Bioactivity and hemocompatibility study of amorphous hydrogenated carbon coatings produced by pulsed magnetron discharge

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ Esta es la versión aceptada del artículo publicado en: This is a accepted manuscript of a paper published in: Journal of Biomedical Materials Research Part A (2013) n.101A DOI: 10.1002/jbm.a.34489 Copyright: 1999-2024 John Wiley & Sons El acceso a la versión publicada del artículo puede requerir la suscripción de la revista. Access to the published version may require subscription. “This is the peer reviewed version of the following article: Lopez-Santos, C.; Colaux JL, Laloy J; Fransolet, M; Mullier, F; Michiels, C; Dogné J-M, Lucas, S. 2013. Bioactivity and hemocompatibility study of amorphous hydrogenated carbon coatings produced by pulsed magnetron discharge. J Biomed Mater Res Part A (2013):101A:1800–1812 which has been published in final form at https:// doi.org/ 10.1002/jbm.a.34489. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions. This article may not be enhanced, enriched or otherwise transformed into a derivative work, without express permission from Wiley or by statutory rights under applicable legislation. Copyright notices must not be removed, obscured or modified. The article must be linked to Wiley’s version of record on Wiley Online Library and any embedding, framing or otherwise making available the article or pages thereof by third parties from platforms, services and websites other than Wiley Online Library must be prohibited." 1 BIOACTIVITY AND HEMOCOMPATIBILITY STUDY OF AMORPHOUS HYDROGENATED CARBON COATINGS PRODUCED BY PULSED MAGNETRON DISCHARGE C. Lopez-Santos 1 , J.L. Colaux 1 , J. Laloy 2 , M. Fransolet 3 , F. Mullier 2 , C. Michiels 3 , J.-M. Dogné 2 , S. Lucas 1 1) Namur Research Institute for Life Sciences (NARILIS), Research Centre for the Physics of Matter and Radiation (PMR), University of Namur (FUNDP), Rue de Bruxelles, 61, B-5000 Namur, Belgium 2) Namur Research Institute for Life Sciences (NARILIS), Namur Thrombosis and Hemostasis Center (NTHC), Department of Pharmacy, University of Namur (FUNDP), Namur, Belgium 3) Namur Research Institute for Life Sciences (NARILIS), Research Unit on Cell Biology (URBC), University of Namur (FUNDP) Rue de Bruxelles, 61, B-5000 Namur, Belgium Abstract Litterature contains very few data about the potential biomedical application of amorphous hydrogenated carbon (a-C:H) thin films deposited by reactive pulsed magnetron discharge even so it is one of the most scalable plasma deposition technique. In this paper, we show that such a C 2 H 2 pulsed magnetron plasma produces high quality coating with good hemocompatibility and bioactive response: no effect on hemolysis and hemostasis were observed, and proliferation of various cell types such as endothelial, fibroblast and osteoblast-like cells was not affected when the deposition conditions were Page 1 of 43 Journal of Biomedical Materials Research: Part A 2 varied. Cell growth on a-C:H coatings is proposed to take place by a two-step process: the initial cell contact is affected by the smooth topography of the a-C:H coatings, whereas the polymeric-like structure, together with a moderate hydrophilicity and a high hydrogen content, directs the posterior cell spreading while preserving the hemocompatible behavior. Keywords: cytocompatibility, hemocompatibility, a-C:H coating, polymeric-like structure, C 2 H 2 magnetron discharge Introduction Amorphous hydrogenated carbon (a-C:H) thin films are interesting materials due to the singular structure composed of carbon bonded in sp 2 and sp 3 electronic configurations and hydrogen 1 . Firstly, their special properties, such as a good hardness, high wear resistance, chemical inertness and low friction coefficient make them very useful as protective coatings. Moreover, good bio-tolerance together with the absence of cytotoxicity make them suitable for biomedical applications 2 . As a result, medical devices such as medical guidewires, surgical needles, intraocular lenses, coronary stents and heart valves are covered with a-C:H coatings 3 . Other similar thin films like diamondlike coatings (DLC) are commonly used in orthopaedic devices as hip, vascular and dental prostheses 4-9 , showing the versatility of these carbon-based materials. There are several methods for the deposition of a-C:H coatings: plasma-enhanced chemical vapour deposition 10 , ion beam deposition 11 or pulsed laser ablation 12 . Recently, the pulsed reactive magnetron process has attracted a high interest in many industrial Page 2 of 43Journal of Biomedical Materials Research: Part A 3 applications 13,14 . The versatility of this deposition method allows to fine tune the chemical composition of the a-C:H