Composite biocompatible hydroxyapatite–silk fibroin coatings for medical implants obtained by...

8
Materials Science and Engineering B 169 (2010) 151–158 Contents lists available at ScienceDirect Materials Science and Engineering B journal homepage: www.elsevier.com/locate/mseb Composite biocompatible hydroxyapatite–silk fibroin coatings for medical implants obtained by Matrix Assisted Pulsed Laser Evaporation F.M. Miroiu a,, G. Socol a , A. Visan a , N. Stefan a , D. Craciun a , V. Craciun a , G. Dorcioman a , I.N. Mihailescu a , L.E. Sima b , S.M. Petrescu b , A. Andronie c , I. Stamatin c , S. Moga d , C. Ducu d a National Institute for Lasers, Plasma, and Radiation Physics, 409 Atomistilor Street, RO-77125, MG-36, Magurele-Ilfov, Romania b Institute of Biochemistry, Romanian Academy, 296 Splaiul Independentei, 060031 Bucharest, Romania c 3Nano-SAE Alternative Energy Sources-University of Bucharest, Faculty of Physics, 409 Atomistilor Street, RO-77125, Magurele-Ilfov, Romania d University of Pitesti, Targul din Vale Str, no. 1, 110040 Pitesti, Romania article info Article history: Received 10 June 2009 Received in revised form 5 October 2009 Accepted 7 October 2009 Keywords: Biomimetic Thin films Hydroxyapatite Fibroin Implants Matrix Assisted Pulsed Laser Evaporation (MAPLE) abstract The aim of this study was to obtain biomimetic inorganic–organic thin films as coatings for metallic med- ical implants. These contain hydroxyapatite, the inorganic component of the bony tissues, and a natural biopolymer – silk fibroin – added in view to induce the surface functionalization. Hydroxyapatite (HA), silk fibroin (FIB) and composite HA–FIB films were obtained by Matrix Assisted Pulsed Laser Evaporation (MAPLE) in order to compare their physical and biological performances as coatings on metallic pros- theses. We used an excimer laser source (KrF*, = 248 nm, = 25 ns) operated at 10 Hz repetition rate. Coatings were deposited on quartz, Si and Ti substrates and then subjected to physical (FTIR, XRD, AFM, SEM) analyses, correlated with the results of the cytocompatibility in vitro tests. The hybrid films were synthesized from frozen targets of aqueous suspensions with 3:2 or 3:4 weight ratio of HA:FIB. An appro- priate stoichiometric and functional transfer was obtained for 0.4–0.5 J/cm 2 laser fluence. FTIR spectra of FIB and HA–FIB films exhibited distinctive absorption maxima, in specific positions of FIB random coil form: 1540 cm 1 amide II, 1654 cm 1 amide I, 1243 cm 1 amide III, while the peak from 1027 cm 1 appeared only for HA and composite films. Osteosarcoma SaOs2 cells cultured 72 h on FIB and HA–FIB films showed increased viability, good spreading and normal cell morphology. The well-elongated, flat- tened cells are a sign of an appropriate interaction with the MAPLE FIB and composite HA–FIB coatings. © 2009 Elsevier B.V. All rights reserved. 1. Introduction It is already known that a composite hydroxyapatite–fibroin material should meet bioaffinity, enhanced osteoinductivity, ade- quate mechanical strength and flexibility for use as implant material. Hydroxyapatite (HA), with the chemical formula (Ca 10 (OH) 2 (PO 4 ) 6 ), constitutes the majority mineral part of nat- ural bones and teeth, providing innate excellent biocompatibility, osteoconductivity and bioactivity [1,2]. Therefore, it is largely stud- ied and used in different applications like bone fillers, maxillofacial reconstructions, dental applications or biomedical ceramic coatings [3–6]. On the other hand, silk fibroin (FIB), the main constituent of the natural silk, is a natural biopolymer, a protein spun in fibers by some lepidoptera larvae, as silkworms or spiders [7]. It shows three different conformations of peptide chains and a crystalline dimor- phism: amorphous random coil, crystalline -sheet and crystalline Corresponding author. Tel.: +40 21 4574491; fax: +40 21 4574243. E-mail address: marimona.miroiu@inflpr.ro (F.M. Miroiu). -helix forms [8]. The silk fibroin is known and proved to be bio- compatible, biodegradable, having unusual high tensile strength and elasticity. It has applications in scaffolds aimed to manipu- late the osseous growth in the desired forms, sutures, artificial skin, artificial tendons, and substrate for cell culture, bone tissue engineering, coatings [9–13]. Silk fibroin–hydroxyapatite is a quite new composite partic- ularly studied as 3D scaffolds, nanocomposites or thick coatings [14–18]. In our studies, silk fibroin is added to the hydroxyapatite in view to induce the surface functionalization of the coating and to improve the mechanical properties (elasticity). The HA–polymer composites mimic the natural bone, formed by the inorganic phase (nanometric biological HA) and organic compounds (mainly col- lagen). The intimate synergy between inorganic and organic phase provide the hard tissues with the providential features as high frac- ture toughness, flexibility and strength. The nanometric dimension of the inorganic element, of high specific surface, similar to the one in the bony apatite is important from the point of view of the mechanical and biological properties [19,20]. The aim of our work was to obtain biomimetic ceramic–polymer composite coatings for metallic medical implants. We have cho- 0921-5107/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.mseb.2009.10.004

