Synergistic Effect of rhBMP-2 Protein and Nanotextured Titanium Alloy Surface to Improve Osteogenic Implant Properties

One of the major limitations during titanium (Ti) implant osseointegration is the poor cellular interactions at the biointerface. In the present study, the combined effect of recombinant human Bone Morphogenetic Protein-2 (rhBMP-2) and nanopatterned Ti6Al4V fabricated with Directed irradiation synthesis (DIS) is investigated in vitro. This environmentally-friendly plasma uses ions to create self-organized nanostructures on the surfaces. Nanocones (≈36.7 nm in DIS 80◦) and thinner nanowalls (≈16.5 nm in DIS 60◦) were fabricated depending on DIS incidence angle and observed via scanning electron microscopy. All samples have a similar crystalline structure and wettability, except for sandblasted/acid-etched (SLA) and acid-etched/anodized (Anodized) samples which are more hydrophilic. Biological results revealed that the viability and adhesion properties (vinculin expression and cell spreading) of DIS 80◦ with BMP-2 were similar to those polished with BMP-2, yet we observed more filopodia on DIS 80◦ (≈39 filopodia/cell) compared to the other samples (<30 filopodia/cell). BMP-2 increased alkaline phosphatase activity in all samples, tending to be higher in DIS 80◦. Moreover, in the mineralization studies, DIS 80◦ with BMP-2 and Anodized with BMP-2 increased the formation of calcium deposits (>3.3 fold) compared to polished with BMP-2. Hence, this study shows there is a synergistic effect of BMP-2 and DIS surface modification in improving Ti biological properties which could be applied to Ti bone implants to treat bone disease.


Introduction
Bone loss affects more than half a million patients in the United States and represents over $2.5 billion in health costs. Indeed, trauma, tumor recessions or developmental defects limit bone s ability to self-repair after an injury, creating large non-healing fragments that require the necessary use of implants and other medical devices [1]. Current treatments interstitials). The movement of these defects in the lattice may form complex structures or result in phase changes due to species accumulation in an area. Thus, ion irradiation may result in substantial chemical and morphological changes on the surface [30,34].
DIS shows multiple advantages compared to conventional nanopatterning techniques or new sintering methods, it is a fast process, reliable, with strong capacity to tune small nanofeatures (10 to 100 s nm) without the use of masks [29], high temperature [35] or toxic reagents [32,33]. In addition, this bottom-up technique has previously shown the capability to tailor nanofeatures on Ti surfaces keeping the bulk properties stable when using low fluences. This bioactive nanotopography has been shown to modulate cytoskeleton orientation and cell adhesion, as well as cell viability, guiding the tissue regeneration process [32,33].
Based on previous results in our group, in this study, we have characterized the surface properties of DIS-treated titanium samples irradiated using high fluences and different incidence angles and evaluated the effect of combining these active nanotopographies with effective biologics, such as BMP-2, to help elicit a cellular biological response, which may have a synergistic effect in promoting osteoblast differentiation in a BMP-2 responsive cellular model such as C2C12 cells [36]. To the best of our knowledge, this is the first study that does a systematic study using ion-induced surface patterning techniques in synergy with biologics compared with industry-leading anodized and sandblast surface treatment technologies.

Materials and Methods
For this purpose, we will first characterize the surface properties (topography, chemistry and wettability) ( Figure 1a) and evaluate the effect of the different samples on cell adhesion and spreading (number of filopodia, cell and nucleus shape, vinculin expression, total cells attached), cell viability by measuring cell metabolic activity, osteogenic differentiation and surface mineralization) (Figure 1b).

