Ta

ECa

b

c

a

ARR1AA

KTPPPC

1

mcaGHwgst2nse2

mT

0h

International Journal of Pharmaceutics 459 (2014) 1– 9

Contents lists available at ScienceDirect

International Journal of Pharmaceutics

j o ur nal ho me page: www.elsev ier .com/ locate / i jpharm

hermosensitive hydrogels of poly(methyl vinyl ether-co-maleicnhydride) – Pluronic® F127 copolymers for controlled protein release

sther Morenoa, Juana Schwartza,b, Eneko Larranetab, Paul A. Nguewaa,armen Sanmartína,c, Maite Agüerosb, Juan M. Iracheb, Socorro Espuelasa,b,∗

Tropical Health Institute, University of Navarra, Irunlarrea 1, E-31008 Pamplona, SpainPharmacy and Pharmaceutical Technology Department, University of Navarra, Irunlarrea 1, E-31008 Pamplona, SpainOrganic and Pharmaceutical Chemistry Department, University of Navarra, Irunlarrea 1, E-31008 Pamplona, Spain

r t i c l e i n f o

rticle history:eceived 2 August 2013eceived in revised form1 November 2013ccepted 18 November 2013vailable online 4 December 2013

eywords:

a b s t r a c t

Thermosensitive hydrogels are of a great interest due to their many biomedical and pharmaceuticalapplications. In this study, we synthesized a new series of random poly (methyl vinyl ether-co-maleicanhydride) (Gantrez® AN, GZ) and Pluronic® F127 (PF127) copolymers (GZ–PF127), that formed ther-mosensitive hydrogels whose gelation temperature and mechanical properties could be controlled bythe molar ratio of GZ and PF127 polymers and the copolymer concentration in water. Gelation tempera-tures tended to decrease when the GZm/PF127 ratio increased. Thus, at a fixed GZm/PF127 value, sol–geltemperatures decreased at higher copolymer concentrations. Moreover, these hydrogels controlled the

hermosensitive hydrogelsluronicsolyanhydrideroteinontrolled release

release of proteins such as bovine serum albumin (BSA) and recombinant recombinant kinetoplastidmembrane protein of Leishmania (rKMP-11) more than the PF127 system. Toxicity studies carried outin J774.2 macrophages showed that cell viability was higher than 80%. Finally, histopathological anal-ysis revealed that subcutaneous administration of low volumes of these hydrogels elicited a tolerableinflammatory response that could be useful to induce immune responses against the protein cargo in thedevelopment of vaccine adjuvants.

© 2013 Elsevier B.V. All rights reserved.

. Introduction

Hydrogels are polymer-based systems that have received muchore attention because of their numerous biomedical and pharma-

eutical applications, including drug delivery, cell encapsulationnd tissue repair (Appel et al., 2012; Jeong et al., 2000; Ruelariépy and Leroux, 2004; Wang et al., 2010; Yu et al., 2010).ydrogels that exhibit the specific property of increased viscosityith increased temperature are known as thermosensitive hydro-

els. These hydrogels have shown easier application and longerurvival periods at the site of application as compared to non-hermosensitive hydrogels (Cheaburu et al., 2013; Douglas et al.,013). However, there is an important need in the development ofew systems with shorter gelation times, higher biodegradability,

tronger mechanical strength, better bioadhesive properties andxtended sustained compounds release (Liu et al., 2007; Park et al.,009; Sosnik and Cohn, 2004).

∗ Corresponding author at: Pharmacy and Pharmaceutical Technology Depart-ent, University of Navarra, Irunlarrea 1, E-31008 Pamplona, Spain.

el.: +34 948425600; fax: +34 948425619.E-mail address: sespuelas@unav.es (S. Espuelas).

378-5173/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.ijpharm.2013.11.030

Pluronic® F127 (PF127), also named poloxamer, is apoly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide)(PEO–PPO–PEO) triblock copolymer that has interest because it canform a gel in situ depending on its concentration and temperature(Alexandridis et al., 1995). Due to the presence of hydrophobicpropylene oxide blocks, concentrated solutions of PF127 (20–30%w/v) can pass from low viscosity solutions (sol) to solid gels (gel)upon heating to body temperature, resulting in a reverse thermalgelation (Moore et al., 2000; You and Van Winkle, 2010). Thus,it has been reported to be one of the less toxic of commerciallyavailable copolymers and its biocompatibility makes PF127 anattractive candidate as a pharmaceutical vehicle for drugs throughdifferent ways of administration (Escobar Chávez et al., 2006; Leeand Tae, 2007). Despite its multiple benefits, PF127 has yet weakmechanical properties and this limits its application because it isnot stable and easily destroyed when diluted after its adminis-tration into the human body (Chun et al., 2005; Sosnik and Cohn,2004).

To circumvent these problems, physical or chemical modifica-

tions of PF127 have been previously reported. Hydrogels composedof linoleic acid (Guo et al., 2009), hyaluronic acid (Hsu et al., 2009)or alginate (Fang et al., 2009) linked to Pluronic F127 sustainedthe release of paclitaxel, cisplatin and carboplatin, respectively, for

2 ourna

lpwPaecte

iapupiaahtdear(tcp

apvd

2

2

AS

E. Moreno et al. / International J

onger period of time than Pluronic alone. V.g. at 6 h 20% of cis-latin was released from hyaluronic acid–Pluronic F127 hydrogelshile PF127 hydrogels released more than 60% (Hsu et al., 2009).

ark et al. developed a chitosan–Pluronic F127 hydrogel suitables an injectable chondrocite delivery carrier for cartilage regen-ration (Park et al., 2009) whereas hydrogels based on conjugatedatechol-functionalized chitosan to PF127 showed strong adhesiono soft tissues and mucous layers that could be useful for tissuengineering (Ryu et al., 2011).

In this study, PF127 was crosslinked with Gantrez® AN (GZ) aim-ng to preserve the thermogelling properties of the former as wells to reinforce its mechanical stability and bioadhesiveness. Thisoly(methyl vinyl ether-co-maleic anhydride) copolymer has beensed for pharmaceutical and medical purposes due to its biocom-atibility and good bioadhesive properties (Arbós et al., 2003). It is

ncluded within the list of GRAS excipients and used in denturedhesive bases, enteric coatings, ostomy adhesives, toothpastesnd mouthwashes (Andrade Acevedo et al., 2009). Moreover, GZas been employed to manufacture bioadhesive nanoparticles ableo improve the gastrointestinal retention and oral absorption ofrugs with poor oral biodisponibility (Agüeros et al., 2010; Arbóst al., 2002; Porfire et al., 2010). Due to the reactivity of the poly-nhydride groups, GZ can easily react with different moleculeseaching novel derivatives that present interesting new abilitiesAgüeros et al., 2010; Yoncheva et al., 2012). Furthermore, whenhe anhydride group degrades hydrolytically, the resulting productontains two carboxylic acid groups which enhance its bioadhesiveroperties (Arbós et al., 2002).