coatings that is usually required to ensure a good cell growth. For example, while fibroblast cells do not require special tuning to proliferate on a-C:H material deposited by laser or plasma techniques 15,16 , osteoblast cell growth is improved by the incorporation of elements like titanium 3,17 , boron 18 or silicon 19 . Moreover, good bioactivity response of endothelial cells has been observed after doping with nitrogen or silicon of such material grown by plasma systems 3,20-23 . Biocompatibility is not only dependent on the chemical composition since other physicochemical properties such as the wettability, polarity or roughness are also of importance 3,6 . For example, an hydrophilic surface seems to enhance the adsorption of several cell-adhesion mediating proteins, hence cell adhesion itself 3 and the surface polarity can improve cell proliferation 24-26 . Surface topography is one of the critical factors because it has an influence on cell-material interactions. Thus, in-vitro studies have shown that osteoblast-like cells attach preferentially to surfaces with a rougher microtopography 27 . Therefore, the bioactive response of the material may be dependent on the cell line as the same surface properties can be an advantage or a problem for the cell growth and proliferation. This fact could limit the potential biomedical applications of the a-C:H coating. If one thinks to use a-C:H coatings for potential applications involving contact with blood, it must fulfil at least three criteria to be hemocompatible: no destruction of the red blood cells (RBCs) (hemolysis), no or limited impact on platelet function (adhesion, activation and aggregation) and no induction on coagulation (secondary hemostasis). It is usually accepted that the main parameters affecting the hemocompatibility of a coating is Page 3 of 43 Journal of Biomedical Materials Research: Part A 4 the wettability of the surface, modulated by the surface chemical composition and roughness. Indeed, hydrophilic coatings (very low water contact angle) favour the adsorption of albumin against fibrinogen, which is suitable for the inhibition of thrombus formation 28-30 . Particularly, hydrophilic polymeric-like materials have also shown nonthrombogenic properties 31 . But literature shows that samples partially hydrophilic, with water contact angles between 56-90°, presented the best hemocompatible behavior 32,33 . Other contradictory examples are also found: non-thrombogenic properties have been observed for hydrophobic diamond-like carbon coatings doped with fluorine or silicon 31 . Therefore, the hemocompatibility of a-C:H coatings is not only dependent on the surface wettability. Especially, platelet activation can be also affected by the morphology 29 and, more recently, it has been proposed that the level of carbon sp 3 /sp 2 hybridization of doped a-C:H layers has an influence on the hemocompatibility through the platelet adhesion 34 . This work aims at the evaluation of the reactive magnetron sputtering system to deposit bioactive a-C:H thin films. Literature contains only a few biocompatible and hemocompatible data on the properties of a-C:H layer grown by this technique, despite the fact that it is probably one of the cheapest involving plasma and with the best industrial scalability. Therefore, we present here a detailed analysis of the in-vitro hemocompatibility (effects on hemolysis, platelet function and coagulation) and bioactive responses of endothelial, fibroblast and osteoblast-like cells on the a-C:H coatings. Finally, the characterization of the physico-chemical properties of these a-C:H coatings is studied to contribute to the understanding of the material behavior and the definition of the suitable properties for a potential biomedical application. Page 4 of 43Journal of Biomedical Materials Research: Part A 5 Material and methods Coating deposition and characterization The reactor consists of a vacuum chamber (<10 -4 Pa of residual pressure) equipped with two confocal pulsed magnetron sputtering sources loaded with 2” graphite targets (99.9% purity). A reactive mixture of argon, acetylene and hydrogen gasses at a ratio of 20:4:1 sccm was used to generate the reactive plasma. Different values of power density from 7 to 21 W cm -2 were applied simultaneously on both targets (one power supply per magnetron) in current mode configuration. The pressure was maintained at 0.4 Pa during 50 min of deposition, whereas the pulse-on and pulse-off times were both set to 50 µs. Well cleaned glass coverslips of 13mm diameter, through ultrasonic baths with bidistilled water and washing with ethanol, were used as substrates. The a-C:H composition depth profiles were determined by ion beam analysis combining Elastic Recoil Detection Analysis (ERDA) and