P

ACa

b

a

ARRAA

K�TCO�P

1

rA[fosfuViprCaev

0h

Colloids and Surfaces B: Biointerfaces 103 (2013) 550– 557

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces B: Biointerfaces

jou rna l h om epa g e: www.elsev ier .com/ locate /co lsur fb

oloxamine micellar solubilization of �-tocopherol for topical ocular treatment

ndreza Ribeiroa, Isabel Sandez-Machob, Matilde Casasb, Susana Alvarez-Pérezb,armen Alvarez-Lorenzoa, Angel Concheiroa,∗

Departamento de Farmacia y Tecnología Farmacéutica, Facultad de Farmacia, Universidad de Santiago de Compostela, 15782 Santiago de Compostela, SpainDepartamento de Química Física, Facultad de Farmacia, Universidad de Santiago de Compostela, 15782 Santiago de Compostela, Spain

r t i c l e i n f o

rticle history:eceived 30 July 2012eceived in revised form 4 October 2012ccepted 18 October 2012vailable online 14 November 2012

eywords:-Tocopheroletronic 1107ataractscular drug delivery–A isotherm

a b s t r a c t

Ophthalmic delivery of �-tocopherol (TOC), which is the most active and cost/effective form of vitaminE, is receiving increasing attention as a way of preventing and treating glaucoma, cataracts, and dryeye syndrome, among other ocular pathologies. The aim of this work was to elucidate the possibilityof using poly(propylene oxide) (PPO) and poly(ethylene oxide) (PEO) block copolymers of poloxaminefamily (namely, Tetronic® 1107) to develop polymeric micelles that can host TOC, enhance the apparentsolubility and sustain the release of this vitamin in lachrymal fluid. The interactions of Tetronic 1107 withTOC were analyzed at the air–water interface recording the �–A isotherms at various temperatures, indi-cating favorable interactions as temperature increased from 10 to 29 ◦C. In 0.9% NaCl aqueous medium,a sharp increase in TOC solubility was observed when T1107 surpasses the critical micellar concentra-tion (CMC); the apparent solubility in 20% T1107 being more than 600-fold and 6000-fold that observedin the absence of copolymer at 4 and 25 ◦C, respectively. Micelles were characterized before and after

olymeric micelles loading by means of dynamic light scattering (DLS) and transmission electronic microscopy (TEM). TOCsustained release profiles were recorded in Franz-Chien diffusion cells. After storage for 3 months at 4 ◦C,TOC-loaded T1107 10% micellar system retained 84% TOC solubilized, which maintained the antioxidantactivity. Furthermore, the rheological properties of the micellar systems were not altered either; the vis-coelastic parameters being dependent on T1107 concentration, which opens the possibility of developing

ops to

from free-flowing eye-dr

. Introduction

Vitamin E is the common designation of eight naturally occur-ing closely related compounds, of which �-tocopherol (TOC;ppendix Fig. A.1) is the most active and cost/effective form

1]. This lipophilic vitamin, which can be found in a variety ofoods, is unstable when exposed to environment factors, such asxygen, light and temperature [2]. TOC is commonly added astabilizer to manufactured plastics, cosmetics and drug dosageorms [3]. In the human body tissues, the levels of TOC are reg-lated through specific transport proteins and receptors [4,5].itamin E is rapidly absorbed from the intestinal lumen and read-

ly distributed between lipids and proteins in the cell membranes,rotecting the cells from oxidative lysis and other free radicaleactions [6,7]. TOC also has antiproliferative and protein kinase-suppressing effects, which makes it useful for the prevention of

therosclerosis, glaucoma, cataracts, and posterior capsule prolif-ration [8]. Oral and parenteral administration of high doses ofitamin E, up to 1000–1200 IU (i.e., 100-fold the recommended

∗ Corresponding author. Tel.: +34 981563100x14886; fax: +34 981547148.E-mail address: angel.concheiro@usc.es (A. Concheiro).

927-7765/$ – see front matter © 2012 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.colsurfb.2012.10.055

in situ gelling systems.© 2012 Elsevier B.V. All rights reserved.

daily allowance), leads to TOC levels at retina and vitreous suit-able for protecting the eye from light-induced pathologies [5,9].However, these systemic routes do not lead to therapeutic lev-els in aqueous humor and lens [10]. The direct ocular applicationof TOC has been shown to be effective for the prophylaxis andrelief of symptoms of glaucoma, age-related and diabetic cataracts,dry eye syndrome, and photorefractive keractomy, and for pro-tection against anti-inflammatory drug-induced cataracts [11–19].In healthy eyes, vitamin E is distributed at a higher concentra-tion in the nucleus of the ocular lens than in the cortex. Duringcataract development, this level ratio becomes the opposite inorder to respond to the higher concentrations of hydrogen per-oxide in the aqueous and vitreous humors surrounding the lens[17]. Instillation of vitamin E liposomal and emulsion-based for-mulations retards cataract progression in humans and animalswith galactosemia or diabetes [18,20]. Topic formulations are alsobeneficial for treatment of eye fatigue and dry-eye syndromeof patients wearing contact lenses [14,21]. Nevertheless, despitethe recent increase in patents [14,21,22], the number of studies

with ophthalmic formulations is still limited probably becausesolubilization of vitamin E at sufficient concentration in stableformulations is a quite challenging issue. In fact, vitamin E hasbeen proposed as an enduring hydrophobic coating for medicated

es B: B

c[

oospmtmqtp(eTcroimaitsstsywpap[olsumso

2

2

pUPUwM

2

mpavcbwa

A. Ribeiro et al. / Colloids and Surfac

ontact lenses to regulate drug release on the corneal surface23].

