Tissue Engineering and Biomaterials Division, National Institute of Genetic Engineering and Biotechnology (NIGEB), Tehran 14965/161, IranNanobiomaterials Laboratory, Faculty of Biomedical Engineering, Amirkabir University of Technology, Tehran 15875/4413, IranHelmerich Advanced Technology Research Center, School of Material Science and Engineering, Oklahoma State University, OK 74106, USADepartment of Anatomy and Cell Biology, McGill University, 3640 University Street, Montréal, Québec H3A 0C7, Canada
r t i c l e i n f o
rticle history:eceived 7 July 2012eceived in revised form 11 October 2012ccepted 15 October 2012vailable online 23 October 2012
This study investigates the effects of crystallographic orientation of titanium substrates on the atomicstructure and biological characteristics of hydroxyapatite (HA) coatings. Samples are prepared fromextruded rod and rolled sheet of commercially pure titanium having distinct distribution of crystallo-graphic planes. Electrophoresis is used to coat titanium substrates having different microstructures. Thebiological performance of both HA-coated and non-coated samples is assessed by osteoblast cell attach-ment, proliferation, differentiation and morphological studies. X-ray diffraction (XRD) analysis of theHA-coated samples indicates the predominant orientation of (0 0 2) for HA-coated sheets compared to(2 1 1) for the HA-coated rod samples. The numbers of attached and grown cells are higher on the surfaceof the HA-coated sheet samples. There is also a significant difference in alkaline phosphatase activity onthe HA-coated sheet samples. Scanning electron microscopy (SEM) analysis of osteoblast cells grown on
HA-coated and non-coated samples demonstrates differences in morphology with respect to spreadingand attachment patterns. We believe that the specific atomic structure that is induced in the HA coating bythe crystallographic orientation of the sheet substrate causes orientation-dependent coordination withbiomolecules and improves cellular interactions. This suggests that crystal orientation of the substratecan be used to both influence the structure of the coating material and improve and control cell–substrateinteractions.
. Introduction
The main cause of bone-related implant failure is poor initialsseointegration at the implant–tissue interface [1,2]. The qualityf the prosthesis–tissue interface and immediate osseointegrationoon after implantation would secure initial stability and long-erm durability of the implant which in turn will reduce the rate ofevision surgeries and health care costs [3]. Many parameters arenvolved in improving osseointegration including implant-relatedactors such as surface chemistry and structure [3]. Controllingell–substrate interactions at the molecular and atomic level isritical to establishing a stable interface and successful long-termiologic implant fixation [4]. One approach to modify the surface
hemistry of metallic substrates in order to improve the interac-ions at the interface is the application of a hydroxyapatite (HA)oating [5]. While various coating techniques have been used for
deposition of HA on metallic substrates [6–8], their limitationsinclude non-uniform coating over geometrically complex surfaces,thermal decomposition of HA, sluggishness of the process and pooradhesivity of the coating to the substrate [9].
Titanium and its alloys, which are widely used in orthope-dics and dentistry, possess a polycrystalline structure consistingof grains with different crystallographic orientation. Characteriz-ing the crystallographic orientation of all the grains is crucial assome key material characteristics are texture-specific [10]. Whengrain distributions are random, there is no preferred orientationand the material has no texture. In most materials, however, thereis a pattern to the grain orientation which confers a preferred orien-tation or crystallographic texture to the material and is representedby pole figures (PF) and orientation distribution function (ODF)[10,11].
In the development of hard tissues, such as the formation of boneand teeth, there are complicated interactions between proteins and
ultimately cells and inorganic mineral phases. In these interac-tions, the crystallographic direction needs to be highly regulatedfor biomineral nucleation and growth [12–14]. In an assay to deter-mine oseteocalcin crystal structure and its binding ability to HA, it
as been shown that osteocalcin can only coordinate with calciumons in the specific crystallographic orientation of the HA crystalattice. This coordination keeps the binding region of the proteinn a chemotactic position for binding of osteoblast surface recep-ors [13]. It has also been reported that cell attachment, spreading,
otility and cell–cell aggregation differs on various crystal faces ofhe substrate [15]. The formation of calcium phosphate crystallitesrom simulated body fluids appears to develop the specific crystalrientation of (1 0 0) [16]. A similar phenomenon has been foundn thermally deposited calcium phosphate coatings on metallicubstrates [17]. One of the main problems with calcium phosphate-oated implants is the dissolution of the coating in the human body,hich causes loosening of the implant. The dissolution rate of cal-
ium phosphate coatings is another parameter that is dependentn crystallographic directions [18].
It is clear from these findings that the organic–inorganicnterface is controlled by interactions between macromoleculesroduced by the cells and atomic structure of the substrate definedy crystallographic orientations. In this study, differently processeditanium substrates, consisting of extruded rod and rolled sheetaving similar chemical composition and distinct crystallographicrientations, are used as substrates and electrophoretically coatedith synthesized nanometer-sized HA powder. It is expected that
he crystallographic orientation of the substrates will affect thetructure of the HA coatings and, in turn, their biological per-ormance as assessed by cell attachment, proliferation, alkalinehosphatase activity and morphological studies.
