Synergistic effects of elastic modulus and surface topology of Ti-based implants on early osseointegration

Abstract

Early osseointegration plays a crucial role in determining the therapeutic efficacy of orthopedic implants. Some factors responsible for early osseointegration, such as inherent mechanical properties and surface topology of implants, are well-characterized. However, the synergistic effects of elastic modulus and surface topology of implants on the osteogenic differentiation of stem cells and early osseointegration have not been thoroughly investigated. In this study, the titanium (Ti) and β-titanium alloy Ti–24Nb–4Zr–8Sn (TNZS) were used to evaluate the synergistic effects of elastic modulus and surface topology on the biological performance of these materials in vitro and in vivo. Scanning electron microscopy imaging confirmed the presence of a micro-scale porous oxide layer on the Ti and TNZS surfaces upon treatment with microarc oxidation (MAO), which resulted in increased surface roughness, enhanced surface wettability and favourable mechanical properties. As compared with Ti-MAO, the TNZS-MAO samples with lower elastic modulus displayed increased cell attachment, alkaline phosphatase activity, collagen secretion, osteogenic marker expression, and mineralization of rat bone marrow mesenchymal stem cells. Upon implantation in rat femoral condylar defects, an inherently low elastic modulus could cooperatively accelerate early osseointegration and maturation of trabecular bone after 4 weeks implantation with the MAO modified surface. These results demonstrated that elastic modulus and the surface micro-scale topographical structure of Ti alloy implants have a synergistic effect on their osseointegration.

---

Introduction

Osseointegration, involving close interaction between host bone and the implant surface, plays a crucial role in determining the therapeutic efficacy of orthopedic implants.^1–3 Recently, several clinical trials have demonstrated that insufficient osseointegration often resulted in a high rate of implant loosening and subsequent failure.^4,5 To achieve improved osseointegration, various surface modifying approaches including sand blasting, acid etching and surface coating have been investigated. Sand blasting has been demonstrated to promote the osseointegration of implants through imparting a rough surface topography. However, after treatment with sand blasting, residual constituents such as Al₂O₃ and Ta₂O₅, have been proven to be detrimental to normal bone mineralization.⁶ To address this issue, surface modification by acid etching is usually used to eliminate the residual constituents resulting from sand blasting.^7,8 It has been well-documented that such surface modifications enhanced osteoblast differentiation and increased new bone formation.^9,10 However, the bioactivity of such modified surface was often inadequate to facilitate osseointegration. Imparting a surface coating has been recognized as an effective technique to endow the implants with appropriate bioactivity. Plasma spraying with hydroxyapatite, the most commonly reported technique, could induce new bone formation around the implants.¹¹ Nevertheless, delamination of the surface coating occasionally occurs during implant placement or upon load-bearing, which hinder their widespread clinical application.¹² Therefore, implants should ideally have both stable surface hierarchical structure and adequate bioactivity for clinical applications in regenerative medicine and orthopedic surgery.

Recently, microarc oxidation (MAO) technique has been used successfully to impart highly stable and bioactive coatings onto various material surfaces.^13–15 Moreover, implant surfaces treated by MAO exhibited microscale porous structure which was conducive for osteogenic differentiation of stem cells.^16,17 More recently, our co-operator Wei et al. has demonstrated that MAO treatment could significantly enhance new bone formation and promote the osseointegration of titanium implants.¹⁶ Therefore, MAO coating may be one of the best methods for modifying titanium-based implant surface to enhance their bioactivity.

Titanium (Ti)-based implant materials have been utilized widely in dental and orthopedic research. Numerous studies have shown that progressive bone resorption around implants often occurs, which could affect the long-term stability of implants and lead to their eventual failure.^18,19 Based on previous studies, mismatch in the mechanical properties of implants and host bone tissue has been identified as a major cause of bone resorption and implant loosening due to stress shielding.²⁰ Therefore, to avoid bone resorption arising from stress shielding, recent studies have focused on the design of implants with low elastic modulus as similar as possible to natural bone. Currently, a novel β-titanium alloy Ti–24Nb–4Zr–8Sn (TNZS) has been developed with a better balance of high mechanical strength and low elastic modulus than most other titanium alloys reported so far.²¹ The low elastic modulus of TNZS could help to evenly distribute compression force during loading, so as to maintain a well-adapted and stable implant–bone interface.²¹ By significantly diminishing stiffness mismatch (stress shielding) associated with the risk of aseptic loosening, TNZS is thus a promising candidate to enhance biomechanical compatibility to facilitate implantation.

Based on the advantages of MAO coating and the excellent mechanical properties of TNZS, the objective of this study was to investigate the synergistic effects of micro-scale surface topography and low elastic modulus on osteogenic induction and osseointegration of Ti-based implants. The interactive response of bone marrow mesenchymal stem cells (BMSCs) to pure Ti, TNZS, MAO treated Ti (Ti-MAO) and MAO treated TNZS (TNZS-MAO) in terms of adhesion, proliferation, osteogenic differentiation and mineralization were investigated in vitro. Additionally, the efficacy of these materials in promoting early osseointegration in situ within rat femur was also evaluated comparatively.

Experimental

Sample preparation

Ti and TNZS disks with 20 mm in diameter and 2 mm thick were polished with different grade abrasive papers, washed with acetone and distilled water. The MAO was performed as described in a previous study.¹⁵ Briefly, the polished Ti and TNZS disks were immersed in fresh electrolyte containing Ca(CH₃COO)₂·H₂O (8.8 g L⁻¹), Na₂SiO₃ (7.1 g L⁻¹), Ca(H₂PO₄)₂·H₂O (6.3 g L⁻¹), EDTA-2Na (15 g L⁻¹) and NaOH (5 g L⁻¹) at 400 V. Ti and TNZS disks were used as anodes, and stainless steel plates were used as cathodes in the electrolytic bath. The applied frequency, work cycle and oxidizing time were 600 Hz, 8.0% and 5 min, respectively.

