deionized water. Then Nile Red was

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

added at 2.5 µmol.L-1 and the fluorescence was measured upon excitation at 530 nm with the Shimadzu RF-5000 spectrofluorimeter. The maximum emission intensity at 630 nm was plotted as a function of TBCP concentrations.

Isothermal titration calorimetry (ITC). CMC and thermodynamic parameters of TBCPs were determined by microcalorimetric measurements with MCS-ITC instrument (MicroCal Inc). Two hundred µL of a micellar solution of TBCP (7% w/v) were injected via a syringe assisted by a computer into a cell containing pure water (cell volume: 1.449 mL). For each injection, 10 µL of TBCP solution were delivered over 20 seconds with a 600 seconds interval between injections to allow complete equilibration. The sample cell was continuously stirred at 220 rpm throughout the experiments. When the micellar solution was injected into the cell, heat change was detected and a plot heat flow as a function of time was obtained. Once the titration was completed, the individual peaks were integrated by MicroCalORIGIN 7 software and the titration curve was obtained. CMC value corresponds to the extremum of the first derivative of the titration curve.

Dynamic and Static light scattering (DLS). TBCPs were directly dissolved in deionized water and solutions were filtered through MilliporeMillex filters (0.45 µm pore size). TBCP solutions were then kept overnight. Dynamic light scattering measurements were performed on a Zetasizer Nano Series ZS (Malvern) equipped with a laser He-Ne 4 mW operating at 633 nm. The data were acquired using a BI-9000AT digital correlator fitted with the instrument. Autocorrelation functions were measured at a 173° back-scattering angle and analyzed by the constrained regularized CONTIN method to obtain distributions of decay rates (Γ). The decay rates afford the determination of the distributions of apparent diffusion coefficients (D = Γ/q2) and the apparent hydrodynamic radii (Rh,app). The last parameter was obtained from the Stokes-

ACS Paragon Plus Environment

Page 8 of 36

Page 9 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Einstein equation: Rh =

kT 6πη D

where k is the Boltzmann constant and η the solvent

viscosity at the temperature T (K). The weight-average molar mass (Mw) of polymer micelles were obtained from the Debye plots determined by concentration dependence scattering intensity data (micelle concentration: 2, 5, 8, and 10 mg/mL), Mw was determined from the intercept of the plot: KC/R(q) = 1/Mw + 2A2C where K = 4π2(n.dn/dC)2/(λ04Nav) is the optical contrast, n the refractive index of solvent, λ0 the incident wavelength, Nav the Avogadro number, C the TBCP concentration and R(q) the Rayleigh ratio. For pMeOxz-b-pTHF-b-pMeOxz triblock copolymers, the dn/dC was 0.135 mL.g-1 in water at 25 °C. The density of pTHF at amorphous state was 9.15 105 g.m-3.16

Curcumin solubilization. TBCP stock solutions were prepared by adding 1 mL distilled water to 25 mg TBCP under vigorous stirring until complete dilution. Curcumin (Cur; Mw: 368 g.mL-1) was recrystallized in ethanol. Cur (0.1 mg) in 2 mL eppendorf tube was mixed with various quantities of TBCP stock solutions. The mixture was complete to 1 mL with distilled water, vortexed for 5 seconds and kept for 6 hours under magnetic stirring in the dark. The solution was centrifuged for 5 minutes at 14,000 rpm, the supernatant was passed through a Millex-HV 0.45µm PVDF membrane (Millipore) and the absorbance was measured at 425 nm. The pellet was solubilized in an ethanol/water mixture (5%: 95% v/v), vortexed and the absorbance was measured at 425 nm. The percentage of solubilized curcumin was expressed as absorbance of sample at 425 nm to absorbance of total curcumin solubilized in ethanol/water mixture (5%:95%; v/v).

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 36

Cells and cell culture. Human epithelial ovarian carcinoma HeLa (CCL-2), Human embryonic kidney HEK293 (CRL-1573) and adenocarcinomic human alveolar basal epithelial A549 (CCL-185) cells lines were from ATCC (Rockville MD, USA). HeLa cells

were

grown

in

MEM

(PAA

Laboratories),

0.4

%

Penicillin

(40

Units/ml)/Streptomycin (40 µg/ml) (PAA Laboratories), 2 mM L-Glutamine (PAA Laboratories), supplemented with 10 % non-heat-inactivated fetal bovine serum (PAA Laboratories). HEK293 and A549 cells were cultured in DMEM containing 10 % heatinactivated fetal bovine serum, 2 mM L-glutamine, 1 mM sodium pyruvate (PAA Laboratories), 100 units.mL-1 penicillin, 100 units/mL streptomycin. 16HBE14onormal human bronchial epithelial

17

and the ΣCFTE29o- CF human tracheal epithelial

(homozygous for the ∆F508 mutation)

18

cell lines, generous gifts from Dr. D.C.