description

The aim of this study was to obtain biomimetic inorganic–organic thin films as coatings for metallic medicalimplants. These contain hydroxyapatite, the inorganic component of the bony tissues, and a naturalbiopolymer – silk fibroin – added in view to induce the surface functionalization. Hydroxyapatite (HA),silk fibroin (FIB) and composite HA–FIB films were obtained by Matrix Assisted Pulsed Laser Evaporation(MAPLE) in order to compare their physical and biological performances as coatings on metallic prostheses.We used an excimer laser source (KrF*, = 248 nm, = 25 ns) operated at 10 Hz repetition rate.Coatings were deposited on quartz, Si and Ti substrates and then subjected to physical (FTIR, XRD, AFM,SEM) analyses, correlated with the results of the cytocompatibility in vitro tests. The hybrid films weresynthesized from frozen targets of aqueous suspensions with 3:2 or 3:4 weight ratio of HA:FIB. An appropriatestoichiometric and functional transfer was obtained for 0.4–0.5 J/cm2 laser fluence. FTIR spectraof FIB and HA–FIB films exhibited distinctive absorption maxima, in specific positions of FIB randomcoil form: 1540cm−1 amide II, 1654cm−1 amide I, 1243cm−1 amide III, while the peak from 1027cm−1appeared only for HA and composite films. Osteosarcoma SaOs2 cells cultured 72 h on FIB and HA–FIBfilms showed increased viability, good spreading and normal cell morphology. The well-elongated, flattenedcells are a sign of an appropriate interaction with the MAPLE FIB and composite HA–FIB coatings

Transcript of Composite biocompatible hydroxyapatite–silk fibroin coatings for medical implants obtained by...

Page 1: Composite biocompatible hydroxyapatite–silk fibroin coatings for medical  implants obtained by Matrix Assisted Pulsed Laser Evaporation

Ci

FIa

b

c

d

a

ARRA

KBTHFIM(

1

mqm(uoir[

tsdp

0d

Materials Science and Engineering B 169 (2010) 151–158

Contents lists available at ScienceDirect

Materials Science and Engineering B

journa l homepage: www.e lsev ier .com/ locate /mseb

omposite biocompatible hydroxyapatite–silk fibroin coatings for medicalmplants obtained by Matrix Assisted Pulsed Laser Evaporation

.M. Miroiua,∗, G. Socola, A. Visana, N. Stefana, D. Craciuna, V. Craciuna, G. Dorciomana,.N. Mihailescua, L.E. Simab, S.M. Petrescub, A. Androniec, I. Stamatinc, S. Mogad, C. Ducud

National Institute for Lasers, Plasma, and Radiation Physics, 409 Atomistilor Street, RO-77125, MG-36, Magurele-Ilfov, RomaniaInstitute of Biochemistry, Romanian Academy, 296 Splaiul Independentei, 060031 Bucharest, Romania3Nano-SAE Alternative Energy Sources-University of Bucharest, Faculty of Physics, 409 Atomistilor Street, RO-77125, Magurele-Ilfov, RomaniaUniversity of Pitesti, Targul din Vale Str, no. 1, 110040 Pitesti, Romania

r t i c l e i n f o

rticle history:eceived 10 June 2009eceived in revised form 5 October 2009ccepted 7 October 2009

eywords:iomimetichin filmsydroxyapatiteibroin

a b s t r a c t

The aim of this study was to obtain biomimetic inorganic–organic thin films as coatings for metallic med-ical implants. These contain hydroxyapatite, the inorganic component of the bony tissues, and a naturalbiopolymer – silk fibroin – added in view to induce the surface functionalization. Hydroxyapatite (HA),silk fibroin (FIB) and composite HA–FIB films were obtained by Matrix Assisted Pulsed Laser Evaporation(MAPLE) in order to compare their physical and biological performances as coatings on metallic pros-theses. We used an excimer laser source (KrF*, � = 248 nm, � = 25 ns) operated at 10 Hz repetition rate.Coatings were deposited on quartz, Si and Ti substrates and then subjected to physical (FTIR, XRD, AFM,SEM) analyses, correlated with the results of the cytocompatibility in vitro tests. The hybrid films weresynthesized from frozen targets of aqueous suspensions with 3:2 or 3:4 weight ratio of HA:FIB. An appro-

2

mplantsatrix Assisted Pulsed Laser Evaporation

MAPLE)

priate stoichiometric and functional transfer was obtained for 0.4–0.5 J/cm laser fluence. FTIR spectraof FIB and HA–FIB films exhibited distinctive absorption maxima, in specific positions of FIB randomcoil form: 1540 cm−1 amide II, 1654 cm−1 amide I, 1243 cm−1 amide III, while the peak from 1027 cm−1

appeared only for HA and composite films. Osteosarcoma SaOs2 cells cultured 72 h on FIB and HA–FIBfilms showed increased viability, good spreading and normal cell morphology. The well-elongated, flat-

n app

tened cells are a sign of a

. Introduction

It is already known that a composite hydroxyapatite–fibroinaterial should meet bioaffinity, enhanced osteoinductivity, ade-

uate mechanical strength and flexibility for use as implantaterial. Hydroxyapatite (HA), with the chemical formula

Ca10(OH)2(PO4)6), constitutes the majority mineral part of nat-ral bones and teeth, providing innate excellent biocompatibility,steoconductivity and bioactivity [1,2]. Therefore, it is largely stud-ed and used in different applications like bone fillers, maxillofacialeconstructions, dental applications or biomedical ceramic coatings3–6].

