Osteoblast response on Ti- and Zr-based bulk metallic glass surfacesafter sand blasting modification

H. F. Li,1 Y. B. Wang,1 Y. F. Zheng,1 J. P. Lin2

1State Key Laboratory for Turbulence and Complex System and Department of Materials Science and Engineering,

College of Engineering, Peking University, Beijing 100871, China2State Key Laboratory for Advanced Metals and Materials, University of Science and Technology Beijing, Beijing 100083, China

Received 5 November 2011; revised 24 March 2012; accepted 29 April 2012

Published online 18 July 2012 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jbm.b.32738

Abstract: The present study aimed to evaluate the osteoblast

response on Ti- and Zr-based BMG surfaces sand blasted

with different grit corundums for implant application, with

mechanically polished disks before sand blasting as control

groups. The surface properties were characterized by scan-

ning electron microscopy (SEM), contact angle, and rough-

ness measurements. Further evaluation of the surface

bioactivity was conducted by MG63 cell attachment, prolifera-

tion, morphology, and alkaline phosphatase (ALP) activity on

the sample surfaces. It was found that corundum sand blast-

ing surfaces significantly increased the surface wettability

and MG63 cell attachment, cell proliferation, and ALP activity

in comparison with the control group surfaces. Besides, the

sample surface treated by large grit corundum is more favor-

able for cell attachment, proliferation, and differentiation

than samples treated by small grit corundum, indicating that

it might be effective for improving implant osseointegration

in vivo. VC 2012 Wiley Periodicals, Inc. J Biomed Mater Res Part B:

Appl Biomater 100B: 1721–1728, 2012.

Key Words: Ti/Zr based bulk metallic glasses, osteoblast, sur-

face modification, sand blasting

How to cite this article: Li HF, Wang YB, Zheng YF, Lin JP. 2012. Osteoblast response on Ti- and Zr-based bulk metallic glasssurfaces after sand blasting modification. J Biomed Mater Res Part B 2012:100B:1721–1728.

INTRODUCTION

Bulk metallic glasses (BMGs) are metallic alloys that caneasily avoid crystallization during solidification and there-fore vitrify at quite low cooling rates (less than 100 K/s).1

They have captured huge interest in recent years due totheir remarkable properties such as high strength, high elas-ticity, good toughness, excellent corrosion resistance, lowmodulus, and unique processing capabilities.2–4

BMGs are recently considered as biomaterials since theydisplay an admirable combination of properties and proc-essing capabilities desired for universal implant applica-tions. The unique properties of BMG alloys such as lowerelastic modulus (50–100 GPa) (The elastic modulus of theTi- and Zr-based BMGs used in this study are 92.6 GPa5 and96 GPa6, respectively) and most uniquely an extremely highelastic limit of 2% as compared to that of a typical metal,namely �0.2%,7,8 make them extremely attractive for bio-medical applications. BMGs are unique in their ability toflex elastically with the bending of the natural bones, whichhas an elastic limit of �1%, and therefore distribute

stresses more uniformly. Furthermore, faster healing ratesresult from reduced stress shielding effects while minimiz-ing stress concentrations.8,9 Because of the unique mechani-cal properties of the BMGs, screws made of the BMGs canhave a thinner shank and deeper threads yielding greaterholding power.9 For a given load, compared with the lowestmodulus developmental titanium alloy, the BMG will requireonly one-fourth the cross-section to carry the load and willundergo twice the deflection and compared to stainlesssteel, the area will be one-third to carry the load, and theBMG would have more than five times the deflection.8,9

Potential applications of BMGs include fracture fixationscrews, rods, pins, hip joint wear surfaces and shafts, aneu-rysm clips, endodontic files, and orthodontic arch wires aswell as components of devices such as pacemakers, neuro-stimulators, medicine-metering pumps, and equipment forremotely viewed microsurgery.9

Among the various kinds of BMGs, Ti-based10 and Zr-based7,11–14 BMGs are the two mostly investigated alloy sys-tems as potential implant biomaterials. Oak et al. carried

Correspondence to: Y. F. Zheng; e-mail: yfzheng@pku.edu.cn

Contract grant sponsor: National Basic Research Program of China (973 Program); contract grant number: 2012CB619102

Contract grant sponsor: State Key Lab of Advanced Metals and Materials; contract grant number: 2011-ZD01

Contract grant sponsor: National High Technology Research and Development Program of China (863 Program); contract grant numbers:

2011AA030101, 2011AA030103

Contract grant sponsor: National Natural Science Foundation of China; contract grant number: 31170909

Contract grant sponsor: Research Fund for the Doctoral Program of Higher Education; contract grant number: 20100001110011