Gruenert (San Francisco, CA, U.S.A.) were cultured at 37 °C in a 5 % CO2-humidified incubator in 20 mL MEM with Earle's Salts (PAA Laboratories), 0.4 % Penicillin (40 Units/ml)/Streptomycin (40 µg/ml) (PAA Laboratories), 2 mM L-Glutamine (PAA Laboratories), supplemented with 10 % non-heat-inactivated foetal bovine serum (PAA Laboratories). Tissue culture plastic wares (75 cm2) were coated for 20-30 minutes at 37 °C with MEM with Earle's Salts containing fibronectin (0.01 mg.mL-1), collagen (0.03 mg/mL) and bovine serum albumin (BSA) (0.1 mg.mL-1). The culture medium was changed every 2 days. On day 8, cells were harvested by treatment with trypsin at 37 °C for 10 minutes. The absence of mycoplasma in cell cultures was determined by using MycoAlert® Mycoplasma Detection Kit (Lonza, Levallois Perret, France).

Patch-clamp and electrophysiological measurements For patch-clamp recordings, cells were cultured in a glass coverslip that was transferred to the experimental chamber of an upright microscope (BX51WI, Olympus Corporation,

ACS Paragon Plus Environment

Page 11 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Tokyo, Japan). Experiments were performed at room temperature (21-23 °C). Cells were placed in continuously flowing (1-2 ml.min-1) bath solution containing (mM): 150 NaCl; 6 CsCl; 1 CaCl2; 1 MgCl2; 10 D-glucose; 10 HEPES (adjusted to pH 7.4 with Tris). Cells were identified at 60x magnification with a CCD camera (XC-ST70CE, Sony, Paris, France). Somatic whole-cell recordings were performed, as previously described. 19 Briefly, low resistance (4-6 MΩ) patch-pipettes pulled on a vertical puller (PB-7, Narishige, Tokyo, Japan) from borosilicate capillaries (Clark Electromedical Instruments, Edenbridge, UK) were filled with internal solution, containing (mM): 100 L-aspartic acid; 94 CsOH; 26 CsCl; 14 NaCl; 1 MgCl2; 3 MgATP; 1 EGTA; 10 HEPES (adjusted to pH 7.3 with Tris). Signals were amplified using the Multiclamp 700B amplifier (Axon Instruments, Foster City, CA, USA). Series resistance was monitored continuously and was typically compensated by 60-70 % in whole-cell configuration. Voltage-clamp recordings were filtered at 4 kHz, sampled at 10 kHz using a data acquisition board (Digidata 1322A, Axon Instruments) operated by Pclamp10 software (Axon Instruments). Off-line analysis was performed using Clampfit10 (Axon Instruments) and Origin8 (Origin Lab Corporation, Northampton, MA, USA).

Results and discussion Synthesis and structural characterization of the triblock copolymers A family of poly(2-methyl-2-oxazoline)-b-poly(tetrahydrofuran)-b-poly(2-methyl-2oxazoline) (TBCPs) was synthesized by cationic ring opening polymerization (Table S1). The sequential addition of tetrahydrofuran (THF), then oxazoline (MeOx) to a cationic initiator allowed for the synthesis of an ABA tribloc copolymer. The length of the various blocks was adjusted by controlling the polymerization time. The structures of TBCPs composed of poly(2-methyl-2-oxazoline) (pMeOx) as hydrophilic block and

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

polytetrahydrofuran (pTHF) as hydrophobic block are illustrated on Figure 1. Taking into account the Pluronic series, we synthesized a TBCP series with molar mass in the range of 1500 - 4000 g.mol-1 with various hydrophilic-hydrophobic (pMeOx/pTHF) balances. NMR was used to determine their composition and their molar mass. The expected shifts of the pTHF blocks were observed at 1.61 and 3.40 ppm and the pMeOx ones were evidenced by two peaks at 2.13 and 3.42 ppm. The telechelic CH2 groups adjacent to the end hydroxyl functions were identified at 3.78 ppm (Figure S1). We have to point out that when only one MeOx unit is added to the living growing oxonium, then the chemical shift of the resulting hydroxyl function is observed at 3.70 ppm. Furthermore, a complete coupling between the telechelic oxonium functions of the growing pTHF and the oxazoline units is evidenced by the absence of signal at 30.10 and 62.80 ppm on the 13C NMR spectra(S2), attributed to the HO-CH2-CH2-CH2-CH2function, is evidenced (S3). 20 This observation demonstrates the absence of homopTHF or pTHF-b-pMeOx diblock copolymer. The molar mass was determined from the ratio of the integrals of the protons of both the THF unit at 1.61 ppm and the methyl of the oxazoline units at 2.13 ppm to that of the telechelic hydroxyl functions. Derivatization of the hydroxyl functions by adding trichloromethylisocyanate in the NMR tube containing TBCP was performed to confirm the molar mass determination. Trichloromethylisocyanate allows derivatization of the hydroxyl function into isocyanate function and shifts the telechelic functions to 4.39 ppm, where no other signal could disturb integration measurement. When the NMR molar mass was compared to results obtained by SEC a fair agreement was observed ( S4 ; Table S1). The use of PMMA standards for SEC calibration may explain the slight discrepancies observed by SEC. Thus NMR molar mass will be used in the following discussion. In order to evidence the purity and architecture of TBCPs, Diffusion-Ordered