The general aim of this work was the design, synthesis and char-cterization of new GZ–PF127 hydrogels. Thermosensitive sol–gelhase transitions and viscosity properties were examined and initro release studies were carried out. In vitro and in vivo studiesetermined their toxicity and biocompatibility.

. Materials and methods

.1. Materials

Poly(methy vinyl ether-co-maleic anhydride) (Gantrez®

N-119, MW = 213,000) was kindly gifted by ISP (Barcelona,pain). Pluronic® F127 and 3-(4,5-dimethylthiazol-2-yl)-2,

Fig. 1. Reaction for the synthesis of t

l of Pharmaceutics 459 (2014) 1– 9

5-diphenyltetrazolium bromide (MTT) were purchased fromSigma (St. Louis, MO, USA). Anhydrous tetrahydrofuran (THF),ethyl ether and sodium hydroxide were obtained from Panreac(Barcelona, Spain). Amicon® Ultra tubes were purchased fromMillipore (Cork, Ireland). Transwell® Permeable Supports wereobtained from Corning (NY, USA). All other reagents were ofanalytical grade and were used without further purification.

2.2. Synthesis of Gantrez® AN–Pluronic® F127 colopymer

Briefly, 2.0 g of PF127 (MW = 12,500) were added into aone-necked glass flask and dissolved in 150 mL of anhydroustetrahydrofuran with a mechanical stirrer. Then, different amountsof GZ (MW = 213,000) were added to the solution (GZm/PF127molar ratios = 1, 2, 5, 10 and 20; GZm = GZ monomer) and reactedfor 48 h at room temperature under magnetic stirring. A reactionbetween the hydroxyl groups of PF127 and the carboxyl groups ofthe maleic anhydride of GZ resulted in the formation of ester groups(Fig. 1). The organic solvent was eliminated by evaporation and theresidue was mixed with cold ethyl ether to precipitate the copoly-mer. Later, the copolymer precipitate was filtered and dried at roomtemperature under vacuum. Finally, the different copolymers werecharacterized without further purification.

2.3. FT-IR analysis

IR spectra were obtained on a Thermo Nicolet FT-IR Nexus spec-trophotometer in a scanning range of 600–4000 cm−1 for 32 scansat a spectra resolution of 4 cm−1. Spectra were run as dry KBr(Aldrich, Germany) pellets and were analyzed using OMNIC E.S.P.software.

2.4. NMR analysis

1H NMR spectra were recorded on a Bruker 400 Ultrashield TM

spectrometer (Rheinstetten, Germany) using Tetra-methylsilane(TMS) as the internal standard. (CD3)2SO and CDCl3 were used assolvents. Chemical shifts were expressed as parts per million, ppm(ı).

he new GZ–PF127 copolymers.

ourna

2

mir

2

swtu

2

afiTi

2t

caisT6ecitcp

2

rrmpw2r

2

1NabMB3(i7o

E. Moreno et al. / International J

.5. Gel permeation chromatography (GPC) analysis

GPC measurements were carried out with a Water LC separationodule equipped with a Aquagel OH-30 column and a refractive

ndex detector. Water was used as a mobile phase and the flowate was 1 mL/min at a room temperature.

.6. Differential thermal analysis (DTA)

Thermal analysis data were obtained using a simultaneous TGA-DTA 851 Mettler Toledo thermoanalyzer. Samples (10–20 mg)ere sealed in pierced aluminum crucible pans and were subjected

o a heating program from 25 to 300 ◦C at a heating rate of 5 ◦C/minnder a static air atmosphere and N2 (20 mL/min) as purge gas.

.7. X-ray diffractometry (XRD) analysis

Power X-ray diffraction patterns were collected on a X Brukerxs D8 Advance diffractometer (Karlsruhe, Germany) using a Nilter, CuK� radiation, a voltage of 40 kV and a current of 30 mA.he scanning rate was 1◦/min, the time constant 3 s/step over a 2�nterval of 2–40◦.

.8. Preparation of GZ–PF127 hydrogels and sol–gel phaseransition measurements in vitro

Solutions of different molar ratios (GZm/PF127) of the newolopymers (1, 2, 5, 10 and 20) were prepared in 10 mL glass vialst concentrations of 0.5, 1, 2, 2.5, 3, 5, 7.5 and 10 wt.% in deion-zed water to a final volume of 2 mL. Vials were kept in continuoustirring at room temperature until complete dissolution (24–48 h).hen, 0.2 N NaOH was added carefully in order to adjust pH between

and 7.5. Finally, solutions were kept in a water bath at differ-nt temperatures to allow gel formation. Temperatures at whichopolymer solutions stopped flowing upon the vial inversions, dur-ng at least 30 s, were visually observed and recorded as gelationemperatures. The new hydrogels were compared with differentoncentrations of physical mixtures of GZ and PF127. Data wereerformed in triplicate and results were expressed as mean ± SD.

.9. Viscosity analysis

Viscosity measurements of GZ–PF127 hydrogels were car-ied out at different temperatures using a Haake Viscotester 550otational viscometer with a SV2 rotor and equipped with a ther-ostatic bath Thermo Phoenix II. 6 mL of hydrogel samples were

repared in water and placed on the plate of the viscometerhere temperature was increased at 2 ◦C intervals over the range

5–45 ◦C. After a stabilization time of 30 s, 100 data points wereecorded during a measuring period of 60 s.

.10. In vitro release studies

Release from GZ–PF127 hydrogels (molar ratio = 5; 5 and0 wt.%) was evaluated using dextran (Dextran-Texas Red®, 10 kDa,eutral; Invitrogen) and two different proteins, bovine serumlbumin (BSA, 66 kDa, Sigma) and recombinant kinetoplastid mem-rane protein of Leishmania (rKMP-11 kDa, kindly provided byanuel Soto), as examples of molecules with different properties.

riefly, copolymers were dissolved in 2 mL of deioned water in0 mL vials (diameter 2.5 cm). After total dissolution, a 1% w/w

drug weight vs. colopymer weight) of each drug was added andncubated for 5 min at 37 ◦C on a water bath shaker. Then, PBS (pH.4) at 37 ◦C was carefully added to the samples to a final volumef 20 mL. The content in PBS was analyzed at different times using

l of Pharmaceutics 459 (2014) 1– 9 3

a spectrofluorometer (FP-6300, Jasco) for dextran samples (� exci-tation: 595 nm; � emission: 615 nm; band excitation width: 5 nm;band emission width: 10 nm; number of cycles: 3) and Microbicin-choninic acid (MicroBCA) protein assay kit (obtained from PierceUSA) for BSA and rKMP-11 samples. These samples were incubatedwith the MicroBCA reagent for 2 h at 37 ◦C. After that, solutionswere measured in a spectrophotometer (iEMS Reader MS, Labsys-tems) at 570 nm and compared with the absorbance data obtainedfor a BSA and rKMP-11 control calibration curve (0–200 �g/mL) dis-solved in PBS. Samples were assessed in duplicate and results wereexpressed as mean of cumulative release (%) ± SD.

