Pre-osteoblastic cell response on three-dimensional, organic-inorganichybrid material scaffolds for bone tissue engineering

Konstantina Terzaki,1,2 Maria Kissamitaki,1,2 Amalia Skarmoutsou,3 Costas Fotakis,4,2

Costas A. Charitidis,3 Maria Farsari,2 Maria Vamvakaki,1,2 Maria Chatzinikolaidou1,2

1Department of Materials Science and Technology, University of Crete, P.O. Box 2208, GR-71303 Heraklio, Greece2Institute of Electronic Structure and Laser (IESL), Foundation for Research and Technology Hellas (FORTH), Greece3School of Chemical Engineering, National Technical University of Athens, Greece4Department of Physics, University of Crete, Greece

Received 15 June 2012; revised 28 October 2012; accepted 29 October 2012

Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.a.34516

Abstract: Engineering artificial scaffolds that enhance cell

adhesion and growth in three dimensions is essential to suc-

cessful bone tissue engineering. However, the fabrication of

three-dimensional (3D) tissue scaffolds exhibiting complex

micro- and nano-features still remains a challenge. Few materi-

als can be structured in three dimensions, and even those have

not been characterized for their mechanical and biological

properties. In this study, we investigate the suitability of three

novel materials of different chemical compositions in bone tis-

sue regeneration: a hybrid material consisting of methacrylox-

ypropyl trimethoxysilane and zirconium propoxide, a hybrid

organic–inorganic material of the above containing 50 mole%

2-(dimethylamino)ethyl methacrylate (DMAEMA) and a pure

organic material based on polyDMAEMA. More specifically, we

study the mechanical properties of the aforementioned materi-

als and evaluate the biological response of pre-osteoblastic

cells on them. We also highlight the use of a 3D scaffolding

technology, Direct femtosecond Laser Writing (DLW), to fabri-

cate complex structures. Our results show that, while all three

investigated materials could potentially be used as biomateri-

als in tissue engineering, the 50% DMAEMA composite exhibits

the best mechanical properties for structure fabrication with

DLW and strong biological response.VC 2013 Wiley Periodicals, Inc.

J Biomed Mater Res Part A: 00A:000–000, 2013.

Key Words: MC3T3-E1 pre-osteoblasts, three-dimensional

scaffold fabrication, hybrid material, nanomechanical charac-

terization, cell adhesion

How to cite this article: Terzaki K, Kissamitaki M, Skarmoutsou A, Fotakis C, Charitidis CA, Farsari M, Vamvakaki M,Chatzinikolaidou M. 2013. Pre-osteoblastic cell response on three-dimensional, organic-inorganic hybrid material scaffolds forbone tissue engineering . J Biomed Mater Res Part A 2013:00A:000–000.

INTRODUCTION

In tissue engineering, a number of parameters significantlyinfluence the cellular response on a functional cell-scaffoldconstruct, such as the material chemistry, the material po-rosity and pore size, interconnectivity, mechanical proper-ties, cell seeding density, and various exogenous growth fac-tors.1 Scaffold characteristics related to their mechanicaland chemical properties are considered crucial. It has beenacknowledged that cells seeded on scaffolds can recognizedifferences related to the materials’ physical and mechanicalproperties and subsequently change their response andfunctions. Earlier studies have reported the influence of thescaffold stiffness on cell adhesion,2 morphology,3,4 prolifera-tion,5,6 and differentiation.5

Considering in situ bone repair, current strategies for sur-gical intervention include the use of autografts and allografts.

Examples of commercially available allografts are InfuseVR

Bone Graft (http://www.medtronic.com/patients/lumbar-de-generative-disc-disease/surgery/index.htm), a collagen car-rier sponge with bone morphogenetic protein 2 designed totreat degenerative disc disease, and TrinityV

R

EvolutionTM

(http://www.orthofix.com/common_products.asp?pid¼90&cid¼45), a cancellous bone containing viable adult stem cells,osteoprogenitor cells and a demineralized bone component.Each approach has limitations, such as cost, variability inosteogenic capacity, and lot-to-lot variability. Previous workhas described7 donor-site morbidity in the use of autograftsand the risk of immunogenic rejection and disease transmis-sion in the use of allografts. To overcome these inherentlimitations of autografts and allografts, synthetic bone-graftsubstitutes have been developed as an alternative. There is amajor clinical need for versatile biomaterial systems for bone

Correspondence to: Maria Chatzinikolaidou; e-mail: mchatzin@materials.uoc.gr

Contract grant sponsor: European Union (European Social Fund—ESF)

Contract grant sponsor: Greek national funds (Operational Program ‘‘Education and Lifelong Learning’’ of the National Strategic Reference

Framework (NSRF) - Research Funding Program: Heracleitus II-Investing in knowledge society through the European Social Fund)

Contract grant sponsor: ITN TOPBIO; contract grant number: PITN-GA-2010-264362

Contract grant sponsor: Special Account for Research Fund of the University of Crete

VC 2013 WILEY PERIODICALS, INC. 1

repair that mimic the architecture and mechanical and bio-stimulating functions of native bone.8

Zirconium propoxide (ZPO) is a particularly suitable bio-material due to its advantageous mechanical propertiessuch as high strength, toughness, and stability. Although ourwork considers ZPO as a component in a hybrid biomaterialfor a tissue engineering application, most prior work usingit derives from the area of prosthetic substitution.9–12

Organic materials possess certain functionalities thataffect their interactions with cells. Numerous organic coat-ings on solid substrates have been employed for the devel-opment of cell culture scaffolds as well as antibacterialsurfaces, whereas organic nanoparticles have been used indrug and gene delivery agents. Among them, poly(2-(dime-thylamino ethyl)methacrylate) (PDMAEMA)-based materialshave been extensively used on surfaces and coatings andwere shown to exhibit antibioadherent properties againstbacteria, macrophages, and fibroblasts similarly to other cat-ionic biocides.13–16 A recent study on PDMAEMA17 proposedthat polymers with a branched architecture and an interme-diate molecular weight are promising candidates for effi-cient gene delivery, since they combine low cytotoxicitywith acceptable cell transfection.

