In vitro biological effects of titanium rough surface obtained by calcium phosphate grid blasting

Anthony Citeau ^a, Jerome Guicheux ^{, , a}, Claire Vinatier ^a, Pierre Layrolle ^a, Thien P Nguyen ^b, Paul Pilet ^c and Guy Daculsi <sup>a

a</sup> INSERM EM 9903, Research Centre on Materials with Biological Interest, School of dental surgery, Nantes University, 1 place Alexis Ricordeau, Nantes Cedex 1, 44042, France
^b LPC, Institute of Materials Jean Rouxel, UMR CNRS 6502, 2 rue de la Houssinière, Nantes Cedex 3, 44322, France
^c Department of imagery and microcharacterization, IFR 26, University Hospital, 1 place Alexis Ricordeau, Nantes Cedex 1, 44042, France

Received 8 October 2003;  accepted 9 February 2004.  Available online 16 March 2004.
Biomaterials
Article in Press, Corrected Proof - Note to users

1. Abstract

Surface roughness modulates the osseointegration of orthopaedic and dental titanium implants. High surface roughness are currently obtained by blasting of titanium implants with silica or aluminium oxide abrasive particles. This process may cause the release of cytotoxic silicium or aluminium ions in the peri-implant tissue. To generate a biocompatible roughened titanium surface, we currently develop an innovative grid-blasting process using biphasic calcium phosphate (BCP) particles. Titanium alloy (Ti6Al4 V) discs were either polished, BCP grid-blasted or left as-machined. BCP grid-blasting created an average surface roughness of 1.57±0.07 
m compared to the original machined surface of 0.58±0.05 
m. X-ray photoelectron spectroscopy indicated traces of calcium and phosphorus and relatively less aluminium on the BCP grid-blasted surface than on the initial titanium specimen. Scanning electronic microscopy observations and measurement of mitochondrial activity (MTS assay) showed that osteoblastic MC3T3-E1 cells were viable in contact with the BCP grid-blasted titanium surface. In addition, our results indicate that MC3T3-E1 cells expressed ALP activity and conserved their responsiveness to bone morphogenetic protein BMP-2. The overall results clearly indicate that this calcium phosphate grid-blasting technique increases the roughness of titanium implants and provides a non-cytotoxic surface with regard to mouse osteoblasts.

Author Keywords: Author Keywords: Titanium; Surface roughness; Osteoblast; Calcium phosphate; Biocompatibility

2. Article Outline

1. Introduction

2. Materials and methods

2.1. Materials

2.2. Preparation of titanium discs

2.3. Surface roughness analysis

2.4. X-ray photoelectron spectroscopy

2.5. Cell culture

2.6. Cell morphology

2.7. Cell viability

2.8. Alkaline phosphatase (ALP) activity

2.9. Statistical analysis

3. Results

3.1. Surface roughness

3.2. XPS analysis

3.3. Cell morphology

3.4. Cell viability

3.5. ALP activity

4. Discussion

Acknowledgements

References

3. 1. Introduction

With regards to their biological and physicochemical properties, titanium-based biomaterials have been successfully used in orthopaedic, dental and maxillo-facial surgery mainly as endosseous implants. Clinical success has been achieved not only because of mechanical strength [1] or excellent biocompatibility [2] of titanium alloys but also because of other characteristics such as surface properties. Among them, surface roughness is one of the most acknowledged properties improving the bone anchorage of titanium implants [3, 4 and 5].

One way to obtain a roughened surface is to blast the implants with Al₂O₃ and SiC particles. However, the use of such abrasive particles leads to a surface contamination which may induce a local toxic or inflammatory reaction by dissolution of Al₂O₃ ions into host bone [6]. This local release of remnants from blasting materials has been suggested to impair bone mineralization and repair through a competition between aluminium and calcium ions [7].

In parallel, a large body of evidence indicate that surface roughness of titanium implants strongly affects the behaviour of bone cells. On the one hand, numerous in vitro studies have demonstrated that increased surface roughness enhances cell attachment, proliferation and differentiation of osteogenic cells [8, 9 and 10]. On the other hand, several reports indicate that increased surface roughness not only fails to stimulate but alters cell function and bone formation [11, 12 and 13].

Calcium phosphate biomaterials, in particular hydroxyapatite (HA) and
-Tricalcium phosphate (
-TCP), has been contemplated as a potential abrasive materials. Indeed HA and
-TCP are biocompatible and well known to form a direct bond with surrounding tissue after bone implantation. In addition, it has been clinically evidenced that plasma-sprayed HA coating greatly enhances the early osseointegration and long term success rate of titanium prosthesis [14 and 15]. Viewing these biological properties, we questioned whether a recently developed calcium phosphate biomaterials [16], namely biphasic calcium phosphate (BCP) consisting of a mixture of HA and
-TCP, could be used as abrasive particles to prepare biocompatible roughened titanium surface.