2.11. Analysis of release data

Data obtained from the in vitro release experiments were fittedto different mathematical models of drug release. Models used forthis study were the Korsmeyer–Peppas equation (Eq. (1)) and theHiguchi equation (Eq. (2)). Korsmeyer–Peppas et al. (Ritger et al.,1987) developed a simple semiempirical model which exponen-tially relates drug release with the elapsed time (Eq. (1)).

Mt

M∞= KKP · tn (1)

where Mt/M∞ is the drug release fraction at time t, KKP is a con-stant incorporating the structural and geometric characteristics ofthe matrix tablets and n is the release exponent indicative of thedrug release mechanism. The value of n indicates the mechanismof the release (Peppas and Sahlin, 1989). If this value is around0.5 (the exact value depends of the geometry) the mechanism isCase I (Fickian) diffusion and a value between 0.5 and 0.89 indi-cates anomalous (non-Fickian) diffusion. Values of n equal to 0.89indicate Case II transport.

Mt

M∞= KH · t1/2 (2)

When the obtained release mechanism is mainly a Fickian dif-fusion, a dimensionless expression of the Higuchi model was used(Eq. (2)) (Costa et al., 2001). Mt/M∞ is the drug release fraction attime t, and KH is the Higuchi constant. In the original Higuchi equa-tion the Mt (amount of drug released at time t) term is found insteadof Mt/M∞.

2.12. In vitro cytotoxicity of GZ–PF127 copolymer in J774.2macrophages

For the determination of the hydrogels cytotoxicity, a modifi-cation of ISO 10993-5:2009(E) was carried out. An indirect contactmethodology was developed using Transwell® microplates (Trudeland Massia, 2002). The use of Transwell® inserts to contain hydro-gels and polymeric solutions is an excellent method for screeningnew materials. GZ/PF127 = 5 copolymer aqueous solutions (3 and5 wt.%) were obtained by dissolving the copolymer in water andadjusting the pH around 6–7.5. Then, Transwell® inserts (8 �mpore size polycarbonate membrane filter) were filled with 1.5 mLof the aqueous solutions and, after heating for 5 min at 37 ◦C in theincubator, hydrogels were formed.

J774.2 macrophages were grown and maintained in RPMI 1640medium (Invitrogen) supplemented with 10% fetal bovin serum(Gibco), 2 mM L-glutamine, 100 units/mL penicillin and 100 mg/mLstreptomycin (Gibco) at 37 ◦C in 5% CO2. 1.5 mL of cultured cellswere carefully seeded in the lower chamber of 6 well plates(7.5 × 105 cells per well) using the pipette tip access ports of the

Transwell® inserts. After that, cultured plates were returned to theincubator for 20 h. As negative control no hydrogel was added towells. Thereafter, hydrogels were removed and the MTT assay wasperformed. Briefly, the MTT reagent was solubilized in PBS at a

4 ournal of Pharmaceutics 459 (2014) 1– 9

cratpwLe

2

uoAtNeg(nMRaiwFbatss

3

3

orhim

deagTv

fr7Pfowrn

tPip

Table 1Characteristics of the synthesized copolymers.

GZm/PF127 molarratio

Free PF127 (%) Degree ofsubstitution (%)

Tma (◦C)

1 70.81 29.12 532 76.63 13.11 505 54.23 8.13 53

10 24.01 6.71 5220 13.92 4.13 52

E. Moreno et al. / International J

oncentration of 2 mg/mL and sterile-filtered. A 300 �L portion ofeagent was added to each well and cells were incubated for 3 ht 37 ◦C. To ensure complete solubilization of the formazan crys-als, 1.5 mL of 10% SDS in 0.01 M HCl lysing buffer was added andlates were gently shaken for 30 min. Finally, sample absorbancesere measured at 570 nm using a plate reader (iEMS Reader MS,

absystems). Data were obtained from at least three independentxperiments.

.13. Gel integrity and biocompatibility in vivo

Female Wistar rats (average weight 200 g; Harlan, Spain) weresed to study gel integrity and biocompatibility of GZ–PF127 aque-us copolymer solutions (molar ratio = 5; 5 wt.%, pH 6.5–7) in vivo.nimal experiments were performed in compliance with regula-

ions of the responsible Ethical Committee of the University ofavarra in strict accordance with the European legislation in animalxperiments. Rats were anesthetized with 3–4% isofluorane in oxy-en. After shaving and disinfecting, different volumes of hydrogel50, 100, 250, 500, 750 and 1500 �L) at 20 ◦C were injected subcuta-eously into the sides of the back using a syringe with a 22G needle.oreover, 1500 �L of PBS or PF127 25% (w/v) were also injected.

atsı bulges were measured every 2 days during the experimentnd animals were killed 1 day, 7 days, 14 days or 30 days after thenjection of the hydrogel. Then, tissues containing the implants as

ell as the surrounding tissue were harvested and cut into slices.or histological examination, samples were fixed in 4% phosphate-uffered formaldehyde solution for 48 h and dehydrated through

graded ethanol series. After embedding in paraffin, tissue sec-ions (4–5 �m thick) were stained with hematoxylin and eosin bytandard techniques. In order to assess the histological changes,amples were evaluated using light microscopy.

. Results

.1. Synthesis and characterization of the copolymers

GZ–PF127 copolymers were synthesized by means of ring-pening polymerization of the anhydride groups of GZ through theeaction with the hydroxyl groups of PF127 (Fig. 1). PF127, whichas two interconnectable functional groups, is used as a crosslink-

ng agent and allows the formation of chemical bonds between theacromolecular chains.FT-IR spectra of the crosslinked GZ showed the C–H vibrations

istributed between 2951 and 2886 cm−1. The typical band of thester carbonyl, that confirmed the introduction of PF127 blocks,ppeared at 1735 cm−1 and the absorption band of the carboxylicroup, resulting from the opened maleic anhydride, at 1770 cm−1.he high intensity of a band at 1112 cm−1 is due to the C O Cibration (data not shown).

1H-NMR and GPC analysis of GZ–PF127 copolymers were per-ormed in order to calculate total and free amounts of PF127,espectively. Protons assignments for crosslinked GZ in CDCl3 (ı.28 ppm) used in the determination were: ı 1.13 ppm = CH3 ofF127 and ı 2.25 ppm = CH2 of GZ. The same analysis was conductedor all the GZ–PF127 copolymers and the degree of substitutionbtained for the copolymers appears in Table 1. It was observed thathen molar ratios increased the degree of substitution decreased,

evealing that although more GZ was added the reaction yield didot increase.