VC 2012 WILEY PERIODICALS, INC. 1721

out a series of studies about the corrosion behavior and bio-compatibility of the Ti45Zr10Pd10Cu31Sn4 metallic glass,and their results demonstrated that this kind of metallicglass exhibited good corrosion resistance and excellent bio-compatibility in osteoblast culture test.10 Horton et al.8 per-formed a series of tests on the biocompatibility of Zr-10Al-5Ti-17.9Cu-14.6Ni BMG (BAM-11) through direct evaluationof the viability and metabolic activity of cells on the BMGsurface. It was found that BAM-11 displays as good biocom-patibility as the titanium and polyethylene controls. Corro-sion measurements, which were directly compared to 316Lstainless steel specimens and to commercially pure titaniumspecimens, showed that the BAM-11 material has adequatecorrosion rates for biomedical application. Liu et al.7 investi-gated the corrosion behaviors and evaluated the potentialcytotoxicity of Zr60Nb5Cu22.5Pd5Al7.5 and Zr60Nb5Cu20-Fe5Al10 BMGs. Their result shows that the two BMGs ex-hibit excellent corrosion resistance and exhibit as good bio-compatibility as Ti-6Al-4V alloy, and thus show a promisingpotential for biomedical applications.

In our previous studies, we have studied the mechanicalproperties, corrosion resistance, in vitro and in vivo biocom-patibility of Ti41.5Zr2.5Hf5Cu37.5Ni7.5Si1Sn5 (TZHCNSS)BMG.15 It was found that comparing with pure Ti, TZHCNSSBMG shows superior mechanical properties with higherhardness and better wear resistance. Due to the oxide filmformed on its surface, TZHCNSS BMG shows great corrosionresistance close to pure Ti in electrochemical measure-ments. The in vitro and in vivo testing results proved thatTZHCNSS BMG shows excellent biocompatibility. We alsoinvestigated the corrosion resistance and in vitro biocom-patibility of Zr-based BMGs16,17 with chemical compositionsof Zr41Ti14Cu12Ni10Be23 (LM1), Zr44Ti11Cu10Ni10Be25(LM1b), and Zr57Nb5Cu15.4Ni12.6Al10 (LM106) with pureTi and pure Zr as control groups. Our results demonstratedthat these Zr-based BMGs show superior corrosion resist-ance and are comparable with pure Ti and pure Zr. The cy-totoxicity results show that they have no cytotoxicity toL929 and NIH3T3 cells, and the cells could adhere andgrow well on these Zr-based BMG surfaces.

However, the success of an implant is determined by itsintegration into the tissue surrounding the material. Cell ad-hesion and cell spreading is an important parameter forimplant engineering. Low biomaterial efficiency is often dueto poor integration of the implant with surrounding tis-sue.18 The long-term stability of an endosseous implant inits receptor site is ensured by a direct structural and func-tional connection between ordered, living bone and the sur-face of a load carrying implant, which is called osteointegra-tion.19 Optimal bone regeneration usually requires anoptimized surface of the implant itself.20

Surface roughness is considered to influence the proper-ties of adherent cells, and the best design for an implant isa surface with a rough microtopography.21 Upon implanta-tion, the implant surface is conditioned by proteins, ions,sugars, and lipids present in the blood and tissue fluids. Im-portant surface properties affecting this process include to-pography, chemistry, surface charge, and hydrophilicity.22

Previous studies show that Ti implants with greater rough-ness can enhance bone-to-implant contact and increase theremoval torque forces.23,24 Tissue integration is conditionedby the adhesion and spreading ability of cells on implantsurfaces. Cell behavior on biomaterial surfaces depends onimplant-cell interactions, correlating with surface properties.Surface hydrophilicity, roughness, texture, chemical composi-tion, charge and morphology strongly affect cellularresponses in contact with the implants.25 However, to theauthor’s knowledge, previous studies about BMGs’ potentialbiomedical applications had neglected the effect of surfaceroughening treatment, and there is no report about theinfluence of BMGs’ surface roughness and wettability onosteoblast response or osseointegration as yet.

The aim of the present investigation is to show the roleof the surface roughness of Ti- and Zr-based bulk metallicglasses (BMGs) on the cell attachment, proliferation, mor-phology, and differentiation. For this study, an osteoblast-like cell line MG63 has been chosen as target cell for anymedical device for bone ingrowth. Surfaces with differentroughness are obtained by sand blasting using corundumwith various grit sizes.

MATERIALS AND METHODS

Materials preparationTi- and Zr-based BMGs with nominal compositionsTi40Zr25Ni3Cu12Be205 and Zr50.7Cu28Ni9Al12.326 wereprepared by arc melting the mixtures of highly pure constit-uent elements at 2000�C in a Ti-gettered argon atmosphere.In order to ensure the homogeneity of the composition,each ingot was re-melted at least four times, followed bycasting the melted alloys into a water-cooled copper mold,forming cylindrical samples with a length of about 70 mmand a diameter of 10 mm.