. Experimental methods
.1. Preparation of titanium substrates
To obtain substrates having different crystallographic texture,ommercially pure titanium (CP-Ti) extruded rod and rolled sheetMcMaster-Carr Company, Los Angeles, CA) was used and cut in aay to have the same surface area. Slices of the CP-Ti rod wererepared by cutting the rod normal to its z-axis. Samples wereechanically ground with SiC papers followed by vibratory polish-
ng with a vibrometer (Buehler, Lake Bluff, IL) to achieve a mirrorurface finish. Subsequently the samples were cleaned with ethanolnd distilled water, respectively, in an ultrasonic bath for 20 min. Toncrease surface roughness, the CP-Ti samples were etched and oxi-ized with a solution consisting of 10% nitric acid, 10% hydrochloriccid, 10% sulfuric acid and 10% H2O2 for 1 h. This treatment cre-ted small, nano-sized pits that increases the potential adherencef the HA powder on the surface of CP-Ti samples [19]. The sam-les were then submerged in a 5.0 mol/L NaOH aqueous solution at0 ◦C for 48 h, followed by washing with deionized water. Finally,efore electrophoretic deposition, the samples were heated in anlectric furnace at a heating rate of 1 ◦C/min, maintained at 600 ◦Cor 1 h, and then the furnace allowed to cool to room tempera-ure.
.2. Crystallographic texture analysis
Crystallographic texture is represented by pole figures (PF)nd orientation distribution function (ODF) measurements whichhow the major orientations present in the polycrystalline mate-ial. Pole figures were measured by X-ray diffraction (XRD) andore than one pole figure is needed to calculate ODF [20]. The
ole figures of rod and sheet titanium samples were obtainedsing a Siemens D-500 X-ray diffractometer (Siemens, Munich,ermany). Six pole figures were measured (0 0 2), (1 0 0), (1 0 1),
1 0 2), (1 0 3), (1 1 0) and the ODF was calculated using TexToolsoftware [21].
iointerfaces 103 (2013) 200– 208 201
2.3. Synthesis and characterization of hydroxyapatite powder
The HA powder was synthesized by the metathesis method[22] in a solution containing tetra hydrated calcium nitrate(Ca(NO3)2·4H2O) and di-ammonium hydrogen phosphate(NH4)2HPO4 with Ca/P ratio of 1.667. The reaction was asfollows:
10Ca(NO3) + 6(NH4)2HPO4 + 8NH4OH → Ca10(PO4)6OH2
+ 6H2O + 20NH4NO3
To obtain the HA powder, NH4+ and NO3
− ions were removed bywashing the precipitate repeatedly with water followed by dryingin an oven at 100 ◦C for 24 h. The powder was calcinated at 1000 ◦C(heating rate of 5 ◦C/min) for 1 h in an air atmosphere. HA powderwas obtained by grinding with an agate mortar and pestle for 1 hand characterized by XRD (EQuniox 3000, INEL, France) and Fouriertransform infrared spectroscopy (FTIR; Bruker IFS 48, Germany).
2.4. Electrophoretic deposition of hydroxyapatite coatings
Calcined HA powder was used as the coating material. A sus-pension was made by adding 5 g of HA powder into 500 mL ofethanol (99.99%, Merck, Germany) and dispersed by adding 0.25%of carboxymethyl cellulose (CMC, [C6H7O2(OH)2OCH2COONa]n).The suspension was stirred and sonicated in an ultrasonic bath for30 min to ensure a homogeneous suspension [23,24]. The suspen-sion was left for 30 min to eliminate agglomerated particles throughsedimentation. Electrophoretic deposition was performed in a glassbeaker equipped with a holder to fix the electrodes [25]. The plat-inum counter electrode and titanium samples were used as anodeand cathode, respectively. As HA particles acquire positive chargeand their deposition occurs at the cathode at a pH 4 [26], the pHof the HA suspension was adjusted to 4 by the addition of smallquantities of NaOH (0.1 M) and HNO3 (0.1 M). A constant voltageof 10 V for 45 s at 25 ◦C was applied to coat the rod and sheet tita-nium samples. Samples were dried at room temperature for 24 hand sintered at 800 ◦C for 2 h to densify the coatings.
2.5. Characterization of HA-coated substrates
The HA-coated rod and sheet titanium substrates were char-acterized by XRD, scanning electron microscopy (SEM) and watercontact angel measurements. To determine the effects of crys-tallographic texture of the substrate on the structure of the HAcoatings, grazing-angle XRD was performed using a Philips X’pertMPD System (Philips, Eindhoven, Netherlands) with a copper anode(Ka = 0.154 nm) and scanning rate of 0.75◦ s−1. Data were acquiredbetween 2� = 20◦ and 80◦ at 0.01◦ increments. The experimentswere carried out at room temperature (25 ◦C) at 1◦ takeoff angleand operating at 50 kV and 40 mA. The surface morphology ofthe HA-coated titanium samples before and after sintering wasexamined by SEM (AIS2100, Seron Technology). Energy-dispersivespectroscopy (EDS; Thermo Noran) was used to determine theelements present in the deposit and estimate the Ca/P ratio. Thewettability of HA-coated and non-coated rod and sheet titaniumsamples was determined by the placing a droplet of distilled wateron the surface of the sample and measuring the contact angle after10 s using a light microscope (OCA 15 plus, Dataphysics) equippedwith two CCD color cameras and an image analyzer system.