Differential thermal analyses were carried out to examine the

hermal behavior of the new copolymers. The DTA thermograms ofF127 showed a sharp endothermic peak corresponding to its melt-ng temperature (Tm) at 55 ◦C. Furthermore, the same endothermiceak was obtained in the case of the physical mixture of PF127

a Tm: melting temperature; PF127 Tm = 55 ◦C; GZ Tm > 300 ◦C; physical mixture ofGZm/PF127 = 55 ◦C.

and GZ. In contrast, GZ samples exhibited no thermal signals in thetemperature interval studied, revealing a great thermal stability(Table 1).

Additionally, the thermograms for the synthesized copolymersshowed a different behavior than the initial polymers and the phys-ical mixture, confirming that a reaction between PF127 and GZoccurred. These samples melted at a slightly lower temperature,between 50 and 53 ◦C (Table 1), and attained wider endothermicpeaks in the case of copolymers with a molar ratio of 10 and 20(data not shown).

X-ray diffractograms indicated that the different GZ–PF127copolymers maintained a crystal structure similar to PF127 anddifferent from the amorphous structure of GZ, suggesting that thecrystallinity is, apparently, not affected by the reaction (data notshown).

3.2. Sol–gel transition phases

Sol–gel transitions of GZ–PF127 copolymer solutions weredetermined by the vial inversion method using deionized water.All synthesized copolymers showed pH-dependent gelation behav-ior. After complete dissolution of the copolymers in water, an acidaqueous medium (pH 2–4.5) was obtained due to the hydrolysisof unreacted anhydride groups and the appearance of carboxylicacids. In these aqueous conditions of pH, the hydroxyl groups wereionized and the hydrogels could not be formed although the solu-tion temperature was increased until 50 ◦C.

It was necessary to add an appropriate amount of NaOH to formthe hydrogels. This change in pH was needed because a small pro-portion of the carboxylic groups of the polymer were dissociatedat low pH values, resulting in a flexible spiral. By adding a base,the dissociation of carboxylic groups was enhanced and an electro-static repulsion was created between charged regions. Moleculeswere expanded, the system was more rigid and gelation occurred.Gelations took place at a pH between 6 and 7.5 (near the physiolog-ical pH) in all cases. A change in the behavior of the hydrogels wasalso observed when the pH rose above 8.5. This pH increase did notproduce hydrogel gelation due to an excess of alkali and the disap-pearance of electrostatic charges when the carboxylic groups wereneutralized, as well as a viscosity loss. Moreover, it was seen that thegelation temperature remained constant if the pH varied between6 and 7.5. In all cases, gels reverted to solution upon cooling. Asshown in Fig. 2C, the system was liquid at 20 ◦C and, upon heatingaround 37 ◦C, the system in the test vial was physically gelled andcould not flow out even by inversion of the tube.

Fig. 2A shows that gel formation could be modulated by manip-ulation of the molar ratios and concentrations of the copolymers.It was observed that gelation occurred at lower weight percent-ages of the copolymers when molar ratios increased, e.g. for a 2molar ratio copolymer gelation occurred at a concentration of 7.5%

(w/v) whereas only a concentration of 2% (w/v) was needed inthe case of copolymers with a molar ratio of 10. Moreover, gela-tion temperatures tended to decrease when the GZm/PF127 ratioincreased till GZm/PF127 of 20, when gelation occurred at room

E. Moreno et al. / International Journal of Pharmaceutics 459 (2014) 1– 9 5

Fig. 2. (A) Influence of GZm/PF127 molar ratio and concentration in sol–gel transition phases. (B) Gelation temperatures as function of GZm/PF127 molar ratio and con-c .%) wag

twwctcwg

ptisc(t

3

mfmw(atep

F(

entration. (C) The GZ–PF127 composite (Gm/PF127 = 5 at a concentration of 2.5 wtel).

emperature (Fig. 2B). For example, a gelation temperature of 37 ◦Cas obtained for copolymers with GZm/PF127 = 2 at 7.5% (w/v)hile copolymers with a molar ratio of 10, at the same con-

entration, gelled at 26.2 ◦C. At a fixed GZm/PF127 value, sol–gelemperatures decreased at higher copolymer concentrations, e.g.opolymers with GZm/PF127 = 5 at 2.5% (w/v) gelled at 36.3 ◦Chereas if the weight percentage was increased until 10% (w/v),

elation occurred at 26.2 ◦C.Interestingly, gelation for the initial polymer GZ (Fig. 3A) and

hysical mixtures of these polymers (GZ and PF127) without reac-ion at a final concentration of 10% w/v (abbreviated as 10% mixn Fig. 3B and C) did not happen in a range between 20 and 50 ◦C,howing a different behavior than the corresponding synthesizedopolymers. Moreover, although there was an amount of free PF127Table 1), the new copolymers exhibited distinct properties thanhe PF127 alone.

.3. Viscosity studies

Gelation temperatures were confirmed by viscosity measure-ents. Fig. 3 presents the viscosity vs. temperature profile obtained

or the different copolymers. While GZ (10% w/v) and polymerixtures without reaction in a molar ratio of 2 and 5 (10%/v) attained a viscosity of 0 Pa·s at the temperatures tested

Fig. 3A–C), new copolymers presented a gradual viscosity increase

s a function of temperature. This increase clearly indicates thathe liquid solution turned into a gel in the vicinity of 37 ◦C. Asxpected, viscosities sharply increased around the gelation tem-erature, reaching higher values at high polymer concentrations

504030200

1

2

3

4 GZ(10%) PF127(25%)

Vis

cosi

ty (

Pa.

s)

T(ºC)

504030200

1

2

3

4 10% 10% mix

T(ºC)

BA

ig. 3. Viscosities vs. temperature profile at different concentrations: (A) for GZ and PF127D) GZm/PF127 = 10 copolymers. Physical mixtures of the polymers without reaction at 1

s fluid at 20 ◦C but was physically gelled at 37 ◦C (© indicates sol. and � indicates

(e.g. 3.5 Pa s for the GZm/PF127 = 5 copolymer at 10 wt.% vs. 1.2 Pa sat 3 wt.%, Fig. 3C). Therefore, it has been observed that the sys-tem reaches a point where the viscosity decreases although thetemperature increases. This could be due to the fact that therotational viscosimeter used in the determination breaks the gelcontinuously, being this effect more appreciable at higher viscosi-ties.