Surface preparationThe Ti- and Zr-based BMG cylindrical samples were cut intodisks with 10-mm diameter and 1.5-mm thickness by aspark discharging machine. The samples were polished with2000# silicon carbide sand paper as a control surface(referred as Ti BMG untreated and Zr BMG untreated here-inafter). Some of them were then sand blasted at the pres-sure of 0.5 MPa with small-grit (180#(grit) 20� 50 lm) co-rundum and large-grit corundum (60#(grit) 300� 500 lm)in order to obtain different roughness for each group; allthe samples were then cleaned ultrasonically in acetone,ethanol, and deionized water in turn for 15 min, and finallydried in air. Prior to use in cell experiments, the untreatedTi- and Zr-based BMGs and their sand-blasted samples werecleaned by sonication in detergent and ultra-pure water andfollowed by autoclave sterilization.

Surface and structure analysisThe surface structure of the BMG samples before and aftersand blasting was identified by thin film X-ray diffraction(TFXRD) with Cu Ka radiation (Rigaku RINT 2500, Japan)operated at 50 kV and 300 mA at room temperature. The

1722 LI ET AL. OSTEOBLAST RESPONSE ON Ti- AND Zr-BASED BULK METALLIC GLASS SURFACE

glancing angle of the incident beam against the surfaces ofthe BMG samples was fixed at 2� .

The changes in the surface morphologies and the micro-structures of the samples before and after surface modifica-tion were characterized by scanning electron microscope(SEM, Hitachi S-4800, Tokyo, Japan). The water contactangle was measured using an OCA20 (Dataphysics Instru-ment, Germany) contact angle system, with a digital cameraand image analysis software. Ultra-pure water was used asthe wetting liquid, with a drop size of 2 lL. An average of 6measurements was taken for each group. The surface rough-ness was measured with a TR200 handheld roughmeter(TIME Group Inc., China). Six samples were selected andmeasured for each group.

Cell–surface interactionsMG63 human osteoblast-like cells were cultured in mini-mum essential medium (MEM, Sigma, America), supple-mented with 10% fetal bovine serum (FBS) and antibiotics(100 U mL�1 penicillin, 100 lg mL�1 streptomycin) at 37�Cin a humidified atmosphere of 5% CO2.

The cells were detached from culture flasks by trypsini-zation and centrifuged. Their number and viability weredetermined by the trypan blue dye exclusion test. After cellcounting, MG63 osteoblast-like cells were plated at a densityof 105 cells mL�1 in 24-well plates containing sterilesamples.

The surface of cell-adhered experimental samples wasobserved by SEM (Hitachi S-4800, Japan). For SEM observa-tion, after 4, 8, 12, and 24 h of incubation, the cells wererinsed with PBS three times gently to remove unattachedcells, and then the cells were fixed with 2.5% glutaralde-hyde in PBS for 60 min at room temperature. After dehydra-tion in graded series of alcohols (50%, 60%, 70%, 80%,90%, and 100%) for 10 min each and dried in hexamethyl-disilazane (HMDS) solution, the samples were sputter-coated with gold in a precision etching coating system(Model 682, Gantan).

The initial attachment and proliferation of cells wereevaluated by measuring the quantity of the cells attached tothe samples by the MTT assay. After 4, 8, 12, 24, and 72 hof incubation, the cells were rinsed with PBS three timesgently to remove unattached cells and then 1 mL of freshcell culture medium was added; 100 lL of MTT was addedto each well for a further 4 h incubation. And then 1 mL offormazan solubilization solution was added to each well. Af-ter 10-min low speed vibration, the measurement was car-ried out spectrophotometrically at 490 nm by an ELISAreader (Elx-800, bio-Tek Instruments, USA).

For alkaline phosphatase activity (ALP) test, after 3, 7,and 14 days of cultivation, the culture medium wasremoved and then 900 lL/well 10 mM p-nitrophenylphos-phate (Amresco, USA) containing 2 mM MgCl2 in a carbon-ate buffer solution (pH ¼ 10.2) was added. After incubatingat 37�C for 60 min, 100 lL/well of 1M NaOH was added tostop the reaction. The ALP activity was evaluated as theamount of p-nitrophenol (PNP) released through the enzy-matic reaction and measured at 405 nm wavelength using

an ELISA reader (Elx-800, bio-Tek Instruments, USA). Fornormalization, the total protein content was measured usinga bicinchoninic acid (BCA) assay kit (Sigma). Thus, the ALPactivity was expressed in nmol of PNP produced per minuteper mg protein.