In view of the above-mentioned data, the main objectives of this work were (i) to develop a new technique of blasting using BCP particles to provide a roughened surface which does not release cytotoxic elements and (ii) to assess the effects of such a roughened surface with respect to osteoblasts.

In this attempt, we examined the roughness of Ti6Al4 V discs prepared by our innovative BCP grid-blasting technique. Then, the cytocompatibility of such titanium implants was studied by examining the morphology, viability, alkaline phosphatase activity and BMP-2 responsiveness of osteoblastic MC3T3-E1 cells cultured in contact with BCP grid-blasted titanium discs.

4. 2. Materials and methods

2.1. Materials

Commercially Titanium alloy (Ti6Al4 V) bars were machined in discs of 14 mm in diameter and 2 mm in thickness (Biomatlante, Vigneux de Bretagne, France). Paper Silicon Carbide and diamond suspension were obtained from Buehler (Limonest, France). Abrasive biphasic calcium phosphate (BCP) particles (40-80 mesh), containing hydroxyapatite (HA) and
-tricalcium phosphate (
-TCP) in a weight ratio of 75/25 were obtained from Biomatlante. The average size of BCP particles was 40.8 
m as determined by scanning electron microscopy (SEM, JEOL 6300, Tokyo, Japan) and image analysis (Quantimet 500MC, Leica, Cambridge, Great Britain). Cell culture plastic ware was purchased from Falcon (Becton-Dickinson, Franklin Lakes, NJ, USA). Foetal calf serum (FCS),
-MEM,
-Glutamine, penicillin/streptomycin, trypsin/EDTA and phosphate buffered salt (PBS) were purchased from Invitrogen Corporation (Paisley, UK). Nonidet P-40 was obtained from Amersham Biosciences (Orsay, France). P-nitrophenylphosphate (p-NPP), 2-amino-2-methyl-1-propanol and MgCl₂ were purchased from Sigma (St. Louis, MO, USA). 3-(4,5-Dimethylthiazol-2-gl)-5-(3-carboxymethoxylphenyl)-2-(4-sulphophenyl-2 H) tetrazolium inner salt (MTS) was from Promega (Madison, WI, USA). Recombinant human BMP-2 was generously provided by Genetics Institute (Cambridge, Massachusetts, USA). All other chemicals were from standard laboratory suppliers and were of the highest purity available.

2.2. Preparation of titanium discs

Discs of titanium alloy were cleaned on their two faces with acetone. Discs were thereafter randomly divided into three groups.

A mirror polished titanium surface (Tipolish group) was obtained by polishing with silicon carbide paper. Decreased grind paper P320, P1200 and P4000 were successively used on a rotative polisher at 125 rotations per minute (rpm) for 30 s. Discs were then hand-polished for 1 min with a suspension of diamond particles (1 
m in diameter) at 250 rpm.

The BCP grid-blasted surface (Tiblast group), was prepared by blasting BCP particles of 40-80 mesh with a microhardness higher than 510HV and a density of 1.2 g/cc onto both faces of the discs (Jetstream, International Guyson, Skipton, UK). The BCP particles were blasted at a distance of 10-15 cm of titanium discs at air pressure of 8 bars for 5 min. BCP grid-blasted discs were then passivated by incubation for 1 h in 26% nitric acid at room temperature and rinsed with deionised water according to ASTM F86-01.

The Tipassiv group, was obtained by passivating the left as-machined titanium discs for 1 h in 26% nitric acid at room temperature.

After drying, all discs were double packaged and sterilized by steam for 30 min at 121°C.

2.3. Surface roughness analysis

Six titanium discs of each group were randomly selected and roughness measurements were carried out with a digital profilometer (Surtronic 3+, Taylor Hobson SA, Courtaboeuf, France) linked to an acquisition software TalyProfile. This profilometer system consists of a fine needle which scanned the surface for 10 mm in length. Results were expressed as average roughness (Ra) which is the arithmetic mean of the height variation on the roughness profiles, as well as RSm which is the mean spacing of surface peaks. These parameters, Ra and RSm expressed in
m, gave a general description of surface roughness [13]. Titanium surface was also examined using scanning electron microscopy SEM (JEOL 6300, Tokyo, Japan) at 15 kV.

2.4. X-ray photoelectron spectroscopy

XPS analysis were performed with a Leybold ESCA LH 12 spectrometer (University of Nantes—CNRS) at a low pressure of 10⁻⁹ mbar, using MgK
radiation (h
=1256.4 eV). The power of the X-ray source was 120 W (12 kV and 10 mA). The energy pass was set at 25 eV for all the experiments. Absolute binding energy (BE) was calibrated using the Au 4f line at 84.4 eV from a gold dot deposited on the surface of the samples.

