Relevant aspects in the surface properties in titanium dental implants for the cellular viability

Materials Science and Engineering C 64 (2016) 1–10

Contents lists available at ScienceDirect

Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec

Relevant aspects in the surface properties in titanium dental implants for the cellular viability E. Velasco-Ortega a, C.A. Alfonso-Rodríguez b, L. Monsalve-Guil a, A. España-López a, A. Jiménez-Guerra a, I. Garzón b, M. Alaminos b, F.J. Gil c,d,⁎ a

Department of Stomatology, Faculty of Dentistry, University of Seville, Seville, Spain Tissue Engineering Group, Department of Histology, Faculty of Medicine and Dentistry, University of Granada, Granada, Spain Biomaterials, Biomechanics and Tissue Engineering Research Group, Dept. Ciencia de los Materiales e Ingeniería Metalúrgica, ETSEIB, Universitat Politécnica de Catalunya, Barcelona, Spain d School of Dentistry, Universitat Internacional de Catalunya, Barcelona, Spain b c

a r t i c l e

i n f o

Article history: Received 25 July 2015 Received in revised form 2 January 2016 Accepted 14 March 2016 Available online 17 March 2016 Keywords: Roughness treatments Titanium Wettability Surface free energy Cellular behavior Bone index contact

a b s t r a c t Roughness and topographical features are the most relevant of the surface properties for a dental implant for its osseointegration. For that reason, we studied the four surfaces more used in titanium dental implants: machined, sandblasted, acid etching and sandblasted plus acid etching. The roughness and wettability (contact angle and surface free energy) was studied by means 3D-interferometric microscope and sessile drop method. Normal human gingival fibroblasts (HGF) were obtained from small oral mucosa biopsies and were used for cell cultures. To analyze cell integrity, we first quantified the total amount of DNA and LDH released from dead cells to the culture medium. Then, LIVE/DEAD assay was used as a combined method assessing cell integrity and metabolism. All experiments were carried out on each cell type cultured on each Ti material for 24 h, 48 h and 72 h. To evaluate the in vivo cell adhesion capability of each Ti surface, the four types of discs were grafted subcutaneously in 5 Wistar rats. Sandblasted surfaces were significantly rougher than acid etching and machined. Wettability and surface free energy decrease when the roughness increases in sand blasted samples. This fact favors the protein adsorption. The DNA released by cells cultured on the four Ti surfaces did not differ from that of positive control cells (p N 0.05). The number of cells per area was significantly lower (p b 0.05) in the sand-blasted surface than in the machined and surface for both cell types (7 ± 2 cells for HGF and 10 ± 5 cells for SAOS-2). The surface of the machined-type discs grafted in vivo had a very small area occupied by cells and/or connective tissue (3.5%), whereas 36.6% of the sandblasted plus acid etching surface, 75.9% of sandblasted discs and 59.6% of acid etching discs was covered with cells and connective tissue. Cells cultured on rougher surfaces tended to exhibit attributes of more differentiated osteoblasts than cells cultured on smoother surfaces. These surface properties justify that the sandblasted implants is able to significantly increase bone contact and bone growth with very good osseointegration results in vivo. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Over the last four decades dental implants have changed the face of dentistry. Dental implants are used as artificial replacements for the teeth roots and most of them consist of titanium (Ti). Dental implants are used to restore function and aesthetics in fully or partially edentulous patients [1–3]. Ti is considered to be an excellent material due to its physical properties such as stability, resistance and elasticity [4]. In addition, Ti implants are very stable in vivo and biocompatible, which can improve osseointegration [5]. However, cell adhesion to this type of material is not always strong, and new formulations and surface modifications should be developed to increase cell attachment to Ti ⁎ Corresponding author at: School of Dentistry, Universitat Internacional de Catalunya, Immaculada 22, Barcelona, Spain. E-mail address: [email protected] (F.J. Gil).

http://dx.doi.org/10.1016/j.msec.2016.03.049 0928-4931/© 2016 Elsevier B.V. All rights reserved.

implants [6]. In fact, it has been previously demonstrated that surface characteristics are one of key factors that determine the long-term success of dental implants [7]. In this regard; novel Ti-based biomaterials for use in dentistry include some modifications of the implant surface to increase biocompatibility and cell adhesion. In general, human bone and stromal cells have very low attachment capability to smooth metal surfaces, and the use of these materials could lead to infections, inflammation and low cell viability [8–10]. For this reason, modifications affecting roughness, topography and chemistry of the implant surface have been proposed to increase cell viability and biocompatibility [11]. The most common surface modifications used in dentistry are based on mechanical abrasion, sandblasting and acid etching [12]. Although some previous studies tried to evaluate the effects of these surface modifications on biocompatibility and functionality of dental implants by determining adhesion, migration and cell proliferation [13–14], the most appropriate modification for use on Ti dental

2

E. Velasco-Ortega et al. / Materials Science and Engineering C 64 (2016) 1–10

implants has not been elucidated to the date. In addition, most of these studies are carried out on one single cell type such as fibroblasts, osteoblasts and epithelial cells [15–16], and most studies are restricted to the use of a limited number of methods and techniques to evaluate cell viability and cell function. In this work, we have carried out an ex vivo analysis of cell behavior and function on four surface modifications of Ti implants using two different human cell types - stromal fibroblasts and a bone cell line - by using a combined array of methods and techniques to determine which Ti material modification could be more appropriate for future in vivo use.

2. Materials and methods 2.1. Titanium materials The material studied was a commercial titanium alloy (Ti-6Al-4V ELI extra low interstitial medical grade). In this research, we used four different types of Ti materials subjected to surface modifications. Each material consisted in a disc of 12 mm diameter and 4 mm height whose surfaces had been treated by (A) mechanical abrasion, (B) sandblasting plus acid etching, (C) sandblasting or (D) acid etching. All Ti6Al4V discs were sterilized by gamma irradiation and each cell type was cultured on the top surface of each disc type at a concentration of 88,500 cells/cm2 (100,000 cells per disc) using culture medium. Each

cell type was maintained in each Ti material for 24 h, 48 h and 72 h at 37 °C with 5% CO2 in a cell incubator. 2.2. Roughness Roughness was determined by means of a white light interferometer microscopy (Wyko NT1100, Veeco). The surface studied was 459.9 × 604.4 μm2 for all samples. Data analysis was performed with Wyko Vision 232TM software (Veeco, USA). A Gaussian filter was used to separate waviness and form from the roughness of the surface. The following cut-off values were applied: λc = 0.8 mm, for micro-rough surfaces and λc = 0.25 mm for control surfaces. The measurements were realised in four different surfaces for each type of treatment to characterize the amplitude and spacing roughness parameters Sa and Pc, respectively. Sa and Pc were calculated by averaging the values of each profile that were evenly distributed along the surface analysed. Sa (the average roughness) is the arithmetic average of the absolute values of the distance of all points of the profile to the mean line. Pc is the number of peaks in the profile per length of analysis. Finally, the index area is the calculated [(real area) / (nominal area)] ratio. The hybrid parameter was calculated from the total real surface instead of from the profiles, as the index area parameter should be used for the Wenzel correction introduced below. Scanning electron microscope (Jeol 6400, Japan) was used to qualitatively to analyse the surface topography of the implants before being implanted.

