Materials Science and Engineering C 31 (2011) 1809–1818

Contents lists available at SciVerse ScienceDirect

Materials Science and Engineering C

j ourna l homepage: www.e lsev ie r .com/ locate /msec

Nanostructured implant surface effect on osteoblast gene expression andbone-to-implant contact in vivo

Gustavo Mendonça a,b,c,⁎, Daniela Baccelli Silveira Mendonça a,b,Luis Gustavo Pagotto Simões d, André Luis Araújo d, Edson Roberto Leite d,Alexsander Luiz Golin e, Francisco J.L. Aragão a,f, Lyndon F. Cooper b

a Universidade Católica de Brasília, Pós-Graduação em Ciências Genômicas e Biotecnologia, SGAN Quadra 916, Módulo B, Av. W5 Norte 70.790-160-Asa Norte Brasília/DF, Brazilb Bone Biology and Implant Therapy Laboratory, Department of Prosthodontics, University of North Carolina at Chapel Hill, 404 Brauer Hall, CB #7450, Chapel Hill, NC 27511, USAc Universidade Católica de Brasília, Curso de Odontologia, Taguatinga/DF, Brazild Departmento de Química, Universidade Federal de São Carlos-UFSCAR, Rod. Washington Luiz, 13565-905 São Carlos, SP, Brazile Departmento de Engenharia Mecânica, Faculdade de Engenharia Mecânica, Pontifícia Universidade Católica de Curitiba, Curitiba, PR, Brazilf Embrapa Recursos Genéticos e Biotecnologia, Laboratório de Introdução e Expressão de Genes, PqEB W5 Norte, 70770-900, Brasília, DF, Brazil

⁎ Corresponding author at: Bone Biology and Implant Tof Prosthodontics, University of North Carolina at ChapelChapel Hill, NC 27511, USA. Tel.: +1 919 966 4579; fax: +

E-mail addresses: gustavo_mendonca@dentistry.unclyndon_cooper@dentistry.unc.edu (L.F. Cooper).

0928-4931/$ – see front matter © 2011 Elsevier B.V. Alldoi:10.1016/j.msec.2011.08.021

a b s t r a c t

a r t i c l e i n f o

Article history:Received 27 July 2010Received in revised form 17 May 2011Accepted 30 August 2011Available online 10 September 2011

Keywords:Titanium oxideZirconia oxideNanostructuredImplant surfaceGene expression

The aim of this study was to investigate the response of nanostructured implant surfaces at the level ofosteoblast differentiation and its effects in bone-to-implant contact (BIC) and removal-torque values(RTV). CpTi grade IV implants (1.6×4.0 mm) were machined or machined and subsequently coated withan oxide solution. The surfaces were divided into: machined (M), titania-anatase (An), titania-rutile (Ru),and zirconia (Zr). Surfaces were examined by scanning electron microscopy, atomic force microscopy, andby X-ray microanalysis. Implants were inserted in rat tibia and harvested from 0 to 21 days for measurementof Alkaline Phosphatase, Bone Sialoprotein, Osteocalcin, Osteopontin, and RUNX-2 mRNA levels by real timePCR; from 0 to 56 days for RTV; and from 0 to 56 days for BIC. The roughness parameter (Sa) was comparedby one-way ANOVA followed by Tukey Test. Comparison of Torque removal values and histomorphometricmeasurements on implants in vivo was performed by Kruskal–Wallis test and the significance level for allstatistical analyses was set at p≤0.05. mRNA levels on all nanostructured surfaces were increased comparedto M. At 56 days, the mean RTV in Ncm was 11.6±2.5, 11.3±2.4, 11.1±3.5, 9.7±1.4 for An, Ru, Zr, and M,respectively. Higher BIC (%) was measured for all the nanostructured surfaces versus M at 21 and 56 days(pb0.05). Nanostructured topographic features composed of TiO2 or ZrO2 applied to machined cpTi implantpromoted greater mesenchymal stem cell commitment to the osteoblast phenotype and associated increasedBIC and physical association with bone.

herapy Laboratory, DepartmentHill, 404 Brauer Hall, CB #7450,1 919 966 3821.

.edu (G. Mendonça),

rights reserved.

© 2011 Elsevier B.V. All rights reserved.

1. Introduction

Advances in materials science and biotechnology have led to manyimprovements in implant and bone regeneration therapy [1–3].Modifying the implant surface is one way of improving bone-to-implant contact and the host response to a better and more reliableosseointegration process [1–3]. Implant surface technology has expe-rienced many advances during the last two decades [4,5]. In the last10 years many novel implant surfaces have been developed andbecame available in the market [1,2,6]. Nanotechnology allows themanipulation of materials at the atom level creating a surface that is

more interactive to the molecules and structures that are related tothe host cells. Many studies have demonstrated the effect of nano-technology by means of improved cell attachment, proliferation,differentiation, and in the case of bone cells, the deposition and min-eralization of the bone matrix [3,6–15].