After determination of the sol–gel transition temperaturesand viscosities, the copolymer with GZm/PF127 of 5 was cho-sen for further investigation due to its properties. Purificationof the copolymer was carried out by spin centrifugation usingcentrifugal concentrator tubes at 5000 rpm for 30 min. Finally,the purified copolymer was obtained by lyophilization. It wasobserved that purified copolymer showed different behaviorthan the corresponding copolymer without purification and alsodifferent properties than its physical mixture, forming a gel atroom temperature. For this reason, the following studies weredone with the copolymer without purification.

3.4. In vitro release studies

BSA, rKPM-11 and dextran were used as examples of moleculeswith different molecular weights, molecular volumes and com-positions (a 66 KDa protein, a 11 KDa glycoprotein and 10 KDapolysaccharide, respectively) in order to investigate the release

behavior from GZm/PF127 = 5 hydrogels in PBS (pH 7.4) at 37 ◦C.Three distinct systems were prepared, two using the synthesizedhydrogel at two different concentrations (5 and 10% w/v) and acontrol hydrogel made of PF127 (at 25% w/v).

504030200

1

2

3

4 3% 5% 10% 10% mix

T(ºC)

504030200

1

2

3

4

C D

2% 5%

T(ºC)

polymers and hydrogels prepared with (B) GZm/PF127 = 2, (C) GZm/PF127 = 5 and0% (w/v) were abbreviated as 10% mix.

6 E. Moreno et al. / International Journal of Pharmaceutics 459 (2014) 1– 9

504030201000

20

40

60

80

100

PF127 (25%)BS

A r

elea

se (

%)

Time (h)

1508060402000

20

40

60

80

100

5% 10%B

SA

rel

ease

(%

)

Time (h)

30020010000

20

40

60

80

100 5% 10%

KM

P-1

1 re

leas

e (%

)

Time(h)

1508060402000

20

40

60

80

100

DCBA

5% 10%

Dex

tran

rel

ease

(%

)

Time (h)

F ◦ m PF1G

eafiiwgd

hcr3ssgGvtefbtd(hcrp

ttctdi

wwierg

ha

i(p

the negative control, more than 80% of cells treated with hydro-gels were viable. Furthermore, an increase in the concentration ofcopolymer (from 3 to 5% w/v) did not affect cell viability.

5(5%)5(3%)Control0

20

40

60

80

100

Cel

l via

bilit

y (%

)

ig. 4. In vitro release studies in PBS at 37 C. (A) Cumulative release of BSA froZm/PF127 = 5; 5 and 10 wt.%). The results represent the means ± SD for n = 2.

All the studied systems presented a sustained release, with thexception of the PF127 hydrogel loaded with BSA that showedlmost 70% cumulative release after 1 h (Fig. 4). A distinctive pro-le could be observed for the three molecules. At the initial time,

t is important to point out that a burst release of BSA and dextranas observed because of their rapid diffusion through the hydro-

els when the buffer solution was put. However, no burst effect wasetermined in the case of rKMP-11 samples.

Fig. 4A and B shows the release profiles of BSA from newydrogels (5 and 10% w/v) and PF127 (25% w/v). After 30 min, theumulative release of BSA was 23.1 and 18.6% for 5 and 10 wt.% gels,espectively. However, a 66.7% was obtained for PF127 samples.

h later, the cumulative release of BSA found for the synthe-ized hydrogels was around 35% while 86% was obtained for PF127ystems. After 24 h, a complete release was reached in PF127 hydro-els whereas lower percentages, 83.8 and 66.3%, were found forZ–PF127 hydrogels (at 5 and 10 wt.%, respectively). These obser-ations confirm that new synthesized hydrogels could sustainhe release of BSA more than uncrosslinked PF127. This can bexplained taking into account the nature of PF127 hydrogels. PF127orms gels by the weak hydrophobic interactions that take placeetween the poly(propylene oxide) blocks at elevated tempera-ures. Because of these weak interactions, hydrogels can be easilyissolved when they are in contact with a certain amount of waterCohn et al., 2006). On the contrary; the cross-linked GZ–PF127ydrogels may possess greater ability to preserve its structure andould retain the protein longer. Furthermore, quantitative (100%)elease of entrapped protein was observed, meaning that BSA is notrecipitated or aggregated.

The release of dextran at 37 ◦C appears in Fig. 4D. After 30 min,he releases for 5 and 10 wt.% hydrogels were 14.8 and 8.4%, respec-ively. A 52% release was reached within 24 h for both polymeroncentrations. Furthermore, dextran was released gradually upo 80–90% of the initial load. By quantifying the amount of residualextran, it was demonstrated that a fraction (10–20%) was trapped

nside the hydrogel network.At 37 ◦C and pH 7.4, the rKMP-11 diffused from the hydrogels

ith more difficulty (Fig. 4C). After 2 weeks, only a 35.5 and a 42.3%ere released, respectively. It was determined that a quantity of the

nitial protein (57–71%) was inside the hydrogels at the end of thexperiment. The recombinant protein rKMP-11 presents 6 histidineesidues in its terminal C that could strongly interact with free acidroups within the hydrogel hampering its release.

In contrast to expectations, it was observed that, although BSAas a higher molecular weight, it was released faster from hydrogelst all polymer concentrations than dextran or rKMP-11.

Another important fact is that the release was slightly quickern all cases when the concentration of the synthesized gel was 5%w/v). The hydrogel matrix was more consistent when it was pre-ared using a 10% (w/v) of the product. This fact slows the release

27 (25 wt.%). Cumulative release of (B) BSA, (C) rKMP-11 and (D) dextran from

because the gel matrix presents a more entangled structure, whichhinders the diffusion of the loaded macromolecule.

3.5. In vitro release studies from the hydrogels

Release profiles of the synthesized hydrogels loaded with BSAand dextran were fitted to different mathematical release models(Table 2). The results for the Korsmeyer–Peppas equation showedthat the release mechanism of BSA and dextran from hydrogelsprepared with a 10% (w/v) concentration could occur through dif-fusion because of the values found for the n parameter (ca. 0.5).On the other hand, the n values obtained for the release of thesecompounds from gels prepared with a 5% (w/v) concentration werelower than 0.5 so they were beyond the limits of Korsmeyer–Peppasmodel, probably due to observed faster release in the first part ofthe curves.rKMP-11 protein release curves were not fitted to anymathematical release models because its cumulative release wasfar from reaching the 100% and only a 40% of release was foundafter 14 days.

3.6. In vitro cytotoxicity on J774.2 macrophages

The effect of GZm/PF127 = 5 copolymer with a molar ratio of 5was assessed in vitro on J774.2 macrophages with the MTT assay.After 20 h, samples under test were removed and cell survival wasmeasured. The results (Fig. 5) indicated that, when compared with

Fig. 5. Cell viability (%) in J774.2 macrophages after incubation of 1.5 mL ofGZm/PF127 = 5 hydrogel at two different concentrations (3 and 5 wt.%). Cells in theabsence of hydrogel were used as control samples. The results were obtained fromthe MTT assay and represent the means ± SD for n = 4.

