a novel method for hydrogel nanostructuring

ruc,d, Hechno, Haifagy, Hanology

Keywords:Hydrogel

drogeft maolvin

as well as its swelling and mechanical properties. Small Angle X-ray Scattering (SAXS) and

reasinructu

after synthesis in order to create a nanoporous structure[4,5]. The incorporation of nanometric scaled structures

gree of hyellar lyotrdiacrylate

properties including network swelling, mechanics, anddegradation [7]. Nanostructuring of hydrogels during theirformation has recently been reported by our group [8]. Thisapproach utilizes the self-assembly ability of biocompati-ble, amphiphilic block-copolymers of poly(ethyleneoxide)/poly(propylene oxide) (Pluronic) in order to

0014-3057/$ - see front matter 2014 Elsevier Ltd. All rights reserved.

Corresponding author at: Department of Chemical Engineering,Technion-Israel Institute of Technology, Haifa 32000, Israel. Tel.: +972 48293588; fax: +972 4 8295672.

E-mail address: [email protected] (H. Bianco-Peled).

European Polymer Journal 52 (2014) 137145

Contents lists available at ScienceDirect

European Poly

journal homepage: www.else

MACR

OMOLECU

LARNANOTE

CHNOLO

GYhttp://dx.doi.org/10.1016/j.eurpolymj.2014.01.004(PCL-b-PEO-b-PCL) hydrogels were created by PCL removal DA) hydrogels was found to impact their physicalcopolymers driven by hydrophobic interactions betweenthe blocks [13]. For example, nanostructured poly(e-cap-rolactone)-poly(ethylene oxide)-poly(e-caprolactone)

micelles was found to reduce the deweight gain. Combining hexagonal or lamquid crystals with poly(ethylene glycol)drogelopic li-(PEG-control over the material characteristics. In principle, byoptimizing the nanostructure, materials tailor-made for aspecic application can be created with structural deni-tion to a sub-cellular level. The development of hydrogelnanostructuring methods are still challenging becausethe high water content excludes the use of lithographictechniques. One approach for nanostructuring hydrogelsutilizes the self-assembly capability of block and graft

core cross-linked poly(ethylene glycol)-block-poly(e-cap-rolactone) (PEG-b-PCL) micelles having spherical or rod-like morphologies were likewise prepared and evaluatedfor use as drug-eluting soft contact lenses [6]. Integrationof micelles with crosslinked cores into pHEMA hydrogelsled to the formation of different internal nanostructureswhich were dependent on the amount and morphologyof the embedded micelles. Incorporation of cross-linkableStructureSAXSNanostructuring

1. Introduction

Nanostructuring is becoming incthe design of precisely dened 3D stCryogenic Transmission Electron Microscopy (cryo-TEM) revealed that photochemicalcrosslinking of the hydrogel caused immobilization of the nanostructured micelles.Mechanical and weight gain experiments demonstrated a signicant impact of these nano-structures on the hydrogels elastic modulus as well as the transient and equilibriumweight gain ability of the material.

2014 Elsevier Ltd. All rights reserved.

gly important inres which enable

with different morphologies into the hydrogels contrib-uted to the control over the hydrogel nanostructure,mechanical and physical properties [68]. Poly(2-hydroxy-ethyl methacrylate) (pHEMA) hydrogels incorporated withAccepted 5 January 2014Available online 11 January 2014

lated blockcopolymer molecules, which enable the attachment of these micelles to thehydrogel matrix through their endgroups. This design impacts the hydrogel nanostructureMacromolecular Nanotechnolgy

A novel method for hydrogel nanost

Ortal Yom-Tov a, Ilya Frisman b, Dror Seliktar c

a Inter-Departmental Program for Biotechnology, Technion-Israel Institute of TbDepartment of Chemical Engineering, Technion-Israel Institute of TechnologycDepartment of Biomedical Engineering, Technion-Israel Institute of Technolod The Russell Berrie Nanotechnology Institute, Technion-Israel Institute of Tech

a r t i c l e i n f o

Article history:Received 25 August 2013Received in revised form 17 December 2013

a b s t r a c t

Nanostructured hycreation of novel sonanostructuring invturing

avazelet Bianco-Peled b,d,logy, Haifa 32000, Israel32000, Israel

ifa 32000, Israel, Haifa 32000, Israel

ls tailor-made for specic applications are new grounds for theterials. The current research presents a new method for hydrogelg the incorporation of Pluronic F127 micelles mixed with acry-

mer Journal

vier .com/locate /europol j

fact, the gradual escape of Pluronic molecules from the

138 O. Yom-Tov et al. / European Polymer Journal 52 (2014) 137145

MACR

OMOLECU

LARNANOTE

CHNOLO

GYF127-DA and PEG-DA were prepared from Pluronic

F127 (BASF) and poly(ethylene glycol) (PEG, molecularweight 10 kDa), respectively, as described elsewhere [9].Briey, acrylation was carried out under Argon by reactingthe polymers with Acryloyl-chloride (Merck, Darmstadt,Germany) and triethylamine (TEA) (Fluka) at a molar ratioof 150% relative to the hydroxyl groups. The resultingproduct was precipitated and dried under vacuum for 48 h.