Deposition of nanoparticles onto the titanium surface representsone of many approaches to imparting nanofeatures to a titaniumdental implant [4]. Sol-gel techniques achieve deposition of nanome-ter scale accretions to the implant surface [16,17] usually in an aniso-tropic fashion. Alumina, titania, zirconia, hidroxyapatite, and othermaterials can be applied using this technique [18,19]. Owing totheir resultant atomic-scale interactions, the accretions display strongphysical interactions with the underlying surface [18–21].

The exact mechanism by which nanostructured surfaces allow thisimproved cell response is still not fully understood. Some studieshave demonstrated an increase in protein adsorption on the surfacewhen nanostructures are present [10]. Other studies have shown an

1810 G. Mendonça et al. / Materials Science and Engineering C 31 (2011) 1809–1818

upregulation in osteoblast specific genes that drives the osteoblasticpathway [3,22]. Imparting nanofeatures to the surface can create ahydrophilic surface that helps protein adsorption and increased celladhesion [10].

The outcomes of osseointegrated implants have been very predict-able in many different situations, such as full, partial or singleedentulism not only in delayed healing but also in immediate loading[23–25]. However, in poor bone quality conditions, these outcomesmay become unpredictable [26]. By changing the implant surface,the way bone heals around the implant can be modified and thiscan assure a better bone-to-implant contact in poor bone qualitysituations.

Based on the fact that Titanium implant surfaces imparted withnanofeatures can modulate adherent cellular responses, the aim ofthis study was to evaluate the effects of TiO2 and ZrO2 anisotropicnanofeatures on the surface of machined cpTi implants. This studysought to evaluate molecular and mechanical behavior in vivo usingthe rat tibiae model at the level of osteoblast differentiation and atthe level of interfacial bone formation by measuring gene expression,removal torque values and bone-to-implant contact. These anisotrop-ic nanofeatures on these surfaces improved the differentiation of themesenchymal stem cells into osteoblasts and it improved the biome-chanical performance of those implants (increased bone-to-implantcontact and removal torque values). Several researches have beendone in vitro but just a few have been done in vivo. The presence ofnanofeatures, either TiO2 or ZrO2 was able to improve the biologicalresponses of the machined titanium. In this work it was not possibleto separate the effect of the topography from the chemical composi-tion of the surfaces. However, other studies have already demonstrat-ed that the nanotopography could be more important than thechemical composition when dealing with titanium or aluminum sub-strates [7,8].

Fig. 1. AFM and SEM evaluation of the Machined implant surface. A — Surface roughness (Achined surface. At 50,000× and 100,000× magnification the nanofeatures are not observed

2. Materials and methods

2.1. Implant design and surface analysis

Commercially pure grade IV titanium disks (6.0×1.0 mm)(Neodent Implante Osteointegravel Ltda., Curitiba, PR, Brazil) wereprepared by machining. Other disks were subsequently treated bycoating with a Titanium or Zirconium oxide (TiO2 or ZrO2) nanocoat-ing. The coated surfaces were prepared by dip coating the implant in aTitanium or Zirconium containing solution [27]. The disks weredivided into four groups: machined (M), Titania-Anatase (An),Titania-Rutile (Ru), and Zirconia nanocoating (Zr). The surface ofthe disks was examined by a high-resolution scanning electronmicroscope (Supra 35, Carl Zeiss, Oberkochen, Germany) and atomicforce microscopy (Nanoscope IIIA atomic force microscope (DigitalInstruments, Santa Barbara, CA, United States)). Observations weremade at three different points on the disk surfaces, and averagevalues were calculated. The chemical composition of the surfaceswas analyzed by X-ray microanalysis (EDX) in a Zeiss DSM 940A(Carl Zeiss, Oberkochen, Germany) scanning electron microscopeoutfitted with an Oxford INCA X-sight microanalysis system (OxfordInstruments, the United Kingdom). The XPS spectra were recordedon a Kratos Axis Ultra spectrometer with a concentric hemisphericalanalyzer and a delay line detector. Monochromatic Al Ka X-rayswere used at 150 W, and the chamber base pressure was less than10−8 Torr. Survey spectra were obtained at a pass energy of 80 eVand a step size of 1 eV, while high resolution elemental scans weretaken at a pass energy of 20 eV and a step size of 0.1 eV. All spectrawere corrected for the adventitious C 1s peak at 284.6 eV. The Anand Ru surfaces were also examined by X-ray diffraction (XRD)(XRD-6100, Shimadzu, Tokyo, Japan) to confirm their crystallinestructure.