Sample characterization

The surface and representative cross-sectional morphology of samples was examined by scanning electron microscopy (SEM, Hitachi S-4800, Japan) at an accelerating voltage of 15 kV. Energy-dispersive X-ray spectroscopy (EDS) was performed to validate the chemical elements composition using an EMAX EX-300 system (Horiba, Japan) connected to the scanning electron microscope. The phase composition and structure of samples were analyzed by X-ray diffraction (XRD; D8 Advance, Japan) using Co Kα as the radiation source, at a scanning speed of 0.5 min⁻¹. The mechanical properties were determined by a nanoindentation testing system (Nanoindenter XP, MTS, USA) with a well-calibrated berkovich diamond indenters. The maximum indentation depth and displacement resolution are 500 μm and 0.01 nm, respectively. The maximum load and load resolution were 500 mN and 50 nN, respectively. The change of temperature was kept below 1 °C and the diamond tip drift limit was set at 0.25 nm s⁻¹. All continuous stiffness measurements were carried out under displacement control mode with tip-displacement rate of 10 nm s⁻¹. According to the Oliver and Pharr method, the elastic modulus and hardness were analyzed for a constant depth of 1000 nm.²² Topographical analysis with an interferometer (AEP, USA) was performed on discs of four groups. Surface water contact angles of Ti materials were measured using a drop shape analysis system (Filderstadt OCA15, Germany).

Protein adsorption assay

The samples were immersed in 1 mg mL⁻¹ bovine serum albumin (Solarbio, China) for 24 h at 37 °C. After incubation, the samples were washed by phosphate buffered saline (PBS) three times. Then the adsorbed protein were detached by 1% sodium dodecyl sulfate (Biolink, China) and measured using MicroBCA protein assay kit (Beyotime, China).

Cell culture

Rat BMSCs (rBMSCs) were cultured in DMEM basal medium supplemented with 10% (v/v) fetal bovine serum (FBS), 100 IU per mL penicillin–streptomycin, and 2 mM glutamine (all from Cyagen Biosciences, Inc., China). After two passages, cells were seed at 5 × 10⁴ cells per mL.

Cell adhesion and proliferation

To assess the initial adhesion of rBMSCs, the cells were incubated with Ti and TNZS disks for 0.5, 1 and 2 h. After incubation, the cells were fixed with 4.0% paraformaldehyde and stained with 40,60-diamidino-2-phenylindole (DAPI, Sigma, USA) for 10 min. Then the adherent cells were counted in five random fields within each sample under laser scanning confocal microscopy (Zeiss, LSM 780, Germany).²³

To evaluate proliferation of rBMSCs on various samples, the cell counting kit-8 (CCK-8) assay was performed on 1, 3, 5, 7 days after seeding, according to the manufacturer's instructions. Briefly, the cell were incubated for 2 h in culture medium supplemented with 10% (v/v) CCK-8 solution, and the absorbance was measured with a microplate reader (BioRad, Hercules, CA, USA).

Morphological observation

The cell morphology on the various samples were observed with SEM. After culture for 1 day, the samples were washed three times with PBS, fixed in 2.5% glutaraldehyde (w/v), incubated with 0.18 mol L⁻¹ sucrose and dehydrated through a graded ethanol series (30–100%), followed by air drying.

For focal adhesion staining, adherent cells were fixed with 4.0% (w/v) paraformaldehyde, incubated with the primary antibody of vinculin (Abcam, ab18058, UK) for 1 h and then stained with FITC-conjugated goat anti-rabbit antibody (Abcam, ab6825, UK) for 1 h followed by DAPI staining. The images were captured by a laser scanning confocal microscope (Zeiss, LSM 780, Germany). Integrated optical density (IOD) of vinculin on samples was analyzed via the Image-pro Plus 6.0 software.

Osteogenic differentiation and mineralization

Alkaline phosphatase (ALP) activity. After culture for 3 and 7 days, the adherent cells were stained with the BCIP/NBT alkaline phosphatase color development kit (C3206, Beyotime, China) according to manufacturer's instructions. To quantify the ALP activity, an Alkaline Phosphatase Assay Kit (Abcam, UK) was utilized. Culture supernatants were incubated with alkaline buffer and p-nitrophenyl phosphate for 60 min and then the reaction was terminated with the stop solution. Finally, the absorbance of the reaction mixture was measured at 405 nm.

Collagen secretion. After culture for 10 days, the cells were washed with PBS, fixed in 4.0% (w/v) paraformaldehyde and then stained with 0.1% (w/v) sirius red (Sigma, USA) to detect the secreted collagen. For quantitative analysis, the sirius red was dissolved in 0.2 M NaOH/methanol and the absorbance was measured at 540 nm.

Alizarin red staining. To evaluate ECM mineralization, alizarin red staining was performed after 14 days of osteogenic induction culture. After washing and fixation, the co-cultured cells was incubated with 1% (w/v) alizarin red S (pH 4.1–4.5, Sigma, USA) for 30 min at room temperature. For quantitative analysis, 10% (w/v) cetylpyridinium chloride (Sigma, USA) in 10 mM sodium phosphate was used to dissolve the alizarin red stain on the specimen and the absorbance was measured at 620 nm.

Osteogenic gene expression. The quantitative reverse transcription polymerase chain reaction (qRT-PCR) was performed to assay the expression levels of osteogenesis related genes. rBMSCs cultured on samples were harvested after 4 and 7 days of culture by TRIzol (Invitrogen, USA). The extracted RNA was reverse transcribed into complementary DNA (cDNA) utilizing reverse transcriptase in accordance with the manufacturer's instructions (Toyobo, Japan). The qRT-PCR analysis of genes including ALP, osteopontin (OPN), bone sialoprotein (BSP) and bone morphogenetic protein 2 (BMP-2) was performed on the Applied Biosystems 7500 thermocycler utilizing the Quantitect Sybr Green Kit (Roche, Switzerland). The primers for the target genes are listed in Table 1.