Hybrid organic–inorganic materials combining the abovementioned mechanical, chemical, and biological materialproperties have emerged recently as a new class of materi-als in tissue engineering. These hybrid materials possessdifferent properties compared to their component materialsand constitute high-performance and multifunctional materi-als with an excellent balance between strength, toughness,and tunable chemical and mechanical characteristics.18 Com-posite scaffolds based on hydroxyapatite, the most widelystudied hybrids so far, were shown to possess improvedmechanical properties and osteoconductivity.19,20 ZrO2

incorporated within poly(e-caprolactone) matrices wasshown to develop advanced composite substrates withimproved mechanical and biological performance.21

In tissue engineering, the ability to control tissue forma-tion in three dimensions is essential.22 Although several fab-

rication methods have been used to produce scaffolds, thosemethods are unable to produce three-dimensional (3D) sub-micron and nanoscale scaffolds with precise control of thegeometry, a crucial factor for the recent developments inthe field of tissue engineering. Femtosecond laser-inducedtwo-photon polymerization is a promising technique thatfulfills these requirements. In Direct fs Laser Writing (DLW),the beam of an ultrafast laser is tightly focused into the vol-ume of a photosensitive material, initiating multiphoton po-lymerization within the material.23 By moving the beamfocus three-dimensionally, arbitrary 3D, high-resolutionstructures can be written into the volume of the material(Fig. 1). A variety of materials have been structured usingDLW including purely organic polymers,24 organic-inorganichybrids,25 biodegradable materials,26 and proteins.27,28

In this article, we present our investigations into thesuitability of materials that can be structured in complex, 3Dgeometries using DLW for bone tissue scaffolds. First, wedescribe the synthesis and characterization of chemical andmechanical properties of three different materials compris-ing methacryloxypropyl trimethoxysilane (MAPTMS), ZPO,and/or 2-(dimethylamino)ethyl methacrylate (DMAEMA).Next, we use these materials to fabricate 2D films, and inves-tigate the cell viability and proliferation of pre-osteoblasticcells on them. Additionally, we explore the influence of thematerials’ chemical composition on cell proliferation. Finally,we report on the pre-osteoblastic cell adhesion on 3D scaf-folds fabricated from the hybrid organic–inorganic materialcomposition containing 50% DMAEMA, within the first hourand up to 3 days in culture.

MATERIALS AND METHODS

Material synthesisAll the chemicals used in this work were obtained fromSigma-Aldrich (Germany) and were used as received unlessotherwise stated. The material used for the fabrication of2D films and 3D structures is an organic–inorganic compos-ite, produced by the addition of MAPTMS (99%) to ZPO(70% in propanol). DMAEMA (>99%) was also added

FIGURE 1. DLW allows the fabrication of readily assembled, fully 3D structures. (a) a typical porous structure (b) a micro ballerina.

2 TERZAKI ET AL. PRE-OSTEOBLASTIC CELL RESPONSE

which was copolymerized with MAPTMS upon photopoly-merization. ZPO and the alkoxysilane groups of MAPTMSserved as the inorganic network forming moieties. 4,4-bis(-diethylamino)benzophenone was used as the photoinitiator.

MAPTMS was first hydrolyzed using HCl solution (0.1M) at a 1:0.1 ratio. After 5 min, the ZPO was slowly addedto the hydrolyzed MAPTMS at a 3:7 ZPO:MAPTMS molar ra-tio. After stirring for 15 min, DMAEMA was added at aDMAEMA/MAPTMS molar ratio 5:5. Following another 30min, water was added to the mixture at a 2.5:5 MAPTM-S:H2O molar ratio. Finally, the photoinitiator, at a 1% (w/w)concentration with respect to photopolymerizable methacry-late moieties was added to the mixture. After stirring for afurther 15 min, the composite was filtered using a 0.22 lmsyringe filter.

A hybrid material comprising only MAPTMS and ZPO inthe absence of DMAEMA was also prepared following a pro-cedure similar to that described above. This material wasused as a control material to assess the effect of DMAEMAon the chemical, mechanical and biological properties of thefilms and 3D cell scaffolds.

Sample preparationThree types of specimens, all prepared on glass substrates,were employed in this study: (i) thin films for the quantifica-tion of cell proliferation (ii) 3D square blocks with dimen-sions 200 � 200 � 10 lm3 (l � w � h) for the investigationof cell adhesion and cell morphology by immunocytochemi-cal staining and (iii) 3D scaffolds with bar distances of 1, 2,and 5 lm for the investigation of cell adhesion by SEM.Materials specimens used for the adhesion, viability and pro-liferation experiments were incubated for 1 h in ethanol, air-dried under sterile conditions in a laminar flow and rinsedbriefly with alpha-minimal essential medium (MEM) cell cul-ture medium without fetal bovine serum (FBS) prior to cellseeding.

Hybrid thin film preparation. Thin films were prepared bydrop-casting or spin-coating onto 100-lm thick silanizedglass substrates, and they were dried in the oven at 50 �Cfor 5 min before the photopolymerization. The heating pro-cess led to the condensation of the alkoxide groups and theformation of the inorganic matrix. Next, the methacrylatemoieties were polymerized using a KrF excimer laser, oper-ating at 248 nm, resulting in the formation of irreversibleand fully saturated aliphatic CAC covalent bonds that fur-ther increase the connectivity of the material. Finally, thesamples were developed for 30 min in a 50:50 solution of1-propanol:isopropanol, and were further rinsed withisopropanol.

Fabrication of hybrid blocks and 3D scaffolds byDLW. The experimental setup employed for 3D structurefabrication has been previously described extensively.29 ATi:Sapphire femtosecond laser (Femtolasers Fusion, 800 nm,75 MHz, <20 fs) was focused into the photopolymerisablecomposite using a microscope objective lens (20�, N.A. ¼0.65 and 40� N.A. ¼0.95, Zeiss, Plan Apochromat). Sample

movement was achieved using piezoelectric and linearstages, for fine and step movement, respectively (PhysikInstrumente GmbH, Germany). The whole DLW setup wascomputer-controlled using the 3DPoli (3DPoli@gmail.com)software. The average power used for the fabrication of thehigh-resolution structures was 80 mW, measured before theobjective, while the average transmission was 20%. Thescanning speed was always set at 2 mm/s.

