Enhanced biocompatibility and osteogenic ability on amino-modified polyetheretherketone

Polyetheretherketone (PEEK) is a potential implant material for dental application due to its excellent mechanical properties and low elasticity modulus. However, its biological inertia results in weak osseointegration between implants and bone tissue, which limited its clinical application. In this study, amino groups were covalently grafted on the PEEK surface using a simple facile self-assembly method to address its poor osteogenic ability. The surface characterization, cell adhesion, proliferation and osteogenic differentiation of MC3T3-E1 cells on bare PEEK and amino-modified PEEK (PEEK-APTES) were studied. After grafting amino groups onto the PEEK, the surface morphology changed, the contact angle decreased significantly. The PEEK-APTES showed boosted cell adhesion, proliferation, alkaline phosphate (ALP) activity, extracellular matrix (ECM) mineralization, and expression of osteogenic genes in MC3T3-E1 cells. These findings suggested that amino modification significantly improved the biocompatibility and osteogenic ability of PEEK in vitro.


Introduction
Polyetheretherketone (PEEK) has an elastic modulus (approximately 3-4 GPa) similar to that of human bone (approximately 17 GPa), which can reduce the stress shielding effect caused by the mismatch of elastic modulus between metal materials and bone [1][2][3]. In addition, PEEK has many excellent properties such as high mechanical strength, abrasion resistance, corrosion resistance, and non-toxicity, and is approved by the U. S. Food and Drug Administration (FDA) as an implantable biomaterial [4,5]. Therefore, PEEK is expected to replace medical titanium and its alloys for dental and orthopedic implants. However, PEEK is a biologically inert material. Its poor osteogenic activity makes it unable to form effective osseointegration with the surrounding bone tissue, which limits its biomedical application. Fortunately, surface modification on PEEK is a promising method to improve its osteogenic ability and has become one of the research focuses.
Although PEEK is hard to be chemically modified due to its extreme stability, O. Noiset et al found that the carbonyl group in benzophenone could be reduced to a hydroxyl group by a strong reducing agent so that the PEEK surface can be further chemically modified [6]. Several studies indicated that cell adhesion, cell spreading, and proliferation on the surface of biomaterials could be promoted by introducing certain chemical groups, such as carboxyl groups [7,8], hydroxyl groups [9], amino groups [10], and sulfonic acid groups [11]. Moreover, the amino-modified surface is more able to stimulate the osteogenic differentiation of cells [12]. Thus, we expect to introduce amino groups onto PEEK surface using silane coupling agents to improve the biocompatibility and osteogenic capacity of PEEK implants. We hope that this study will contribute to the application of PEEK implants and provide a new idea for the surface modification of PEEK implants and other implant materials. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.

Materials and methods
2.1. Preparation of amino-modified PEEK PEEK (GEHR ® PEEK-1000, Germany) samples were cut into a 15 mm diameter and a 1 mm (±0.1 mm) thickness. All samples were polished using waterproof silicon carbide sandpaper (400, 800, 1200, 3000, 5000 grit), ultrasonically cleaned for 20 min in absolute ethyl alcohol and ultrapure water, and finally dried at 60°C. Amino-modification was following the schedule illustrated in scheme 1. In order to reduce the carbonyl group in PEEK to hydroxyl, a total of 200 mg sodium borohydride (NaBH 4 , Jinshan Chemical Reagent Co., LTD, China) was dissolved in 100 ml dimethylsulfoxide (DMSO, CHRON CHEMICALS, China), and PEEK samples were immersed in the solution for 3 h at 120°C. Then, the samples were cleaned in absolute ethyl alcohol (CHRON CHEMICALS, China) and ultrapure water, dried for 3 h at 60°C, and named PEEK-OH. The introduction of amino groups was achieved through silanization. Briefly, PEEK-OH samples were immersed in the solution containing 5% 3-ammoniumpropyltriethoxysilane (APTES, Macklin, China), 5% ultrapure water, 90% absolute ethyl alcohol for 3 h at home temperature. After being cleaned in absolute ethyl alcohol and ultrapure water, the samples were dried for 3 h at 110°C. The samples were named PEEK-APTES. Before cellular experiments, all samples were sterilized with ultraviolet radiation for 30 min.