Table 1 Primers sequences utilized for real time RT-PCR analyses

Target gene	Forward primer sequence (5′–3′)	Reverse primer sequence (5′–3′)
ALP	CACCAACGTGGCCAAGAACAT	AGGGGAACTTGTCCATCTCCA
OPN	TTTGCTTTTGCCTGTTCGGC	AGTCATCCGTTTCTTCAGAGGAC
BSP	GAGAACGCCACACTCTCAGG	GAGCCTTGCCCTCTGCATCT
BMP-2	TGCTCAGCTTCCATCACGAAG	TCTGGAGCTCTGCAGATGTGA
GAPDH	GGTCGGTGTGAACGGATTTGG	GCCGTGGGTAGAGTCATACTGGAAC

---

Immunofluorescence analysis. Cells were fixed with 4.0% (w/v) paraformaldehyde, permeabilized with 0.1% (w/v) Triton-X 100 and blocked with 5% (w/v) bovine serum albumin (BSA). The cells were incubated with primary antibodies against BSP (Abcam, ab52128, UK) and BMP-2 (Abcam, ab14933, UK) for 2 h, respectively. Subsequently, TRITC-conjugated goat anti-rabbit antibody (Abcam, ab50598, UK) was used as the secondary antibody. Finally, cell nuclei were stained with DAPI. The images were captured by a laser scanning confocal microscope (Zeiss, LSM 780, Germany). Integrated optical density (IOD) of images was analyzed via Image-pro Plus 6.0 software.

In vivo osseointegration evaluation

Animals and surgical procedures. All animal experiments were conducted in accordance with the Animal Care and Use Committee of Peking University. A total of twelve 10 week-old male Sprague-Dawley rats were used in this study. All the rats were randomly divided to four groups corresponding to Ti, Ti-MAO, TNZS and TNZS-MAO. Firstly, the rats were anesthetized by intraperitoneal injection of 40 mg kg⁻¹ pentobarbital sodium solution. Teeth plant equipment was used to machine a hole of dimension 2 × 4 mm³.

After removing the bone chippings, the samples were implanted and the muscular fascia, subcutaneous tissue and skin were sutured in sequence. The rats were sacrificed at 4 weeks post-implantation. The femurs were then detached from the rats and fixed in 10% (v/v) neutral buffered formalin, prior to further investigation.

Microcomputed tomography (μ-CT) analysis. The specimens were placed and scanned under μ-CT as previous studies.^24,25 Files were reconstructed using a modified Feldkamp algorithm, which was created using a microtomographic analysis software (Tomo NT; Skyscan, Belgium). Next, three phases: bone, bone marrow and implant were obtained by filtering and segmenting μ-CT data. With the chosen concentric 500 μm radius ring, the new bone formation surrounding the cylindrical implant were calculated.

Histological analysis

Tissue processing and sectioning were carried out as previously described.²⁶ Briefly, after dehydration in a graded ethanol series (70–100%), tissue samples were embedded in methyl methacrylate (MMA) and sectioned at 20 μm thickness. Then toluidine blue staining was performed separately on tissue sections, and histological analysis was carried out under light microscopy (CX21, Olympus, Japan). Bone volume fraction (bone volume/total volume, BV/TV) measurements were performed on the chosen concentric 500 μm radius ring and bone-to-implant contact (BIC) values were calculated.

Statistical analysis

All data were expressed as mean ± standard deviation (SD). The one way ANOVA was performed for statistical analysis. The threshold for statistically-significant differences between data sets was set at *p ≤ 0.05.

Results

Characteristics of samples

SEM images are shown in Fig. 1. Some minor scratches and pits due to the polishing treatment, were clearly seen on the Ti surface (Fig. 1Aa). Nevertheless, both Ti-MAO and TNZS-MAO displayed rough surface morphology with porous micro-scale structure (insets of Fig. 1Ab and 1Ad). EDS spectra showed that Ca, P and Si elements were detected in both the Ti-MAO and TNZS-MAO samples, and that there were no obvious differences in the element content between the two groups (Fig. 1Bb and 1Bd).

Fig. 1 SEM images (A) and EDS spectra (B) of different samples: (a) Ti, (b) Ti-MAO, (c) TNZS, and (d) TNZS-MAO.

Then the phase composition of samples were analyzed by XRD, as presented in Fig. 2A. The peaks of titanium, anatase and amorphous phase were detected on MAO coatings. The results revealed that the MAO coating of Ti-MAO sample were mainly consisted of anatase and amorphous phase, while the MAO coating of TNZS-MAO was mainly consisted of amorphous phase. Fig. 2B showed the representative cross-sectional morphologies of MAO coating on the TNZS substrate. The interface between MAO coating and TNZS substrate displayed good bonding that no cracks and visible demarcation line presented. The thickness of MAO coating on TNZS plates was about 4–6 μm.

Fig. 2 XRD patterns of different samples (A) and the representative cross-sectional SEM images of TNZS-MAO (B). The yellow dashed line indicate the interface between TNZS and MAO coating.

The surface elastic modulus and surface hardness of different Ti samples was further investigated as shown in Table 2. As expected, TNZS showed lower surface elastic modulus and surface hardness than pure Ti. After MAO treating, the surface elastic modulus and surface hardness of Ti-MAO and TNZS-MAO samples were dramatic decreased. Interestingly, the TNZS-MAO remained lower elastic modulus and hardness than Ti-MAO, which was mainly due to the distribution of amorphous phase.