Polymeric thin films. Polymer thin films based onDMAEMA were also prepared on the tissue engineeringglass slides using surface initiated atom transfer radical po-lymerization. First, the surface-bound initiator, 3-(2-bromoi-sobutyramido)propyl(triethoxy)silane was synthesized bythe reaction of 3-aminopropyl(triethoxy)silane with 2-bro-moisobutyrylbromide. The self-assembled monolayer of theinitiator was formed by immersing the glass substrates in aflask containing a tetrahydrofuran (THF) solution of the ini-tiator for 24 h. After the incubation period the substratewas rinsed extensively with THF and ethanol, dried under astream of nitrogen and transferred to the polymerizationflask. The PDMAEMA brushes were synthesized on the ini-tiator-functionalized glass substrates. In a typical synthesis,DMAEMA (10 g, 63.16 mmol) was dissolved in a 4/1 metha-nol/water mixture (10 mL). Next, copper (I) bromide (0.14g, 0.07 mmol) and 2,20-bipyridyl (0.32 g, 0.14 mmol) wereadded under nitrogen. 0.1 mol% of Cu(II) deactivator rela-tive to Cu(I) was also added with the catalyst system toensure a controlled growth of the polymer chains. The reac-tion mixture was then transferred to a nitrogen-purgedreaction flask containing the initiator coated glass sub-strates. The polymerization was allowed to proceed at roomtemperature for 5 h, after which the substrate was removedand was rinsed thoroughly with water and methanol andwas dried under a nitrogen flow. These polymeric filmswere used as control samples and their chemical, mechani-cal and biological properties were compared to those of thehybrid materials investigated in this work.

CharacterizationEllipsometry. A variable angle spectroscopic ellipsometer(model VASE, J. A. Woollam) was used to determine thethickness and the refractive index of the polymeric andhybrid films. The measurements were performed at threedifferent angles of incidence 65�, 70�, and 75� in the wave-length range 450–1200 nm.

Atomic force microscopy. The surface morphology of thehybrid and polymer films in the dry state was imaged byatomic force microscopy (AFM). AFM studies were per-formed at ambient conditions on a multimode Nanoscope IIIinstrument (Digital Instruments, Veeco) operating in thetapping mode at 1 Hz scan rate.

Contact angle measurement. The static contact angleswere determined by the sessile drop method on a home-made instrument using a 10 lL drop of nanopure water(18.2 MX). Images of the droplet on the meterial spin-

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2013 VOL 00A, ISSUE 00 3

coated or grown onto a glass substrate were recorded usinga camera and the contact angles were calculated from theseimages using the appropriate software. For each sample, atleast three measurements from different surface locationswere averaged.

Nanomechanical properties of hybrid and organic coat-ings. The nanoindentation test was used as a static methodto determine the mechanical behaviour of the material thinfilms when deformed at the sub-micron scale. The methoddeveloped by Oliver and Pharr30,31 allows determining thehardness (H) and the reduced modulus (Er) from the nano-indentation load-displacement data.32 The evaluation of theelastic properties of the coatings is based on the precisemeasurement of the penetration depth of an indenter intothe coating.33,34 In order to characterize the coatings’ me-chanical properties, the indenter must have a well-definedgeometry and avoid substrate effects. The investigated pene-tration depths should not exceed 10% of the coatingsthickness.

The indentation tests were performed using a HysitronTriboLabV

R

nanomechanical test instrument. The transducerused to perform the nanoindentation tests, has the ability toapply loads in the range of 1–10,000 lN with a high loadresolution of 1 nN, while the maximum penetration depththat can be recorded is 3000 nm (3 lm) with a resolutionof 0.04 nm. A three-sided pyramidal Berkovich diamond in-denter (120 nm radius of curvature) was used for the mea-surement of hardness, stiffness and elastic modulus. The tipradius was calculated prior to the experimental procedure,by calibration on fused quartz. Experiments were performedin a clean area environment with �45% humidity and 23�C ambient temperature.35 The nanomechanical instrumentused in this study was equipped with a Scanning ProbeMicroscope (SPM), in which the mounted Berkovich probetip was moved in a raster scan pattern across a sample sur-face using a three-axis piezo positioner.

Indentation tests were performed on a 100 lm silanizedglass substrate, and the H and Er of the substrate weremeasured. Additionally, films with thicknesses in the orderof a few micrometers, of the three materials, PDMAEMA,Hybrid 1 and Hybrid 2 (see Table II), deposited onto silan-ized glass substrates by spin coating, were tested at severalapplied loads, ranging from 10 to 6000 lN before and aftersubmersion into the cell culture medium (MEM alpha modi-fication) for 2 h and following rinsing with H2O. Loadingand unloading segments of the performed indentation testshad a course of 10 s, respectively. The holding time at themaximum loads for all tests in this work was 1 s, since nocreep phenomena were observed at any sample with excep-tion of the organic coating. For this reason, different inden-tation tests were performed for the organic coating, withloading, hold time and unloading segments of 5 s each, toeliminate creep phenomenon and avoid the nose effect inthe upper part of the unloading curve. Repeatable tests (sixtests) at each applied load were performed at the center ofthe samples and at four different points across the samplesdiameter. The distance between neighboring indentation

sites was larger than 6 lm in order to avoid the lateral sizeof the plastic zone around the largest indents and excludethe mutual influence of individual indents.36

Cell cultureCell culture media and reagents were purchased from Invitro-gen. Early passages 3–15 of the mouse calvaria pre-osteoblas-tic cell line MC3T3-E1 were grown in cell culture flasks usingalpha-MEM, supplemented with glutamine (2 mM), penicillin(50 IU/mL), streptomycin (50 g/mL) and 10 vol % FBS in ahumidified atmosphere and 5% CO2 at 37 �C in a cell incuba-tor (Thermo Scientific).37 Confluent cells were washed withphosphate buffer saline (PBS) and passaged after trypsination(0.25% trypsin in 1 mM EDTA), seeded at 60–80% confluenceand allowed to grow for 4–5 days before the next passage. Forcell adhesion and proliferation experiments on the compositematerial-coated specimens, cells were cultured in alpha-MEMwith only 1% FBS, to reduce the effect of serum on cell attach-ment and growth.

Cell viability and proliferation assay10 � 103 cells in alpha-MEM with 1% FBS were seeded onspecimens with film coatings consisting of our testing mate-rials and placed in the cell culture incubator at 37 �C. Ondays 1, 3, and 7 post seeding cell viability and proliferationassay was performed with the resazurin-based PrestoBlueTM

reagent (Invitrogen) according to the manufacturer’sinstructions. The reagent was incubated on the cells at 37�C for 60 min. The absorbance was measured in a spectro-photometer (Molecular Devices SpectraMax M2) and cellnumber quantification was performed by means of a cali-bration curve. Error bars representing the average of tripli-cates 6 STDV in three independent experiments were calcu-lated. (**) symbols denote significant differences accordingto statistical analysis by one-way analysis of variance(ANOVA) Dunnett’s test.

