Bone Mimetic Polyetheretherketone Implant Coating Facilitates Early Osseointegration

Aging society demands advanced bone implants that can achieve long‐term success through osseointegration at the bone‐implant interface. Polyetheretherketone (PEEK), a bioinert implant material, often struggles with connecting to existing bone tissue, leading to inflammation and implant replacement. Following nature's lessons, a new strategy is introduced that masks the surface of bio‐ and chemically inert Polyetheretherketone implants with a covalently anchored bone‐mimetic surface. This method transforms PEEK into a bone‐integrative material by grafting a ≈300 nm thick gelatin layer onto its surface, subsequently mineralized with calcium phosphate. Herein, it is showed that this surface modification for implant materials combines excellent bulk implant properties with the characteristic structure and functional properties of natural bone. In vitro biocompatibility assays, employing NIH‐3T3 fibroblast and MC3T3‐E1 osteoblast cell lines, confirm the enhanced biocompatibility of the modified material. This strategy offers promising prospects for improving bone implants and exemplifies the adaptation of a non‐osseointegrative material to a bone‐like interface.


Introduction
The use of bone and joint implants has witnessed a steady increase, driven by an aging population and the aspiration for an active, extended quality of life.3][4] The concern of implant loosening, in particular, poses a significant challenge in the field.In recent years, over 30000 revision surgeries were required in a single reported year, comprising 10% of total surgeries in Germany. [5,6]This issue has profound implications for both patient health and the financial burden on healthcare systems, necessitating the exploration of advanced implant materials and design strategies.
[9] Its mechanical properties closely resemble those of bone, mitigating the issues associated with stiffer materials like titanium.Additionally, PEEK exhibits low friction, corrosion resistance, and radiolucency, making it a compelling choice for various clinical applications. [9,10]espite these advantages, PEEK's inherent bioinertness can pose challenges in achieving optimal osseointegration, a crucial factor for implant success. [11]It is generally accepted that bulk properties of the implant materials are not the only critical determinants for performance, as the bone-implant interface also plays a crucial role in this process. [12,13]This interface is essential for developing new materials with improved biocompatibility and osseointegration, aiming to enhance the implant's stability in the host and reduce ingrowth failures.[16][17][18] Although many of these modifications demonstrated improvements in the biological performance of the implant materials, some of the resulting coatings possess several inherent disadvantageous properties.Delamination or diffusion of particles from the surface has been reported for coatings with a non-covalent character after long-term use. [19,20]Furthermore, such coatings neglect the biomimicry of their biological surrounding in terms of chemical composition and structure, such as titanium coatings, which can corrode and therefore reduce the implant's biocompatibility.
This study aims to address these challenges by developing innovative approaches to enhance PEEK's biocompatibility and osseointegration, ultimately improving its performance as a bone implant material.
The following sections will present our research ideas, methodologies, and results related to enhancing PEEK's potential for clinical applications.We acknowledge that titanium remains a valuable implant material in certain clinical contexts and do not seek to undermine its merits but rather focus on the strengths and challenges of PEEK.
The surface modification was obtained through chemical activation of a PEEK surface in the first step, on which gelatin, the degradation product of collagen as natural structural bone matrix, was covalently attached as a thin layer.In order to mimic the protein-calcium phosphate character of bone, the surface-bound gelatin was slowly mineralized with calcium phosphate to form a nanocomposite layer (PEEK-CaP, Figure 1).

Introduction of Carboxylic Acid Groups
The chemical activation of the PEEK-based implant surface was accomplished by reducing the keto-to hydroxy-groups (PEEK-OH) as a first step.The success of the reaction was monitored by 1 H NMR spectroscopy (see Figure S1a,b, Supporting Information).Native unmodified PEEK should only display aromatic protons.The appearance of a non-aromatic singlet-signal in 1 H-NMR spectra should directly correlate with the amount of modified PEEK material.However, due to the impurities in the sample, determining the correct benzhydryl-protons is complicated.A 1 H-spectrum displayed a singlet at  = 8.56 ppm, typical for carbocationic species, [21][22][23] like Ar-CH + -Ar, with an intensity of 0.37, indicating that 37% of the monomeric PEEK-units exhibit such a proton.For the NMR-measurements a very strong acid was used, which could have dehydrated the PEEK-OH, leaving a stable carbocation.Another signal at 3.41 ppm could be the benzhydrylproton.It is a doublet with a coupling constant of 3.1 Hz, which could be attributed to 4J spin coupling of the benzhydryl proton and aromatic protons.