Fig. 1. Titanium surface analysis by scanning electron microscopy (SEM). In the first three rows, the four types of Ti discs (A (mechanical abrasion), B (sandblasting plus acid etching), C (sandblasting) and D (acid etching)) were analyzed at different augmentations. The lowest panel represents the three-dimensional representation of the analysis of surface plot for each material.

E. Velasco-Ortega et al. / Materials Science and Engineering C 64 (2016) 1–10

2.3. Wettability The static contact angle (SCA) was carried out through the sessile drop method. Eventual kinetic effects have not been observed because the final drop shape does not change with the time. Drops were generated with a micro-metric syringe and deposited on the substrate surface. The measurements were obtained at room temperature (T = 25 °C) in an environmental PMMA chamber that was saturated with the study liquid. This chamber avoids evaporation of the liquid of the dispensed drop during the contact angle measurement. The SCA were determined with ultra-pure distilled water (MilliQ). At least four measurements were carried out with four different samples in each series. The contact angle measurements were performed with a contact angle video based system (Contact Angle System OCA15plus, Dataphysics, Germany) and analyzed with SCA20 software (Dataphysics, Germany). Wenzel equation was applied to account for the effect of roughness on wettability [17–18]: cosðCA0Þ ¼ rIA  cosðCAÞ

ð1Þ

where, CA′ is the apparent or Wenzel contact angle measured with the contact angle meter, rIA is the index area measured by white light interferometry, and CA is the intrinsic or Young contact angle. Other corrections for the apparent contact angle of heterogeneous surfaces, such as the Cassie-Baxter equation, were not applied due to the theoretical calculations reported by Sheng et al. [19]. Topographical features of the type obtained in grit-blasted surfaces with contact angles below 90° resulted in rough surfaces under a wetting state in the Wenzel regime. The apparent contact angles were analysed, and the intrinsic contact angles were calculated from Equation 1 with three different liquids on each material: ultra-pure distilled water (MilliQ), di-iodomethane and formamide [17]. Subsequently, to compare the different grit-blasted surfaces, the total surface free energy (γS, SFE), ‘London’ or ‘dispersion’ component (γdS ), and ‘polar’ component (γpS ) of SFE for all series were calculated. The SFE and its components were obtained by means of the Owens and Wendt equation [20–23]. γS ¼ γdS þγpS γL  ð1 þ cosðCAÞÞ ¼ 2 

ð2Þ   1=2
1=2 γdL  γdS þ γpL  γpS

ð3Þ

where, S and L refer to the solid and liquid, respectively. 2.4. Cell culture Normal human gingival fibroblasts (HGF) were obtained from small oral mucosa biopsies obtained during surgical removal of impacted third molars from healthy donors as previously reported [13]. Briefly, samples were washed in PBS and digested in type I collagenase for 6 h at 37 °C. Then, stromal fibroblasts were harvested by centrifugation and cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% antibiotics/ antimycotics. SAOS-2 bone cell line was purchased from Sigma-

Table 1 Mean ± standard deviation of surface roughness parameters Ra and Pc for the different types of cp Ti implant surfaces. Implant surface

Ra (μm)

Pc (cm−1)

A B C D

0.33 ± 0.1 2.64 ± 0.2 2.78 ± 0.2 1.69 ± 0.1

150.9 ± 6.1 92.1 ± 13.1 82.1 ± 10.0 198.3 ± 3.41

3

Table 2 Apparent contact angles for the three liquids used on the different cp Ti surfaces. Values are mean ± standard deviation. Statistical differences vs. smooth surfaces for each column are indicated by single and double asterisk-symbols (p b 0.05) (19). Surface

Water CA′ [°]

Di-iodomethane CA′ [°]

Formamide CA′ [°]

A B C D

63.3 ± 5 82.0 ± 5* 75.9 ± 6* 65.8 ± 7

49.5 ± 2* 37.0 ± 3** 57.8 ± 1 36.9 ± 4**

50.8 ± 1 356 ± 1* 59.3 ± 2 34.0 ± 5*

Aldrich and maintained in culture in the same culture medium. All cell types were cultured at 37 °C with 5% CO2 and medium was renewed every three days. Cells were subcultured using trypsin-EDTA (0.5 g/L of trypsin and 0.2 g/L of EDTA) (Sigma Aldrich). Patients gave their consent to participate in this study and the Ethics committee approved this study. 2.5. Scanning electron microscopy (SEM) To characterize the surface structure of each Ti material and to analyze cell morphology, we used a scanning electron microscopy (SEM). For this purpose, Ti materials containing the stromal and bone cells cultured for 24 h, 48 h and 72 h were fixed in 2.5% glutaraldehyde for 6 h, washed in cacodylate buffer and dehydrated in alcohol series (50%, 70%, 96% and 100%). Finally, samples were dried at room temperature and coated with carbon. Cell morphology was analyzed with a FEI Quanta 200 scanning electron microscope using the high vacuum mode. For each cell type and each Ti surface, the number of cells was quantified in an area of 200 × 200 μm (40,000 μm2) after 72 h of ex vivo culture. 2.6. Analysis of cell viability To analyze cell integrity, we first quantified the total amount of DNA and LDH released from dead cells to the culture medium. Then, LIVE/ DEAD assay was used as a combined method assessing cell integrity and metabolism. All experiments were carried out on each cell type cultured on each Ti material for 24 h, 48 h and 72 h. As positive controls (PC), the same cell type was cultured on culture dishes at the same confluence and cell number as the experimental groups cultured on Ti discs. As negative controls (NC), the same cells were treated with Triton X100 to induce 100% cell death. • In order to quantify the cell DNA released to the culture medium as a consequence of cell death, 10 μL of culture medium were taken from each experimental group after the established incubation time. Each aliquot was diluted in distilled nuclease-free water (Ambion-Life Technologies, Austin, TX) until a final volume of 100 μL and free DNA was quantified by using a SmartSpec plus Spectrophotometer (BIO-RAD, Hercules, CA) at 260–280 nm wavelengths. Three different replicates (three independent Ti discs) were analyzed by this method, and 3 technical replicate measures were taken from each sample. • Cytoplasmic membrane integrity was assessed by monitoring the presence of a cytoplasmic molecule - lactate dehydrogenase LDH - in Table 3 Water contact angle, surface free energy and its components for the different Ti surfaces. Values are mean ± standard deviation. Statistical differences vs. smooth surfaces for each column are indicated by single and double asterisk-symbols (p b 0.05). Surface