Characterization
The surface chemical composition of the samples was examined by x-ray photoelectron spectroscopy (XPS, Kratos, AXIS Ultra DLD, UK) with a monochromatic Al Kα source. The binding energies were calibrated by the C1s peak at 285.0 eV. The surface morphology of the samples was examined by Scanning electron microscopy (SEM, FEI, Inspect F50, USA). Before being examined, all samples were vacuum-dried and coated with gold. The water contact angle of the samples was measured to investigate the wettability using a contact angle goniometer (Dataphysics, OCA20, Germany).

Cell adhesion
After culturing for 4 h, the samples were fixed with 4% paraformaldehyde (Biosharp, China) for 30 min followed by washed three times with phosphate buffered saline (PBS). Then the samples were stained with 4′,6′-Scheme 1. Schematic diagram of APTES grafting. diamidino-2-phenylindole (DAPI, Solarbio, China) for 8 min before washed three times with PBS. The cell adhesion was observed using an inverted fluorescent microscope.
A cell counting kit (CCK-8, Dojindo, Japan) was also used to assess cell adhesion. After culturing for 4 h, the samples were washed three times with PBS. CCK-8 reagent was added to each well according to the manufacturer's guidelines. After incubation at 37°C away from light for 4 h, the optical density (OD) values at 450 nm were measured with a microplate reader (BioTeke, China).
The cell adhesion morphology was examined by a confocal laser scanning microscope (CLSM, Leica, Germany). After culturing for 4 h, the samples were fixed with 4% paraformaldehyde for 30 min, permeabilized with 0.1% Triton X-100 (Beyotime, China) for 10 min, and blocked with 5% bovine serum albumin (BSA, Beyotime, China) for 1 h at room temperature. Each well received a 200 μl solution of Actin-Tracker Green (Beyotime, China) solution. After incubation for 1 h at room temperature, the samples were washed three times with PBS containing 0.1% Triton X-100, stained with DAPI for 8 min, washed three times with sterile PBS, then imaged using a CLSM.

Cell proliferation
A CCK-8 kit was used to assess the cell growth on the samples. As described above, cell growth was measured on days 1, 4 and 7.
2.6. Alkaline phosphatase (ALP) activity assay The ALP activity was performed by staining on days 7 and 14. After the medium was removed, the samples were fixed in 4% paraformaldehyde for 30 min, stained with an NBT/BCIP ALP color development kit (Sangon Biotech, China), and examined under a stereomicroscope.
The ALP activity was also evaluated using an alkaline phosphate activity assay kit (Nanjing Jiancheng Biotechnology, China) according to the manufacturer's guidelines on days 7 and 14. The absorbance of ALP was evaluated by measuring the OD values at 520 nm. For standardization, the total protein concentration was evaluated using a bicinchoninic acid (BCA) protein assay kit (BioTeke, China). According to the manufacturer's guidelines, the OD values were measured at 562 nm and the total protein concentration was calculated according to the standard curve. Finally, the ALP activity was expressed as an activity unit (U/gprot).

Extracellular matrix (ECM) mineralization assay
The mineralized nodule formation was assessed by Alizarin Red S (ARS, Solarbio, China) staining on days 21 and 28. Briefly, the samples were fixed in 4% paraformaldehyde for 30 min, stained with 200 μl ARS solution (2%) for 20 min at room temperature, and washed three times with PBS. The calcium nodules were photographed using a stereomicroscope. For quantification, the stained samples were submerged in a 1% (w/v) hexadecyl pyridinium chloride (Solarbio, China) for 2 h under slow shaking to dissolve the absorbed alizarin red dye. The OD value of the solution at 550 nm was measured.

Real-time polymerase chain reaction (RT-PCR) analysis
The expression of osteogenic genes was evaluated by RT-PCR analysis on days 7 and 14. Total RNA was extracted using an RNAsimple Total RNA Kit (TIANGEN, China), the cDNA was reverse-transcribed from total RNA using a ReverTra Ace ® qPCR RT Master Mix (TOYOBO, Japan) according to the manufacturer's protocols.  The expression levels of runt-related transcription factor 2 (Runx2), alkaline phosphatase (Alp), and type I collagen (Col1α1) were quantified using a QPCR system (DNA Engine Opticon 2, Bio-rad, USA) with SYBR Green Realtime PCR Master Mix (TOYOBO, Japan). β-actin was used as the housekeeping gene for normalization. The primers for different genes are listed in table 1. Quantification of the gene expressions was based on the CT (cycle threshold) values. The relative RNA expression levels were expressed as a log scale.