On the other hand, silk fibroin (FIB), the main constituent of

he natural silk, is a natural biopolymer, a protein spun in fibers byome lepidoptera larvae, as silkworms or spiders [7]. It shows threeifferent conformations of peptide chains and a crystalline dimor-hism: amorphous random coil, crystalline �-sheet and crystalline

∗ Corresponding author. Tel.: +40 21 4574491; fax: +40 21 4574243.E-mail address: [email protected] (F.M. Miroiu).

921-5107/$ – see front matter © 2009 Elsevier B.V. All rights reserved.oi:10.1016/j.mseb.2009.10.004

ropriate interaction with the MAPLE FIB and composite HA–FIB coatings.

© 2009 Elsevier B.V. All rights reserved.

�-helix forms [8]. The silk fibroin is known and proved to be bio-compatible, biodegradable, having unusual high tensile strengthand elasticity. It has applications in scaffolds aimed to manipu-late the osseous growth in the desired forms, sutures, artificialskin, artificial tendons, and substrate for cell culture, bone tissueengineering, coatings [9–13].

Silk fibroin–hydroxyapatite is a quite new composite partic-ularly studied as 3D scaffolds, nanocomposites or thick coatings[14–18]. In our studies, silk fibroin is added to the hydroxyapatitein view to induce the surface functionalization of the coating andto improve the mechanical properties (elasticity). The HA–polymercomposites mimic the natural bone, formed by the inorganic phase(nanometric biological HA) and organic compounds (mainly col-lagen). The intimate synergy between inorganic and organic phaseprovide the hard tissues with the providential features as high frac-ture toughness, flexibility and strength. The nanometric dimension

of the inorganic element, of high specific surface, similar to theone in the bony apatite is important from the point of view of themechanical and biological properties [19,20].

The aim of our work was to obtain biomimetic ceramic–polymercomposite coatings for metallic medical implants. We have cho-

Page 2: Composite biocompatible hydroxyapatite–silk fibroin coatings for medical  implants obtained by Matrix Assisted Pulsed Laser Evaporation

152 F.M. Miroiu et al. / Materials Science and

Table 1Selected samples of hydroxyapatite, fibroin and composite thin films deposited byMAPLE.

Code Sample type

FIB5 Fibroin (5 wt% in suspension)HA5 Hydroxyapatite (5 wt%)FIBNaOH Fibroin (5% suspension) with NaOH added

ssonLt

2a

2

a

Fts

FIBNaCl Fibroin (5% suspension) with NaCl addedHA3–FIB2 Hydroxyapatite (3 wt%)–fibroin (2 wt%)HA3–FIB4 Hydroxyapatite (3 wt%)–fibroin (4 wt%)C Control–borosilicate glass

en to synthesize the hybrid hydroxyapatite–polymer coatings andimple hydroxyapatite and simple polymer coatings, respectivelyn titanium substrates (as basis of the orthopedical implants) by aovel fabrication method for this composite, Matrix Assisted Pulsedaser Deposition (MAPLE), as being the most appropriate for syn-hesis of large organic molecule thin films [21].

. Experimental procedure: methods, materials andnalyses

.1. Matrix Assisted Pulsed Laser Evaporation (MAPLE) method

MAPLE is an advanced laser technique, based on a cryogenicpproach, which has been developed since 1998, to produce a “pro-

ig. 1. FTIR spectra of (a) HA, FIB and hybrid HA–FIB films on silicon, comparedo those of fibroin and HA powders; (b) fibroin thin films (from different fibroinolutions) on silicon compared with the original powder.

Engineering B 169 (2010) 151–158

tected” accurate transfer of organic and polymeric materials in formof thin films [22,23].

The material subjected to the laser irradiation – which is called“target” – is a frozen composite. It is obtained by freezing fol-lowing the dissolution of the “active” biomaterial (up to 5 wt%) inan appropriate volatile solvent, highly absorbing the laser wave-length, but not reacting under laser exposure. The laser pulsesintensities are adjusted to avoid the biomolecules damage. Thestructural and functional fidelity is preserved inducing a non-directlaser–material interaction in a vacuum chamber. Due to the lowconcentration of biomolecules in the frozen target, the laser pho-tons preponderantly interact with the matrix (the solvent), which isvaporized. The complex biomolecules are released undamaged and,by means of the collisions with the other molecules, moved towardthe substrate, where they form a uniform thin film. In the sametime, the volatile solvent is pumped away by the vacuum system.During the deposition process the target is kept at low temperatureby a cooler.

As its precursor, Pulsed Laser Deposition, MAPLE is a layer-by-layer growing method, which can deposit thin films with dopantsand may also be applied with masks, in order to induce a controlleddistribution of the local structure parameters by the nature andtype of the coating.