Rutherford Backscattering Spectroscopy (RBS), using an impinging beam of 4 He at 2.2 MeV and a posterior spectra fitting with SIMTarget 35 and SIMNRA 36 codes. X-ray photoelectron spectroscopy (XPS) study was performed in an SSX-100 spectrometer (Surface Science Instruments) with an Al K α radiation at 1486.6 eV as excitation source. Raman spectroscopy was carried out by a LabRAM HR High Resolution 800 UV Confocal Raman microscope with a green laser (He-Ne 532.14 nm), in backscattering configuration. Thickness of the a-C:H layer was measured with a stylus profilometer from Veeco Instruments (Surface Profile Measuring System Dektak). Surface topography was analyzed by a Multimode 8 AFM (Veeco diNanoScope V), working in tapping mode, with Si tips of 350-380 kHz resonant frequency and 8-20 N/m spring constant. Page 5 of 43 Journal of Biomedical Materials Research: Part A 6 Tribological analysis was done in a Hysitron Inc. Triboindenter system with the load and displacement resolutions of 0.1 µN and 0.1 nm, respectively. Hardness and reduced Young’s modulus were estimated using a Berkovich indenter with a face angle of 65.3°. Wettability study was performed through static water contact angle measurements with OCA20 equipment and SCA20 software (Dataphysics), according to the Young’s method 37 . Bidistilled water sessile drops of 1 µl were employed. Young’s method relates the angle of a liquid droplet in contact to a solid surface with the interfacial tensions at equilibrium. The reported values are an average of ten measurements taken for each examined surface. To estimate the surface tension of the a-C:H coating, diiodomethane contact angles were measured in a similar way, according to the Owens-Wendt-Kaelble approximation 38 . This model presents a surface tension composed of two contributions: one due to polar interactions and one due to non-polar or dispersive interactions between the liquid and the solid phases. Therefore, the measurement of contact angles with at least two liquids is required, where diiodomethane is chosen to emphasize the non-polar interactions unlike water 39,40 . Biocompatibility test - Cell culture and viability assay Samples were sterilized in an autoclave (20 min at 121 °C and 1,2 x10 5 Pa). Posterior XPS and wetting measurements have confirmed that there were no significant changes after sterilization: only a slight increase of 3 % of the oxygen surface concentration of aC:H and a slight increase in the water contact angle of about 3° were found. These variations can be considered within the error margins of the techniques. EAhy926 human Page 6 of 43Journal of Biomedical Materials Research: Part A 7 endothelial cell line was seeded on the samples at a density of 10,000 cells/well in 0.5 ml of the culture medium Dulbecco’s Modified Eagle Medium (DMEM, Gibco) supplemented with 10% (v:v) fetal bovin serum (FBS, Gibco, US) in 24-well plates. Endothelial cells were maintained in culture at 37 °C under 5 % CO 2 and incubated from 1 to 3 days. At each day, the number of viable cells was determined with a MTT assay using Thiazolyl Blue Bromide MTT (2.5 mg/ml of [3-(4.5-dimethylthiazol-2-yl)-2.5diphenyltetrazolium bromide], Sigma-Aldricht, Germany, in Phosphate buffered saline (PBS), 6.7 mM phosphate, pH 7.4) which allows measuring the cell viability on a-C:H coatings. 500 µl of MTT solution were added to the medium for 2 h at 37 °C in a 5% CO 2 atmosphere. Formazan crystals are lysed in a "lysis buffer" composed of SDS 30% in H 2 O d + N,N-dimethyl-formamide 2:1 (v:v) at pH: 4.7. The optical absorbance was measured with a plate spectrophotometer (xMark TM , Bio-Rad, US) at 570 nm. MTT protocol is also applied to two other cell types. The human skin fibroblast AG04431 (Coriell Institute for Medical Research, USA) and the human MG-63 osteoblast-like cells (ATCC, USA), were respectively grown in Basal Medium Eagle (BME, Gibco, US) supplemented with 10 % FBS and 2 mM L-glutamine and in DMEM supplemented 10 % FBS. Statistical analysis was carried out using one-way analysis of variance (ANOVA), presenting the mean value from 3 repeats and the error bars referring to the standard error of the mean. In what follows, glass substrate was used as positive control because of the good cell adhesion and spreading on it. Meanwhile a copper (Cu) thin film deposited in the same system than a-C:H coatings was used as negative control because of its cytotoxicity. Page 7 of 43 Journal of Biomedical Materials Research: Part A 8 - Cell morphology staining Cell morphology was observed after cell protein staining with Coomassie Blue. Cells were washed twice with PBS, and fixed with 3.7% paraformaldehyde (PFA) in distilled water for 10 min at room temperature (RT). Subsequently, they were permeabilized with 