The aim of this work was to elucidate the possibilities of devel-ping a polymeric micellar system for the ocular administrationf TOC. Polyoxyethylated amphiphilic polymers are particularlyuited for this purpose since they can increase drug solubility,rolong the time on the precorneal area, and enhance cornea per-eability, resulting in an increased ocular bioavailability compared

o other ophthalmic liquid formulations [24]. Unlike other nano-etric systems, the spontaneous formation of micelles allows large

uantities to be prepared in reproducible manner. Specifically,he work was carried with Tetronic 1107 (Appendix Fig. A.1), aoloxamine block copolymer formed by four poly(propylene oxide)PPO) and poly(ethylene oxide) (PEO) chains bonded to an ethyl-ne diamine central group [25]. Compared to related copolymers,etronic 1107 combines two advantageous features: (i) its provenompatibility with ocular and other tissues [26,27]; in fact, it is cur-ently being used at 1% in FDA-approved multipurpose solutionsf contact lenses to prevent protein adsorption [28,29]; and (ii)ts ability to self-associate in hydrophobic core-hydrophilic shell

icellar structures that solubilize and stabilize apolar drugs indverse aqueous environments [30–34]. The CMC of Tetronic 1107n 0.9% NaCl is as low as 0.4 mM (i.e., 0.6% w/w) and micelles ofhis poloxamine at 10% can increase up to 27-fold the apparentolubility of ethoxzolamide, a carbonic anhydrase inhibitor, anduccessfully pass the HET-CAM biocompatibility test according tohe NICEATM-ICCVAM protocol [35]. To the best of our knowledge,olubilization of TOC in polymeric micelles has not been reportedet. Thus, as a first step, the interaction of Tetronic 1107 with TOCas analyzed at the air–water interface with the aim of obtainingredictive information on the performance of the mixed systemst bulk, namely the capability of the micelles to host TOC and thehysical stability of the vitamin E-loaded micelles under dilution36]. Then, the effect of copolymer concentration on the solubilityf TOC was examined. Micelles were characterized, before and afteroading, by means of dynamic light scattering (DLS) and transmis-ion electronic microscopy (TEM). TOC release rate was evaluatedsing Franz-Chien diffusion cells. Finally, physical stability of theicelles and antioxidant activity of TOC throughout three months

torage were evaluated in order to further elucidate the feasibilityf poloxamine-based systems as ocular formulations.

. Materials and methods

.1. Materials

DL-�-tocopherol (TOC) (purity >98%) and 2,2-diphenyl-1-icrylhydrazyl (DPPH) were from Sigma Aldrich (St. Louis, MO,SA). Tetronic 1107 (T1107, Mw 15,000, 70% PEO, 60 EO and 20O units per each four arms) was from BASF (New Milford, CT,SA). Tween 80 was from Panreac (Barcelona, Spain). Purified wateras obtained by reverse osmosis (resistivity >18.2 M� cm; MilliQ®,illipore, Spain). Other reagents were of analytical grade.

.2. �–A isotherms

The surface pressure was measured using a Wilhelmy plateade from chromatography paper (Whatman Chr1, UK) as the

ressure sensor (accuracy of ±0.1 mN/m) in a single barrierpparatus (NIMA 611, UK) of total area 550 cm2, placed on an anti-ibration table. Prior to experiments, the trough was cleaned with

hloroform and ethanol and rinsed with water. The monolayer sta-ility was verified by monitoring the change in surface pressurehile holding the area constant. The subphase was purified water

nd the temperature was set at 10, 20, 25 or 29 ◦C. Solutions of

iointerfaces 103 (2013) 550– 557 551

TOC (0.1 mg/mL) and of copolymer (0.1 mg/mL) were prepared inchloroform, to which a drop of amyl alcohol was added to favorthe extensibility. To record the �–A isotherms of each single com-ponent, a volume of ca. 150 �L of the solution was deposited bymeans of a syringe (Hamilton, USA) at the air–water interface andallowed to stand for at least 10 min in order to ensure the completeevaporation of the solvent. The monolayers were compressed at aspeed of 15 cm2/min. To record the �–A isotherms of the binary sys-tems, adequate volumes of each solution were previously mixed.For example, T1107–TOC mixture with a TOC molar fraction of 0.4was prepared by mixing 1 mL of 0.1 mg/mL T1107 solution and19.5 �L of 0.1 mg/mL TOC solution. Then, the experiments werecarried out in the same way as for the single components.

2.3. Solubilization of TOC in polymeric micelles

Solutions of T1107 (0–20% w/w) were prepared by adding thecopolymer to cold 0.9% NaCl aqueous solution under stirring and,then, the solutions, were stored for 24 h at 4 ◦C or 25 ◦C. Aliquots(5 mL) of each solution were transferred into vials containing TOCin excess (∼240 mg) and kept under magnetic stirring at 4 ± 0.5 ◦Cor 25 ± 1 ◦C for 72 h. The samples were then filtered through PTFEmembranes of 0.45 �m pore size (Sartorius, Gottingen, Germany)to remove insoluble TOC. The concentration of dissolved TOC wasspectrophotometrically quantified at 292 nm (Agilent 8453, Wald-bronn, Germany), after adequate dilution with ethanol, using acalibration curve obtained with TOC solutions (0.01–0.1 mg/ml) inethanol. The filtered solutions were stored for three months at 4 ◦Cprotected from light and then the amount of TOC that remaineddissolved was again quantified as explained above.

2.4. Dynamic light scattering (DLS)

DLS was used to determine the apparent hydrodynamic diam-eter (Dh) of the micelles before and after loading of TOC. DLSmeasurements were carried out at 90◦ scattering angle using anALV-5000 F optical system equipped with CW diode-pump Nd:YAGsolid-state laser (400 mW) operated at a wavelength of 532 nm(Coherent Inc., Santa Clara, USA). The experiments were carriedout in triplicate and the apparent hydrodynamic radius (Rh,app) ofthe micelles was calculated from the apparent diffusion coefficientapplying the Stokes–Einstein equation [37].