2.2. Materials, set-up and MAPLE experiments

We used in our experiments commercial 2 �m granulation poly-mer powder of degummed (i.e. sericin-free) Bombyx mori fibroin(Wuxi Allied Technologies, Inc., China), and commercial Merckhydroxyapatite powder.

For MAPLE experiments we prepared different fibroin andhybrid solutions or suspensions. According to literature, thewater-insoluble fibroin can be solubilised in organic substances(N-methyl morpholine N-oxide, MMNO N-dimethyl acetamide,copper-ethylendiamine, Ca(NO3)2/MeOH) or in saturated solutionsof calcium chloride or lithium bromide, simple or with ethylic alco-hol [24–26]. Thus, a number of films were grown from fibroinsubjected firstly to a solubilisation process, and then to one ofseparation.

In view to homogenise them, the calcium chloride and lithiumbromide solutions were heated at 95 ◦C [27] and 60 ◦C [28] and keptat these temperatures for 8 and 4 h, respectively. Subsequently, toseparate the fibroin from salt, the solutions were subjected to achemical dialysis process in bidistilled water, for 4 days. In the end,

Fig. 2. XRD spectra of the fibroin and HA–fibroin thin films on silicon substratesdeposited by MAPLE.

Page 3: Composite biocompatible hydroxyapatite–silk fibroin coatings for medical  implants obtained by Matrix Assisted Pulsed Laser Evaporation

F.M. Miroiu et al. / Materials Science and Engineering B 169 (2010) 151–158 153

F es (ups

t1

ramscr

M2w

a3hpw

fip

aottn

1

TRe

ig. 3. AFM images of fibroin and HA–fibroin MAPLE thin films on quartz substratamples, 60 �m × 60 �m).

he dialysis membranes kept only the fibroin molecules larger than0,000 Da, i.e. the aqueous fibroin solutions.

Alternatively, a quantity of salty-fibroin solutions were sepa-ated by centrifugation with a Universal 320R Hettich Zentrifugen,t 4 ◦C and 6000 rpm, for 30 min. Supplementary, the lithium bro-ide solution was centrifuged 20 min at 9000 rpm, obtaining a pure

olubilised fibroin. The calcium chloride solution was additionallyentrifuged for 30 min at 9000 rpm, however without being sepa-ated.

The third method to get the polymer (fibroin) frozen target forAPLE technique was to obtain fibroin aqueous suspensions, of

–5 fibroin wt%. The very homogenous aspect of the suspensionas kept by using a mechanical agitator.

For the hybrid HA–fibroin films, we used polymer solutions,dding the hydroxyapatite up to a fixed HA/polymer weight ratio,/2 or 3/4. The solutions were mechanically stirred, in view of theomogenisation. In order to reach the optimum, neutral pH 7.4,referred by cells, in some samples a few drops of NaOH or NaClere added, neutralizing the slightly acid fibroin suspensions.

By comparison, HA coatings were deposited by MAPLE frombroin (5 wt%) aqueous suspension, using the same experimentalarameters as for the fibroin-containing coatings.

The codes of the samples are mentioned in the Table 1. Inll cases, according to MAPLE technology, the solid targets werebtained by solutions/suspensions frozen in liquid nitrogen andheir maintenance, for the period of the deposition process, under

he melting point, by means of a cooler device, fed also with liquiditrogen.

An excimer laser source KrF*, � = 248 nm, �FWHM = 25 ns, with0–15 Hz pulses repetition rate was used in our MAPLE experi-

able 2oughness values of MAPLE fibroin and HA–fibroin coatings deposited on quartz,stimated by AFM.

Sample/roughness, 15 �m × 15 �m Fibroin HA3–FIB4 HA3–FIB2

Peak-to-peak (nm) 106 220 397Average (nm) 26 95 171

: FIB, HA3–FIB4, HA3–FIB2 samples, 15 �m × 15 �m; down: HA3–FIB4, HA3–FIB2

ments. The laser fluence incident on the sample was set between0.4 and 0.5 J/cm2 and the pressure inside the vacuum chamberkept at 7 Pa. The deposition substrates – medical grade etched tita-nium, quartz or double polished Si (1 1 1) – were placed at 4 cmdistance, parallel to the target and maintained at 30 ◦C. A seriesof 20,000 or 50,000 laser pulses was applied for each depositedfilm.

The coatings have been estimated to have about 900 nmthickness for 50,000 subsequent pulses, as resulted from theUV–vis transmission spectra by measurements of the interferencefringes.

2.3. Analyses

The stoichiometry of the polymer and ofpolymer–hydroxyapatite coatings was assessed by Fouriertransform infrared spectroscopy and X-ray diffraction analyses,and the morphology by atomic force microscopy and scanningelectron microscopy investigations, respectively. Following thephysico-chemical characterization, all coatings were subjected toin vitro biological characterizations, in order to evaluate both theoptimal behaviour and their most favourable deposition regime.

