- Email: [email protected]
Preparation and in vitro evaluation of mesoporous hydroxyapatite coated β-TCP porous scaffolds
Materials Science and Engineering C 33 (2013) 5001–5007
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Preparation and in vitro evaluation of mesoporous hydroxyapatite coated β-TCP porous scaffolds Xinyu Ye a, Shu Cai a,⁎, Guohua Xu b,⁎⁎, Ying Dou a, Hongtao Hu b, Xiaojian Ye b a b
Key Laboratory for Advanced Ceramics and Machining Technology of Ministry of Education, School of Materials Science and Engineering, Tianjin University, Tianjin 300072, People's Republic of China Shanghai Changzheng Hospital, Shanghai 200003, People's Republic of China
a r t i c l e
i n f o
Article history: Received 20 May 2013 Received in revised form 22 July 2013 Accepted 23 August 2013 Available online 31 August 2013 Keywords: Mesoporous hydroxyapatite Coatings β-tricalcium phosphate (β-TCP) scaffold Cell response Protein expression
a b s t r a c t A mesoporous hydroxyapatite (HA) coating was prepared on a β-tricalcium phosphate (β-TCP) porous scaffold by a sol-gel dip-coating method using the block copolymer Pluronic F127 (EO106PO70EO106) as the template. For application as a bone graft, in vitro cell response and bone-related protein expression of mesoporous HA coated β-TCP scaffold were investigated, using the non-mesoporous HA coated scaffold as the control group, to evaluate the influence of the mesoporous structure on the biological properties of HA coating. It was found that the increased surface area of the mesoporous HA coating greatly affected the response of MC3T3-E1 osteoblasts and the expression of proteins. An enzyme-linked immunosorbent assay recorded a significantly higher expression of alkaline phosphatase (ALP) and bone sialoprotein (BSP) in the mesoporous group than those in the control group (*p b 0.05) after different incubation periods. The introduction of mesopores enhanced the expression of ALP and BSP in the cells grown on the mesoporous HA coatings, on the premise of maintaining the protein expression in a sequence to ensure the correct temporo-spatial expression in osteogenesis. These results indicated that the mesoporous HA coating would provide a good environment for cell growth, suggesting that it could be used as the coating material for the surface modification of the tissue engineering scaffolds. © 2013 Elsevier B.V. All rights reserved.
1. Introduction The bone-forming ability of biomaterials is related to chemical composition as well as textural properties, such as pore size and pore volume [1]. Increasing the specific surface area and pore volume of biomaterials would accelerate the kinetic deposition process of hydroxycarbonate apatite (HCA) and consequently enhance the boneforming bioactivity [2]. Owing to the large surface area and pore volume, mesoporous materials have been proposed for use in biological applications such as drug release, bioadsorption and biosensing [3]. Recently the research related to mesoporous materials for bone tissue regeneration has been performed by several research groups [4–7]. For example, mesoporous SiO2–CaO–P2O5 scaffolds or powders have been prepared and demonstrated the superior bone-forming bioactivities in vitro compared to normal bioglass [5]. Mesoporous titania layers could lead to a significant enhancement of apatite formation compared with non-coated or even nano-tubular Ti surface coatings [6]. As compared to other bone repairing materials, hydroxyapatite (HA) possesses the favorable biocompatibility and osteoconductivity, and HA coating has been considered to be the optimized choice for surface modification
⁎ Corresponding author. Tel.: +86 22 27425069. ⁎⁎ Corresponding author. Tel.: +86 21 81886999. E-mail addresses: [email protected] (S. Cai), [email protected] (G. Xu). 0928-4931/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2013.08.027
by combining with other materials [8]. Therefore, HA coating with mesostructure would display the suitable properties for use as the biomaterials for bone repair. In our preliminary work, the uniform and crack-free mesoporous HA coating was synthesized on the glass substrates by a sol-gel method, using F127 (EO106PO70EO106, EO is ethylene oxide, PO is propylene oxide) as the template [9]. In vitro tests revealed that MC3T3-E1 cells cultured on the mesoporous HA coating showed better surface wettability and cell spreading than on the non-mesoporous HA coating. The presence of mesopores was preliminarily confirmed to be a key parameter influencing the cell response to HA coating. Besides that, the differences in cell behavior may affect the expression of bone-related proteins, which plays an important role in cell differentiation and de novo mineralization in bone. Therefore, it is necessary to evaluate the expression of bone-related proteins in the cells grown on the mesoporous HA coating to further clarify the effect of mesopores on the cell response and bone-forming ability. In this study, we focused on the cell response and bone-related protein expression in MC3T3-E1 cells of the mesoporous HA coating (denoted as M-HA), using the non-mesoporous HA coating (denoted as N-HA) with a similar surface morphology and phase composition to those of the control group. Considering the potential use for the surface modification of implants, the porous β-TCP scaffolds were selected as the substrates to investigate the effect of mesoporous HA coating on their biological properties.
5002
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
2. Materials and methods
2.3. In vitro evaluation
2.1. Sample preparation
Osteoblast-like cells MC3T3-E1 [10] were cultured in Dulbecco's Modified Eagle's Medium (DMEM) containing 10% fetal bovine serum (FBS), and maintained at 37 °C in a saturated humidified atmosphere of 5% CO2. The cells were detached with a 0.1% trypsin-EDTA solution in phosphate-buffered saline (PBS, pH 7.4), centrifuged at 1000 rpm for 5 min, washed and resuspended. Samples N-HA and M-HA were cleaned by physiological saline, continually rinsed with distilled water, and sterilized using Cobalt-60 radiation. The cells were seeded on the samples in a 24-well plate at a concentration of 1.0 × 105 cells/mL in 1.5 mL, and incubated with 5% CO2 at 37 °C for different culture periods. The culture medium was renewed twice in a week. After being cultured for 3 and 7 days, the samples were washed with phosphate buffered saline (PBS), fixed using 2% glutaraldehyde for 2 h, washed by PBS for 20 min after drawing off the fixed agent, dehydrated through a graded alcohol solution, and then critical point drying was performed with an apparatus (HCP-2, Hitachi Koki Ltd., Japan). After being coated with gold, the attachment of MC3T3-E1 cells on the coatings was observed by SEM (XL-30, Philips). For a protein production assay, the cells were harvested at 3, 7 and 14 days, and transferred into a 96-well plate. The plates were centrifuged at 3000 rpm for 10 min and the medium was aspirated from the wells. After being placed in an incubator at 50 °C for 30 min, the dried plates were wrapped with Clingwrap and stored at room temperature for the further tests. The production of alkaline phosphatase (ALP) and bone sialoprotein (BSP) in MC3T3-E1 cells were detected by using an enzyme-linked immunosorbent assay (ELISA) kit (EYSIN Co., Shanghai, China). Briefly, standards or analytes were added into the wells, and biotin-antibodies were then added and incubated. Subsequently, horseradish peroxidase avidin (HRP-avidin) was added to bind the detection antibodies, and a substrate solution was used to develop the color. The reaction was terminated by the addition of an acidic stop solution, and the optical density (OD) of the yellow color was immediately determined at 450 nm. Because the absorbance was proportional to the amount of analyte, the concentration of ALP and BSP was measured in reference to the standard curve.