FM) for Machined. B, C and D — SEM images at low and high magnification for the Ma-on the surface (C and D).

Fig. 2. AFM and SEM evaluation of the Titania-Anatase coated implant surface. A — Surface roughness (AFM). B, C and D — SEM images at low and high magnification. At 50,000×and 100,000× magnification the nanofeatures present on the surface are evident (C and D).

Fig. 3. AFM and SEM evaluation of the Titania-Rutile coated implant surface. A — Surface roughness (AFM). B, C and D— SEM images at low and high magnification. At 50,000× and100,000× magnification the nanofeatures present on the surface are evident (C and D).

1811G. Mendonça et al. / Materials Science and Engineering C 31 (2011) 1809–1818

Fig. 4. AFM and SEM evaluation of the Zirconia coated implant surface. A — Surface roughness (AFM). B, C and D — SEM images at low and high magnification. At 50,000× and100,000× magnification the nanofeatures present on the surface are evident (C and D).

Table 1Roughness parameters of the surfaces.

Roughness parameters

M An Ru Zr

Sa (nm) 86.52b 162.21c 60.08a 89.71b

Sq (nm) 111.10 204.34 81.22 109.25Rp, μm) 0.49 0.91 0.36 0.23

Different letters mean statistically significant different groups at p≤0.05.

1812 G. Mendonça et al. / Materials Science and Engineering C 31 (2011) 1809–1818

Threaded implants for the rat tibia were fabricated from commer-cially pure grade IV titanium (1.6×4.0 mm) (Neodent ImplanteOsteointegravel Ltda.). The implants were kept as machined or trea-ted as described above by dip coating the implant in a Titanium orZirconium containing solution. All implants and disks were subse-quently cleaned and sterilized according to standard procedures formanufacture of dental implants.

2.2. Animal surgery

A rat tibia model of osseointegration was used [28]. All procedureswere performed according to a University of Brasília approved proto-col. Male Wistar rats (12-week-old) were purchased and acclimatedfor 7 days prior to initiation of studies. Anesthesia was achievedusing ketamine/xylazine (80–100 mg/kg and 10–20 mg/kg respec-tively) and supplemental local anesthesia was obtained using lido-caine 2% with epinephrine (1:100,000). Each surgical site wasshaved, prepared using betadine scrub and then isolated for implantplacement by a full thickness myocutaneous flap. Implants wereplaced distal to the tibial diaphysis using a stepwise drilling proce-dure performed with dental drills and using sterile saline irrigation.Implants were placed into the osteotomies by a self-tapping proce-dure. The sites were closed using resorbable sutures. Buprenorphine(0.01–0.05 mg/kg) was administered as post-surgical analgesia and5 mg/kg of ketoprofen subcutaneously 24 h post-operatively. Animalswere evaluated continuously following surgery and ambulation usingthe implanted limbs was the defined criteria for immediate recovery.The animals were sedated and euthanized with an overdose ofthiopental.

For the molecular analysis, two implants were placed in each tibiato provide sufficient RNA for each experimental sample. The implantswere randomly placed in sites 7 and 12 mm distal to the articularsurface of each hind limb of a rat. The time points for molecular

analyses were 3, 7, 14 and 21 days after implant placement. Forevery time point, three rats were used for each implant surface(n=3 rats).

For removal torque analysis, two implants were placed in eachtibia. The implants were inserted as described above. The time pointsfor removal torque analysis were 7, 21 and 56 days after implantplacement. For every time point, three rats were used for each im-plant surface (n=3 rats; 6 implants). After sacrifice, implant resis-tance to torque was measured using a torque-gauged wrench(1200ATG-NS; Tochnichi, Tokyo, Japan).

For the histological analysis, one implant was placed in each tibia.The implant was inserted 7 mm away from the knee joint. The timepoints for histological analyses were 7, 21 and 56 days after implantplacement. For every time point, three rats were used for each im-plant surface (n=3 rats; 3 implants). The tibias were removed andfixed in 4% phosphate-buffered formalin (pH 7.0), for 10 days, andthen transferred to a solution of 70% ethanol until processing. Thespecimens were dehydrated in increasing concentrations of ethanolup to 100%, infiltrated and embedded in methylmethacrylate resin(London Resin Company, Berkshire, England), according to the tech-nique described by Donath & Breuner [29]. The sections were pre-pared for histomorphometry and stained with Toluidine Blue foroptic microscopic analysis.