2.5. Cell culture

MC3T3-E1 cells is a non-transformed cell line established from newborn mouse calvaria. These cells exhibit an osteoblastic phenotype as evidenced by the expression of ALP activity [17], the synthesis of extracellular matrix (ECM) components such as osteocalcin and type-1 collagen [18] and their ability to mineralise the ECM. MC3T3-E1 cells were routinely grown in alpha MEM medium supplemented with 10% FCS, 1% penicillin/streptomycin and 1%
-Glutamine. Cells were subcultured once a week using Trypsin/EDTA and maintained at 37°C in a humidified atmosphere of 5% CO₂ in air. Medium was completely renewed every two days.

2.6. Cell morphology

Cells were seeded onto materials in 24-multiwell plates at a final density of 10,000 cells/cm². After indicated times, media were removed and specimens were fixed with 4% glutaraldehyde in PBS (pH 7.2) for 1 h 30 min at 4°C. After dehydratation in graded alcohols, specimens were treated with graded mixture of ethanol/trichlorotrifluoroethane (75/25, 50/50, 25/75 and 0/100). They were then sputter-coated with gold-palladium. The surface of the specimens was finally examined with backscattered (BSE) and secondary electrons (SE) in scanning electron microscopy (SEM) under a 15 kV voltage [19].

2.7. Cell viability

Cell viability was measured as mitochondrial NADH/NADPH-dependent dehydrogenase activity, resulting in the cellular conversion of the tetrazolium salt MTS into a soluble formazan dye [20 and 21] with the CellTiter 96 Aqueous non-radioactive cell proliferation assay (Promega). Briefly, MC3T3-E1 cells were cultured either onto the various titanium discs or in the absence of material (Plastic) in 24 multiwell plates at a density of 10,000 cells/cm². After indicated times, culture media were removed and 100 
l of MTS solution was added in each well for 2-3 h according to the manufacturer's instructions. Finally, colorimetric measurement of formazan dye was performed on a spectrophotometer with an optical density reading at 490 nm. Results were expressed as relative MTS activity as compared to control conditions (cells cultured in the absence of materials).

2.8. Alkaline phosphatase (ALP) activity

ALP activity was evaluated, as previously described [22], in MC3T3-E1 cells cultured either onto the various titanium discs or in the absence of material (Plastic) in 24 multiwell plates (10,000 cells/cm²). After indicated times, cells were washed twice with ice-cold PBS and scraped in a 0.2% aqueous solution of Nonidet P-40. Cell suspension was sonicated on ice for 30 s and centrifugated for 5 min at 4°C. Aliquots of supernatants were subjected to protein assay with the Pierce Coomassie Plus assay reagent (Pierce, Rockford, USA) and to ALP activity measurement. ALP activity was assessed at pH 10.3 in 0.1 
2-amino-2-methyl-1-propanol containing 1 m
MgCl₂. P-NPP (10 m
) was used as a chromogenic substrate for an optical density reading at 405 nm. Results were expressed as relative ALP activity compared with control conditions (cells cultured in the absence of titanium discs).

Finally to test the osteoblastic responsiveness to BMP-2, one of the most potent factor stimulating ALP activity [22], MC3T3-E1 cells were incubated in low-serum (1%) medium for 24 h prior to stimulation with BMP-2. BMP-2 (100 ng/ml) or its vehicle (PBS containing 0.5% bovine serum albumin) was then added for an additional 48 h before ALP determination as described above.

2.9. Statistical analysis

Each experiment was repeated at least twice with similar results. Results are expressed as mean± SEM of triplicate determinations. Comparative studies of means were performed using one-way ANOVA followed by a post hoc test (Fisher projected least significant difference) with a statistical significance at p < 0.05.

5. 3. Results

3.1. Surface roughness

SEM indicates a clear discrepancy between each of the three groups. Polished titanium samples (Fig. 1A) display a mirror polished surface. On the contrary, both Tipassiv (Fig. 1B) and Tiblast ( Fig. 1C) samples exhibited a highly rugged and irregular surface.

(73K)

Fig. 1. SEM micrographs showing the surface roughness of the various tested titanium discs. (A) mirror-polished titanium discs (Tipolish); (B) as-machined and passivated titanium discs (Tipassiv); (C) BCP grid-blasted titanium discs (Tiblast). Original magnification ×1000; Bar = 10 
m.

The roughness profiles are presented in Fig. 2, and the values of Ra and RSm are reported in Table 1. The polishing process (Tipolish) was found to induce a significant decrease (p<0.01) in surface roughness and a significant increase (p<0.05) in RSm. The Tiblast discs, revealed a significant increase of the average roughness (p<0.005), while RSm was significantly decreased by about two-fold.