M-HA sol was prepared by using Ca(NO3)2·4H2O and P2O5 as raw materials, and Pluronic F127 (Sigma, USA) as the template for introducing the mesostructure. As described in our previous study [9], 0.033 g/mL of a P2O5 (Kermel, Tianjin) ethanol solution and 0.136 g/mL of a Ca(NO3)2 · 4H2O (Damao, Tianjin) aqueous solution were successively added into 0.192 g/mL of an F127 ethanol solution at 40 °C under vigorous stirring for producing the sol. For comparison, N-HA sol was prepared at the same conditions but without F127. β-TCP porous scaffolds made in our lab (Ф = 12.0 mm × 6.0 mm, porosity: 70.0%–75.0%) were used as the substrates, cleaned in acetone, alcohol and distilled water for 10 min using an ultrasonic cleaner, and then dried in air. The coatings were synthesized by dip-coating on the substrates at a constant rate of 1.0 mm/s, and this process was repeated for five times. After aging at room temperature for 48 h and drying at 60 °C for 12 h, the as-synthesized coatings were calcined at 500 °C for 3 h to remove the template. In addition, N-HA and M-HA powders were also synthesized as the above process for the phase composition and porous structure analyses. 2.2. Characterization Morphological and microstructural observations of specimens were carried out by scanning electron microscopy (SEM, XL-30, Philips) and transmission electron microscopy (TEM, JEM-2010HR, JEOL). The phase composition of samples N-HA and M-HA was analyzed by using X-ray diffraction (XRD, D/Max-2500, Philips) using Cu Kα radiation over the range from 10° to 60°. N2 adsorption–desorption isotherms were measured with a Quantachrome NOVA analyzer. The pore size distribution of the non-mesoporous and mesoporous samples was calculated from the adsorption–desorption data using the density functional theory (DFT) model. The specific surface area of HA was calculated according to the Brunauer–Emmett–Teller (BET) equation. For each group, five tests were performed and the results were averaged.
Fig. 1. N2 adsorption–desorption isotherm and pore size distribution curve of samples N-HA and M-HA.
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
5003
Statistical analysis was performed using SPSS 11.0 to evaluate the differences between the M-HA and N-HA groups. Each data point was presented as means ± deviation (X SD) from five independent experiments and a p value of b0.05 represented a significant difference. 3. Results 3.1. Characterization of mesoporous structure The characterization of the mesoporous structure in sample M-HA was carried out by N2 adsorption–desorption measurements. Fig. 1 shows the adsorption–desorption isotherms and the pore size distribution curves based on the DFT model for samples N-HA and M-HA. For sample N-HA, no hysteresis loops were exhibited in N2 isotherms, revealing the non-mesoporous structure in the coating. In the case of sample M-HA, N2 isotherms presented a type IIb curve with an H3 hysteresis loop for the typically non-rigid and disordered mesoporous structure. Moreover, the pore size distribution curves present the introduction of the mesopores of ~3.17 nm in the M-HA group, as compared to the micropores of ~0.42 nm in the N-HA group. Meanwhile, as shown in Fig. 2, the specific surface area of sample M-HA increased from 9.17 m2/g to 211.7 m2/g due to the presence of mesopores. 3.2. Phase composition and surface morphology XRD patterns shown in Fig. 3 indicate that samples N-HA and M-HA have the same phase composition, including the major phase hydroxyapatite (HA, PDF#09-0432) and the minor phases tricalcium phosphate (β-TCP, PDF#09-0169) and calcium pyrophosphate (CPP, PDF#090346). The crystallization degree of samples N-HA and M-HA were calculated to be 82.45% and 80.81% respectively using MDI JADE 5.0. There is no significant difference in the crystallization degree between the two groups. Meanwhile, as seen from the XRD patterns, the intensities of some diffraction peaks are higher and others are lower for sample M-HA, as compared to sample N-HA. It should be attributed to the crystal growth along different planes due to the presence of the template and mesoporous structure [11].
Fig. 3. XRD patterns of samples N-HA (a) and M-HA (b).
Fig. 4 presents the surface morphology of β-TCP porous scaffolds before and after coating with the mesoporous HA. As shown in Fig. 4a, the surface micrograph of the uncoated scaffold shows the clear demarcation of grain boundaries. For the mesoporous HA coated sample (Fig. 4b), it could be found that the surface of β-TCP scaffold was fully covered by a crack-free coating. While the coating was prepared by five cycles of dip-coating in sol, the grain morphology under the coating can be dimly observed, suggesting that the thickness of the obtained coating is below micron scale. 3.3. In vitro cell growth To evaluate the in vitro cytocompatibility of the mesoporous HA coated β-TCP scaffolds, the cell growth on the surfaces of samples N-HA and M-HA was observed after the incubation for different periods. As shown in Fig. 5, the cells adhered to the non-mesoporous and
Fig. 2. BET adsorption plot of samples N-HA and M-HA. Inset reveals the specific surface area.