70% ethanol in distilled water for 10 min at RT and dried at 37 °C. The cells were finally stained with Coomassie Blue for 30 s, washed with distilled water and dried at 37 °C, before being observed with an inverted microscope (Labovert FS, Leitz). In order not to repeat similar obtained information, only EAhy926 endothelial cells morphology staining is presented in the next Results section. - Cytoskeleton analysis Cytoskeleton morphology was studied in cells by immunofluorescence staining. As example of the similar results, we presented only the images corresponding to the EAhy926 endothelial cells. Cells were fixed with 4 % PFA for 10 min at 37 °C. Subsequently, the samples were washed with PBS and permeabilized with 1 % Triton-X100 (Sigma-Aldricht, Germany). Actin was stained with Alexa Fluor 546 phalloïdin (Molecular Probes) and nuclei with To-Pro-3 (Molecular Probes) diluted (1:80) in PBSRNase (2 mg/ml). The samples were again washed three times before being embedded in Mowiol (Aldrich) and photographed with a confocal microscope TCS (Leica SP1) using a 409 objective and a constant photomultiplier. - Inflammatory cytokine secretion Page 8 of 43Journal of Biomedical Materials Research: Part A 15 Cell growth analysis Cell adhesion was evaluated at the 24h after seeding on the a-C:H coatings. Coomassie blue staining, presented on the endothelial cells as example, corroborated the good attachment and cell spreading on the a-C:H surface (Fig. 3). Furthermore, actin cytoskeleton morphology was observed 24 h after seeding of the endothelial cells on the different coatings (Fig. 4). Both on the positive control and on the different a-C:H coatings, cells presented a well spread with a well-structured cytoskeleton with the development of actin fibers, establishing an organized fibers meshwork distributed throughout the cell body and preferentially oriented along the main nucleus axis. Cell proliferation was followed through MTT assay 2 and 3 days after seeding. The small values are due to the fact that a low cell density was seeded due to the small size of the samples, trying to avoid the cell confluence at 3 days of culture. The a-C:H thin films did not induce any detrimental or toxic effect for EAhy926 endothelial, AG04431 skin fibroblasts, or MG63 osteoblast-like cells at 2 days of culture, as can be observed in Fig. 5. The results demonstrate that there were as much viable cells on a-C:H coatings as on the positive control, while cells seeded on the negative control were affected by toxicity. The a-C:H coatings had not statistical effects compared to the positive control. Thus, the optical density values from the a-C:H coatings respect to those corresponding to the positive control, taking account the statistical error bars, are very similar. In the worst cases, they represent more than the 85% of the positive control result, what is considered a good bioactive response. Also, well-spreading of the cells on the a-C:H surface was observed, as represented the results from EAhy926 endothelial after 3 days of seeding, similar to those from the other cultured cells. Cell morphology on the a-C:H surface was Page 15 of 43 Journal of Biomedical Materials Research: Part A 16 similar to the one observed for the positive control and opposite to the negative control, as shown in Figs. 6 - 8. Indeed, there was no survival of any of the three types of cells when seeded on the negative control surface. Additionally, IL-8 production is known to be caused by cells undergoing on inflammatory response 48 . As observed in Fig. 9, cells seeded on the different coatings did not produce more IL-8 than the cells seeded on glass (positive control). This indicates that no inflammation response was taking place on the a-C:H coatings. Hemocompatibility test The effect of a-C:H coating deposited at the intermediate power density, on the lysis of human RBC, is presented in Table 3. The a-C:H coating as well as the other negative controls showed insignificant hemolytic effect (< 5 % in comparison to the positive control) 49 . Primary hemostasis analysis through the study of platelet function (adhesion, activation and aggregation) in Fig. 10A reveals the adhesion of only few platelets on the a-C:H thin film. Adherent platelets remained not activated on the a-C:H surface and kept their conformation (absence of pseudopodia). No platelet aggregation was observed. Fig. 10B shows the results of the cTGT on the a-C:H coating according to 41 , compared to negative and positive controls. After 1 h of incubation without 4 µM of PL inducer, first traces of active thrombin appeared in NPP after a lag time of 17 min. The C max was around 70 nM and was used as reference for the following results. Conversely, glass sample used as positive control revealed a