(19K)

Fig. 2. Roughness profiles of the various tested titanium discs. (A) mirror-polished titanium discs (Tipolish); (B) as-machined and passivated titanium discs (Tipassiv); (C) BCP grid-blasted titanium discs (Tiblast). Representative profiles, obtained with a digital profilometer, are shown.

Table 1. Quantitative results of roughness measurements: Ra, average roughness; RSm, mean spacing of surface peaks

3.2. XPS analysis

Fig. 3 shows the XPS spectra of as-received (Tipassiv) and BCP grid-blasted (Tiblast) titanium specimen. In the standard sample (Tipassiv), besides the titanium and oxygen lines, we noted the presence of carbon contamination on the surface of the film (Fig. 3a). In addition, a small amount of aluminium oxide (Al 2p peak at ~75 eV) was also detected on the surface of the sample. In the BCP grid-blasted sample (Tiblast), the concentrations of both titanium and carbon were reduced, probably resulting from the blasting procedure. We observed the onset of a new peak at around 345 eV which is assigned to calcium-bonding (Fig. 3b). Traces of phosphorus were also found at ~136 eV.The blasting process seems to favour the incorporation of BCP on the surface of the titanium oxide and the elimination of contamination impurities.

(4K)

Fig. 3. XPS survey spectra of (a) machined and passivated titanium (Tipassiv) and (b) BCP grid-blasted titanium (Tiblast).

3.3. Cell morphology

Cell morphology has been examined by SEM and results are illustrated in Fig. 4A-F. On tipolish samples (Fig. 4A and B) cells exhibited large spreading with numerous dorsal microvili and numerous specula-like lamelipodes. On Tipassiv samples ( Fig. 4C and D) MC3T3-E1 had an almost spread-out aspect, with multiple peripheral filopodia, dorsal microvili and cytoplasmic extensions. On Tiblast samples, MC3T3-E1 cells had a round shape, displayed dorsal microvili but exhibited few cytoplasmic extensions ( Fig. 4E and F). Cellular interconnections were frequently observed whatever the surface treatment ( Fig. 4 B, D and F). Cellular distribution appears to be obvious only on the polished surface ( Fig. 4A) with the appearance of multicellular spindles. Cells showed particular distribution neither on the surface of Tipassiv ( Fig. 4C) nor Tiblast samples ( Fig. 4E).

(85K)

Fig. 4. SEM observation of MC3T3-E1 morphology after a two days culture in direct contact on titanium discs with different surface roughness. (A) and (B) Group Tipolish , (C) and (D) Group Tipassiv, (E) and (F) Group Tiblast. A, C, E: samples were observed using back-scattered electrons. Original magnification ×250, BAR=100 
m. B, D, F: samples were observed using secondary electrons. Original magnification ×3000; BAR=10 
m.

3.4. Cell viability

To examine the viability of osteoblasts in contact with titanium discs of the various groups, we carried out a measurement of MTS activity on MC3T3-E1 after 4, 8 and 15 days in culture.

Results indicated that cell viability was decreased in contact with titanium discs at day 4, whatever the surface treatment. MTS activity in treated groups was significantly reduced by about 30, 33 and 50% (p<0.001), respectively for Tipolish, Tipassiv and Tiblast as compared to plastic conditions (Fig. 5A). After 8 days, MC3T3-E1 viability did not show a significant difference between control conditions, Tipassiv and Tipolish. Conversely, a significant reduction in MTS activity of nearly 30% (p<0.01) was still observed when Tiblast was compared to control (Fig. 5B). After 15 days in culture, we did not find significant differences in the cellular viability of osteoblasts cultured on the various surface tested ( Fig. 5C).

(18K)

Fig. 5. MTS activity of MC3T3-E1 cells cultured in direct contact on titanium discs with different surface roughness for 4 days (A), 8 days (B) or 15 days (C). MTS activity was determined as indicated in materials and methods. Results are expressed in relative MTS activity compared with control conditions (plastic). ^**p<0.0001 and ^*p<0.01 as compared to plastic.

These results show that osteoblast viability decreased as early as 4 days in culture on titanium discs of the three groups as compared to the plastic. This reduction was almost completely recovered as early as 8 days and totally abolished after 15 days in culture.

3.5. ALP activity

To check the expression of one of the osteoblastic differentiation markers, ALP activity was determined in osteoblasts cultured for 15 days in contact with the various titanium surfaces. On Tipolish and Tipassiv samples, we did not observe a significant modification of ALP activity as compared to plastic. On the contrary, a 15-day culture of osteoblasts on Tiblast discs triggered a slight reduction of 37±11% (p<0.01) of ALP activity compared with plastic (Fig. 6).

(4K)

Fig. 6. ALP activity of MC3T3-E1 cells after a 15-day culture in direct contact on titanium discs with different surface roughness. ALP activity was measured as described in materials and methods. Results are expressed in relative ALP activity compared with control conditions (plastic). ^*p<0.01 as compared to plastic.