2.6. Cell culture
Human osteoblast-like cells (MG63) provided by American TypeCulture Collection (ATCC, Manassas, VA, USA) were used for thisstudy. The cells were cultured in T25 plastic flasks (Nunc) in
2 es B: Biointerfaces 103 (2013) 200– 208
atp3mtef
2a
onTwpubtrsMtatAdctacHa
2
casgbl
2
spm
3
3
onsetipw
Table 1Water contact angle of HA-coated and non-coated titanium samples. The valueswere calculated by image analyzer software and reported as mean ± SD (n = 5).
lpha minimum essential medium (�-MEM, Invitrogen, Corpora-ion, USA) supplemented with 10% fetal bovine serum, 100 U/mlenicillin and 100 mg/ml streptomycin. Cells were incubated at7 ◦C in a humidified atmosphere of 5% CO2 and 95% air. The growthedium was changed every 48 h. Cultured cells were detached by
rypsinization, suspended in fresh culture medium and used for thexperiments. Conventional culture well plates were used as controlor each set of experiments.
.7. Cell attachment, proliferation and alkaline phosphatasectivity
Cell attachment experiments were performed on the surfacesf HA-coated and non-coated titanium samples by counting theumber of attached cells after 60, 120, and 240 min of incubation.he samples were placed into a 24-well culture plate and seededith a density of 5 × 104 cells/ml. After each time point, the sam-les were washed with a phosphate buffer solution (PBS) to removenattached cells. Adherent cells were removed from the samplesy incubation with 0.25% trypsin in EDTA (Invitrogen Corpora-ion, USA). Trypsin was removed by centrifugation and the cellse-suspended in fresh growth medium. The number of cells in theolution was counted with a hemocytometer. The proliferation ofG63 cells on titanium samples and plastic culture plates (con-
rols) was investigated by counting the number of cells after 2, 5,nd 7 days of culture. Cells were seeded onto the samples and con-rol in a 24-well culture plate with a density of 5 × 103 cells/ml.fter each time point the grown cells were rinsed twice with PBS,etached from the samples by trypsin and counted using a hemo-ytometer. The alkaline phosphatase activity (ALP) was assayed byhe hydrolysis of p-nitrophenyl phosphate (Sigma, St. Louis, MO)s the release of p-nitrophenyl from p-nitrophenolphosphate. Theolor changes of the products were measured via Autoanalyzer-902itachi (Germany) at 405 nm. The enzyme activity was expresseds unit mg−1 min−1 of protein.
.8. Cell morphology
Morphological characteristics of cells on the surfaces of HA-oated and non-coated titanium samples were investigated with
Philips XL30 field emission SEM (FE-SEM). Cells grown on theamples for 3 days were first washed with PBS, fixed with 2.5%lutaraldehyde and dehydrated in graded series of ethanol–wateraths. The surface of the samples was sputter coated with a 15 nm
ayer of Pt/Pd and examined in FE-SEM.
.9. Statistical analysis
Data were first analyzed by analysis of variance (ANOVA). Whentatistical differences were detected, student’s t-test for com-arisons between groups was performed. Data are reported asean ± standard deviations at a significance level of p < 0.05.
. Results
.1. Orientation distribution functions analysis
Based on the six pole figures measured for each substrate, therientation distribution functions (ODFs) of the rod and sheet tita-ium substrates were calculated using TexTools software (data nothown). This software was also used to identify the preferred ori-ntation of the grains in the titanium substrates. As orientation of
he crystallographic planes exposed at the surface of the substrates important in cell culture experiments, we only considered planesarallel to the sample surfaces, which according to their ODF data,as the (1 0 −1 0) plane for the rod samples and the (0 0 0 2) plane
* p < 0.01 compared to the non-coated sheet sample.** p < 0.01 compared to the HA-coated rod sample.
for the sheet samples. The (0 0 2) and (1 0 0) pole figures for the rodand sheet titanium substrates are shown in Fig. 1. It is clear thatthe distribution of these two crystallographic planes in differentlyprocessed rod and sheet substrates are significantly different.
3.2. Characterization of HA powder and HA-coated titaniumsamples
The Fourier transform infrared spectroscopy (FT-IR) spectrumand XRD patterns of the synthesized HA powder are shown inFig. 2. The OH− bands at 3555 cm−1 and 622 cm−1 as well asbands at 561 cm−1, 603 cm−1, and 1040 cm−1 identified in the FT-IR spectra correspond to PO4
3−. The synthesized HA powder hadan average grain size diameter of ∼104 nm as determined from theSEM images using ImageJ software (data not shown). After elec-trophoretic deposition and sintering of the HA coatings on the rodand sheet titanium substrates, the samples were characterized bySEM, XRD, contact angle measurements and EDS. SEM observationof the surface of the titanium samples after etching and oxidiz-ing with a mixed solution of HNO3–HF–H2SO4–H2O2 shows smallnano-sized pits (Fig. 3a). After electrophoretic deposition, a porousHA coating containing constitutional water is formed (Fig. 3b andc), which partially dehydrates after sintering and creates a homo-geneous coating with no observable cracks on the surfaces of eitherthe rod or sheet substrate (Fig. 3d).