remarkable increase in the catalytic activity of thrombin (increase of C max to 148 nM and reduction of lag time to 13 min), indicating Page 16 of 43Journal of Biomedical Materials Research: Part A 17 activation of coagulation. The thrombin generation profile of the a-C:H thin film appeared very similar to the one of the negative control (i.e. silicon) and to the NPP plasma curves. Discussion The special physico - chemical properties of a-C:H coatings have previously been associated to a good protective function, raising the interest in potential biomedical applications, as presented in the introduction section. Recently, the reactive magnetron process has been recognized as a well established production system of a-C:H coatings in many industrial fields. Therefore, a-C:H coatings deposited by reactive magnetron sputtering for a biomedical function must be biocompatible as well as hemocompatible to ensure good living cells - material interactions. Moreover, the possibility to obtain different surface properties of a-C:H coatings makes interesting the analysis of the influence of parameters such as the chemical composition, structure and topography in their bioactive response. We showed that the present a-C:H coatings contain more than 30 at. % of hydrogen and exhibit a I D /I G ratio between 0.77 - 0.92 and a very low roughness. The a-C:H coatings present a moderate hydrophilic character (water contact angles around 40 - 74°) and the surface-tension stays below 62 mN/m. The bioproperties of such surfaces are a good cell proliferation (fibroblasts, endothelial cells and osteoblastlike cells) and hemocompatibility (no hemolysis, prevention of platelet adhesion/activation/aggregation, and prevention of thrombin generation). Our discussion of the relationship between the bioactive response and the physico-chemical properties Page 17 of 43 Journal of Biomedical Materials Research: Part A 18 will be presented in light of: i) wettability and surface tension, ii) surface topography and iii) surface structure. I) Wettability and surface tension The wetting behavior of the material has already been related to the cell proliferation. For example, Van Wachem et al. 50 have shown that a moderate hydrophilicity of polymeric surfaces improved the adsorption of serum proteins, affecting positively the cell attachment and spreading. Crespin et al. 48 have confirmed that endothelial cell density was higher on a-C:H coatings with water contact angle values between 40 - 75°. This tendency is in agreement with the wettability observed for the present a-C:H coatings. Polar oxygen species in the a-C:H coating have been suggested to be responsible for the induced hydrophilicity. The presence of oxygen polar groups on the a-C:H surface would facilitate cell adhesion through ionic interactions with functional groups such as NH 2 and COOH from cell proteins, that are positively or negatively charged at physiological pH, respectively 43,44 . These oxygen polar groups affect the polar component value of surface tension of the a-C:H coatings, around 4 - 23 mN/m, and consequently the polarity. Hallab et al. 51 have also proposed that polarity values below 0.5 are beneficial for the cell adhesion, which is the case for the a-C:H coatings as described in the Table 2. Thus, results from Tables 1 and 2 show that the a-C:H coatings exhibit a change in the water contact angle and in the polar component of the surface tension without a change in the in-depth oxygen content but only in the surface composition, the increase of the power density facilitating the generation of oxygen-containing groups at the surface of the coating. Page 18 of 43Journal of Biomedical Materials Research: Part A 19 Another contribution to take into account is the reduction of the water contact angle due to the presence of transpolyacetylene chains in the microstructure 52 , since they favor the incorporation of polar oxygen functional groups to the surface. Moreover, the influence of the roughness on the wettability can be observed from Fig. 2 and Table 2, where a lower water contact angle is associated with a rougher surface. Finally, Fig. 11A tries to correlate the cell viability and the hydrophilicity and, indirectly the surface oxidization of the a-C:H coatings. Wettability as well as surface tension and surface oxygen content provided the same relationships to the viability of the three cell lines. However, statistical error of the cell viability results does not allow to consider clear tendencies between the growth of each cell line on the samples and their hydrophilicity. Only for the case of MG63 cells could be thought about an improvement of the cell viability on a surface with a lower water contact angle. It has also been shown that wettability and surface tension