Scaffolds with isolated carbon nanowalls promote osteogenic differentiation through Runt-related transcription factor 2 and osteocalcin gene expression of osteoblast-like cells Open Access

Recent reports have paid much attention to application of micro- and nano-structures as cell culture scaffolds, including titanium with nanoscale roughness,¹ nanofibers,² hydroxyapatite,³ and carbon nanomaterials.^4,5 The adhesion of osteosarcoma Saos-2 cells is improved by the roughness of nanocrystalline diamonds, and their intercellular molecular mechanisms are modified by the surface morphology of the scaffold, as reported by Broz et al.⁴ Saos-2 cell differentiation is modified by scaffold nanostructures, such as variation in the diameters of titanium dioxide nanotubes, as reported by Voltrova et al.,⁵ suggesting the importance of identifying other micro- or nano-structures for culturing cells.

Carbon nanowalls (CNWs) are self-organized carbon nanomaterials with a wall-like aggregation of nanographenes oriented vertically on a substrate and are characterized by high surface area, high aspect ratio, and high conductivity.^6,7 CNWs retain their shape when immersed in liquid, unlike other carbon nanomaterials such as carbon nanotubes, which cohere due to van der Waals force.⁸ Thus, CNWs have potential application in electrodes for fuel cells, biosensors, and scaffolds for culturing cells.^9–16 

The biocompatibility of CNWs has been explored for their use as scaffolds for cell culture for biomedical applications.^10–14 The morphology of the cells changed due to changes in cell adhesion when the CNW surface was chemically terminated with hydrogen, oxygen, fluorine, or nitrogen.^9,13 The high conductivity of CNWs promoted the proliferation of cells cultured on CNW scaffolds upon electrical stimulation.¹⁰ 

We recently reported that the differentiation of Saos-2 cells was significantly up-regulated when the cells were incubated on CNW scaffolds with average wall-to-wall distances ranging from 200 to 400 nm.¹¹ 

Typically, it is only possible to control narrow wall-to-wall distances using a radical-injection plasma-enhanced chemical vapor deposition (RI-PECVD) system.¹⁷ However, very recently, isolated CNWs were fabricated with average wall-to-wall distances of over 1 μm by the application of nanosecond pulse voltages onto the growth substrate.¹⁸ This technology allows the control of wall-to-wall distances,¹⁹ with CNWs prepared with wall-to-wall distances ranging from 100 to 3300 nm that were subsequently employed as scaffolds for cell culture.

In this study, we demonstrated the utility of sparse isolated CNWs with average wall-to-wall distances from 100 to 3300 nm as scaffolds for culturing osteosarcoma Saos-2 cells. We observed cell morphology, proliferation, and differentiation, and we evaluated the gene expression of proliferating and differentiating osteogenic cells using the real-time polymerase chain reaction (PCR).

All scaffolds were prepared using CNWs synthesized by a RI-PECVD system. Controlling the repetition rates of the nanosecond pulses and maintaining a constant electrostatic input provided CNWs with average wall-to-wall distances ranging from 100 to 3300 nm. The substrate holder (32 cm²) was electrically connected to the inductive energy storage (IES) circuit. Details of the RI-PECVD system with the IES circuit were reported previously.¹⁸ The synthesis conditions for the CNW scaffolds used for cell culture are shown in Table I. Differences in cell morphology and cell viability using CNWs with various wall-to-wall distances were studied.

Saos-2 cells were incubated in Dulbecco’s modified Eagle’s medium (DMEM) (Sigma, D5796) in a humidified atmosphere of 5% CO₂/95% air at 37 °C. The experimental setup for culturing cells on CNW scaffolds was previously described.¹⁰ A 48-well cell culture cluster (Corning, 3548) made from polystyrene (PS) was used for incubation as a control. Before cell culture, all components were washed with pure water and sterilized by exposure to ultraviolet light for 20 min. The wells were filled with 300 μl phosphate-buffered saline (PBS; Gibco Life Technologies) for 20 min, and then the PBS was replaced with fresh PBS for 15 min. Thereafter, each well was filled with 300 μl DMEM supplemented with 10% fetal bovine serum (Life Technologies, 10437-028), 100 units/ml penicillin, and 100 μg/ml streptomycin (Life Technologies, 15140-122) and left for 1 h at room temperature. A total of 3000 cells were seeded in each well.