Grazing-angle XRD analysis of the HA-coated substrates showsthe presence of characteristic peaks of HA on both rod and sheetsubstrates. Clear differences in the HA atomic structure were,however, detected between surfaces of HA-coated rod and sheettitanium samples (Fig. 4). The (2 1 0), (1 1 2), and (1 0 2) peaks ofHA were similar on the surfaces of both the rod and sheet samples.The (0 0 2), (4 0 0), and (2 0 3) peaks were the main peaks on the HA-coated sheet sample, whereas on the HA-coated rod only the (2 1 1)peak was detected. There was no (4 0 0) peak in the HA-coated rodsample and the (3 0 0), (0 0 2) and (2 0 3) peaks showed much lessintensity compared to the HA-coated sheet sample.
The wettability of HA-coated and non-coated titanium sam-ples was measured by water contact angel measurement (Table 1).Among all the samples, the highest contact angle was measured forthe non-coated sheet. Hydrophilicity was highest on the HA-coatedsamples. The surfaces of the HA-coated rods had a higher aqueouscontact angle suggesting a less hydrophilic surface compared to theHA-coated sheet samples. EDS analysis of the HA-coated samplesafter the cell culture experiments indicated a mole mass ratio of Caand P, n(Ca)/n(P) of ∼1.667 in both the HA-coated rod and sheet sam-ples suggesting that no decomposition of HA had occurred duringelectrophoretic deposition. There also was a significantly higherconcentration of silicon on the surfaces of the HA-coated sheetscompared to the HA-coated rod samples as revealed by EDS (Fig. 5).
3.3. Cell attachment, proliferation and alkaline phosphatase
activity
MG63 cell attachment was performed on the surface of theHA-coated and non-coated titanium samples after 1, 2, and 4 h
A.T. Rad et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 200– 208 203
F -Ti (1s betw
occiHMlsp
cTca
aiaah7
3
adootsfiTnma
ig. 1. Pole figure measurements of (a) extruded rod CP-Ti (1 0 0) (b) rolled sheet CPhow clear differences in the texture and distribution of the crystallographic planes
f incubation (Fig. 6). After the first hour, the amount of MG63ells adhered on the non-coated substrates was significantly higherompared to the coated samples. There were, however, signif-cantly higher amounts of cells attached to the surfaces of theA-coated sheet samples than the coated rods. The amount ofG63 attached cells on the HA-coated samples reached the same
evels as the non-coated samples at 2 h. After 4 h, the HA-coatedheet titanium sample showed the highest cell attachment com-ared to the HA-coated rod, non-coated samples and control.
Proliferation of MG63 cells was tested on HA-coated and non-oated titanium samples after 2, 5 and 7 days of culture (Fig. 7).he number of grown cells was significantly higher on the HA-oated sheets compared to HA-coated rods, non-coated samples,nd control in all time points tested.
Alkaline phosphatase (ALP) activity increased in all samples as function of time (Fig. 8). The ALP activity did not show any signif-cant difference between the non-coated sheet and rod samples,lthough there were significant differences between HA-coatednd non-coated samples. The HA-coated sheet samples showed theighest ALP activity compared to other samples and controls at both
and 14 days.
.4. Cell morphology
SEM observation of MG63 osteoblasts grown on the HA-coatednd non-coated rod and sheet titanium samples for 72 h showedifferences in cell density and spreading patterns. The majorityf cells, however, exhibited their phenotypic morphology. MG63steoblasts were flat with a polygonal configuration and attachedo the substrate by cellular extensions (Fig. 9a–f). On HA-coatedheets, cells were more flat and well-spread out with extendedlopodia to surrounding areas and other cells (Fig. 9b and d).
hese features were clearly different from cells grown on theon-coated samples (Fig. 9e and f). MG63 cells began to form aonolayer on both types of samples after 3 days incubation. In
greement with data obtained from cell proliferation experiments,
0 0) (c) rod CP-Ti (0 0 2) and (d) sheet CP-Ti (0 0 2) calculated by TexTools softwareeen the substrates.
SEM observation showed the numbers of cells were highest on thesurfaces of HA-coated sheets. No morphological differences weredetected between HA-coated rod and HA-coated sheet samples.
4. Discussion
In this study we introduce a new approach involving the use ofcrystallographic texture of underlying substrate to manipulate theatomic structure of HA coating for affecting cell–substrate inter-actions. Differently processed titanium samples having diversepreferred grain orientations are coated with nanometer-sized HApowder using electrophoretic deposition. The substrates are thenused to assess the MG63 cell response. Cell attachment and pro-liferation experiments show significantly higher cell density onthe coated samples compared to the non-coated samples. TheHA-coated sheet sample shows the best overall biological per-formance in comparison between the HA-coated and non-coatedsamples.