A B C D

Contact angle (°)

Surface free energy (mJ/m2) Total surface free energy

Dispersive component

Polar component

67.29 ± 4.62 72.11 ± 5.15* 78.33 ± 2.94* 66.84 ± 6.97

43.78 ± 1.34 42.08 ± 1.96 42.67 ± 1.18 49.52 ± 3.11*

31.19 ± 1.96 42.00 ± 1.33* 35.99 ± 0.83** 41.10 ± 2.34*

12.70 ± 2.95 8.18 ± 1.24** 7.19 ± 0.92* 9.64 ± 3.23*

4

E. Velasco-Ortega et al. / Materials Science and Engineering C 64 (2016) 1–10

Table 4 Surface energy and its components for the different Ti surfaces. Values are mean ± standard deviation. Statistical differences for each column are indicated by point, singleasterisk and double-asterisk symbols (p b 0.05). Surface

A B C D

Surface energy (mJ/m2) Total

Dispersive component

Polar component

42.3 ± 29.0 27.9 ± 1.4* 28.5 ± 1.3* 37.7 ± 2.1

24.6 ± 2.0** 23.5 ± 0.9•* 18.7 ± 1.2 25.7 ± 2.1**

16.3 ± 5.1* 5.1 ± 0.8 10.5 ± 2.1• 12.0 ± 3.2•

the culture medium. Quantification was carried out by using a commercial LDH cytotoxicity detection kit (Roche, Germany) by following the manufacturer's recommendations. Three different replicates (three independent Ti discs) were analyzed by this method, and 3 technical replicate measures were taken from each sample. • To simultaneously analyze cell viability and functionality by a dual cytoplasmic and nuclear assay, the calcein/AM-ethidium homodimer-1 Viability/Cytotoxicity assay kit LIVE/DEAD (Life Technologies, Carlsbad, CA) was used. Cells were cultured on each Ti material, washed in PBS, detached from the discs using trypsin-EDTA and seeded in culture chambers (Lab-Tek II, Nunc, Denmark) with culture medium. After 15 min at 37 °C in a cell incubator, medium was removed and

calcein/AM-ethidium homodimer-1 were added as indicated by the manufacturer. After 15 min, cells were examined by using a fluorescence microscope (Nikon Eclipse 90i, Nikon, Japan) and the number of red (dead) cells and green (alive and metabolically active) cells was determined. For each experimental condition (cell types, Ti material and incubation time), five different images were taken, and the number of dead and alive cells was determined by using the ImageJ software (MacBiophotonics, Ontario, Canada) as previously reported [13]. For each condition, an average of 1000 cells was analyzed using this system.

2.7. In vivo analysis To evaluate the in vivo cell adhesion capability of each Ti surface, the four types of discs - A, B, C and D - were grafted subcutaneously in 5 Wistar rats. Briefly, each animal was anesthetized with ketamine and acepromazine and four incisions were made in the back skin. Then, a subcutaneous pocket was dissected on each incision and a disc of each type was implanted in each animal. Finally, the skin surgical injuries were repaired using 5/0 non-absorbable stitches. Each rat was euthanatized 72 h later and the four discs of each type were surgically removed from the animals, fixed in 2.5% glutaraldehyde for 6 h, washed in

Fig. 2. Morphological analysis of cells cultured on the different titanium surfaces using scanning electron microscopy (SEM). The four types of Ti discs (A (mechanical abrasion), B (sandblasting plus acid etching), C (sandblasting) and D (acid etching)) were analyzed after HGF and SAOS-2 cells were cultured on each surface for 24, 48 and 72 h.

E. Velasco-Ortega et al. / Materials Science and Engineering C 64 (2016) 1–10

cacodylate buffer and dehydrated in alcohol series (50%, 70%, 96% and 100%). Finally, samples were dried at room temperature, coated with carbon and analyzed with a FEI Quanta 200 scanning electron microscope. Morphology of the cells and extracellular matrix were evaluated and the percentage of area of each disc covered by cells and tissue was calculated.

2.8. Statistical analysis For DNA and LDH quantification, all results were first normalized by considering the values obtained for the positive control as 0% and the values obtained for the negative control as 100%. For LIVE/DEAD™, normalization was performed by taking the results obtained for positive controls as 100% and the results of the negative controls as 0%. For each method, results obtained for each experimental condition were compared with the positive and negative controls using MannWhitney statistical test, and the same test was used to compare the number of cells grown on different surfaces.

3. Results 3.1. Surface analysis The analysis of the different Ti discs using SEM showed some differences among the different discs (Fig. 1). First, the surface analysis of the native Ti material showed that the surface treated by mechanical abrasion was smooth and slightly undulated, with some small pits on its surface and parallel superficial strips at higher augmentation (A). In contrast, the material treated with sandblasting and acid etching (B) was highly rough; a skeletal nano-porous structure resulting from the acid-etching treatment is superimposed to the micro-rough structure resulting from the initial blasting treatment with a coarse and uneven surface consisting in excavated areas (depressions) and raised mountain-like areas, and had evident strips or bands etched on its surface. Ti discs subjected to sandblasting (C) had, a heterogeneous surface structure with peaks and valleys of varied geometry showing several flat facets. The facets also had small irregularities appearing as pits and stripes. Finally, the material treated with acid etching (D) presented a myriad of small craters and groves. The walls of the craters had a micro-patterned structure and pitting at the bottom of the craters was observed. Roughness and topographical features are the most relevant of the surface properties for a bone substitute implants for its clinical success. For this reason, we studied here the surface topography by means 3Dinterferometric microscope and the values of surface roughness obtained are shown in Table 1. Sandblasted plus acid etching (B) and sandblasted (C) surfaces were significantly rougher than acid etching (C) and Control (A) surfaces, and D surfaces were significantly rougher than A according to others authors [24–25]. B and C surfaces did not have significantlydifferent values of roughness. The sizes of those nano-topographic features are smaller than the lateral resolution of the technique used to measure roughness. Table 2 shows the apparent contact angles (CA’) for the three liquids used on all the surfaces tested. Table 3 shows the Surface Free Energy (SFE) calculated according to the Owens and Wendt approach by using the corresponding intrinsic contact angles on all different surfaces. Table 4 also shows the values for the dispersive and polar component parts of the Surface Free Energy (SFE) and these values are important for the electrostatic charges density on the biomaterial surfaces. Overall, the sandblasted treatment decreased surface wettability with water, i.e. it increased the contact angles. As the roughness increased, the SFE decreased. These differences were statistically significant.