E. Moreno et al. / International Journal of Pharmaceutics 459 (2014) 1– 9 7

Table 2Constants of the Korsmeyer–Peppas and the Higuchi equations calculated from the in vitro release studies.

Korsmeyer–Peppas Higuchi

Formulation k × 10−6 (h−n) n R2 k × 10−6 (h−0.5) R2

Dextran (GZ/PF127 = 5; 5%) 19 ± 1 0.31 ± 0.01 0.998 12 ± 1 0.861 ± 0.0 ± 0.0 ± 0.0

3

witt3c

tit

(wtp

pdttio(miwr

o

Fsa

Dextran (GZ/PF127 = 5; 10%) 10 ± 1 0.48BSA (GZ/PF127 = 5; 5%) 27 ± 1 0.30BSA (GZ/PF127 = 5; 10%) 22 ± 2 0.51

.7. Gel integrity and biocompatibility in vivo

Different polymer solution volumes (molar ratio = 5; 5% v/w)ere subcutaneously injected into the ratsı back and, as shown

n Fig. 6A, at body conditions the polymer solution changed instan-aneously to a gel due to an increase of the temperature. Capsulehicknesses were measured every 2 days during the experiment.0 days after implantation, the capsule thicknesses had hardlyhanged (Fig. 6B and C).

Histopathological analysis at different time points revealed thathe capsule thickness was produced by the persistence of thenjected material and the infiltration of inflammatory cells duringhe course of a foreign body reaction to the hydrogel (Fig. 7).

As compared with negative control group (1.5 mL of PBS)Fig. 7A) tissue samples of mice injected with PF127 (1.5 mL; 25%/v) presented normal epidermis and dermis with discrete subcu-

aneous tissue fibrosis after 1 month of injection, considering thisolymer as well tolerated (Fig. 7B).

However, an acute inflammatory reaction was observed 1 dayost implantation of 50 �L of GZ–PF127 hydrogel. Epidermis andermis layers of the skin were normal while, in the subcutaneousissue, the injected material was surrounded by necrotic neu-rophils (Fig. 7C). The acute reaction was followed by sub-acutenflammatory signs 7 and 15 days after administration. At the sitef infection, there were rests of materials surrounded by a discreteafter 7 days, Fig. 7D) or more severe (after 15 days, Fig. 7E) halo of

ononuclear cells, macrophages and foreign-body giant cells. Themmune response was resolved 30 days after implantation (Fig. 7E)

ith the formation of a fibrous capsule around the almost negligibleemaining material.

It was observed that the inflammation intensity dependedf the administered volumes, being this inflammation higher at

ig. 6. (A) Subcutaneous gel implants after the injection of 50, 100, 250 and 500 �L, respite of injection 1 month post-implantation. (C) Implant areas after the injection of 1500

t 5 wt.% measured at different times after implantation (0, 1, 7, 14 and 30 days).

2 0.997 10 ± 1 0.9974 0.983 21 ± 2 0.8955 0.983 23 ± 1 0.987

higher volumes of injected hydrogel. Histopathological studiesafter the implantation of 500 �L of hydrogel showed a very intenseincrease in the subcutaneous tissue thickness due to an enor-mous chronic inflammatory response. A great inflammatory halowith mononuclear cells, macrophages and foreign-body giant cellsphagocyting the material was also observed at the different timesof sacrifice (Fig. 7G–J).

The degree of inflammation for tissues injected with 100, 250,750 and 1500 �L of the hydrogel was also strong and showed a deepforeign body reaction (data not shown).

4. Discussion

Pluronic® F127 is one of the most widely studied copolymers.However, its poor mechanical properties, high permeability tosolutes and short residence time have limited its use (He et al.,2008; Ruel Gariépy and Leroux, 2004). With the aim of improv-ing its characteristics, in the present work, we have developednovel thermosensitive hydrogels based in the reaction betweenPF127, used as cross-linker, and GZ. The reason for introducing GZwas to enhance the mechanical properties of PF127 as well as toincrease its bioadhesiveness. The reverse thermal gelation behaviorof PF127 has been attributed to temperature-dependant changesin water structure around the PPO blocks (Fusco et al., 2006; Heet al., 2008). This PPO blocks start to aggregate when tempera-ture increases and water becomes a poor solvent, reaching a gel.The new PF127-GZ hydrogels characterized in the current studypossess thermogelation properties that may enhance the biomed-

ical and pharmaceutical applications of PF127 alone (Glatter et al.,1994). Thus, a near instantaneous gelation was also observed. Thisproperty could prevent runoff and allow a more precise application.All the systems, with the exception of GZm/PF127 = 20 copolymer

ectively, of GZm/PF127 = 5 hydrogel at 5% (w/v). (B) Gel implants remained in the�L of PBS and PF127 (25 wt.%) and 50, 500 and 1500 �L of GZm/PF127 = 5 hydrogel

8 E. Moreno et al. / International Journal of Pharmaceutics 459 (2014) 1– 9

Fig. 7. Hematoxylin-eosin stains (30×). (A) Tissue injected with 1.5 mL of PBS 1 month after injection. (B) Tissue injected with 1.5 mL of PF127 at 25 wt.% 1 month post-i r 1, 7,o spect

trtdwoe

ittmvGach(GnrlrPtfft0crdref

irdchi

njection. (C–F) Tissues injected with 50 �L of GZm/PF127 = 5 hydrogel at 5 wt.% aftef GZm/PF127 = 5 hydrogel at 5 wt.% after 1, 7, 15 and 30 days post-implantation, re

hat gels at room temperature, underwent a thermodynamicallyeversible transition. They were flowing solutions at low tempera-ures and turned into gels at temperatures between 24 and 37 ◦C,epending on the concentration and molar ratio used (Fig. 2B). It isell known that PF127 gels at concentration >20% (w/v) whereas

nly a 2.5% (w/v) concentration of GZm/PF127 = 5 copolymer wasnough to produce a gel.