An ideal HA coating on the surface of a substrate should beresistant to dissolution and exhibit high bond strength with thesubstrate. The bond strength of the coating can usually be achievedby sintering. Sintered HA coatings are relatively dense and adher-ent to the substrate while unsintered HA coatings can be easilyremoved. Owing to densification during sintering, shrinkage andcracking of the coatings may occur. Also, thermal stresses inducedby differences in thermal expansion coefficients between the sub-strate and coating during sintering and cooling could lead tocracking particularly with thicker coatings [27]. Sintering temper-ature is also crucial since low sintering temperatures may lead toa weakly bonded, low-density coating while high sintering tem-peratures induces irreversible dehydroxylation of HA to anhydrouscalcium phosphates, such as tricalcium phosphate, resulting in
the collapse and decomposition of the HA structure. As titaniuminduces decomposition of HA above 1050 ◦C, to minimize the degra-dation of the HA coating the sintering temperature in the currentstudy was 800 ◦C [28].
204 A.T. Rad et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 200– 208
(b) of
pinotsha(asciiu
ccwt
Fig. 2. FTIR spectra (a) and XRD pattern
There is no evidence of the HA coating decomposing to calciumhosphate in our HA-coated substrates as revealed by the XRD. It
s apparent that the coating process used in this study allows calci-ation of HA at a temperature low enough to avoid decompositionf the coating. The sharpness of the XRD peaks indicates good crys-allinity for the HA coatings on the surfaces of both the rod andheet substrates. The crystallographic structure of the HA coatings,owever, showed clear differences between the surfaces of the rodnd sheet titanium samples as determined from low-angle XRDFig. 4). The HA-coated sheet sample shows (0 0 2), (3 0 0) and (4 0 0)s the main crystallographic planes compared to the HA-coated rodample having (2 1 1) as the main structural plane. The needle-likerystals of HA on the sheet substrate are similar to the HA foundn the human body [29]. The alteration in the structure of the HAs likely due to the different crystallographic orientations of thenderlying sheet and rod titanium substrates.
Both cell attachment and proliferation show that osteoblast cells
an attach to HA-coated samples more efficiently than on the non-oated samples. Among the 4 types of samples the best cell activityas observed on the HA-coated sheet titanium sample. We believe
hat these cell–substrate interactions are directly correspond to the
the synthesized hydroxyapatite powder.
sequential events happening at the organic–inorganic interface.Development of this interface is complex and involves sophisti-cated mechanisms of interacting water molecules, proteins, andultimately, cells with inorganic substrates. These mechanisms aredirectly affected by material-related factors and the host responseto the material. The initial event after insertion of a biomaterialin the body is the interaction of water molecules on the materialsurface [30]. The arrangement of these water molecules is sensi-tive and strongly depends on the surface characteristics and atomicstructure of the material. They may dissociate to form hydroxyl ions(OH−) or strongly bind to the surface of hydrophilic materials. How-ever, on hydrophobic materials, water molecules can loosely bindwith less force than hydrogen bonding, such as electrostatic forces,and known as hydration water. Thus, the concepts of hydropho-bic and hydrophilic are directly related to the binding strength ofwater to the surface at a molecular scale. Once the water layerhas formed, natural ions like Cl− and Na+, enter the interface and
incorporate into the water overlayer as hydrated ions. The specificarrangement of these ions, and their water shells, is influenced bythe properties of the material surface [30,31]. Following formationof the water overlayer, different biomolecules from body fluids and
A.T. Rad et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 200– 208 205
F h (a)
8
sTsofatobtt
ig. 3. SEM images of titanium samples after etching with an acidic solution for 100 ◦C in air atmosphere for 2 h (d).
erum that surround the material reach the surface and adsorb.here are many different proteins present, therefore; the compo-ition of the protein layer will depend on their presence at the sitef implantation and their binding affinity to the surface. The sur-ace characteristics of the surface controls protein conformationnd orientation based on its surface properties, which also affectshe organization of water molecules [15,30,32,33]. As an example
f this dependency, it has been commonly observed that proteinsind to hydrophilic surfaces with their hydrophilic areas towardhe surface through the presence of water molecules, while pro-eins on hydrophobic surfaces are more likely to bind with their
Fig. 4. Low-angle XRD patterns of HA-coated sheet (top) and HA-coated rod (bo
after electrophoretic deposition of the HA coating (b, c) and after the sintering at
hydrophobic regions closest to the surface without interveningwater molecules. Once the protein layer has been established, cellsbegin adhering to the material surface. Bound proteins interactwith cell surface receptors, primarily from the integrin superfa-mily [34]. The binding processes between proteins and the material,and between the proteins and cells, may be interdependent sincethe composition and conformation of adsorbed proteins may be
affected as they adsorb to the material [35,36]. Thus, cell adhe-sion and growth of anchorage-dependent cells may depend directlyor indirectly on the nature of the surface to which the proteinsadsorb. Owing to the differences in free energy of solid–solid
ttom) show clear differences in crystallographic structure of HA coatings.