These results indicate that osteoblasts cultured in direct contact with titanium discs of different surface roughness maintained their capability to express ALP, while a BCP grid-blasted one seems to slightly reduce this parameter.

To further document the osteoblast behaviour, we questioned whether MC3T3-E1 cells exhibit a differential responsiveness to BMP-2, with respect to the various materials. As previously described [22], BMP-2 induced a significant 8-fold increase in ALP activity when MC3T3-E1 cells were cultured on plastic. In addition, BMP-2 was found to significantly (p<0.005) stimulate ALP activity in MC3T3-E1 cells after a 15-day culture on Tipolish, Tipassiv as well as Tiblast nearly of 5-, 6- and 3-fold, respectively (Fig. 7). These results indicate that MC3T3-E1 cells cultured for 15 days on the various titanium surfaces maintained a significant responsiveness to BMP-2, even if a BCP grid blasted one seems to reduce BMP-2 induced ALP activity.

(4K)

Fig. 7. BMP-2-induced ALP activity in MC3T3-E1 cells after a 15-day culture in direct contact on titanium discs with different surface roughness. Cells were treated with BMP-2 (100 ng/ml) or its vehicle (control) 48 h prior to ALP activity determination. ALP activity was determined as precised in materials and methods. Results are expressed in relative ALP activity compared with respective control conditions (without BMP-2 stimulation). ^*p<0.0001 and ^*p<0.001 compared with respective control.

6. 4. Discussion

Surface roughness is one of the key factors for the osseous integration of orthopaedic and dental titanium implants. It improves the mechanical anchoring of calcium phosphate coating that acts as a support to colonization and attachment of osteogenic cells, thereby providing a scaffold for bone tissue formation [23]. Surface roughness is currently obtained by grid-blasting using silica or aluminium abrasive particles. This grid-blasting procedure has been proven to cause a contamination of implant surface which can have unfavourable effects on its bone integration [6]. With regards to this drawback, we aimed at developing an innovative strategy for sandblasting titanium surface based on the use of a specific abrasive calcium phosphate powder, a recognized biocompatible and osteoconductive material. It should further prevent the surface contamination by cytotoxic elements. To our knowledge, this study is the first report demonstrating that BCP grid-blasting process makes possible the increase of roughness of titanium surface while preserving its biocompatibility with regard to osteoblasts. In addition, this increased-roughened surface allowed the maintenance of ALP expression as well as BMP-2 responsiveness, two hallmarks of osteoblastic phenotype.

Three different surface roughnesses were produced on titanium by using mechanical polishing (Tipolish), nitric acid passivation (Tipassiv), and BCP grid-blasting (Tiblast). The three microstructures are well characterized by the two roughness parameters Ra and RSm that correspond to average peak to valley distance and average distance between peak to peak, respectively. The average roughness (1.5 
m) obtained with BCP grid-blasting appears to be comparable to those obtained by aluminium oxide or silica grid-blasting process [13 and 24]. Previous reports have clearly demonstrated that primary bone anchorage of titanium implants is markedly improved by surface roughness with Ra ranging from 0.5 to 1.5 
m [25 and 26]. In the light of these data, our roughened titanium surface seems to exhibit a satisfactory Ra. In addition, we could plan to improve the surface roughness by acting on the particular size of the BCP powder as well as on the projection force.

In this work, we were also interested in examining the in vitro cytocompatibility of titanium surface treated with our method. According to standards (ISO 10993-5: Biological evaluation of medical devices—Part 5: Tests for in vitro cytotoxicity) we focused on a cytotoxicity (morphology and cell viability) test by direct contact with osteoblasts as target cells for any medical device intended for implantation in bony site. Among the cellular models mimicking the various stages of osteoblastic differentiation in a temporally regulated manner, the clonal osteogenic cell line MC3T3-E1 is one of the most widely used since it is a non-transformed cell line that expresses a high alkaline phosphatase activity in the confluent state and mineralizes the extracellular matrix in vitro [18].

SEM analyses of MC3T3-E1 morphology indicate that cells cultured on mirror polished surface (Tipolish) were largely spread out on the surface of materials and exhibited large cytoplasmic extensions suggesting an active cell migration. On the contrary, osteoblasts expanded on rugged and irregular surface (Tipassiv and Tiblast) exhibit a round shape with large pseudopodia thus strongly suggesting quite adhesion capabilities. The presence of multiple dorsal microvilli, in particular with cells cultured on mirror polished surface, indicated a great cell activity. Taken together, these results indicate that cells seem to be more adherent on the BCP grid-blasted surface while conserving their biological activity. Since surface of titanium implants must allow an optimal cell and tissue adhesion to induce a satisfactory early and late bone anchorage [24], it seems reasonable to assume that BCP grid-blasted process is a promising technology. In addition, the molecular mechanisms underlying signalling that is induced by cell-substrate (ECM or materials) interactions have been recently deciphered. The molecules of focal adhesion contacts have been shown to interact with transcription factors thereby regulating the expression of specific target genes [27]. Viewing these data, it could be helpful to clearly analyse the cell adhesion process, to get further insights in the role of surface roughness in cell migration, proliferation or gene expression.