Titanium Sample Preparation
Titanium alloy samples (Ti4Al6V, area 0.25 cm 2 , Stryker, Kalamazoo, MI, USA) were polished to mirror finished and cleaned before DIS irradiation. The samples were grinded using 320, 1200 and 2400 sandpaper in an Ecomet III grinder (Buehler, Lake Bluff, IL, USA) then were mirror-polished using a ChemoMet cloth (Buehler, Lake Bluff, IL, USA) with a 0.05 µm silica solution (MasterMet, Buehler, Lake Bluff, IL, USA). Afterwards, the samples were ultrasonically cleaned in acetone, isopropanol, ethanol, and water for 15 min each. As controls sandblasted/acid-etched (SLA, commercially pure Ti grade IV, area 0.21 cm 2 ) and acid-etched/anodized (Anodized), area 0.25 cm 2 ) Ti6Al4V surfaces were used. SLA surfaces were provided by Tissue Engineering Group (TEG) of the Complutense University of Madrid, Spain. DIS samples were irradiated with argon ions at 1000 eV of energy and 1 × 10 18 ions/cm 2 using two different incidence angles: 60 • and 80 • degrees naming the samples DIS 60 • and DIS 80 • , respectively. All samples were ultrasonically cleaned and autoclaved before using them for the in vitro studies.

Surface Characterization of Titanium Samples
The surface topography was analyzed by scanning electron microscopy (SEM, S-4800, Hitachi, Tokyo, Japan) at two ranges of magnification: 10 k-22 k to detect microstructures and 70 k-100 k to detect submicron and nanostructures. The surface chemistry was examined by X-ray diffraction (XRD, PANalytical Phillips X pert MRD system #2, Malvern Panalytical, Malvern, United Kingdom) with Cu Kα radiation wavelength (λ = 0.15418 nm), generated at a voltage of 45 kV and a filament emission of 40 mA. Theta angle (2θ) was collected from 2θ = 30-80 • , with a step size of 0.02 • , and the analysis was performed with Origin and Jade software. The surface hydrophilicity was determined by contact angle (CA, Ramé-hart 250 Contact Angle Goniometer, Ramé-hart, Succasunna, NJ, USA) with DROPimage Advanced Software. The sessile method of CA analysis was employed, using 3 µL of deionized water drops to measure the CA of each sample; 4 samples per condition were used for the measurements.
2.3. In Vitro Cell Culture 2.3.1. C2C12 Cell Line C2C12 cells are multipotent cells that can differentiate, in addition to myotubes, into osteoblasts and adipocytes under specific culture conditions [37]. They have been widely used to study BMP-2 bioactivity since they display low basal BMP signaling activity and show good BMP-2 responsiveness. e.g., expressing bone differentiation markers (early phase such as alkaline phosphatase (ALP) or late-stage such as osteocalcin (OCN)), and inducing the formation of calcium nodules in ascorbic acid, β-glycerol phosphate and dexamethasone rich media [36,38,39]. Murine C2C12 myoblasts were purchased from the American Type Culture Collection (ATCC ® CRL-1772). This cell line was maintained in Dulbecco s Modified Eagle Medium (DMEM) (ATCC ® 30-2002™) supplemented with 10% Fetal Bovine Serum (FBS) (Invitrogen, Waltham, MA, USA) in a 37 • C, 5% CO 2 incubator. At 80% cellular confluence, cells were trypsinized in 0.25% Trypsin/EDTA (Invitrogen, Waltham, MA, USA). C2C12 passages <20 were cultured in Ti samples at densities of 60,000 cell/cm 2 (adhesion assay) or 100,000 cell/cm 2 (viability, osteogenic differentiation and mineralization assay) in 48 well-plates (Corning Costar) with DMEM media, 10% FBS and 3 µg/mL of BMP-2. The cells were initially cultured in 10 µL drops with BMP-2 on Ti surfaces for 4 h to promote cell attachment and then the rest of the media was added (490 µL).

