Enhanced differentiation of human osteoblasts on Ti surfaces pre-treated with human whole blood

Acta Biomaterialia xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Acta Biomaterialia

journal homepage: www.elsevier .com/locate /actabiomat

Enhanced differentiation of human osteoblasts on Ti surfaces pre-treatedwith human whole blood

http://dx.doi.org/10.1016/j.actbio.2015.03.0221742-7061/� 2015 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

⇑ Corresponding author at: Laboratory for Biointerfaces, Lerchenfeldstrasse 5,Empa, Swiss Federal Laboratories for Materials Science and Technology, CH-9014 St.Gallen, Switzerland. Tel.: +41 58 765 74 47.

E-mail address: [email protected] (K. Maniura-Weber).

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of human osteoblasts on Ti surfaces pre-treated with human whole blooBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

Brigitte S. Kopf a, Angela Schipanski a,b, Markus Rottmar b,c, Simon Berner d, Katharina Maniura-Weber a,b,⇑a Laboratory for Materials Biology Interactions, Empa, Swiss Federal Laboratories for Materials Science and Technology, St. Gallen, Switzerlandb Laboratory for Biointerfaces, Empa, Swiss Federal Laboratories for Materials Science and Technology, CH-9014 St. Gallen, Switzerlandc Department of Urology, University Hospital Zurich, CH-8091 Zurich, Switzerlandd Institut Straumann AG, CH-4002 Basel, Switzerland

a r t i c l e i n f o a b s t r a c t

Article history:Received 24 December 2014Received in revised form 2 March 2015Accepted 19 March 2015Available online xxxx

Keywords:Titanium implant surfacesHydrophilicityBlood material interactionOsteogenic differentiationCell material interaction

Early and effective integration of a metal implant into bone tissue is of crucial importance for its long-term stability. While different material properties including surface roughness and wettability but alsoinitial blood-implant surface interaction are known to influence this osseointegration, implications ofthe latter process are still poorly understood. In this study, early interaction between blood and theimplant surface and how this affects the mechanism of osseointegration were investigated. For this, bloodcoagulation on a micro-roughened hydrophobic titanium (Ti) surface (SLA-Hphob) and on a hydrophilicmicro-roughened Ti surface with nanostructures (SLActive-HphilNS), as well as the effects of wholehuman blood pre-incubation of these two surfaces on the differentiation potential of primary humanbone cells (HBC) was assessed. Interestingly, pre-incubation with blood resulted in a dense fibrin networkover the entire surface on SLActive-HphilNS but only in single patches of fibrin and small isolated fibrecomplexes on SLA-Hphob. On SLActive-HphilNS, the number of HBCs attaching to the fibrin network wasgreatly increased and the cells displayed enhanced cell contact to the fibrin network. Notably, HBCs dis-played increased expression of the osteogenic marker proteins alkaline phosphatase and collagen-I whencultivated on both surfaces upon blood pre-incubation. Additionally, blood pre-treatment promoted anearlier and enhanced mineralization of HBCs cultivated on SLActive-HphilNS compared to SLA-Hphob.The results presented in this study therefore suggest that blood pre-incubation of implant surfacesmimics a more physiological situation, eventually providing a more predictive in vitro model for theevaluation of novel bone implant surfaces.

� 2015 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction

The long-term stability of titanium (Ti) dental implants highlydepends on an early and effective integration of the metal deviceinto bone tissue. It has been shown that the surface properties ofthe implant, such as roughness, chemical composition, surfaceenergy and wettability regulate the response of osteoblasts [1–4].On rough surfaces, the osteoblasts show improved initial cell reac-tions including the production of a collagen (Col) -rich extracellularmatrix (ECM) [5], increased alkaline phosphatase (ALP) activity andelevated levels of osteocalcin expression [6]. Also, surface rough-ness has been shown in vitro to promote the formation of physio-logic minerals compared to dystrophic mineral structures on flat

surfaces [7]. Notably, an enhanced bone formation was demon-strated in vivo around hydrophilic micro-structured implant sur-faces when compared to surfaces with the same micro-structurebut with a hydrophobic surface [8–11]. Many in vitro studies usingeither human osteoblasts, mesenchymal stem cells (MSCs)[2,12,13] or osteoblast-like cell lines [4,14,15] have investigatedthe underlying mechanism that may explain the regulatory effectof surface roughness on in vivo bone formation. Integrin signallingwas found to be one of the pathways involved in osteoblastic dif-ferentiation on Ti, especially when the cells are cultivated onhydrophilic micro-rough surfaces [16,17].

It has been furthermore suggested that the initial blood-implantsurface interactions are important determinants of peri-implanthealing [18,19]. During surgery, the implant surfaces transientlyget in contact with blood of the patient. Within a short timevitronectin, immunoglobulins, fibrinogen and fibronectin areadsorbed from blood on the implant surface, building a layer of

d. Acta

2 B.S. Kopf et al. / Acta Biomaterialia xxx (2015) xxx–xxx

plasma proteins [20,21] that consists of important cell-adhesion-promoting and cell-activating ligands [20,22]. Interestingly, theexact composition of the adsorbed protein layer was found to bedependent on the materials surface [23–25]. When plasma pro-teins such as fibrinogen and fibronectin get in contact with theimplant surface, several biological processes are initiated includingthe coagulation cascade, fibrinolysis and activation of the comple-ment system [26,27]. Of special importance is fibrinogen, a precur-sor molecule of fibrin, which plays an important role not only inclot formation and fibrinolysis, but also for cell–matrix interactions[28], integrin-receptor-based cell adhesion and proliferation onsurfaces [19,29]. Fibrin provides a temporary matrix at the site ofinjury that can rapidly be remodelled, invaded and replaced bycell-associated proteolytic activity [30]. This is of crucialimportance, since migration of osteoblastic cells occurs throughthe provisional fibrin scaffold to reach the protein layer on theimplant [31].

Several studies investigated the influence of topographical [32–35] and physiochemical [36–38] Ti surface properties on theadsorption of plasma proteins, the induction of the blood coagula-tion cascade and blood-cell adhesion. In particular it was shownthat blood clot formation was enhanced on super-hydrophilic Tisurfaces, leading to a dense fibrin network, whereas on hydropho-bic Ti surfaces only a thin, non-structured network was observed[37,39–41]. Increased platelet activation, known to be involved inthe first step of wound healing [42], stronger complement activa-tion, and reduced numbers of macrophages/monocytes have alsobeen reported on hydrophilic surfaces compared to hydrophobicsurfaces with the same topographical features [40]. Platelets fur-thermore play an important role in haemostasis and inflammation,and their activation, and thus the release of growth factors, isenhanced by surface microtopography [18,19,43]. Interestingly,several growth factors including transforming growth factor b, pla-telet-derived growth factor, insulin growth factor, basic fibroblastgrowth factor and vascular endothelial growth factor, which areall released by blood platelets in response to injury [42], are knownto support revascularization and osseointegration [44]. Therefore, astructured layer of blood components on implant surfaces may beof major importance in the early phase of bone healing arounddental implants to promote osseointegration.