5

3.2. Analysis of cell morphology and viability When human cells were cultured on the surface of each material for increasing periods of time, we found some differences between both cell types but very few differences among the different Ti discs surfaces (Fig. 2). In general, cells cultured on the surface of all materials displayed a normal elongated, spindle-shaped or star-shaped morphology compatible with a normal cell type, although this phenotype was slightly modified in some of the surfaces. No morphological alterations were found at any time, and a defined cytoplasm and nuclei was identifiable in most cells. In the first place, HGF cultured on the surface of Ti materials subjected to mechanical abrasion (surface A) tended to show a typical morphology at all times (24, 48 and 72 h), and cells were flattened with many cytoplasmic extensions and lamellipodia. The same behavior was found in SAOS-2 cells cultured on A surfaces, where cells were regular, polygonal and elongated. The number of cells found per 40,000 μm2 area was 23 ± 6 for HGF and 29 ± 11 for SAOS-2. In the second place, HGF cells cultured on B surfaces showed some initial minor morphological changes at 24 h of culture as identified by a less elongated shape with shorter prolongations. However, cells cultured for 48 and 72 h acquired the typically elongated shape as found in A surfaces. In turn, SAOS-2 cells also had some morphological changes at 24 h, with more rounded cells with fewer prolongations, and these changes tended to increase after 48 and 72 h. The number Table 5 Analysis of cell viability as determined by DNA and LDH quantification and LIVE/DEAD methods in human gingival fibroblasts (HGF) and SAOS-2 cells. Normalized values are shown as average ± standard deviations. Incubation time

Cell type

Culture surface

DNA release

LDH analysis

LIVE/DEAD™

24 h 24 h 24 h 24 h 24 h 24 h

HGF HGF HGF HGF HGF HGF

0.9 ± 9.1 8.4 ± 0.1 83.3 ± 3.8 1.3 ± 32.7 7.0 ± 0.1 96.3 ± 1.1 0.8 ± 5.2 3.6 ± 0.0 97.0 ± 2.1 0.5 ± 5.4 1.0 ± 0.0 92.2 ± 2.2 0.0 ± 14.1 0.0 ± 0.1 100.0 ± 0.3 100.0 ± 14.0 100.0 ± 0.2 0.0 ± 0.1

48 h 48 h 48 h 48 h 48 h 48 h

HGF HGF HGF HGF HGF HGF

72 h 72 h 72 h 72 h 72 h 72 h

HGF HGF HGF HGF HGF HGF

24 h 24 h 24 h 24 h 24 h 24 h

SAOS-2 SAOS-2 SAOS-2 SAOS-2 SAOS-2 SAOS-2

48 h 48 h 48 h 48 h 48 h 48 h

SAOS-2 SAOS-2 SAOS-2 SAOS-2 SAOS-2 SAOS-2

72 h 72 h 72 h 72 h 72 h 72 h

SAOS-2 SAOS-2 SAOS-2 SAOS-2 SAOS-2 SAOS-2

Ti-A Ti-B Ti-C Ti-D Positive control Negative control Ti-A Ti-B Ti-C Ti-D Positive control Negative control Ti-A Ti-B Ti-C Ti-D Positive control Negative control Ti-A Ti-B Ti-C Ti-D Positive control Negative control Ti-A Ti-B Ti-C Ti-D Positive control Negative control Ti-A Ti-B Ti-C Ti-D Positive control Negative control

0.0 ± 16.1 4.6 ± 0.1 96.0 ± 4.5 0.0 ± 19.7 3.4 ± 0.0 95.3 ± 2.2 0.0 ± 37.6 0.0 ± 0.0 96.9 ± 1.4 0.0 ± 20.4 0.0 ± 0.0 91.4 ± 5.9 0.0 ± 26.5 0.0 ± 0.0 100.0 ± 0.2 100.0 ± 55.3 100.0 ± 0.1 0.0 ± 0.1 0.0 ± 10.3 1.0 ± 0.0 98.4 ± 1.0 0.0 ± 4.3 0.0 ± 0.1 98.2 ± 1.5 0.0 ± 9.5 0.0 ± 0.0 98.9 ± 1.4 0.0 ± 16.8 0.0 ± 0.1 96.6 ± 4.1 0.0 ± 30.6 0.0 ± 0.0 100.0 ± 1.3 100.0 ± 31.1 100.0 ± 0.5 0.0 ± 0.1 0.0 ± 17.2 13.5 ± 0.1 93.7 ± 2.5 0.0 ± 13.2 8.9 ± 0.1 86.6 ± 6.2 0.2 ± 9.4 3.1 ± 0.0 95.4 ± 1.1 0.0 ± 10.3 0.8 ± 0.0 79.1 ± 4.9 0.0 ± 13.8 0.0 ± 0.1 100.0 ± 0.6 100.0 ± 27.3 100.0 ± 0.2 0.0 ± 0.1 0.0 ± 10.1 3.3 ± 0.0 98.2 ± 2.9 0.0 ± 8.3 0.0 ± 0.0 95.5 ± 2.7 0.0 ± 8.8 0.0 ± 0.0 100.5 ± 2.6 0.0 ± 5.7 0.0 ± 0.0 95.3 ± 2.6 0.0 ± 7.6 0.0 ± 0.1 100.0 ± 3.8 100.0 ± 30.6 100.0 ± 0.1 0.0 ± 0.1 0.2 ± 21.1 4.4 ± 0.0 97.7 ± 2.9 0.0 ± 11.1 1.9 ± 0.0 97.4 ± 2.5 0.0 ± 18.1 0.9 ± 0.0 98.2 ± 12.3 0.0 ± 22.9 0.0 ± 0.0 92.3 ± 6.9 0.0 ± 15.6 0.0 ± 0.0 100.0 ± 2.3 100.0 ± 22.7 100.0 ± 0.1 0.0 ± 0.1

6

E. Velasco-Ortega et al. / Materials Science and Engineering C 64 (2016) 1–10

Fig. 3. Analysis of cell viability as determined by DNA and LDH quantification and LIVE/DEAD Viability/Cytotoxicity assay. In each case, HGF (in blue) and SAOS-2 cells (in red) were cultured on each Ti surface A (mechanical abrasion), B (sandblasting plus acid etching), C (sandblasting) and D (acid etching))and the amount of released DNA and LDH along with the percentage of live cells were analyzed at 24, 48 and 72 h. PC: positive controls, NC: negative controls. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 4. Illustrative images of the analysis of cell viability as determined by LIVE/DEAD Viability/Cytotoxicity assay. HGF and SAOS-2 cells were cultured on the surface of A (mechanical abrasion), B (sandblasting plus acid etching), C (sandblasting) and D (acid etching) Ti materials for 24, 48 and 72 h. PC: positive controls, NC: negative controls.