Optical microscopyA suspension of 2 � 104 cells in alpha-MEM with 1% FBSwere seeded on specimens with 3D scaffolds or blocks con-sisting of the material Hybrid 2 and placed in the cell cul-ture incubator at 37 �C. Cells on the specimens were exam-ined daily for 7 days and visualized by optical microscopyby means of a Zeiss Axiovert 200 microscope. Images weretaken by a ProgResV

R

CFscan Jenoptik camera (Jena, Ger-many) using the ProgResV

R

CapturePro 2.0 software andobjective lenses for 10 and 20-fold magnifications.

Scanning electron microscopy2 � 104 cells in alpha-MEM with 1% FBS were seeded onspecimens with 3D scaffolds consisting of the materialHybrid 2 and placed in the cell culture incubator at 37 �Cfor 2 h. Specimens were then removed from the incubatorand rinsed three times with PBS, fixed with 2% paraformal-dehyde for 1 h, post-fixed with 1% osmium tetroxide, anddehydrated in increasing concentrations (from 30 to 100%)of ethanol. The specimens were then dried in a critical pointdrier (Baltec CPD 030), sputter-coated with a 15-nm thick

4 TERZAKI ET AL. PRE-OSTEOBLASTIC CELL RESPONSE

layer of gold (Baltec SCD 050) and observed under a scan-ning electron microscope (JEOL JSM-6390 LV) at an acceler-ating voltage of 15 kV.

Laser scanning confocal microscopyA suspension of 2 � 104 cells in alpha-MEM with 1% FBSwere seeded on specimens with 3D scaffolds consisting ofthe material Hybrid 2 and placed in the cell culture incuba-tor at 37 �C for 24 h. Cells on specimens were rinsed withPBS and fixed with 4% paraformaldehyde for 15 min andpermeabilized with 0.1% Triton X-100 (Merck, Darmstadt,Germany) in PBS for 5 min on ice. The non-specific bindingsites were blocked with a 2% BSA solution in PBS for 60min. Actin and vinculin focal adhesion complexes werestained by incubating cells-on-specimens in 100 lL dilutedfluorescein isothiocyanate-conjugated anti-vinculin primaryantibody (FITC-conjugated anti-vinculin, Sigma-AldrichChemie GmbH, Munich, Germany) in blocking solution(1:100) for 1 h and subsequently staining them in simulta-neous incubation with 100 lL 1 mg/mL tetramethyl rhoda-mine isothiocyanate-conjugated phalloidin (TRITC-phalloidinconjugate, Sigma-Aldrich Chemie GmbH, Munich, Germany)for 5 min. The samples were then washed with PBS andobserved under a Leica DM IRBE laser scanning confocalmicroscope.

Statistical analysisStatistical analysis was performed using the one-way ANOVADunnett’s test. To statistically evaluate the difference in cellproliferation after certain time points (1, 3, and 7 days), wecompared all three material chemical compositions at eachtime point against the control glass surface.

RESULTS AND DISCUSSION

Table I summarizes the polymer and hybrid film character-istics. A PDMAEMA layer with a dry film thickness of 50 nmwas obtained after 5 h polymerization time. As imaged byAFM, the surface-anchored film covered the substrate sur-face completely and homogeneously, suggesting that theATRP polymerization proceeded uniformly on the substrate,and a surface roughness of the dry polymeric film of 0.5 6

0.1 nm was found, verifying that the prepared polymer sur-face is very smooth [Fig. 2(a)]. A contact angle of 70 6 2�

was measured for the PDMAEMA coated glass substratesuggesting the preparation of a moderately hydrophilic sur-face. On the other hand, the thickness of the two films pre-pared by spin coating was with 6.3 lm for Hybrid 1 and 2.4lm for Hybrid 2 much higher compared to that of thePDMAEMA layer as measured by ellipsometry. However,

AFM verified the preparation of a uniform and smooth filmin both cases with average roughness of 0.4 6 0.1 and 0.36 0.1 nm, for the Hybrid 1 and 2 materials, respectively[Fig. 2(b,c)]. Finally, static contact angle measurements sug-gested that the Hybrid 1 film exhibited a similar hydrophi-licity to the PDMAEMA film (CA ¼ 71 6 2�) attributed tothe presence of remaining surface hydroxyl groups ofMAPTMS, whereas, a higher contact angle of 85� wasobtained for the Hybrid 2 film suggesting a lower fractionof surface AOH groups as discussed below.

The glass substrate revealed elastoplastic behavior at allapplied loads, and the measured H and Er values werefound 0.75 6 0.06 GPa and 1.74 6 0.53 GPa, respectively.Figure 3 presents load-unload curves of the synthesizedcoatings, before and after submersion of the samples intothe medium. All samples were submerged in 12 mL of cellculture medium with the exception of the organic coating,because PDMAEMA is soluble in water and the film woulddissolve in the medium.38 Moreover, the nanomechanicalproperties of the organic coating were studied at a thickerspin coated polymer film (0.5 lm, Table II). The calculationof the nanomechanical properties of soft thin film depositedon hard substrates is not accurate, because of the enhancedstresses derived by the glass substrate, the creep effect andthe pile-up formations.39 Fig. 3 presents load-unload curvesof the hybrid coatings before and after submersion in thecell culture medium. No load–unload of the organic coatingare presented because of the lower applied load performedto study the nanomechanical properties of the coating. How-ever, it is observed (presented by the H and Er values in Ta-ble II and Figure 5) that the organic coating is the softestmaterial of all tested samples. Moreover, load–unload curvesof the Hybrid 1 coating, before submersion into themedium, revealed that the material exhibited an elasticbehavior at low applied loads and elastoplastic behavior athigher applied loads. However, after submersion into thecell culture medium, a softening and an increase of theplastic behavior of the Hybrid 1 coating was observed.Conversely, the Hybrid 2 coating revealed a plastic behaviorat low penetration depths and an elastoplastic behavior athigher depths, while a hardening after submersion into thecell culture medium was obtained.

It was noted that hybrid coatings revealed discontinuitiesin the loading curves after submersion into the cell culturemedium, which were attributed to stronger interactionsbetween the tip and the film surface. The plastic behaviorobserved in the first surface layers of the Hybrid 1 coatingafter submersion was attributed to the influence of remain-ing non-condensed AOH groups, due to incomplete

TABLE I. Organic and Hybrid Film Characteristics

Film CompositionThickness by

Ellipsometry (nm)Roughnessby AFM (nm)

Static ContactAngle (�)

Organic PDMAEMA 50 6 2 0.5 6 0.1 70 6 2Hybrid 1 MAPTMS þ ZPO 6300 6 2 0.3 6 0.1 71 6 2Hybrid 2 50 mol% MAPTMS þ

ZPO þ 50 mol% DMAEMA2365 6 2 0.4 6 0.1 85 6 2

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2013 VOL 00A, ISSUE 00 5

formation of the inorganic network, which re-orient towardsthe surface of the hybrid coating in the aqueous medium.40

The enhanced resistance to all applied loads found for theHybrid 2 coating suggested a lower concentration of remain-ing AOH groups and thus a higher conversion of the conden-sation reaction. This is due to the presence of the liquidDMAEMA monomer, which reduces the viscosity of the

hybrid material in the spin-coated film, allowing for a higherflexibility of the inorganic network formed during the con-densation reaction upon heating.