In the present study, the influence of blood components inter-acting with Ti surfaces with different wettability (hydrophilic withnanostructures (NS), hydrophobic without NS) on the differentia-tion potential of primary human bone cells (HBC) was assessedin vitro. For this, the commercially available, unmodifiedhydrophobic (Hphob) Ti SLA dental implant surface and the super-hydrophilic (Hphil) SLActive surface were subjected to humanwhole blood. Both the effect of the chemical properties of thedifferent surfaces on the blood clot formation and the effect ofpre-incubation of these two surfaces with whole human blood ondifferentiation and mineralization of primary human osteoblastswere evaluated. Blood pre-incubation of implant surfaces poten-tially mimics a more physiological situation and therefore mightbe more predictive for evaluating novel implant surfaces prior toanimal experiments.

2. Materials and methods

2.1. Titanium discs

All samples and surface modifications were prepared byStraumann (Institut Straumann AG, Basel, Switzerland). The sur-faces of the SLA-Hphob and SLActive-HphilNS samples in this studycorrespond to the commercially available SLA� and SLActive�

implant surfaces, respectively. Discs of 5 mm diameter were

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of huBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

prepared from a c.p. titanium sheet grade 2. For both sample typeslarge grit sandblasting (with corundum, particle size 250–500 lm)was applied to generate a macro roughness (lateral feature sizeabout 20–40 lm), and subsequent acid etching was used to gaina micro topography (lateral feature size about 2–4 lm). SLA-Hphob samples were sandblasted, acid-etched in a boiling mixtureof HCl and H2SO4, followed by cleaning in HNO3 and rinsing indeionized water according to a proprietary process byStraumann. Finally the discs were air dried and packed in alu-minum foil. SLActive-HphilNS samples were prepared employingthe same sandblasting and acid-etching process as for SLA,however, further treatment took place under nitrogen to preventexposure to air. The samples were rinsed in 0.9% NaCl and finallystored in 0.9% NaCl solution at pH 5. Notably, it was reported thatnanostructures (NS) spontaneously form on Ti SLActive surfaces[45]. All discs were c-sterilized (25–42 kGy) before being usedfor investigations. Wettability of SLA-Hphob and SLActive-HphilNShas been investigated extensively in previous studies by our groupand others [46,47].

2.2. Partially heparinized whole human blood

Whole human blood from healthy volunteers (ethical approvalwas obtained from the local ethics committee; EKSG 12/111) waspartially heparinized upon withdrawal by using 9 ml S-Monovette tubes (neutral S-Monovette�, Sarstedt, Switzerland)with 3 IU/ml sodium heparin (Carl Roth, Karlsruhe, Germany)resulting in a final concentration of 0.5 IU heparin/ml blood.Blood samples were kept at room temperature and used within1 h after withdrawal.

2.3. Teflon model and blood incubation

The Ti samples were placed into a custom made device madeout of polytetrafluoroethylene (PTFE, Teflon), designed as a roundchamber with an inner diameter of 50 mm and 4 mm height, con-taining 21 cavities with an inner diameter of 5 mm and a height of1 mm. The device was filled with 7.8 ml of blood, closed with aPTFE lid, sealed off with parafilm and placed on a wave shaker(Polymax, Heidolph�, Germany) at 10 rpm and 37 �C for 10 min.The different sample types were always incubated in separatechambers. Blood was carefully removed after incubation and sam-ples were rinsed gently 4 times in pre-warmed phosphate-bufferedsaline (PBS, Sigma) before usage for further experiments. Eachexperiment was carried out 5 times using blood from differentdonors.

2.4. Primary human bone cells

Bone marrow samples were obtained from patients undergoingsurgical hip replacement after informed consent (ethical approvalwas obtained from the local ethics committee; EKSG 08/14).Femur-derived bone marrow samples containing pieces of trabecu-lar bone were incubated in isolation medium containing 25 mMHEPES (Fluka, Switzerland), 128.5 mM NaCl (Fluka), 5.4 mM KCl(Fluka), 5.5 mM D(+)- (Sigma, Switzerland), 51.8 mM D(+)-saccha-rose (Fluka) and 0.1% BSA (Sigma) overnight at 4 �C. After cen-trifugation at 110�g for 15 min at 4 �C the supernatant wasdiscarded. The residual pellet that included pieces of trabecularbone was repeatedly washed with isolation medium under micro-scopic observance to rinse out tissue and cells from the bonepieces. For primary human bone cell (HBC) cultures, several piecesof bone with a total weight of approximately 25 mg were trans-ferred into a T75 culture flask containing proliferation mediumconsisting of a-MEM (22561, Gibco, Switzerland), 10% Fetal calfserum (FCS; Lonza, Switzerland), 1% PSN (penicillin (5 mg/ml),

man osteoblasts on Ti surfaces pre-treated with human whole blood. Acta

B.S. Kopf et al. / Acta Biomaterialia xxx (2015) xxx–xxx 3

streptomycin (5 mg/ml), neomycin (10 mg/ml) all from Gibco,Switzerland) and 1 ng/ml basic fibroblast growth factor (FGF-2,Sigma). After 1 week of cultivation at 37 �C in humidified 5% CO2

atmosphere, bone cells started to grow out onto the tissue cultureflask. At a confluence between 80 and 90% (approximately after3 weeks), cells were seeded into new T75 flasks (1 � 105 cells/flask) and cultured in proliferation medium until they were usedfor cell culture experiments on the test samples. All experimentswere performed with HBCs of passage 2 as described above.

2.5. Cell seeding

Sub-confluent primary HBCs were incubated with trypsin solu-tion (0.5 g/l trypsin, 0.2 g/l EDTA, Sigma) for 4 min, resuspended inproliferation medium before centrifugation at 110�g for 15 min at4 �C. Prior to cell seeding, the different Ti samples, eitherpreviously incubated with blood for 10 min or not, were placedinto commercial 96-well cell culture plates (TPP, Switzerland).Cells were seeded in proliferation medium on Ti samples (±bloodpre-incubation) at a cell seeding density of 2 � 104 cells/cm2 and5 � 103 cells/cm2 on tissue culture polystyrene (TCPS) as a control.Ti samples were transferred into new wells with differentiationmedium after 24 h. Differentiation medium consisted of a-MEM,10% FCS, 1% PSN, 50 lM ascorbic acid phosphate (Sigma), 2 mMb-glycerophosphate (Sigma), 10 nM 1,25-dihydroxy-vitamine D3

(Sigma) and 10 nM dexamethasone; the two latter chemicals wereprepared and added freshly each time when medium was changed.Additionally, cells on TCPS were cultivated in differentiation med-ium as positive control and in proliferation medium as negativecontrol. All samples were incubated at 37 �C in humidified 5%CO2 atmosphere.

2.6. Immunofluorescence stainings

For confocal laser scanning microscopy and fluorescence micro-scopy, cells on samples were washed 3 times in PBS, fixed for20 min in an aqueous solution of 4% PFA, 65 mM PIPES, 25 mMHEPES, 10 mM EGTA and 3 mM MgCl2 and subsequently permeabi-lized for 10 min in PBS containing 0.1% Triton X 100.