The success of the reaction was further monitored with Attenuated Total Reflectance-InfraRed (ATR-IR) spectroscopy (see Figure S1d, Supporting Information).In comparison to native PEEK, the ATR-IR spectrum of PEEK-OH revealed a broad band near 3440 cm −1 , assigned to O─H stretch, a new band at 2863 cm −1 , that was assigned to C─H-stretch and a band at 1037 cm −1 that was assigned to C─O stretch of the alcohol.
The implant's surface has a significant interaction with the surrounding tissues and is influenced by the local environment.Wettability plays a crucial role in evaluating the impact of the implant on tissue integration.Therefore, in order to gain a deeper understanding of the wettability and surface free energy of the various PEEK surfaces examined in this study, the water contact angle was measured.Moreover, the surface free energies and their corresponding dispersive and polar fractions were calculated by conducting additional contact angle measurements using diiodo methane.Water contact angle measurements displayed a change in contact angle from native PEEK to PEEK-OH of 93.3 ± 1.5°to 65.2 ± 0.4°, indicating a notably more hydrophilic surface (Figure S2 and Table S1, Supporting Information).This was further indicated by the increase in the polar fraction of the surface free energy from 0.1 ± 0.1 mN m −1 of non-modified PEEK to 11.8 ± 0.2 mN m −1 of PEEK-OH (Table S1, Supporting Information).
The carboxylic acid functional group was then introduced to the surface (PEEK-COOH) by coupling succinic acid to the surface, which was prompted by N,N'-dicyclohexylcarbodiimide (DCC), and 4-dimethylaminopyridine (DMAP).The appearance of two non-aromatic triplet signals in the 1 H-NMR spectrum (Figure S1c, Supporting Information) at 2.8 and 5.2 ppm with identical integration confirmed the success of the coupling reaction.Comparing to the integration singlet at 35 ppm, representing the CH of the reduced PEEK materials, we could roughly estimate that 35% of hydroxyl groups were converted to carboxylic acid.A new peak at 1738 cm −1 was observed in the IR spectrum (Figure S1d, Supporting Information), corresponding to the ester bond formation.The PEEK-COOH surface showed a similar wettability to PEEK-OH with a water contact angle of 49.9 ± 0.2°(Figure S2 and Table S1, Supporting Information).Thus, a continuous drop in the contact angles was observed after the surface activation.This was further confirmed through the increase of the polar fraction of the surface free   S1, Supporting Information).

Attachment of Gelatin
The chemical anchorage of the protein component of the bonemimetic coating was performed next.For this, gelatin was used as an inexpensive, easy to handle and water-soluble alternative to collagen, the organic matrix in natural bone.Gelatin is derived from collagen by acid or base treatment and still retains a lot of the properties of native collagen with regard to biocompatibility and biomineralization.
X-ray photoelectron spectroscopy (XPS) measurements revealed a signal of the N1s at 403 eV (Figure 2a), which stems from the gelatin matrix on the surface.The water contact angle, which was ≈55.3 ± 1.8°(Figure S2 and Table S1, Supporting Information), did not show a large change after coupling gelatin compared to the angle of PEEK-COOH.The surface free energy showed a polar fraction of 16.57 ± 1.11 mN m −1 .
Scanning electron microscopy (SEM) (Figure 2b) and atomic force microscopy (AFM) (Figure 2c,d) revealed gelatin as a homogeneously distributed, non-ordered network on the surface.SEM images of non-modified PEEK and PEEK-gelatin, displaying larger surface areas are shown in Figure S3 (Supporting Information).
The corresponding AFM phase contrast image (Figure 2d) showed no distinct phase shift, also suggesting a homogeneously covered surface.An elemental mapping of N (Figure 2e) and staining with Cy-5 (Figure 2f) further demonstrated a homogeneous coverage of the PEEK surface with gelatin.Thermogravimetric analysis (TGA) revealed ≈2.16 wt% gelatin on the PEEK-Gel material (Figure S4, Supporting Information).

Mineralization to PEEK-CaP
In order to fully mimic the protein-calcium phosphate character of natural bone, the surface-bound gelatin layer was then mineralized to form a continuous composite layer (PEEK-CaP).A buffered solution containing CaCl 2 , K 2 HPO 4 , and pAsp as a nucleation inhibitor realized this in a slow mineralization reaction.