Table 2 The surface elastic modulus and hardness of different Ti samples

Samples	Elastic modulus (GPa)	Hardness (GPa)
Ti	140 ± 15	4.6 ± 0.4
Ti-MAO	56 ± 5	2.2 ± 0.2
TNZS	80 ± 12	4.0 ± 0.4
TNZS-MAO	40 ± 10	2.0 ± 0.2

---

The surface roughness of different Ti samples was measured qualitatively by white-light interference analyses. As shown in Table 3, the micro-roughness of the Ti-MAO and TNZS-MAO samples were significantly higher than that of the control polished Ti and TNZS samples. Furthermore, the surface roughness of Ti-MAO was higher than that of TNZS-MAO.

Table 3 Comparison among Ti, TNZS, Ti-MAO and TNZS-MAO in terms of surface roughness, water contact angles and protein adsorption

Samples	Ra (nm)	Contact angle (deg.)	Protein adsorption (μg mL^−1)
Ti	332 ± 5.65	63.8 ± 2.52	20.8 ± 0.99
Ti-MAO	963 ± 2.83	56.5 ± 3.48	24.1 ± 0.89
TNZS	265 ± 28.28	65.5 ± 3.48	20.6 ± 1.74
TNZA-MAO	850 ± 51.61	42.5 ± 5.28	25.6 ± 0.37

---

Surface wettability of different samples were characterized by water contact angles. As shown in Table 3, polished Ti and TNZS samples displayed similar contact angles of 63.8 ± 2.52° and 65.5 ± 3.48° respectively. Nevertheless, both Ti-MAO and TNZS-MAO displayed significant increases in hydrophilicity with contact angle of 56.5 ± 3.48° and 42.5 ± 5.28°, respectively. The results indicated that MAO treatment effectively increased the surface wettability of Ti and TNZS samples, rather than elastic modulus.

The protein adsorption capacity of different samples was investigated. As shown in Table 3, higher protein adsorption content were present on Ti-MAO and TNZS-MAO, as compared to the untreated Ti samples, which indicated that porous structure could promote protein adsorption. Although there was no significant difference, higher protein adsorption content could be observed on TNZS-MAO compared to Ti-MAO.

Evaluation of BMSCs responses in vitro

The initial cell adhesion results are displayed in Fig. 3C. More adherent cells were observed on Ti-MAO and TNZS-MAO compared to polished Ti and TNZS. Moreover, TNZS-MAO induced a larger adherent cell number after 1.0 h of culture compared with Ti-MAO. In addition, after incubation for 2.0 h, the other three different specimens still exhibited more cell number than the Ti samples. These results are mostly consistent with the protein adsorption analysis data.

Fig. 3 The response of rBMSCs to various Ti samples in vitro. (A) Focal adhesion formation detected by vinculin staining (green) triple-labeled with actin (red) and nuclei (blue) for rBMSCs cultured on the samples for 4 hours: (a) Ti, (b) Ti-MAO, (c) TNZS, and (d) TNZS-MAO. (B) Cell morphology on the specimens after 1 day of culture imaged by SEM: (a and e) Ti, (b and f) Ti-MAO, (c and g) TNZS, and (d and h) TNZS-MAO. The blue arrow indicate cell filopodia. (C) The initial adherent numbers of rBMSCs counted by DAPI staining. The quantitative analysis of vinculin (D) and cell spreading area (E) on different samples. (F) Cell proliferation on different samples evaluated by CCK8 assay after 1, 3, 5, 7 days of culture. *p < 0.05 compared to the Ti samples, #p < 0.05 compared to the Ti-MAO samples, % p < 0.05 compared to the TNZS samples.

Meanwhile, cytoskeletal and focal adhesion (FA) formation were investigated by immunofluorescence after 4.0 h of culture. As shown in Fig. 3A, the cell on the Ti samples appeared minor cytoskeletal reorganization while those on MAOs mostly displayed abundant reassembly of the cytoskeleton and more extended appearance. Moreover, from the quantitative analysis, the cell spreading area on MAOs were significantly greater than those of the untreated samples, and TNZS-MAO with low elastic modulus displayed the greatest cell spreading area (Fig. 3D). Consistent with the trend of cell spreading area, it was obvious that the presence of FA on MAOs were greater than that of Ti (Fig. 3A). With decreasing elastic modulus, more FA were present on TNZS-MAO, as compared to Ti-MAO (Fig. 3A and E).

The cell morphology was further investigated by SEM. As seen in Fig. 3B, notable differences in cell shape were also observed on different specimens after culture for 1 day. The cells on Ti and TNZS displayed an elongated shape that possessed a small amount of slender filopodia, whereas those on Ti-MAO and TNZS-MAO mostly exhibited a polygonal osteoblastic-like shape with multiple filopodia.²⁷ Additionally, cells on TNZS-MAO displayed more extended appearance, as compared to Ti-MAO.

As shown in Fig. 3F, cell proliferation was assessed by CCK-8 assay after 1, 3, 5 and 7 days of culture. It could be observed that the cell number increased steadily as time progressed. On day 1, the cell proliferation on specimens exhibited similar trend as the cell adhesion data. However by day 3, cell proliferation on Ti-MAO, TNZS and TNZS-MAO was distinctly lower than that on Ti. After 5 days of culture, it could be observed that cell proliferation was reduced on Ti-MAO. Additionally, TNZS-MAO exhibited lower cell proliferation compared to TNZS. At day 7, cell viability on Ti-MAO and TNZS-MAO remained lower trend than untreated Ti and TNZS. Additionally, significantly higher cell viabilities were observed on TNZS-MAO compared to Ti-MAO on day 5 and day 7, thus indicating that TNZS-MAO offered a more favourable environment for cell growth.