Cellular Adhesion
To determine the combined effect of BMP2 and DIS surface treatment during initial cellular attachment of C2C12, cells were cultured as mentioned above on the different Ti surfaces (DIS 60 • , DIS 80 • , and polished, SLA, and Anodized as controls). After 4 h, Ti samples were prepared for SEM and confocal laser scanning microscopy. For SEM, the samples were washed twice with Phosphate-buffered saline (PBS) and fix with 2.5% glutaraldehyde (Sigma, St. Louis, MO, USA) overnight at 4 • C. Then, samples were dehydrated using an ethanol gradient in PBS (30%, 50%, 70%, 80%, 90% and 100%), each step for 15 min. Afterwards, they were critically point dried, sputter with gold-palladium and observed using a scanning electron microscope (JSM-6490LV, Jeol, Tokyo, Japan) at 2 k-3.5 k magnification. For confocal microscopy, the Ti samples were washed twice with PBS and fixed with 5% Formalin (Sigma, MO, USA) for 10 min, permeabilized with 0.1% Triton X-100 (Sigma, St. Louis, MO, USA) for 5 min, washed with 0.01% BSA/PBS and incubated for 30 min at room temperature with Texas red phalloidin (Invitrogen, Waltham, MA, USA) (1:75), DAPI (Thermofisher, Waltham, MA, USA) (1:1000) and Alexa 647 vinculin (Invitrogen, Waltham, MA, USA) (1:125) to stain actin filaments, nuclei and vinculin, respectively. Finally, the samples were examined via confocal microscopy (Leica SP8 Laser Confocal Microscope Microsystems, Wetzlar, Germany). The number of filopodia, vinculin intensity, total cells, cell area and nucleus area were quantified using FIJI software.
Cell Viability C2C12 cells were seeded on titanium discs as mentioned above. After 3 days, the Ti samples were incubated for 3 h with Alamarblue (Invitrogen, Waltham, MA, USA) following the manufacturer s recommendations. Briefly, media was removed to avoid counting unattached cells and fresh media with Alamarblue in a 1:10 ratio was added to the wells. Alamarblue is a resazurin-based solution, a cell-permeable compound that upon entering living cells is reduced to resorufin, a fluorescent compound. After incubating the samples in the dark for 3 h, the fluorescence signal was measured using a microplate reader (Synergy HT, BioTek, Winooski, VT, USA) at 530 nm/590 nm Ex/Em.

Evaluation of Cell Differentiation: Alkaline Phosphatase Activity
C2C12 cells were seeded on titanium discs as mentioned above. After 3 days, cells were washed twice with PBS, and lysed by the addition of 100 µL of buffer lysate and subjecting the samples to 3 cycles at −80 • C/37 • C for 30 min. ALP activity was measured according to the manufacturer's instructions using a commercial ALP kit (ab83369, Abcam, Cambridge, MA, USA). ALP hydrolyses phosphate esters in alkaline conditions, generating an organic radical and an inorganic phosphate. This ALP kit uses p-nitrophenyl phosphate (pNPP) as a phosphatase substrate which is dephosphorylated by ALP, generating a yellow compound (p-nitrophenol). The absorbance was measured at 405 nm on a microplate reader (Synergy HT, BioTek, Winooski, VT, USA). ALP enzyme activity was expressed U/L.

Evaluation of Cell Mineralization: Calcium Deposits Production
C2C12 cells were seeded on Ti discs in a 48 well-plate at 37 • C in a 5% CO 2 incubator at a density of 100,000 cell/cm 2  The mineralization percentage (%) was calculated by dividing the area covered by the red deposits to cell culture area times 100.

Statistical Analysis
All experiments were done in triplicates using two to three samples per condition, except for the confocal and SEM experiments where we examined 3-5 different fields of a sample. Analysis of variance (ANOVA) and Bonferroni s Multiple Comparison Test were used to determine statistically significant differences at 0.05 level of significance by using Origin lab and Graph Prism 5 software.