Preparation of PEG-brinogen (PF) precursor solution isdescribed elsewhere [8]. Briey, bovine plasma brinogen(SigmaAldrich) was dissolved in phosphate buffer saline(PBS) containing 8 M Urea. Tris (2-carboxyethyl) phos-phine hydrochloride (TCEP) (SigmaAldrich) was addedto completely dissolve the protein dissolution, the pHwas adjusted to 8, and PEG-DA (10 kDa) in PBS solutionwas added while maintaining a ratio of 4:1 PEG-DA to pro-tein cysteines. After the reaction was completed, the prod-uct was diluted, precipitated, redissolved, homogenizedand dialyzed against 150 mM PBS at 4 C for 24 h withtwo changes of PBS (Spectrum, 1214-kDa MW cutoff, Cal-ifornia, USA). Protein concentration was determined byBis-Cinchoninic Acid (BCA) protein assay.

2.2. Nanostructured PF hydrogels

Mixtures of Pluronic F127 and F127-DA at different ra-tios were prepared with PF precursor solution (proteinhydrogel resulted in a signicant change to the nanostruc-ture of the polymer network.

The main objective of the current study was to design anovel method for nanostructuring of hydrogels. This meth-od is based on embedding Pluronic micelles in a hydrogelwhile anchoring some of their molecules to the surround-ing network through their endgroups. Our hypothesis wasthat in this design, the anchored molecules could provide ameans to further crosslink the hydrogel and stabilize thenetwork. In contrast to traditional low molecular weightcrosslinkers, the covalently bound molecule can stretchas the unbound Pluronic diffuse out of the hydrogel andthe micellar structures are lost. We postulated that this un-ique design will allow further manipulation of the gelproperties. Herein we describe a comprehensive analysisof hydrogels prepared using this new nanostructuringmethodology. Small angle X-ray scattering (SAXS) andtransmission electron microscopy at cryogenic tempera-ture (cryo-TEM) were used for structural characterization,whereas mechanical and weight gain experiments wereused to explore the impact of nanostructure alterationson these properties.

2. Experimental

2.1. Synthesis of PEG diacrylate (PEG-DA) and F127 diacrylate(F127-DA)impart nanostructured organization in the hydrogels dur-ing photopolymerization. Limited stability of these mi-celles was observed, mostly attributed to their diffusionout of the hydrogels after four days in aqueous buffer. Inconcentration 8.5 0.5 mg/ml) at 4 C until complete dis-solution of the block-coloymers was achieved. Total con-centration of the block-copolymer (Pluronic F127 plusF127-DA) was kept constant at 10% (w/v). The solutionwas mixed with 0.1% (v/v) photoinitiator stock solutioncontaining 10% (w/v) Irgacure2959 in 70% ethanol and30% deionized water. The hydrogel precursor solutionwas heated to 37 C for 10 min in order to induce micelleformation, followed by irradiation with UV light (365 nm,45 mW/m2) for 5 min in order to achieve a chemicallycrosslinked hydrogel.

Small Angle X-ray Scattering (SAXS) experiment wereperformed using a small-angle diffractometer (MolecularMetrology SAXS system) with Cu Ka radiation from asealed microfocus tube (MicroMax-002+S), two Gbel mir-rors, and three-pinhole slits (generator powered at 45 kVand 0.9 mA). The scattering patterns were recorded by a20 20 cm two-dimensional position sensitive wire detec-tor (gas lled proportional type of Gabriel design with200 lm resolution) that was positioned 150 cm behindthe sample. The resolution of the SAXS system was approx-imately 23 nm1. The scattered intensity I(q) was re-corded in the interval 0.07 < q < 2.7 nm1, where q is thescattering vector dened as q = (4p/k)sin(h), 2h is the scat-tering angle, and k is the radiation wavelength(0.1542 nm). The sample under study was sealed in athin-walled glass capillary of about 2 mm diameter and0.01 mm wall thickness, and measured under vacuum atconstant temperature. The I(q) was normalized to the fol-lowing parameters: time, solid angle, primary beam inten-sity, capillary diameter, transmission, and the Thompsonfactor. Scattering from the solvent, empty capillary andelectronic noise were subtracted. SAXS curves were mea-sured at q vs. I, where I has the units of (1/nm3).

Cryogenic Transmission Electron Microscopy (cryo-TEM)micrographs were obtained from ultra-fast cooled vitriedcryo-TEM specimen prepared under controlled conditions37 C and 100% relative humidity as described elsewhere[10]. Specimens were examined in a Philips CM120 cryo-TEM operating at 120 kV, using an Oxford CT3500 cool-ing-holder system that kept the specimens at about180 C. Low electron-dose imaging was performed witha Gatan Multiscan 791 CCD camera, using the Gatan DigitalMicrograph 3.1 software package.