E. Velasco-Ortega et al. / Materials Science and Engineering C 64 (2016) 1–10

7

Fig. 5. Scanning electron microscopy evaluation of Ti discs implanted in vivo in laboratory animals. The four types of Ti discs (surfaces A (mechanical abrasion), B (sandblasting plus acid etching), C (sandblasting) and D (acid etching)) were analyzed 72 h after in vivo grafting using SEM. The area corresponding to attached cells and connective tissue has been highlighted in the images at lower magnification.

of cells was 8 ± 9 for HGF and 11 ± 8 for SAOS-2 per analyzed area (differences with the A surface were statistically significant). On the other hand, HGF cells cultured on the surface of materials treated with sandblasting (C materials), showed some minor modifications as compared to the A surface, with cells showing less prolongations, but very few differences were found in general. However, SAOS-2 cells grown on C surfaces were more rounded and showed very few cell prolongations as compared with cells cultured on A surfaces. The number of cells per area was significantly lower (p b 0.05) in the C surface than in the A and surface for both cell types (7 ± 2 cells for HGF and 10 ± 5 cells for SAOS-2). Finally, cells cultured on D surfaces were morphologically very similar to cells cultured on C Ti materials. As shown in Fig. 1, HGF showed some minor modifications as compared to A surfaces, whereas SAOS-2 cells were more rounded and the number of cell prolongations was lower. The statistical analysis showed that the number of cells grown per area was similar to the B and C surfaces and lower to A surface (8 ± 4 for HGF and 11 ± 7 for SAOS-2). 3.3. Analysis of cell viability as determined by DNA quantification To determine cell viability of the cells cultured on each surface, we first quantified the amount of nuclear DNA released by dead cells to

the culture medium (Table 5 and Fig. 3A). The results showed that the concentration of DNA released by negative control cells was significantly higher than that of the rest of study groups (p b 0.05). However, the DNA released by cells cultured on the Ti surfaces A, B, C and D did not differ from that of positive control cells (p N 0.05). In addition, the free DNA detected in the culture medium was very low in all Ti surfaces at all incubation times, and no statistical differences were found among the four Ti discs. Differences between HGF and SAOS-2 cells were not significant (p b 0.05). 3.4. Analysis of cell viability as determined by LDH quantification The analysis of LDH released to the culture medium by cells with altered cell viability showed several differences among groups (Table 5 and Fig. 3B). First, cells cultured on each Ti surface had significantly less LDH release than negative control cells (p b 0.05). Then, comparison with positive control cells showed significant differences for the A (mechanical abrasion) (p = 0.00209) and B (sandblasting plus acid etching) (p = 0.02223) Ti surfaces, but not for C (sandblasting) and D (acid etching) (p N 0.05). In fact, the highest LDH levels corresponded to cells cultured on discs with the A surface for both cell types. Differences for the different periods of time (24, 48 and 72 h) and between both cell types

Table 6 ALP activity for all the investigated samples measured at 7, 14, and 22 days of incubation. ALP activity per cell (mol/L min−1). Surface

7 days

17 days

22 days

A B C D

1.2 · 10−11 ± 0.3 · 10−11 4.2 · 10−11 ± 0.9 · 10−11* 3.9 · 10−11 ± 1.0 · 10−11* 1.8 · 10−11 ± 0.7 · 10−11

1.9 · 10−11 ± 0.7 · 10−11* 6.1 · 10−11 ± 1.3 · 10−11** 5.7 · 10−11 ± 1.0 · 10−11 2.4 · 10−11 ± 0.9 · 10−11**

1.7 · 10−11 ± 0.6 · 10−11 4.5 · 10−11 ± 0.9 · 10−11* 5.3 · 10−11 ± 0.8 · 10−11 2.0 · 10−11 ± 0.5 · 10−11*

8

E. Velasco-Ortega et al. / Materials Science and Engineering C 64 (2016) 1–10

were not statistically significant (p N 0.05), although LDH values tended to be higher at 24 h. Cell differentiation was determined after 1, 7, 17, and 22 days in culture and was established by measuring the ALP activity. Because cell differentiation is of major interest for biomedical applications, we induced osteoblastic differentiation with osteogenic medium and determined the alkaline phosphatase activity as early marker for osteoblast differentiation. 3.5. Analysis of cell viability as determined by the dual LIVE/DEAD viability/ cytotoxicity assay Determination of cell viability by the dual cytoplasmic and nuclear assay LIVE/DEAD revealed that the cell viability and function of both cell types was high in general (above 78% in all cases), although differences among groups were detected (Table 5 and Figs. 3C and 4). First, the statistical analysis demonstrated that cell viability as determined by this assay was higher than negative control cells for all groups (p b 0.05). In the second place, comparison with positive control cells revealed statistical differences for the A (p = 0.00329), B (p = 0.00329) and D (p = 0.00329) surfaces, but not for the C material (p N 0.05). No differences were found between HGF and SAOS-2 (p b 0.05) and for the three time periods analyzed here (p b 0.05 for the 24, 48 and 72 h comparisons). 3.6. In vivo analysis When the four types of Ti discs were grafted in laboratory animals, we found that cells tended to attach and adhere to rough surfaces rather than to smooth A surfaces (Fig. 5). In general, the SEM evaluation of each surface revealed that the areas covered with cells and tissue in vivo had numerous spindle-shaped or star-shaped cells with abundant cell prolongations and lamellipodia, along with a high concentration of collagen fibers and red blood cells. The structure and composition of this in vivo-formed connective tissue was very similar among the four Ti surfaces, but the amount of this tissue was different in each case. Specifically, the surface of the A-type discs grafted in vivo had a very small area occupied by cells and/or connective tissue (3.5% ± 0.9%), whereas 36.6% ± 2.5% of the surface of the B discs, 75.9% ± 5.2% of C discs and 59.6% ± 7.6% of D discs was covered with cells and connective tissue (Fig. 5). Differences were not significant. 4. Discussion Ti implants have been extensively used in dentistry and orthopaedic surgery. However, biocompatibility of implants varies according to numerous factors, and one of the most important parameters affecting osseoconductivity and osseointegration is the morphology of the implant surface [5]. In this regard, it has been previously demonstrated that the degree of roughness of the implant surface, its chemistry, topography, and wettability may affect cell function, adhesion and viability [26]. The development of bone-implant interfaces depend on the direct interactions of osteoblasts and the subsequent deposition of bone matrix. Therefore, the proper cell adhesion and the formation of osteoblast extracellular matrix (ECM) are essential steps for the successful osseointegration of biomaterials [27–28]. Cell adhesion on 2D surfaces constrains to a certain extent the natural interaction with ECM. It starts with the adsorption of adhesive matrix proteins from the surrounding medium, followed by recognition of these proteins by the cells [29–30] which triggers specific cellular responses [31–32]. It is now well documented that the biomaterial properties, such as wettability [33], charge [34], chemistry and surface topography [35] play a critical role in the establishment of cell-biomaterial contacts [36]. Although controversial results exist [37], the fact that surface topography strongly influences the behavior of adhering cells is widely accepted in the literature [31–35], and particularly for the Titanium (Ti)