A feasible viscosity of a thermosensitive vehicle can play anmportant role in maintaining a non-flowing organization in bodyissues. Viscosity studies revealed a sharp viscosity increase aroundhe gelation temperatures in the samples tested (Fig. 3). Further-

ore, it is important to point out that, at 37 ◦C, similar viscosityalues were obtained for a 25% (w/v) PF127 solution and a 5% (w/v)Zm/PF127 = 5 copolymer solution (around 2 Pa s), which implies

noticeable reduction of the amount needed for the gelation pro-ess (Fig. 3A and B). Despite the similar viscosities, GZm/PF127 = 5ydrogel sustained the BSA release longer than PF127 systemFig. 4A and B). As compared with PF127, the cross-linked PF127-Z copolymers should form more stable and entangled hydrogeletworks, showing more mechanical strength to sustain BSA,KMP-11 and dextran release. For example, after 30 min, the cumu-ative release of BSA was 23.1 and 18.6% for 5 and 10 wt.% gels,espectively (Fig. 4A and B). However, a 66.7% was obtained forF127 samples. 3 h later, the cumulative release of BSA found forhe synthesized hydrogels was around 35% while 86% was obtainedor PF127 systems (Fig. 4A and B). In deep, BSA and dextran releaserom hydrogels prepared with a 10% (w/v) concentration was con-rolled by a diffusion mechanism as reflected the n values (ca..5) obtained with the Korsmeyer–Peppas model (Table 2). Loweropolymer concentration (5% w/v) showed more noticeable burstelease effect at the beginning. V.g. 5% (w/v) hydrogels presented aextran cumulative release of 28% after 1 h whereas the cumulativeelease in hydrogels at 10% (w/v) was 15%. The observation couldxplain the n values <0.45 obtained on fitting the release studiesrom 5 wt.% GZ–PF127 hydrogels to the Korsmeyer–Peppas model.

Thereafter, we evaluated the biocompatibility of the hydrogelsn vitro and in vivo. MTT assays addressed with J447.2 macrophagesevealed that, when GZ is crosslinked at a 5 molar ratio, gel extracts

id not significantly reduce cell viability and demonstrated goodytocompatibility (Fig. 5). In vivo, the subcutaneous injection of theydrogels elicits in a sequential fashion acute, subacute, chronic

nflammatory responses and, finally, a foreign body reaction whose

15 and 30 days post-implantation, respectively. (G–J) Tissues injected with 500 �Lively.

severity at high volumes of injection could compromise its biomed-ical application as sustained release devices (Fig. 7).

However, the inflammatory response produced after the admin-istration of 50 �L hydrogel with the presence of a discretegranuloma 1 month later (Fig. 7F) could be exploited to elicit spe-cific immune responses against the protein loaded in the hydrogel(i.e. rKMP-11). The inflammation and innate immune response isbehind of mechanism of action of all known vaccine adjuvants(Levitz and Golenbock, 2012). Therein, rKMP-11 is a surface proteinfound in the cytoplasmic membrane of Leishmania and is expressedboth in the promastigote and amastigote forms (Matos et al., 2010).Importantly, vaccination with rKMP-11 was able to prevent diseasedevelopment in different experimental models of leishmaniasis(Bhaumik et al., 2009).

Furthermore, by varying the GZ-Pluronic F127 molar ratio andconcentration we can manipulate the gelation temperature (Fig. 2).The hydrogels that gel at 32–34 ◦C could find application after top-ical or mucosal administration.

5. Conclusions

The current study details the development of new thermosensi-tive hydrogels from GZ and PF127. Both polymers were cross-linkedwith the aim to improve the weak mechanism properties ofPF127 hydrogels. Viscosity testing demonstrated that GZ–PF127hydrogels exhibit excellent flexible and thermally induced gellingproperties. GZ–PF127 was a solution at room temperature andbecame a very rapid reverse thermal gelation and also a sharpincrease in viscosity upon heating in the vicinity of 37 ◦C at lowerconcentrations than PF127 alone (v.g. 5% vs. 25%). This fact couldbe advantageous for maintaining the solution form before applica-tion to the human body and allow their ease of injection. We alsodemonstrated that by varying the molar ratio and concentration ofcopolymers we could modify the gelling temperature and viscosityof the resulting hydrogels. GZ–PF127 composites allow sustainedrelease of proteins and induce an inflammatory response that canbe useful for vaccination purposes.

Acknowledgments

We would like to thank the ADA Foundation (University ofNavarra) for the grant that was awarded to J. S. and the Tropical

ourna

HA1

R

A

A

A

A

A

A

B

C

C

C

C

D

E

F

F

G

G

E. Moreno et al. / International J

ealth Institute and FIMA (Fundación para la Investigación Médicaplicada) for supporting this work. The recombinant protein rKMP-1 was kindly provided by Dr. M. Soto (CBM-UAM, Madrid, Spain)

eferences

güeros, M., Zabaleta, V., Espuelas, S., Campanero, M.A., Irache, J.M., 2010. Increasedoral bioavailability of paclitaxel by its encapsulation through complex formationwith cyclodextrins in poly(anhydride) nanoparticles. J. Control. Release 145, 2–8.

lexandridis, P., Athanassiou, V., Hatton, T.A., 1995. Pluronic-P105 PEO-PPO-PEOblock copolymer in aqueous urea solutions: micelle formation, structure andmicroenvironment. Langmuir 11, 2442–2450.

ndrade Acevedo, R., Machon, L., Chávez, N., 2009. Effectiveness of a mouthwashcontaining Triclosan and Gantrez in the reduction of biofilm and gingivitis: aclinical pilot study. J. Contemp. Dent. Pract. 10, E033–E040.

ppel, E.A., Loh, X.J., Jones, S.T., Dreiss, C.A., Scherman, O.A., 2012. Sustained releaseof proteins from high water content supramolecular polymer hydrogels. Bioma-terials 6, 4646–4652.

rbós, P., Wirth, M., Arangoa, M.A., Gabor, F., Irache, J.M., 2002. Gantrez® AN as anew polymer for the preparation of ligand-nanoparticle conjugates. J. Control.Release 83, 321–330.

rbós, P., Campanero, M.A., Arangoa, M.A., Renedo, M.J., Irache, J.M., 2003. Influenceof the surface characteristics of PVM/MA nanoparticles on their bioadhesiveproperties. J. Control. Release 89, 19–30.

haumik, S., Basu, R., Sen, S., Naskar, K., Roy, S., 2009. KMP-11 DNA immunization sig-nificantly protects against L. donovani infection but requires exogenous IL-12 asan adjuvant for comparable protection against L. major. Vaccine 27, 1306–1316.

heaburu, C., Ciocoiu, O., Staikos, G., Vasile, C., 2013. Thermoresponsive sodiumalginate-g-poly(N-isopropylacrylamide) copolymers. III. Solution properties. J.Appl. Polym. Sci. 127, 3340–3348.

hun, K.W., Lee, J.B., Kim, S.H., Park, T.G., 2005. Controlled release of plasmid DNAfrom photo-cross-linked Pluronic hydrogels. Biomaterials 26, 3319–3326.

ohn, D., Lando, G., Sosnik, A., Garty, S., Levi, A., 2006. PEO–PPO–PEO-basedpoly(ether ester urethane)s as degradable reverse thermo-responsive multi-block copolymers. Biomaterials 27, 1718–1727.

osta, P., Manuel, J., Lobo, S., Sousa Lobo, J.M., 2001. Modeling and comparison ofdissolution profiles. Eur. J. Pharm. Sci. 13, 123–133.