Statistical analysisAll data were expressed as the mean 6 standard deviation(SD) of six replicates. Statistical procedures were performedwith SPSS 17.0. The significance of the obtained data wasperformed using one-way analysis of variance, followed byTukey’s test for multiple comparison procedure with a con-fidence level of 95% (p < 0.05) considered statisticallysignificant.

RESULTS

Microstructural characterizationFigure 1 shows the TFXRD patterns of Ti- and Zr-basedBMGs before and after surface modification. It can be seenfrom Figure 1 that there is no diffraction peak correspond-ing to crystalline phases, indicating the amorphous state ofthe samples. Besides, after sand blasting, there is no newphase appearing in the TFXRD patterns, indicating that thesand-blasting treatment did not influence the structure ofBMGs.

Surface characterizationFigure 2 presents the surface morphologies and roughnesschanges of Ti- and Zr-based BMGs before and after sandblasting. It can be seen from Figure 2 that after sand blast-ing by the two different grits of corundums, the surfacemorphology of BMGs have changed, and there are variousmicro-rough surfaces of the sand-blasted surfaces comparedwith the smooth control surface. The complex microstruc-ture of the sand-blasted surface arises from the corundumgrit sand-blasting process, which produces cavities with anaverage diameter of about 2 and 20 lm for small grit corun-dum and large grit corundum, respectively. Accordingly, thevalues of Ra increases with corundum grit sizes as shownin Figure 3. Before sand blasting, the average roughness(Ra) for mechanically polished Ti- based BMG and Zr-basedBMG are 0.075 6 0.004 lm and 0.073 6 0.006 lm, respec-tively. The Ra for Ti-based BMG and Zr-based BMG increaseto 0.750 6 0.017 lm and 0.757 6 0.017 lm, respectively,after sand blasting by small grit corundum, whereas the Rafor Ti-based BMG and Zr-based BMG are 2.118 6 0.015 lmand 2.14560.069 lm, respectively, after sand blasting bylarge grit corundum The change of surface roughness afterdifferent grit corundum sand blasting further confirm thatdifferent corundum grits are able to adjust the surface to-pography of BMG samples.

Water contact angle measurements (Figure 4) indicatethat the Ti- and Zr-based BMG samples before sand blastingexhibits a slightly hydrophilic behavior, and the contactangles are 80� 6 6.6� for Ti-based BMG and 76.4� 6 6.2�

for Zr-based BMG, and after sand-blasting surface modifica-tion the Ti- and Zr-based BMG surfaces show enhancedhydrophilic behavior, with the contact angle 43.1� 6 5.4�

ORIGINAL RESEARCH REPORT

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | OCT 2012 VOL 100B, ISSUE 7 1723

for Ti-based BMG and 41.6� 6 0.7� for Zr-based BMG forsmall grit corundum blasting cases and 28.3� 6 4.1� for Ti-based BMG and 24.4� 6 2.6� for Zr-based BMG for largegrit corundum blasting cases, respectively.

Cell-surface interactionsIt can be seen from Figures 5 and 6 that the cell adhesion(4, 8, and 12 h), proliferation (24 and 72 h), and ALP activ-ity increase dramatically with increasing micropore size androughness of the Ti- and Zr-based BMG surfaces. As shownin Figure 5, the cells exhibit different shapes on untreatedcontrol group, 180# corundum grit blasting group and 60#corundum grit blasting group surfaces. Scanning electronmicrographs of the cells on the untreated control groupsubstrates showed that at 4 hours after seeding, MG63 cellshave a relatively spherical morphology [Figure 5(a) foruntreated Zr-BMG and Figure 5(m) for untreated Ti-BMG].

However, after 4 h culture, cells on 180# and 60# corun-dum grit blasting group spread out quite well and cells arein spindle or polygon shape [Figure 5(b) for 180# Zr-BMG,Figure 5(c) for 60# Zr-BMG, Figure 5(n) for 180# Ti-BMGand Figure 5(o) for 60# Ti-BMG]. Moreover, cells firmlyadhere to 60# corundum grit blasting sample surfaces [Fig-ure 5(c) for 60# Zr-BMG and Figure 5(o) for 60# Ti-BMG]and many filopodia and lamellipodia appear after 4 h seed-ing. At 8 h after seeding, spherical cells on untreated controlgroup substrates [Figure 5(d) for untreated Zr-BMG and Fig-ure 5(p) for untreated Ti-BMG] start to extend on the sur-face while more extension and filopodia and lamellipodiaform on the sand-blasted sample surfaces [Figure 5(e,f) for180# and 60# Zr-BMG and Figure 5(q,r) for 180# and 60#Ti-BMG]. At 12 hours after seeding [see Figure 5(g) foruntreated Zr-BMG and Figure 5(s) for untreated Ti-BMG],the cells on untreated surfaces have spread and have mor-phologies that start reaching confluence on substrates, andafter sand-blasting surface modification, the cells anchortightly on the rough surfaces [Figure 5(h,i) for180# and

FIGURE 1. TFXRD patterns of Ti- and Zr- based BMGs before and af-

ter sand blasting. [Color figure can be viewed in the online issue,

which is available at wileyonlinelibrary.com.]