1813G. Mendonça et al. / Materials Science and Engineering C 31 (2011) 1809–1818

2.3. RNA isolation and analysis

Immediately following sacrifice, the implants were unscrewedand the surrounding bone tissue was removed with a 2.0 mm tre-phine. The tissues were kept frozen in liquid nitrogen until the RNAisolation procedure. Tissue was ground with mortar and pistil inliquid nitrogen. Total RNA in the cell lysates was isolated using theTrizol (Invitrogen, Carlsbad, CA) according to the manufacturer's pro-tocol and collected by ethanol precipitation. Total RNA was quantifiedusing UV spectrophotometry. From each total RNA sample, cDNA wasgenerated using SuperScript reverse transcriptase (Invitrogen, Carls-bad, CA) in a standard 20 μL reaction. All cDNAs were subjected to po-lymerase chain reaction (PCR) for GAPDH mRNA as a test of RNAintegrity and cDNA synthesis. Subsequently, equal volumes of cDNAwere used to program real time PCR reactions specific for mRNAsencoding ALP, BSP, OCN, OPN and RUNX-2. Reactions were performedusing TaqMan Universal PCR Master Mix (Applied Biosystems, FosterCity, CA) and thermocycling in an ABI 7200 real time thermocycler(Applied Biosystems, Foster City, CA). Relative mRNA abundancewas determined by the 2−ΔΔCt method and reported as fold induc-tion. GAPDH abundance was used for normalization.

2.4. Histomorphometry

Longitudinal histological sections from each implant were cap-tured through a video camera Sony (Sony Corp, Tokyo, Japan) joinedto a light microscope. The images were analyzed through the Bio-quant Nova (Bioquant Image Analysis Corporation, Nashville, TN),where the percentages of bone-to-implant contact (BIC) were deter-mined. Through linear measurements, the percentages of mineralizedbone in direct contact with the implant surface were determined [28].

2.5. Statistical analysis

Descriptive statistics were calculated using SPSS. The roughnessparameter (Sa) was compared by one-way ANOVA followed by TukeyTest. Comparison of Torque removal values and histomorphometric

Fig. 5. X-ray microanalysis (EDX) of the chemical composition of the im

measurements on implants in vivo was performed by Kruskal–Wallistest. For all statistical analyses, the significance level was set at p≤0.05.

3. Results

3.1. Surface analysis

SEM and AFM evaluation reveal marked topographic differences be-tween M, and nanostructured surfaces. For the An, Ru and Zr implants(Figs. 1–4) at 5000×magnification the coating is visible (change inmor-phology compared to machined surface). At 50,000× and 100,000×magnification nanoscale topography is distributed on the surface withfeatures of approximately 20 nm in diameter (Figs. 2–4). For the Ma-chined (Fig. 1) there is no evidence of nanofeatures on the surface at50,000× or 100,000× magnification.

Surface roughness parameters were obtained from the AFM analysis(Table 1). The roughness profile of each surface is shown in Figs. 1 to 4.Differences between machined surface and all nanostructured surfaceswere evident demonstrating that the coating was able to cover theoriginal machined surface creating a new topography.

The EDX and XPS analyses showed the general composition ofeach implant surface and are shown in Fig. 5. The EDX surface scan-ning also demonstrated that the expected material was distributedover the entire implant surface (Fig. 6). The XPS demonstrated tracesof different chemical components on each surface (Table 2 andFig. 7a). On the M surface traces of Mg, Zn, Na, Ca, S and Si were prob-ably due to the machining, polishing and cleaning processes. A high-resolution analysis showed presence of Ti metallic and titanium oxide(TiO2) on this surface (Fig. 7b). For the nanostructured surfaces, thehigh-resolution scanning demonstrated that the titanium on thesesurfaces was in oxide groups and no traces of titanium metallicwere found (Fig. 7b). They also demonstrated that zirconium foundon Zr surfaces was in oxide groups (Fig. 7d). For the Zr group, ahigh level of zirconium was found on this surface. The small amountof Mg, Zn, N, Ca, P and Si found on these surfaces were attributed tothe cleaning/coating process. The amount of titanium observed onZr surfaces also demonstrates that the oxide surface is composed oftitanium oxide and zirconium oxide for this surface. The X-ray

plant surface. A — Machined. B — Anatase. C — Rutile. D — Zirconia.

Fig. 6. X-ray microanalysis (EDX) of the coated implant. A, B and C — SEM image of the implants. Ti, O and Zr — Scanning of the surface demonstrates the presence of Titanium (Ti),Oxygen (O) and Zirconium (Zr) on the surfaces. (1) Shows an area used for the EDX analysis.

1814 G. Mendonça et al. / Materials Science and Engineering C 31 (2011) 1809–1818

diffraction confirmed that the crystalline form of titanium was eitheranatase or rutile in the respective implant surfaces (data not shown).