Samples for water weight gain experiments were pre-pared by using round Teon molds with diameter of14 mm, whereby PF hydrogels with the addition of Pluron-ic F127/F127-DA mixtures at different ratios were tested.The control groups included PF hydrogels with the additionof PEG-DA at different percentages. Each sample was madeby transferring precursor solution (0.5 ml) into the Teonmold, heating the solution to 37 C for 10 min and thencrosslinking the hydrogel with UV light for 5 min as de-scribed before. The hydrogels were subsequently sub-merged in 150 mM PBS containing 0.2% sodium azide inPetri dishes. The dishes were incubated at 37 C and thewater weight gain ratio was determined gravimetricallyas follows:

%Weight gain mW mDmW

100 1

in the amount of molecules bound to the matrix and con-

aggregation is likely owing to the nature of the PF

equation describing the scattering as a cylinder withlength L and radius of gyration Rg [14]:

Icyl KcylLpq eq2R2g2 2

where Kcyl is a constant.Pluronic F127 at concentration of 10% (w/v) was previ-

ously shown to form closely packed micelles [1517]. Be-cause the corona of these micelles consisted of PEGchains with low contrast, the scattering from the micellesImicelle is calculated by multiplying the form factor P(qR),which describes the shape of a spherical core with radiusR, and the structure factor S(qRHS) which describes particleinterference with interaction radius of RHS between neigh-boring hard spheres. The PercusYevick approximation forhard spheres was used to calculate the structure factor[18]. Taking both the form factor and the structure factorinto account, the following equations were used to de-scribe scattering from Pluronic F127 micelles, shownschematically in Fig. 1B [18]:

Imicelle u 4p3 R3 Dq2 PqR SqRHS 3

where u is the volume fraction of the cores and Dq is theelectron density difference between the core and the med-ium. The shape factor is given by [18]

PqR 3 sinqR qR cosqRqR3

" #24

O. Yom-Tov et al. / European Polymer Journal 52 (2014) 137145 139

MACR

OMOLECU

LARNANOTE

CHNOLO

GYprecursor, a semi-synthetic molecule composed of ahydrophobic Fibrinogen backbone with grafted hydro-philic PEG chains [13]. The grafted chains on the PF carrysaturated endgroups that allow network formation viaphotopolymerization reaction.

Our previous study [12] demonstrated that SAXS curvesonly reect the contribution of the brinogen backboneand not that of the attached PEG chains because the elec-tron density of PEG (3.45 107 3) is almost identical tothat of the solvent (water, 3.34 107 3). The scatteringpatterns of the PF solution alone was thus tted to ansequently, the alteration of stable nanostructural featuresthat resulted in the hydrogels, including mesh size andmicellar arrangement.

3.1. Precursor solutions with added Pluronic F127/F127-DAmixtures

Precursor solutions containing Pluronic F127/F127-DA at different ratios were investigated in order to ob-serve if the block-copolymer mixtures self-assemble at37 C in a similar fashion to solutions of precursor with10% (w/v) unreactive Pluronic F127. The structure ofthe three-component PF/Pluronic F127/F127-DA systemis rather complex, nevertheless the structure of each ofthe individual components was previously subject to asubstantial analysis. Our previous SAXS and SANS analysisof PF precursor showed that these molecules aggregatedin solution due to proteinprotein interactions, andformed elongated cylinder-like objects with grafted PEGchains [12], as schematically illustrated in Fig. 1A. ThisMechanical characterization was performed using aLloyd Tensile machine in compression mode. The samplepreparation protocol, compositions and size were identicalto the ones used in the swelling experiments. Thesamples were compressed at a rate of 1 mm/min. TheYoungs modulus was calculated from the slope between1% and 10% extensions in the linear region of the stressstrain curve [11]. The average values and standard devia-tions were obtained from the analysis of at least 8measurements.

3. Results and discussion

The current research explores the relationship betweenarchitecture and properties using immobilized Pluronic

F127 micelles in PEG-brinogen hydrogels. Covalent link-age of the micelles into the matrix was achieved by replac-ing part of the Pluronic F127 molecules with an acrylatedderivative, F127-DA, which can react with the PF precursormolecule during the free-radical polymerization reactionthat leads to hydrogel formation. In order to isolate theinuence of Pluronic F127 acrylation from other parame-ters, such as micelle size and density, the overall block-copolymer concentration was kept constant at 10% (w/v).Therefore, the main variable in the experimental designof this study was the ratio between Pluronic F127 andF127-DA. This experimental setup allowed for variationsFig. 1. (A) Schematic representation of aggregated PF. (B) Schematicrepresentation of Pluronic F127 micelles. (C) Normalized scatteringintensity curves of PF hydrogels with (s) 2.5% F127-DA and 7.5%Pluronic F127, (D) 5% F127-DA and 5% Pluronic F127, (e) 7.5% F127-DA and 2.5% Pluronic F127 and (h) 10% F127-DA. The intensity curveswere multiplied by a factor for better visualization. Fits to Eq. (11),calculated from the best-t parameters summarized in Table 1, are shownas solid lines.

in the micrographs of these materials, as can be seen inFig. 2.