FIGURE 2. SEM images of Ti- and Zr-based BMGs before and after sand blasting: (a) Ti BMG untreated; (b) Ti BMG 180#; (c) Ti BMG 60#; (d) Zr

BMG untreated; (e) Zr BMG 180#; (f) Zr BMG 60#.

FIGURE 3. Roughness of Ti- and Zr-based BMGs before and after

sand blasting. [Color figure can be viewed in the online issue, which

is available at wileyonlinelibrary.com.]

1724 LI ET AL. OSTEOBLAST RESPONSE ON Ti- AND Zr-BASED BULK METALLIC GLASS SURFACE

60# Zr-BMG and Figure 5(t,u) for 180# and 60# Ti-BMG].At 24 h after seeding, the cells on untreated surfaces [Fig-ure 5(j) for untreated Zr-BMG and Figure 5(v) for untreatedTi-BMG] spread completely and start to appear with a fewfilopodia, while cells on the rough surfaces after sand blast-ing [Figure 5(k,l) for180# and 60# Zr-BMG and Figure5(w,x) for 180# and 60# Ti-BMG] show tremendous filopo-dia and extracellular matrix(ECM) and bind tightly to thesubstrates. In short, MG63 cells cultured on rough surfacestreated by sand blasting extend and spread out earlier thanthat of untreated samples. After 4, 8, 12, and 24 h culture,MG63 cells on the rough surfaces have more filopodia andcytoplasmic extensions, stronger attachment and adhesionat the same time point, and cells appeared grow preferen-tially into the cavities on the surfaces created by sand blast-ing, forming attachments in the micropits.

The cell attachment (4, 8, and 12 h incubation) and pro-liferation (1 and 3 days incubation) results were expressedas the number of cells normalized to that of the control sur-face after a certain time culture on sample surfaces. After 4,8, and 12 h incubation, the number of MG63 cells attachedon 180# corundum grit blasting group and 60# corundumgrit blasting group surfaces were greater than that onuntreated control surfaces (Figure 6) (p < 0.05). The cellproliferation results (Figure 7) indicated that the roughsurfaces had better cytocompatibility with the proliferationrate higher than that of the untreated control group at alltesting time points (p < 0.05). The cell attachment and pro-liferation of Ti-based BMG and Zr-based BMG are close andthere were no statistical difference between them (p >

0.05). Besides, the cells on the large corundum grit blastinggroup surfaces had a significantly higher proliferation ratethan that on the small corundum grit blasting group surfa-ces. Thus, large corundum grit blasting is more effectivethan small corundum grit blasting.

In osteoblasts, ALP expression is an early marker ofosteogenic differentiation. Factors influencing ALP expres-sion may further affect cell osteogenic mineralization ability.Figure 8 demonstrates that the ALP activity of MG63 cellson sand-blasting surfaces was significantly higher (p <

0.05) than that of untreated control groups after 3, 7, and14 days culture. The ALP activity of Ti-based BMGs isslightly lower than Zr-based BMGs in the same surface con-dition, without statistical difference (p > 0.05). The 60# co-rundum grit blasting surface seemed to have the best abilityto enhance ALP activity, which coincides well with the cellattachment and proliferation tests and scanning electronmicroscope observation. From the results of surface topog-raphy [Figure 2(a–f)] and roughness (Figure 3) measure-ments and contact angle tests (Figure 4), it can be drawn asa conclusion that rougher and relative hydrophilic BMGsurfaces are more favorable for cell attachment, prolifera-tion, and differentiation. Our results suggest that the surfaceroughness and hydrophilicity may play a major role forosteointegration and bone remodeling.

DISCUSSION

In this work, Ti- and Zr-based bulk metallic glasses (BMGs)were sand blasted with different grit corundums, yieldingrough surfaces with different roughness. These rough surfa-ces dramatically increased the wettability of these twoBMGs, thus enhancing cell attachment, cell proliferation, andALP activity in comparison with the control group surfaces,indicating further improvement of implant osseointegrationprocess in vivo.

Influence of surface roughness on wettabilityand interface adhesionWettability is one of the most important properties on solidsurfaces. The surface wettability is generally evaluated bythe contact angle. The contact angle (y) is defined as theangle between the solid surface and the tangent line of theliquid phase at the interface of three phases. For the sim-plest case, the wettability of the solid surface is commonlyevaluated by the contact angle given by Young’s equation27:

cSV ¼ cSL þ cLV cos h (1)

where cSV, cSL and cLV are the interfacial free energies perunit area of the solid-gas, solid-liquid, and liquid-gas interfa-ces, respectively.