3.2. Gene expression analysis

Surface modifications had different effects on osteoblast gene ex-pression. Osteoblast-specific mRNA levels were greater in bone sur-rounding all nanostructured surfaces versus M implants (Figs. 8 and 9).

Table 2Ion composition data from XPS analyses.

Atomic concentration %

Mg 1s Zn 2p Na 1s N 1s Ca 2p S

Machined 0.7 3.66 0.61 0Anatase 0.55 0.43 0.28Rutile 0.24 0.66Zirconia 0.74

3.3. Removal torque and histomorphometric analysis

For the removal torque values, all surfaces were similar at day 7, andthen higher for the Ru implant group at day 21 (Fig. 10). No statisticallysignificant differences were observed between the groups; however, allthe nanostructured surfaces presented a higher RTV compared toMachined at day 56 (Fig. 10). The histomorphometric analysis revealeda higher bone-to-implant contact for all nanostructured surfaces versusM at 21 and 56 days (Fig. 11). A more rapid accrual of interfacial bone

2p P 2p Si 2p Zr 3d O 1s C 1s Ti 2p

.59 3.15 54.68 26.61 10.0161.11 13.63 24

0.93 60.5 18.63 19.051.28 11.61 57.04 21.28 8.05

Fig. 7. A — Representative wide-scan XPS spectra of Machined, Anatase, Rutile, and Zirconia treated Titanium implants. B — High-resolution scan of Ti2p spectra showing the com-ponent peaks. C — High-resolution scan of O1s spectra showing the component peaks. D — High-resolution scan of Zr3d spectra showing the component peaks.

1815G. Mendonça et al. / Materials Science and Engineering C 31 (2011) 1809–1818

was suggested by the increase in BIC from days 7 to 21 compared to Mimplant groups. There were significant differences between all nano-structured (An, Ru and Zr) and Machined groups at days 21 and 56(n=3, pb0.05; Kruskal–Wallis test) (Fig. 11). No statistically signifi-cant differences were found among the nanostructured implantsregarding the BIC (Fig. 11).

4. Discussion

AFM and SEM evaluation demonstrated a marked difference be-tween a machined implant surface and implant surfaces that were ac-crued with a coating that imparted nanofeatures to the surface. In thisstudy, a machined implant was coated with three different coatingmaterials, based on Titanium or Zirconium. The difference betweenthe two TiO2 coatings is regarding to its crystallographic form. Inthe last few years, Zirconia ceramics have also been used as implantmaterial because of its biocompatibility and mechanical propertiesthat make them suitable as materials for dental implants [30–32].TiO2 or ZrO2 was found over the entire surface for the nanostructuredsamples due to the coating process. Little is known about how surfacemodification influences the stability and bone tissue response tozirconia implants, and how a Titanium implant modified with a Zirco-nia coating would behave. Even though the different materials addedto the smooth surface of the Titanium implants improved responsesat the molecular and biomechanical level, proof that one surfaceparameter (nanoscale features alone) affected the bone bondingbehavior of the implant requires isolation of individual variables,such as chemical composition. The main limitation of this investiga-tion involves the superimposition of nanoscale topography (TiO2 orZrO2 coating) on a macro scale structure (machined titanium screw-shaped implant). The present investigation did not experimentally

segregate nanoscale feature effects from potential influences of sur-face chemistry and micron scale topography. This cannot be fully ac-complished using the complex (nano/micron) topographic characterof the implant design under current study. Other studies have alreadypresented positive effects of nanostructured Alumina, Titania andHydroxyapatite when compared to the same materials in micronscale, strongly implicating the role of nanofeatures in control of oste-oblast function [7,8].

One of the main concerns related to coating the implant surface isthe risk of coating detachment and toxicity of related debris. Thisquestion was addressed by Gutwein and Webster [33] who evaluatedthe relationship of particle size and cell viability and proliferationcompared to micron-particles. Nanoparticles of titania and aluminahad less negative impact in cell viability and proliferation. However,owing to their resultant atomic-scale interactions, the accretions dis-play strong physical interactions with the underlying surface.

The roughness (Sa) values as measured by the AFM for the nano-structured and machined surfaces were below 100 nm, except forthe An group. The features for the An and Ru coated implants werecomposed of small nanofeatures of around 100 nm with a centralpore of about 10–15 nm(Figs. 2 and 3). The features for the Zr coatedimplants were composed of small nanofeatures of various shapes(mostly rounded structures) of about 50–100 nm in diameter(Fig. 4). These features were added to the previously machined sur-face as seen in Fig. 1. The mean roughness is just one parameterused to classify the surfaces. The chemical composition and character-istics of the coatings were evaluated by EDX and XPS analyses reveal-ing that the material of interest was deposited evenly on the implantsurfaces (Figs. 5–7).