surfaces [29,37]. For example fibroblasts adhering on ground Ti surface extend their body in the direction of surface groves and these aligned cells attach better than the spherical ones [38]. Similar orientations on grooved surfaces undergo osteoblasts [39]: on smooth substrate they were found randomly oriented, while they line up in parallel to the groves of 5 μm deep. Conversely, the osteoblasts did not assess topography of 0,5 μm [39]. The attempts for quantitative characterization of roughness however, are rather weak in many of these studies. Indeed, it is now increasingly admitted that the roughness has to be considered in terms of amplitude and organization [40–42], and both factors influence the osteoblasts behavior. Grit blasting of Ti to produce defined surface roughness may facilitate the development of bioactive interface, which may improve in vitro osteoblast differentiation [11–14] as well as osseointegration [14]. However, the mechanisms that underscore such topographical response are still poorly understood. The concave valleys on a Ti surface will accumulate a higher density of hydroxil with negative charge and highly polar groups than convex peaks [43,44]. The possibility of air pockets at the bottom of the topography in the early stages of contact with the protein solution, due to dynamic effects during wetting, is a possible explanation for the preferential accumulation of the protein on the top of the topographical features. Consequently, the SFE values must be put into perspective and considered only for comparative purposes between surfaces with similar topographical characteristics, i.e. those that are sandblasted. Moreover, some of the abrasive particles remain adhered to the titanium surface after the shot blasting treatment. These remaining abrasives particles (in general, alumina) influence SFE but they were not heterogeneously distributed along the surfaces, so they probably did not influence fibronectin distribution on the sandblasted surfaces. However, they may have affected the total amount of protein adsorbed through their influence on the wettability and SFE [41–42]. A similar pattern of adsorption was observed for other proteins, such as albumin and fibrinogen, i.e. one globular and one fibrillar protein. This means that the heterogeneity of adsorption has nothing to do with the protein structure. Moreover, there was no correlation between the blasting particle size, i.e. the surface roughness, and the corresponding changes in SFE [43–45]. All these factors lead to the conclusion that the observed heterogeneity in protein adsorption must be attributed to a property or feature of the blasted Ti surface, regardless of the protein structure or the way the roughness is formed [17]. To assess cell viability, we used three different analysis methods. This combinatorial approach was able to increase the sensitivity and accuracy of the results as previously stated by Stoddart [46] and MartínPiedra et al. [47]. In general, our results showed that all Ti surfaces were very biocompatible and the number of cells whose viability was compromised by culturing on the different Ti surfaces was very low. However, some differences were found among the four Ti surfaces. On the one hand, the methods based on intracellular components release (DNA and LDH) showed that the amount of free DNA in the culture medium was always very low as compared to control cells cultured on culture flasks. These results suggest that the number of cells whose cytoplasmic and nuclear membranes have been disrupted as a consequence of the interaction with the surfaces was very low, and confirm the safety of Ti materials for clinical use [48] Furthermore, identification of cytoplasmic cell components in the culture medium as determined by LDH quantification demonstrated certain degree of cytotoxicity in cells cultured on A (mechanical abrasion) and B (sandblasting plus acid etching) surfaces, but not in C (sandblasting) and D (acid etching). It is well known that LDH methods are more sensible than DNA quantification, and cells showing limited cell damage may be positive for LDH methods and negative for DNA quantification. On the other hand, the application of a highly-sensitive dual dye exclusion and metabolic cytoplasmic assay (LIVE/DEAD) allowed us to evaluate two crucial parameters of the living cells: the integrity of the cell membrane pumps and the functionality of cytoplasmic enzymes [49]. By using this accurate method, we were able to detect that A (mechanical abrasion), B (sandblasting

E. Velasco-Ortega et al. / Materials Science and Engineering C 64 (2016) 1–10

plus acid etching) and D (acid etching) surfaces generated certain degree of damage on the cells cultured on Ti discs, and the C (sand blasting) surface was the only one that was not associated to cell damage. Globally, all these results suggest that C (sandblasting) surfaces and, to a lesser extent, also the D (acid etching) surfaces, could be the most biocompatible ones, and very few differences were found between the two cell types analyzed here. These results are in agreement with previous reports demonstrating that Ti is very biocompatible in general, but sandblasted Ti may be associated to high cell viability as compared to other Ti surfaces [14]. After analyzing the cell viability, we evaluated the capability of each biomaterial to induce ex vivo cell differentiation on Ti surfaces. As previously demonstrated, Ti implants are able to actively regulate cells rather than undergoing a passive healing process and surface conditions can affect cell morphology and differentiation [20]. In this milieu, our results showed that human cells cultured on B (sandblasting plus acid etching), C (sandblasting) and D (acid etching) Ti surfaces displayed some signs of cell differentiation as compared to A (mechanical abrasions) surfaces. In fact, our results are in agreement with reports by [14,20], who demonstrated that cells cultured on rougher surfaces tended to exhibit attributes of more differentiated osteoblasts than cells cultured on smoother surfaces. The fact that cells were more rounded, with fewer prolongations, and cell viability was reduced, points out the possibility that cells cultured on B (sandblasting plus acid etching), C (sandblasting) and D (acid etching) surfaces may be undergoing a process of osteoblastic differentiation, especially in the case of the bone-like SAOS-2 cells. ALP activity was much higher (p-value b 0.001) on the B (sandblasting plus aid etching) and C (sandblasting) surfaces than on all the A (mechanical abrasion) or D (acid etching) during the whole period of experimentation (Table 6). No significant differences were observed between the two B and C that peaked in ALP activity after 17 days in culture. Cells on the A and D samples exhibited similar response for the 22 days of incubation, which indicates a delay in their early differentiation. Finally, in the present study we demonstrated that the four Ti surfaces analyzed in this work had different cell adhesion potential in vivo. Thus, the subcutaneous implant of these Ti materials in laboratory animals revealed that smooth A (mechanical abrasion) materials had very few cells and connective tissue adhered to their surface, whereas rough B (sandblasting plus acid etching), C (sandblasting) and D (acid etching) discs had a significant percentage of their surface covered by cells, collagen fibers and connective tissue. Although differences were not significant, our results point out the possibility that living cells and tissues may tend to adhere preferably to C (sandblasting) surfaces than to B (sandblasting plus acid etching) and D (acid etching) surfaces. New works should be carried out with higher number of animals to confirm this hypothesis, but several studies already demonstrated that roughened Ti surfaces behave in vivo much better than smooth surfaces, and superior histomorphometry and stronger bone responses have been found [50]. In fact, previous reports demonstrated that sandblasting is able to significantly increase bone contact and bone growth with very good osseointegration results in vivo [51]. In summary, in the present work we demonstrated that biocompatibility and cell adhesion can be improved by modifying the characteristics of the Ti surface. Although differences among the three rough surfaces were minor, the surface associated to highest cell viability was the sandblasted Ti. If these results can be confirmed by independent studies including in vivo testing on mandibular bone, we hypothesize that sandblasting may improve osteointegration and shorten the edentulous period of a patient [52,53]. 5. Conclusions Sandblasted treatment decreased surface wettability and it increased the contact angles, as the roughness increased, the Surface Free Energy decreased. Besides, the negative charge produced by the