3.2. Nanostructured hydrogels with added Pluronic F127/F127-DA mixtures

Nanostructured PF hydrogels containing Pluronic

F127/F127-DA at various ratios were studied with the pur-pose of determining whether the attachment of micelles tothe hydrogel alters the nanostructures. SAXS patterns ofthe nanostructured hydrogels are shown in Fig. 3A.Increasing the relative percentage of F127-DA resulted ina smearing and a shifting of the inter micellar peak tosmaller scattering angles. This outcome hinted to analteration in the nanostructure of the hydrogels (aftercrosslinking) which could be ascribed to the binding ofthe micelles onto the PF network. With an increase in the


MACR

OMOLECU

LARNANOTE

CHNOLO

GYThe structure factor takes the form:

SqRHS 1

1 24hbGA=A 5

where hb is the volume fraction of the hard spheres and

A 2qRHS 6

GAa sinAAcosAA2

b 2AsinA2A2cosA2

A3

cfA4 cosA43A26cosAA36AsinA6g

A5

7

a 1 2hb2

1 hb48

b 6hb 1 hb=22

1 hb29

c ha2

10

Finally, the scattering from the PF solution containingPluronic F127/F127-DA is described by summing Eqs.(2) and (3):

Iq Icyl Imicelle

KcylLpq eq2R2g2 u 4p

3R3 Dq2 PqR SqRHS 11

A good t of Eq. (11) to the experimental SAXS data(Fig. 1C) conrmed that micelles are formed in solutionat 37 C, as initially assumed. Close packing of the micellesis evident from the structural peak observed at small scat-tering angles. As can be expected from the similarities be-tween the scattering patterns, there were nodistinguishable differences between the calculated param-eters of the different solution samples, which indicatedthat altering the ratio between Pluronic F127 and F127-DA has no remarkable inuence on the nanostructure.The tted core radius is about 80 , in agreement with pre-viously reported values for Pluronic F127 in water at37 C [10].

Cryo-TEM was used to directly observe micelle forma-tion in solutions of PF containing F127. Because the con-trast between Pluronic F127 molecules to thebackground was very low, increased exposure time of theirradiated electron beam was applied to the tested sam-ples. This technique was previously used in cryo-TEM stud-ies of Pluronic micelles to enhance contrast [10]. In orderto ensure that the nanostructures in the sample were notas a result of damage caused to the sample by overexpo-sure to radiation, short exposure times were also used incontrol measurements. The selective electron beam etch-ing exposed micellar structures in the PF samples contain-ing Pluronic F127. Because the PF solutions containing thedifferent percentages of F127-DA demonstrated similarscattering patterns, only PF solution with 10% w/v addedF127-DA was studied by Cryo-TEM as a representativeexample. A well-ordered micelles structure was observedFig. 2. Cryo-TEM images of PF solution with 10% F127-DA at magnica-tion of (A) X66 and (B) X175.

When tting the experimental SAXS data of the PF andF127-DA materials to Eq. (11), we found that the modelwhich does not account for any variations in the cylindersdimensions did not result in as good a t to this data whencompared to the t for PF materials alone. In contrast, asimilar model with variations in the cylinders dimensionsresulted in a much better t of the experimental SAXS datafor the PF and F127-DA materials. Therefore, it was as-sumed that the PF protein size variations are related tothe increasing concentrations of F127-DA in the precursorsolution [12]. This could be attributed to the aggregation ofthe protein backbone occurred during crosslinking; conse-quently two populations of proteins with different sizesare now present: one population preserves the original ra-dius of approximately 2 nm and a second populationwhose size varies as a function of crosslinker type and con-centration. For simplicity, the revised model approximated


Yrelative F127-DA concentration, the percentage of boundmicelles to the hydrogel increased and as a result, the dis-tance between the micelles increased and their degree ofordering was reduced.

Fig. 3. (A) Normalized scattering intensity curves of PF hydrogelscontaining 10% block copolymer at various Pluronic F127/F127-DAratios (s) 2.5% F127-DA and 7.5% Pluronic F127, (D) 5% F127-DA and 5%Pluronic F127, (e) 7.5% F127-DA and 2.5% Pluronic F127 and (h) 10%F127-DA. The intensity curves were multiplied by a factor for bettervisualization. Fits to Eq. (11), calculated from the best-t parameterssummarized in Table 1, are shown as solid lines. (B) Normalizedscattering intensity curves of PF hydrogels with (e) 1%PEG-DA, 1.5% (h)PEG-DA and (D) 2% PEG-DA. The intensity curves were multiplied by afactor for better visualization. Fits to Eq. (2), calculated from the best-tparameters summarized in Table 1, are shown as solid lines.