ACS Paragon Plus Environment

Page 12 of 36

Page 13 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

SpectroscopY NMR (DOSY NMR) experiments were carried out at 25 °C, in D2O. This two dimensional NMR method is based on pulse-field gradient spin-echo NMR experiments, in which components are submitted to diffusion, leading to measurements of chemical shifts and diffusion coefficients. 22 The diffusion coefficient (D) values of TBCPs were determined by fitting the exponential decay of the resonance signals intensity versus the gradient strength 22 and compared to that obtained with the pMeOx homopolymer, to demonstrate that no pMeOx is present in the triblock copolymer (Table S2). A pMeOx of a similar molar mass of the block of the triblock copolymers is taken as a reference. For the homopolymer pMeOx, D is higher than that of TBCP3 containing the same MeOx unit number per block, showing that the triblock copolymer has a higher hydrodynamic volume than the homopolymer. This is a first indication that the block copolymers are not a mixture of two homopolymers. The 1H NMR spectrum, projected on the x-axis, exhibits signals corresponding to the pMeOx blocks, at 2.1 ppm and 3.3 ppm, and the pTHF block at 3.4 and 1.6 ppm. The triblock architecture is confirmed by the pseudo 2D NMR of the amphiphilic copolymer showing clearly that all the proton signals belonging to THF and MeOx units have the same diffusion coefficient (1.40.10-10 m2s-1) which is lower than that of D2O (1.9.10-9 m2s-1) (Figure S5). The absence of other signals indicates that the products are not a mixture of pTHF and pMeOx, neither diblock copolymers.

Determination of the critical micellar concentration (CMC) The self-assembling properties of TBCPs in aqueous solution were determined by different

techniques

including

fluorescence

spectroscopy,

isothermal titration

calorimetry (ITC), dynamic light scattering (DLS) and DOSY NMR (Table S1). Nile Red (NR) is a lipophilic phenoxazone dye, which is widely used to evidence micelle formation.23 As shown in Figure S6A, the emission spectra of NR increase in the

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

presence of increasing concentrations of TBCPs. When the maximum emission intensities are plotted versus TBCP concentrations, a sigmoidal micellization curve is observed, and the CMC is determined as the inflexion point of the curve (Figure S6B ; Table S1). CMCs were also determined by using ITC. 24 Figure S7 represents the typical titration curve obtained by ITC where the CMC value corresponds to the extreme of the first derivative of the curve. CMCs measured by ITC and NR fluorescence spectroscopy were similar for most of TBCPs, except for TBCP6, TBCP2 and TBCP3. For these latter, CMCs determined by NR fluorescence spectroscopy were much higher than those measured from ITC. The difference could result from the sample preparation. For ITC measurements, copolymer solution was continuously stirred in the cuvette during measurements allowing micelles to reach an equilibrium state. In contrast, with NR fluorescence spectroscopy, micelles could dissociate slowly during dilution and the fluorescent probe could be detected for higher TBCP concentrations. These values were in the range 0.5-5 mg.L-1, depending on the TBCP molar mass and composition. The usual values of low molar mass amphiphilic copolymers were around 10 mg.L-1.25 Thus the TBCP family may provide a good structural stability of the drug-loaded micelles after dilution in the body. CMCs were confirmed by DLS from correlation function diagrams of TBCPs measured by DLS plotted as a function of the concentration (Figure S8). When they were unimers at low concentration, a single exponential fit the correlation function. Increasing TBCP concentration, the decay of the correlation function changed to a bi-exponential form that witness for the appearance of micelles. CMCs determined by DLS were comparable to those measured by ITC and fluorescence spectroscopy. DOSY NMR allowed determining the hydrodynamic radii and diffusion coefficients of a TBCP selection (Table S3). For the pMeOx homopolymer (Mn=870 g.mol-1), D was