In the present investigation, cell viability was investigated through the evaluation of mitochondrial deshydrogenase activity (MTS activity). Our results have shown that BCP grid blasted discs induce an early decrease in cell viability that is totally recovered in a long-term culture. One of the explanations to this transient alteration of cell viability could be that osteoblasts are sensitive not only to the implant surface roughness but also to its chemical composition. Aluminium-containing titanium alloys such as Ti6Al4 V, exhibit an oxidized layer on their surfaces containing aluminium oxides [12]. The formation of this layer, which is due to a chemical reaction between oxygen and metal is facilitated by grid blasting [28]. Since it was previously reported that aluminium oxides may exert some cytotoxic effects [7], it will be necessary to investigate whether our BCP grid blasted surface may release some cytotoxic derivatives at least in the early phase of culture. In addition, Xavier et al. have recently reported that nitric acid treatment of titanium surface (passivation) increase the release of chemical elements from the surface of titanium [29]. We have therefore performed XPS analysis of both passivated (Tipassiv) and BCP grid blasted (Tiblast) titanium specimen. The blasting process seems to favour the incorporation of biocompatible BCP on the surface of the oxide and the elimination of contamination impurities. Although a comparison between Al₂O₃ and BCP blasting has not been performed in the present study, recent reports have shown that alumina particles are often embedded on the titanium surface and these impurities might hamper the osseointegration of titanium implants [7, 30, 31 and 32]. Another work concluded the opposite showing that the osseointegration of titanium dental implants is not affected by residual aluminium oxide [33]. In order to address this issue, we plan to compare the behaviour of osteoblast cells as well as the osseointegration of both alumina and BCP-blasted titanium implants with comparable surface roughness.

While the effects of surface roughness remains controversial with respect to osteoblastic phenotype expression [13 and 34], we aimed at testing whether MC3T3-E1 preserved their phenotype in direct contact with BCP grid-blasted surface. This phenotypic characterization can be based on the analysis of various osteoblastic markers including type I collagen, ALP, osteocalcine, and cbfa-1/OSF-2. In our work, we focused our investigation on ALP activity. BCP grid-blasted surface induced a slight decrease in ALP activity, as well as in BMP-2 responsiveness. Once again, this reduction in osteoblastic activity may be related to the release of trace elements [29]. However, the positive correlation between surface roughness and ALP activity reported by Boyan et al. was limited to RA values ranging from 5 to 7 
m [9]. Since the Ra values obtained by BCP grid-blasting process remained definitely lower, it will be interesting to increase the roughness obtained by BCP grid-blasting in order to observe some valuable cellular effects.

In conclusion, our results demonstrate for the first time that biphasic calcium phosphate abrasive particles can be used to create titanium surface roughness. This new grid-blasting process increases surface roughness of titanium implants and offers a noncytotoxic surface with regard to mouse osteoblasts. Today, bone implantation is required to clearly decipher the efficacy of our BCP grid-blasted surface in increasing early bone anchorage. This work is part of an effort to better understand the complex processes of cell/material interactions that could be of clinical relevance to the development of more advanced technology for bone, dental and joint tissue engineering.

7. Acknowledgements

Authors thank Genetic institute Corp. (Cambridge, USA) for providing rhBMP-2 and Dr. X. Bourges (Biomatlante SA, Vigneux de Bretagne, France) for supplying BCP abrasive particles and titanium discs. Authors also thank F. Castells (Nantes University Institute of Technology, France) for help with roughness measurement. We gratefully acknowledge Dr. P. Weiss for the critical reading of the manuscript. This work was partially supported by grants from INSERM EM 9903, "Fondation de l'Avenir pour la Recherche Médicale Appliquée", and CPER Biomatériaux Pays de Loire 2000-2004.