Table 1Best t parameters for the SAXS data shown in Fig. 3.

F127-DA (%w/v) 2.5% 5%

PF hydrogels with Pluronic F127/F127-DA mixtures, total blockcopolymer conceRH () 96.1 0.3 101.1 hb 0.289 0.002 0.253 R () 89.4 0.4 91.9 R1cyl () 17.0 0.3 16.4 R2cyl () 27.2 0.5 30.6 Dq2micelle (e

6) [15] 6.25 104 6.25 Dq2cyl (e

6) [12] 0.106 0.106

PEG-DA (%w/v) 1%

PF hydrogels with PEG-DAR1cyl () 15.9 0.4R2cyl () 37.4 0.8Dq2 (e 6) [12] 0.106

MACR

OMOLECU

LARNANOTE

CHNOLO

Gthe cylinder size distribution using two populations of cyl-inders with different radii, and Eq. (11) was adjustedaccordingly. The best t parameters to the revised Eq.(11) are summarized in Table 1, and supported the qualita-tive observations made based on the shape of the curves.Increasing the relative percentage of F127-DA interfereswith the ability of the Pluronic F127 molecules to self-assemble into spherical micelles. As a result, the distancebetween neighboring existing micelles RHS increased andtheir volume fraction hb decreased. Comparing the tparameters to those obtained from the corresponding solu-tions showed that the size of the protein backbone wasalso affected by the chemical crosslinking reaction (i.e.photopolymerization). Specically, the fact that two differ-ent sizes of cylinders with different radii had to be used toachieve a better t to the experimental SAXS data under-scores the observations regarding F127-DA. Moreover, lar-ger aggregates were evidently formed when more F127-DAwas present. Thus, enhancing the micelle stability byimmobilizing them in the PF network resulted in furtheraggregation of the protein backbone and the formation oflarger, more size dispersed aggregates. As stated above,additional changes to the network due to the increasedpercentage of micelles bound to the PF network includelarger distances between the micelles and reduced degreeof ordering. These main ndings are summarized in the

7.5% 10%

ntration 10 w/v%0.1 109 1 125 10.002 0.205 0.003 0.197 0.002

0.2 94.6 0.8 85.3 0.30.3 17.2 0.3 18.0 0.20.4 39.1 0.4104 6.25 104 6.25 104

0.106 0.106

1.5% 2%

14.3 0.6 14.3 0.433.3 0.4 33.3 0.20.106 0.106

hypothesized structure of the nanostructured PF hydrogelsshown schematically in Fig. 4.

Cryo-TEMmicrographs of crosslinked PF hydrogels con-taining F127-DA (Fig. 5) supported the hypothesis that theinuence of F127-DA on the hydrogel nanostructure is pro-nounced only after photopolymerization of these solutions.This is despite the fact that the nanostructure of the PFwith F127-DA was not profoundly affected in the precursorsolutions. The crosslinked specimens demonstrated dis-tinct regions with ordered micelles, which co-existed withregions of randomly distributed micelles. The density ofthe micelles in the random regions was relatively small(Fig. 5). No distinguishable differences in micelle size couldbe observed between micrographs from different sampleswith increasing concentrations of F127-DA. These

observations were also in agreement with the SAXS results.Moreover, a rough estimation of the core radius of the mi-celles could be obtained from the micrographs and wasfound to be roughly 10 nm, which is in accordance withthe radius tted by the SAXS analysis.

One may argue that the changes to the size of the aggre-gated protein backbone are a consequence of the crosslink-ing process itself, rather than the binding of micelles to thePF network. In order to examine this possibility, SAXS pat-terns of PF solutions and hydrogels crosslinked with PEG-DA (rather than F127-DA) were experimentally obtained(Fig. 3B). PEG-DA is a nonionic hydrophilic polymer chainthat does not exhibit micelle formation, but does containacrylate functional groups for participating in the photo-chemical reaction. The experimental data from these sam-ples were tted to Eq. (2) describing scattering from a longcylinder, where the equation was adjusted to highlight thesize distribution aspects of the data. Similarly to PF solu-tions with added Pluronic F127, no dissimilarities wereobserved between the scattering patterns from the differ-ent solutions (data not shown). However, the scatteringcurves of the hydrogels are also similar, regardless of thedegree of crosslinking (Fig. 3B). The best t parameterswere also in agreement with the trends in the scatteringcurves and approximately the same for all samples (Ta-ble 1). This nding indicated that no signicant alterationsto the nanostructures occurred due to crosslinking, andbinding of micelles was indeed the reason for the observedstructural alteration in the PF/F127-DA samples.Fig. 4. Schematic representation of PF hydrogel embedded with a

mixture of Pluronic F127 and F127-DA.