Evaluation of Surface Topography of Irradiated Titanium Samples
Ion irradiation transfers energy and momentum via ion-atom collisions. This results in erosive and diffusive regimes, which drive the surfaces to self-nanopatterned, generating surfaces with attractive topographies [40]. Previous work conducted on Ti alloy and pure titanium (porous scaffolds) revealed that changing the incidence angle, the nanopatterning process was governed by two regimes: diffusion and erosion. At normal incidence angle (0 • ) and low fluences (ion/cm 2 ), ion diffusive process predominates, generating short nanoripples and nanorod-like structures. Moving from 0 to 60 • , in small or low off-normal angles, there is a combination of diffusive and erosive regimes where ripples and nanorods are combined on the surface, increasing the uniformity of the surface treatment and length. At highly oblique off-normal angles (≥80 • ), an erosion regime predominates, in which ions from the source crash and sputter the atoms of the outmost layer of the surface. This complex process showed more elongated nanoripples and no nanorods [32,33]. In this study, we have observed that at higher fluences and off-normal incidence angles, nanoripples have grown in height, turning into nanowalls/nanocones. As observed in Figure 2, scanning electron microscopy (SEM) images of the studied specimens surface topography, DIS 60 • generated nanowalls homogeneously distributed on the surface of 16.5 ± 1 nm of thickness, and 41.2 ± 2.9 nm of spacing distance between nanowalls (white arrow); whereas DIS 80 • samples presented nanocone-like structures of a width of 36.7 ± 9.9 nm (white arrow) spaced throughout the surface.

Evaluation of Surface Chemistry and Wettability of Irradiated Titanium Alloy Specimens
Our samples are (α + β) titanium alloy that contains α stabilizer element aluminum and β stabilizer element vanadium. α phase could be observed using (100), (002) and (101) α peaks. β phase could be observed at (110) β peaks (black arrow). We did not observe any major changes in the α and β phase of the modified DIS Ti samples, which confirms the similar surface crystalline structure of Ti6Al4V modified samples as polished Ti (Figure 3) [41,42]. These surface modifications do not change the bulk crystallographic orientation of Ti (microstructure) due to the low ion penetration around 3-4 nm depth [32,33]. On the other hand, we did observe a reduction of CA values on DIS samples compared to polished surfaces. DIS surfaces were slightly more hydrophilic than polished Ti even though no statistical differences were detected (p > 0.05). It should be noticed that SLA (40.83