Figure 4 presents SPM images of the tested areas for allsynthesized films before and after submersion into the me-dium. No cracks around or underneath the indents for anycoating was observed, before and after submersion of the

FIGURE 2. AFM images of the (a) PDMAEMA, (b) Hybrid 1, and (c) Hybrid 2 surfaces. [Color figure can be viewed in the online issue, which is

available at wileyonlinelibrary.com.]

FIGURE 3. Load-unload curves at (a) low and (b) high applied loads before and after submersion into the cell culture medium. The organic film

coating was not submerged in the cell culture medium. [Color figure can be viewed in the online issue, which is available at

wileyonlinelibrary.com.]

6 TERZAKI ET AL. PRE-OSTEOBLASTIC CELL RESPONSE

samples into the medium, revealing good adhesion to thesubstrate for the organic coating and low brittleness of thetwo hybrid films.41 However, pile-ups around the performedindents are observed for organic and Hybrid 2 coatings,denoted with circles in Figure 4. The pile-up phenomenon isoften excessive in situations of a soft coating onto a hardsubstrate, because the indenter often penetrates well intothe substrate. When pile-up occurs, the real contact area ofthe indenter is larger than the one calculated by the Oliverand Pharr model, and lower values of hc are found, resultingin an overestimation of the H values.39

The H and Er values calculated by the Oliver and Pharrmethod are presented in Table II for all coatings, before and

after submersion into the culture medium, at the 10% ofcoatings’ thickness. The data shown in Table II wereobtained by six repeatable tests at the same applied load,reaching only the 10% of coatings thickness, at four differ-ent points across the diameter of the samples. The H and Ervalues presented in Figure 5 as a function of indentationdepth are the mean values of six repeatable tests at eachapplied load, located at the center of the sample, wherehigher homogeneity was observed. The organic coatingrevealed the lowest H and Er values (0.028 GPa and O.502GPa, respectively) of the non-submerged samples. The Hand Er values at higher indentation depths increased to�0.062 GPa and �1.1 GPa, respectively, due to substrate’seffect. H values for the Hybrid 1 coating were measured�0.8 GPa at the near surface (50 nm) and as the indenterfurther penetrated into the coating, H values, at firstincreased, reaching a maximum value of 1.5 GPa, and finallydecreased to 0.45 GPa. Finally, the H values for Hybrid 2coating decreased with increasing penetration depth, andwas found H �0.58 GPa for the near surface and �0.27 GPafor the bulk. The two hybrid coatings revealed the samecourse in Er values before submersion into the cell culturemedium, as the indenter penetrated further into the

TABLE II. Hardness and Reduced Modulus Values of the

Synthesized Samples at 10% Coating Thickness

Film H (GPa) Er (GPa)

Organic (�0.5 lm) 0.0283 6 0.0015 0.505 6 0.013Hybrid 1 (6.3 lm) 1.114 6 0.0227 2.281 6 0.0672Submerged 0.297 6 0.016 1.268 6 0.0132Hybrid 2 (2.4 lm) 0.527 6 0.029 3.725 6 0.0619Submerged 0.638 6 0.0025 7.98 6 0.0666

FIGURE 4. SPM images of tested areas for all studied samples. Pile-ups are denoted with circles around the performed indents. [Color figure

can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2013 VOL 00A, ISSUE 00 7

samples. At the near surface, Er values were �7.5 GPa, andfor the bulk �1 GPa. Table II presents H and Er values ofthe three films studied, at a 10% coating thickness. Itshould be noted that there are no previous studies on thenanomechanical properties of the materials studied in thiswork. Instead, the mechanical properties of a water swollenPDMAEMA hydrogel containing 3.2% polymer has beenmeasured by tensile testing for which a shear modulus of32.3 kPa was found, which is much lower than the valueobtained for the dry PDMAEMA film in this study (0.028GPa) and is attributed to the high water content of thehydrogel sample.42 Conversely, the hardness and elasticmodulus of a MAPTMS film were found 0.6 and 4.5 GPa,respectively.43 The differences in the H and Er valuesobtained for the Hybrid 1 coating compared to the valuesreported for the MAPTMS film are attributed to the absenceof ZPO in the hybrid coating reported in the literature andalso to possible differences in the degree of condensationand polymerization of the two samples. Finally, the H and Ervalues of hybrid films comprising MAPTMS, ZPO, and MAA,on glass and silicon substrates were measured by nanoin-dentation. Similar values (H ¼ 0.37 GPa and Er ¼ 1.6 GPa)to those measured for the Hybrid 2 coating in this study(Fig. 5) were obtained, whereas, possible deviations wereattributed to differences in the chemical composition andthe use of DMAEMA vs MAA for the two samples.44

A very important characteristic of polymeric materialsthat plays a crucial role in several applications such as per-sonal care products,45 coatings, composite materials, mem-branes, sensors,46 and biomedical products,47–49 is theirability to uptake moisture. Water can change the polymerchain conformation and results in variation of the mechani-cal properties of the material. Consequently, the load-unloadcurves of the hybrid materials after the submersion of thesamples into the cell culture medium were measured, andthe H and Er values were calculated. After submersion, theHybrid 1 coating revealed lower H and Er values, whichwere attributed to possible re-orientation of the remainingAOH groups towards the Hybrid 1 coating surface in thecell culture medium.50 On the contrary, an increase in the H

and Er values of the Hybrid 2 coating was calculated, asexpected from the load-unload curves, presented in Figure3. This increase was attributed to some kind of materialreorganization in the presence of a lower fraction of remain-ing AOH groups upon heat treatment of the spin-coatedfilm, as discussed above. As the surface roughness of allfilms synthesized in this study is similar for the three sam-ples, these results are consistent with the contact anglemeasurements presented above. A more hydrophilic surfacewas found for the Hybrid 1 coating and agrees with anincreased number of surface AOH groups in the coating,whereas, the higher contact angle and thus higher hydro-phobicity measured for the Hybrid 2 coating is in goodagreement with the deficiency of the film in remaining AOHgroups.