5004
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
Fig. 4. Surface morphology of β-TCP porous scaffold before (a) and after (b) coating with the mesoporous HA.
mesoporous HA coated scaffolds after culture for 3 days, and presented different degrees of spreading after 7 days of culture. On day 3 (Fig. 5a and b), it could be found that MC3T3-E1 cells attached to the surfaces of both samples N-HA and M-HA, and no obvious difference was observed between the two groups. On day 7, as compared to sample N-HA (Fig. 5c), the cells have spread, connected with each other and almost covered the surface of sample M-HA (Fig. 5d). In addition, MC3T3-E1 cells on sample N-HA exhibited a relatively round morphology with sphere-like surface evaginations (Fig. 6a, the arrows pointing to). In contrast, the cells on sample M-HA spread in an elongated and flat shape (Fig. 6b), indicating the better attachment of MC3T3-E1 cells to the mesoporous HA coating. An in vitro cell assay revealed that the presence of the mesoporous structure was beneficial for the spreading of the cells on HA coating. 3.4. Bone-related protein expression Bone maturation is related to the expression of bone-related proteins, so it is useful to evaluate the effect of the mesoporous structure on the protein expression, such as alkaline phosphatase (ALP) and bone sialoprotein (BSP), which were studied in this work. As a membrane-binding protein with the catalytic domain on the osteoblastic plasmalemma, ALP is considered to be a marker of early osteogenic
development [12]. Fig. 7 shows that the production of ALP in both groups increased with the culture time, while ALP values of sample M-HA were higher than those of sample N-HA over the entire assay period of 14 days. For mesoporous coating sample M-HA, after 3 and 7 days of incubation, ALP values reached 137.0 U/L and 209.3 U/L respectively, markedly higher than those of samples N-HA (81.0 U/L and 128.0 U/L) at the corresponding incubation times (⁎p b 0.05). Increasing the incubation time to 14 days, the ALP value of sample M-HA (228.6 U/L) was still higher than that of sample N-HA (172.6 U/L), but no statistical differences were detected between samples N-HA and M-HA, displaying that the difference of the ALP values in the two groups decreased in the final stage of the incubation. BSP, one of the non-collagenous proteins, would act in the nucleation of HA crystals and its temporo-spatial expression would be identified with de novo mineralization in bone [13]. As shown in Fig. 8, the production of BSP in both groups was enhanced with the increased time, and BSP values in the mesoporous group were higher than those in the non-mesoporous group over the assay period. At day 3 and day 7, BSP values of sample M-HA were measured to be 6.9 nmol/L and 8.9 nmol/L respectively, being higher than those of sample N-HA (4.9 nmol/L and 6.3 nmol/L). Moreover, there were no statistical differences between the two groups after 3 and 7 days of incubation. Thereafter, the BSP value of sample M-HA rapidly increased to 15.5 nmol/L
Fig. 5. SEM micrographs of the cells on samples N-HA (a, c) and M-HA (b, d) after 3 days and 7 days of culture.
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
5005
Fig. 6. Morphology of the cells on samples N-HA (a) and M-HA (b) after 7 days of culture.
at day 14, and was markedly higher than the 10.0 nmol/L value of sample N-HA (⁎p b 0.05). It could be stated that the difference of BSP values between the non-mesoporous and mesoporous groups increases as the incubation time increases. The results show that the mesoporous structure would influence the protein production in MC3T3-E1 cells grown on HA coatings, and the comparison of protein expression in N-HA and M-HA groups could be summarized as follows: (1) the cells cultured on the mesoporous HA coating exhibited higher protein expression than those on the non-mesoporous HA coating at the corresponding incubation times; (2) the protein expression displayed in certain sequence for both NHA and M-HA groups, that is, the expression of ALP is more significantly increased in the initial periods of the incubation (0–7 days), whereas that of BSP production is higher in the final stage (7–14 days).
4. Discussion In this work, mesoporous HA coated β-TCP scaffolds were prepared by a dip-coating technique using a mesoporous HA precursor as a solution, and followed heat treatment at a temperature of 500 °C. Mesoporous HA powders are usually prepared using calcium nitrate and phosphates as raw materials and CTAB or F127 as the templates [14]. In this study, Ca(NO3)2·4H2O and P2O5 were selected as inorganic species and F127 was used as the template. Herein, alkyl phosphate with a low amount of phosphoric acid was obtained by dissolving P2O5 into anhydrous ethanol, and phosphoric acid would mainly transform into dihydrogen phosphate ions by the ionization [15]. In order to
Fig. 7. ALP expression of MC3T3-E1 on the non-mesoporous and mesoporous HA coated β-TCP scaffolds after different incubation times. The asterisk indicates the significant difference in protein expression between the N-HA and M-HA groups (⁎p b 0.05).
better understand the formation process of mesoporous HA, a possible process is schematically illustrated in Fig. 9. The poly-(ethylene oxide) (PEO) and poly-(propylene oxide) (PPO) blocks of F127 were both hydrophilic at temperatures below 15 °C. With the increase of temperature, the PPO block tended to be hydrophobic and acted as the core of F127 micelles, and the PEO block existed as the shells of the micelles. Initially, the PEO block of F127 micelles attracted the hydroxyl groups in alkyl phosphate and dihydrogen phosphate ions by hydrogen bonding to form F127-phosphate micelles (reaction (1)). Then, the F127-phosphate micelles interacted with Ca ions in the solution by electrostatic force to form the mesoporous HA precursor micelles (reaction (2)). Corresponding to the theory of the intendance to get an optimum geometrical morphology with the most stable state [14], the spherical precursor micelles would induce the formation of spherical mesoporous HA particles after the calcination, as the TEM micrograph shown in Fig. 9. When the above solution was coated on β-TCP scaffolds, followed by gelation and heat treatment, a mesoporous structure was formed in the coatings. As bone graft materials, their biological properties would be dependent on the phase composition as well as surface topography and properties. The preceding results using XRD analysis shown in Fig. 3 have confirmed that the non-mesoporous and mesoporous coatings have the same phase composition including the major phase HA and minor phases β-TCP and CPP. Meanwhile, the coated scaffolds presented similar porous structure to that of the β-TCP scaffolds. Therefore, the differences in cell growth and protein expression between samples N-HA and M-HA should be attributed to the introduction of a mesoporous structure in the HA coatings. To clarify whether the introduction of a mesoporous structure in the HA coatings affected the biocompatibility and the ability to promote tissue formation of the coated β-TCP scaffolds, in vitro assays such as cell adhesion and spreading, and ALP and BSP expression were performed using osteogenic cell line MC3T3-E1 as cellular models. As shown in Fig. 5a and b, MC3T3-E1 cells attached to samples N-HA and M-HA presented no obvious difference after 3 days of incubation, while the differences of the cell attachment, spreading and proliferation were detected after 7 days of culture (Fig. 5c and d). As the widely used bone repair materials, both the non-mesoporous and mesoporous HA coatings with no cytotoxicity could provide the tolerable surfaces for cell adhesion; therefore, MC3T3-E1 cells were found to attach to the surfaces of both samples N-HA and M-HA at day 3. However, the subtle changes in local micro-environment caused by the introduction of mesopores in the HA coatings resulted in significant discrepancies in cell spreading and proliferation after culture for 7 days. It is well known that the adsorption of proteins onto the surfaces of biomaterials would directly influence the subsequent cell behavior, and the protein adsorption onto the solid surfaces would be affected by the interactions between proteins, water and interface [16]. For the hydrophilic materials, such as HA coatings in this work, electrostatic force and hydrogen bonding would play an important role in the protein adsorption [17]. Considering the identical phase composition of samples N-HA and M-HA, the electrostatic interaction would have a similar influence on
5006
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
coatings, its wetting mechanism should follow the Wenzel model [18] described as follows: cos θW ¼ r cos θY ;
Fig. 8. BSP expression of MC3T3-E1 on the non-mesoporous and mesoporous HA coated β-TCP scaffolds after different incubation times. The asterisk indicates the significant difference in protein expression between the N-HA and M-HA groups (⁎p b 0.05).