XPS and Energy-Dispersive -X-ray Spectroscopy (EDXS) (Figure 3a,b) investigations of the implant surface revealed that the main constituents in the composite layer are carbon, oxygen, nitrogen, calcium, and phosphate.The presence of nitrogen on the surface, as was detected by the weak XPS signal, indicates that the material has indeed a composite character.Grazing incidence X-Ray diffraction (giXRD) investigations were performed to get insight into the phase composition of the mineral component of the composite.The giXRD spectrum showed no distinct reflexes next to the PEEK-related signals, indicating that the calcium phosphate has an amorphous character or is a poorly crystalline calcium phosphate phase (Figure 3f).This suggests that the crystallization of calcium phosphate was inhibited during the mineralization process.
The precipitation of amorphous calcium phosphate (ACP) is highly beneficial for the osseointegration of bone implants since osteoblasts remodel calcium phosphate during the formation of new bone, and in this respect, ACP is better soluble and thus better suited than any other calcium phosphate phase.26] After mineralization, the surface of the PEEK showed a water contact angle of 64.4°± 3.3°, and the polar fraction of the surface free energy decreased to a value of 10.82 ± 1.87 mN m −1 (Figure 3g; Figure S2 and Table S1, Supporting Information).It has been shown that the adhesion of bone cells is higher on moderate hydrophilic substrates as they allow for higher protein adsorption on the surface.[29] Thus, the biomimetic coating has the potential for a high cytoadhesion and resulting differentiation of bone-related cells.
The SEM and AFM images (Figure 3c,d; Figure S5, Supporting Information) revealed very small and randomly oriented particles with defined grain boundaries on the surface.This is an important finding in terms of application because bigger crys-tals on the surface, which are only physically attached, tend to detach already after low mechanical stress, and thus can lead to inflammation. [30,31]SEM investigations on a surface, which was sputter coated with gold under controlled conditions and was cut via FIB, demonstrated a coating thickness ranging between roughly 200-300 nm (see Figure 3e) and is not characterized by a layer-by-layer structure of calcium phosphate on gelatin.This indicates that the CaP precipitated inside the gelatin layer.Furthermore, these investigations revealed that the coating possesses a nanoporous inner structure.Control experiments by the mineralization on the PEEK-COOH surface resulted in either several big and randomly distributed aggregates, which display a flake-like and bent topography (Figure S6a,b, Supporting Information), or randomly distributed, small needle-like aggregates (Figure S6c,d, Supporting Information).Both large structures can potentially easily detach from the implant surface and are thus an undesired mineralization outcome.These results suggest that the gelatin layer itself strongly influences the mineralization process on the implant materials surface and prevents the formation of micrometer-sized structures on the surface of the implants.
A residual mass between 0.6-0.9wt% was consistently found after a gradually increased temperature to 1000 °C in the TGA measurements (Figure S7, Supporting Information) compared to 0.3 wt% for the PEEK-Gel samples.This increase suggests the presence of calcium phosphate, which should remain as calcium oxide at the end of the measurements.
To assess the adhesion strength and durability of the biomimetic coating, a standardized scotch tape test was performed that follows established procedures in materials science.In this test, a piece of adhesive tape was applied to the surface of the coated PEEK samples and then removed fast.The test was performed with multiple replicates to ensure consistency and reliability.The purpose of this test is to evaluate the ability of the coating to withstand mechanical stresses and remain firmly adhered to the surface.The scotch tape test is a common technique used in materials science to estimate the adhesion quality of various coatings.It is based on the principle that strong adhesion between the coating and the substrate should result in minimal damage or detachment of the coating during the tape's application and removal.
The outcomes of the scotch tape test in our study demonstrated the remarkable adhesion strength of the biomimetic coating on PEEK, as evidenced by the coating's ability to maintain its integrity even after the tape test (Figure 4a).A further example of the surface on a different sample in different regions on the surface after scotch tape test are displayed in Figure S8 (Supporting Information).This was further confirmed by subsequent XPS measurements, which showed that although the signal intensity decreased after the scotch tape test, the main elements of the coating could still be detected (Figure 4b).
For materials used for medical applications in the long term, this test is a way to assess the durability and resistance against mechanical strains and abrasion, which are decisive properties.The results of the scotch tape test strengthen the applicability of the biomimetic coating for bone implants, particularly because it shows its tight bonding to the PEEK substrate.This characteristic will guarantee the reliability and efficiency of the implant covering on a practical scale.Adhesive strength can be conveniently estimated through the Scotch tape test, which mainly quantifies peel forces and may not fully convey the performance of the coating under shear.In order to overcome this limitation, we have also performed some additional tests that provide a complete evaluation of the mechanical integrity based on various loading conditions.The abrasion resistance was thus investigated by screwing surface-modified PEEK screws into pig bones.By subsequent removal and staining of the screw surface by a dye, it could be demonstrated that the coating remained stable on the surface.(Figure 4c-e).