2.5. Transmission electron microscopy (TEM)

Drug-free and drug-loaded micellar systems (10%, 10 �L) wereplaced onto a copper grid covered with Formvar film and, after 60 s,the excess was carefully removed and dried at room temperaturefor 15 min. Then, phosphotungstic acid (2% w/v, 10 �L) was addedand the excess removed after other 60 s. Finally, the sample wascarefully washed with distilled water, dried in a closed containerand observed using a Philips CM-12 TEM apparatus (FEI Company,Eindhoven, Netherlands).

2.6. Rheology

Viscoelastic behavior of 10 and 20% (w/w) T1107 dispersionswith and without TOC, freshly prepared and after three monthsstorage protected from light, was evaluated at 10 (the lowest tem-perature permittable with the available equipment), 25 and 37 ◦Cin a Rheolyst AR-1000N rheometer (TA Instruments, New Castle,

DE, USA) equipped with an AR2500 data analyzer, and fitted witha Peltier plate. The storage (G′) and the loss (G′′) moduli wererecorded at 0.5 Pa in the 0.5–50 rad/s angular frequency intervalusing a cone-plate geometry (diameter 6 cm, angle 2◦).

5 es B: Biointerfaces 103 (2013) 550– 557

2

aoeS(t2(s3mrdw

P

rt

A

wCP

2

ucSm8dvut8ebw

3

3

amamimmttpdmt

π (mN/m)0 5 10 15 20 25

Cs-1

(mN

/m)

0

10

20

30

40

50

60

70 20ºC

25ºC29ºC

10ºC

Area (A2/molecu le)0 20 40 60 80 10 0 12 0 14 0 16 0 18 0 20 0

π (m

N/m

)

-5

0

5

10

15

20

25

30

35

20ºC25ºC29ºC

10ºCA

B

52 A. Ribeiro et al. / Colloids and Surfac

.7. Antioxidant activity

The free radical DPPH method [38] was applied to quantify thentioxidant activity of TOC solubilized in 10% T1107 solution afterne day and after three months from the preparation. Aliquots ofach sample were diluted 25-fold with ethanol:water 50:50 (v/v).imilarly, fresh TOC 0.076 mg/mL solutions in ethanol:water 50:50v/v) (positive control) were prepared. The diluted samples andhe reference were further diluted 8, 20 and 40-fold (final volume

ml) in ethanol:water 50:50 (v/v). Then, 2 ml of a DPPH solution30 �g/mL, in ethanol:water 50:50 v/v) were added to each dilutedolution and the mixtures were kept protected from the light for0 min. The reference consisted of 2 ml ethanol:water 50:50 (v/v)ixed with 2 ml of the DPPH solution. Then, the absorbance was

ecorded at 525 nm (Agilent 8453, Waldbronn, Germany). For eachilution of micellar system, the power of inhibition of free radicalsas calculated as [38]:

I(%) =(

1 − Asample

Areference

)× 100

The values of PI of two dilutions of each micellar system thatesulted to be above (PI1) and below (PI2) 50% were used to estimatehe anti-radical power (ARP) as follows

RP (%) =(

1CI50

)× 100

here CI50 = C1 − �C and �C = [(C1 − C2)·(PI1 − 50)]/(PI1 − PI2) with1 and C2 the concentrations of the dilutions that render PI1 andI2 respectively.

.8. In vitro release studies

TOC release from 10% (w/w) T1107 micellar system was eval-ated, in triplicate, using Franz-Chien diffusion cells fitted withellulose acetate membrane filter (0.45 �m, Albet®, Barcelona,pain). The diffusion area was 0.785 cm2. The receptor compart-ent (7 mL) was filled with NaCl 0.9% solution containing Tween

0 0.5% (w/v) [39]. The micellar system (1 mL) was placed in theonor compartment and gently covered with parafilm® to pre-ent evaporation. The receptor solution was kept at 37 ± 0.5 ◦C andnder stirring with a magnetic bar. The concentration of TOC inhe receptor solution was monitored over time at 292 nm (Agilent453, Waldbronn, Germany) by taking 500 �L samples at pre-stablished time points, which were diluted with 500 �L ethanolefore absorbance measurement. The volume sample was replacedith fresh 0.9% NaCl solution containing 0.5% (w/v) Tween 80.

. Results and discussion

.1. �–A isotherms

The few papers that report on the behavior of TOC at their–water interface indicate that this vitamin can form stableonolayers [40–42]. We observed that the isotherm of TOC exhibits

firm influence of temperature (Fig. 1a). At 20 ◦C or below, theonolayer behaves as a condensed liquid. As the temperature

ncreased from 20 to 29 ◦C, a transition in the isotherm becameore evident and the collapse pressure diminished. The plot of theonolayer compressibility vs. surface pressure (Fig. 1b) enabled us

o identify the regions at which the phase transitions occurred bet-er. The local minimum in the compressibility corresponds to the

ressure of the transition [43,44]. The plots corresponding to 10 ◦Cid not show a defined minimum. As the temperature increased, theinimum shifted toward lower pressure. The effect of the tempera-

ure has been attributed to an increase in entropy and enthalpy due

Fig. 1. �–A isotherms of TOC at the air–water interface recorded at various temper-atures (A) and dependence of the monolayer compressibility on surface pressure(B).