Fourier transform infrared spectroscopy (FTIR) analyses wereobtained by means of a FTIR Shimadzu 8400S working in7800–350 cm−1 in transmission mode (samples on double-side pol-ished Si). Atomic force microscopy (AFM) records were performedby an Integrated Platform SPM—NTegra model Prima microscope,in semicontact mode and the scanning electron microscopy (SEM)images were taken with a SEM JEOL 6400 apparatus. X-raydiffraction (XRD) spectra were recorded with a Rigaku Ultima IVdiffractometer, using Cu K� radiation. The acquisition conditionswere the following: �–2� configuration and parallel beam optics,high voltage applied on the X-ray tube U = 40 kV, current intensity

I = 30 mA, angular 2� range of 8–60◦, �(2�) = 0.05◦, fixed angle ofincidence ˛ = 1◦.

MTS assay is a colorimetric method to determine cytotoxic-ity or the number of viable proliferating cells. In vitro tests areperformed by adding a small amount of reagent directly to cul-

Page 4: Composite biocompatible hydroxyapatite–silk fibroin coatings for medical  implants obtained by Matrix Assisted Pulsed Laser Evaporation

154 F.M. Miroiu et al. / Materials Science and Engineering B 169 (2010) 151–158

fibroi

t4eaoictcv

Fig. 4. SEM micrographs of hydroxyapatite, fibroin and HA–

ure wells, incubating for 1–4 h and then recording absorbance at90 nm. The reductions of the MTS salt takes place, in the pres-nce of phenazine methosulfate, only when reductase enzymes arective. Therefore the quantity of formazan measured by the amountf 490 nm absorbance is directly proportional to the number of liv-ng cells in culture. One hundred thousand osteosarcoma SaOs2

ells were seeded on each sample and cultured for 72 h on the sixypes of coated titanium surfaces and on the control borosilicateover glass control, respectively. MTS assay was then performed iniew to compare the viability of cells grown on the samples surface.

Fig. 5. MTS assay on SaOs2 cells grown on HA/Ti, FIB/Ti and HA–FIB/Ti MAPLE

n MAPLE thin films on Si (1 1 1) double polished substrates.

Results of the 490 nm absorbance measurements are representedin the diagram hereunder. Samples were tested in duplicate and foreach duplicate sample the reaction was quantified in triplicate.

The first in vitro fluorescence microscopy experiments were per-formed with the ER-Tracker dye. Identical samples to those testedbefore were incubated in a 1:1000 dye dilution in complete media

at 37 ◦C for 30 min, to allow its uptake and distribution into theendoplasmic reticulum. This experimental approach allows thelive analysis of the cell morphology as it does not require cellsfixation and the fluorescently marked organelle forms a network

deposited structures. C is the standard control borosilicate cover glass.

Page 5: Composite biocompatible hydroxyapatite–silk fibroin coatings for medical  implants obtained by Matrix Assisted Pulsed Laser Evaporation

e and

dPc

scsi

Fp

F.M. Miroiu et al. / Materials Scienc

istributed in almost all the cytoplasm volume. Titanium and HA-LD samples were tested using primary human osteoblasts cellsultured as described before [29].

Then SaOs2 cells were analysed for adhesion molecules expres-ion and localization by immunofluorescence. First, cells wereultured on the biomaterial surface and on the standard controlample of borosilicate cover glass, fixed and permeabilized accord-ng to a previously described protocol [30,31].

ig. 6. Fluorescence microscopy: (A) ER-Tracker staining of SaOs2 cells grown on MAPrimary osteoblast cells grown on Ti and PLD-deposited HA coatings.

Engineering B 169 (2010) 151–158 155

3. Results and discussion

3.1. Physico-chemical characterization

For the fibroin powder (Fig. 1a), the following absorption FTIRmaxima, specific to the fibroin beta-form, are evidenced: 1265(∼1260) cm−1, 1520 cm−1, 1625 (∼1630) cm−1, and 1698 (∼1703sau 1695) cm−1 [24,32–36], as well as an absorption band at

LE deposited fibroin/Ti, HA/Ti and HA–FIB/Ti coatings; (B) ER-Tracker staining of

Page 6: Composite biocompatible hydroxyapatite–silk fibroin coatings for medical  implants obtained by Matrix Assisted Pulsed Laser Evaporation

1 e and

7pifi

iact6

Ft

56 F.M. Miroiu et al. / Materials Scienc

00 cm−1 (amide V) [24]. At 612 cm−1 a rather small maximum ofoly (l-alanine) in beta-sheet form is observed [37]. Another max-

mum at 1235 cm−1, which is attributed to the random coil form ofbroin is also present [38].

The hydroxyapatite powder spectrum (Fig. 1a) exhibits max-

ma at: 564 cm−1 (∼570 cm−1) and 601 cm−1, assigned to thesymmetric bending PO4, 961 (∼954) cm−1 for the symmetri-al PO4 stretching vibration, a sharp maximum at 1031 cm−1 ofhe asymmetrical PO4 stretching vibration and two shoulders at30 cm−1 (∼631 cm−1) and ∼1081 cm−1, respectively, belonging to

ig. 7. Immunofluorescence microscopy: adhesion of SaOs2 cells stained for actin (red) aitanium, FIB5 and HA5 samples.

Engineering B 169 (2010) 151–158

PO4 group, as well as another maximum at 3566 (∼3562) cm−1,attributable to OH [39].

