Polarization of an electroactive functional film on titanium for inducing osteogenic differentiation

To enhance the surface bioactivity of titanium (Ti) prostheses, an electroactive polyvinylidene fluoride (PVDF) film was prepared on a Ti substrate to provide a mimetic of the electrical microenvironment, which facilitated the performance of cell functions. The results of cell proliferation and differentiation assays indicated that polarization of the PVDF-Ti (PTi) altered its surface charge, thus inducing adhesion, proliferation and osteogenic differentiation of cells. The polarized PVDF-Ti (PPTi) may therefore find applications in bone regeneration.

Scientific RepoRts | 6:35512 | DOI: 10.1038/srep35512 organic solvents. PVDF is polymorphic and can form different crystalline structures, such as alpha, beta and gamma structures, depending on the crystallization conditions 16 . Fourier transform infrared (FTIR) spectroscopy is an appropriate tool for identification of the characteristic absorption band phases of PVDF. The attenuated total reflectance-fourier transform infrared (ATR-FTIR) spectra of the PPTi and NPTi are presented in Fig. S2(a). The PPTi and NPTi both exhibited prominent β -phase peaks at 836 cm −1 and 873 cm −1 , with small α -phase peaks being observed at 762 cm −1 and 873 cm − 1 11,17,18 . The fraction of the crystalline form in a crystal can be estimated from the absorbance of the characteristic peaks of all crystalline forms and their absorption coefficients 19 . The absorbance at 762 cm −1 (A α ) and 836 cm −1 (A β ) and the absorption coefficients of the α form (K α = 6.1 × 10 4 cm 2 /mol) and the β form (K β = 7.7 × 10 4 cm 2 /mol) can be used to calculate the mass fraction of the α form (Fα ) in the crystal using Eq. 1 (assuming that the densities of the α and β crystallites are equal) 19 : The relative fraction of the β -phase [F β ] PVDF in the films was calculated. F β of the PPTi and NPTi was 75.95% and 73.93% within the crystalline regions, respectively. The degree of crystallinity of the PPTi and NPTi was 48.5% and 45.3%, respectively, as calculated from the DSC results using the following equation Eq. 2 ( Fig. S2(b)) 17,20 .
Hm/ HmB 100% (2) where Δ Hm is the crystalline-phase melting enthalpy per gram of film from PTi; Δ HmB is the enthalpy value corresponding to the melting of 100% crystalline PVDF film (Δ HmB = 104.7 J/g). X ray diffraction (XRD) analysis of the PPTi and NPTi revealed distinct peaks at 18.56° and 20.36° corresponding to crystal planes at (020) for α -phase PVDF and (110) for β -phase PVDF, respectively (Fig. S2(c)) 11,19,21 . PFM results. Figure 2 shows the PFM amplitude and phase images of the samples within a 5 × 5 μ m scan area.
As demonstrated in Fig. 2(a1,b1), the PFM amplitude images confirmed that the β -phase regions exhibited piezoelectric properties. Compared with the NPTi, a higher piezoelectric response was observed in the PPTi, which exhibited a higher β -phase content within the crystalline region 22,23 . The phase images presented in Fig. 2(a2,b2) revealed distinct contrasts, which may indicate that the poling of the PTi was not complete. Contrast variations in the phase image were the result of the presence of a number of domains in the scan area 22,23 . Consistent with the calculations based on the FTIR analysis, the PPTi displayed a slightly higher proportion of polarized β -phase.
Theoretically, a PVDF polymer serves as an insulator of electricity, particularly in amorphous regions. When the sample is subjected to AC bias during AFM scanning, charge accumulation occurs on the surface 24 . Charge accumulation creates an electrostatic force between the tip and sample, which results in an artificial phase shift toward the positive voltage direction. The SKPM surface potential exhibited two distinct values (− 930 ± 12 mV and − 157 ± 8 mV) for the PPTi and NPTi. The SKPM and zeta potential analyses of the PPTi and NPTi revealed that the surface of the PPTi accumulated more negative charges (Fig. 3).
Cell proliferation and differentiation. The proliferation and differentiation profiles of BMSCs cultured on the PPTi and NPTi are presented in Fig. 4. The proliferation of BMSCs was investigated using the CCK-8 assay. Figure 4(a) showed that the proliferation of BMSCs was greater on the PPTi than NPTi at 4 and 7 days. On day 1, cell proliferation on the NPTi was similar to that on the PPTi. Alkaline phosphatase (ALP) activity was examined to evaluate the osteogenic differentiation of BMSCs on the PPTi and NPTi after 7, 14 and 21 days of cell incubation. The ALP activity of the BMSCs cultured on the PPTi was slightly higher than in the cells cultured on the NPTi at 7 and 14 days ( Fig. 4(b)). After 21 days, a significant difference was observed. This difference may be attributed to the increased negative charge of the PPTi. Therefore, the PPTi exhibited a good biocompatibility and osteogenic ability in vitro. These findings were further supported by the real-time polymerase chain reaction (RT-PCR) analysis of other osteogenic marker genes (ALP, collagen I and osteopontin) (Fig. 5), in which the cells on the PPTi exhibited significantly higher gene expression (*p < 0.05, **p < 0.01) than those on the NPTi after 14 days of incubation.