Very few in vivo studies have been performed to investigate theeffects of a nanostructured implant surface on osseointegration

Fig. 8. Bone-specific mRNA expression in bone tissue adjacent to M, An, Ru or Zr implants. Total RNA was isolated from peri-implant bone tissues after 3, 7, 14 and 21 days. Expres-sion levels (fold change) for ALP BSP, OCN, OPN are compared to Machined day 3 for Machined, Anatase, Rutile and Zirconia coated implants, A, B, C and D, respectively, 3–21 daysfollowing implant placement (2−ΔΔCt method).

1816 G. Mendonça et al. / Materials Science and Engineering C 31 (2011) 1809–1818

[3,34–36]. On the other hand, the effects of nanostructured topogra-phy on adherent osteoblast responses regarding differentiation andmineralization have been addressed in several in vitro studies [7–9].All these studies corroborate the effects of imparted nanotopographyon osteoblast behavior and osseointegration process. This studyemployed the rat model [37] to gain insight into the potential differ-ences that occur during the osseointegration process at Titanium orZirconium oxide coated titanium implant of defined nanometer-scale topography versus implants with definedmachined topography.Together, the in vivo molecular data, the histomorphometry and

Fig. 9. The RNA levels for Runx2 in peri-implant bone tissues surrounding Machined,Anatase, Rutile and Zirconia coated implants are compared as fold change 3–21 daysfollowing implant placement. The results are shown as fold change (2−ΔΔCt method).

the removal torque data suggest that, when compared with machinedcpTitanium implant surfaces, the nanoscale Zirconia or Titania coated im-plant surfaces supported greater osteoblastic differentiation, osteoblast-specific gene expression, related bone accrual andmechanical interactionwith surrounding bone.

The surface-specific gene expression obtained at each time pointdemonstrated a higher level at Runx2 expression for Zr surface.Runx2 is a key transcription gene in osteoblast differentiation andan increase in its levels may be related to the effect of nanostructuredsurfaces on osteoblast [38,39]. Some authors have also shown an

Fig. 10. Removal torque values for Machined, Anatase, Rutile and Zirconia coatedimplants. Removal torque was measured for 6 implants/surface/timepoint.

Fig. 11. Bone-to-implant contact for Machined, Anatase, Rutile and Zirconia coated im-plants. Histomorphometric measurement of bone to implant contact was calculated for3 implants/surface/timepoint. The mean percent of bone to implant contact is shown(7–56 days; *pb0.05).

1817G. Mendonça et al. / Materials Science and Engineering C 31 (2011) 1809–1818

increase in Runx2 levels on Osseospeed implant surface (anothernanostructured surface) [3,22]. In this study the Runx2 mRNA levelsfor Zr and Ru surface increased up to day 14, and for An the levelsstarted higher and decreased after that. This behavior may be relatedto the different chemical composition of each surface.

This Sol-gel technique resulted in an anisotropic distribution ofthe nanofeatures on the surfaces. The effects of the distribution ofnanofeatures have also been evaluated by several authors. Dalbyand co-workers [40] demonstrated that the presence of disorderednanofeatures could initiate, in MSCs, the expression of some osteo-genic markers even without cell culture induction of the cells alongthe osteoblast pathway. In this report, in vivo measurements ofmRNA levels in bone surrounding the various implants showed anincrease in Runx2 mRNA levels as well as concomitant increases inALP, BSP, OCN, OPN levels to Zr, Ru and An, but not M (Figs. 8and 9). Runx2 elevations will increase the expression of otherbone related genes such as Alkaline phosphatase, Collagen type I,Osteocalcin, and Osteopontin [38]. Specific conclusions about the or-ganization, distribution and size of nanofeatures are yet to be deter-mined and have been the focus of other investigations [4,41].

ALP, BSP, OCN and OPN are marked genes of osteoblast differentia-tion and expressed at various time-points related to stage of cell com-mitment [42]. In this study, An surface presented an increase in ALP,OCN and OPN at day 7 (Fig. 8) followed by a higher peak of Runx2expression at day 3 (Fig. 9). The same pattern is noted to Ru and Zr,although they are both one time-point shifted (to days 7 and 14).

For the removal torque analysis similar low valueswere observed forall surfaces at day 7. This is consistent with the early responses of bonetissue to surgical intervention. At day 21 up to day 56, the Ru implantspresented a higher increase in removal torque values compared to Mimplants. An presented a slight increase at day 21 and the highestvalue of all surfaces at day 56. Zr only increased the RTV at day 56. Allthis behavior may be related to the chemical composition of thesurfaces. Statistically significant differences in removal torque valueswere not revealed in this study, however all nanostructured surfacespresented higher removal torque values than M at day 56 (Fig. 10).