FTIR analysis provided fairly similar spectra for all fibroin films,independently of the provenience solution (Fig. 1b). It is knownthat thin films with low degree of crystallinity or stoichiometry

give broadening, shifting or diminishing of the bands’ intensity. Inthis respect, the coatings mostly resembling to the fibroin pow-der are those of the suspension with 5 wt% fibroin concentrationin water. This result suggests that MAPLE target preparation canbe made by the easiest method, of the aqueous fibroin suspen-

nd vinculin (green) markers; best results on HA3–FIB4/Ti sample in comparison to

Page 7: Composite biocompatible hydroxyapatite–silk fibroin coatings for medical  implants obtained by Matrix Assisted Pulsed Laser Evaporation

e and

stfirMrrs

rHctra∼(bprIsiais(

LHtpotpt

Mwfia[

ofit

3

s(dg

cfttwfiattMt

F.M. Miroiu et al. / Materials Scienc

ion, without the previous salty saturated solutions and withouthe corresponding complicated and costly techniques to separatebroin from these solutions. Hence, even if all the samples showedelatively the same positive behaviour of biocompatibility (data ofTS assay not shown herewith), the next series of experiments

esumed only to samples made by simple fibroin suspensions and,espectively, of hybrid hydroxyapatite–fibroin aqueous suspen-ions.

The FTIR spectra of the simple HA, FIB polymer and HA–FIB,espectively, revealed that the films derived from the suspensionsA3–FIB2 or HA3–FIB4 have maxima specific to both materialsomposing the hybrid (Fig. 1a). The fibroin films present distinc-ive amides absorption maxima, in specific positions of fibroinandom coil form, e.g. at 1540 cm−1 amide II, 1654 cm−1 (∼1660)mide I, 1243 cm−1 (∼1240) amide III, and strong absorption band3309 cm−1, due to NH stretching. The skeletal stretching region

1100–900 cm−1) includes a band at 1015 cm−1 and another twinand at 998 and 975 cm−1, from the –(gly-gly)- and –(gly-ala)-eriodic sequences of fibroin [24]. As the FIB and the HA films,espectively, the composite films show absorption for amide (I,I, III) fibroin random coil at 1654, 1540, 1243 cm−1, the NHtretching vibration at 3290 (∼3309) cm−1 and the PO4 stretch-ng vibration 1027–1030 cm−1. The maximum attributable to OH,t 3566 (∼3562) cm−1, appears in all the coatings, but has anntensity hardly visible. The HA film exhibited most visibly theharp maximum around 1031 cm−1 of the PO4 stretching vibrationFig. 1a).

As HA phase is known as amorphous when deposited by Pulsedaser Deposition (PLD) at room temperature [40–42], there are noA peaks in the XRD spectrum of the composite film. The XRD spec-

ra of fibroin and HA–fibroin coatings (HA3–FIB4) show a diffractioneak at 20.5◦, specific to �-sheet crystalline form of fibroin [32,37],ver a broad hallo centred around 21.5◦, specific also to the silk pro-ein, possible to �-helix form [37] (Fig. 2). The very weak emergingeak at 12◦ is attributed also to �-helix fibroin [32,37]. Howeverhe data suggest a predominantly amorphous content of fibroin.

The AFM images (Fig. 3) put in evidence morphology specific toAPLE films: homogeneous, but with a certain degree of roughness,hich is gradually decreasing when the silk particles compose thelms (Table 2). We mention that the topography with extendedctive surface area is favourable to the cell growth and adhesion43].

The SEM images of hydroxyapatite, fibroin and HA–fibroin filmsn Si (1 1 1) double polished substrates reveal a topography con-rming that of the AFM analysis, where the HA presence enhanceshe surface profile (Fig. 4).

.2. Biological in vitro assays

Cell viability does not present major variations between assayedamples. All tested samples are non-toxic during the 72 h of the testFig. 5). However, the samples which most resemble to the stan-ard control sample are FIB5, HA3–FIB2 and FIBNaOH. The sampleenerating the most reduced viability is FIBNaCl.

The data were confirmed by the morphology analysis of fluores-ence microscopy. A uniform cell distribution is observed in Fig. 6or FIB5 and HA3–FIB2 samples. Bone cells are very well spread onhe substrate, elongated and flattened, indicating a good interac-ion with the deposited films. A good interaction with the surface,hich confirms the viability data, is also exhibited by FIBNaOHlms. On the contrary, on the HA5 and FIBNaCl samples cells adopt

heterogeneous morphology, predominantly spherical, as a sign of

he lack of compatibility between their physical-chemical proper-ies and cell adhesion. We might therefore conclude that for these

APLE deposition parameters the pure HA films are less favourableo cell interaction and that the droplets of NaCl solution added to

Engineering B 169 (2010) 151–158 157

change the pH of the fibroin suspension are not favourable. Whenusing PLD HA thin films on titanium, growth of primary humanosteoblasts was not impaired as shown in Fig. 6B. The inset depictsa magnification to prove the optimal dye distribution in the ER. Asan intermediary behaviour, cells grown on HA3–FIB4 samples showa significant spreading on the films surface, but their distributionis less uniform and they are fewer, compared to FIB5, FIBNaOH andHA3–FIB2 samples. Bone cells attached to titanium but their prolif-eration was less efficient then on HA-containing coatings (Fig. 6B).