to the freedom of motion of the phytol hydrophobic side chain ofTOC and to their larger interactions, which leads to more expandedand less rigid monolayers [42]. As can be appreciated from Fig. 1a,the limit area (i.e., the intercept on the X axis of the tangent straightline to the isotherm in the condensed region) notably increasedwhen temperature was raised from 20 to 29 ◦C. The increase inentropy and enthalpy was confirmed applying Motomura equation,which is an adaptation of Clausius–Clapeyron model to bidimen-sional systems [45]:

�H =(

d˘t

dT− d�

dT

)× T × �A

where �H and �A represent the changes in enthalpy and molecu-lar area in the transition region from expanded liquid to condensedliquid, respectively. The difference between the molecular areasoccupied by the monolayer at the end (Ac) and the beginning (Ae)of the pseudoplateau was used to estimate �A. Since the pseu-doplateau region is not rigorously flat and the pressure slightlychanges during the transition, d˘/dT was estimated from the meanvalues of pressure at the beginning and at the end of the transi-

tion. The term d�/dT corresponds to the change of water surfacetension as the temperature increases, which is 0.153 mN/(m K)in the 5–40 ◦C range. At 293 K, �H and �S/T values obtained inthis way were 6796.94 cal/mol and 22.81 cal/mol K (for Ae = 82.23

A. Ribeiro et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 550– 557 553

Area (A2/molecu le)0 200 0 400 0 600 0 800 0 1000 0 1200 0

π (m

N/m

)

0

5

10

15

20

25

30

XTOC

0.0 0.2 0.4 0.6 0.8 1.0

ΔGex

c (KJ

/mol

)

-2

-1

0

1

2

10ºC X=0X=0.2X=0.4X=0.6X=0.8

2.55.010

1520

Fig. 2. �–A isotherms and excess free energy change of mixing of T1107–TOC mono-layers at the air–water interface, recorded at 10 ◦C (XTOC indicates the molar fractionor

ÅvA

vcXfaloo(8pTarussblacabI

Area (A2/molecu le)0 200 0 400 0 600 0 800 0 1000 0 1200 0

π (m

N/m

)

0

5

10

15

20

25

30

XTOC

0.0 0.2 0.4 0.6 0.8 1.0

ΔGex

c (KJ

/mol

)

-2

-1

0

1

2

3

4

5

20ºC X=0X=0.2X=0.4X=0.6X=0.8

2.55.010

152025

Fig. 3. �–A isotherms and excess free energy change of mixing of T1107–TOC mono-

f TOC in each mixture with the copolymer). The coefficient of variation (CV) ofepeated measurements was in all cases below 6.7%.

2/molecule and Ac = 99.87 A2/molecule). At 302 K, �H and �S/Talues resulted to be 9363.85 cal/mol and 31.01 cal/molK (fore = 89.31 A2/molecule and Ac = 113.29 A2/molecule).

The �–A isotherms of poloxamines on HCl aq. have been pre-iously described [30]. Similarly to that observed under acidonditions, the �–A isotherms of T1107 on water (see plots for

= 0 in Figs. 2–4, and Appendix Fig. A.2) showed three well dif-erentiated regions that are attributed to the fact that their long EOrms undergo conformational changes as the pressure increases. Atow pressure, the copolymer adopts a flat (pancake) conformationn the surface. The total PO (17.5 A2) and EO (13–16.5 A2) groupsf T1107 at the air–water interface, plus the water molecules8.5 A2) that are interacting with these groups, roughly occupy000 A2/molecule. When the area becomes smaller, the pressurerogressively increases up to a conformational change occurs.he copolymer adopts a mushroom conformation above 5 mN/mnd a pseudoplateau was observed up to 10 mN/m, due to theearrangement of the PPO coils into loops and the immersion of EOnits in the aqueous subphase. The end of the pseudoplateau corre-ponded to the situation in which all EO units were immersed in theubphase, while the surface was exclusively occupied by the PPOlocks. This pseudoplateau is particularly evident for polymers with

ong EO arms, such as T1107, and leads to monolayers that behaves expanded-like films [36,46–48]. As the monolayer was furtherompressed, a rapid increase in � occurred. The copolymer adopts

brush conformation, characterized by the stretching of the PPOlocks due to space limitations and to increased lateral interactions.n the subphase, the helical PEO chains entangle with neighboring

layers at the air–water interface, recorded at 20 ◦C (XTOC indicates the molar fractionof TOC in each mixture with the copolymer). The coefficient of variation (CV) ofrepeated measurements was in all cases below 6.7%.

copolymer molecules, whilst the PPO blocks at the interface canform loops and even be partially solubilized in the aqueous PEOlayer (this process may be facilitated by the hydrophilic ethylene-diamine group). As compression increased more, both hydrophilicand hydrophobic blocks became stretched (condensed state). Aminor effect of temperature from 10 to 29 ◦C was observed onthe isotherm pattern of T1107: at low pressures, the increase intemperature led to a small increase in pressure (up to 1 mN/m) forthe same area, while at relatively high pressure (� > 14–15 mN/m),the isotherms superimposed. These observations are related to thedecrease in affinity for water of T1107 as temperature increases.