As ALP and collagen are important markers of early osteogenic differentiation,^28,29 the early osteogenic differentiation of rBMSCs on specimens were evaluated by quantifying the secretion of ALP and collagen (Fig. 4). As early as 3 days of culture, ALP secretion can be detected but there were no obvious differences on all the substrata. After culture for 7 days, large ALP positive areas were present on MAOs surface, whereas only a small amount of ALP nodules were scattered on untreated samples. With lower elastic modulus, TNZS-MAO induced more ALP staining than Ti-MAO. Meanwhile, the collagen secretion on substrates exhibited similar trend as the ALP activity. The results of quantitative analysis further confirmed that TNZS-MAO with low elastic modulus displayed the highest ALP activity and collagen secretion.

Fig. 4 ALP activity and collagen secretion of rBMSCs on various Ti samples. (A) ALP and sirius red staining images: (a, e and i) Ti, (b, f and j) Ti-MAO, (c, g and k) TNZS, and (d, h and k) TNZS-MAO. (B) Quantitative analysis of ALP production (C) quantitative analysis of collagen secretion. *p < 0.05 compared to the Ti samples, #p < 0.05 compared to the Ti-MAO samples, % p < 0.05 compared to the TNZS samples.

To further investigate the synergetic effects of elastic modulus and MAO porous structure on osteogenic activity, the differentiation of rBMSCs on specimens were evaluated by analyzing osteogenic gene expression including ALP, OPN, BSP and BMP-2 (Fig. 5C). As expected, TNZS-MAO induced the highest gene expression levels after 14 days of culture, and displayed the most osteogenic activity. Immunocytochemistry was performed to qualitatively assess the expression of BSP and BMP-2 as shown in Fig. 5A and B. The quantitative results demonstrated that MAOs induced significantly higher expression levels of BSP and BMP-2. Moreover, TNZS-MAO with low elastic modulus further enhanced the expression of BSP and BMP-2.

Fig. 5 In vitro osteogenic differentiation of rBMSCs on various Ti samples. (A) Immunocytochemistry for detection of BSP (red) and BMP-2 (red) expression by rBMSCs cultured for 14 days: (a and e) Ti, (b and f) Ti-MAO, (c and g) TNZS, and (d and h) TNZS-MAO. (B) Quantitative analysis of the expression levels of BSP and BMP-2 by different samples. (C) Gene expression levels of ALP, BSP, OPN and BMP-2 in rBMSCs cultured for 7 and 14 days. *p < 0.05 compared to the Ti samples, #p < 0.05 compared to the Ti-MAO samples, % p < 0.05 compared to the TNZS samples.

Next, ECM mineralization was evaluated by alizarin red staining. As shown in Fig. 6A, only small mineralization dots were observed to be scattered on untreated samples. However, abundant mineralization nodules were induced on Ti-MAO, TNZS, TNZS-MAO, especially TNZS-MAO. The quantitative results demonstrated that more ECM mineralization were induced on MAO-treated samples than on untreated samples (Fig. 6B). Moreover, TNZS-MAO with low elastic modulus displayed the most mineralized nodules, thus implying that low elastic modulus could further enhance osteogenic differentiation.

Fig. 6 In vitro mineralization of rBMSCs on various Ti samples. (A) Alizarin red staining of rBMSCs cultured for 21 days: (a) Ti, (b) Ti-MAO, (c) TNZS, and (d) TNZS-MAO. (B) The quantitative analysis of ECM mineralization on different samples. *p < 0.05 compared to the Ti samples, #p < 0.05 compared to the Ti-MAO samples, % p < 0.05 compared to the TNZS samples.

Evaluation of osseointegration in vivo

To further investigate the impact of elastic modulus and hierarchical structure on osseointegration of Ti implants in vivo, a rat femur model was utilized and osseointegration was characterized by μ-CT and histological morphometry. Fig. 7 shows the three-dimensional reconstructions of the femur containing the implants at 4 weeks after surgery. It could be observed that more new bone was formed around MAO treated implants, as compared to untreated Ti implants, with TNZS-MAO implants displaying the most new bone formation.

Fig. 7 Representative three-dimensional μ-CT scan images after 4 weeks of implantation: (a and e) Ti, (b and f) Ti-MAO, (c and g) TNZS, and (d and h) TNZS-MAO.

Additionally, histological morphometry was performed and it was observed that new bone was being formed around all implant surfaces at 4 weeks post-surgery (Fig. 8A and B). Only a small amount of newly loosened bone was scattered at the interfaces of Ti implants and obvious gaps were observed between the newly formed bone and implants. At the interfaces of Ti-MAO, the newly formed bone was relatively dense and only a few gaps were found. Meanwhile, in the TNZS-MAO group, excellent integration was displayed with the newly formed bone being continuous and integrated tightly with the implants. Furthermore, as seen from Fig. 8C, the bone–implant contact value of TNZS-MAO was higher than other groups. Taking into account the μ-CT results, it can be concluded that low elastic modulus and surface hierarchical structure of implants had synergistic effects on early osseointegration of implants in vivo.

Fig. 8 Histological analysis of osseointegration at 4 weeks after implantation. (A) Histological imaging of new bone (NB) formation around the implants. (a) Ti, (b) Ti-MAO, (c) TNZS, and (d) TNZS-MAO. MC: marrow cavity. NB: new bone. (B) Histomorphometry analysis of bone volume/total volume (BV/TV). (C) Quantitative analysis of bone–implant contact values (BIC). *p < 0.05 compared to the Ti implants; #p < 0.05 compared to the Ti-MAO implants.