The optical microscopy images (Figure 6) show an over-view of pre-osteoblastic cells seeded on Hybrid 2 blocks(200 � 200 � 10 lm) for 2 h. Cells adhere and spread onthe material surfaces, at the edges of the material blockstructures, and in the boundary between the material andthe glass substrate underneath. Cells cover the material byforming a polygonal morphology. Particularly in the 20-foldmagnification, we observe elongated cells with cytoplasmicextensions attached on the material block. This is a charac-teristic morphology for cells seeded on a preferable, bio-compatible material surface not showing any adverseeffects. The cell appearance is similar to that of the cellsthat adhere on the glass substrate, which is known as a sub-strate exhibiting good MC3T3-E1 cell adhesion.51

Representative scanning electron microscopy images(Fig. 7) demonstrate cells that adhere on the Hybrid 2 ma-terial, extend protrusions within the material and prolifer-ate after 3 days. Specifically, Figure 7(a) shows an overviewof cells adhered after 1 h on a 3D fabricated scaffold withbar distances of 2 lm. The observed fully spread cell mor-phology with polygonal shape and cytoplasmic extensionssignal good cell adhesion within the first hour after seedingon the material. In high magnifications [Fig. 7(b)] we obtaincells spreading after 2 h on the scaffolds with 2 lm bar dis-tances and extending cytoplasmic protrusions in the

FIGURE 5. a: Hardness (H) and (b) reduced modulus (Er) values as a function of maximum indentation depth for all tested samples. [Color figure

can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

8 TERZAKI ET AL. PRE-OSTEOBLASTIC CELL RESPONSE

material structures. As the scaffolds’ bar distancesincreased up to 5 lm, more pseudopodia are shown toextend within the materials’ structure as depicted in Figure

7(c). Cells grown for 72 h on the Hybrid 2 structured mate-rial [Fig. 7(d)] demonstrate their proliferation and subse-quent coverage of the structure with a dense cell layer,

FIGURE 7. Scanning electron microscopy images of MC3T3-E1 pre-osteoblastic cells cultured on scaffolds of Hybrid 2. (a) Overview of cells

adhered after 1 h on a fabricated scaffold with bar distances of 2 lm shown with a 1200-fold magnification; (b) Cells adhere on the scaffolds af-

ter 2 h and extend protrusions in the material structures (scaffolds’ bar distances 2 lm, magnification 14,000-fold); (c) the bar distances

increased up to 5 lm, more pseudopodia are shown to be extended after 2 h within the materials structure (13,000-fold magnification); (d) After

72 h the adhered cells proliferated and formed a dense cell layer that completely covered the structures (300-fold magnification).

FIGURE 6. Optical microscopy images of pre-osteoblastic cells seeded on 50% DMAEMA blocks (200 � 200 � 10 lm) for 2 h. Cells adhere and

spread on the material surfaces and at the edges between the material and the glass substrate beneath exhibiting a good initial adhesion (left:

10-fold magnification, right: 20-fold magnification).

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2013 VOL 00A, ISSUE 00 9

exhibiting the proliferation increase as a consequence ofgood initial adhesion and thus, validating the biocompati-bility of the hybrid material.

The morphology of adhered cells is shown by confocalfluorescence microscopy images of cells grown for 24 h onblock structures of the Hybrid 2 material. Cells display awell-organized cytoskeleton as reflected by the actin stain-ing shown in red [Fig. 8(a)]. The visualization of vinculinparticipating in focal adhesion points is demonstrated ingreen in Figure 8(b). Figure 8(c) depicts an overlay of thedouble-stain in different magnifications. Arrows display theboundary between the material block structure and theunderlying glass.

A further question is, whether the chemical compositionof the material plays a role in cell viability and proliferation.Figure 9 shows similar cell viability within the first day forall examined surfaces. MC3T3-E1 pre-osteoblasts culturedon all three materials with different chemical compositionshow a proliferation increase after 3 and 7 days. In particu-lar, after 3 days, cell proliferation is significantly higher onthe PDMAEMA and the Hybrid 1 compared to the Hybrid 2surface as indicated by the statistical analysis. Surprisingly,both of the above surfaces showed similar hydrophilicitywith measured contact angles around 70�. In contrast, cells

proliferated on the Hybrid 2 material with measured contactangle of 85� did not exhibit a significant increase after 3days. Additionally, the cell proliferation increase after 3 days

FIGURE 8. Confocal fluorescence microscopy images of cells grown for 24 h on block structures of the Hybrid 2 material. Cell morphology is

shown following double-staining with TRITC-phalloidin (red) and FITC-conjugated anti-vinculin antibody (green). Arrows display the boundary

between the material block structure and the underlying glass surface. (a) shows the fibrillar network of actin cytoskeleton (40-fold magnifica-

tion), (b) demonstrates the vinculin participating in focal adhesion (40-fold magnification), (c) displays an overlay of the double-stain in 40-fold

magnification. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

FIGURE 9. Cell proliferation for the three materials investigated in

this study as shown in a cell viability and proliferation assay by

means of the PrestoBlueTM. Error bars represent the average of tripli-

cates 6 STDV in three independent experiments. Optical density val-

ues are normalized to the cell number according to a calibration

curve. (**) symbols denote significant differences (p < 0.05) according

to statistical analysis by one-way ANOVA Dunnett’s test. [Color figure

can be viewed in the online issue, which is available at

wileyonlinelibrary.com.]

10 TERZAKI ET AL. PRE-OSTEOBLASTIC CELL RESPONSE

on Hybrid 1 material can be attributed to the softening ofthe material following immersion in culture medium. Onday 7, cell proliferation increased on all materials withoutstatistically significant differences. After 7 days we observedcomplete coverage with cells reaching a maximal number.

All cell adhesion, viability, and proliferation experimentswere performed by using alpha-MEM culture medium withonly 1% FBS (instead of the 10% typically used) to avoidsupporting cell adhesion and proliferation. Therefore, ourobservations on the cellular response reflect the pure inter-action between materials and cells without being affectedby the positive influence of high serum concentrations oncell attachment and growth.