In view to evaluate adhesion of osteoblasts cultured in directcontact with the fibroin/Ti and HA–fibroin/Ti surfaces, SaOs2 cellswere stained for actin (red) and vinculin (green) markers and anal-ysed by immunofluorescence microscopy. The most favourablesurface proved to be HA3–FIB4/Ti, where actin and vinculin pat-terns are similar to the cells grown in standard conditions andon bare Ti (Fig. 7). FIB5 alone generates poor vinculin expressionand a decrease in the total number of contact foci. Actin filamentsare less stretched and cells spread less uniformly. Also, no specificstaining was observed on HA5 only coatings, which emit autoflu-orescent signals. This behaviour indicates optimal cell adhesion ofosteoblasts on HA–fibroin hybrid thin film.

4. Conclusions

The FTIR analyses provide similar spectra for all frozen targetsobtained from solutions of polymers, suggesting that preparationcan be done by the simplest method, an aqueous suspension offibroin without prior achievement of saturated salt solutions andsubsequent techniques, like complicated separation of fibroin ofthese solutions.

Compared to the simple fibroin or HA thin films, AFM andSEM analyses show an intermediary topography of the compos-ite HA–fibroin coatings, with micronic droplets which gives anextended surface, favourable to bone cells anchorage and prolif-eration. FTIR and XRD analyses, both for simple polymer films, aswell as for the HA–polymer hybrids, indicate a cvasistoichiometrictransfer of these compounds from the solid target in the form of thinfilms. The presence of the main FTIR absorption maxima, specificto the basic substances, validates that the transfer method chosen,MAPLE, is correct. The fibroin content in the films is a mixture ofpreponderantly random coil with crystalline forms, �-sheet and �-helix. In vitro viability tests prove appropriate bone cell behaviourfor the interaction with the selected MAPLE fibroin and HA–fibroincoatings: non-toxicity, good spreading and normal cell morphol-ogy.

The best performances, in terms of physico-chemical andbiological properties (viability, adhesion), were proved for the com-posite samples of HA3–FIB4 type. Before validation for furtheradvanced in vivo tests, in view of the orthopedical applicationsas hybrid biomimetic coatings, supplementary in vitro studies onosteoblast expression markers will be performed.

Acknowledgements

This work was supported under the Contract PN2 TD287/2007financed by the Romanian Ministry of Education, Research andYouth. F.M. Miroiu acknowledges with thanks the kind help of Mrs.Cristina Hlevca, Eleonora Gheorghiu and Vasilica Zaharachescufrom the National Institute for Chemical Pharmaceutical Researchand Development ICCF Bucharest. We thank Dr. Karine Anselme forthe kind gift of providing the SaOs2 cell line.

References

[1] J.C. Elliott, Structure and Chemistry of the Apatites and Other CalciumOrthophosphates, Elsevier, Amsterdam, 1994.

Page 8: Composite biocompatible hydroxyapatite–silk fibroin coatings for medical  implants obtained by Matrix Assisted Pulsed Laser Evaporation

1 e and

[

[[

[

[

[

[[[

[[

[

[

[

[[

[[

[

[

[[

[[[[[[

[[

[

[41] V. Nelea, C. Ristoscu, C. Chiritescu, C. Ghica, I.N. Mihailescu, A. Cornet, Appl.

58 F.M. Miroiu et al. / Materials Scienc

[2] W. Suchanek, M. Yoshimura, J. Mater. Res. 13 (1) (1998) 94–117.[3] G. Daculsi, O. Laboux, O. Malard, P. Weiss, J. Mater. Sci.: Mater. Med. 14 (3)

(2003) 195–200.[4] J. Lee, L. Rouhfar, O. Beirne, J. Oral Maxillofac. Surg. 58 (12) (2000) 1372–1379.[5] B. Leon, J. Jansen (Eds.), Thin Calcium Phosphate Coatings for Medical Implants,

Springer, 2008.[6] V. Nelea, M. Jelinek, I.N. Mihailescu, in: R. Eason (Ed.), Pulsed Laser Deposition

of Thin Films: Applications-led Growth of Functional Materials, Wiley & Sons,Hoboken, New Jersey, 2006, pp. 421–459.

[7] M. Mondal, K. Trivedy, S. Nirmal Kumar, Caspian J. Environ. Sci. 5 (2) (2007)63–76.

[8] C. Mo, P. Wu, X. Chen, Z. Shao, Appl. Spectroc. 60 (12) (2006) 1438–1441.[9] R. Nazarov, H.-J. Jin, D.L. Kaplan, Biomacromolecules 5 (2004) 718–726.10] S. Hofmnn, H. Hagenmüller, A.M. Koch, et al., Biomaterials 28 (2007)

1152–1162.11] Q. Lu, Q. Feng, K. Hu, F. Cui, J. Mater. Sci.: Mater. Med. 19 (2008) 629–634.12] J. Perez-Rigueiro, C. Viney, J. Llorca, M. Elices, J. Appl. Polym. Sci. 70 (1998)

2439–2447.13] S. Sofia, M.B. McCarthy, G. Gronowicz, D.L. Kaplan, J. Biomed. Mater. Res. 54