ouglas, T.E.L., Skwarczynska, A., Modrzejewska, Z., Balcaen, L., Schaubroeck, D.,Lycke, S., Vanhaecke, F., Vandenabeele, P., Dubruel, P., Jansen, J.A., Leeuwen-burgh, S.C.G., 2013. Acceleration of gelation and promotion of mineralization ofchitosan hydrogels by alkaline phosphatase. Int. J. Biol. Macromol. 56, 122–132.

scobar Chávez, J.J., López Cervantes, M., Naik, A., Kalia, Y.N., Quintanar Guerrero, D.,Ganem Quintanar, A., 2006. Applications of thermo-reversible Pluronic F-127gels in pharmaceutical formulations. J. Pharm. Pharm. Sci. 9, 339–358.

ang, J., Hsu, S., Leu, Y., Hu, J., 2009. Delivery of cisplatin from Pluronic co-polymersystems: liposome inclusion and alginate coupling. J. Biomat. Sci. Polym. E 20,1031–1047.

usco, S., Borzacchiello, A., Netti, P., 2006. Perspectives on: PEO–PPO–PEO triblockcopolymers and their biomedical applications. J. Bioact. Compat. Polym. 21,149–164.

latter, O., Scherf, G., Schillen, K., Brown, W., 1994. Characterization of apoly(ethylene oxide)-poly(propylene oxide) triblock copolymer (EO27-PO39-

EO27) in aqueous solution. Macromolecules 27, 6046–6054.

uo, D., Xu, C., Quan, J., Song, C., Jin, H., Kim, D., Choi, Y., Cho, M., Cho, C., 2009.Synergistic anti-tumor activity of paclitaxel-incorporated conjugated linoleicacid-coupled poloxamer thermosensitive hydrogel in vitro and in vivo. Bioma-terials 30, 4777–4785.

l of Pharmaceutics 459 (2014) 1– 9 9

He, C., Kim, S.W., Lee, D.S., 2008. In situ gelling stimuli-sensitive block copolymerhydrogels for drug delivery. J. Control. Release 127, 189–207.

Hsu, S., Leu, Y., Hu, J., Fang, J., 2009. Physicochemical characterization and drugrelease of thermosensitive hydrogels composed of a hyaluronic acid/Pluronicf127 graft. Chem. Pharm. Bull. 57, 453–458.

Jeong, B., Bae, Y.H., Kim, S.W., 2000. Drug release from biodegradable injectable ther-mosensitive hydrogel of PEG-PLGA-PEG triblock copolymers. J. Control. Release63, 155–163.

Lee, S., Tae, G., 2007. Formulation and in vitro characterization of an in situ gelable,photo-polymerizable Pluronic hydrogel suitable for injection. J. Control. Release119, 313–319.

Levitz, S., Golenbock, D., 2012. Beyond empiricism: informing vaccine developmentthrough innate immunity research. Cell 148, 1284–1292.

Liu, C., Gong, C., Pan, Y., Zhang, Y., Wang, J., Huang, M., Wang, Y., Wang, K., Gou, M.,Tu, M., Wei, Y., Qian, Z., 2007. Synthesis and characterization of a thermosensi-tive hydrogel based on biodegradable amphiphilic PCL-Pluronic (L35)-PCL blockcopolymers. Colloid. Surf. A 302, 430–438.

Matos, D.C.S., Faccioli, L.A.P., Cysne Finkelstein, L., De Luca, P., Corte Real, S.,Armôa, G.R.G., Lemes, E.M.B., Decote-Ricardo, D., 2010. Kinetoplastid membraneprotein-11 is present in promastigotes and amastigotes of Leishmania amazo-nensis and its surface expression increases during metacyclogenesis. Mem. Inst.Oswaldo Cruz 105, 341–347.

Moore, T., Croy, S., Mallapragada, S., Pandit, N., 2000. Experimental investigation andmathematical modeling of Pluronic F127 gel dissolution: drug release in stirredsystems. J. Control. Release 67, 191–202.

Park, K., Lee, S., Joung, Y., Na, J., Lee, M., 2009. Thermosensitive chitosan-Pluronichydrogel as an injectable cell delivery carrier for cartilage regeneration. ActaBiomater. 5, 1956–1965.

Peppas, N., Sahlin, J., 1989. A simple equation for the description of solute release.III. Coupling of diffusion and relaxation. Int. J. Pharm. 57, 169–172.

Porfire, A., Zabaleta, V., Gamazo, C., Leucuta, S., Irache, J., 2010. Influence of dextranon the bioadhesive properties of poly(anhydride) nanoparticles. Int. J. Pharm.390, 37–44.

Ritger, P.L., Ritger, P., Peppas, N., 1987. A simple equation for description ofsolute release I. Fickian and non-fickian release from non-swellable devicesin the form of slabs, spheres, cylinders or discs. J. Control. Release 5,23–36.

Ruel Gariépy, E., Leroux, J., 2004. In situ-forming hydrogels-review of temperature-sensitive systems. Eur. J. Pharm. Biopharm. 58, 409–426.

Ryu, J., Lee, Y., Kong, W., Kim, T., Park, T., Lee, H., 2011. Catechol-functionalizedchitosan/Pluronic hydrogels for tissue adhesives and hemostatic materials.Biomacromolecules 12, 2653–2659.

Sosnik, A., Cohn, D., 2004. Ethoxysilane-capped PEO–PPO–PEO triblocks: a new fam-ily of reverse thermo-responsive polymers. Biomaterials 25, 2851–2858.

Trudel, J., Massia, S.P., 2002. Assessment of the cytotoxicity of photocrosslinkeddextran and hyaluronan-based hydrogels to vascular smooth muscle cells. Bio-materials 23, 3299–3307.

Wang, X., Li, X., Li, Y., Zhou, Y., Fan, C., Li, W., Ma, S., Fan, Y., Huang, Y., Li, N., Yu,L., Zhang, Z., Zhang, H., Ding, J., 2010. Biodegradability and biocompatibility ofthermoreversible hydrogels formed from mixing a sol and a precipitate of blockcopolymers in water. Biomacromolecules 11, 2169–2178.

Yoncheva, K., Calleja, P., Agüeros, M., Petrov, P., Miladinova, I., Tsvetanov, C., Irache,J., 2012. Stabilized micelles as delivery vehicles for paclitaxel. Int. J. Pharm. 436,258–264.

You, S., Van Winkle, D., 2010. Single molecule observation of DNA electrophoresisin Pluronic F127. J. Phys. Chem. B 114, 4171–4177.

Yu, L., Zhang, Z., Zhang, H., Ding, J., 2010. Biodegradability and biocompatibility ofthermoreversible hydrogels formed from mixing a sol and a precipitate of blockcopolymers in water. Biomacromolecules 11, 2169–2178.