The removal torque values are in accordance with the other re-sults in this paper and may be correlated to a significantly greaterbone accrual at the nanostructured implants after 21 days. Suggestedis a relationship between torque removal and bone to implant con-tact. The torque removal analysis in this study appears to be relatedto the other results obtained in this study and also with other papersthat used the same analysis [43]. Although the increase in BIC and thegene expression pattern did not lead to a statistically significant dif-ference in the RTV it may be related to the challenges of the animalmodel used in this study. The acknowledgment of increased bone to

implant contact had occurred in the same time frame as increasedtorque removal and suggests that the surface nanofeatures influencedbone accrual predominantly and potential physicochemical interac-tions with surrounding bone (bone bonding) secondarily. While thiscannot be fully excluded without direct comparisons among similarmaterials with different nanotopographies, the results suggest thatthis Nanostructured surfaces (An, Ru and Zr) promoted greater phys-ical interaction of the implant with bone through a process ofenhanced bone accrual.

According to our results, it is not clear yet how nanotopographyimproves osteoblast response, but it is clear it has an important rolein cell differentiation. A systematic investigation of how nanoscale to-pography of a given bulk chemistry affects the processes underscor-ing the result of osseointegration is indicated.

5. Conclusion

Machined cpTi implant modifiedwith a nanoscale coating of TiO2 orZrO2 promoted greater mesenchymal stem cell commitment to theosteoblast phenotype and associated increase in bone-to-implant con-tact and physical associationwith bone. However,more studies are nec-essary to evaluate the long-termeffects of such surfacemodifications onimplant stability and osseointegration. The presence of TiO2 or ZrO2

nanofeatures applied to a machined cpTitanium endosseous implantmay directly influence cell behavior to enhance osseointegration.

Acknowledgments

The authors would like to thank CAPES (Coordenação de Aperfei-çoamento de Pessoal de Nível Superior), CNPq (Conselho Nacionalde Desenvolvimento Científico e Tecnológico), and Neodent ImplanteOsteointegravel for their contributions to this article through grants.

References

[1] D. Buser, N. Broggini,M.Wieland, R.K. Schenk, A.J. Denzer, D.L. Cochran, B. Hoffmann,A. Lussi, S.G. Steinemann, J. Dent. Res. 83 (2004) 529–533.

[2] L.F. Cooper, Y. Zhou, J. Takebe, J. Guo, A. Abron, A. Holmen, J.E. Ellingsen, Biomaterials27 (2006) 926–936.

[3] J. Guo, R.J. Padilla, W. Ambrose, I.J. De Kok, L.F. Cooper, Biomaterials 28 (2007)5418–5425.

[4] G. Mendonca, D.B. Mendonca, F.J. Aragao, L.F. Cooper, Biomaterials 29 (2008)3822–3835.

[5] T. Albrektsson, A. Wennerberg, Int. J. Prosthodont. 17 (2004) 536–543.[6] V.C. Mendes, R. Moineddin, J.E. Davies, Biomaterials 28 (2007) 4748–4755.[7] T.J. Webster, R.W. Siegel, R. Bizios, Biomaterials 20 (1999) 1221–1227.[8] T.J. Webster, C. Ergun, R.H. Doremus, R.W. Siegel, R. Bizios, Biomaterials 21 (2000)

1803–1810.[9] T.J. Webster, C. Ergun, R.H. Doremus, R.W. Siegel, R. Bizios, J. Biomed. Mater. Res.

51 (2000) 475–483.[10] T.J. Webster, L.S. Schadler, R.W. Siegel, R. Bizios, Tissue Eng. 7 (2001) 291–301.[11] R.L. Price, L.G. Gutwein, L. Kaledin, F. Tepper, T.J. Webster, J. Biomed. Mater. Res. A

67 (2003) 1284–1293.[12] R.L. Price, K.M. Haberstroh, T.J. Webster, Med. Biol. Eng. Comput. 41 (2003)

372–375.[13] G. Mendonca, D.B. Mendonca, L.G. Simoes, A.L. Araujo, E.R. Leite, W.R. Duarte, L.F.

Cooper, F.J. Aragao, Int. J. Oral Maxillofac. Implants 24 (2009) 205–215.[14] G. Mendonca, D.B. Mendonca, L.G. Simoes, A.L. Araujo, E.R. Leite, W.R. Duarte, F.J.