The �–A isotherms of T1107–TOC mixed monolayers (Figs. 2–4,and Appendix Fig. A.2) revealed that as the proportion in TOCincreased, the pressure diminished for the same area value. Thecharacteristic pseudoplateau of T1107 isotherm progressively van-ished as the proportion in TOC rose. Miscibility of the componentswas evaluated through the excess free energy of the mixed mono-layers estimated as the difference between the experimental dataand those values that would be obtained if the behavior of themixture was ideal, as follows [49,50]:

�Gexc = NA

∫ �

�0

A1,2d� − X1

∫ �

�0

A1d� − X2

∫ �

�0

A2d�

where NA is Avogadro’s number, and A1, A2 and A1,2 are the molec-ular areas of the pure components and of the mixture, respectively,at a given surface pressure �. Upper limits of the integral were 2.5,

554 A. Ribeiro et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 550– 557

XTOC

0.0 0.2 0.4 0.6 0.8 1.0

ΔGex

c (KJ

/mol

)

-5

-4

-3

-2

-1

0

1

2

Area (A2/molecu le)0 200 0 400 0 600 0 800 0 1000 0 1200 0

π (m

N/m

)

0

5

10

15

20

25

3025ºC X=0

X=0.2X=0.4X=0.6X=0.8

2015

2.55.0

10

Fig. 4. �–A isotherms and excess free energy change of mixing of T1107–TOC mono-layers at the air–water interface, recorded at 25 ◦C (XTOC indicates the molar fractionof TOC in each mixture with the copolymer). The coefficient of variation (CV) ofr

5ao(

i(wtatbct�afiFimtfaFsl

T1107 co ncent ration (M)

0.00 0 0.00 2 0.00 4 0.00 6 0.00 8 0.01 0 0.01 2 0.01 4

TOC

(M)

0.00

0.02

0.04

0.06

0.08

0.10

0.12

4ºC25ºC

TOC, multimodal distributions were obtained (Fig. 6), which is inagreement with previous reports [35]. The presence of different

epeated measurements was in all cases below 6.7%.

, 10, 15, 20 and 25 mN/m; except for the experiments carried outt 29 ◦C, in which the maximum upper limit was 15 mN/m becausef the collapse of TOC monolayer at temperature below 20 mN/mFig. 1a).

The �Gexc values of isotherms recorded 10 ◦C were slightly pos-tive, except for low pressures and high TOC relative proportionsFig. 2). At 20 ◦C, �Gexc values were in all cases positive (Fig. 3),hich means that T1107–TOC interactions are less favorable than

hose between each component itself. As explained above, TOC has long phytol hydrophobic chain that is not easy to accommodate inhe copolymer monolayer. Positive deviations from linearity haveeen previously observed for mixtures of TOC with phosphatidyl-holines [40], cholesterol [41], and vitamin C derivatives [42]. Onhe contrary, as the temperature increased to 25 and 29 ◦C the

Gexc of the T1107–TOC mixtures became negative, particularlyt low compression pressures. This finding suggests that as thereedom of motion of the phytol hydrophobic side chain of TOCncreases (as indicated by the conformational changes reported inig. 1 and the associated �H and �S/T values), TOC finds it easier tonteract with T1107 macromolecules. Furthermore, T1107 becomes

ore hydrophobic as temperature increases. As a consequence,he mixed monolayer at 25 ◦C is remarkably more stable than thatormed by the components separately. Further increase in temper-ture up to 29 ◦C also rendered negative values of �Gexc (Appendixig. A.2), although smaller than at 25 ◦C probably because the tran-

ition of the TOC component from expanded liquid to condensediquid occurs at lower pressures.

Fig. 5. Phase solubility diagrams for TOC with T1107 at 4 ◦C and 25 ◦C in 0.9% NaCl.The error bars represent the standard deviations (n = 4).

3.2. Micellar solubilization

An important property of micelles with particular consequencesin drug delivery is their ability to host poorly water-soluble activesubstances. Fig. 5 shows the TOC apparent solubility in 0.9% NaCl(isotonic with the lachrymal fluid) as a function of T1107 concentra-tion at 4 and 25 ◦C. These temperatures were chosen to resemblestorage conditions in the fridge and handling at room tempera-ture. Above the CMC of T1107 (ca. 0.0004 M or 0.6%), the solubilityof TOC rose linearly with increasing poloxamine concentration.At 25 ◦C, the micellar solution containing 0.013 M (∼20%) T1107enhanced TOC solubility from 0.006 mg/mL to 40.7 mg/mL, namely6783 times. As expected, the increase in solubility was smaller (661times) when the test was performed at 4 ◦C, which can be attributedboth to a lower tendency of T1107 to aggregate as micelles (asobserved for other poloxamines [34]) and particularly to the loweraffinity of TOC for T1107 at temperature below the transition ofTOC. As the �–A isotherms indicate, the interactions of TOC withT1107 become more favorable as the temperature increases. Theincrease in entropy may also enable TOC to adopt a conforma-tion suitable for accommodation in the hydrophobic cores of T1107polymeric micelles; thus, increasing the apparent solubility.

Despite micellar solubilization in aqueous salt environment isusually less favorable than in pure water [34], the increase in theapparent solubility of TOC is in the range of the values reportedfor the solubilization of other active substances by T1107 micelles;namely, 4.7, 27, 4870, and 15,100-fold the solubility in water ofsimvastatin [30], ethoxzolamide [35], efavirenz [51] and triclosan[52] respectively. Compared to other procedures for increasingTOC solubility, T1107 micelles resulted to be more efficient thanethanol:water 70:30 (v/v) medium (450-fold) [53] or inclusioncomplexes with cyclodextrins (150-fold) [54]. It has been alsoshown that incorporation in liposomes is limited to 5 mol% [55].Subsequent studies were mostly carried out with micellar systemsprepared with 10% T1107 since they can host relevant quantities ofTOC while maintaining a low viscosity adequate for instillation (asdiscussed below).