the protein adsorption in the two groups. Therefore, the significant differences in cell behavior between N-HA and M-HA groups should be attributed to the effect of hydrogen bonding, which is related to the surface wettability of HA coatings. For the hydrophilic surfaces of HA
ð1Þ
where θW is the apparent contact angle, θY is the ideal contact angle (Young's angle) of water on a surface, and r is the roughness factor that is defined as the ratio of actual surface area over the projected area. It could be stated that a high surface area would have a beneficial effect on the promotion of the surface wettability of a HA coating. As mentioned above, the introduction of a mesoporous structure increased the surface area of the HA coating from 9.17 m2/g to 211.7 m2/g, and decreased the water contact angles on the HA coating from 67.8° to 30.1° as a consequence [9]. Therefore, sample M-HA with higher surface hydrophilicity is more suitable for the formation of a water molecule layer on its surface than sample N-HA. Subsequently, hydrogen bonding formed between the water molecule layer and the residues of proteins, in favor of adsorbing the proteins onto the surface of sample M-HA. Arima et al. [19] prepared the self-assembled monolayers (SAMs) functionalized with CH3/COOH groups, and found that the amounts of adsorbed serum proteins increased as the water contact angle on SAMs decreased. It could be indicated that the protein adsorption would be improved with the promotion of surface wettability for sample M-HA due to the mesoporous structure. Because the cells would recognize the motifs within the proteins and adhere to the surface, the enhanced protein adsorption of sample M-HA is beneficial for the attachment of osteoblast on the mesoporous HA coating. Therefore, sample M-HA showed better in vitro cytocompatibility than sample N-
Fig. 9. Schematic illustration of the formation of mesoporous HA particles. Reaction (1) shows the formation of F127-phosphate micelles by hydrogen bonding. Reaction (2) reveals the formation of mesoporous HA precursor micelles by electrostatic force. The spherical precursor micelles induced the formation of spherical mesoporous HA particles after the calcination.
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
HA with the increase in the incubation period, which was confirmed by the cell experiment. For sample N-HA, the cells on its surface exhibited a relatively round shape with sphere-like surface evaginations and the microvilli (Fig. 6a), revealing a low degree of cell spreading. In contrast, more cells were well attached to sample M-HA and spread in an elongated shape (Fig. 6b). The mesoporous HA coating provided a surface with higher availability of cell-anchoring domains, stimulating the cell spreading and proliferation. As anchorage-dependent cells, the attachment of MC3T3-E1 cells would also determine the synthesis and secretion of bone-related proteins in cells, such as ALP and BSP. As shown in Figs. 7 and 8, the results of the protein production assay recorded significant differences on protein expression between the non-mesoporous and mesoporous HA groups. The production of ALP and BSP was higher in the cells cultured on sample M-HA than on sample N-HA at all the designed incubation periods, which is good for the osteogenic development, due to the important role of ALP and BSP in cell differentiation or de novo mineralization. Because the mesoporous HA coating provided a suitable surface for cell growth, the cells on sample M-HA displayed a more elongated shape and a higher extent of proliferation than those on sample N-HA as the incubation time increased. The cells on sample M-HA connected with each other and formed the cell multilayer on the mesoporous HA coating, being beneficial for the cell function including the synthesis and secretion of ALP and BSP. Therefore, the cells cultured on the mesoporous HA coating exhibited higher protein expression than those on the non-mesoporous HA coating at the corresponding incubation times. In addition to that, it could be also found that the ALP value of sample M-HA was markedly higher than that of sample N-HA at day 3 and day 7 (⁎p b 0.05), whereas the BSP value of the mesoporous HA coated scaffold was markedly higher at day 14 (⁎p b 0.05). This phenomenon is consistent with the study on the bone-related gene expression in osteoblasts in which ALP was highly expressed in the early stage of osteogenesis, whereas BSP was expressed late in bone maturation [20]. As bone-related proteins, ALP and BSP must be activated in a correct sequence to ensure the correct temporo-spatial expression in osteogenesis, otherwise the overexpression or underexpression of certain proteins would lead to cellular growth disorders or bone diseases [21]. Thus, in the present study, the presence of the mesoporous structure in the HA coating would improve the production of bonerelated proteins on the premise of maintaining the protein expression in a correct sequence. Based upon these analyses, the influence of mesoporous HA coating on the behavior of MC3T3-E1 cell could be summarized as follows: (1) the introduction of mesopores improved the surface area and surface wettability of the HA coating; (2) the improvement of surface wettability promoted the cell growth on the HA coating, which affected the subsequent protein production in cells; (3) the enhancement of protein expression would improve the bone maturation as a consequence. Therefore, it could be speculated that a mesoporous structure would influence the biological properties of the HA coating and that the mesoporous HA coating with a high surface area might be a candidate for the surface modification of the tissue engineering scaffolds. 5. Conclusions Cell growth and bone-related protein expression in MC3T3-E1 cells on the mesoporous HA coated β-TCP porous scaffold were investigated in this study. The variation of surface topography and properties led to the discrepancies in the behavior of the cells grown on HA coating.