Discussion

The elastic modulus and surface topology of biomaterials have been documented to exert a profound influence on cell and tissue function. Hence, the modification of these parameters could be a promising strategy to achieve better osseointegration. However, the synergistic effects of elastic modulus and surface topology of the implants on osseointegration are currently not well-understood. In this study, we made a comparative evaluation of various Ti-based implants with different elastic modulus and surface morphology through assessment of in vitro osteogenic activity and in vivo osseointegration within bone defects. Our results thus suggest that micro-scale porous hierarchical structure coupled with the biomimetic low elastic modulus of Ti-based implants could synergistically enhance osteogenic differentiation of BMSCs in vitro and improve osseointegration in vivo.

Surface topological structure is well-known to exert a profound effect on the physicochemical properties and biological activity of implants.^30,31 Previous investigations have demonstrated that micro/nanostructures can control stem cell adhesion, proliferation and differentiation.^23,32 In our study, a micro-scale porous hierarchical structural layer was coated on the Ti and TNZS implant surfaces by the MAO technique, which was considered a routine protocol for implant surface modification. The formation of the unique porous hierarchical structure could be explained as arising from the incorporation of gas bubbles and the breakdown of the dielectric barrier layer under a strong electric field.³³ The physicochemical properties of Ti and TNZS samples were obviously modified by MAO treatment. As expected, enhanced surface roughness, decreased water contact angle and improved protein adsorption levels were displayed by Ti-MAO and TNZS-MAO samples. Our results are consistent with previous studies that demonstrated the enhancing effects of MAO treatment on surface roughness, hydrophobicity and protein adsorption.^34,35 These results thus imply that MAO treatment could be an effective approach for further improving the excellent physicochemical properties and biocompatibility of TNZS implantation materials.

It has been shown that the porous hierarchical structure generated by MAO could enhance stem cell adhesion and promote osteogenic differentiation.^23,34 As expected, in this study, both Ti-MAO and TNZS-MAO samples promoted cell attachment, cell spreading and significantly improved osteogenic differentiation, as demonstrated by the higher ALP activity as well as more collagen secretion and ECM nodule formation versus the control polished samples. The increased biological activity and osteogenic inductivity of MAO-treated Ti and TNZS samples can be mostly attributed to the enhanced surface wettability and protein adsorption capacity arising from the modified topographical properties (increased surface roughness and/or porous structure). It is recognized that the Ca/P ratio is an important factor in evaluating the apatite-inducing ability after simulated body fluid immersion. Based on previous research, the apatite-inducing ability of incorporated Ca and P elements in MAO coating was poor, and only a thin film with relatively low Ca (8.9 at%) and P (10.2 at%) content were formed on the surface.³⁶ In this work, there were no obvious differences in the Ca and P element content between the Ti-MAO and TNZS-MAO. This observation is consistent with previous studies that the hydrophilicity plays a vital role in protein adsorption, which could promote cell adhesion at an early stage.^37,38

The mechanical properties of implanted materials, particularly elastic modulus, have been found to play vital roles in influencing the differentiation pathway and lineage fate of stem cells in tissue regeneration.^39–41 In our study, the introduce of MAO coating dramatically reduced the mechanical properties of both Ti and TNZS including elastic modulus and hardness. This may be ascribed to the porous structure of MAO coating and abundant of amorphous phase resulted from Ca and P elements in MAO coating. Interestingly, the TNZS-MAO remained lower elastic modulus and hardness compared with Ti-MAO. This phenomenon could be elucidated that the MAO coating of TNZS-MAO were mainly consisted of amorphous phase while a small amount of anatase existed in Ti-MAO as confirmed by XRD results. Because of the fundamental difference, more protein adsorption and higher initial cell adhesion numbers were displayed by TNZS-MAO, as compared to the Ti-MAO samples. This inherent elastic modulus of Ti based implants may play a dominant role in determining cell behaviour compared to other physicochemical property such as surface roughness, as TNZS-MAO had a lower surface roughness than Ti-MAO in present work. It has been reported that matrix stiffness could regulate stem cell morphology, cytoskeleton and focal adhesion characteristics.^42–44 In this study, well-organized cytoskeleton and larger cell spreading area could be observed on the TNZS-MAO samples, as compared to the Ti-MAO samples. Moreover, it was observed that higher FA numbers were displayed on TNZS-MAO than Ti-MAO samples, which may be attributed to mechanotransduction as reported in previous studies.^45–47 Furthermore, higher ALP activity, more collagen secretion and ECM nodule formation were observed on TNZS-MAO, as compared to Ti-MAO. Hence, it is highly possible that the lower matrix elastic modulus had a synergistic effect on the enhancing effects of MAO treated Ti based materials on the cytoskeletal organization, focal adhesion formation and osteogenic differentiation of MSCs in vitro.

It is believed that primary bone formation around implants is a key factor for effective bone–implant contact and implant stability, which determined the therapeutic efficacy of implants in vivo.^48–50 Additionally, the elastic modulus exert a profound influence on early implant stability and it has also been recognized that large mismatch of modulus might lead to implant loosening.^20,51 As expected, our study showed that the porous hierarchical structure efficiently enhanced new bone formation. TNZS-MAO with low elastic modulus further accelerated osseointegration. Interestingly, the osteogenic capacity of Ti-MAO is superior to TNZS in vitro, while new bone formation around TNZS implants was similar to Ti-MAO in vivo. This may be ascribed to the prominent effect of inherent elastic matching of implants on osseointegration in vivo. It is well known that mechanical matching plays a crucial role in mitigating the effects of stress and strain at the bone–implant interface.⁵² During the process of implantation, there may be interaction between the implant and host bone, which could make the implant generate micro-strain spontaneously to match or adapt to mechanical stimulation arising from the host tissue. Hence, it can be recognized that the mechanical properties of the implants, particularly elastic modulus play a key role in the process of implantation in vivo. This has also been confirmed by a previous study which demonstrated that the amount of new bond surrounding the lower elastic Ti-7.5Mo implant was two-folds of that surrounding the higher elastic Ti–6Al–4V implant in rabbit femur after 26 weeks post-surgery.⁵³ These results thus suggest that lowering the elastic modulus should be seriously considered as a key factor in designing suitable implants for orthopedic applications. The results of this study thus demonstrate that low elastic modulus coupled with porous hierarchical surface structure of implants could achieve the desired early osseointegration.