This comprehensive evaluation further underscores the potential of the biomimetic PEEK implant coating to facilitate early osseointegration and enhance its suitability for bone implant materials.

Cell Culture Studies
Having demonstrated the successful mineralization of PEEK-CaP and introduced its surface characteristics and durability, the logical question of whether the biomimetic implant coating influences biocompatibility toward mammalian cells was investigated.The general cytotoxicity and biocompatibility of the biomimetic surface coatings were evaluated by testing two murine cell lines, a fibroblast (NIH3T3) and an osteoblast (MC3T3-E1) line, regarding their growth behavior, viability, and phenotype.Those two cell lines are commonly used in bone implant biocompatibility studies and represent two cell types typically in contact with bone implants during bone ingrowth.Osteoblasts are responsible for bone formation and play a critical role in the process of osseointegration, which is the fusion of the implant with the surrounding bone.They play a crucial role in the integration of the implant by forming new bone around its surface, which helps to anchor it firmly in place. [32,33]ibroblasts, on the other hand, are a type of connective tissue cell that plays a critical role in the formation of extracellular matrix (ECM) and wound healing.During the process of bone healing and implant integration, fibroblasts play a vital role in the formation of the fibrous tissue that surrounds the implant.This fibrous tissue acts as a scaffold for osteoblasts to deposit new bone tissue and provides mechanical support for the implant during the initial stages of integration. [32]Furthermore, fibroblasts are commonly used for in vitro toxicity testing of medical devices and implants during the initial stages of safety assessment as they are a type of cell that can be easily obtained and cultured in the laboratory.
Overall, the coordinated action of these and other cell types, such as osteoclasts, mesenchymal stem cells, macrophages, and endothelial cells, is crucial for successful implant integration and the formation of new bone tissue. [32]ithin an approach to assess the general cytotoxicity of the bone implant coating, the materials were evaluated for their indirect and contact toxicity on the cells.The indirect influence was assessed by exposing osteoblasts and fibroblasts to culture medium, which was incubated with PEEK-CaP and evaluated for their survival with a Fluorescence-activated Cell Sorting (FACS) based life-death staining.The overall death rate did not differ significantly from the control in both cell lines, suggesting a high treatment tolerance.(Figure S9a,b, Supporting Information) Furthermore, the assay showed no significant difference regarding lethality between PEEK and PEEK-CaP for fibro-and osteoblasts.Similar results were observed in a direct toxicity testing, where both cell types were incubated with PEEK or PEEK-CaP coating.(Figure S9c,d, Supporting Information).Surprisingly, both cell lines showed a significantly increased metabolic activity of ≈1.5 to 3-fold when incubated on PEEK-CaP compared to unmodified PEEK (Figure S9e,f, Supporting Information).This result can be partially, but probably not solely attributed to the elevated proliferation.Thus, these results confirm that no toxic substances were extracted from the material, and the overall cytotolerance of PEEK-CaP toward cells relevant to the implant-bone interface is imminent.
To determine the proliferation difference in PEEK and PEEK-CaP, both cell lines were allowed to adhere, propagate and proliferate on the respective coating for one day.After this, they were removed from the material, and the cell number was determined.While the fibroblasts showed a significantly increased proliferation when grown on PEEK-CaP (Figure 5a, right), this tendency could not be seen for osteoblasts despite an apparent increase in their metabolic activity (Figure 5a, left, and Figure S9e,f, Supporting Information).
Therefore, the distinct adhesion properties of fibroblasts and osteoblasts were investigated through two different methods.First, the cells were incubated on the respective material for 6 h.After this time, non-adhered cells were removed, and the number of adherent ones was determined.For the fibroblasts this assay showed that the cells could adhere significantly better to the modified material by a factor of 2.3 in comparison to the non-modified PEEK (Figure 5b) (P = 0.0001 without serum in the medium and P = 0.0020 with serum).This is consistent with the results of the water contact angle measurements and the determination of the surface free energy.[36] Herein, an optimal contact angle between 60-70°, which fits to the contact angle for PEEK-CaP, has been shown to increase the attachment and proliferation rates for fibroblasts, especially NIH3T3 fibroblasts. [27,28]his assay was performed in the presence and in the absence of serum supplement to the cell culture medium, to estimate if and to which extent the adherence of the cells is influenced by serum proteins, which might adsorb to the surface.The results of this assay showed no significant difference in the cell attachment to the surfaces with or without serum present in the culture medium (P = 0.1682 for the PEEK measurements and P = 0.1682 for PEEK-CaP).However, during the experimental procedure, it became apparent that osteoblast cells adhere firmly to the PEEK-CaP surface, and the detachment process through trypsination was incomplete.Therefore, to further assess osteoblast adherence, the cells remaining on the implant material after trypsination were stained with crystal violet (Figure 5c).Accordingly, osteoblasts were exposed to a surplus of trypsin and incubated for an extended period.Even after this extended trypsination treatment, osteoblasts remained firmly attached on the PEEK-CaP coating, while they lost attachment to the PEEK coating (Figure 5c).