Actin filaments and nuclei were stained with Alexa Fluor 488-labelled phalloidin (1:40, Invitrogen, Switzerland) and 0.01 lg/ml40,6-diamidino-2-phenylindole (DAPI, Sigma) for 1 h. Fibrin fibreswere stained using the monoclonal anti-human fibrinogen anti-body (1:50, Sigma) for 2 h followed by incubation with anti-mouseAlexa 555 (1:300, Molecular Probes, Switzerland). Collagen-I (Col-I) staining was carried out by incubation in a solution of mouseanti-collagen type I antibody (1:1000, Sigma) for 1 h, followed bygoat a-mouse IgG Alexa Fluor 488 (1:400, Molecular Probes,Switzerland) and DAPI staining for 1 h. Alkaline phosphatase(ALP) was stained using anti-human bone alkaline phosphatase(bALP, 1:1000, Developmental Studies Hybridoma Bank, USA) fol-lowed by Alexa Fluor 546 goat anti-mouse (1:400, Invitrogen)and DAPI staining. Unspecific binding sites were blocked beforeany antibody incubation with 5% goat serum and 1% BSA in PBSfor 1 h. All antibodies were dissolved in 1% BSA in PBS and all stain-ings were performed at RT. Secondary antibody incubation wasalways performed in the dark. All incubations of samples with anti-body solutions were followed by 3 washes with PBS.

Microscopic images were taken with 10� or 63� objectiveseither using fluorescence microscopy (Axio Imager.M1, Carl ZeissAG, Switzerland) or laser scanning confocal microscopy (LSM780, Carl Zeiss AG, Switzerland), with filter sets corresponding tothe fluorescence of interest.

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of huBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

2.7. Assessment of cell adhesion

Labelling of HBCs was performed with 1,10-dioctadecyl-3,3,30,30-tetramethylindocarbocyanine perchlorate (DiI; Sigma) accordingto the manufacturer’s instructions. Briefly, 1 � 106 cells/ml werestained in cell culture medium supplemented with 12.5 lg DiIfor 15 min at 37 �C in 5% CO2 atmosphere. Afterwards, cells werewashed 3 times with pre-warmed medium and cultured on Tisamples (±blood pre-incubation) in differentiation medium for24 h. The next day, 30 min before fluorescence microscopy (10�objective), nucleic were stained in fresh differentiation mediumsupplemented with Hoechst 33342 (5 lg/ml; Molecular Probes)at 37 �C in humidified, 5% CO2 atmosphere. Image analysis wasdone with ImageJ [48]. Out of the mosaic images (constituted outof 56 single images), ten single pictures were chosen randomlyand nuclei as well as labelled cells were counted per area (mm2).Afterwards, the ratio of labelled cells per mm2 was assessed.HBCs with DiI and nuclear stain were included into calculationsand cells with DiI stain but without nuclear stain representedblood cells and were excluded from calculations.

2.8. Scanning electron microscopy (SEM), energy-dispersive X-rayspectroscopy (EDX)

Samples undergoing SEM analysis were fixed in Karnovskysolution for 1 h at room temperature. Prior to SEM analysis, cellson samples were dehydrated using a gradient series of ethanol(from 50% to 100%), followed by an incubation in hexamethyldisi-lazane (HMDS, Sigma) for 30 min. Finally, samples were placedinto a new 96-well plate and left to dry overnight at room tem-perature. On the next day, samples were sputter-coated with10 nm gold–platinum (high vacuum coater Leica EM ACE 600,Switzerland). Images were acquired using a Hitachi S-4800 scan-ning electron microscope (Hitachi High-TechnologiesCorporation, Japan) at an accelerating voltage of 5 kV and 10 lAcurrent with magnifications between 700 and 5000 times. EDXanalysis was performed using a Hitachi S-4800 at an acceleratingvoltage of 20 kV and 10 lA current with a magnification of10,000 to 15,000.

2.9. Assessment of cell mineralization

2.9.1. Xylenol orange stainingHBCs were cultivated on Ti samples (±blood pre-incubation) in

differentiation medium and on TCPS in differentiation or prolifera-tion medium as positive and negative control for 14 and in parallelfor 21 days. On day 13 and day 20 xylenol orange (XO; Sigma) wasadded to the cell culture medium to a final concentration of 20 lM.After incubation overnight at 37 �C in humidified, 5% CO2 atmo-sphere, XO-containing medium was changed by the respectivefresh medium containing Hoechst 33342 (5 lg/ml; MolecularProbes) for staining nuclei. After 30 min, cell culture mediumwas again replaced by the respective fresh medium and two sam-ples per Ti type were analysed. Images were taken with a fluores-cence microscope with a 10� objective. Samples were furtherprocessed for quantification of mineralization using QuantiChrom™ Calcium mineralization assay (DICA-500, Gentaur,Switzerland).

2.9.2. Quanti Chrom™ Calcium mineralization assayHBCs were cultivated on different Ti sample types (±blood

pre-incubation) in differentiation medium, as well as on TCPS inproliferation or differentiation medium as controls for 14 and21 days. In order to quantify mineralization, the Ca2+ content of

man osteoblasts on Ti surfaces pre-treated with human whole blood. Acta

4 B.S. Kopf et al. / Acta Biomaterialia xxx (2015) xxx–xxx

HBCs was correlated to the cell number on the respective surface.The number of metabolically active cells in culture was determinedby alamar blue (AB) assay (alamarBlue� Cell Viability Reagent,Invitrogen). Briefly, 10% of medium volume in the well AB wasdirectly added to culture medium followed by an incubation of15 h at 37 �C in a humidified 5% CO2 atmosphere. Fluorescenceintensity was measured at 530 and 635 nm with BiotekInstruments Flx 800, (Witec AG, Switzerland). Cell numbers werecalculated by interpolating fluorescence intensity readings froman 8-point standard curve, measured from different numbers ofcells after 1 day in culture. The standard curve was evaluated inparallel to each experiment with the corresponding HBCs.Afterwards, the Ca2+ content on the same samples was assessedemploying the Quanti Chrom™ Calcium Assay according to themanufacturer’s instructions. Briefly, all samples were washedtwice with pre-warmed Ca2+- and Mg2+-free PBS and lysed in100 ll, 1 M HCl for 3 h at 37 �C under constant agitation.Subsequently, 10 ll of each supernatant were transferred to a96-well plate and 90 ll of working reagent (equal volumes ofQuanti Chrom™ solutions A and B) containing a phenolsul-phonephthalein dye, were added. The dye forms a stable bluecoloured complex with free Ca2+. After 3 min, the absorbance ofthe solution was measured at 595 nm (Biotek Instruments Elx800, Witec AG, Switzerland). Samples were analysed with up to 8samples per group if available but at a minimal number of tripli-cates. The Ca2+ concentrations were calculated by the means ofan 8-point standard curve, which was performed according tothe manufacturer’s instruction in parallel to each experiment.Finally, the ratio of Ca2+ concentration per metabolically active cellwas assessed for each sample. Further, for each Ti sample type andcultivation period 2 samples, previously incubated with or withoutblood, were incubated without cells in medium only and used asblanks for the respective sample type in the different mineraliza-tion assays.

Fig. 1. Fibrin network formation on Ti surfaces after whole blood incubation. SLA-Hphob

human blood before evaluation by (A) SEM imaging and (B) confocal microscopy after swhereas planar, compact patches of fibrin, with scattered single fibres (white arrowsindependent experiments (scale bar 20 lm).