8. References

1. M. Long and H.J. Rack, Titanium alloys in total joint replacement: a materials science perspective. Biomaterials 19 (1998), pp. 1621-1639. Abstract | PDF (341 K)

2. J.A. Disegi, Titanium alloys for fracture fixation implants. Injury 31 Suppl 4 (2000), pp. 14-17. Abstract-MEDLINE  

3. K.A. Thomas and S.D. Cook, An evaluation of variables influencing implant fixation by direct bone apposition. J Biomed Mater Res 19 (1985), pp. 875-901. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex  

4. P. Predecki, J.E. Stephan, B.A. Auslaender, V.L. Mooney and K. Kirkland, Kinetics of bone growth into cylindrical channels in aluminum oxide and titanium. J Biomed Mater Res 6 (1972), pp. 375-400. Abstract-Compendex | Abstract-MEDLINE  

5. L. Carlsson, T. Rostlund, B. Albrektsson and T. Albrektsson, Removal torques for polished and rough titanium implants. Int J Oral Maxillofac Implants 3 (1988), pp. 21-24. Abstract-MEDLINE  

6. U. Gbureck, A. Masten, J. Probst and R. Thull, Tribochemical structuring and coating of implant metal surfaces with titanium oxide and hydroxyapatite layers. Mat Sci Eng C 23 (2003), pp. 461-465. SummaryPlus | Full Text + Links | PDF (381 K)

7. M. Esposito, J.M. Hirsch, U. Lekholm and P. Thomsen, Biological factors contributing to failures of osseointegrated oral implants. (II). Etiopathogenesis. Eur J Oral Sci 106 (1998), pp. 721-764. Abstract-MEDLINE   | Full Text via CrossRef

8. J. Lincks, B.D. Boyan, C.R. Blanchard, C.H. Lohmann, Y. Liu, D.L. Cochran, D.D. Dean and Z. Schwartz, Response of MG63 osteoblast-like cells to titanium and titanium alloy is dependent on surface roughness and composition. Biomaterials 19 (1998), pp. 2219-2232. Abstract | PDF (752 K)

9. B.D. Boyan, V.L. Sylvia, Y. Liu, R. Sagun, D.L. Cochran, C.H. Lohmann, D.D. Dean and Z. Schwartz, Surface roughness mediates its effects on osteoblasts via protein kinase A and phospholipase A2. Biomaterials 20 (1999), pp. 2305-2310. SummaryPlus | Full Text + Links | PDF (181 K)

10. D.D. Deligianni, N. Katsala, S. Ladas, D. Sotiropoulou, J. Amedee and Y.F. Missirlis, Effect of surface roughness of the titanium alloy Ti-Al-4 V on human bone marrow cell response and on protein adsorption. Biomaterials 22 (2001), pp. 1241-1251. SummaryPlus | Full Text + Links | PDF (427 K)

11. K. Anselme, M. Bigerelle, B. Noel, E. Dufresne, D. Judas, A. Iost and P. Hardouin, Qualitative and quantitative study of human osteoblast adhesion on materials with various surface roughnesses. J Biomed Mater Res 49 (2000), pp. 155-166. Abstract-MEDLINE | Abstract-EMBASE | Abstract-Compendex   | Full Text via CrossRef

12. K. Anselme, Osteoblast adhesion on biomaterials. Biomaterials 21 (2000), pp. 667-681. SummaryPlus | Full Text + Links | PDF (224 K)

13. A. Wennerberg, T. Albrektsson and B. Andersson, Bone tissue response to commercially pure titanium implants blasted with fine and coarse particles of aluminum oxide. Int J Oral Maxillofac Implants 11 (1996), pp. 38-45. Abstract-MEDLINE  

14. G. Daculsi, O. Laboux and R.Z. LeGeros, Outcome and perspectives in bioactive coatings: what's new, what's coming. ITBM-RBM 23 (2002), pp. 317-325. SummaryPlus | Full Text + Links | PDF (2071 K)

15. R.Z. LeGeros, Properties of osteoconductive biomaterials: calcium phosphates. Clin Orthop 395 (2002), pp. 81-98. Abstract-MEDLINE | Abstract-EMBASE   | Full Text via CrossRef

16. G. Daculsi, Biphasic calcium phosphate concept applied to artificial bone, implant coating and injectable bone substitute. Biomaterials 19 (1998), pp. 1473-1478. Abstract | PDF (321 K)

17. A. Suzuki, J. Guicheux, G. Palmer, Y. Miura, Y. Oiso, J.P. Bonjour and J. Caverzasio, Evidence for a role of p38 MAP kinase in expression of alkaline phosphatase during osteoblastic cell differentiation. Bone 30 (2002), pp. 91-98. SummaryPlus | Full Text + Links | PDF (299 K)

18. H. Sudo, H.A. Kodama, Y. Amagai, S. Yamamoto and S. Kasai, In vitro differentiation and calcification in a new clonal osteogenic cell line derived from newborn mouse calvaria. J Cell Biol 96 (1983), pp. 191-198. Abstract-EMBASE | Abstract-MEDLINE   | Full Text via CrossRef

19. D. Heymann, J. Guicheux and A.V. Rousselle, Ultrastructural evidence in vitro of osteoclast-induced degradation of calcium phosphate ceramic by simultaneous resorption and phagocytosis mechanisms. Histol Histopathol 16 (2001), pp. 37-44. Abstract-EMBASE | Abstract-MEDLINE | Abstract-Elsevier BIOBASE  

20. B. Relic, J. Guicheux, F. Mezin, E. Lubberts, D. Togninalli, I. Garcia and W.B. van den Berg, Guerne PA Il-4 and IL-13, but not IL-10, protect human synoviocytes from apoptosis. J Immunol 166 (2001), pp. 2275-2282.