206 A.T. Rad et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 200– 208
F
ivwva
Fcmet
Fig. 7. Histogram showing cell proliferation expressed as cell number/cm2 after 2, 5,
ig. 5. EDS spectra of HA-coated sheet (a) and HA-coated rod (b) titanium samples.
nterface and free energy of solid–liquid interface anisotropy, indi-
idual grain faces in a polycrystalline material display differingetting behavior. Even within the grain boundaries, due to local
ariations in free energy resulting from atomic arrangement, therere differences in wetting behavior [37]. Water molecules that
ig. 6. Histogram showing the cell density of MG63 osteoblasts attached to HA-oated and non-coated rod and sheet titanium substrates. Each bar represents theean of attached osteoblasts ±SD (n = 3 in each group) in one of three identical
xperiments. *p < 0.05 compared to the HA-coated rod sample; **p < 0.05 comparedo the non-coated sheet sample.
and 7 days. Each bar represents the mean of attached osteoblasts± SD (n = 3 in eachgroup) in one of three identical experiments. *p < 0.05 compared to the HA-coatedrod sample; **p < 0.05 compared to the non-coated sheet sample.
attach to the surface differently enable the interaction of specificmolecules (proteins) involved in the cell substrate interaction [32].Increased surface wettability or hydrophilicity has also been relatedto enhance protein adsorption [38,39] and cell spreading [40,41]on biomaterials. The surface water contact angle is highest on non-coated sheet titanium samples while the HA-coated sheet samplesshow the lowest water contact angel suggesting the highest sur-face wettability. There was a significant difference in water contactangles between the HA-coated and non-coated titanium samples.As surface chemistry is similar between each set of HA-coatedand non-coated samples, different surface wettability is related tothe diverse crystallographic orientations and atomic structures ofthe samples [42]. Proteins such as ECM proteins, cytoskeletal pro-teins, and membrane receptors (integrins) are generally involvedin cell–substrate interactions. Interactions between these proteinsand their specific receptors induce signal transduction, which con-sequently influences cell adhesion, growth, and differentiation [43].Since the degree of wettability is different among substrates withdifferent crystallographic faces, this could explain the increasedcell adhesion and proliferation observed on HA-coated sheet tita-
nium substrates. It is expected that the cell–substrate interactionsobserved in our study were controlled by adsorption affinity of theproteins involved with the different crystal faces, leading to var-ied cell responses. This is in agreement with reports suggesting
Fig. 8. Histogram showing alkaline phosphatase activity of osteoblast cells after 7and 14 days. Data are reported as mean ± SD (n = 3 in each group). *p < 0.05 comparedto the non-coated rod sample; **p < 0.05 compared to the HA-coated rod sample;***p < 0.05 compared to the non-coated rod sample.
A.T. Rad et al. / Colloids and Surfaces B: Biointerfaces 103 (2013) 200– 208 207
ted sh
tf
lcshcdasfspsHh
stticr
Fig. 9. SEM images of MG63 cells cultured on HA-coated rod (a and c); HA-coa
hat different crystallographic faces in HA have different affinityor macromolecule and protein adsorption [13].
It has been reported that biological responses, including cell pro-iferation and organization of cytoskeleton, are affected by surfacerystallinity of the biomaterial [4]. The greater amount of crystallineurfaces that develop the better organized cytoskeleton and theigher degree of osteoblast spreading [44]. XRD patterns of the HA-oated rod and sheet samples show sharp peaks that indicate a highegree of crystallinity, which would promote better cell stretchingnd more lamellipodia and filopodia existence on the HA-coatedamples. Moreover, it has been reported that calcium ions in HAorm sites of positive charge on the surface which aids in the adhe-ion of fibronectin and vitronectin [45]. It is well known that theresence of these two proteins positively affect cell attachment andpreading [46]. Therefore, higher MG63 spreading on the surface ofA-coated samples in comparison to the non-coated ones couldave been influenced by the presence of Ca2+ ions [47,48].
EDS spectra of the HA-coated samples after cellular attachmenthowed higher concentrations of Si and Ca on the HA-coated sheetitanium samples (Fig. 5). Silica has been shown to contribute to
he formation of biosilica, an inorganic polymer, and can inducen vitro biomineralization. At the same time, it is an inorganicomponent in the early steps of bone formation [49,50]. It iseported that the formation of biosilica begins with the expression
eet (b and d); non-coated rod (e) and non-coated sheet (f) titanium samples.
of silicatein enzyme intracellularly and continues to grow aroundan axial filament composed of this enzyme. It is extruded fromthe cells and completes its final form and size extracellularly.Even though the first steps of biosilica formation within thecells are becoming increasingly understood, it is unclear howthe extracellular silicatein strings are formed [51]. It is, however,documented that the biogenic silicon (biosilica) is colocalized withHA in bone tissue [52] and its formation has been directly corre-lated to osteoblast activity [53,54]. Therefore, the higher amountsof silica and calcium detected on the HA-coated sheet couldthe result of higher osteoblast activity, improved cell–substrateinteraction and/or enhanced degree of biomineralization observedon the surface of this sample. This could be another indication fororientational dependency of the interactions that occurred on thesurface of HA-coated sheet titanium samples.