Statistical analysis
The statistical analysis was performed using IBM SPSS Statistics 22.0. The data were expressed as means ± standard deviations (SD) and analyzed statistically using the student's t-test. p < 0.05 was considered statistically significant.

Surface characterization
To confirm that amino groups were successfully introduced onto the surface of PEEK, XPS examination was carried out. As shown in figure 1, the C1s and O1s peaks were shown both in the survey spectrum of PEEK, PEEK-APTES, but the N1s and Si2p peaks were shown only in the survey spectrum of PEEK-APTES. In the C1s spectrum of PEEK, the peaks at 284.8 eV, 286.4 eV and 287.1 eV were assigned to aliphatic carbon (C-C/C-H), C-O and carbonyl group (C=O), respectively [13,14]. However, the carbonyl group peaks disappeared in the C1s spectrum of PEEK-APTES. This indicated that the carbonyl groups in the testing region were reduced to hydroxyl groups. The O=C peak at 531.5 eV and O-C peak at 533.3 eV were detected in the O1s spectrum of PEEK, the O-C/O-Si peak and a -OH peak at 532.4 eV were detected in the O1s spectrum of PEEK-APTES [13,15]. No nitrogen peak was found in the N1s spectrum of PEEK, but a new -NH 2 peak at 399.5 eV was detected in the N1s spectrum of PEEK-APTES. These results suggested that APTES replaced some hydroxyl groups and were successfully grafted on the surface of PEEK. The surface morphologies of PEEK and PEEK-APTES are shown in figure 2(A). There were many grooves on the surface of PEEK, which might be caused by polishing. However, the surface of PEEK-APTES became smooth and homogeneous after grafting APTES. Figure 2(B) shows that the static water contact angle of PEEK was significantly higher than that of PEEK-APTES (p < 0.05). PEEK had a contact angle of 87.97 ± 1.54°, while PEEK-APTES had a contact angle of 72.23 ± 2.44°. This indicated that amino modification could increase the hydrophilicity of PEEK surface. The amino groups and the unreacted hydroxyl groups on the surface of PEEK-APTES specimens were responsible for the increase in surface hydrophilicity [6,16]. Hydrophilic surfaces were more beneficial to cell attachment. In order to verify this, we further carried out cytological experiments.

Cell adhesion and proliferation
Cell adhesion and proliferation could reflect the biocompatibility of biomaterials. As shown in figures 3(A) and (B), the PEEK-APTES showed more cell adhesion than PEEK when examined by DAPI staining. Similarly, figure 3(C) shows that the cell adhesion on the surface of PEEK-APTES was substantially higher than that on the surface of PEEK according to the CCK-8 assay (p < 0.05). In figure 3(D), the nucleus was blue fluorescent and F-actin microfilaments were green fluorescent. The cells on the surface of PEEK-APTES extended well and presented a well-developed cytoskeleton with numerous and obvious F-actin microfilaments in the cytoplasm, while the cells on the surface of PEEK were with poor extension and showed fuzzy F-actin microfilaments. As shown in figure 3(E), the cells proliferated as the culture time extended. The cell growth PEEK-APTES group was considerably more significant than that PEEK group on days 1, 4, and 7 (p < 0.05). The cell adhesion and proliferation both significantly enhanced on the surface of PEEK-APTES, which might be related to the surface wettability. Yu W et al used the plasma-enhanced chemical vapor deposition (PECVD) technique to modify carbon fiber reinforced PEEK (CPEEK) with amino groups. The amino group-modified CPEEK showed a smaller contact angle, at around 75°, which is consistent with our study [17]. There was research demonstrated that a hydrophilic surface was more beneficial to cell adhesion and proliferation in comparison with a hydrophobic surface. The attachment of cells to biomaterial surfaces was achieved through the interplay between the integral proteins on the cell surface and the attachment proteins on biomaterial surfaces. This interplay could be promoted on a hydrophilic surface [9,18]. It was also revealed that cells could change their morphology according to the surface properties, then promote the cell adhesion and proliferation through activating FAK signal pathway and up-regulating the expression of integrin α2 [19,20]. However, in our study, the surface topography might not affect the cell adhesion and proliferation as surface wettability did because the smooth surface of PEEK-APTES was a disadvantage to cell adhesion according to the previous studies [11,[21][22][23]. There were studies suggesting that the amino groups on the surface of PEEK-APTES also could affect the cell adhesion, spreading and cell proliferation because the extracellular matrix fibronectin playing a major role in osteoblast adhesion was extremely sensitive to amino groups [24]. Moreover, the MC3T3-E1 cells carrying negative charges could electrostatically adsorb onto the surface of PEEK-APTES specimens which was positively charged due to the protonation of amino groups in the aqueous solution [25].