21. D. Magne, G. Bluteau, C. Faucheux, G. Palmer, C. Vignes-Colombeix, P. Pilet, T. Rouillon, J. Caverzasio, P. Weiss, G. Daculsi and J. Guicheux, Phosphate is a specific signal for ATDC5 chondrocyte maturation and apoptosis-associated mineralization: possible implication of apoptosis in the regulation of endochondral ossification. J Bone Miner Res 18 (2003), pp. 1430-1442. Abstract-MEDLINE  

22. J. Guicheux, J. Lemonnier, C. Ghayor, A. Suzuki, G. Palmer and J. Caverzasio, Activation of p38 mitogen-activated protein kinase and c-Jun NH2-terminal kinase by BMP-2 and their implication in the stimulation of osteoblastic cell differentiation. J Bone Miner Res 18 (2003), pp. 2060-2068. Abstract-MEDLINE  

23. N. Aebli, J. Krebs, H. Stich, P. Schawalder, M. Walton, D. Schwenke, H. Gruner, B. Gasser and J.C. Theis, In vivo comparison of the osseointegration of vacuum plasma sprayed titanium- and hydroxyapatite-coated implants. J Biomed Mater Res 66A (2003), pp. 356-363. Abstract-MEDLINE | Abstract-Compendex   | Full Text via CrossRef

24. P. Linez-Bataillon, F. Monchau, M. Bigerelle and H.F. Hildebrand, In vitro MC3T3 osteoblast adhesion with respect to surface roughness of Ti6Al4 V substrates. Biomol Eng 19 (2002), pp. 133-141. SummaryPlus | Full Text + Links | PDF (628 K)

25. Cook SD, Thomas KA, Kay JF, Jarcho M. Hydroxyapatite-coated titanium for orthopedic implant applications. Clin Orthop 1988:225-43.

26. H.J. Ronold and J.E. Ellingsen, Effect of micro-roughness produced by TiO₂ blasting—tensile testing of bone attachment by using coin-shaped implants. Biomaterials 23 (2002), pp. 4211-4219. SummaryPlus | Full Text + Links | PDF (708 K)

27. A. Ben-Ze'ev, Cytoskeletal and adhesion proteins as tumor suppressors. Curr Opin Cell Biol 9 (1997), pp. 99-108.

28. K. Anselme, P. Linez, M. Bigerelle, D. Le Maguer, A. Le Maguer, P. Hardouin, H.F. Hildebrand, A. Iost and J.M. Leroy, The relative influence of the topography and chemistry of TiAl6V4 surfaces on osteoblastic cell behaviour. Biomaterials 21 (2000), pp. 1567-1577. SummaryPlus | Full Text + Links | PDF (1813 K)

29. S.P. Xavier, P.S. Carvalho, M.M. Beloti and A.L. Rosa, Response of rat bone marrow cells to commercially pure titanium submitted to different surface treatments. J Dent 31 (2003), pp. 173-180. SummaryPlus | Full Text + Links | PDF (288 K)

30. A. Toni, C.G. Lewis, A. Sudanese, S. Stea, F. Calista, L. Savarino, A. Pizzoferrato and A. Giunti, Bone demineralization induced by cementless alumina-coated femoral stems. J Arthroplasty 9 (1994), pp. 435-444. Abstract

31. L. Savarino, E. Cenni, S. Stea, M.E. Donati, G. Paganetto, A. Moroni, A. Toni and A. Pizzoferrato, X-ray diffraction of newly formed bone close to alumina or hydroxyapatite-coated femoral stem. Biomaterials 14 (1993), pp. 900-905. Abstract

32. B.W. Darwell, N. Samman, W.K. Luk, R.K. Clark and H. Tideman, Contamination of titanium castings by aluminium oxide blasting. J Dent 23 (1995), pp. 319-322.

33. A. Piatelli, M. Degidi, M. Paolantonio, C. Mangano and A. Scarano, Residual aluminium oxide on the surface of titanium implants has no effect on osseointegration. Biomaterials 24 (2003), pp. 4081-4089.

34. D.D. Deligianni, N.D. Katsala, P.G. Koutsoukos and Y.F. Missirlis, Effect of surface roughness of hydroxyapatite on human bone marrow cell adhesion, proliferation, differentiation and detachment strength. Biomaterials 22 (2001), pp. 87-96. SummaryPlus | Full Text + Links | PDF (1271 K)

Corresponding author. Tel.: [phone]; fax: [phone]

---