F127-


MACR

OMOLECU

LARNANOTE

CHNOLO

GYFig. 5. PF hydrogels with (A) 2.5% F127-DA and 7.5% Pluronic F127, (B) 5%(D) 10% F127-DA.DA and 5% Pluronic F127, (C) 7.5% F127-DA and 2.5% Pluronic F127 and

3.3. Weight gain properties

Transient stability of micelles was investigated usingweight gain studies. The time-dependent weight gain abil-ity of PF hydrogels crosslinked with F127-DA at differentconcentrations is presented in Fig. 6A. A maximum is seenfor all the studied hydrogels at about 2 h. At longer times,the weight decreases until equilibrium is reached. The ini-tial rapid increase in weight could be attributed to wateruptake, which is induced by osmotic pressure and thehydrophilic nature of Pluronic F127. The decay of weightgain at longer times could be due to diffusion of unboundPluronic F127 molecules out of the hydrogel. The equilib-rium weight gain depends on the ratio between F127-DAto Pluronic F127: the larger the F127-DA percentage,the smaller the weight gain. This observation is in line withthe suggestion that F127-DA integrates within the PF net-work, thus enhancing its crosslinking density and dimin-ishing its weight gain ability. In order to further examinethe hypothesis that unbound Pluronic F127 moleculesdiffuse out from the hydrogels, a mass balance was per-formed. The calculated amount of solids extracted duringswelling experiments, normalized to the total amount ofsolids in the hydrogel, was found to increase as the relativeconcentration of F127-DA in the block-copolymer mixturedecreased (Fig. 6B). This outcome could also explain themaximum observed in Fig. 6A; after 2 h of swelling, the un-bound Pluronic F127 molecules start to diffuse out from

the hydrogels which results in a decrease in the swellingratio.

3.4. Mechanical properties

The different degrees of weight gain observed for differ-ent samples indicate that the effective degree of crosslink-ing depends on the concentration of F127-DA which, inturn, is expected to alter the mechanical properties of thehydrogels. The Youngs modulus of hydrogels which werefully hydrated was determined using a tensile-testinginstrument. As expected, an increase in F127-DA percent-age leads to higher values of Youngs modulus (Fig. 7); alinear relationship is observed between the crosslinkerconcentration and Youngs modulus values.

3.5. Effect of Pluronic F127/F127-DA ratio on fully hydratedPF hydrogels


MACR

OMOLECU

LARNANOTE

CHNOLO

GYFig. 6. (A) Weight gain vs. time of PF hydrogels where F127-DAconcentration is () 2.5%, (N) 5%, () 7.5% and (j) 10%. (B) Extraction ofsolids from hydrogels with the addition of F127-DA vs. crosslinkerconcentration.The concentration of F127-DA in the crosslinked PFhydrogels affected the equilibrium weight gain of the gelsto the extent that the nal water content in the network(after 4 days) was noticeably reduced when micelles whereimmobilized within. SAXS studies were thus performed onhydrogels with variations in Pluronic F127/F127-DA ratioafter incubation in PBS buffer at 37 C for four days, inorder to examine the differences in nanostructure that re-sulted when micelles were immobilized. The data was tto Eq. (11), taking size distribution into account. The bestt parameters are summarized in Table 2. A detailedcomparison between the day 0 and day 4 hydrogels (Fig. 8,Tables 1 and 2) revealed smearing of the inter micellar peakfor all compositions, and a shift of this peak to smaller scat-tering angles. This could be ascribed to the release of un-bound Pluronic F127 molecules from the hydrogel, whichwas also detected bymass balance analysis; releasedPluron-ic F127 likely leads to fewer micelles with larger distancesbetween them. The increase in intensity of the fully hydratedhydrogels (day 4) compared with day 0 hydrogels could bedue to an increase in the cylinders radius of gyration from30 to approximately 50 . Consequently, data for day 4

Fig. 7. Young modulus vs. crosslinker concentration of fully hydrayed PFhydrogels with F127-DA.

Indeed, as the network imbibes more water, the resultingrelease of Pluronic F127 molecules from the hydrogel en-ables the precursor molecules free interaction with one an-other, leading to the formation of larger proteinproteinaggregates and subsequently larger nanostructures.

3.6. Relationship between nanostructure to physical andmechanical properties

The ability of the nanostructures formed by Pluronic

F127-DA to alter the physical and mechanical propertiesof the hydrogels was explored using correlation analysisbetween the protein radius of gyration R2 and the Youngsmodulus at day 4 and weight gain at equilibrium. An in-crease in the protein radius of gyration values caused a lin-ear increase in the Youngs modulus values and a decreasein the weight gain values (Fig. 9A and 9B accordingly). The

% 7.5% 10%

08.9 0.3 123.9 0.1 114.4 0.5.230 0.002 0.199 0.001 0.241 0.00201.7 0.2 94.6 0.4 53.7 0.319.5 0.1 19.0 0.2 14.8 0.255.4 0.2 57.2 0.2 Not available


MACR

OMOLECU

LARNANOTE

CHNOLO

GYTable 2Best t parameters for the SAXS data shown in Fig. 6.