9

debris of alumina on the titanium surface favors the protein adsorption and consequently the cell adhesion. However, we did not find statistical differences in the cell differentiation for the four surfaces studied. The surface of the machined-type discs grafted in vivo had a very small area occupied by cells and/or connective tissue for the machined samples and around 75% for sandblasted samples. Cells cultured on rougher surfaces tended to exhibit attributes of more differentiated osteoblasts than cells cultured on smoother surfaces.

Acknowledgements The authors are grateful to Spanish Government for the CICYT (Ministerio Economía y Competitividad) project MAT 2012-30706.

References [1] D. Buser, Titanium for dental applications (I), in: Brunette D.M., Tengvall P., Textor M., Thomsen P. (Eds.), Titanium in Medicine, Springer Verlag, Berlin-New YorkHeidelberg 2001, pp. 875–888. [2] A. Bagno, C. Di Bello, Surface treatments and roughness properties of Ti-based biomaterials, J. Mater. Sci. Mater. Med. 15 (2004) 939–945. [3] T.B. Kardos, Cellular responses to metal ions released from implants, J. Oral Implantol. 40 (2014) 294–298. [4] O.E. Ogler, Implant surface, materials, design and osseointegration, Dent. Clin. North Am. 28 59 (2) (2015) 505–520. [5] E. Velasco, J. Pato, A. Cameán, J.J. Segura, In vitro evaluation of cytotoxicity and genotoxicity of a commercial titanium alloy for dental implantology, Mutat. Res. 702 (2010) 17–23. [6] L. Le Guehennec, Surface treatments of titanium dental implants for rapid osseointegration, Dent. Mater. 23 (2007) 844–854. [7] T. Albrektsson, P.I. Branemark, H.A. Hansson, J. Lindstrom, Osseointegrated titanium implants. Requirements for ensuring a long-lasting, direct bone-to-implant anchorage in man, Acta Orthop. Scand. 52 (1981) 155–170. [8] H. Kim, S. Choi, J. Ryu, S. Koh, J. Park, I.S. Lee, The biocompatibility of SLA-treated titanium implants, Biomed. Mater. 3 (2008) 234–240. [9] D. Li, S.J. Ferguson, T. Beutler, D.L. Cochran, C. Sittig, H.P. Hirt, D. Buser, Biomechanical evaluation of the interfacial strength of a chemically modified sandblasted and acid-etched titanium surface, J. Biomed. Mater. Res. A 76 (2) (2006) 323–334. [10] D.D. Deligianni, N.D. Katsala, P.G. Koutsoukos, Y.F. Missirlis, Effect of surface roughness of hydroxyapatite on human bone marrow cell adhesion, proliferation, differentiation and detachment strength, Biomaterials 22 (1) (2001) 87–96. [11] C. Aparicio, D. Rodriguez, F.J. Gil, Variation of roughness and adhesion strength of deposited apatite layers on titanium dental implants, Mater. Sci. Eng. C 31 (2011) 320–324. [12] A. Wennerberg, T. Albrektsson, Suggested guidelines for the topographic evaluation of implant surfaces, Int. J. Oral Maxillofac. Implants 15 (2000) 331–344. [13] K. Anselme, Osteoblast adhesion on biomaterials, Biomaterials 21 (2000) 667–681. [14] C. Aparicio, F.J. Gil, J.A. Planell, E. Engel, Human osteoblast proliferation and differentiation on grit-blasted and bioactive titanium for dental applications, J. Mater. Sci. Mater. Med. 13 (2002) 1050–1111. [15] K. Donath, G. Breuner, A method for the study of undecalcified bones and teeth with attached soft tissues. The Sage-Schliff (sawing and grinding) technique, J. Oral Pathol. 11 (1989) 318–326. [16] J. Guillem, L. Delgado, M. Godoy-Galardo, M. Pegueroles, M. Herrero, F.J. Gil, Fibroblast adhesion and activation onto micro-machined titanium surfaces, Clin. Oral Implants Res. 24 (2013) 770–780. [17] M. Pegueroles, F.J. Gil, J.A. Planell, C. Aparicio, The influence of blasting and sterilization on static and time-related wettability and surface-energy properties of titanium surfaces, Surf. Coat. Technol. 202 (2008) 3470–3479. [18] R.N. Wenzel, Resistance of solid surfaces to wetting by water, Ind. Eng. Chem. 28 (1936) 988–994. [19] Y.J. Sheng, S. Jiang, H.K. Tsao, Effects of geometrical characteristics of surface roughness on droplet wetting, J. Chem. Phys. 127 (2007) 234704-1-7. [20] E.A. Vogler, Structure and reactivity of water at biomaterial surfaces, Adv. Colloid Interf. Sci. 74 (1998) 69–117. [21] J. Drelich, Static contact angles for liquids at heterogeneous rigid solid surfaces, Pol. J. Chem. 71 (1997) 525–549. [22] R.J. Good, Contact-angle, wetting, and adhesion - a critical-review, J. Adhes. Sci. Technol. 6 (1992) 1269–1302. [23] M. Morra, C. Cassinelli, Bacterial adhesion to polymer surfaces: a critical review of surface thermodynamic approaches, J. Biomater. Sci. Polym. Ed. 9 (1997) 55–74. [24] M. Pegueroles, C. Aparicio, M. Bosio, E. Engel, F.J. Gil, J.A. Planell, G. Altankov, Spatial organization of osteoblast fibronectin matrix on titanium surfaces: effects of roughness, chemical heterogeneity and surface energy, Acta Biomater. 6 (2010) 291–301. [25] C. Aparicio, D. Rodriguez, F.J. Gil, Variation of roughness and adhesion strength of deposited apatite layers on titanium dental implants, Mater. Sci. Eng. C 31 (2011) 320–324. [26] L. Feller, Y. Jadwat, R.A. Khammissa, R. Meyerov, I. Schechter, J. Lemmer, Cellular responses evoked by different surface characteristics of intraosseous titanium implants, Biomed. Res. Int. 171945 (2015), http://dx.doi.org/10.1155/2015/171945.