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of huBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

2.10. Statistical analysis

The results are expressed as mean ± SD. The unit of analysis wasa sample (one disc) and the level of significance was set at p < 0.05.Statistical significance was calculated using nonparametric ANOVAon ranks using Graph Pad Prism 6 software (GraphPad Software,Inc., CA, USA).

3. Results

3.1. Interaction of blood components with Ti surfaces SLA-Hphob andSLActive-HphilNS

Representative SEM images of the blood incubationexperiments are presented in Fig. 1A. Early and enhanced bloodcoagulation was observed on SLActive-HphilNS (Fig. 1A), showinga dense fibrin network of individual fibres as well as erythrocytesand leukocytes integrated into the fibrin network. The blood coa-gulation on SLA-Hphob was less pronounced and the surface waspartially covered with single compact fibrin fibres (Fig. 1A).Assessing fibrin deposition on the two different surfaces byimmunohistochemical staining with a specific antibody for fibrin,the fluorescent micrographs show clear differences in the structureand amount of the fibrin fibre network between SLA-Hphob andSLActive-HphilNS (Fig. 1B). Only single fibrin fibres, among compactfibrin patches, were observed on SLA-Hphob (Fig. 1B, arrow heads)compared to SLActive-HphilNS, where area-wide, a more looselyarranged fibrin fibrous network was detectable.

3.2. Bone cell attachment and morphology on surfaces pre-incubatedwith blood

To assess cell attachment on SLA-Hphob and SLActive-HphilNSsurfaces, HBCs were labelled with DiI dye and stained for the nuclei

and SLActive-HphilNS were incubated for 10 min with partially heparinized wholetaining fibrin fibres (red). A dense fibrin network was present on SLActive-HphilNS

) were present on SLA-Hphob. Selected images show representative results of 15

man osteoblasts on Ti surfaces pre-treated with human whole blood. Acta

B.S. Kopf et al. / Acta Biomaterialia xxx (2015) xxx–xxx 5

before evaluating cell number after 1 day of culture (Fig. 2).Qualitative (Fig. 2A) as well as quantitative (Fig. 2B) results showedan increased number of attached cells when both surfaces werepre-incubated with blood compared to surfaces without previouscontact with blood. Significantly more cells attached on SLActive-

Fig. 2. Cell adhesion on Ti surfaces after whole blood incubation. (A) Primary human bSLActive-HphilNS pre-incubated with (blood + HBC) or without whole human blood (microscopic images (scale bar 200 lm). (B) Quantitative assessment of HBCs attached trepeats. Significance was defined as p < 0.05; +significant to SLA-Hphob (HBC + blood); #s

Fig. 3. Cell morphology on Ti surfaces after whole blood incubation. Primary human boincubated with (blood + HBC) or without whole human blood (HBC) before evaluation by(red), the actin cytoskeleton (green) and nuclei (blue). Scale bars in (A) represent 20 lm

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of huBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

HphilNS (HBC + blood) compared to SLA-Hphob (HBC) (p < 0.0001),on SLActive-HphilNS (HBC + blood) compared to SLActive-HphilNS(HBC) (p = 0.001) and on SLActive-HphilNS (HBC + blood) comparedto SLA-Hphob (HBC + blood) (p < 0.05) (Fig. 2B). On SLA-Hphob

(HBC + blood) significantly more cells attached compared to

one cells (HBCs) were stained with DiI (pink), cultivated for 24 h on SLA-Hphob andHBC) and stained for cell nuclei (blue) before acquiring confocal laser scanningo the different surfaces after 24 h. Data is presented as mean ± SD; n = 3 technicalignificant to SLActive-HphilNS (HBC); ⁄significant to SLA Hphob (HBC).

ne cells (HBCs) were cultivated for 4 days on SLA-Hphob and SLActive-HphilNS pre-(A) SEM imaging and (B) confocal microscopy after staining for the fibrin network

, in the smaller insert 2.5 lm and for (B) 10 lm.

man osteoblasts on Ti surfaces pre-treated with human whole blood. Acta

6 B.S. Kopf et al. / Acta Biomaterialia xxx (2015) xxx–xxx

SLA-Hphob (HBC) (p = 0.001). Additionally, significantly more cellsattached on SLActive-HphilNS (HBC) compared to SLA-Hphob (HBC)(p = 0.001).

The cell morphology of HBCs on SLA-Hphob and SLActive-HphilNSpre-incubated with or without blood on day 4 was monitored bySEM imaging (Fig. 3A) and confocal laser scanning microscopy(Fig. 3B). Qualitatively, more contacts of the filopodias of theHBCs with the fibrin network were observed on SLActive-HphilNSand SLA-Hphob pre-incubated with blood (Fig. 3A, lower row) com-pared to cells attached on surfaces which had not previously beenincubated with blood (Fig. 3A, upper row). Overall much moreinteractions of filopodias of the cells with the fibrin scaffold wereobserved on SLActive-HphilNS than on SLA-Hphob when pre-incubated with blood (Fig. 3A, lower row, small inserts). Z-stacksof CLSM micrographs showed in the XZ as well as in the YZ planea continuous fibrin network providing a matrix, in which HBCsare embedded (Fig. 3B).

3.3. Protein expression of Col-I and ALP and on day 7 and day 14

To evaluate the effect of pre-incubation with or without bloodon the osteogenic differentiation of HBCs cultivated on SLA-Hphob

and SLActive-HphilNS, protein expression of Col-I and ALP wasassessed by immunohistochemical staining. On SLA-Hphob andSLActive-HphilNS pre-incubated with blood, overall expression of

Fig. 4. ALP- and Col-I protein expression of HBCs on Ti surfaces after whole blood incubHphob and SLActive-HphilNS pre-incubated with (blood + HBC) or without whole human bconfocal laser scanning images (scale bar 200 lm).

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of huBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

ALP and Col-I was increased on day 7 and day 14 compared tosurfaces without blood pre-treatment (Fig. 4). Notably, bloodpre-incubation induced the Col-I expression much stronger onSLActive-HphilNS compared to SLA-Hphob on both time-points. Toassess the differentiation potential of HBCs, cells from each patientwere also cultured on TCPS in proliferation medium as a negativecontrol and in differentiation medium as a positive control, show-ing ALP and Col-I expression only in differentiation medium butnot in proliferation medium (Fig. S1).

3.4. Mineralization of HBCs on surfaces with and without blood pre-incubation

To analyse the effect of blood pre-treatment of SLA-Hphob andSLActive-HphilNS on calcification of bone cells, HBCs were culturedfor 14 and 21 days on surfaces with or without blood pre-incuba-tion and mineralization was qualitatively assessed by xylenolorange staining (Fig. 5) and quantitatively by measuring the Ca2+

content of HBCs in correlation to the number of metabolicallyactive cells (Fig. 6). Calcification of HBCs was assessed with EDXanalysis (Fig. 7).