Pluronic F127-DA 2.5% 5

RHS () 120.2 1.5 1hb 0.180 0.002 0R () 107.4 0.2 1R1cyl () 19.4 0.2R2cyl () 50.6 0.2

Fig. 8. Scattering from fully hydrated PF hydrogels with a total blockco-polymer concentration of 10% (w/v), where F127-DA concentration is ()2.5%, (N) 5%, () 7.5% and (j) 10%. Full symbols are for as-preparedhydrogels, empty symbols for fully hydrated hydrogels. The intensitycurves were multiplied by a factor for better visualization; the samefactor was used for the curves of the as-prepared and swollen hydrogelswith the same F127-DA content. Fits to Eq. (11), calculated from the best-of hydrogels with 10% F127-DA did not give a good t to Eq.(11), possibly due to deformation of the micelles leading toexistence of non-spherical objects.

It should also be noted that the position of the intermicellar peak was similar for all Pluronic F127-containinghydrogels after equilibrium weight gain, which indicatesthat the distances between the micelles were similar. Asnoted above, increasing the ratio of F127-DA to Pluronic

F127 reduces the number of initially formed micelles. Onthe other hand, the number of unbound Pluronic F127molecules that diffuse out of the hydrogel during the equi-librium weight gain process is smaller at higher F127-DApercentage. Thus, the balance of these two opposing effectsmay have led to an approximately constant micelle con-centration in the fully swollen hydrogels.

Another structural change induced by the equilibriumweight gain process was related to the size of the proteinbackbone. The model used to analyze the data approxi-mates the size distribution of the cylindrical protein aggre-gates as two distinct cylinders sizes with radii of gyrationof R1 and R2. For both day 0 and day 4 hydrogels, the valueR1 does not appear to depend on the F127-DA concentra-tion (Tables 1 and 2), whereas R2 increased with increasingproportions of F127-DA. For hydrogels containing anyamount of F127-DA, for example, the value of R2 increasesfrom about 30 at day 0 to approximately 50 followingequilibrium weight gain at day 4. This could very well becaused by the formation of larger protein aggregates asso-ciated with an increased crosslink density of the network.

t parameters summarized in Tables 1 and 2, are shown as solid lines.Fig. 9. Protein radius of gyration vs. (A) Young modulus and (B) weightgain at equilibrium of PF hydrogels with different F127-DAconcentrations.

most likely explanation for these observations ties back tothe diffusion of the block-copolymer molecules that leachout of the hydrogel and create cavities within it; in thefully hydrated hydrogel, the proteins radii (which aremodeled as cylinders) are able to self-aggregate thus creat-ing longer fragments and consequently larger nanostruc-tures. Those fragments enable high water uptake whichresults in diminished Youngs modulus values. Conse-quently, we can conclude that alterations in nanostructure,

Acknowledgement

This research was supported by the Singapore NationalResearch Foundation under CREATE programme: TheRegenerative Medicine Initiative in Cardiac RestorationTherapy (NRF-Technion).

References


MACR

OMOLECU

LARNANOTE

CHNOLO

GYachieved through different degrees of crosslinking, have agreat impact on the physical and mechanical propertiesof the hydrogel.

4. Conclusions

A new method for the manipulation of hydrogelmechanical and physical properties is presented throughdifferent levels of nanostructuring. By ne tuning theamount of F127-DA relative to Pluronic F127 molecules,different degrees of crosslinking are achieved which inu-ence the hydrogels weight gain ability and Youngs modu-lus. SAXS and Cryo-TEM measurements revealed twoimportant features of the nanostructured hydrogels.Firstly, as the hydrogel imbibes water to reach equilibriumweight gain, the Pluronic F127 molecules leach out of thehydrogel creating cavities and network imperfections. Sec-ondly, proteinprotein aggregation occurs in these hydro-gels during the equilibrium weight gain process, resultingin an increased size of the protein backbone. Moreover,SAXS and cryo-TEM data revealed closely packed micellesat 37 C with well-ordered structures. Nevertheless, thephotochemical crosslinking caused immobilization of themicelles in proportion to the F127-DA concentration,which in turn caused the distance between the micellesto increase and their degree of ordering to be reduced.After equilibrium weight gain of the nanostructuredhydrogels, the protein backbone increased in size in pro-portion to F127-DA concentration. Ultimately, the nano-structures in the hydrogel impacted the physical andmechanical properties; the equilibrium weight gain de-pends on the ratio between F127-DA to Pluronic F127contents: larger F127-DA percentages leads to smallerweight gain ability. Moreover, an increase in F127-DA per-centage leads to higher values of Youngs modulus. Theequilibrium weight gain ability decreased and the Youngsmodulus linearly increased with the increase in proteinsize after four days of swelling. Accordingly, it is likely thatthe higher porosity and mesh size resulting from the re-lease of unbound Pluronic F127 molecules from the fullyhydrated hydrogels had the most pronounced impact onthe hydrogels characteristics.[1] Tae G, Korneld JA, Hubbell JA. Sustained release of human growthhormone from in situ forming hydrogels using self-assembly ofuoroalkyl-ended poly(ethylene glycol). Biomaterials2005;26:525966.