ACS Paragon Plus Environment

Page 14 of 36

Page 15 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

around 2.2 10-10 m2s-1 and was taken as a reference through all the following discussion. Below their CMC determined by DLS, D varied from 1.1 10-10 m2s-1 to 1.5 10-10 m2s-1 and were in the same range as that of the homopolymer, proving that TBCPs and the homopolymer had the same conformation in solution. Consequently, their hydrodynamic radii varied from 0.6 nm to 1.4 nm. D values calculated from the StokesEinstein equation with DLS data were 1.8 10-10 m2s-1 to 4.1 10-10 m2s-1, and were slightly different from those obtained by DOSY NMR. Above CMC, the DOSY NMR analyses evidenced two populations with different D. The first population (D between 1.1 10-10 m2s-1 and 1.5 10-10 m2s-1) corresponded to unimers. The second one (D between 2.8 10-11 m2s-1 and 7.9 10-11 m2s-1) corresponding to micelles was ten times lower than that of the unimers. D determined from DLS (7.4 10-12 m2s-1 to 4.8 10-11 m2s-1) were slightly different from the DOSY NMR ones. These micelles exhibited a hydrodynamic radius (Rh) comprised between 5 nm and 32 nm. Alexandridis et al.26 reported such differences between the Rh values determined from DLS and DOSY NMR for the copolymer (pOE-b-pOP-b-pOE) which were ascribed to the polymer Đ. The combination of four techniques permitted to determine an accurate CMC for each TBCP, which is an important parameter regarding their potent biological activities. When considering CMC from DLS, the triblock architecture indicates that CMC decreases with the pTHF molar mass when that of pMeOx remains constant. Moreover, TBCP hydrophobicity increases with the pTHF molar mass favoring association of the hydrophobic blocks into the micelle core. Conversely, the CMC decreases when the pTHF molecular increases. Figure 2 shows the CMC variation of TBCPs as a function of the pTHF molar mass for a constant pMeOx molar mass. As expected, the CMC decreases rapidly when the pTHF block length increases and reaches a value of 2.6 10-5 mol.L-1 for a polymerization degree of 40. Very interestingly, the curve of log(CMC)

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 36

versus NTHF 1/3 (Figure 2) showed a linear relationship which is in accordance with the mixed micelle model suggested by Eisenberg et col. 27 The slope, close to -1.6, suggests a high dependence of the CMC with the pTHF molar mass.

Determination of the structure of micelles To design efficient systems in terms of drug delivery, one should take into account various important physicochemical parameters such as the size, morphology, charge, hydrophobicity and drug loading.28-30 A deeper characterization of micelles was performed for TBCP4, TBCP7 and TBCP8 having different molar masses, but a same pTHF length. This hydrophobic block being responsible for drug loading, micelles formed with TBCP2 were also characterized (nMeOx is close to that of TBCP4, but with a larger nTHF). The approach consisted in studying the aggregation number, core radius, corona thickness and stacking parameter β of micelles considering the copolymer chemical composition, 31,25 and the molar fraction of pMeOx length f =

NA where NA + NB

NA was NMeOx and NB was NTHF .25 The density of single pMeOx chain inside the micelle corona was determined and reported in Table S4. The molar mass of micelles (Mn,mic) at a TBCP concentration ten times higher than its CMC was first determined via static light scattering using the Debye equation. The aggregation number Nagg was given by:

N agg =

Mnmic where Mn,unim was the unimer molar mass determined from 1H-NMR. The Mnunim

hydrophobic core radius was given by: Rc = 3

3N agg Mn pTHF where MnpTHF was the pTHF 4π N avo ρ

molar mass, Navo the Avogadro number and ρ the density of the pTHF block equal to 0.915g.mL-1. The corona thickness L was determined by  = ℎ −  . The packing parameter β of a surfactant was defined by

β=

ν lc a0

where v was the volume occupied

ACS Paragon Plus Environment

Page 17 of 36

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

by the pTHF block, lc the pTHF length and ao the surface area of pMeOx (Scheme 1). Accordingly for a given TBCP, ν =

4π Rc3 4π Dh2 , lc = Rc and a0 = . For a complete 3N agg N agg

representation of the TBCP self-organization, the density of a single pMeOx chain inside the corona of the micelle was determined following: ρ MeOx =

3N MeOx N agg 4π (Rh3 − Rc3 )

. The

data are presented in Table S4. It appears that Nagg decreases with the length of pMeOx, the pTHF length remaining equal to 12-13. This observation is consistent with results already reported for copoly(oxyalkylene) copolymers.33 The pMeOx length does not affect the core radius of the micelles in the molar mass range studied. Different morphologies can be expected for lipids micelles according to the packing parameter β.22 When β is 0