Evaluation of the Cellular Attachment on Irradiated Titanium Samples
After the implantation-derived immune response, mesenchymal stem cells (MSC) migrate to the implant site, where growth factors are released, and start their differentiation process into bone-forming cells [6,43]. In light of this, we evaluated cellular attachment of undifferentiated bone-forming cells 4 h after seeding on the BMP-2 conjugated modified Ti alloy surfaces through the analysis of SEM microphotographs (compiled in Figure 5). We observed the presence of attached cells in all samples. However, SEM images suggested that with the addition of BMP-2, the filopodia number per cell increases, particularly for DIS 80 • and Anodized samples and slightly for SLA samples. We observed a 2.25fold increase on DIS 80 • and 0.6-fold increase on Anodized samples compared to their counterpart without BMP-2. However, DIS 80 • samples seemed to have more filopodia (39 ± 20 filopodia) compared to the other samples averaging less than 30 filopodia per cell (see Figure 5a,b). Although these results are promising, it will be interesting in future studies to measure the adhesion strength of cells cultured on these surfaces to corroborate this data. Filopodia protrusions are composed of bundles of actin filaments which play a role in the initial cell adhesion, spreading and migration. Cells use filopodia to sense and tether to their surroundings, which require the development of strong tensile forces. As time progresses, these tensile forces will further stabilize the cells, recruiting cell adhesion receptors (integrins) and force-regulating proteins (vinculin, tailin and zyxin), which participate in forming mature adhesions [44][45][46].
Due to the values observed on titanium surfaces regarding cellular adhesion structures (filopodia prolongations), we decided to evaluate the adhesion process by the immunostaining and quantification of vinculin protein expression. Vinculin protein participates in focal adhesion complexes and plays a fundamental role in cell-cell and cell-matrix interactions and regulates adhesion through binding, polymerizing and remodeling actin fibers [47,48]. Figure 6a shows the confocal images of C2C12 cells, in which vinculin proteins appeared in green, actin in red, and cell nuclei in blue color. Cell number and vinculin intensity from these images were quantified using FIJI software and the results were compiled in Figure 6b. We observed that the addition of BMP-2 increased vinculin expression on cells growing on polished (465%), DIS 80 • (200%) and Anodized samples (110%) compared to the samples without BMP-2, although no statistical differences between DIS 80 • with BMP-2 and polished BMP-2 were found.
In addition, cell spreading was evaluated through the analysis of the confocal microphotographs (compiled in Figure 7). The cell cytoskeleton, which is composed of actin filaments, was measured to determine the area of the cell (Figure 7a). We observed that the presence of BMP-2 in cells growing on DIS 80 • samples showed higher surface area, increasing cellular spreading by 120% compared to its counterpart without the protein, but no statistical differences were found compared to polished samples with BMP-2 ( Figure 7b). Although we did not observe significant differences in nucleus area among the different samples, it seemed that the addition of BMP-2 increased the nucleus area of cells cultured on DIS 80 • samples (Figure 7c). Figure 8 shows cell viability results after 3 days of cell culture of all studied surfaces measured by the metabolic activity of C2C12 cells. Percentages of cell viability were calculated using polished without BMP-2 as reference (100%), observing that all surfaces achieved cell viability percentages from 80 to 120%.    Integrins α and β-subunit are linked to the actin cytoskeleton via adaptor proteins (talin and vinculin). Integrins agglomerate and form focal adhesions, which translate mechanical stimuli into biochemical signals that will start gene expression either by activating signaling pathways, such as extracellular-signal-regulated kinase (ERK)/mitogen-activated protein kinase (MAPK) or through actomyosin contractility, which will distort the shape of the nucleus and allow the translocation of transcription factors [47][48][49]. Fourel and collaborators observed an increase in β3 integrin expression in C2C12 cultured on soft polymeric films with matrix-bound BMP-2, which promoted cell spreading and adhesion [50]. BMP-2 enhancement of C2C12 cellular adhesion has also been reported in hydroxyapatite and magnesium surfaces by Huang and collaborators [51]. In this study, we used nanopatterned metallic substrates, which might serve as BMP-2 nanoreservoirs, promoting localized BMP-2 signaling. We observed an increase in vinculin intensity and changes in cell area in surfaces with BMP-2, particularly in DIS 80 • treated surfaces. Although we did not measure integrin expression, the vinculin and cell spreading results suggest that integrin-mediated signaling is involved in this process and could explain the changes we observed in cellular attachment and cell fate.
Moreover, topographical features have been found to influence cell attachment [13,49]. For example, Pan and collaborators evaluated the behavior of MSC on a macropore/ nanowire structure fabricated via vacuum plasma spraying and etching with sodium hydroxide (NaOH), which mimics bone hierarchical microenvironment. These cells were elongated and spread in multiple directions with well-developed focal adhesions. This led to an increase in cytoskeletal tension and yes-associated protein (YAP) activity and nuclear translocation. YAP is a member of the Hippo Signaling pathways that shuttles between the cytoplasm and the nucleus under specific physical cues, e.g., stiffness and topography, acting as a promoter for osteogenic transcription factors; thus, inducing MSC differentiation [52]. Other structures such as nanotubes (diameters of 10 nm and 30 nm), which have spacings of less than 50 nm provide an effective length promoting integrin clustering/focal contact formation and increasing detachment forces [53,54]. It is hypothesized that this is caused by the structures matching the integrin diameters (10 nm) [55]. Alternatively, this phenomenon might be caused by an increase in surface contact area to which more integrins can bind to. For example, Babchenko and collaborators have found that nanopatterned surfaces with nanocones, which were densely packed and distributed homogeneously on the surface, increased vinculin signaling and promoted focal adhesion kinase (FAK) activation in osteoblastic-like cells (Saos-2) [56,57]. Considering this, the dimensions of our nanostructures could further improve integrin-mediated adhesion and traction forces; and thus, their involvement in mechanotransduction pathways. Integrin expression and its role on cell morphology and behavior is the focus of a future study which will include cells at different stages of osteogenic differentiation and different stages of cellular adhesion.
Other factors that influence osteoblast cell spreading and proliferation are surface texturing/patterning [8,58]. Although we did not measure roughness (Sa) in this study, similar Ar + irradiation conditions (with lower fluence 2.5 × 10 17 ions/cm 2 ) were performed on titanium alloy and showed Sa of 10 to 50 nm reaching higher values on SLA (49 nm) [32]. However, it will be interesting in further studies to correlate these measurements with the DIS conditions used in this study.