On SLActive-HphilNS previously incubated with blood, strongmineralization could be seen already on day 14 and was furtherincreased by day 21 (Fig. 5). In contrast, HBCs cultivated onSLA-Hphob showed almost no mineralization by day 14 and no

ation. Primary human bone cells (HBCs) were cultivated for 7 and 14 days on SLA-lood (HBC) and stained for ALP (red), Col-I (green) and nuclei (blue) before acquiring

man osteoblasts on Ti surfaces pre-treated with human whole blood. Acta

Fig. 5. Mineralization of HBCs on Ti surfaces after whole blood incubation. Primary human bone cells (HBCs) were cultivated for 7 and 14 days on SLA-Hphob and SLActive-HphilNS pre-incubated with (blood + HBC) or without whole human blood (HBC) and stained for mineral deposition (red) and nuclei (blue) before acquiring fluorescenceimages. Samples incubated with blood only served as negative control (scale bar 200 lm).

B.S. Kopf et al. / Acta Biomaterialia xxx (2015) xxx–xxx 7

further increase at the later time-point could be observed.Similarly, HBCs grown on SLActive-HphilNS and SLA-Hphob withoutblood pre-treatment displayed almost no mineralization both onday 14 and day 21. Also, no mineralization was observed whensamples were incubated for 14 and 21 days with whole humanblood only.

When quantifying the mineralization of HBCs from 5 differentdonors (I–V), cells grown on SLActive-HphilNS pre-incubated withblood showed much higher Ca2+ levels on day 14 and day 21 com-pared to SLA-Hphob pre-incubated with blood and compared toSLA-Hphob and SLActive-HphilNS without pre-incubation with blood(Fig. 6). The significant influence of blood pre-treatment ofSLActive-HphilNS surfaces was observed with HBCs of all donorsexcept for donor IV on day 14 and donor II on day 21, where theCa2+ level was increased but without statistical significance.Notably, the Ca2+ level of HBCs cultured on SLA-Hphob andSLActive-HphilNS without blood pre-incubation did not show a sig-nificant difference on both time-points when compared to theircorresponding surfaces that were pre-incubated with blood.Samples of HBCs from donor II cultured on SLActive-HphilNS with-out blood pre-incubation on day 14 could not be measured due to acontamination. Additionally, no Ca2+ was detected at all when SLA-Hphob and SLActive-HphilNS were pre-incubated with blood alone orincubated in cell culture medium alone and treated under other-wise identical conditions (Fig. S2). Representative images of theEDX analysis of samples on day 21 are shown in Fig. 7 and con-firmed the calcification of HBCs when cultured on SLActive-HphilNS pre-incubated with blood. Elemental analysis showed dis-tinct calcium-phosphate signals in crystal formations associatedwith HBCs on blood pre-incubated SLActive-HphilNS samples. Nocalcium-phosphate was detectable of HBCs cultured on SLA-Hphob

pre-incubated with blood. Also no calcium-phosphate was mea-sured when SLA-Hphob and SLActive-HphilNS were pre-incubatedwith blood alone for 21 days.

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of huBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

4. Discussion

An important part of the cellular microenvironment is the ECM,which, with its major proteins fibronectin, fibrinogen, vitronectin,laminin and Col-I, plays a decisive role in regulation of differentia-tion and development [49,50]. Fibronectin and vitronectin arepresent not only in the ECM but also in the blood serum and areknown to promote cell adhesion and spreading [51,52].Interestingly, it was shown that fibronectin and fibrinogen adsorbin higher quantities on hydrophilic than on hydrophobic surfacesand that the pre-adsorbed proteins lead to an increased initialattachment of osteoblastic cells [53,54].

Many studies have been performed using fractionated blood,such as platelet enriched plasma or serum, to study the blood/im-plant interface. However, these studies do not reflect thephysiological situation of a material implantation as they do nottake into account the complex composition of whole blood andthe possible interplay between different blood components suchas erythrocytes and platelets [18,55]. Using human whole bloodpre-incubations, this study therefore examined the blood clot for-mation on two otherwise well-defined surfaces, the hydrophobicSLA-Hphob and hydrophilic SLActive-HphilNS with nanostructuresand how this affects the differentiation potential of primary humanbone cells.

Looking at the blood coagulation on SLActive-HphilNS and SLA-Hphob, a three-dimensional fibrin network of several layers and sin-gle fibres among scattered compact patches of fibrin were found onthe two surfaces, respectively. Such enhanced blood-coagulationand fibrin scaffold deposition on super-hydrophilic surfaces hasbeen also reported previously [39,40]. While different structuralfibrin arrangements could be seen on the two surfaces, similaroverall fluorescence intensity of the Z-stack projections wereobserved, indicating that the amount of fibrin is possiblycomparable on both types of surfaces.

man osteoblasts on Ti surfaces pre-treated with human whole blood. Acta

Fig. 6. Quantification of mineralization. Primary human bone cells (HBCs) of fivedifferent donors (I–V) were cultivated for (A) 14 and (B) 21 days on SLA-Hphob andSLActive-HphilNS pre-incubated with (blood + HBC) or without whole human blood(HBC) and mineral deposition was quantified using Quanti Chrom™ Calciummineralization assay. Data is presented as mean ± SD; n = 3–8 technical repeats.Significance was defined as p < 0.05; ⁄significant to SLA-Hphob (HBC + blood);#significant to SLActive-HphilNS (HBC); +significant to SLA-Hphob (HBC).

8 B.S. Kopf et al. / Acta Biomaterialia xxx (2015) xxx–xxx

Migration of osteogenic cells occurs first through a provisionalfibrin matrix and it has been proposed that this process is influ-enced by the release of soluble factors from cellular componentsof the blood clot [31]. In this study, pre-incubation with wholehuman blood enhanced the attachment of primary human bonecells on both surfaces, showing increased cell numbers after24 h (Fig. 2) as well as after 4 days (data not shown).Additionally, SEM analysis showed that more cell contacts ofthe filopodias with the fibrin scaffold were formed on SLActive-HphilNS compared to SLA-Hphob. This is in good agreement withprevious studies that have shown that hydrophilic surfacesadsorb higher quantities of fibrinogen and fibronectin thanhydrophobic surfaces, leading to an increased initial attachmentof osteoblasts [53,54,56]. Another factor positively influencingadhesion and differentiation of osteoblastic cells is a nanoscaletopography of surfaces [11,57,58]. Interestingly, after the surfacemodification which renders the surfaces hydrophilic, it wasreported that nanostructures spontaneously form on Ti SLActivesurface [45]. Therefore the significantly increased numbers of

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of huBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

attached cells on SLActive-HphilNS pre-incubated with blood,might be due to a combined effect of the increased fibrin scaffoldon the hydrophilic surface and the inherent nanoscale topographyof the Ti surface [46].

Osteogenic differentiation is a gradual process and expressionof specific osteoblastic marker proteins such as Col-I and ALP isup- or down-regulated over time. One of the most frequentlyused markers to demonstrate osteoblast differentiation is ALP,and although the exact role in osteogenesis has not been clearlyelucidated [59,60], has its expression been described as a charac-teristic feature of osteoblasts [61]. However, assignment of ALPexpression to specific phases appears difficult. On the one hand,ALP has been reported to be expressed in the early developmentof osteoblastic differentiation while later on, when other genessuch as osteocalcin are up regulated, ALP expression declines[59]. On the other hand, differentiation of MSCs to osteoblastson honeycomb collagen scaffolds showed enhanced ALP expres-sion starting on day 14 with further significant increase on day21 and day 28 [62]. Several studies investigating differentiationof MC3T3 osteoblast cells [63] or human MSC (hMSC) [64] cul-tured on rough Ti surfaces reported even constant ALP expressionlevels over 15 days [63] and over 28 days [64] without signifi-cantly different expression levels. In this study, ALP proteinexpression levels of primary human bone cells cultured on bothsurfaces without blood pre-incubation were very low on day 7and day 14. In contrast, pre-incubation with blood led to a strongALP signal already on day 7, which was further enhanced on day14. Interestingly, this enhanced ALP expression on surfaces withblood pre-treatment was independent of the surface wettability,suggesting that blood clot formation is the dominant factor posi-tively influencing osteoblast differentiation.