The DLS analysis of T1107 (10%) micellar systems prepared in0.9% NaCl aqueous solution were carried out before and after theloading with TOC at 25 ◦C. All samples showed clean autocorrela-tion diagrams, with a high number of counts. Both with and without

peaks in the population distribution obtained from DLS measure-ments provided useful information about the existence of several

A. Ribeiro et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 550– 557 555

Inte

nsity

(a.u

.)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

rh (nm)0.1 1 10 10 0 100 0 1000 0

Inte

nsity

(a.u

.)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Non-loaded T1107 10%

TOC-loaded T1107 10%

FT

sietwpwalnrpwiagOn(t(ts

awtlt1atoTatw

Angular frequ enc y (rad /s)0.1 1 10 10 0

G' a

nd G

'' (P

a)

0.001

0.01

0.1

1

10

100

1000

10000

G'' 10%, 10, 25 and 37ºC

G'' 20%, 10ºC

G' 20%, 25ºC

G'' 20%, 25ºC

G'' 20%, 37ºC

G' 20%, 37ºC

Fig. 7. Storage (G′) and loss (G′′) moduli of T1107 10% and 20% dispersions at 10 ◦C(G′′ blue lines), 25 ◦C (G′′ green lines, G′ gray lines), and 37 ◦C (G′′ red lines, G′ black

ig. 6. DLS analysis of T1107 10% micellar systems before and after loading withOC at 25 ◦C. Three independent measurements were registered for each system.

pecies with sufficient distinct sizes to be distinguished coexistingn solution. In the present case, Tetronic copolymers systems (eithermpty or loaded with vitamin E) display well differentiated popula-ions, which can be assigned in the following way: (i) the populationith mean sizes at ca. 2–3 nm might correspond to singly dispersedolymeric chains, that is, polymer unimers; (ii) polymeric micellesith sizes ranging from 10 to 30 nm (PDI = 0.468; SD = 2.77), in

greement with previous reports [32]; and (iii) other two popu-ations at ca. 200–300 nm and at even larger sizes (ca. 2000 nm),ot fully resolved when the micelles are empty, which would cor-espond to insoluble material residuals coming from the syntheticrocess of the copolymers (these copolymers are commercial andere used as received). This material can be partially incorporated

n the micelles forming larger micelle clusters (the peak centeredt ca. 200 nm), or to promote the formation of very large aggre-ates of insoluble species (population centered at ca. 2000 nm).nce TOC was loaded, the micellar population became predomi-ant and the size of the micelles slightly increased up to 30–40 nmPDI = 0.475; SD = 1.33), suggesting that TOC favors the micelliza-ion process. These results were further confirmed by means of TEMAppendix Fig. A.3). All micelles showed spherical morphology andhe presence of TOC did not modify their shape, but increased theize from ca. 36 to 62 nm.

The stability of TOC–T1107 10% micellar systems after stor-ge for 3 months at 4 ◦C, protected from light, was tested in twoays: (a) the amount of TOC that remained solubilized, and (b)

he antioxidant activity; which are two critical issues for the shelf-ife of a micellar system intended for ocular application. Regardinghe first issue, the initial TOC concentration 1.9 mg/ml decayed to.6 mg/ml; namely more than 84% TOC still remained in solutionfter three months. This is in agreement with previous results onhe stability of T1107 micelles against intense dilution [30]. On thether hand, the antioxidant power (ARP) of the freshly prepared

OC–T1107 micellar system was 13.0%, close to that of an equiv-lent solution of TOC in ethanol:water 50:50 (v/v) (13.1%). Afterhe storage, the ARP decayed to 12.7%, which is in close agreementith the small decrease in the concentration of TOC solubilized.

lines). (For interpretation of the references to color in Figure legend, the reader isreferred to the web version of the article.)

Therefore, T1107 effectively protected TOC in the micellar coreseven after storage under the less favorable conditions forTOC–T1107 interactions.

3.3. Viscoelasticity

Another relevant practical aspect refers to the easiness of appli-cation of the micellar systems as eye-drops. T1107 dispersions atboth 10% and 20% behaved as free flowing fluids at 10 ◦C, onlyexhibiting small G′′ values and negligible G′ (Fig. 7). At 10%, no rel-evant effect of temperature on the viscous behavior was observed.By contrast, 20% dispersions behaved as viscoelastic fluids withrelaxation times (i.e., the inverse of the angular frequency at whichG′ crossovers G′′) close to 0.15 s at 25 ◦C and longer than 20 s at37 ◦C. This temperature dependence could be exploited for prepar-ing in situ gelling formulations that are instilled as eye-drops andtransform into gel depots on the corneal surface [26]. It should bealso noted that loading of TOC and storage for 3 months did notchange the viscoelastic profile of the T1107 dispersions.

3.4. TOC release

Diffusion of TOC from the T1107 10% micellar systems, as freshlyprepared and after three months storage, was recorded using Franz-Chien diffusion cells filled with NaCl 0.9% medium containingTween 80 0.5% in order to create sink conditions. Diffusion stud-ies with TOC alone could not be carried out due to its extremelylow aqueous solubility, which prevents homogenous dispersion inwater. The micellar systems sustained the release for one day; 80%released in the first 12 h (Fig. 8). Under the eye physiological con-ditions, more efficient control of the release is expected. Moreover,the micelles could enter into the corneal cells while the unimers ofT1107 inhibited the efflux pumps; both favoring the ocular absorp-tion of TOC and prolonging the therapeutic effects [35,56].