Osteogenic differentiation
Cells' osteogenic differentiation to osteoblasts is the key to successful osseointegration around implants. ALP is an early biomarker of osteogenic differentiation, reflecting the maturity of osteoblasts. As shown in figure 4(A), the PEEK-ATPES group presented deeper ALP staining than the PEEK group on days 7 and 14. As for ALP quantification ( figure 4(B)), the ALP activity of the PEEK-APTES group was significantly higher than that of the PEEK group on day 14 (p < 0.05), while there was no statistical difference between the two groups on day 7 (p > 0.05). The reason was that only partial MC3T3-E1 cells on PEEK-APTES specimens were differentiating to osteoblasts on day 7, which might be caused by the low efficiency of the osteogenic differentiation medium. However, we indeed observed a significantly higher ALP activity in PEEK-APTES groups than that in PEEK groups on day 14, which would further promote the mineralization of ECM, an essential index of osteogenic differentiation in late stages. As demonstrated in figures 5(A) and (B), the PEEK-APTES group showed more ARS-staining mineralized matrix as well as higher quantification of ECM mineralization than the PEEK group on days 21 and 28 (p < 0.05). The expression levels of osteogenic genes are essential to evaluate the osteogenic differentiation of MC3T3-E1 cells on implants at a molecular level. As shown in figure 6, the gene expressions of Alp and Col1α1 in the PEEK-APTES group were remarkably higher than the PEEK group on days 7 and 14 (p < 0.05). These results were consistent with the results of ALP staining and quantification. However, there was no statistical difference in the gene expression of Runx2 between the PEEK-APTES group and the PEEK group (p > 0.05). The reason might be that Runx2 is a regulator of osteogenic differentiation at an early stage and the MC3T3-E1 cells used for RT-PCR assay had differentiated into mature osteoblasts at the testing time, resulting in a low mRNA level of Runx2. Anyhow, the MC3T3-E1 cells on PEEK-APTES specimens presented higher ALP activity, ECM mineralization and expression levels of osteogenic genes, indicating that amino group modification could improve the osteogenic ability of PEEK, which was consistent with the previous studies. W Yu et al constructed amino group-modified titanium surface via the PECVD technique with an allylamine monomer and found that the cell adhesion, proliferation, ALP activity, and expressions of osteogenesis-related genes substantially increased on amino group-modified titanium in contrast to those on original titanium [20]. W Yu et al also fabricated the amino group-modified carbon reinforced PEEK, which possessed enhanced bioactivity and osteogenic property [17]. The better cell spreading of the MC3T3-E1 cells on PEEK-APTES specimens might contribute to the enhanced osteogenic differentiation because a stretched cell morphology was the prerequisite of the MC3T3-E1 cells' proliferation and osteogenic differentiation [26][27][28][29].

Conclusion
Amino groups could be introduced onto the PEEK surface using a facile self-assembly method. Amino modification substantially enhanced the cell adhesion, proliferation and osteogenic differentiation of MC3T3-E1 cells on PEEK-APTES specimens in vitro. This indicated that amino modification could improve the biocompatibility and osteogenic ability of PEEK in vitro. This surface modification strategy may provide ideas for the surface modification of PEEK implants and other implant materials.

Data availability statement
The data generated and/or analysed during the current study are not publicly available for legal/ethical reasons but are available from the corresponding author on reasonable request.

Supporting information
The work was supported by Sichuan Medical Association [S19023].