Conclusions

In this study, Ti-MAO and TNZS-MAO implant samples were successfully fabricated to comparatively evaluate the synergistic effects of elastic modulus and surface topological structure of implants on the osteogenic differentiation of stem cells and early osseointegration. MAO treatment resulted in increased surface roughness, improved surface wettability and favourable mechanical properties. In vitro studies including the immunofluorescence staining and RT-PCR analyses indicated that the TNZS-MAO samples with low elastic modulus displayed the most cell attachment, alkaline phosphatase activity, collagen secretion, osteogenic marker expression, and mineralization of rBMSCs, as compared with the other samples. Moreover, μ-CT and histological analyses of the implants revealed that inherently low elastic modulus could cooperatively accelerate early osseointegration and the maturation of trabecular bone upon implantation with MAO modified Ti implants at 4 weeks post-surgery. These results thus demonstrated that elastic modulus and surface micro-scale structure had a synergistic effect on the osseointegration of Ti implants for orthopedic applications.

Acknowledgements

This work was financially supported by the National Basic Science Research Program (2012CB33900), the National Natural Science Foundation of China (81425007, 51502006), the National Scientific and Technological Support Program of China (2013BAD16809) and the National High Technology Research and Development Program of China (2015AA033601).

Notes and references

1. L. Le Guehennec, A. Soueidan, P. Layrolle and Y. Amouriq, Dent. Mater., 2007, 23, 844–854.
2. R. Agarwal and A. J. Garcia, Adv. Drug Delivery Rev., 2015, 94, 53–62.
3. S. P. Pilipchuk, A. B. Plonka, A. Monje, A. D. Taut, A. Lanis, B. Kang and W. V. Giannobile, Dent. Mater., 2015, 31, 317–338.
4. S. Kotsovilis, I. K. Karoussis and I. Fourmousis, Clin. Oral Implants Res., 2006, 17, 587–599.
5. M. Retzepi and N. Donos, Clin. Oral Implants Res., 2010, 21, 673–681.
6. U. Gross and V. Strunz, J. Biomed. Mater. Res., 1985, 19, 251–271.
7. H. Schliephake, A. Aref, D. Scharnweber, S. Rosler and A. Sewing, Clin. Oral. Implan. Res., 2009, 20, 38–44.
8. L. Salou, A. Hoornaert, G. Louarn and P. Layrolle, Acta Biomater., 2015, 11, 494–502.
9. D. Buser, N. Broggini, M. Wieland, R. K. Schenk, A. J. Denzer, D. L. Cochran, B. Hoffmann, A. Lussi and S. G. Steinemann, J. Dent. Res., 2004, 83, 529–533.
10. I. Abrahamsson, T. Berglundh, E. Linder, N. P. Lang and J. Lindhe, Clin. Oral Implants Res., 2004, 15, 381–392.
11. J. L. Ong, D. L. Carnes and K. Bessho, Biomaterials, 2004, 25, 4601–4606.
12. A. Kamali, A. Hussain, C. Li, J. Pamu, J. Daniel, H. Ziaee, J. Daniel and D. J. McMinn, J. Bone Jt. Surg., Br. Vol., 2010, 92, 717–725.
13. W. M. Weng, W. Q. Yu, W. L. Shan and F. Q. Zhang, Shanghai J. Stomatol., 2010, 19, 621–625.
14. X. N. Gu, N. Li, W. R. Zhou, Y. F. Zheng, X. Zhao, Q. Z. Cai and L. Ruan, Acta Biomater., 2011, 7, 1880–1889.
15. R. Zhou, D. Wei, J. Cao, W. Feng, S. Cheng, Q. Du, B. Li, Y. Wang, D. Jia and Y. Zhou, Mater. Sci. Eng., C, 2015, 49, 669–680.
16. R. Zhou, D. Wei, J. Cao, W. Feng, S. Cheng, Q. Du, B. Li, Y. Wang, D. Jia and Y. Zhou, ACS Appl. Mater. Interfaces, 2015, 7, 8932–8941.
17. Y. T. Sul, C. Johansson, E. Byon and T. Albrektsson, Biomaterials, 2005, 26, 6720–6730.
18. H. Kröger, P. Venesmaa, J. Jurvelin, H. Miettinen, O. Suomalainen and E. Alhava, Clin. Orthop. Relat. Res., 1998, 66–74.
19. T. Albrektsson and C. Johansson, Eur. Spine J., 2001, 10(2), S96–S101.
20. R. Huiskes, H. Weinans and B. van Rietbergen, Clin. Orthop. Relat. Res., 1992, 124–134.
21. Y. L. Hao, S. J. Li, S. Y. Sun, C. Y. Zheng and R. Yang, Acta Biomater., 2007, 3, 277–286.
22. A. Oyane, H. M. Kim, T. Furuya, T. Kokubo, T. Miyazaki and T. Nakamura, J. Biomed. Mater. Res., Part A, 2003, 65, 188–195.