After three months storage, TOC was released at a slightlygreater rate and with more variability than the freshly prepared

systems. This finding can be related to the presence of a greater pro-portion of free vitamin outside the micelles or adsorbed on to theirsurfaces after storage, which can diffuse faster than TOC hosted

556 A. Ribeiro et al. / Colloids and Surfaces B: B

Tim e (h)0 2 4 6 8 10 12 24

TOC

rele

ased

(mg/

cm2 )

0.0

0.5

1.0

1.5

2.0

Freshly prepa red3 mon ths storage

Ffs

ia

wafi(1

4

Tmmaaoerfspai

S

ima1

A

(e

[

[[

[

[

[

[

[[

[[

[

[

[[[

[

[

[[

[

[[

[

[

[

[

[[[

ig. 8. Diffusion profiles at 37 ◦C of TOC from T1107 10% micellar dispersions asreshly prepared and after three months storage. The error bars represent thetandard deviations (n = 3).

nside the micelles. Drug diffusion coefficient, D, was estimatedpplying the Higuchi model [57]:

Q

A= 2 × C0 ×

(D × t

�

)0.5

here Q/A represents the amount of TOC released per unit area,nd C0 is the initial TOC concentration in the micellar system. Thettings rendered R2 values above 0.99. D values were 1.23 × 10−4

0.08 × 10−4) cm2/s for freshly prepared micellar systems and.41 × 10−4 (0.20 × 10−4) cm2/s after the storage.

. Conclusion

The analysis of the �−A isotherms of TOC with poloxamine1107 at the air–water interface provides useful predictive infor-ation on the feasibility of solubilizing TOC in the polymericicelles. In close agreement with the �Gexc values, the increase in

pparent solubility is larger (one order of magnitude) at 25 ◦C thant lower temperature. Hosted TOC molecules favor self-aggregationf T1107 unimers and make the micelles physically stable for sev-ral months at 4 ◦C. Tuning T1107 concentration also enables theegulation of the viscoelastic behavior of the micellar systems fromree flowing liquids to in situ gelling systems. Overall, the enhancedolubility and stability of TOC as well as the sustained release profilerovided by T1107 micellar systems point toward them being suit-ble drug carriers for the attainment of therapeutic concentrationsn the ocular structures.

upplementary data

Chemical structure of TOC and Tetronic 1107 (Fig. A.1), �–Asotherms and excess free energy change of mixing of T1107–TOC

onolayers at the air–water interface recorded at 29 ◦C (Fig. A.2),nd TEM micrographs of T1107 10% micelles and TOC-loaded T11070% micelles (Fig. A.3).

cknowledgments

Work supported by MICINN (SAF2011-22771), Xunta de Galicia10CSA203013PR) and FEDER (Spain) and Fundac ão para Ciência

Tecnologia (FCT, Praxis grant SFRH/BD/40947/2007, Portugal).

[

[

[

iointerfaces 103 (2013) 550– 557

Authors thank P. Taboada and B. Blanco-Fernandez for help withthe DLS analysis and DPPH tests, and BASF Corporation (VerenaGeiselhart and Valerie Akavi) for providing poloxamine (Tetronic)samples.

Appendix A. Supplementary data

Supplementary data associated with this article canbe found, in the online version, at http://dx.doi.org/10.1016/j.colsurfb.2012.10.055.

References

[1] H. Sies, M.E. Murphy, J. Photochem. Photobiol. B: Biol. 8 (1991) 211.[2] C. Sabliov, C. Fronczek, C. Astete, M. Khachaturyan, L. Khachatryan, C. Leonardi,

J. Am. Oil Chem. Soc. 86 (2009) 895.[3] F. Manzanarez-López, H. Soto-Valdez, R. Auras, E. Peralta, J. Food Eng. 104

(2011) 508.[4] H.J. Kayden, M.G. Traber, J. Lipid Res. 34 (1993) 343.[5] K.N. Engin, Mol. Vis. 15 (2009) 855.[6] M. Drach, J. Narkiewicz-Michałek, A. Sienkiewicz, M. Szymula, C. Bravo-Díaz,

Colloids Surf. A: Physicochem. Eng. Asp. 379 (2011) 79.[7] G.S. Patil, D.G. Cornwell, J. Lipid Res. 19 (1978) 416.[8] N.K. Özer, Ö. S irikc i, S. Taha, K.N. Engin, D. Boscobionik, S. Clement, A. Stocker,

A. Azzi, in: T. Özben (Ed.), Free Radicals, Oxidative Stress and Antioxidants,Plenum Press, New York, 1998, p. 333.

[9] R. Bhat, T. Raju, A. Barrada, M. Evans, Pediatr. Res. 22 (1987) 16.10] M. Nagata, K. Kawazu, Y. Midori, M. Kojima, E. Shirasawa, K. Sasaki, Jpn. J.

Ophthalmol. 45 (2001) 125.11] H. Aonuma, K. Koide, K. Masuda, I. Watanabe, Jpn. J. Ophthalmol. 41 (1997) 160.12] K. Bilgihan, U. Adiguzel, C. Sezer, G. Akyol, B. Hasanreisoglu, Ophthalmologica

215 (2001) 192.13] T. Sakamoto, D.R. Hinton, H. Kimura, C. Spee, R. Gopalakrishna, S.J. Ryan, Graefes

Arch. Clin. Exp. Ophthalmol. 234 (1996) 186.14] V. Wong, L.L. Wood, Composition and method for treating dry eye syndrome,

PCT Int. Appl. (2010), WO 2010042843 A2 20100415.15] M. Kojima, Y.B. Shui, H. Murano, M. Nagata, O. Hockwin, K. Sasaki, N. Takahashi,

Invest. Ophthalmol. Vis. Sci. 43 (2002) 1116.16] M. Kojima, Y.B. Shui, H. Murano, K. Sasaki, Ophthalmic. Res. 28 (Suppl. 2) (1996)

64.17] Y. Ohta, J. Clin. Biochem. Nutr. 35 (2004) 35.18] Y. Ohta, T. Yamasaki, T. Niwa, Y. Majima, I. Ishiguro, Exp. Eye Res. 68 (1999)

747.19] M. Nagata, M. Kojima, K. Sasaki, J. Ocul. Pharmacol. Ther. 15 (1999) 345.20] M. Kaneda, Y. Majima, H. Hattori, H. Torii, T. Akiyama, M. Kosaka, J. Eye 8 (1991)

93.21] C. Inoue, T. Okumura, H. Sugafuji, Ophthalmic preparations for treatment of eye

fatigue of patients wearing contact lenses, Patent JP 2011246383.22] M.H. Friedlaender, H. Takruri, Ultraviolet absorbing ophthalmic compositions,

PCT Int. Appl. (2008), WO 2008002582 A2 20080103.23] C.C. Peng, J. Kim, A. Chauhan, Biomaterials 31 (2010) 4032.24] J. Jiao, Adv. Drug Deliv. Rev. 60 (2008) 1663.25] C. Alvarez-Lorenzo, A. Rey-Rico, A. Sosnik, P. Taboada, A. Concheiro, Front.