[2] Scalfani VF, Bailey TS. Access to nanostructured hydrogel networksthrough photocured body-centered cubic block copolymer melts.Macromolecules 2011;44:655767.

[3] Guo C, Bailey TS. Highly distensible nanostructured elastic hydrogelsfrom AB diblock and ABA triblock copolymer melt blends. SoftMatter 2010;6:480718.

[4] Kang J, Beers KJ. Synthesis and characterization of PCL-b-PEO-b-PCL-based nanostructured and porous hydrogels. Biomacromolecules2006;7:4538.

[5] Kang J, Beers KJ. Macromolecular transport through nanostructuredand porous hydrogels synthesized using the amphiphilic copolymer,PCL-b-PEO-b-PCL. Eur Polym J 2009;45:30049.

[6] Lu C, Mikhail AS, Wang X, Brook MA, Allen C. Hydrogels containingcore cross-linked block Co-polymer micelles. J Biomater Sci Polym Ed2012;23:106990.

[7] Clapper JD, Guymon CA. Nanostructured hydrogels viaphotopolymerization in lyotropic liquid crystalline systems. MolCryst Liq Cryst 2009;509:77280.

[8] Frisman I, Seliktar D, Bianco-Peled H. Nanostructuring of PEG-brinogen polymeric scaffolds. Acta Biomater 2010;6:251824.

[9] Halstenberg S, Panitch A, Rizzi S, Hall H, Hubbell JA. Biologicallyengineered protein-graft-poly(ethylene glycol) hydrogels: a celladhesive and plasmin-degradable biosynthetic material for tissuerepair. Biomacromolecules 2002;3:71023.

[10] Mortensen K, Talmon Y. Cryo-TEM and SANS microstructural studyof pluronic polymer solutions. Macromolecules 1995;28:882934.

[11] Krupa I, Nedelcev T, Racko D, Lacik I. Mechanical properties of silicahydrogels prepared and aged at physiological conditions: testing inthe compression mode. J SolGel Sci Technol 2010;53:10714.

[12] Frisman I, Orbach R, Seliktar D, Bianco-Peled H. Structuralinvestigation of PEG-brinogen conjugates. J Mater Sci MaterMed 2010;21:7380.

[13] Shachaf Y, Gonen-Wadmany M, Seliktar D. The biocompatibility ofPluronic F127 brinogen-based hydrogels. Biomaterials2010;31(10):283647.

[14] Glatter O, Kratky O. Small angle X-ray scattering; 1982. 515pp.[15] Glatter O, Scherf G, Schillen K, Brown W. Characterization of a

poly(ethylene oxide)poly(propylene oxide) triblock copolymer(EO27PO39EO27) in aqueous solution. Macromolecules1994;27:604654.

[16] Wanka G, Hoffmann H, Ulbricht W. Phase diagrams and aggregationbehavior of poly(oxyethylene)poly(oxypropylene)poly(oxyethylene) triblock copolymers in aqueous solutions.Macromolecules 1994;27:41459.

[17] Mortensen K, Brown W. Poly(ethylene oxide)poly(propyleneoxide)poly(ethylene oxide) triblock copolymers in aqueoussolution. The inuence of relative block size. Macromolecules1993;26:412835.

[18] Pedersen JS. Analysis of small-angle scattering data from colloidsand polymer solutions: modeling and least-squares tting. AdvColloid Interface Sci 1997;70:171210.

A novel method for hydrogel nanostructuring1 Introduction2 Experimental2.1 Synthesis of PEG diacrylate (PEG-DA) and F127 diacrylate (F127-DA)2.2 Nanostructured PF hydrogels

3 Results and discussion3.1 Precursor solutions with added Pluronic F127/F127-DA mixtures3.2 Nanostructured hydrogels with added Pluronic F127/F127-DA mixtures3.3 Weight gain properties3.4 Mechanical properties3.5 Effect of Pluronic F127/F127-DA ratio on fully hydrated PF hydrogels3.6 Relationship between nanostructure to physical and mechanical properties

4 ConclusionsAcknowledgementReferences

a novel method for hydrogel nanostructuring

Documents