5007
After 7 days of culture, the cells with an elongated and flat shape were observed on the surface of the mesoporous HA coated sample. Additionally, there are significant differences between the non-mesoporous and mesoporous HA coating groups in the expression of ALP at 3 and 7 days, and BSP at 14 days (⁎p b 0.05). The expression of ALP and BSP for the mesoporous HA coated scaffold was within a correct sequence of the bone-related protein expression. It could be concluded that the introduction of the mesoporous structure changed the surface properties and enhanced the biological properties of the HA coating. Acknowledgments The authors acknowledge the financial support by the National Nature Science Foundation of China (Grant Nos. 51072129, 81271954), and the Natural Science Foundation of Tianjin (Grant No. 11 JCYBJC02600). References [1] Y.P. Guo, Y. Zhou, D.C. Jia, C.Q. Ning, Y.J. Guo, Mesoporous structure and evolution mechanism of hydroxycarbonate apatite microspheres, Mater. Sci. Eng. C 30 (2010) 472. [2] M. Vallet-Regi, C.V. Rages, A.J. Salinas, Glasses with medical applications, Eur. J. Inorg. Chem. 6 (2003) 1029. [3] V.L. Chavez, L. Song, S. Barua, X. Li, Q. Wu, D. Zhao, K. Rege, B.D. Vogt, Impact of nanopore morphology on cell viability on mesoporous polymer and carbon surfaces, Acta Biomater. 6 (2010) 3035. [4] C. Wu, W. Fan, M. Gelinsky, Y. Xiao, P. Simon, R. Schulze, T. Doert, Y. Luo, G. Cuniberti, Bioactive SrO–SiO2 glass with well-ordered mesopores: characterization, physiochemistry and biological properties, Acta Biomater. 7 (2011) 1797. [5] C. Wu, Y. Zhang, Y. Zhou, W. Fan, Y. Xiao, A comparative study of mesoporous glass/silk and non-mesoporous glass/silk scaffolds: physiochemistry and in vivo osteogenesis, Acta Biomater. 7 (2011) 2229. [6] T. Dey, P. Roy, B. Fabry, P. Schmuki, Anodic mesoporous TiO2 layer on Ti for enhanced formation of biomimetic hydroxyapatite, Acta Biomater. 7 (2011) 1873. [7] H. Tang, Y. Guo, D. Jia, Y. Zhou, Preparation and in vitro characterization of crack-free mesoporous titania films, Surf. Coat. Technol. 206 (2011) 8. [8] A.A. Ribeiro, R.F.C. Marques, A.C. Guastaldi, J.S. de Carvalho Campos, Hydroxyapatite deposition study through polymeric process on commercially pure Ti surfaces modified by laser beam irradiation, J. Mater. Sci. 44 (2009) 4056. [9] X. Ye, S. Cai, G. Xu, Y. Dou, H. Hu, Synthesis of mesoporous hydroxyapatite thin films using F127 as templates for biomedical applications, Mater. Lett. 85 (2012) 64. [10] S. Zhang, J. Li, Y. Song, C. Zhao, X. Zhang, C. Xie, Y. Zhang, H. Tao, Y. He, Y. Jiang, Y. Bian, In vitro degradation, hemolysis and MC3T3-E1 cell adhesion of biodegradable Mg-Zn alloy, Mater. Sci. Eng. C 29 (2009) 1907. [11] L. Song, D. Zhu, X. Sun, S. Wang, D. Sun, Hierarchical assembly and synthesis of mesoporous spherical hydroapatite nanoparticles, Acta Chim. Sin. 67 (2009) 2697. [12] K.W. Lee, C.M. Bae, J.Y. Jung, G.B. Sim, T.R. Rautray, H.J. Lee, T.Y. Kwon, K.H. Kim, Surface characteristics and biological studies of hydroxyapatite coating by a new method, J. Biomed. Mater. Res. B Appl. Biomater. 98B (2011) 395. [13] L. Malaval, F. Liu, P. Roche, J.E. Aubin, Kinetics of osteopro-genitor proliferation and osteoblast differentiation in vitro, J. Cell. Biochem. 74 (1999) 16. [14] Y.F. Zhao, J. Ma, Triblock co-polymer templating synthesis of mesostructured hydroxyapatite, Microporous Mesoporous Mater. 87 (2005) 110. [15] A.K. Lynn, W. Bonfield, A novel method for the simultaneous, titrant-free control of pH and calcium phosphate mass yield, Acc. Chem. Res. 38 (2005) 202. [16] H. Zeng, K.K. Chittur, W.R. Lacefield, Analysis of bovine serum albumin adsorption on calcium phosphate and titanium surface, Biomaterials 20 (1999) 377. [17] J.W. Shen, T. Wu, Q. Wang, H.H. Pan, Molecular simulation of protein adsorption and desorption on hydroxyapatite surfaces, Biomaterials 29 (2008) 513. [18] E. Martines, K. Seunarine, H. Morgan, N. Gadegaard, C.D.W. Wilkinson, M.O. Riehle, Superhydrophobicity and superhydrophilicity of regular nanopatterns, Nano Lett. 5 (2005) 2097. [19] Y. Arima, H. Iwata, Effect of wettability and surface functional groups on protein adsorption and cell adhesion using well-defined mixed self-assembled monolayers, Biomaterials 28 (2007) 3074. [20] C. Wang, Y. Duan, B. Markovic, J. Barbara, C.R. Howlett, X. Zhang, H. Zreiqat, Phenotypic expression of bone-related genes in osteoblasts grown on calcium phosphate ceramics with different phase compositions, Biomaterials 25 (2004) 2507. [21] M.M. Stevens, J.H. George, Exploring and engineering the cell surface interface, Science 310 (2005) 1135.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Preparation and in vitro evaluation of mesoporous hydroxyapatite coated β-TCP porous scaffolds Xinyu Ye a, Shu Cai a,⁎, Guohua Xu b,⁎⁎, Ying Dou a, Hongtao Hu b, Xiaojian Ye b a b
Key Laboratory for Advanced Ceramics and Machining Technology of Ministry of Education, School of Materials Science and Engineering, Tianjin University, Tianjin 300072, People's Republic of China Shanghai Changzheng Hospital, Shanghai 200003, People's Republic of China
a r t i c l e
i n f o
Article history: Received 20 May 2013 Received in revised form 22 July 2013 Accepted 23 August 2013 Available online 31 August 2013 Keywords: Mesoporous hydroxyapatite Coatings β-tricalcium phosphate (β-TCP) scaffold Cell response Protein expression
a b s t r a c t A mesoporous hydroxyapatite (HA) coating was prepared on a β-tricalcium phosphate (β-TCP) porous scaffold by a sol-gel dip-coating method using the block copolymer Pluronic F127 (EO106PO70EO106) as the template. For application as a bone graft, in vitro cell response and bone-related protein expression of mesoporous HA coated β-TCP scaffold were investigated, using the non-mesoporous HA coated scaffold as the control group, to evaluate the influence of the mesoporous structure on the biological properties of HA coating. It was found that the increased surface area of the mesoporous HA coating greatly affected the response of MC3T3-E1 osteoblasts and the expression of proteins. An enzyme-linked immunosorbent assay recorded a significantly higher expression of alkaline phosphatase (ALP) and bone sialoprotein (BSP) in the mesoporous group than those in the control group (*p b 0.05) after different incubation periods. The introduction of mesopores enhanced the expression of ALP and BSP in the cells grown on the mesoporous HA coatings, on the premise of maintaining the protein expression in a sequence to ensure the correct temporo-spatial expression in osteogenesis. These results indicated that the mesoporous HA coating would provide a good environment for cell growth, suggesting that it could be used as the coating material for the surface modification of the tissue engineering scaffolds. © 2013 Elsevier B.V. All rights reserved.