have also a strong influence on the hemocompatibility of a-C:H thin films. A low hemolysis rate has been related to a hydrophilic surface that would improve the compatibility to biofluids 29,53 . When blood enters in contact with the a-C:H coating, plasma proteins are adsorbed on the surface and platelets adhered to the plasma-protein coated material. Fibrinogen plays a major role in platelet aggregation and fibrin formation in coagulation, and subsequently in the formation of thrombus. Roy et al. 29 have mentioned that when the electrostatic interaction between a negatively charged hydrophilic surface and the positively charged domains of fibrinogen is weak, there is a reduction of the fibrinogen adsorption and the platelet adhesion. Also, it has been found that platelets were less attached on hydroxyl-containing polymer surfaces 28 . However, other works have presented a low hemolysis rate in Page 19 of 43 Journal of Biomedical Materials Research: Part A 20 hydrophobic polymers 54 . In similar materials like DLC coatings, water contact angles around 60 - 70° resulted in the reduction of the plasma proteins adsorption. Thus, an interfacial tension between the coating and albumin lower than between the coating and blood or fibrinogen, reduces the degradation of protein structure and the activation of blood coagulation mechanism 30 . II) Surface topography In general, it is known that roughness promotes a good effect on the cell anchorage on polymeric materials 51,55,56 . Nevertheless, fibroblasts adhered better on smoother surfaces 2,57 . According to the present results, it is not possible to think about a direct correlation between the increase in the cell viability and the roughness of the a-C:H coatings as Fig. 11B shown. At the low levels of smoothening shown by the a-C:H surfaces, cell viability is practically independent for any cell line. Even if a better MG63 cell viability could be suspected in rougher samples, the statistical error cannot ensure it. On the contrary, surface topology, i.e. the shape of topographic features, has an influence on the cell adhesion 48,58,59 . Typical dimensions of biocompatible nanostructures have been found to have a height of about 10 nm and a lateral spacing below 75 nm 60 . For example, surfaces with topographic features of low dimension (less than 100 nm) are suitable for the adhesion of osteoblast cells, since they can mimic the structure of natural tissues of proteins and extracellular parts (with dimensions around tens of nanometers) 3 . The present results have showed the generation of features on the a-C:H surface with dimensions in a similar range, justifying the generalized good bioactive response. Smooth topographies have also improved the hemocompatible behavior, through the plasma proteins absorption. The size of human serum albumin molecule is around 3x8 Page 20 of 43Journal of Biomedical Materials Research: Part A 21 nm 2 , whereas for the fibrinogen molecule is of 7x46 nm 2 . For this reason, surface feature dimensions more similar to the albumin size were found to favor its absorption rather than the one of fibrinogen 61 . III) Surface structure This work links the cell spreading on a-C:H coatings deposited by reactive pulsed magnetron sputtering to the structure of the coating (I D /I G ratio) at the same level of importance than those of the wettability, the chemical composition or the topography. A previous work established such a relationship for N-doped DLC and Si-doped DLC coatings 21 . In that case, a high hydrogen concentration together with a low I D /I G value (0.2 - 0.5) had a good influence on endothelial cell adhesion. At the present case, the I D /I G ratio should be in the range of 0.77 - 0.92 to get a good cell proliferation on a-C:H coatings without any additional chemical element, as Fig. 11C indicates. It seems that a higher I D /I G ratio, i. e. a minor carbon sp 3 bonds concentration or a more ordered a-C:H structure, favor the proliferation of MG63 cells, unlike Eahy926 cells. However, taking account the margin of statistical variation, it is not possible to extract a clear conclusion about tendencies between the cell viability and the a-C:H structure. Recently, endothelial cell adhesion by a two-step process has been proposed 62 : in the initial cell contact, the topography is the main factor of influence, whereas the surface chemistry governs the following cell spreading. An extrapolation can be made for the fibroblast and osteoblast-like cell growth on the deposited a-C:H coatings. This two steps process is an interesting concept, but has to be slightly modified in the view of our findings: good cytocompatibility results have been obtained with different levels of hydrophilicity, surface roughness and chemical composition. Therefore, we