CONCLUSIONS

Novel hybrid and organic materials of different chemicalcomposition have been investigated, and their mechanicalproperties and biological response in pre-osteoblastic cellswere demonstrated. All materials were found to formsmooth films of a moderately hydrophilic character andindicated good mechanical properties. On these films, cellswere observed to adhere well and demonstrated increasedproliferation. Although all three investigated materials couldpotentially be used as biomaterials in tissue engineering,the Hybrid 2 composite containing 50% DMAEMA exhibitsthe best mechanical properties for structuring by DirectLaser Writing and strong biological response. In particular,the Hybrid 2 material exhibited excellent mechanical prop-erties, especially after submersion into the culture medium.A good pre-osteoblastic cell adhesion on the Hybrid 2 mate-rials scaffolds with spread cell morphology from the firsthour of observation until up to several days, together with aproliferation increase after 3 and 7 days reflect the biocom-patibility of these materials. Finally, we demonstrated theability to fabricate 3D structures by two-photon polymeriza-tion using the Hybrid 2 material, opening new possibilitiesfor the development of high-precision scaffolds with com-plex geometries and architectures for engineered tissues.

ACKNOWLEDGMENTS

Authors acknowledge Prof. H.P. Jennissen and Dr. M. Laub forthe generous donation of MC3T3-E1 cells and Maria Manousi-daki for structuring the ballerina by the DLW technology. Wewould like to thank Mrs. Aleka Manousaki and Ms. AlexandraSiakouli for expert technical assistance with SEM.

REFERENCES1. Kim K, Dean D, Wallace J, Breithaupt R, Mikos AG, Fisher JP. The

influence of stereolithographic scaffold architecture and composi-

tion on osteogenic signal expression with rat bone marrow stro-

mal cells. Biomaterials 2011;32:3750–3763.

2. Takai E, Costa KD, Shaheen A, Hung CT, Guo XE. Osteoblast Elas-

tic Modulus Measured by Atomic Force Microscopy Is Substrate

Dependent. Annals of Biomedical Engineering 2005;33:963–971.

3. Docheva D, Padula D, Popov C, Mutschler W, Clausen-Schau-

mann H, Schieker M. Researching into the cellular shape, volume

and elasticity of mesenchymal stem cells, osteoblasts and osteo-

sarcoma cells by atomic force microscopy. J Cell Mol Med 2008;

12:537–552.

4. Hanjaya-Putra D, Yee J, Ceci D, Truitt R, Yee D, Gerecht S. Vascu-

lar endothelial growth factor and substrate mechanics regulate in

vitro tubulogenesis of endothelial progenitor cells. J Cell Mol

Med 2010;14:2436–2447.

5. Kong HJ, Polte TR, Alsberg E, Mooney DJ. FRET measurements

of cell-traction forces and nano-scale clustering of adhesion

ligands varied by substrate stiffness. P Natl Acad Sci USA 2005;

102:4300–4305.

6. Khatiwala CB, Peyton SR, Putnam AJ. Intrinsic mechanical prop-

erties of the extracellular matrix affect the behavior of pre-osteo-

blastic MC3T3-E1 cells. Am J Physiol-Cell Ph 2006;290:C1640–C50.

7. Betz RR. Limitations of autograft and allograft: New synthetic sol-

utions. Orthopedics 2002;25:S561–S70.

8. Khan Y, Yaszemski MJ, Mikos AG, Laurencin CT. Tissue engineer-

ing of bone: Material and matrix considerations. J Bone Joint

Surg Am 2008;90A:36–42.

9. Balamurugan A, Kannan S, Rajeswari S. Structural and electro-

chemical behaviour of sol-gel zirconia films on 316L stainless-

steel in simulated body fluid environment. Materials Letters 2003;

57:4202–4205.

10. Filiaggi MJ, Pilliar RM, Yakubovich R, Shapiro G. Evaluating sol-

gel ceramic thin films for metal implant applications .1. Process-

ing and structure of zirconia films on Ti-6Al-4V. J Biomed Mater

Res 1996;33:225–238.

11. Chevalier J, Grandjean S, Kuntz M, Pezzotti G. On the kinetics and

impact of tetragonal to monoclinic transformation in an alumina/

zirconia composite for arthroplasty applications. Biomaterials

2009;30:5279–5282.

12. Lim HB, Oh KS, Kim YK, Lee DY. Characteristics of hydrothermal

stability and machinability of t-ZrO2/Al2O3 composites as a femo-

ral head for total hip replacements. Mat Sci Eng a-Struct 2008;

483–84:297–301.

13. Lowe AB, Vamvakaki M, Wassall MA, Wong L, Billingham NC, Armes

SP, et al. Well-defined sulfobetaine-based statistical copolymers as

potential antibioadherent coatings. J BiomedMater Res 2000;52:88–94.

14. Green JBD, Bickner S, Carter PW, Fulghum T, Luebke M, Nord-

haus MA, et al. Antimicrobial Testing for Surface-Immobilized

Agents With a Surface-Separated Live-Dead Staining Method. Bio-

technol Bioeng 2011;108:231–236.

15. Rawlinson LAB, Ryan SM, Mantovani G, Syrett JA, Haddleton

DM, Brayden DJ. Antibacterial Effects of Poly(2-(dimethylamino

ethyl)methacrylate) against Selected Gram-Positive and Gram-

Negative Bacteria. Biomacromolecules 2010;11:443–453.

16. Jung KH, Kim Y, Lim Y. Preparation of Antibacterial Polyester

Glass Mat Sheets Containing Pdmaema-Functionalized Mwnt

Nanocomposites. International Journal of Modern Physics B 2011;

25:4311–4314.

17. Synatschke CV, Schallon A, Jerome V, Freitag R, Muller AHE.

Influence of polymer architecture and molecular weight of poly(2-

(dimethylamino)ethyl methacrylate) polycations on transfection

efficiency and cell viability in gene delivery. Biomacromolecules

2011;12:4247–4255.

18. Ma PX. Biomimetic materials for tissue engineering. Adv Drug

Deliver Rev 2008;60:184–198.

19. Laurencin CT, Attawia MA, Elgendy HE, Herbert KM. Tissue engi-

neered bone-regeneration using degradable polymers: The forma-

tion of mineralized matrices. Bone 1996;19:S93–S9.

20. Thomson RC, Yaszemski MJ, Powers JM, Mikos AG. Hydroxyapa-

tite fiber reinforced poly(alpha-hydroxy ester) foams for bone

regeneration. Biomaterials 1998;19:1935–1943.

21. Russo T, Gloria A, D’Anto V, D’Amora U, Ametrano G, Bollino F,

et al. Poly(epsilon-caprolactone) reinforced with sol-gel synthe-

sized organic-inorganic hybrid fillers as composite substrates for

tissue engineering. J Appl Biomater Biom 2010;8:146–152.