1. Introduction The bone-forming ability of biomaterials is related to chemical composition as well as textural properties, such as pore size and pore volume [1]. Increasing the specific surface area and pore volume of biomaterials would accelerate the kinetic deposition process of hydroxycarbonate apatite (HCA) and consequently enhance the boneforming bioactivity [2]. Owing to the large surface area and pore volume, mesoporous materials have been proposed for use in biological applications such as drug release, bioadsorption and biosensing [3]. Recently the research related to mesoporous materials for bone tissue regeneration has been performed by several research groups [4–7]. For example, mesoporous SiO2–CaO–P2O5 scaffolds or powders have been prepared and demonstrated the superior bone-forming bioactivities in vitro compared to normal bioglass [5]. Mesoporous titania layers could lead to a significant enhancement of apatite formation compared with non-coated or even nano-tubular Ti surface coatings [6]. As compared to other bone repairing materials, hydroxyapatite (HA) possesses the favorable biocompatibility and osteoconductivity, and HA coating has been considered to be the optimized choice for surface modification
⁎ Corresponding author. Tel.: +86 22 27425069. ⁎⁎ Corresponding author. Tel.: +86 21 81886999. E-mail addresses: [email protected] (S. Cai), [email protected] (G. Xu). 0928-4931/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msec.2013.08.027
by combining with other materials [8]. Therefore, HA coating with mesostructure would display the suitable properties for use as the biomaterials for bone repair. In our preliminary work, the uniform and crack-free mesoporous HA coating was synthesized on the glass substrates by a sol-gel method, using F127 (EO106PO70EO106, EO is ethylene oxide, PO is propylene oxide) as the template [9]. In vitro tests revealed that MC3T3-E1 cells cultured on the mesoporous HA coating showed better surface wettability and cell spreading than on the non-mesoporous HA coating. The presence of mesopores was preliminarily confirmed to be a key parameter influencing the cell response to HA coating. Besides that, the differences in cell behavior may affect the expression of bone-related proteins, which plays an important role in cell differentiation and de novo mineralization in bone. Therefore, it is necessary to evaluate the expression of bone-related proteins in the cells grown on the mesoporous HA coating to further clarify the effect of mesopores on the cell response and bone-forming ability. In this study, we focused on the cell response and bone-related protein expression in MC3T3-E1 cells of the mesoporous HA coating (denoted as M-HA), using the non-mesoporous HA coating (denoted as N-HA) with a similar surface morphology and phase composition to those of the control group. Considering the potential use for the surface modification of implants, the porous β-TCP scaffolds were selected as the substrates to investigate the effect of mesoporous HA coating on their biological properties.
5002
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
2. Materials and methods
2.3. In vitro evaluation
2.1. Sample preparation
Osteoblast-like cells MC3T3-E1 [10] were cultured in Dulbecco's Modified Eagle's Medium (DMEM) containing 10% fetal bovine serum (FBS), and maintained at 37 °C in a saturated humidified atmosphere of 5% CO2. The cells were detached with a 0.1% trypsin-EDTA solution in phosphate-buffered saline (PBS, pH 7.4), centrifuged at 1000 rpm for 5 min, washed and resuspended. Samples N-HA and M-HA were cleaned by physiological saline, continually rinsed with distilled water, and sterilized using Cobalt-60 radiation. The cells were seeded on the samples in a 24-well plate at a concentration of 1.0 × 105 cells/mL in 1.5 mL, and incubated with 5% CO2 at 37 °C for different culture periods. The culture medium was renewed twice in a week. After being cultured for 3 and 7 days, the samples were washed with phosphate buffered saline (PBS), fixed using 2% glutaraldehyde for 2 h, washed by PBS for 20 min after drawing off the fixed agent, dehydrated through a graded alcohol solution, and then critical point drying was performed with an apparatus (HCP-2, Hitachi Koki Ltd., Japan). After being coated with gold, the attachment of MC3T3-E1 cells on the coatings was observed by SEM (XL-30, Philips). For a protein production assay, the cells were harvested at 3, 7 and 14 days, and transferred into a 96-well plate. The plates were centrifuged at 3000 rpm for 10 min and the medium was aspirated from the wells. After being placed in an incubator at 50 °C for 30 min, the dried plates were wrapped with Clingwrap and stored at room temperature for the further tests. The production of alkaline phosphatase (ALP) and bone sialoprotein (BSP) in MC3T3-E1 cells were detected by using an enzyme-linked immunosorbent assay (ELISA) kit (EYSIN Co., Shanghai, China). Briefly, standards or analytes were added into the wells, and biotin-antibodies were then added and incubated. Subsequently, horseradish peroxidase avidin (HRP-avidin) was added to bind the detection antibodies, and a substrate solution was used to develop the color. The reaction was terminated by the addition of an acidic stop solution, and the optical density (OD) of the yellow color was immediately determined at 450 nm. Because the absorbance was proportional to the amount of analyte, the concentration of ALP and BSP was measured in reference to the standard curve.