propose that Page 21 of 43 Journal of Biomedical Materials Research: Part A 22 in the two step process, cell spreading is governed by the polymeric-like structure of our coatings, described by the level of sp 3 /sp 2 carbon hybridization. Material structure of the a-C:H coatings should have a stronger influence in the cell proliferation in order to level the effects of the different surface chemistry and roughness. Of course, the moderate hydrophilicity, through the oxygen incorporation, and the high hydrogen content are parameters that predispose the surface to a good bioactive response. This proposition is also suitable for hemocompatibility. Indeed, the sp 3 concentration of the present a-C:H coatings is similar to those presented by Logothetidis et al. 61 who observed a high absorption of albumin versus fibrinogen. Hemocompatibility of this type of films became worse when sp 3 /sp 2 ratio increased above 1.1 or the I D /I G ratio was higher than 1.6 30,63 . This is in line with our suggestion that a sp 3 /sp 2 ratio around 0.8 - 0.9 is suitable for cell spreading and to prevent thrombus generation on the a-C:H surface. Conclusion A good bioactive response of a-C:H thin films deposited by reactive pulsed magnetron sputtering with endothelial, fibroblast and osteoblast-like cells have been obtained. At the same time, the a-C:H coatings were hemocompatible, without any hemolytic and hemostastic effects. Surface chemical composition, roughness as well as wettability changes are important parameters, but cannot solely explain the observations. For this reason, we proposed that cell proliferation occurs by a two-step process: in the initial cell contact, the smooth topography is the main factor of influence, whereas the polymericlike character, together with the surface chemistry (moderate hydrophilicity through Page 22 of 43Journal of Biomedical Materials Research: Part A 23 oxygen incorporation and high hydrogen content), governs the following cell spreading and preserves the hemocompatibility. Acknowledgements This work was partially supported by the Tribofutur project (Belgian Walloon Region Funding), the REGPOT-CT-2011-285895-AL-NANOFUNC European project and by the CERUNA funding of the University of Namur. We thank Pf. D. Bonifazi and Pf. B.-L. Su from the Dept. of Chemistry, University of Namur (FUNDP), for the access to the AFM and contact angle equipments respectively, Pf. M.P. Delplancke-Ogletree from the Dept. of Industrial Chemistry, Free University of Brussels (ULB), for the access to the nanoindenter and Dr. J.C. Gonzalez from the Institute of Materials Science of Seville (ICMS) for the Raman measurements. Page 23 of 43 Journal of Biomedical Materials Research: Part A 24 References 1. Robertson J. Diamond-like amorphous carbon. Materials Science & Engineering R-Reports 2002;37(4-6):129-281. 2. Rodil SE, Olivares R, Arzate H, Muhl S. Properties of carbon films and their biocompatibility using in-vitro tests. Diamond and Related Materials 2003;12(37):931-937. 3. Vandrovcova M, Bacakova L. Adhesion, Growth and Differentiation of Osteoblasts on Surface-Modified Materials Developed for Bone Implants. Physiological Research 2011;60(3):403-417. 4. Hauert R, Muller U. An overview on tailored tribological and biological behavior of diamond-like carbon. Diamond and Related Materials 2003;12(2):171-177. 5. Dorner-Reisel A, Gartner G, Reisel G, Irmer G. Diamond-like carbon films for polyethylene femoral parts: Raman and FT-IR spectroscopy before and after incubation in simulated body liquid. Analytical and Bioanalytical Chemistry 2008;390(6):1487-1493. 6. Ma WJ, Ruys AJ, Mason RS, Martin PJ, Bendavid A, Liu ZW, Ionescu M, Zreiqat H. DLC coatings: Effects of physical and chemical properties on biological response. Biomaterials 2007;28(9):1620-1628. 7. Louro C, Moura CW, Carvalho N, Stueber M, Cavaleiro A. Thermal stability in oxidative and protective environments of a-C:H cap layer on a functional gradient coating. Diamond and Related Materials 2011;20(2):57-63. 8. Meunier C, Stauffer Y, Daglar A, Chai F, Mikhailov S, Hildebrand HF. Comparison of hydrogenated and unhydrogenated carbon films obtained by FCVA onto Ti6Al4V: Structure, hardness and biocompatibility study. Surface & Coatings Technology 2006;200(22-23):6346-6349. 9. Hang RQ, Ma SL, Chu PK. Corrosion behavior of DLC-coated NiTi alloy in the presence of serum proteins. Diamond and Related Materials 2010;19(10):12301234. 10. C. Casiraghi FP, A.C. Ferrari, D. Grambole, J. Robertson Diamond & Related Materials 2005;14. 11. A Ibenskas AG, S Meskinis, M Andrulevicius, S Tamulevicius. Diamond & Related Materials 2011;20. 12. J Eskusson RJ, E Lust. Applied Surface Science 2009;255. 13. Czyzniewski A. Surface & Coatings Technology 2009;203. 