Representative scanning electron microscope (SEM) images of BMSCs fixed at specific time points were used to evaluate differences in cell morphology on the PPTi and NPTi (Fig. 6). Cells typically required 2-4 h to adhere to the TCPS surface but required up to 12 h for attachment on the PTi. Because our PTi samples were hydrophobic, the cells were weakly attached to the samples and may have been washed off due to the rinsing and dehydration procedures performed on day 1. Therefore, the assays were carried out on day 7. As shown in Fig. 6, the morphologies of BMSCs on the PPTi and NPTi were different. Spindle and triangular cells were observed on the PPTi, and more BMSCs exhibited larger filopodia on the PPTi.

Discussion
Piezoelectric effects have been explored in bone-tissue regeneration since the first observation of these effects in bone. Piezoelectric films can induce in vivo formation of periosteal bone 25 . Electrical osteogenesis has been proved by X ray photographs and histological studies performed during a first series of implantations of piezoelectric and non-piezoelectric PVDF films. Ficat et al. had studied the osteogenic power of two kinds of PVDF films (monomorph PVDF films and bimorph PVDF films), and they attributed the formation of periosteal bone induced by piezoelectric PVDF films to a piezoelectric effect 26 . Previous studies of osteoblast-PVDF interactions have demonstrated that these interactions can be exploited clinically to promote tissue growth 13,27 . The different phase types of PVDF films can affect the adhesion and proliferation of cells in different ways 28 . Surface topography and surface wettability also greatly influence cell adhesion and proliferation. Charged surface promote osteoblast adhesion and proliferation. Thus, the combination of surface properties and piezoelectricity is a key factor in the induction of cell adhesion, proliferation and osteogenic differentiation on Ti with a layer of piezoelectric material.
We hypothesized that the cellular behaviours on the surface of the PPTi and NPTi would differ based on their surface charges and intrinsic properties. After polarization, the dipoles of the piezoelectric PTi become oriented, resulting in the distribution of negative charges on the surface. Consequently, among the ionic/biological constituents present in the growth medium, cations adhere strongly via electrostatic interactions with the negatively charged surface. Proteins (e.g., integrin and fibronectin), and the negatively charged cytomembrane are attracted (Fig. 1). Consequently, the PPTi induced cell adhesion, proliferation and osteogenic differentiation. FTIR, DSC and XRD characterization revealed that the PPTi and NPTi displayed a majority of β -phase PVDF. The polarization process resulted in no change in the chemical composition, but a small change in the molecular structure was revealed by a small change in the phase content and a slight decrease in hydrophobicity (Fig. S3). The SKPM and zeta potential analyses of the PPTi and NPTi revealed that the PPTi surface accumulated more negative charges. PFM analysis demonstrated that the PPTi exhibited a greater piezoelectric area, whereas the NPTi exhibited weaker piezoelectric properties. The results of our cellular experiments performed in vitro demonstrated that the PPTi induced a more homogeneous distribution of BMSCs and osteogenic differentiation. In conclusion, the polarization process improved the surface distribution of negative charge and the piezoelectric response, resulting in inducing BMSC adhesion, proliferation and osteogenic differentiation.
In summary, an electroactive PTi was studied as a suitable material for bone regeneration applications due to its piezoelectric effect. To study the effect of the polarization of the PTi on the cellular response, BMSC adhesion, proliferation and differentiation were evaluated on the PPTi and NPTi. BMSC adhesion was influenced by the  Preparation of the PTi. A PVDF membrane was prepared on the Ti sheet surface using PVDF solution (10%). In a typical procedure, PVDF was dissolved in a mixture of DMF and acetone (4:6) with stirring for 4 h at 60 °C. After cooling to room temperature, approximately 50 μ L of the solution was spread on the Ti surface and dried at 80 °C for 2 h to remove the solvent via evaporation and to permit isothermal crystallization of PVDF. Electrical poling of the PVDF films was achieved through corona discharge at 100 °C. The specimen was heated from room temperature to 100 °C while the voltage applied between the electrodes was gradually increased from 0 to 1.5 kV at a rate of ~100 V/min. The sample was maintained at the elevated temperature and an applied voltage of 1.5 kV for 1 h. The sample was then cooled to room temperature (20-25 °C) under continued application of a voltage of 1.5 kV. The piezoelectric response (d33) of the poled samples was then verified using a wide-range d33 meter (YE2730A, Sinoceramics, Inc., China). The obtained piezoelectric d33 coefficient was approximately − 28 pC N −1 .  For the in vitro assays, circular PTi with a diameter of 10 mm was sterilized via immersion several times in 75% ethanol for 30 min. Next, the samples were washed for 5 min in sterile phosphate-buffered saline (PBS) 5 times to eliminate residual ethanol. The samples were then exposed to ultraviolet light (UV) for 30 min. The obtained samples were PPTi and NPTi.