10

E. Velasco-Ortega et al. / Materials Science and Engineering C 64 (2016) 1–10

[27] M. Jayaraman, U. Meyer, M. Buhner, U. Joos, H.P. Wiesmann, Influence of titanium surfaces on attachment of osteoblast-like cells in vitro, Biomaterials 25 (2004) 625–631. [28] R. Lange, F. Luthen, U. Beck, J. Rychly, A. Baumann, B. Nebe, Cell-extracellular matrix interaction and physico-chemical characteristics of titanium surfaces depend on the roughness of the material, Biomol. Eng. 19 (2002) 255–261. [29] C.J. Wilson, R.E. Clegg, D.I. Leavesley, M.J. Pearcy, Mediation of biomaterial-cell interactions by adsorbed proteins: a review, Tissue Eng. 11 (2005) 1–18. [30] D.A. Puleo, A. Nanci, Understanding and controlling the bone-implant interface, Biomaterials 20 (1999) 2311–2321. [31] M.C. Siebers, P.J. der Brugge, X.F. Walboomers, J.A. Jansen, Integrins as linker proteins between osteoblasts and bone replacing materials. A critical review, Biomaterials 26 (2005) 137–146. [32] A.J. Garcia, Get a grip: integrins in cell-biomaterial interactions, Biomaterials 26 (2005) 7525–7529. [33] L. Ponsonnet, K. Reybier, N. Jaffrezic, V. Comte, C. Lagneau, M. Lissac, C. Martelet, Relationship between surface properties (roughness, wettability) of titanium and titanium alloys and cell behaviour, Mater. Sci. Eng. C 23 (2003) 551–560. [34] B.D. Boyan, R. Batzer, K. Kieswetter, Y. Liu, D.L. Cochran, S. Szmuckler-Moncler, D.D. Dean, Z. Schwartz, Titanium surface roughness alters responsiveness of MG63 osteoblast-like cells to 1 alpha,25-(OH)(2)D-3, J. Biomed. Mater. Res. 39 (1998) 77–85. [35] Z. Schwartz, J.Y. Martin, D.D. Dean, J. Simpson, D.L. Cochran, B.D. Boyan, Effect of titanium surface roughness on chondrocyte proliferation, matrix production, and differentiation depends on the state of cell maturation, J. Biomed. Mater. Res. 30 (1996) 145–155. [36] K. Anselme, M. Bigerelle, B. Noel, E. Dufresne, D. Judas, A. Iost, Hardouin P., Qualitative and quantitative study of human osteoblast adhesion on materials with various surface roughnesses, J. Biomed. Mater. Res. 49 (2000) 155–166. [37] R.G. Richards, The effect of surface roughness on fibroblast adhesion in vitro, Inj.-Int. J. Care Injured 27 (1996) 38–43. [38] J.Y. Martin, Z. Schwartz, T.W. Hummert, D.M. Schraub, J. Simpson, J. Lankford, D.D. Dean, D.L. Cochran, B.D. Boyan, Effect of titanium surface-roughness on proliferation, differentiation, and protein-synthesis of human osteoblast-like cells (Mg63), J. Biomed. Mater. Res. 29 (1995) 389–401. [39] E. Eisenbarth, P. Linez, V. Biehl, D. Velten, J. Breme, H.F. Hildebrand, Cell orientation and cytoskeleton organisation on ground titanium surfaces, Biomol. Eng. 19 (2002) 233–237.

[40] K.D. Chesmel, C.C. Clark, C.T. Brighton, J. Black, Cellular-responses to chemical and morphologic aspects of biomaterial surfaces 2. The biosynthetic and migratory response of bone cell-populations, J. Biomed. Mater. Res. 29 (1995) 1101–1110. [41] M. Wieland, P. Hanggi, W. Hotz, M. Textor, B.A. Keller, N.D. Spencer, Wavelengthdependent measurement and evaluation of surface topographies: application of a new concept of window roughness and surface transfer function, Wear 237 (2000) 231–252. [42] M. Bigerelle, K. Anselme, Statistical correlation between cell adhesion and proliferation on biocompatible metallic materials, J. Biomed. Mater. Res. A 72A (2005) 36–46. [43] X.X. Wang, S. Hayakawa, K. Tsuru, A. Osaka, A comparative study of in vitro apatite deposition on heat-, H2O2-, and NaOH-treated titanium surfaces, J. Biomed. Mater. Res. A 54 (2001) 172–178. [44] C. Aparicio, J.M. Manero, F. Conde, M. Pegueroles, J.A. Planell, M. Vallet-Regí, F.J. Gil, Acceleration of apatite nucleation on microrough bioactive titanium for bone replacing implants, J. Biomed. Mater. Res. A 82 (2007) 521–529. [45] T.B. Kardos, Cellular responses to metal ions released from implants, J. Oral Implantol. 40 (3) (2014) 294–298. [46] M.J. Stoddart, Cell viability assays: introduction, Methods Mol. Biol. 740 (2011) 16–24. [47] M.A. Martin-Piedra, I. Garzon, A. Celeste, C.A. Alfonso-Rodriguez, M.C. SanchezQuevedo, A. Campos, M. Alaminos, Average cell viability levels of human dental pulp stem cells: an accurate combinatorial index for quality control in tissue engineering, Cytotherapy 23 (2013) 507–518. [48] V. Vijayaraghavan, A.V. Sabane, Hypersensitivity to titanium: a less explored area of research, J. Indian Prosthodont. Soc. 12 (4) (2012) 201–207. [49] A.C. Oliveira, I.A. Rodríguez, I. Garzón, M.A. Martín-Piedra, A. Rodríguez, J.M. García, M.C. Sánchez-Quevedo, M. Alaminos, An early and late cytotoxicity evaluation of lidocaine on human oral mucosa fibroblasts, Exp. Biol. Med. 239 (1) (2014) 71–82. [50] I.S. Yeo, J.S. Han, J.H. Yang, Biomechanical and histomorphometric study of dental implants with different surface characteristics, J. Biomed. Mater. Res. B Appl. Biomater. 87 (2008) 303–311. [51] A. Piattelli, L. Manzon, A. Scarano, M. Paolantonio, M. Piattelli, Histologic and histomorphometric analysis of the bone response to machined and sandblasted titanium implants: an experimental study in rabbits, Int. J. Oral Maxillofac. Implants 13 (1998) 805–810. [52] A. Wennerberg, T. Albrektsson, On implant surfaces: a review of current knowledge and opinions, Int. J. Oral Maxillofac. Implants 25 (2010) 63–74. [53] A. Wennerberg, T. Albrektsson, Effects of titanium surface topography on bone integration: a systematic review, Clin. Oral Implants Res. 20 (Suppl. 4) (2009) 172–184.