14. M Keunecke KW, K Bewilogua, R Cremer, H-G Fuss. Thin Solid Films 2009;518. 15. Jelinek M, Smetana K, Kocourek T, Dvorankova B, Zemek J, Remsa J, Luxbacher T. Biocompatibility and sp(3)/sp(2) ratio of laser created DLC films. Materials Science and Engineering B-Advanced Functional Solid-State Materials 2010;169(1-3):89-93. 16. Singh A, Ehteshami G, Massia S, He JP, Storer RG, Raupp G. Glial cell and fibroblast cytotoxicity study on plasma-deposited diamond-like carbon coatings. Biomaterials 2003;24(28):5083-5089. Page 24 of 43Journal of Biomedical Materials Research: Part A 31 List of figure captions Fig. 1.- A) XPS spectra of the a-C:H coating deposited by reactive pulsed magnetron sputtering (12W cm -2 ); B) Raman spectra of the a-C:H coating deposited by reactive pulsed magnetron sputtering (12W cm -2 ). Fig. 2AFM topography images (1µmx1µm) of the control samples used in the different cytocompatibility and hemocompatibility tests: A) glass coverslip, B) copper thin film, C) silicon thin film, and of the a-C:H coatings at different applied power density values and thickness: D) 7 W cm -2 / 200nm, E) 12 W cm -2 / 260nm, F) 21 W cm -2 / 454nm. Height scale is indicated in the grey color scale and the roughness is represented as the mean square roughness coefficient R ms . Fig. 3.- Morphology study of EAhy926 endothelial cells adhesion on the a-C:H coatings analyzed by Coomassie Blue staining and optical microscopy 24 h after cell seeding. Results are compared to those of the positive (glass) and negative (Cu) controls. Fig. 4.- Actin cytoskeleton analysis of the EAhy 926 endothelial cells growth on the aC:H coatings. Nuclei and actin stress fiber formation were observed in blue and red, respectively. Results are compared to those of the positive (glass) and negative (Cu) controls. Fig. 5.- A) MTT viability test at 2 and 3 days after EAhy926 endothelial cell seeding on the deposited a-C:H coatings; B) MTT viability test at 2 days after fibroblast (FS) and MG63 osteoblast-like cell seeding on the deposited a-C:H coatings. Results are expressed as mean ± 1 SD (n = 3). *; **; ***: p<0.05; p<0.01; p<0.001 versus glass. Fig. 6.- Cell morphology of EAhy926 endothelial cells seeded on the a-C:H thin deposited with different applied power densities, after 3 days of cell culture. Images were Page 31 of 43 Journal of Biomedical Materials Research: Part A 32 taken in phase contrast microscopy. Results are compared to those of the positive (glass) and negative (Cu) controls. Fig. 7.- Cell morphology of fibroblasts seeded on the a-C:H thin deposited with different applied power densities, after 3 days of cell culture. Images were taken in phase contrast microscopy. Results are compared to those of the positive (glass) and negative (Cu) controls. Fig. 8.- Cell morphology of MG63 osteoblast-like cells seeded on the a-C:H thin deposited with different applied power densities, after 3 days of cell culture. Images were taken in phase contrast microscopy. Results are compared to those of the positive (glass) and negative (Cu) controls. Fig. 9.- Evaluation of the pro-inflammatory IL-8 cytokine secretion by EAhy926 endothelial cells cultured for 48 h on the a-C:H coatings, compared to the positive (glass) and negative (Cu) controls. Results are expressed in pg of IL-8/µg of proteins, as mean ± 1 SD (n = 3). *; **; ***: p<0.05; p<0.01; p<0.001 versus glass. Fig 10.- A) SEM images of the platelets on the a-C:H coating (12W cm -2 ). Scale bares = 10 and 1 µm. On the left, homogenous repartition of platelets on the surface; on the right, view of adherent platelets; B) Calibrated thrombin generation test on the a-C:H coating, compared to a positive control (glass), a negative control (Si) and the NPP plasma, using the contact pathway thrombin inducer. Curves represent the mean of three individual measurements. Fig 11.- Cell viability results at 2 days of cell culture of the a-C:H coatings in function of: A) hydrophilicity (WCA), B) roughness (R MS ) and C) structure (I D /I G ratio). Page 32 of 43Journal of Biomedical Materials Research: Part A 209x297mm (150 x 150 DPI) Page 33 of 43 Journal of Biomedical Materials Research: Part A 209x297mm (150 x 150 DPI) Page 34 of 43Journal of Biomedical Materials Research: Part A 209x297mm (150 x 150 DPI) Page 35 of 43 Journal of Biomedical Materials Research: Part A 209x297mm (150 x 150 DPI) Page 36 of 43Journal of Biomedical Materials Research: Part A 209x297mm (150 x 150 DPI) Page 37 of 43 Journal of Biomedical Materials Research: Part A 209x297mm (150 x 150 DPI) Page 38 of 43Journal of Biomedical Materials Research: Part A 209x297mm (150 x 150 DPI) Page 39 of 43 Journal of Biomedical Materials Research: Part A 209x297mm (150 x 150 DPI) Page 40 of 43Journal of Biomedical Materials Research: Part A