Aragao, L.F. Cooper, Biomaterials 30 (2009) 4053–4062.[15] G. Mendonca, D.B. Mendonca, F.J. Aragao, L.F. Cooper, J. Biomed. Mater. Res. A 94

(2010) 169–179.[16] H.W. Kim, Y.H. Koh, L.H. Li, S. Lee, H.E. Kim, Biomaterials 25 (2004) 2533–2538.[17] S.H. Lee, H.W. Kim, E.J. Lee, L.H. Li, H.E. Kim, J. Biomater. Appl. 20 (2006) 195–208.[18] B. Ben-Nissan, A.H. Choi, Nanomedicine (Lond) 1 (2006) 311–319.[19] A.H. Choi, B. Ben-Nissan, Nanomedicine (Lond) 2 (2007) 51–61.[20] L.D. Piveteau, B. Gasser, L. Schlapbach, Biomaterials 21 (2000) 2193–2201.[21] J.L. Arias, M.B. Mayor, J. Pou, Y. Leng, B. Leon, M. Perez-Amor, Biomaterials 24

(2003) 3403–3408.[22] Z.M. Isa, G.B. Schneider, R. Zaharias, D. Seabold, C.M. Stanford, Int. J. Oral Maxillofac.

Implants 21 (2006) 203–211.[23] R. Adell, U. Lekholm, B. Rockler, P.I. Branemark, Int. J. Oral Surg. 10 (1981)

387–416.[24] R. Adell, B. Eriksson, U. Lekholm, P.I. Branemark, T. Jemt, Int. J. OralMaxillofac. Implants

5 (1990) 347–359.[25] L.F. Cooper, I.J. De Kok, F. Rojas-Vizcaya, P. Pungpapong, S.H. Chang, Compend.

Contin. Educ. Dent. 28 (2007) 216–225 quiz 226.

1818 G. Mendonça et al. / Materials Science and Engineering C 31 (2011) 1809–1818

[26] F.D. das Neves, D. Fones, S.R. Bernardes, C.J. do Prado, A.J. Neto, Int. J. Oral Maxillofac.Implants 21 (2006) 86–93.

[27] M. Pechini. U.S. Patent; 3,300,697; 1967.[28] T. Masuda, G.E. Salvi, S. Offenbacher, D.A. Felton, L.F. Cooper, Int. J. Oral Maxillofac.

Implants 12 (1997) 472–485.[29] K. Donath, G. Breuner, J. Oral Pathol. 11 (1982) 318–326.[30] M. Gahlert, T. Gudehus, S. Eichhorn, E. Steinhauser, H. Kniha, W. Erhardt, Clin.

Oral Implants Res. 18 (2007) 662–668.[31] J. Oliva, X. Oliva, J.D. Oliva, Int. J. Oral Maxillofac. Implants 22 (2007) 430–435.[32] H.J. Wenz, J. Bartsch, S. Wolfart, M. Kern, Int. J. Prosthodont. 21 (2008) 27–36.[33] L.G. Gutwein, T.J. Webster, Biomaterials 25 (2004) 4175–4183.[34] J.E. Ellingsen, C.B. Johansson, A. Wennerberg, A. Holmen, Int. J. Oral Maxillofac.

Implants 19 (2004) 659–666.[35] L. Meirelles, L. Melin, T. Peltola, P. Kjellin, I. Kangasniemi, F. Currie, M. Andersson,

T. Albrektsson, A. Wennerberg, Clin. Implant. Dent. Relat. Res. 10 (2008) 245–254.

[36] L. Meirelles, F. Currie, M. Jacobsson, T. Albrektsson, A. Wennerberg, Int. J. OralMaxillofac. Implants 23 (2008) 641–647.

[37] A. Abron, M. Hopfensperger, J. Thompson, L.F. Cooper, J. Prosthet. Dent. 85 (2001)40–46.

[38] H. Harada, S. Tagashira, M. Fujiwara, S. Ogawa, T. Katsumata, A. Yamaguchi, T.Komori, M. Nakatsuka, J. Biol. Chem. 274 (1999) 6972–6978.

[39] S. Harada, G.A. Rodan, Nature 423 (2003) 349–355.[40] M.J. Dalby, N. Gadegaard, R. Tare, A. Andar, M.O. Riehle, P. Herzyk, C.D. Wilkinson,

R.O. Oreffo, Nat. Mater. 6 (2007) 997–1003.[41] S. Lavenus, G. Louarn, P. Layrolle, Int. J. Biomater. 2010 (2010) 915327.[42] L.F. Cooper, P.K. Yliheikkila, D.A. Felton, S.W. Whitson, J. Bone Miner. Res. 13

(1998) 620–632.[43] S. Narai, S. Nagahata, Int. J. Oral Maxillofac. Implants 18 (2003) 218–223.