A close look at the spreading and morphology of the osteoblast (Figure 5d) and fibroblast (Figure S9g,h, Supporting Information) cells by SEM further corroborated our observation from the crystal violet staining experiments.Osteoblasts, which were grown on PEEK-CaP were flat, densely grown, and spread on the substrate, suggesting that the cells strongly adhered to the surface.In contrast, on the pure PEEK surface, there was no full surface coverage, the cells were smaller in size and some cells were rounded (Figure 5d).Similarly, fibroblasts were flat and were very densely grown on the modified PEEK-CaP, while several spots with detached fibroblasts could be observed on the pure PEEK surface (Figure S9g,h, Supporting Information).Some cells may have been washed away during the sample preparation due to reduced attachment of the fibroblasts on uncoated PEEK (Figure S9g,h, Supporting Information).This altered growth behavior could be attributed to the hydrophobic PEEK surface compared to the hydrophilic PEEK-CaP coating.Interestingly, we observed strong extracellular matrix deposition in osteoblast cultures when grown for 24 h on the PEEK-CaP surface (Figure 5d), indicating an early differentiation stage. [37,38]This remarkable functionalization of the surface by the osteoblasts appears to be the first step toward natural bone formation on the PEEK-CaP surface.

Conclusion
In conclusion, our study has successfully employed a costeffective wet-chemical approach to create a covalently bonded 3D continuous biomimetic coating, with a sub-micrometer thickness, on the bio-inert polymer PEEK.This coating has exhibited an excellent combination of surface-free energy and mechanical stability, which are the essential criteria for long-term use.Significantly, the viability-, adhesion-and proliferation assays with fibroblast and osteoblast cell lines showed a significant increase in biocompatibility after coating compared with pure PEEK.This now enables the use of PEEK as osseointegrative bone implant material without the disadvantage of bio-inertness in the past.That way, full advantage can be taken of its mechanical compatibility with bone and the absence of any corrosion or other chemical reactions in the body, which handicap titanium implants.
Taken together, our results provide a first step toward the design and engineering of covalently bonded biomimetic "bonelike" coatings on synthetic polymeric biomaterials and pave the way for the preparation of biomimetic biomaterials for long-term use.

Experimental Section
Materials: A thin PEEK film (0.25 mm thickness) was obtained from Invibio Ltd.UK.Polyaspartic acid (pAsp, molecular weight 3800 g mol −1 ) was obtained from Bayer chemicals.Gelatin (type A, bloom ≈500) was obtained from Gelita.The other chemicals were purchased from Sigma-Aldrich and used directly for the reaction without further purification.
NMR Spectroscopy: NMR measurements were performed with a Bruker Avance III 400 with a B-ACS60 autosampler, BCU05 VT cooling unit and automation with Bruker ICON-NMR software, operating at frequencies of 400 MHz for 1H nuclei.Acquired spectra were evaluated with MestReNova-software.The measurements were performed in PTFE-NMRtubes with CF 3 SO 3 D as a solvent.
Attenuated Total Reflection Infrared (ATR-IR) Spectroscopy: ATR-IR measurements were conducted with a Perkin Elmer Spectrum 100 FT-IR Spectrometer equipped with a Universal ATR-Sampling Accessory.
X-Ray Photoelectron Spectroscopy (XPS): XPS measurements were performed either with a PHI Quantera SXM or a PHI 5800 MultiTechnique ESCA System.The PHI Quantera SXM was equipped with an aluminum anode (15 kV, 1486.6 eV) and a quartz monochromator.The direction of photoelectron collection was made in 45 °angles normal to the sample and the incident X-ray beam.To compensate for charging effects, the binding energies from the Mo1S component of the sample holder was used as a reference.The spectra were analyzed using the CasaXPS software package.The PHI 5800 MultiTechnique ESCA System was operated with monochromatised Al-K (1486.6 eV) radiation.The detection angle of the measurement was 45°and pass energies of 93.9, and 29.35 eV were used for survey and detailed spectra, respectively.Sample charging was neutralized with an electron flood gun, and the XP spectra were calibrated to the signal of the main C1s peak at 284.8 eV.