However, Young’s equation is applicable only to an idealsurface, that is, homogeneous, rigid, insoluble, and flat. Realsurfaces have surface roughness and surface heterogeneity.Therefore, Wentzel modified Young’s equation consideringthe surface roughness to obtain the following equation.27

cos h0 ¼ c cos h (2)

where y0 is the apparent contact angle on rough surface, cis the surface roughness ratio between the actual surfacearea and the apparent surface area.27 The Wentzel equationcommendably described the influence of surface roughnesson contact angles. In the case of the hydrophilic surface, therougher the surface, the smaller the contact angle and viseversa in the case of hydrophobic surfaces, the rougher thesurface, the greater the contact angle, i.e., roughnessdecreases the water contact angle if its value on the

FIGURE 4. Contact angles before and after sand-blasting surface

modification. [Color figure can be viewed in the online issue, which is

available at wileyonlinelibrary.com.]

ORIGINAL RESEARCH REPORT

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | OCT 2012 VOL 100B, ISSUE 7 1725

unroughened surface of the same material is <90 andincreases it if the angle is >90. This generally means thatroughness makes hydrophilic surfaces more hydrophilic,

and hydrophobic ones more hydrophobic. In our presentstudy, because the initial contact angle before sand blastingis less than 90�, the contact angles after sand-blasting

FIGURE 5. SEM morphologies of MG63 cells cultured on different surfaces (a)–(l): Zr BMG; (m)–(x): Ti BMG for 4 h, 8 h, 12 h, and 24 h.

1726 LI ET AL. OSTEOBLAST RESPONSE ON Ti- AND Zr-BASED BULK METALLIC GLASS SURFACE

surface modification reduce and the rougher the surface,the smaller the contact angle.

Adhesion to rough surfaces may be effective because ofthe intrinsically high surface energy of atoms on an asperitysurface. The increase in surface area, possibly by a veryhigh factor, also raises the surface energy when expressedper unit nominal area.

Rough surfaces, when stressed, may redistribute thestress so as to increase energy dissipation during the failureof the joint. The strengthening of an interface resulting fromincreasing roughness may change the mechanism of fracturefrom a less to a more energetic mode. With increasing inter-facial roughness between two incompatible polymers, themechanism may change from chain pull out to crazing orother forms of plastic deformation.28

Gent and Lai have convincingly demonstrated the effectof roughness on adhesion in careful experiments with rub-ber adhesion. In comparing adhesion to smooth and to grit-blasted steel, they observed increases in peel energy by fac-tors of two to three times, which they ascribed to theincrease in surface area.29

Effect of surface roughness and wettability onbiological responseThere is a great deal of evidence that surface roughnessplays an important role in determining successful osseointe-gration of implants. Osteoblast differentiation, proliferation,and matrix production30 as well as their production of localgrowth factors and cytokines31 are affected by surface to-pography and roughness.

Roughness has been shown to enhance osteoblastic de-velopment. The surface roughness itself can increase theosteoblastic proliferation and differentiation in cell cultures.Hatano et al.32 investigated the effect of surface roughnesson proliferation and alkaline phosphatase expression of ratcalvarial cells cultured on polystyrene and their resultsdemonstrated that the proliferation and gene expression ofalkaline phosphatase (ALP) and osteocalcin of the calvarialcells increased on the rough-surfaced cover strips.

Lampin et al.33 studied the influence of roughness ofpoly(methyl methacrylate) (PMMA), obtained by sand blast-ing, on chick embryos’ corneal and vascular endothelia celladhesion, and their results indicated that cell adhesionpotential seemed to increase with roughness. Osteoblastsare sensitive to the surface modifications.34 In the study ofZhang et al.,35 osteoblast MC3T3-E1 cells were cultured onpure titanium (grade 2) with different surface topographies.It was observed that cell adhesion, proliferation, morphol-ogy, ALP activity, and calcium deposition increased withincreasing roughness. Garcia et al.36 suggested that

FIGURE 7. MG63 cell proliferation on Zr BMG surfaces and Ti BMG

surfaces (*p < 0.05 compared with untreated surfaces). [Color figure

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

wileyonlinelibrary.com.]

FIGURE 8. ALP activity normalized to the protein content of osteo-

blastic cells after 3 d, 7 d, and 14 d cultured on Zr BMG surfaces and

Ti BMG surfaces (*p < 0.05 compared with untreated surfaces). [Color

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

wileyonlinelibrary.com.]