The most important marker associated with the formation ofinorganic matrix of bone is collagen [65]. A primary function ofosteoblasts is the synthesis of collagen [66], which is a majorcomponent of bone and contributes to its mechanical stability.Studies investigating mRNA expression levels of Col-I of MC3T3osteoblasts cultured on rough Ti surfaces reported the highestexpression levels after 10 days [9], whereas hMSCs displayed con-stant expression from day 7 to day 28 [64]. In the present study,an increase in Col-I protein expression could be observed on bothsurfaces upon pre-incubation with whole human blood whencompared to surfaces without blood contact. This indicates thatblood enhances the capability of differentiated primary humanbone cells in culture to synthesize Col-I. Interestingly, Col-Iexpression levels of primary human bone cells were higher whencultivated on SLActive-HphilNS than when cultivated on SLA-Hphob.This was especially prominent at day 7 and much less clear at day14, suggesting that wettability has an influence on Col-I expres-sion. The positive effect of blood components on osteogenesiswas previously demonstrated, showing that hMSCs seeded on 3-D fibrin scaffolds containing high fibrinogen complex concentra-tions undergo osteogenic differentiation [67]. The 3-D fibrin net-work, which was observed on SLActive-HphilNS, may therefore beresponsible for the up-regulation of the Col-I production and thuspromote osteogenic differentiation. However, when used as a sea-lant, fibrin has not been shown to have a direct effect on osteoin-duction [68]. On the other hand, many studies showed enhancedosteogenesis upon the use of platelet-rich fibrin (reviewed in[69]), suggesting that cytokines embedded in the fibrin matrixrather than the fibrin network itself promoted the osteogenic dif-ferentiation of HBCs on SLActive-HphilNS in this study.

The final phase of the osteoblast differentiation is themineralization, where mineral matrix containing mainly calciumphosphate in the form of hydroxyapatite, is secreted and depositedby mature osteoblasts [70]. In the present work, mineral deposi-tion could be observed starting on the second week of osteoblast

man osteoblasts on Ti surfaces pre-treated with human whole blood. Acta

Fig. 7. Characterization of mineralization. Primary human bone cells (HBCs) were cultivated for 21 days on SLA-Hphob and SLActive-HphilNS pre-incubated with (blood + HBC)or without whole human blood (HBC) before evaluation by SEM imaging and EDX analysis. (SEM images: scale bar 10 lm; EDX images scale bar 20 lm; arrows indicatemeasured spectrums for EDX).

B.S. Kopf et al. / Acta Biomaterialia xxx (2015) xxx–xxx 9

differentiation but only on cells cultured on SLActive-HphilNS pre-incubated with blood. No formation of minerals was observed onsamples incubated with blood only indicating that the observedmineralization is cell-associated. While simple stainings have pre-viously been shown to be insufficient to accurately demonstratemineralization [71], quantitative assessment of the mineralizationalso showed significantly increased amounts of calcium deposit-formation starting on day 14 when cells where cultured onSLActive-HphilNS pre-incubated with blood. This was further con-firmed by SEM-EDX, which revealed that only cells on SLActive-HphilNS surfaces pre-incubated with blood exhibit accumulatedgranular materials containing calcium phosphate. Importantly,the results presented in this study correlate well with previousin vivo studies reporting enhanced bone apposition on the hydro-philic SLActive-HphilNS 2 weeks after implantation when comparedto the performance of the hydrophobic SLA-Hphob surface [8,9].

5. Conclusions

This study could show that whole human blood coagulation isinfluenced by the wettability of Ti implant surfaces and that pre-incubation of Ti surfaces with whole human blood leads to anincreased number of attached cells, enhanced ALP and Col-I proteinexpression and strong mineralization. Overall, more primaryhuman osteoblasts attached to the hydrophilic surface, displayingstrong interactions with the fibrin scaffold and also a faster andenhanced osteogenic differentiation. This early and enhancedmineralization on the hydrophilic surface pre-incubated withblood also correlates well with previously published in vivo stud-ies. Therefore our results suggest that blood pre-incubation ofimplant surfaces mimics a more physiological situation, therebyproviding a more predictive in vitro model that can be employedfor the evaluation of novel implant surfaces.

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of huBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

Disclosures

The authors declare no conflict of interest. Simon Berner is cur-rently an employee of Institut Straumann AG and he participated inthe study as contributing scientist. The study was funded by CTI(Grant No. 13747.1). Institut Straumann AG (Basel, Switzerland)solely provided sample discs for the study and had no role in studydesign.

Acknowledgments

We thank the Swiss Commission for Technology and InnovationCTI (Grant No: 13747.1) for financial support. We also like to thankUrsina Tobler for her constant support in the laboratory work dur-ing the study.

Appendix A. Figures with essential color discrimination

Certain figures in this article, particularly Figs. 1–5 and 7 are dif-ficult to interpret in black and white. The full color images can befound in the on-line version, at http://dx.doi.org/10.1016/j.actbio.2015.03.022.

Appendix B. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.actbio.2015.03.022.

References

[1] Gittens RA, Olivares-Navarrete R, Cheng A, Anderson DM, McLachlan T,Stephan I, et al. The roles of titanium surface micro/nanotopography and

man osteoblasts on Ti surfaces pre-treated with human whole blood. Acta

10 B.S. Kopf et al. / Acta Biomaterialia xxx (2015) xxx–xxx

wettability on the differential response of human osteoblast lineage cells. ActaBiomater 2013;9(4):6268–77.

[2] Olivares-Navarrete R, Hyzy SL, Hutton DL, Erdman CP, Wieland M, Boyan BD,et al. Direct and indirect effects of microstructured titanium substrates on theinduction of mesenchymal stem cell differentiation towards the osteoblastlineage. Biomaterials 2010;31(10):2728–35.

[3] Park JH, Olivares-Navarrete R, Wasilewski CE, Boyan BD, Tannenbaum R,Schwartz Z. Use of polyelectrolyte thin films to modulate Osteoblast responseto microstructured titanium surfaces. Biomaterials 2012;33(21):5267–77.

[4] Zhao G, Raines AL, Wieland M, Schwartz Z, Boyan BD. Requirement for bothmicron- and submicron scale structure for synergistic responses of osteoblaststo substrate surface energy and topography. Biomaterials2007;28(18):2821–9.

[5] Lohmann CH, Bonewald LF, Sisk MA, Sylvia VL, Cochran DL, Dean DD, et al.Maturation state determines the response of osteogenic cells to surfaceroughness and 1,25-dihydroxyvitamin D3. J Bone Miner Res2000;15(6):1169–80.

[6] Martin JY, Dean DD, Cochran DL, Simpson J, Boyan BD, Schwartz Z.Proliferation, differentiation, and protein synthesis of human osteoblast-likecells (MG63) cultured on previously used titanium surfaces. Clin Oral ImplantsRes 1996;7(1):27–37.