Contact Angle Measurements: Contact angle measurements were performed on an automated Krüss Advance Drop Shape Analysis system.The surface free energy with polar and disperse fractions was determined by these optical contact angle measurements with water and diiodomethane.All measurements were performed at room temperature without any pre-treatment of the samples.For each sample, at least 20 measurements were performed and averaged for one drop.Water drop volume was 2 μl, and diiodo-methane drop volume was 0.8 μl.The disperse and polar fractions were calculated by the Advance evaluation software package according to Owens, Wendt, Rabel, and Kaelble. [39,40]hermogravimetric Analysis (TGA): Thermogravimetric analyses were performed using a Netzsch simultaneous thermal analyzer 449 F3 Jupiter.The heating rate was kept at 10 K min −1 under oxygen atmosphere using aluminum oxide crucibles.Decomposition steps were analyzed using the Netzsch Proteus 6.1 software package.
Scanning Electron Microscopy (SEM) /Energy Dispersive X-Ray Spectroscopy (EDX) Focused Ion Beam (FIB): All PEEK samples were prepared for scanning electron microscopy by mounting the dried specimens on an aluminum stub covered with double-sided tape.The samples were analyzed with a CrossBeam 1540XB from Zeiss and a Hitachi TM3000.

Sample Preparation of Cells on PEEK for SEM:
To visualize the morphology of osteoblast and fibroblast cells, the cells were grown on different PEEK samples for 24 h.The samples were then washed three times with PBS.Fixation was performed in a three-step way in a 2% glutaraldehyde solution (in 0.1 m Na-cacodylate buffer, supplemented with 0.01 m CaCl 2 , and 0.01 m MgCl 2 , pH 7) at room temperature.The fixation solution was changed after 10, 20 min, and 1 h, respectively.Afterward, the samples were washed three times with PBS.In preparation for critical point drying, the samples were incubated in a graded series of aqueous ethanol solutions (30%, 50%, 70%, 80%, 90%, 96%, and 100%) for 15 min and again two times in 100% ethanol for 30 min each.Critical point drying was performed with a CPD 030 (BAL-TEC AG, Balzers, FL).The samples were coated with a 3 nm platinum layer.SEM analysis was conducted with a ZEISS Auriga TM Crossbeam Workstation and ZEISS CrossBeam 1540XB.Both operated in a high vacuum using variable acceleration voltages.
Atomic Force Microscopy (AFM): AFM measurements were performed on a JPK nano wizard II ultra.AFM scans were performed in tapping mode.Data visualization was performed with the JPKSPM Data Processing software.
Grazing Incidence X-Ray Diffraction (giXRD): The phase composition of the surface composite was determined by giXRD analysis.The data were collected at room temperature with a Bruker D8 Discover with Cu-K radiation equipped with a VÅNTEC-500 2D detector at an incidence angle of Cy-5-Staining Samples: PEEK samples were immersed in a 0.1 mg ml −1 Cy5-solution (in DMSO: MilliQ water 1:9) at room temperature for 4 h in the dark.Afterward, the samples were washed with ethanol 3 times to remove the non-bound dye.
Cy-5-Staining Cells: General cell culture and assay conditions.NIH3T3 mouse fibroblasts were cultured in 7.5 cm 2 culture dishes in Dulbecco's Modified Eagle Medium (DMEM) with 10% fetal calf serum (FCS) and 1% penicillin/streptavidin at 37 °C and 5% CO 2 .Cells were split and verified for normal growth and morphology every two or three days.
All PEEK-test samples were cut into discs with 1.2 mm diameter and were sterilized either by autoclaving or incubation in 70% EtOH (in H 2 O) and dried with an open lid under a sterile hood.All assays were performed in 24-well plates.All growth and incubation periods were unless otherwise stated, performed in a humidified air atmosphere with 5% CO 2 at 37 °C.Unless otherwise stated, all tests were conducted in three technical replicates.
Statistics for Cell Culture: All data were presented as mean ± S.D. (standard deviation) for n = 3 unless stated otherwise.Statistical analyses were performed by t-test.Differences were considered statistically significant for p ≤ 0.05.