Evaluation of Osteogenic Differentiation and Mineralization on Irradiated Titanium Samples
Generally, there are three major stages during MSC osteogenic differentiation: differentiation, matrix maturation, and mineralization. First, attached MSC will differentiate into osteoprogenitor cells expressing Runt-related transcription factor 2 (Runx2), Distal-less homeobox 5 (Dlx5) and Osterix (Osx). Once committed to an osteogenic lineage, these cells (preosteoblasts) will express ALP and collagen I (Col I) and eventually will differentiate into mature osteoblasts, expressing and secreting Col I, OCN, and osteopontin (OPN), which form the matrix. Afterwards, ALP will aid in the mineralization process by releasing phosphate ions, which will be combined with calcium to produce hydroxyapatite. In mouse models, the mineralization phase peeks at 14-21 days [59][60][61][62]. In C2C12, ALP is expressed by stimulating the cells with 300 ng/mL of BMP-2 after 2-3 days in culture, which induces their osteogenic differentiation and inhibits myotube formation [38,63]. From previous studies and what has been established in the literature, BMP-2 UdeA optimal concentration was 3 µg/mL to detect significant ALP production after 3 days of C2C12 cell culture [64].
For this purpose, we cultured BMP-2 and C2C12 cells directly on modified Ti alloy surfaces and evaluated the osteogenic differentiation and mineralization. We observed statistical differences in ALP activity produced by cells on Ti surfaces with BMP-2 versus without BMP-2 due to the intrinsic osteoinductive effect of BMP-2. We observed an increase in ALP production for DIS 80 • (265%), Anodized (195%) and SLA samples (190%), yet we did not observe statistical differences among the different samples with BMP-2. This fact suggests that BMP-2 plays a more dominant role in cell differentiation than the surface nanotopography. Nevertheless, it should be noticed the slightly higher ALP activity on cells growing on DIS 80 • surfaces with BMP-2 compared to the other samples with BMP-2, which suggests a possible synergistic effect between the nanopatterned irradiated surface at 80 • and BMP-2s (Figure 9a). This synergistic effect might be similar to the one MSCs face in their native environment, where they have physical and biological cues, leading to their differentiation into bone cells. On the other hand, the late phase of osteoblast differentiation or cell mineralization phase is characterized by the presence of calcium and phosphate deposits, which form hydroxyapatite and collagen fiber production. Through Alizarin red staining of free calcium nodules, we could determine the extent of the C2C12 cell mineralization process on Ti surfaces in the presence of mineralization media [39]. DIS 80 • and Anodized surfaces with BMP-2 showed a greater mineralization area after 21 days in cell culture compared to other samples, particularly DIS 80 • with BMP-2 increased the percentage of calcium nodules by 329% compared to polished samples with BMP-2 ( Figure 9b). The differences between DIS-based modified surfaces and BMP-2 mineralization results could, in fact, still be within the variance as the statistical sample was limited.
Surface treatment can influence the adsorption of bone extracellular matrix proteins such as BMP-2 and modulate cell behavior. For instance. Xiao and collaborators evaluated BMP-2 adsorption on polished, etched and anodized Ti surfaces. They found that etched samples had the highest BMP-2 absorption, yet the protein changed its conformation and reduced its bioactivity in MSC in the long term, in contrast to BMP-2 adsorbed on anodized samples [65]. They also studied the synergistic effect of BMP-2/fibronectin adsorption on these surfaces and observed that it promoted MSC spreading and osteoblast differentiation [66]. These results agree with those observed in our study in which DIS 80 • in the presence of BMP-2 showed the highest levels of vinculin expression (FAK components) and cell spreading (cell area). Nevertheless, future studies are needed to evaluate how our surfaces modulate protein conformation, adsorption, bioactivity at different concentrations and timepoints and their effects on cellular behavior. Researchers have observed that cells can respond to nanopatterns of less than 13 nm in height or orthogonal or hexagonal patterns of nanopits with diameters from 35 to 200 nm [67]. Nanopits of 22 nm promote osteogenic gene expression via integrin signaling [68]. Our results agree with other studies using different nanopatterning techniques, which stimulate bone formation by targeting integrin signaling [69][70][71] yet DIS presents many advantages over these methods. By modulating DIS irradiation parameters, we can generate reproducible and scalable nanopatterns directly on the material in a short amount of time (usually a few minutes) without using toxic chemicals that pose health risks or require extra steps to discard them. Future studies will focus on evaluating integrin-mediated bone formation on these surfaces.
BMP-2 controlled delivery is one of the main challenges in bone tissue engineering as uncontrolled drug release can cause serious side effects such as ectopic bone formation, bone reabsorption and hematomas in soft tissues [72]. Currently, these and other growth factors can be tethered to the implant surface via physisorption or covalent binding to control their release. Physical adsorption relies on electrostatic interactions, hydrogen bonding, or hydrophobic interactions. In covalent binding, the substrate is treated with plasmas, chemical etching and surface coatings to functionalize the surface. Also, bioconju-gation reactions such as amidation, esterification and click reactions through carbodiimides, silanes, mussel-inspired bioconjugation have been used [73,74]. Another strategy is to encapsulate growth factors in scaffolds, e.g., titania nanotubes, to control the release of BMP-2 and induce bone formation [21,73]. Therefore, it will be interesting in the future to evaluate the functionalization of these surfaces via ion irradiation with non-inert gases (nitrogen and oxygen) to generate functional groups in which BMP-2 can be tether to and control its release.