[7] Boyan BD, Bonewald LF, Paschalis EP, Lohmann CH, Rosser J, Cochran DL, et al.Osteoblast-mediated mineral deposition in culture is dependent on surfacemicrotopography. Calcif Tissue Int 2002;71(6):519–29.

[8] Buser D, Broggini N, Wieland M, Schenk RK, Denzer AJ, Cochran DL, et al.Enhanced bone apposition to a chemically modified SLA titanium surface. JDent Res 2004;83(7):529–33.

[9] Lang NP, Salvi GE, Huynh-Ba G, Ivanovski S, Donos N, Bosshardt DD. Earlyosseointegration to hydrophilic and hydrophobic implant surfaces in humans.Clin Oral Implants Res 2011;22(4):349–56.

[10] Sawase T, Jimbo R, Baba K, Shibata Y, Ikeda T, Atsuta M. Photo-inducedhydrophilicity enhances initial cell behavior and early bone apposition. ClinOral Implants Res 2008;19(5):491–6.

[11] Wennerberg A, Jimbo R, Stübinger S, Obrecht M, Dard M, Berner S.Nanostructures and hydrophilicity influence osseointegration: abiomechanical study in the rabbit tibia. Clin Oral Implant Res2014;25(9):1041–50.

[12] Cai K, Lai M, Yang W, Hu R, Xin R, Liu Q, et al. Surface engineering of titaniumwith potassium hydroxide and its effects on the growth behavior ofmesenchymal stem cells. Acta Biomater 2010;6(6):2314–21.

[13] Dalby MJ, McCloy D, Robertson M, Agheli H, Sutherland D, Affrossman S, et al.Osteoprogenitor response to semi-ordered and random nanotopographies.Biomaterials 2006;27(15):2980–7.

[14] Boyan BD, Batzer R, Kieswetter K, Liu Y, Cochran DL, Szmuckler-Moncler S,et al. Titanium surface roughness alters responsiveness of MG63 osteoblast-like cells to 1a,25-(OH)2D3. J Biomed Mater Res 1998;39(1):77–85.

[15] Zinger O, Zhao G, Schwartz Z, Simpson J, Wieland M, Landolt D, et al.Differential regulation of osteoblasts by substrate microstructural features.Biomaterials 2005;26(14):1837–47.

[16] Olivares-Navarrete R, Raz P, Zhao G, Chen J, Wieland M, Cochran DL, et al.Integrin alpha2beta1 plays a critical role in osteoblast response to micron-scale surface structure and surface energy of titanium substrates. Proc NatlAcad Sci USA 2008;105(41):15767–72.

[17] Wang X, Schwartz Z, Gittens RA, Cheng A, Olivares-Navarrete R, Chen H, et al.Role of integrin alpha beta in mediating osteoblastic differentiation on three-dimensional titanium scaffolds with submicron-scale texture. J Biomed MaterRes A 2014. http://dx.doi.org/10.1002/jbm.a.35323.

[18] Park JY, Davies JE. Red blood cell and platelet interactions with titaniumimplant surfaces. Clin Oral Implants Res 2000;11(6):530–9.

[19] Shiu HT, Goss B, Lutton C, Crawford R, Xiao Y. Formation of blood clot onbiomaterial implants influences bone healing. Tissue Eng B, Rev2014;20(6):697–712.

[20] Keselowsky BG, Collard DM, Garcia AJ. Surface chemistry modulates focaladhesion composition and signaling through changes in integrin binding.Biomaterials 2004;25(28):5947–54.

[21] Nygren H, Tengvall P, Lundstrom I. The initial reactions of TiO2 with blood. JBiomed Mater Res 1997;34(4):487–92.

[22] Gittens RA, Olivares-Navarrete R, Hyzy SL, Sandhage KH, Schwartz Z, BoyanBD. Superposition of nanostructures on microrough titanium–aluminum–vanadium alloy surfaces results in an altered integrin expression profile inosteoblasts. Connect Tissue Res 2014;55(Suppl. 1):164–8.

[23] Deligianni DD, Katsala N, Ladas S, Sotiropoulou D, Amedee J, Missirlis YF.Effect of surface roughness of the titanium alloy Ti–6Al–4V on human bonemarrow cell response and on protein adsorption. Biomaterials 2001;22(11):1241–51.

[24] Rabe M, Verdes D, Seeger S. Understanding protein adsorption phenomena atsolid surfaces. Adv Colloid Interface Sci 2011;162(1–2):87–106.

[25] Vroman L. Problems in the development of materials that are compatible withblood. Biomater Med Dev Artif Organs 1984;12(3–4):307–23.

[26] Chapin JC, Hajjar KA. Fibrinolysis and the control of blood coagulation. BloodRev 2015;29(1):17–24.

[27] Markiewski MM, Nilsson B, Ekdahl KN, Mollnes TE, Lambris JD. Complementand coagulation: strangers or partners in crime? Trends Immunol2007;28(4):184–92.

[28] Mosesson MW. Fibrinogen and fibrin structure and functions. J ThrombHaemost 2005;3(8):1894–904.

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of huBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

[29] Laurens N, Koolwijk P, de Maat MP. Fibrin structure and wound healing. JThromb Haemost 2006;4(5):932–9.

[30] Schmoekel H, Schense JC, Weber FE, Gratz KW, Gnagi D, Muller R, et al. Bonehealing in the rat and dog with nonglycosylated BMP-2 demonstrating lowsolubility in fibrin matrices. J Orthop Res 2004;22(2):376–81.

[31] Davies JE. Mechanisms of enodosseous integration. Int J Prosthodont1988:391–401.

[32] Arvidsson S, Askendal A, Tengvall P. Blood plasma contact activation on silicon,titanium and aluminium. Biomaterials 2007;28(7):1346–54.

[33] Eriksson C, Lausmaa J, Nygren H. Interactions between human whole bloodand modified TiO2-surfaces: influence of surface topography and oxidethickness on leukocyte adhesion and activation. Biomaterials2001;22(14):1987–96.

[34] Hong J, Azens A, Ekdahl KN, Granqvist CG, Nilsson B. Material-specificthrombin generation following contact between metal surfaces and wholeblood. Biomaterials 2005;26(12):1397–403.

[35] Hong J, Kurt S, Thor A. A hydrophilic dental implant surface exhibitsthrombogenic properties in vitro. Clin Implant Dent Relat Res2013;15(1):105–12.

[36] Alfarsi MA, Hamlet SM, Ivanovski S. Titanium surface hydrophilicity enhancesplatelet activation. Dent Mater J 2014;33(6):749–56.

[37] Di Iorio D, Traini T, Degidi M, Caputi S, Neugebauer J, Piattelli A. Quantitativeevaluation of the fibrin clot extension on different implant surfaces: an in vitrostudy. J Biomed Mater Res B Appl Biomater 2005;74B(1):636–42.

[38] Kammerer PW, Gabriel M, Al-Nawas B, Scholz T, Kirchmaier CM, Klein MO.Early implant healing: promotion of platelet activation and cytokine release bytopographical, chemical and biomimetical titanium surface modificationsin vitro. Clin Oral Implants Res 2012;23(4):504–10.