Proliferation Assay: The cells were incubated on the respective PEEK samples for the mentioned time in supplemented -MEM.After incubation, the cell culture medium was removed, and the samples were carefully rinsed twice with PBS.The remaining cells were detached by trypsination (trypsin/EDTA 0.05%/0.02%from Biochrome), spun down at 800 rpm, and resuspended in fresh cell culture medium.The cell concentration was determined using an Innovatis Casy Cell Counter.
Attachment Assay: The cells were incubated on the respective PEEK samples for 6 h in -MEM.After incubation, the cell culture medium was removed, and the samples were carefully rinsed twice with PBS.The remaining cells were detached by trypsination (trypsin/EDTA 0.05%/0.02%from Biochrome), spun down at 800 rpm, and resuspended in fresh cell culture medium.The cell concentration was determined using a counting chamber.
Crystal-Violet Staining: The cells were incubated on the respective PEEK samples for 2 days.After incubation, 100 μL crystal violet solution (5% in ethanol) was added per well of the 24 well plate.Afterward, the samples were washed with PBS until the washing solution remained clear.Cell density and spreading were evaluated using a Nikon Eclipse TS100 microscope.
Indirect Toxicity Assay: Dead cells were detected using a propidium iodide (PI) flow cytometric assay.The different PEEK materials were incubated for 1 day at 37 °C and 5% CO 2 in -MEM medium.The next day, cells were seeded into the wells in 24 well plates and incubated with 1 mL of the extract for 2 days at 37 °C and 5% CO 2 .The supernatant was collected from single wells and transferred into a test tube.The adherent cells were then washed twice with PBS and detached from the surface using trypsin/EDTA mixture.The detachment reaction was stopped using the medium containing serum.The detached cells were also transferred to the corresponding test tubes.They were centrifuged at 800 rpm for 5 min.The supernatant was discarded, and cells were taken up in 1 mL PBS +2% FCS +10 μg mL −1 PI (in PBS) and transferred into a FACS measurement tube.Heat shock-treated cells (incubation at 65-70 °C for 10-15 min) were used as a positive control.The measurement was performed using a BD LSR II FACS utilizing the DiVa software (BD Bioscience) and analyzed with FlowJo software.
Direct Toxicity Assay: Dead cells were detected using a propidium iodide (PI) flow cytometric assay.PEEK and PEEK-CaP slips were washed with acetone, dried, and placed in 24 well plates.The slips were subsequently sterilized with 70% ethanol and dried with an open lid under a sterile hood.Cells were seeded into each well in 24 well plates and incubated for 2 days at 37 °C and 5% CO 2 .The supernatant was collected from the single wells and transferred into a test tube.The adherent cells were then washed twice with PBS and detached from the PEEK surface using trypsin/EDTA mixture.The detachment reaction was stopped using the medium containing serum.The detached cells were then also transferred to the corresponding test tubes.They were centrifuged at 800 rpm for 5 min.The supernatant was discarded, and cells were taken up in 1 mL PBS +2% FCS +10 μg mL −1 PI (in PBS) and transferred into a FACS measurement tube.Heat shock treated cells (incubation at 65-70 °C for 10-15 min) were used as a positive control.The measurement was performed using a BD LSR II flow cytometer using the DiVa software (BD Bioscience) and analyzed with FlowJo software.
Viability Assay: Viability was determined by a 3-(4,5-dimethyl-2thiazolyl)−2,5-diphenyl-2H-tetrazolium bromide (MTT) assay.Viable cells with an active metabolism could convert the MTT into purple colored formazan crystals, which precipitate inside the cells.These crystals were extracted from the cells by the addition of 500 μl isopropanol in each well and incubated for 12 h in the dark.Afterward the absorbance was measured with a spectrophotometer.After incubation with the test samples, the attached cells were incubated with a 5 mg mL −1 MTT solution in PBS for 1-2 h and extracted with isopropanol overnight at room temperature.The OD was analyzed at 540 nm with a Thermo Scientific Varioskan Flashspectral photometer.Cell culture medium without cells was used as a negative control, and cell culture wells with no PEEK-test samples were used as positive control.

Figure 1 .
Figure 1.Schematic representation of the chemical activation and biomimetic coating of the PEEK implant surface.

Figure 2 .
Figure 2. a) XPS spectra of the PEEK and PEEK-Gel surface.b) SEM image of the PEEK-Gel surface; scale bar.c,d) Corresponding AFM height (c) and phase (d) images of the PEEK-Gel surface; scale bars.e) Nitrogen EDXS mapping of the PEEK-Gel surface.f): photographic images of unmodified PEEK i), PEEK-gelatine ii), and Cy-5-stained PEEK-gelatine iii).

Figure 3 .