Conclusions
The combination of growth factors (BMP-2) and active nanotopographies (nanocones and nanowalls-like structures) have demonstrated a synergy in cell adhesion, differentiation and mineralization of a non-osteoblastic cell lineage (C2C12 cells). The following findings can be drawn: DIS allowed the design of nanostructures on titanium alloys surfaces, resulting in nanocones and nanowalls of size below 50 nm by changing the incidence angle from 60 to 80 degrees with high fluences.

2.
The crystalline structure of DIS samples was unmodified; and although no statistical differences were observed in terms of wettability, DIS samples seemed more hydrophilic than polished samples.

3.
The presence of BMP-2 plays an important role in cellular adhesion and spreading.
In this study, BMP-2 addition seemed to increase filopodia number per cell and vinculin expression in most surfaces and cell spreading on DIS 80 • and polished surfaces compared to surfaces without BMP-2. However, surface topography or nanopatterning by itself does not contribute significantly to these processes, except by increasing slightly vinculin expression in DIS 80 • nanocone-patterned surfaces. 4.
DIS 80 • nanocone-patterned surfaces in conjunction with BMP-2 increase almost 1.2fold cell spreading and 2-fold vinculin expression, reaching values similar to polished samples with BMP-2. Moreover, we observed a 2.25-fold increase in the number of filopodia per cells (39 ± 20) in these surfaces compared to all surfaces with or without BMP-2, suggesting a synergistic effect in cell adhesion when we combine DIS 80 • treatment with BMP-2.

5.
Cell differentiation and mineralization, determined by ALP activity and calcium nodules formation, respectively, were enhanced in the presence of BMP-2 for all samples. In particular, we observed that this effect was more pronounced on DIS 80 • and Anodized samples with BMP-2 (>2-fold increase in ALP activity compared to their counterparts without BMP-2 and >3.3-fold increase in cell mineralization compared to polished samples with BMP-2). 6.
Finally, the nanocone-like structures generated at an incidence angle of 80 • in combination with BMP-2 have shown a stronger synergistic effect in modulating cellular processes when compared to DIS 60 • and polished observing this nanocone topography more suitable to improve cellular interactions. Thus, DIS treatment in conjunction with BMP-2 may improve Ti implants osseointegration by guiding cell differentiation toward bone formation.