FIGURE 6. MG63 cell attachment percentage on Zr BMG surfaces and

Ti BMG surfaces (*p < 0.05 compared with untreated surfaces). [Color

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

wileyonlinelibrary.com.]

ORIGINAL RESEARCH REPORT

JOURNAL OF BIOMEDICAL MATERIALS RESEARCH B: APPLIED BIOMATERIALS | OCT 2012 VOL 100B, ISSUE 7 1727

osteoblast differentiation is greater on hydrophilic surfaces.In general, hydrophilic surfaces displayed better affinity forcells than hydrophobic surfaces.

CONCLUSIONS

Ti- and Zr-based BMG surfaces having combined advantagesof surface topography and wettability together were pre-pared by sand blasting and characterized for the first time.Increased wettability of surfaces leads to higher cell attach-ment, cell proliferation, and differentiation properties com-pared with that of control surfaces for osteoblast MG63cells. What is more, the surface treated by large grit corun-dum is more favorable for cell attachment, cell proliferation,and differentiation than that of small grit corundum, indicat-ing its potential for improved implant osseointegrationin vivo.

REFERENCES1. Jin K, Loffler JF. Bulk metallic glass formation in Zr-Cu-Fe-Al

alloys. Appl Phys Lett 2005; 86.

2. Schroers J, Paton N. Amorphous metal alloys form like plastics.

Adv Mater Process 2006;164:61–63.

3. Johnson WL. Bulk glass-forming metallic alloys: science and tech-

nology [1998 MRS Medal Award Lecture, presented at Sympo-

sium MM]. Mater Res Soc 1999;554:311–339.

4. Schuh CA, Hufnagel TC, Ramamurty U. Overview No.144 – me-

chanical behavior of amorphous alloys. Acta Mater 2007;55:

4067–4109.

5. Guo F, Wang H-J, Poon SJ, Shiflet GJ. Ductile titanium-based

glassy alloy ingots. Appl Phys Lett 2005;86:091907.

6. Zheng W, Huang YJ, Wang GY, Liaw PK, Shen J. Influence of

strain rate on compressive deformation behavior of a Zr-Cu-Ni-Al

bulk metallic glass at room temperature. Metall Mater Trans A

2011;42:1491–1498.

7. Liu L, Qiu CL, Chen Q, Chan KC, Zhang SM. Deformation behav-

ior, corrosion resistance, and cytotoxicity of Ni-free Zr-based bulk

metallic glasses. J Biomed Mater Res A 2008;86A:160–169.

8. Horton JA, Parsell DE. Biomedical potential of a zirconium-based

bulk metallic glass. MRS Proc 2003;754:179–184.

9. Horton JA PD. Bulk metallic glass medical instruments, implants,

and methods of using same. US Patent Appl. 10/256,751 2002.

10. Oak J-J, Hwang G-W, Park Y-H, Kimura H, Yoon S-Y, Inoue A.

Characterization of surface properties, osteoblast cell culture in

vitro and processing with flow-viscosity of Ni-free Ti-based bulk

metallic glass for biomaterials. J Biomed Sci Eng 2009;4:384–391.

11. Liu L, Qiu CL, Huang CY, Yu Y, Huang H, Zhang SM. Biocompati-

bility of Ni-free Zr-based bulk metallic glasses. Intermetallics

2009;17:235–240.

12. Qin FX, Wang XM, Inoue A. Observation of bone-like apatite on

Ti-coated Zr55Al10Ni5Cu30 bulk metallic glass after alkali treat-

ment. Intermetallics 2008;16:917–922.

13. Oak JJ, Louzguine-Luzgin DV, Inoue A. Investigation of glass-

forming ability, deformation and corrosion behavior of Ni-free Ti-

based BMG alloys designed for application as dental implants.

Mater Sci Eng C 2009;29:322–327.

14. Buzzi S, Jin KF, Uggowitzer PJ, Tosatti S, Gerber T, Loffler JF. Cy-

totoxicity of Zr-based bulk metallic glasses. Intermetallics 2006;14:

729–734.

15. Zheng YF, Wang YB, Wei SC. Corrosion behavior and cytotoxicity

evaluation on Ti-Zr-Hf-Cu-Ni-Si-Sn bulk metallic glass, 22nd Euro-

pean Conference on Biomaterials, 07–11 September, 2009, Lau-

sanne, CH, Book of Abstract T9:290.

16. Wang YB, Li HF, Zheng YF, Wei SC, Li M. Correlation between

corrosion performance and surface wettability in ZrTiCuNiBe bulk

metallic glasses. Appl Phys Lett 2010;96:251909.

17. Wang YB, Zheng YF, Wei SC, Li M. In vitro study on Zr-based

bulk metallic glass as potential biomaterial. J Biomed Mater Res

B 2011;96B:34–46.