(2001) 139–148.14] T. Furuzono, S. Yasuda, T. Kimura, S. Kyotani, J. Tanaka, A. Kishida, J. Artif. Organs

7 (3) (2004) 137–144.15] X.D. Kong, F.Z. Cui, X.M. Wang, M. Zhang, W. Zhang, J. Cryst. Growth 270 (1–2)

(2004) 197–202.16] X.D. Kong, X.M. Wang, X. Yu, F.Z. Cui, Key Eng. Mater. 288–289 (2005) 191–194.17] R. Kino, T. Ikoma, S. Yunoki, J. Biosci. Bioeng. 103 (2007) 514–520.18] T. Gde, T. Nindhia, Y. Koyoshi, A. Kaneko, H. Sawada, M. Ohta, S. Hirai, M. Uo,

Trends Biomater. Artif. Organs 22 (1) (2008) 25–29.19] J.Y. Rho, L. Kuhn-Spearing, P. Zioupos, Med. Eng. Phys. 20 (1998) 92–102.20] Y. Cai, Y. Liu, W. Yan, Q. Hu, J. Tao, M. Zhang, Z. Shic, R. Tang, J. Mater. Chem. 17

(2007) 3780–3787.

21] M. Jelinek, T. Kocourek, J. Remsa, R. Cristescu, I.N. Mihailescu, D.B. Chrisey, Laser

Phys. 17 (2) (2007) 66–70.22] A. Pique, Deposition of polymers and biomaterials using the Matrix-Assisted

Pulsed Evaporation (MAPLE) process, in: R. Eason (Ed.), Pulsed Laser Depositionof Thin Films. Applications-Led Growth of Functional Materials, Wiley-Interscience, Hoboken, New Jersey, 2007, pp. 63–84.

[

[

Engineering B 169 (2010) 151–158

23] R. Cristescu, I.N. Mihailescu, M. Jelinek, D.B. Chrisey, in: R. Kassing, P. Petkov, W.Kulisch, C. Popov (Eds.), Functionalized Properties of Nanostructured Materials,NATO Science Series by Springer, Series II: Mathematics, Physics and Chemistry,223, 2006, pp. 211–226.

24] G. Freddi, G. Pessina, M. Tsukada, Int. J. Biol. Macromol. 24 (1999) 251–263.25] Y.-Q. Zhang, W.-L. Zhou, W.-D. Shen, Y.-H. Chen, X.-M. Zha, K. Shirai, K. Kiguchi,

J. Biotechnol. 120 (2005) 315–326.26] B. Zuo, L. Liu, Z. Wu, J. Appl. Polym. Sci. 106 (2007) 53–59.27] J.-H. Yeoa, K-g. Leea, Y.-W. Leea, S.Y. Kim, Eur. Polym. J. 39 (6) (2003) 1195–

1199.28] A. Vasconcelos, G. Freddi, A. Cavaco-Paulo, Biomacromolecules 9 (2008)

1299–1305.29] L.E. Sima, A. Filimon, R.M. Piticescu, G.C. Chitanu, D.M. Suflet, M. Miroiu, G.

Socol, I.N. Mihailescu, J. Neamtu, G. Negroiu, J. Mater. Sci. Mater. Med. (2009June 20), 10.1007/s10856-009-3800-7.

30] G. Negroiu, R.A. Dwek, S.M. Petrescu, J. Biol. Chem. 278 (2003) 27035–27042.31] G. Negroiu, R.M. Piticescu, G.C. Chitanu, I.N. Mihailescu, L. Zdrentu, M. Miroiu,

J. Mater. Sci.: Mater. Med. 19 (4) (2008) 1537–1544.32] Q. Lv, C. Cao, H. Zhu, Polym. Int. 54 (2005) 1076–1081.33] Y. Iridag, M. Kazanci, J. Appl. Polym. Sci. 100 (2006) 4260–4264.34] S. Nakayama, S. Nagare, M. Senna, Thin Solid Films 515 (2006) 2582–2586.35] X. Chen, D.P. Knight, Z. Shao, F. Vollrath, Polymers 42 (2001) 09969–09974.36] M. Yuji, H. Jianen, Food Sci. Technol. Res. 11 (1) (2005) 37–42.37] G. Freddi, P. Monti, M. Nagura, Y. Gotoh, M. Tsukada, J. Polym. Sci. B: Polym.

Phys. 35 (1997) 841–847.38] K. Jamada, J. Tsuboi, Thin Solid Films 440 (2003) 208–216.39] P. Layrolle, G. Daculsi, in: B. Leon, J. Jansen (Eds.), Thin Calcium Phosphate

Coatings for Medical Implants, Springer, 2008, p. 13, p27.40] E.N. Antonov, V.N. Bagratashvili, V.K. Popov, E.N. Sobol, S.M. Howdle, Spec-

trochim. Acta A: Mol. Biomol. Spectrosc. 52 (1) (1996) 123–127.

Surf. Sci. 168 (1–4) (2000) 127–131.42] M. Katto, M. Nakamura, T. Tanaka, T. Nakahyama, Appl. Surf. Sci. 197–198

(2002) 768–771.43] D.M. Brunette, P. Tengvall, M. Textor, P. Thomsen (Eds.), Titanium in Medicine,

Springer, Berlin, 2001.