23. L. Zhao, L. Liu, Z. Wu, Y. Zhang and P. K. Chu, Biomaterials, 2012, 33, 2629–2641.
24. B. Peter, O. Gauthier, S. Laib, B. Bujoli, J. Guicheux, P. Janvier, G. H. van Lenthe, R. Muller, P. Y. Zambelli, J. M. Bouler and D. P. Pioletti, J. Biomed. Mater. Res., Part A, 2006, 76, 133–143.
25. X. Zhang, M. Xu, L. Song, Y. Wei, Y. Lin, W. Liu, B. C. Heng, H. Peng, Y. Wang and X. Deng, Biomaterials, 2013, 34, 9103–9114.
26. M. Cheng, Y. Qiao, Q. Wang, G. Jin, H. Qin, Y. Zhao, X. Peng, X. Zhang and X. Liu, ACS Appl. Mater. Interfaces, 2015, 7, 13053–13061.
27. R. McBeath, D. M. Pirone, C. M. Nelson, K. Bhadriraju and C. S. Chen, Dev. Cell, 2004, 6, 483–495.
28. S. H. Mirmalek-Sani, R. S. Tare, S. M. Morgan, H. I. Roach, D. I. Wilson, N. A. Hanley and R. O. Oreffo, Stem Cells, 2006, 24, 1042–1053.
29. Y. Zhao, H. M. Wong, W. Wang, P. Li, Z. Xu, E. Y. Chong, C. H. Yan, K. W. Yeung and P. K. Chu, Biomaterials, 2013, 34, 9264–9277.
30. Y. Li, W. Yang, X. Li, X. Zhang, C. Wang, X. Meng, Y. Pei, X. Fan, P. Lan, C. Wang, X. Li and Z. Guo, ACS Appl. Mater. Interfaces, 2015, 7, 5715–5724.
31. S. Facca, D. Lahiri, F. Fioretti, N. Messadeq, D. Mainard, N. Benkirane-Jessel and A. Agarwal, ACS Nano, 2011, 5, 4790–4799.
32. D. D. Deligianni, N. D. Katsala, P. G. Koutsoukos and Y. F. Missirlis, Biomaterials, 2001, 22, 87–96.
33. W. Li, W. Li, Z. Yan and W. Fu, Surf. Coat. Technol., 2010, 205, 1651–1658.
34. Y. Tian, S. Ding, H. Peng, S. Lu, G. Wang, L. Xia and P. Wang, Appl. Surf. Sci., 2012, 261, 25–30.
35. R. A. Gittens, T. McLachlan, R. Olivares-Navarrete, Y. Cai, S. Berner, R. Tannenbaum, Z. Schwartz, K. H. Sandhage and B. D. Boyan, Biomaterials, 2011, 32, 3395–3403.
36. R. Zhou, D. Wei, W. Feng, S. Cheng, H. Yang, B. Li, Y. Wang, D. Jia and Y. Zhou, Surf. Coat. Technol., 2014, 252, 148–156.
37. J. Takebe, S. Itoh, J. Okada and K. Ishibashi, J. Biomed. Mater. Res., 2000, 51, 398–407.
38. J. H. Lee and H. B. Lee, J. Biomater. Sci., Polym. Ed., 1993, 4, 467–481.
39. M. B. Keogh, F. J. O'Brien and J. S. Daly, Acta Biomater., 2010, 6, 4305–4313.
40. A. J. Engler, S. Sen, H. L. Sweeney and D. E. Discher, Cell, 2006, 126, 677–689.
41. F. Guilak, D. M. Cohen, B. T. Estes, J. M. Gimble, W. Liedtke and C. S. Chen, Cell Stem Cell, 2009, 5, 17–26.
42. S. Fusco, V. Panzetta, V. Embrione and P. A. Netti, Acta Biomater., 2015, 23, 63–71.
43. Y. Navaro, N. Bleich-Kimelman, L. Hazanov, I. Mironi-Harpaz, Y. Shachaf, S. Garty, Y. Smith, G. Pelled, D. Gazit, D. Seliktar and Z. Gazit, Biomaterials, 2015, 49, 68–76.
44. B. Trappmann, J. E. Gautrot, J. T. Connelly, D. G. Strange, Y. Li, M. L. Oyen, M. A. Cohen Stuart, H. Boehm, B. Li, V. Vogel, J. P. Spatz, F. M. Watt and W. T. Huck, Nat. Mater., 2012, 11, 642–649.
45. J. C. Kuo, J. Cell. Mol. Med., 2013, 17, 704–712.
46. L. B. Case, M. A. Baird, G. Shtengel, S. L. Campbell, H. F. Hess, M. W. Davidson and C. M. Waterman, Nat. Cell Biol., 2015, 17, 880–892.
47. N. Huebsch, E. Lippens, K. Lee, M. Mehta, S. T. Koshy, M. C. Darnell, R. M. Desai, C. M. Madl, M. Xu, X. Zhao, O. Chaudhuri, C. Verbeke, W. S. Kim, K. Alim, A. Mammoto, D. E. Ingber, G. N. Duda and D. J. Mooney, Nat. Mater., 2015, 14, 1269–1277.
48. V. Goriainov, R. Cook, J. M. Latham, D. G. Dunlop and R. O. Oreffo, Acta Biomater., 2014, 10, 4043–4057.
49. R. M. Wazen, J. A. Currey, H. Guo, J. B. Brunski, J. A. Helms and A. Nanci, Acta Biomater., 2013, 9, 6663–6674.
50. J. Karrholm, P. Herberts, P. Hultmark, H. Malchau, B. Nivbrant and J. Thanner, Clin. Orthop. Relat. Res., 1997, 94–110.
51. M. Geetha, A. K. Singh, R. Asokamani and A. K. Gogia, Prog. Mater. Sci., 2009, 54, 397–425.
52. H. W. Wiskott and U. C. Belser, Clin. Oral Implants Res., 1999, 10, 429–444.
53. D. J. Lin, C. C. Chuang, J. H. Chern Lin, J. W. Lee, C. P. Ju and H. S. Yin, Biomaterials, 2007, 28, 2582–2589.

---

Footnote

† These authors contributed equally to this work.

---