As a final note, ALP activity is conventionally used as the initialmarker for osteoblast differentiation [55]. Our results showed thatthe ALP activity is significantly higher on the surface of HA-coatedsamples being highest on the HA-coated sheet samples comparedto the non-coated samples at all time points. There is no significantdifference in ALP activity between non-coated sheet and rod
samples, which is in agreement with previous studies showingthe independency of ALP activity from crystallographic texture[56,57]. It is, however, generally accepted that ALP transcription
2 es B: B
aimoAes
5
ishiibitec
C
R
tch
A
oD(fEE
R
[
[[[[
[[[[[
[
[[[[
[
[[
[[
[[[[[[[
[
[
[
[[
[
[[
[[
[[[[
[
[
[
[
[
[[
08 A.T. Rad et al. / Colloids and Surfac
nd activity would increase on any calcium–phosphate coating orn the presence of Ca2+ ions and improve osteoblast activity and
ineralization [54,58]. In addition, it is possible that the formationf biosilica on the surfaces of HA-coated sheet samples increasesLP activity [58,59]. Therefore, the effect of this polymer on thelevated ALP activity that detected on the surface of HA-coatedheet sample cannot be excluded.
. Conclusions
We have shown that crystallographic texture of the underly-ng substrate can be used to affect the atomic structure of theurface coating and to improve cellular response. These findingsave implications to enhance the quality of HA-coated biomaterials
ncluding dental and other bone related implants. Adhesivity, phys-cal properties and mechanical performance could also be improvedy inducing a specific crystallographic orientation in the HA coat-
ng. Future studies should be directed toward the investigation ofhe mechanical behavior of coatings having different preferred ori-ntations and the molecular mechanisms involved in the observedell–substrate interactions.
onflict of interest
We declare that we have no conflicts of interest.
ole of the funding source
The sponsor of the study had no role in study design, data collec-ion, data analysis, data interpretation, or writing of the report. Theorresponding author had full access to all the data in the study andad final responsibility for the decision to submit for publication.
cknowledgements
We wish to thank Dr. Jerzy A. Szpunar for scientific supportf this work and appreciate the assistance of Dr. Kelly Sears andr. Huolong Li (Pole figure measurement), Slawomir Poplawski
XRD), Leila Daneshmandi and Fatemeh Fayyazbakhsh. We grate-ully acknowledge the financial support of this work by Tissuengineering and Biomaterials Division, National Institute of Geneticngineering and Biotechnology (NIGEB).
eferences
[1] J. Duyck, I. Naert, Clin. Oral Investig. 2 (1998) 102.[2] Z. Schwartz, B. Boyan, J. Cell. Biochem. 56 (1994) 340.[3] A. Stoch, A. Brozek, G. Kmita, J. Stoch, W. Jastrzebski, A. Rakowska, J. Mol. Struct.
596 (2001) 191.[4] N. Eliaz, S. Shmueli, I. Shur, D. Benayahu, D. Aronov, G. Rosenman, Acta Bio-
mater. 5 (2009) 3178.[5] O. Albayrak, O. El-Atwani, S. Altintas, Surf. Coat. Technol. 202 (2008) 2482.[6] M. Manso, S. Ogueta, P. Herrero-Fernandez, L. Vázquez, M. Langlet, J. Garcia-
Ruiz, Biomaterials 23 (2002) 3985.
[7] J. Ma, H. Wong, L. Kong, K. Peng, Nanotechnology 14 (2003) 619.[8] C.W. Yang, T.M. Lee, T.S. Lui, E. Chang, Mater. Sci. Eng. C 26 (2006) 1395.[9] M. Ogiso, J. Long. Term Eff. Med. Implants 8 (1998) 193.10] O. Engler, V. Randle, Introduction to Texture Analysis: Macrotexture, Microtex-
ture, and Orientation Mapping, CRC, Florence, KY, 2009.
[
[
iointerfaces 103 (2013) 200– 208
11] C.S. Barrett, T.B. Massalski, Structure of Metals, Pergamon Press, Oxford, 1980.12] G. Falini, S. Albeck, S. Weiner, L. Addadi, Science 271 (1996) 67.13] Q.Q. Hoang, F. Sicheri, A.J. Howard, D.S.C. Yang, Nature 425 (2003) 977.14] W.J. Shaw, A.A. Campbell, M.L. Paine, M.L. Snead, J. Biol. Chem. 279 (2004)