M-HA sol was prepared by using Ca(NO3)2·4H2O and P2O5 as raw materials, and Pluronic F127 (Sigma, USA) as the template for introducing the mesostructure. As described in our previous study [9], 0.033 g/mL of a P2O5 (Kermel, Tianjin) ethanol solution and 0.136 g/mL of a Ca(NO3)2 · 4H2O (Damao, Tianjin) aqueous solution were successively added into 0.192 g/mL of an F127 ethanol solution at 40 °C under vigorous stirring for producing the sol. For comparison, N-HA sol was prepared at the same conditions but without F127. β-TCP porous scaffolds made in our lab (Ф = 12.0 mm × 6.0 mm, porosity: 70.0%–75.0%) were used as the substrates, cleaned in acetone, alcohol and distilled water for 10 min using an ultrasonic cleaner, and then dried in air. The coatings were synthesized by dip-coating on the substrates at a constant rate of 1.0 mm/s, and this process was repeated for five times. After aging at room temperature for 48 h and drying at 60 °C for 12 h, the as-synthesized coatings were calcined at 500 °C for 3 h to remove the template. In addition, N-HA and M-HA powders were also synthesized as the above process for the phase composition and porous structure analyses. 2.2. Characterization Morphological and microstructural observations of specimens were carried out by scanning electron microscopy (SEM, XL-30, Philips) and transmission electron microscopy (TEM, JEM-2010HR, JEOL). The phase composition of samples N-HA and M-HA was analyzed by using X-ray diffraction (XRD, D/Max-2500, Philips) using Cu Kα radiation over the range from 10° to 60°. N2 adsorption–desorption isotherms were measured with a Quantachrome NOVA analyzer. The pore size distribution of the non-mesoporous and mesoporous samples was calculated from the adsorption–desorption data using the density functional theory (DFT) model. The specific surface area of HA was calculated according to the Brunauer–Emmett–Teller (BET) equation. For each group, five tests were performed and the results were averaged.
Fig. 1. N2 adsorption–desorption isotherm and pore size distribution curve of samples N-HA and M-HA.
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
5003
Statistical analysis was performed using SPSS 11.0 to evaluate the differences between the M-HA and N-HA groups. Each data point was presented as means ± deviation (X SD) from five independent experiments and a p value of b0.05 represented a significant difference. 3. Results 3.1. Characterization of mesoporous structure The characterization of the mesoporous structure in sample M-HA was carried out by N2 adsorption–desorption measurements. Fig. 1 shows the adsorption–desorption isotherms and the pore size distribution curves based on the DFT model for samples N-HA and M-HA. For sample N-HA, no hysteresis loops were exhibited in N2 isotherms, revealing the non-mesoporous structure in the coating. In the case of sample M-HA, N2 isotherms presented a type IIb curve with an H3 hysteresis loop for the typically non-rigid and disordered mesoporous structure. Moreover, the pore size distribution curves present the introduction of the mesopores of ~3.17 nm in the M-HA group, as compared to the micropores of ~0.42 nm in the N-HA group. Meanwhile, as shown in Fig. 2, the specific surface area of sample M-HA increased from 9.17 m2/g to 211.7 m2/g due to the presence of mesopores. 3.2. Phase composition and surface morphology XRD patterns shown in Fig. 3 indicate that samples N-HA and M-HA have the same phase composition, including the major phase hydroxyapatite (HA, PDF#09-0432) and the minor phases tricalcium phosphate (β-TCP, PDF#09-0169) and calcium pyrophosphate (CPP, PDF#090346). The crystallization degree of samples N-HA and M-HA were calculated to be 82.45% and 80.81% respectively using MDI JADE 5.0. There is no significant difference in the crystallization degree between the two groups. Meanwhile, as seen from the XRD patterns, the intensities of some diffraction peaks are higher and others are lower for sample M-HA, as compared to sample N-HA. It should be attributed to the crystal growth along different planes due to the presence of the template and mesoporous structure [11].
Fig. 3. XRD patterns of samples N-HA (a) and M-HA (b).
Fig. 4 presents the surface morphology of β-TCP porous scaffolds before and after coating with the mesoporous HA. As shown in Fig. 4a, the surface micrograph of the uncoated scaffold shows the clear demarcation of grain boundaries. For the mesoporous HA coated sample (Fig. 4b), it could be found that the surface of β-TCP scaffold was fully covered by a crack-free coating. While the coating was prepared by five cycles of dip-coating in sol, the grain morphology under the coating can be dimly observed, suggesting that the thickness of the obtained coating is below micron scale. 3.3. In vitro cell growth To evaluate the in vitro cytocompatibility of the mesoporous HA coated β-TCP scaffolds, the cell growth on the surfaces of samples N-HA and M-HA was observed after the incubation for different periods. As shown in Fig. 5, the cells adhered to the non-mesoporous and
Fig. 2. BET adsorption plot of samples N-HA and M-HA. Inset reveals the specific surface area.
5004
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
Fig. 4. Surface morphology of β-TCP porous scaffold before (a) and after (b) coating with the mesoporous HA.
mesoporous HA coated scaffolds after culture for 3 days, and presented different degrees of spreading after 7 days of culture. On day 3 (Fig. 5a and b), it could be found that MC3T3-E1 cells attached to the surfaces of both samples N-HA and M-HA, and no obvious difference was observed between the two groups. On day 7, as compared to sample N-HA (Fig. 5c), the cells have spread, connected with each other and almost covered the surface of sample M-HA (Fig. 5d). In addition, MC3T3-E1 cells on sample N-HA exhibited a relatively round morphology with sphere-like surface evaginations (Fig. 6a, the arrows pointing to). In contrast, the cells on sample M-HA spread in an elongated and flat shape (Fig. 6b), indicating the better attachment of MC3T3-E1 cells to the mesoporous HA coating. An in vitro cell assay revealed that the presence of the mesoporous structure was beneficial for the spreading of the cells on HA coating. 3.4. Bone-related protein expression Bone maturation is related to the expression of bone-related proteins, so it is useful to evaluate the effect of the mesoporous structure on the protein expression, such as alkaline phosphatase (ALP) and bone sialoprotein (BSP), which were studied in this work. As a membrane-binding protein with the catalytic domain on the osteoblastic plasmalemma, ALP is considered to be a marker of early osteogenic
development [12]. Fig. 7 shows that the production of ALP in both groups increased with the culture time, while ALP values of sample M-HA were higher than those of sample N-HA over the entire assay period of 14 days. For mesoporous coating sample M-HA, after 3 and 7 days of incubation, ALP values reached 137.0 U/L and 209.3 U/L respectively, markedly higher than those of samples N-HA (81.0 U/L and 128.0 U/L) at the corresponding incubation times (⁎p b 0.05). Increasing the incubation time to 14 days, the ALP value of sample M-HA (228.6 U/L) was still higher than that of sample N-HA (172.6 U/L), but no statistical differences were detected between samples N-HA and M-HA, displaying that the difference of the ALP values in the two groups decreased in the final stage of the incubation. BSP, one of the non-collagenous proteins, would act in the nucleation of HA crystals and its temporo-spatial expression would be identified with de novo mineralization in bone [13]. As shown in Fig. 8, the production of BSP in both groups was enhanced with the increased time, and BSP values in the mesoporous group were higher than those in the non-mesoporous group over the assay period. At day 3 and day 7, BSP values of sample M-HA were measured to be 6.9 nmol/L and 8.9 nmol/L respectively, being higher than those of sample N-HA (4.9 nmol/L and 6.3 nmol/L). Moreover, there were no statistical differences between the two groups after 3 and 7 days of incubation. Thereafter, the BSP value of sample M-HA rapidly increased to 15.5 nmol/L
Fig. 5. SEM micrographs of the cells on samples N-HA (a, c) and M-HA (b, d) after 3 days and 7 days of culture.