18. Gristina AG. Biomaterial-centered infection – microbial adhesion

versus tissue integration. Science 1987;237:1588–1595.

19. Stanford CM, Keller JC. The concept of osseointegration and

bone matrix expression. Crit Rev Oral Biol Med 1991;2:83–101.

20. Linez-Bataillon P, Monchau F, Bigerelle M, Hildebrand HF. In vitro

MC3T3 osteoblast adhesion with respect to surface roughness of

Ti6A14V substrates. Biomol Eng 2002;19:133–141.

21. Lincks J, Boyan BD, Blanchard CR, Lohmann CH, Liu Y, Cochran

DL, Dean DD, Schwartz Z. Response of MG63 osteoblast-like cells

to titanium and titanium alloy is dependent on surface roughness

and composition. Biomaterials 1998;19:2219–2232.

22. Zhao G, Schwartz Z, Wieland M, Rupp F, Geis-Gerstorfer J, Coch-

ran DL, Boyan BD. High surface energy enhances cell response to

titanium substrate microstructure. J Biomed Mater Res A 2005;

74A:49–58.

23. Cooper LF. A role for surface topography in creating and main-

taining bone at titanium endosseous implants. J Prosthet Dent

2000;84:522–534.

24. Li DH, Ferguson SJ, Beutler T, Cochran DL, Sittig C, Hirt HP, Buser

D. Biomechanical comparison of the sandblasted and acid-etched

and the machined and acid-etched titanium surface for dental

implants. J Biomed Mater Res 2002;60:325–332.

25. Ponsonnet L, Reybier K, Jaffrezic N, Comte V, Lagneau C, Lissac

M, Martelet C. Relationship between surface properties (rough-

ness, wettability) of titanium and titanium alloys and cell behav-

iour. Mater Sci Eng C 2003;23:551–560.

26. Sun YJ, Qu DD, Huang YJ, Liss KD, Wei XS, Xing DW, Shen J.

Zr–Cu–Ni–Al bulk metallic glasses with superhigh glass-forming

ability. Acta Mater 2009;57:1290–1299.

27. Irie H, Hashimoto K. Photocatalytic active surfaces and photo-

induced high hydrophilicity/high hydrophobicity. Hdb Environ

Chem 2005;2M:425–450.

28. Packham DE. Surface energy, surface topography and adhesion.

Int J Adhes Adhes 2003;23:437–448.

29. Gent AN, Lai SM. Adhesion and autohesion of rubber compounds

– effect of surface-roughness. Rubber Chem Technol 1995;68:

13–25.

30. Martin JY, Schwartz Z, Hummert TW, Schraub DM, Simpson J,

Lankford J, Jr, Dean DD, Cochran DL, Boyan BD. Effect of titanium

surface roughness on proliferation, differentiation, and protein

synthesis of human osteoblast-like cells (MG63). J Biomed Mater

Res 1995;29:389–401.

31. Kieswetter K, Schwartz Z, Hummert TW, Cochran DL, Simpson J,

Dean DD, Boyan BD. Surface roughness modulates the local pro-

duction of growth factors and cytokines by osteoblast-like MG-63

cells. J Biomed Mater Res 1996;32:55–63.

32. Hatano K, Inoue H, Kojo T, Matsunaga T, Tsujisawa T, Uchiyama

C, Uchida Y. Effect of surface roughness on proliferation and alka-

line phosphatase expression of rat calvarial cells cultured on poly-

styrene. Bone 1999;25:439–445.

33. Lampin M, WarocquierClerout R, Legris C, Degrange M, SigotLui-

zard MF. Correlation between substratum roughness and wett-

ability, cell adhesion, and cell migration. J Biomed Mater Res

1997;36:99–108.

34. Gittens RA, McLachlan T, Olivares-Navarrete R, Cai Y, Berner S,

Tannenbaum R, Schwartz Z, Sandhage KH, Boyan BD. The effects

of combined micron-/submicron-scale surface roughness and

nanoscale features on cell proliferation and differentiation. Bio-

materials 2011;32:3395–3403.

35. Zhang EW, Wang YB, Shuai KG, Gao F, Bai YJ, Cheng Y, Xiong

XL, Zheng YF, Wei SC. In vitro and in vivo evaluation of SLA tita-

nium surfaces with further alkali or hydrogen peroxide and heat

treatment. Biomed Mater 2011;6:025001.

36. Garcia AJ, Keselowsky BG. Biomimetic surfaces for control of cell

adhesion to facilitate bone formation. Crit Rev Eukaryot Gene

Expr 2002;12:151–162.

1728 LI ET AL. OSTEOBLAST RESPONSE ON Ti- AND Zr-BASED BULK METALLIC GLASS SURFACE