40263.15] E. Zimmerman, L. Addadi, B. Geiger, J. Struct. Biol. 125 (1999) 25.16] K. Sato, T. Kogure, Y. Kumagai, J. Tanaka, J. Colloid Interface Sci. 240 (2001) 133.17] C.M. Roome, C.D. Adam, Biomaterials 16 (1995) 691.18] W. Xue, X. Liu, X. Zheng, C. Ding, J. Biomed. Mater. Res. Part A 74 (2005) 553.19] F. Chen, W. Lam, C. Lin, G. Qiu, Z. Wu, K. Luk, W. Lu, J. Biomed. Mater. Res. Part
B Appl. Biomater. 82 (2007) 183.20] V. Randle, O. Engler, Microtexture and Orientation Mapping, vol. 270, Gordon
and Breach Science Publishers, Amsterdam, 2000.21] R. Corp. TexTools, (2002).22] N. Eliaz, T. Sridhar, Cryst. Growth Des. 8 (2008) 3965.23] C. Kwok, P. Wong, F. Cheng, H. Man, Appl. Surf. Sci. 255 (2009) 6736.24] M. We, A. Ruys, M. Swain, B. Milthorpe, C. Sorrell, J. Mater. Sci. Mater. Med. 16
(2005) 101.25] M. Javidi, S. Javadpour, M. Bahrololoom, J. Ma, Mater. Sci. Eng. C 28 (2008)
1509.26] O. Albayrak, C. Oncel, M. Tefek, S. Altintas, Rev. Adv. Mater. Sci. 15 (2007) 10.27] P. Ducheyne, S. Radin, M. Heughebaert, J. Heughebaert, Biomaterials 11 (1990)
244.28] A. Buys, C. Sorrell, A. Brandwood, B. Milthorpe, J. Mater. Sci. Lett. 14 (1995) 744.29] P. Ducheyne, W. Van Raemdonck, J. Heughebaert, M. Heughebaert, Biomaterials
7 (1986) 97.30] J. Israelachvili, H. Wennerström, Nature 379 (1996) 219.31] B. Kasemo, J. Prosthet. Dent. 49 (1983) 832.32] B. Kasemo, J. Gold, Adv. Dent. Res. 13 (1999) 8.33] V. Hlady, J. Buijs, Curr. Opin. Biotechnol. 7 (1996) 72.34] S.K. Akiyama, K. Nagata, K.M. Yamada, Biochim. Biophys. Acta 1031 (1990) 91.35] E. Brynda, V. Hlady, J. Andrade, J. Colloid Interface Sci. 139 (1990) 374.36] S. Hattori, J. Andrade, J. Hibbs, D. Gregonis, R. King, J. Colloid Interface Sci. 104
(1985) 72.37] V. Traskine, P. Protsenko, Z. Skvortsova, P. Volovitch, Colloids Surf. A Physic-
ochem. Eng. Aspects 166 (2000) 261.38] J.G. Steele, C. McFarland, B. Dalton, G. Johnson, M.D.M. Evans, C. Rolfe Howlett,
P.A. Underwood, J. Biomater. Sci. Polym. Ed. 5 (1994) 245.39] T.J. Webster, C. Ergun, R.H. Doremus, R.W. Siegel, R. Bizios, J. Biomed. Mater.
Res. 51 (2000) 475.40] G. Altankov, F. Grinnell, T. Groth, J. Biomed. Mater. Res. 30 (1998) 385.41] J. Schakenraad, H. Busscher, C.R.H. Wildevuur, J. Arends, J. Biomed. Mater. Res.
20 (1986) 773.42] V. Traskine, P. Protsenko, Z. Skvortsova, P. Volovitch, Colloids Surf. A: Physic-
ochem. Eng. Aspects 166 (2000) 261.43] K. Anselme, Biomaterials 21 (2000) 667.44] M. Ball, S. Downes, C. Scotchford, E. Antonov, V. Bagratashvili, V. Popov, W.J. Lo,
D. Grant, S. Howdle, Biomaterials 22 (2001) 337.45] B. Feng, J. Weng, B. Yang, S. Qu, X. Zhang, Biomaterials 25 (2004) 3421.46] R. Narayanan, S.Y. Kim, T.Y. Kwon, K.H. Kim, J. Biomed. Mater. Res. Part A 87
(2008) 1053.47] H.S. Cheung, D.J. McCarty, Exp. Cell Res. 157 (1985) 63.48] H.S. Cheung, M.T. Story, D.J. Mccarty, Arthritis Rheum. 27 (1984) 668.49] E.M. Carlisle, J. Nutr. 106 (1976) 478.50] W.E.G. Müller, X. Wang, B. Diehl-Seifert, K. Kropf, U. Schloßmacher, I. Lieber-
wirth, G. Glasser, M. Wiens, H.C. Schröder, Acta Biomater. (2011).51] W.E.G. Müller, S.I. Belikov, W. Tremel, C.C. Perry, W.W.C. Gieskes, A. Boreiko,
H.C. Schröder, Micron 37 (2006) 107.52] W. Landis, D. Lee, J. Brenna, S. Chandra, G. Morrison, Calcif. Tissue Int. 38 (1986)
52.53] W.E.G. Müller, A. Boreiko, X. Wang, A. Krasko, W. Geurtsen, M.R. Custódio, T.
Winkler, L. Lukic-Bilela, T. Link, H.C. Schröder, Calcif. Tissue Int. 81 (2007) 382.54] M. Wiens, X. Wang, U. Schloßmacher, I. Lieberwirth, G. Glasser, H. Ushijima,
H.C. Schröder, W.E.G. Müller, Calcif. Tissue Int. (2010) 1.55] D. Wang, K. Christensen, K. Chawla, G. Xiao, P.H. Krebsbach, R.T. Franceschi, J.
Bone Miner. Res. 14 (1999) 893.56] C. Eriksson, H. Nygren, K. Ohlson, Biomaterials 25 (2004) 4759.57] S. Faghihi, F. Azari, H. Li, M.R. Bateni, J.A. Szpunar, H. Vali, M. Tabrizian, Bioma-
terials 27 (2006) 3532.58] S. Maeno, Y. Niki, H. Matsumoto, H. Morioka, T. Yatabe, A. Funayama, Y. Toyama,
T. Taguchi, J. Tanaka, Biomaterials 26 (2005) 4847.59] B.R. Genge, G.R. Sauer, L. Wu, F.M. McLean, R. Wuthier, J. Biol. Chem. 263 (1988)