Cell morphology was observed using a fluorescent microscope (BX53, Olympus) with 4′,6-diamidino-2-phenylindole (DAPI) and fluorescein isothiocyanate (FITC) filters after incubation for 2 days. The cells were treated sequentially with 4% paraformaldehyde in PBS (Wako, 163-20145) for 30 min, 0.1% TritonXTM-100 (Sigma, X100) in PBS for 15 min, and 5% normal goat serum (Life Technologies, 50062Z) in PBS for 10 min. Actin cytoskeleton was stained with 100 nM Acti-stainTM 488 phalloidin (Cytoskeleton, Inc., PHDG1-A), and nuclei were stained with 200 nM DAPI (Sigma, D9542).

Cell viability was evaluated after 4 days’ incubation. The culture medium was replaced with 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium and an inner salt (MTS) reagent (Promega Corp., G5430); the cells were incubated for 1 h, and then the light absorption of each well was measured at 490 nm. The absorbance after MTS treatment reflected the relative number of live cells, and thus, the absorbance was normalized to that of a control sample.

Cell adhesion was observed using a scanning electron microscope (SU-8200, Hitachi-High-Technologies, Japan) after incubation for 2 days. A Si scaffold with a smooth surface was prepared for comparison with the CNW scaffolds. Before observation, the cells were treated sequentially with PBS, 70% ethanol, 90% ethanol, 95% ethanol, and 99.5% ethanol, each for 10 min. Real-time PCR was used to detect mRNAs encoding Runt-related transcription factor 2 (Runx2) and osteocalcin (OC) after 10 days’ incubation. Cells were seeded at ∼1 × 10⁵/well. The details of the real-time PCR measurement setup were previously described.¹¹ 

CNW scaffolds were prepared with average wall-to-wall distances (D) ranging from 100 to 3300 nm. Figure 1 shows top and tilted views of SEM images of the synthesized CNWs: (a) and (d) condition I (D = 3300 nm), (b) and (e) condition II (D = 700 nm), and (c) and (f) condition III (D = 100 nm). The heights of the CNWs were ∼500 nm. The a-C film uniformly covered the substrate in conditions I and II, as reported previously.¹⁸ The water contact angles on all CNWs were about 40–70 deg. Dowling et al. reported that highly hydrophilic or hydrophobic surfaces led to a progressive reduction in the level of adhesion of osteoblast-like cells.²⁰ Hence, all CNWs were expected to adhere cells.

Saos-2 cells were cultured on the CNW scaffolds. Figure 2(a) shows fluorescence microscope images of cell nuclei and F actin in Saos-2 cells grown on PS and the CNW scaffolds with different D’s (D = 3300, 700, and 100 nm) after incubation for 2 days. The cells adhered on all the CNW scaffolds. The fluorescent areas on the microscopic images were counted using the image processing program ImageJ. Figure 2(b) shows the cell-fluorescent areas of the PS support and the CNW scaffolds with different D’s (D = 3300, 700, and 100 nm) after incubation for 2 days. The cell-fluorescent areas indicated the extensibility of the cells. The CNW scaffolds with D = 100 nm had the smallest area of cell growth of the conditions tested. Cell growth on the sparse CNW scaffolds with D = 3300 nm was similar to that on flat PS. The CNW surfaces showed good biocompatibility because cell extensibility and adhesion were unchanged, suggesting that the CNW edges determine cell extension and cell culture was successful. Figure 2(c) shows the number of cells incubated on the CNW scaffolds (D = 3300, 700, and 100 nm) after 4 days’ incubation. Although the largest number of cells grew on the PS surface (a commercial dish for culturing cells), the number of cells on all the CNW scaffolds was not significantly different, at 1 × 10⁵ to 2 × 10⁵ cells. Thus, sparse, isolated CNW scaffolds do not affect cell proliferation compared to dense CNW scaffolds. Cells on all the CNWs scaffolds adhered similarly regardless of the wall-to-wall distances.