Characterization of the PTi. The morphology of the PTi was assessed after 120s of gold sputtering via field emission scanning electron microscopy (FESEM, Merlin, Germany) and atomic force microscopy (AFM, MFP-3D-S, Asylum Research, USA).
Attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR) was performed for the PPTi and NPTi using a Bruker spectrometer (Vector-22, Switzerland, Bruker Company), from 600 to 1500 cm −1 at a resolution of 1 cm −1 , to evaluate short-range molecular arrangements. The fraction of the β -phase within the crystalline region was calculated using the method first described by Osaki and Ishida [28][29][30] .
The thermal behaviour of the PTi was analysed through differential scanning calorimetry, (DSC, DSC 6, Perkin Elmer, USA). Film samples of 10 mg peeled of the PPTi and NPTi were used for each DSC measurement. The heating rate was 10 °C/min for standard DSC and 2 °C/min for modulated DSC, with a modulation period of 50 s. Melting heat enthalpy integration was performed from 60 °C to 220 °C. Before all DSC experiments, the baseline was calibrated using empty aluminium pans.
X-ray diffraction (XRD) was performed to characterize the crystallinity phase of the PTi. The samples were scanned in a 2θ range of 3 to 60 degrees. XRD was recorded with a Bruker X-ray diffractometer (Bruker D8 Advance) using Cu Kα radiation at λ γ = 15.54 Å.
Piezo-force microscopy (PFM) (MFP-3D, Asylum Research, USA) images provide nanoscale insight into the distribution of polarization within PPTi by detecting its mechanical oscillation in response to the applied AC bias. To quantify the piezoelectric response of the PPTi and NPTi, a consistent AC. bias of 10 V was applied to the tip during sample scanning 31,32 . Piezoresponse imaging was performed over an area of 5 × 5 μ m using a conductive tip with an A.C. bias of 10 V in amplitude (peak-to-peak). PFM detects tip oscillations resulting from piezoelectric vibrations of the sample surface induced by the tip-generated electric field. The PFM amplitude and phase images were recorded.
Cell proliferation. BMSCs were cultured in Gibco Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% foetal bovine serum (FBS) in a humidified incubator with 5% CO 2 at 37 °C. BMSCs at passages P4-P5 were used for the experiments. The culture medium was refreshed every two days. In all experiments, the samples were sterilized with 75% ethanol and ultraviolet light. The experiments were conducted in triplicate; the number of samples was 3 for all analyses (n = 3). For BMSC culture, 500 μ L of BMSC solution per well (2 × 10 4 cells/mL) was added to PPTi and NPTi in 48-well plate. On days 1, 4 and 7, 300 μ L/well of cell counting kit-8 (CCK-8) solution (0.1 mL of CCK-8 per millilitre of medium) was added, followed by incubation for 4 h at 37 °C under 5% CO 2 . Then, 200 μ L of the solution from each well was transferred (in triplicate) to a 96-well plate, and the absorbance was determined at 450 nm using a microplate reader (Biocell lt2, Australia).
Alkaline phosphatase activity. ALP is an early marker of osteogenic differentiation and is related to the production of a mineralized matrix. To determine ALP activity, BMSCs were seeded on PPTi and NPTi at a density of 2 × 10 4 cells/mL in 500 μ L of complete medium without osteogenesis induction media. After incubation for 7, 14 or 21 days, 50 μ L of the cell lysate obtained after treatment with 0.2% Triton − 100 at 4 °C was transferred from each well to a 96-well plate and incubated with 50 μ L of p-nitrophenyl phosphate (pNPP) solution for 30 min at 37 °C. The reaction was terminated by the addition of 0.1 M NaOH solution, and the absorbance was read in a microplate reader at 405 nm.
Cell adhesion. At each incubation time point, the medium was removed from each well, and the samples were washed with PBS. The cells were then fixed with 4% paraformaldehyde (Aladdin) for 1 h at 4 °C. To evaluate cell morphology, the samples were washed with PBS and dehydrated in an ethanol gradient (30, 50, 60, 70, 80, 90 and 100% ethanol in water). The samples were then placed under vacuum at room temperature for 4 h. The dried samples were subsequently subjected to gold sputtering in a vacuum and evaluated via scanning electron microscopy (FESEM, Nova Nano-430, USA).

Statistical analysis.
Data are presented as the mean ± standard error of the mean; at least three replicates were averaged for each data point. Statistical analysis was performed through t-tests with Bonferroni post-hoc analysis using GraphPad Prism Software. A p-value of < 0.05 was considered to indicate a significant difference and p < 0.01 was considered to indicate a very significant difference.