Biosci. (Elite Ed.) 2 (2010) 424.26] A. Rey-Rico, M. Silva, J. Couceiro, A. Concheiro, C. Alvarez-Lorenzo, Eur. Cell.

Mater. 21 (2011) 317.27] A.M. Puga, A. Rey-Rico, B. Magarinos, C. Alvarez-Lorenzo, A. Concheiro, Acta

Biomater. 8 (2012) 1507.28] S. Tonge, L. Jones, S. Goodall, B. Tighe, Curr. Eye Res. 23 (2001) 51.29] L.N. Subbaraman, S. Bayer, M.A. Glasier, H. Lorentz, M. Senchyna, L. Jones,

Optom. Vis. Sci. 83 (2006) 143.30] J. Gonzalez-Lopez, C. Alvarez-Lorenzo, P. Taboada, A. Sosnik, I. Sandez-Macho,

A. Concheiro, Langmuir 24 (2008) 10688.31] C. Alvarez-Lorenzo, A. Concheiro, J. Drug Deliv. Sci. Technol. 20 (2010) 249.32] D.A. Chiappetta, C. Alvarez-Lorenzo, A. Rey-Rico, P. Taboada, A. Concheiro, A.

Sosnik, Eur. J. Pharm. Biopharm. 76 (2010) 24.33] Y. Kadam, K. Singh, D.G. Marangoni, J.H. Ma, V.K. Aswal, P. Bahadur, Colloids

Surf. A: Physicochem. Eng. Asp. 369 (2010) 121.34] P. Parekh, K. Singh, D.G. Marangoni, P. Bahadur, Colloids Surf. B: Biointerfaces

83 (2011) 69.35] A. Ribeiro, A. Sosnik, D.A. Chiappetta, F. Veiga, A. Concheiro, C. Alvarez-Lorenzo,

J. R. Soc. Interface 9 (2012) 2059.36] J. Gonzalez-Lopez, I. Sandez-Macho, A. Concheiro, C. Alvarez-Lorenzo, J. Phys.

Chem. C 114 (2009) 1181.37] R. Pecora, J. Nanopart. Res. 2 (2000) 123.38] P. Goupy, M. Hugues, P. Boivin, M.J. Amiot, J. Sci. Food Agric. 79 (1999) 1625.39] Y.B. Song, J.S. Lee, H.G. Lee, Colloids Surf. B: Biointerfaces 73 (2009) 394.

40] M. Feng-Kui, W. Jing, S. Uematsu, Y. Akahori, Acta Phys. Chim. Sin. 11 (1995)

1077.41] Y.J. Lee, H.S. Rho, D.H. Kim, J.D. Kim, Colloids Surf. A: Physicochem. Eng. Asp.

205 (2002) 173.42] G. Capuzzi, P. Lo Nostro, K. Kulkarni, J.E. Fernandez, Langmuir 12 (1996) 3957.

es B: B

[[

[[[

[[[

[[

[[

A. Ribeiro et al. / Colloids and Surfac

43] M.I. Sández Macho, A. Gil González, A. Suárez Varela, Langmuir 16 (2000) 9347.44] T.J. Joncheray, K.M. Denoncourt, M.A.R. Meier, U.S. Schubert, R.S. Duran, Lang-

muir 23 (2007) 2423.45] K. Motomura, Adv. Colloid Interface Sci. 12 (1980) 1.46] S.M. O’Connor, S.H. Gehrke, G.S. Retzinger, Langmuir 15 (1999) 2580.

47] M.C. Fauré, P. Bassereau, L.T. Lee, A. Menelle, C. Lheveder, Macromolecules 32

(1999) 8538.48] L.C. Chang, Y.Y. Chang, C.S. Gau, J. Colloid Interface Sci. 322 (2008) 263.49] L. Zhao, S.S. Feng, M.L. Go, J. Pharm. Sci. 93 (2004) 86.50] Z. Wang, S. Yang, Chemphyschem 10 (2009) 2284.

[[

[

iointerfaces 103 (2013) 550– 557 557

51] D.A. Chiappetta, C. Hocht, C. Taira, A. Sosnik, Nanomedicine 5 (2009) 11.52] D.A. Chiappetta, J. Degrossi, S. Teves, M. D’Aquino, C. Bregni, A. Sosnik, Eur. J.

Pharm. Biopharm. 69 (2008) 535.53] M.D. Dubbs, R.B. Gupta, J. Chem. Eng. Data 43 (1998) 590.54] T.B. Shea, D. Ortiz, R.J. Nicolosi, R. Kumar, A.C. Watterson, J. Alzheimers Dis. 7

(2005) 297.55] F. Nacka, M. Cansell, P. Meleard, N. Combe, Lipids 36 (2001) 1313.56] C. Alvarez-Lorenzo, A. Sosnik, A. Concheiro, Curr. Drug Targets 12 (2011)

1112.57] R.G. Stehle, W.I. Higuchi, J. Pharm. Sci. 56 (1967) 1367.