[39] Hong J, Nilsson Ekdahl K, Reynolds H, Larsson R, Nilsson B. A new in vitromodel to study interaction between whole blood and biomaterials. Studies ofplatelet and coagulation activation and the effect of aspirin. Biomaterials1999;20(7):603–11.

[40] Milleret V, Tugulu S, Schlottig F, Hall H. Alkali treatment of microroughtitanium surfaces affects macrophage/monocyte adhesion, platelet activationand architecture of blood clot formation. Eur Cell Mater 2011;21:430–44[discussion 44].

[41] Traini T, Caputi S, Gherlone E, Degidi M, Piattelli A. Fibrin clot extension onzirconia surface for dental implants: a quantitative in vitro study. Clin ImplantDent Relat Res 2014;16(5):718–27.

[42] Nurden AT. Platelets, inflammation and tissue regeneration. Thromb Haemost2011;105(Suppl. 1):S13–33.

[43] Davies JE. Bone engineering. In: Davies JE, editor. Toronto: Em Squared Inc;2000.

[44] Schilephake H. Bone growth factors in maxillofacial skeletal reconstruction. IntJ Oral Maxillofac Surg 2002;31(5):469–84.

[45] Wennerberg A, Svanborg LM, Berner S, Andersson M. Spontaneously formednanostructures on titanium surfaces. Clin Oral Implants Res2013;24(2):203–9.

[46] Kopf BS, Ruch S, Berner S, Spencer ND, Maniura-Weber K. The role ofnanostructures and hydrophilicity in osseointegration: in-vitro protein-adsorption and blood-interaction studies. J Biomed Mater Res A 2015. http://dx.doi.org/10.1002/jbm.a.35401.

[47] Rupp F, Scheideler L, Olshanska N, de Wild M, Wieland M, Geis-Gerstorfer J.Enhancing surface free energy and hydrophilicity through chemicalmodification of microstructured titanium implant surfaces. J Biomed MaterRes A 2006;76(2):323–34.

[48] Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years ofimage analysis. Nat Methods 2012;9(7):671–5.

[49] Adams JC, Watt FM. Regulation of development and differentiation by theextracellular matrix. Development 1993;117(4):1183–98.

[50] Allori AC, Sailon AM, Warren SM. Biological basis of bone formation,remodeling, and repair-part II: extracellular matrix. Tissue Eng B Rev2008;14(3):275–83.

[51] Wittmer CR, Phelps JA, Saltzman WM, Van Tassel PR. Fibronectin terminatedmultilayer films: protein adsorption and cell attachment studies. Biomaterials2007;28(5):851–60.

[52] Underwood PA, Bennett FA. A comparison of the biological activities of thecell-adhesive proteins vitronectin and fibronectin. J Cell Sci 1989;93(Pt4):641–9.

[53] Sousa SR, Lamghari M, Sampaio P, Moradas-Ferreira P, Barbosa MA. Osteoblastadhesion and morphology on TiO2 depends on the competitive preadsorptionof albumin and fibronectin. J Biomed Mater Res A 2008;84(2):281–90.

[54] Wei J, Igarashi T, Okumori N, Igarashi T, Maetani T, Liu B, et al. Influence ofsurface wettability on competitive protein adsorption and initial attachmentof osteoblasts. Biomed Mater 2009;4(4):045002.

[55] Thor A, Rasmusson L, Wennerberg A, Thomsen P, Hirsch JM, Nilsson B, et al.The role of whole blood in thrombin generation in contact with varioustitanium surfaces. Biomaterials 2007;28(6):966–74.

[56] Scotchford CA, Gilmore CP, Cooper E, Leggett GJ, Downes S. Protein adsorptionand human osteoblast-like cell attachment and growth on alkylthiol on goldself-assembled monolayers. J Biomed Mater Res 2002;59(1):84–99.

[57] Dalby MJ, Gadegaard N, Tare R, Andar A, Riehle MO, Herzyk P, et al. The controlof human mesenchymal cell differentiation using nanoscale symmetry anddisorder. Nat Mater 2007;6(12):997–1003.

[58] Gittens RA, McLachlan T, Olivares-Navarrete R, Cai Y, Berner S, Tannenbaum R,et al. The effects of combined micron-/submicron-scale surface roughness and

man osteoblasts on Ti surfaces pre-treated with human whole blood. Acta

B.S. Kopf et al. / Acta Biomaterialia xxx (2015) xxx–xxx 11

nanoscale features on cell proliferation and differentiation. Biomaterials2011;32(13):3395–403.

[59] Golub EE, Boesze-Battaglia K. The role of alkaline phosphatase inmineralization. Curr Opin Orthop 2007;18(5):444–8.

[60] Siffert RS. The role of alkaline phosphatase in osteogenesis. J Exp Med1951;93(5):415–26.

[61] Rosa AL, Beloti MM, van Noort R. Osteoblastic differentiation of cultured ratbone marrow cells on hydroxyapatite with different surface topography. DentMater 2003;19(8):768–72.

[62] George J, Kuboki Y, Miyata T. Differentiation of mesenchymal stem cells intoosteoblasts on honeycomb collagen scaffolds. Biotechnol Bioeng2006;95(3):404–11.

[63] Sista S, Wen C, Hodgson PD, Pande G. Expression of cell adhesion anddifferentiation related genes in MC3T3 osteoblasts plated on titanium alloys:role of surface properties. Mater Sci Eng, C 2013;33(3):1573–82.

[64] Mendonça DBS, Miguez PA, Mendonça G, Yamauchi M, Aragão FJL, Cooper LF.Titanium surface topography affects collagen biosynthesis of adherent cells.Bone 2011;49(3):463–72.

Please cite this article in press as: Kopf BS et al. Enhanced differentiation of huBiomater (2015), http://dx.doi.org/10.1016/j.actbio.2015.03.022

[65] Zhu W, Robey PG, Boskey AL. The regulatory role of matrix proteins inmineralization of bone. Fundamentals of osteoporosis. San Diego: ElsevierAcademic Press; 2009.

[66] Koshihara Y, Honda Y. Age-related increase in collagen production in culturedhuman osteoblast-like periosteal cells. Mech Ageing Dev 1994;74(1–2):89–101.

[67] Catelas I, Sese N, Wu BM, Dunn JC, Helgerson S, Tawil B. Human mesenchymalstem cell proliferation and osteogenic differentiation in fibrin gels in vitro.Tissue Eng 2006;12(8):2385–96.

[68] Schwarz N. The role of fibrin sealant in osteoinduction. Ann Chir GynaecolSuppl 1993;207:63–8.

[69] Kumar RV, Shubhashini N. Platelet rich fibrin: a new paradigm in periodontalregeneration. Cell Tissue Banking 2013;14(3):453–63.

[70] Clarke B. Normal bone anatomy and physiology. Clin J Am Soc Nephrol2008;3(Suppl. 3):S131–9.

[71] Bonewald LF, Harris SE, Rosser J, Dallas MR, Dallas SL, Camacho NP, et al. VonKossa staining alone is not sufficient to confirm that mineralization in vitrorepresents bone formation. Calcif Tissue Int 2003;72(5):537–47.

man osteoblasts on Ti surfaces pre-treated with human whole blood. Acta