Figure 3.Chemical characterization of the nanocomposite coating on PEEK.a) XPS and b) EDX spectra of the PEEK-CaP surface after mineralization.c) SEM image of the PEEK-CaP surface d) AFM height image of the PEEK-CaP surface (lower) with corresponding height profile (upper) along the indicated line.e) SEM image of a FIB-cut of the coating.f) giXRD diffractogram of the PEEK and PEEK-CaP surface.g) Water contact angle image of PEEK-CaP.

Figure 4 .
Figure 4. Different mechanical stability tests demonstrated good stability of the coating on PEEK.a) SEM of PEEK-CaP after scotch tape test.b) XPS of PEEK (black) and PEEK-CaP after scotch tape test (red).c) pig bone with two screws within the drilled holes.d) example of a bone screw mounted on the screwdriver.e) images of screws before screwing i), stained with Cy5 before the drilling process ii), non-stained sample after the screwing process iii), and stained sample after the screwing process iV); scale bar:4 mm.

Figure 5 .
Figure 5. Proliferation and attachment of bone-related cells on PEEK and PEEK-CaP.a) Proliferation assay of osteoblasts and fibroblasts by counting the number of cells detached from PEEK and PEEK-CaP surfaces from three replicates; error bars show SD ± SEM; MC3T3-E1-osteoblasts (left) NIH3T3fibroblasts (right) b) Cellular attachment assay of fibroblast on PEEK and PEEK-CaP, with and without serum proteins (three replicates), error bars display SD ± SEM. c) PEEK and PEEK-CaP slips before (left) and after trypsination (right) of osteoblasts.d) SEM micrographs displaying cell spreading of osteoblasts on PEEK (top) or PEEK-CaP (bottom), respectively, after 24 h of cell incubation.Images are pseudo-colored for better visualization Arrow indicates collagen fibers.For gray-scale images, see Figure S10 (Supporting Information).* indicate significant differences of the PEEK-CaP with the unmodified PEEK by student t-test with ns = not statistically significant, *** p < 0.001, ** p < 0.01.
1.5°G eneral Protocols for the Preparation of the Biomimetic Surface Coating: Preparation of PEEK-OH: NaBH 4 (37.8 mg, 1 mmol) was dissolved in DMSO (50 mL) and heated to 120 °C under stirring.The PEEK films (10 films, each 1 × 1 × 0.25 cm) were immersed in the solution and stirred at 120 °C for 4 h.The film samples were then washed with MeOH (30 min), H 2 (30 min), HCl (0.5 m 25 min), and finally H 2 O (30 min).The obtained films of PEEK-OH were dried in a vacuum oven for 4 h at 40 °C and 50 mbar.Preparation of PEEK-COOH: Succinic acid (0.12 g) was first dissolved in tetrahydrofuran (THF), and then N,N'-Dicyclohexylcarbodiimide (DCC, 0.2 g) and 4-dimethylaminopyridine (DMAP, 1 mg) were added to the ice-cold THF solution.After 4 h, the PEEK-OH films (10 films, each 1 × 1 × 0.1 cm) were added to the solution.The reaction was monitored by IR spectroscopy at different time intervals and finished after a reaction time of 4 days.The obtained PEEK-COOH films were washed three times in THF, three times with acetone, and finally dried in the vacuum oven.Preparation of PEEK-Gel: 1-Ethyl-3-(3dimethylaminopropyl)carbodiimide hydrochloride (EDC HCl, 0.89 mmol) and 1-Hydroxy-2,5-pyrrolidinone (N-Hydroxysuccinimide, NHS, 1.07 mmol) were dissolved in 50 mL ice-cold MES-buffer (0.1 m, pH 4.9) PEEK-COOH film samples were added to the solution.After a reaction time of 4 h, PEEK samples were transferred to a solution of gelatin (50 mL, 5 wt%) stirred at 37 °C for 3 days.The film samples were then washed successively with H 2 O (40 °C, 30 min) four times.The obtained PEEK-gelatin samples were freeze-dried in a Christ Alpha 2-4LSC freeze dryer.Preparation of PEEK-CaP: The PEEK-gelatin films were immersed in a solution of HEPES-buffer (10 mm, pH 7.4), containing CaCl 2 (2.7 mm), K 2 HPO 4 (1.35 mm) in the presence of Polyaspartic acid (pAsp, 10 μg ml −1 , molecular weight 3800 g mol −1 ) at 37 °C for 7 d.Afterward, the film was washed with water three times and dried in a vacuum oven for 4 h at 40 °C and 50 mbar.