22. Abbott A. Cell culture: Biology’s new dimension. Nature 2003;424:

870–872.

23. Juodkazis S, Mizeikis V, Misawa H. Three-dimensional microfabri-

cation of materials by femtosecond lasers for photonics applica-

tions. J Appl Phys 2009;106:051101.

24. Tan DF, Li Y, Qi FJ, Yang H, Gong QH, Dong XZ, et al. Reduction

in feature size of two-photon polymerization using SCR500. Appl

Phys Lett 2007;90:071106.

25. Farsari M, Vamvakaki M, Chichkov BN. Multiphoton polymeriza-

tion of hybrid materials. J Opt 2010;12:124001.

ORIGINAL ARTICLE

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH A | MONTH 2013 VOL 00A, ISSUE 00 11

26. Melissinaki V, Gill AA, Ortega I, Vamvakaki M, Ranella A, Haycock

JW, et al. Direct laser writing of 3D scaffolds for neural tissue en-

gineering applications. Biofabrication 2011;3:045005.

27. Engelhardt S, Hoch E, Borchers K, Meyer W, Kruger H, Tovar

GEM, et al. Fabrication of 2D protein microstructures and 3D

polymer-protein hybrid microstructures by two-photon polymer-

ization. Biofabrication 2011;3:025003.

28. Turunen S, Kapyla E, Terzaki K, Viitanen J, Fotakis C, Kellomaki

M, et al. Pico- and femtosecond laser-induced crosslinking of

protein microstructures: evaluation of processability and bioactiv-

ity. Biofabrication 2011;3:045002.

29. Sakellari I, Gaidukeviciute A, Giakoumaki A, Gray D, Fotakis C,

Farsari M, et al. Two-photon polymerization of titanium-contain-

ing sol-gel composites for three-dimensional structure fabrication.

Appl Phys a-Mater 2010;100:359–364.

30. Oliver WC, Pharr GM. An Improved Technique for Determining

Hardness and Elastic-Modulus Using Load and Displacement

Sensing Indentation Experiments. J Mater Res 1992;7:1564–1583.

31. King RB. Elastic Analysis of Some Punch Problems for a Layered

Medium. Int J Solids Struct 1987;23:1657–1664.

32. Ovecoglu ML, Doerner MF, Nix WD. Elastic Interactions of Screw

Dislocations in Thin-Films on Substrates. Acta Metall Mater 1987;

35:2947–2957.

33. Loubet JL, Georges JM, Marchesini O, Meille G. Vickers Indenta-

tion Curves of Magnesium-Oxide (Mgo). J Tribol-T Asme 1984;

106:43–48.

34. Fabes BD, Oliver WC. Mechanical-Properties of Sol-Gel Coatings.

J Non-Cryst Solids 1990;121:348–356.

35. Charitidis CA. Nanomechanical and nanotribological properties of

carbon-based thin films: A review. Int J Refract Met H 2010;28:

51–70.

36. Rabkin E, Deuschle JK, Baretzky B. On the nature of displacement

bursts during nanoindentation of ultrathin Ni films on sapphire.

Acta Mater 2010;58:1589–1598.

37. Chatzinikolaidou M, Lichtinger TK, Muller RT, Jennissen HP. Peri-

implant reactivity and osteoinductive potential of immobilized

rhBMP-2 on titanium carriers. Acta Biomater 2010;6:4405–4421.

38. Vamvakaki M, Unali GF, Butun V, Boucher S, Robinson KL, Bill-

ingham NC, et al. Effect of partial quaternization on the aqueous

solution properties of tertiary amine-based polymeric surfactants:

Unexpected separation of surface activity and cloud point behav-

ior. Macromolecules 2001;34:6839–6841.

39. Saha R, Nix WD. Soft films on hard substrates - nanoindentation

of tungsten films on sapphire substrates. Mat Sci Eng a-Struct

2001;319:898–901.

40. Soppera O, Feuillade M, Croutxe-Barghorn C, Carre C. Mechanical

properties of UV-photopolymerizable hybrid sol-gel films investigated

by AFM in pulsed forcemode. Prog Solid State Ch 2005;33:233–242.

41. Alvarado-Rivera J, Munoz-Saldana J, Ramirez-Bon R. Determina-

tion of fracture toughness and energy dissipation of SiO2-poly(-

methyl metacrylate) hybrid films by nanoindentation. Thin Solid

Films 2011;519:5528–5534.

42. Emileh A, Vasheghani-Farahani E, Imani M. Swelling behavior, me-

chanical properties and network parameters of pH- and temperature-

sensitive hydrogels of poly((2-dimethyl amino) ethyl methacrylate-

co-butyl methacrylate). Eur Polym J 2007;43:1986–1995.

43. Bautista Y, Gomez MP, Ribes C, Sanz V. Relation between the

scratch resistance and the chemical structure of organic-inorganic

hybrid coatings. Prog Org Coat 2011;70:358–364.

44. Etienne-Calas S, Duri A, Etienne P. Fracture study of organic-inor-

ganic coatings using nanoindentation technique. J Non-Cryst Sol-

ids 2004;344:60–65.

45. Gourmand M, Corpart JM. Superabsorbent revolution in hygiene.

Actual Chimique 1999:46–50.

46. Jensen OM, Hansen PF. Water-entrained cement-based materials

I. Principles and theoretical background. Cement Concrete Res

2001;31:647–654.

47. Dong LC, Hoffman AS. A Novel-Approach for Preparation of Ph-

Sensitive Hydrogels for Enteric Drug Delivery. J Control Release

1991;15:141–152.

48. Colombo P. Swelling-Controlled Release in Hydrogel Matrices for

Oral Route. Adv Drug Deliver Rev 1993;11:37–57.

49. Kuzma P, MooYoung AJ, Moro D, Quandt H, Bardin CW, Schlegel

PH. Subcutaneous hydrogel reservoir system for controlled drug

delivery. Macromol Symp 1996;109:15–26.

50. Moreira PJ, Marques PVS, Leite AP. Hybrid sol-gel channel wave-

guide patterning using photoinitiator-free materials. Ieee Photonic

Tech L 2005;17:399–401.

51. Lavenus S, Pilet P, Guicheux J, Weiss P, Louarn G, Layrolle P.

Behaviour of mesenchymal stem cells, fibroblasts and osteoblasts

on smooth surfaces. Acta Biomater 2011;7:1525–1534.

12 TERZAKI ET AL. PRE-OSTEOBLASTIC CELL RESPONSE