The expression levels of Runx2 and OC were used as indicators of cell differentiation. Figure 3(a) shows the relative expression levels of Runx2/RPS18 after incubating the cells under the indicated various conditions for 10 days. The relative expression levels were highest for the D = 700 nm CNW scaffolds. The normalized levels were 1.0 for PS and 3.4 for D = 3300 nm, 8.1 for D = 700 nm, and 2.2 for D = 100 nm for the CNW scaffolds. These differences were significant for all the CNW conditions (p < 0.05). Similarly, Fig. 3(b) shows the relative expression levels of OC/RPS18 after incubation for 10 days. The normalized expression levels were 1.0 for PS and 4.1 for D = 3300 nm, 8.4 for D = 700 nm, and 2.3 for D = 100 nm. These differences were significant for all the CNW conditions (p < 0.05). Interestingly, differentiation of Saos-2 cells was promoted by the middle wall-to-wall distance CNW scaffolds (D = 700 nm).

These results indicate that the effect of the CNW scaffolds on cell differentiation increases with increasing D, consistent with previous research on culturing cells on CNWs with D ranging from 200 to 400 nm.¹¹ However, differentiation decreases as D approaches the size of cells, such as D = 3300 nm. Thus, CNWs with optimal D promote the differentiation of cells.

Figure 4 shows top and tilted SEM images of cells grown on Si and CNW scaffolds with different D’s (D = 3300, 700, and 100 nm). The white arrows indicate the cell regions shown in the top views [rows (a) and (b) of Fig. 4], in color. Rounded cells were observed for the dense CNW scaffolds with D = 100 nm. In the enlarged views [row (b) of Fig. 4], the morphology of the cells shows polygonal shapes and filopodia on the edge of the CNW scaffolds with D = 700 and 100 nm. Smooth curved cell boundaries are seen for the flat Si and the sparse CNW scaffolds with D = 3300 nm. The filopodia act as tentacles to sense the environment. CNWs with D = 700 and 100 nm appear to influence the morphology of the cells, and the cells formed filopodia to sense the walls. Figures 4(c) and 4(d) show tilted views of SEM images of the cells. The cells adhered to the scaffold, especially on the flat Si scaffold surface and the spare CNW scaffold with D = 3300 nm. With dense CNW scaffolds, the wall-to-wall separation was too narrow to allow the cells to adhere to each wall, and thus, adhesion of the cells appeared to be suppressed. Notably, the walls acted like a wedge, and the cells adhered on the CNW scaffold with D = 700 nm, as indicated with red arrows in row (d) of Fig. 4. Thus, the CNW scaffold with D = 3300 nm was not effective for cell adhesion, and the scaffold with D = 100 nm was too narrow to affect the scaffold morphology. The CNW scaffold with D = 700 nm had a large effect on cell adhesion, promoting osteogenic differentiation.

The cytoskeleton affects osteogenic signaling of cells.^21–24 In topography-induced osteogenesis, the nano- or micro-topography of scaffolds activates several pathways of osteogenic differentiation, including the mitogen activated protein kinase (MAPK) pathway, focal adhesion kinase (FAK) activation, bone morphogenetic proteins (BMPs), and the Wnt/β-catenin pathway, as reported by Rougerie et al.²⁴ Growing cells on a scaffold composed of poly(ethylenimine)-conjected graphene oxide (PCL_GO) reduced the cell area and promoted osteogenic differentiation of hMSC cells, as reported by Kumar et al.,²⁵ and the distribution of focal adhesion (FA) was expressed at the tips of cellular protrusions of cells cultured on PCL_GO. Nano-roughed surfaces also reduced the cell area, changed the distribution of FA, and promoted the osteogenic differentiation of cells, as reported by Hasan et al.²⁶ In the present study, cell morphogenesis on the CNW scaffold with D = 700 nm led to osteogenic differentiation. This result suggests that the distribution of FA in cells cultured on the CNW scaffold with D = 700 nm uniquely promotes the differentiation of cells. Figure 5 illustrates the results of this experiment. The morphology of adhesive cells cultured on the CNW scaffold with D = 700 nm favors activation of the pathway of osteogenic differentiation.

In conclusion, osteogenic Saos-2 cells were incubated on CNW scaffolds with different wall-to-wall distances (D = 3300, 700, and 100 nm) synthesized using a RI-PECVD system and application of nanosecond pulses by an IES circuit. Expression of the osteogenic genes Runx2 and OC was promoted using CNWs with D = 700 nm. Isolated CNW scaffolds have potential for the control of bone generation in regenerative medicine.