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
5005
Fig. 6. Morphology of the cells on samples N-HA (a) and M-HA (b) after 7 days of culture.
at day 14, and was markedly higher than the 10.0 nmol/L value of sample N-HA (⁎p b 0.05). It could be stated that the difference of BSP values between the non-mesoporous and mesoporous groups increases as the incubation time increases. The results show that the mesoporous structure would influence the protein production in MC3T3-E1 cells grown on HA coatings, and the comparison of protein expression in N-HA and M-HA groups could be summarized as follows: (1) the cells cultured on the mesoporous HA coating exhibited higher protein expression than those on the non-mesoporous HA coating at the corresponding incubation times; (2) the protein expression displayed in certain sequence for both NHA and M-HA groups, that is, the expression of ALP is more significantly increased in the initial periods of the incubation (0–7 days), whereas that of BSP production is higher in the final stage (7–14 days).
4. Discussion In this work, mesoporous HA coated β-TCP scaffolds were prepared by a dip-coating technique using a mesoporous HA precursor as a solution, and followed heat treatment at a temperature of 500 °C. Mesoporous HA powders are usually prepared using calcium nitrate and phosphates as raw materials and CTAB or F127 as the templates [14]. In this study, Ca(NO3)2·4H2O and P2O5 were selected as inorganic species and F127 was used as the template. Herein, alkyl phosphate with a low amount of phosphoric acid was obtained by dissolving P2O5 into anhydrous ethanol, and phosphoric acid would mainly transform into dihydrogen phosphate ions by the ionization [15]. In order to
Fig. 7. ALP expression of MC3T3-E1 on the non-mesoporous and mesoporous HA coated β-TCP scaffolds after different incubation times. The asterisk indicates the significant difference in protein expression between the N-HA and M-HA groups (⁎p b 0.05).
better understand the formation process of mesoporous HA, a possible process is schematically illustrated in Fig. 9. The poly-(ethylene oxide) (PEO) and poly-(propylene oxide) (PPO) blocks of F127 were both hydrophilic at temperatures below 15 °C. With the increase of temperature, the PPO block tended to be hydrophobic and acted as the core of F127 micelles, and the PEO block existed as the shells of the micelles. Initially, the PEO block of F127 micelles attracted the hydroxyl groups in alkyl phosphate and dihydrogen phosphate ions by hydrogen bonding to form F127-phosphate micelles (reaction (1)). Then, the F127-phosphate micelles interacted with Ca ions in the solution by electrostatic force to form the mesoporous HA precursor micelles (reaction (2)). Corresponding to the theory of the intendance to get an optimum geometrical morphology with the most stable state [14], the spherical precursor micelles would induce the formation of spherical mesoporous HA particles after the calcination, as the TEM micrograph shown in Fig. 9. When the above solution was coated on β-TCP scaffolds, followed by gelation and heat treatment, a mesoporous structure was formed in the coatings. As bone graft materials, their biological properties would be dependent on the phase composition as well as surface topography and properties. The preceding results using XRD analysis shown in Fig. 3 have confirmed that the non-mesoporous and mesoporous coatings have the same phase composition including the major phase HA and minor phases β-TCP and CPP. Meanwhile, the coated scaffolds presented similar porous structure to that of the β-TCP scaffolds. Therefore, the differences in cell growth and protein expression between samples N-HA and M-HA should be attributed to the introduction of a mesoporous structure in the HA coatings. To clarify whether the introduction of a mesoporous structure in the HA coatings affected the biocompatibility and the ability to promote tissue formation of the coated β-TCP scaffolds, in vitro assays such as cell adhesion and spreading, and ALP and BSP expression were performed using osteogenic cell line MC3T3-E1 as cellular models. As shown in Fig. 5a and b, MC3T3-E1 cells attached to samples N-HA and M-HA presented no obvious difference after 3 days of incubation, while the differences of the cell attachment, spreading and proliferation were detected after 7 days of culture (Fig. 5c and d). As the widely used bone repair materials, both the non-mesoporous and mesoporous HA coatings with no cytotoxicity could provide the tolerable surfaces for cell adhesion; therefore, MC3T3-E1 cells were found to attach to the surfaces of both samples N-HA and M-HA at day 3. However, the subtle changes in local micro-environment caused by the introduction of mesopores in the HA coatings resulted in significant discrepancies in cell spreading and proliferation after culture for 7 days. It is well known that the adsorption of proteins onto the surfaces of biomaterials would directly influence the subsequent cell behavior, and the protein adsorption onto the solid surfaces would be affected by the interactions between proteins, water and interface [16]. For the hydrophilic materials, such as HA coatings in this work, electrostatic force and hydrogen bonding would play an important role in the protein adsorption [17]. Considering the identical phase composition of samples N-HA and M-HA, the electrostatic interaction would have a similar influence on
5006
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
coatings, its wetting mechanism should follow the Wenzel model [18] described as follows: cos θW ¼ r cos θY ;
Fig. 8. BSP expression of MC3T3-E1 on the non-mesoporous and mesoporous HA coated β-TCP scaffolds after different incubation times. The asterisk indicates the significant difference in protein expression between the N-HA and M-HA groups (⁎p b 0.05).
the protein adsorption in the two groups. Therefore, the significant differences in cell behavior between N-HA and M-HA groups should be attributed to the effect of hydrogen bonding, which is related to the surface wettability of HA coatings. For the hydrophilic surfaces of HA
ð1Þ
where θW is the apparent contact angle, θY is the ideal contact angle (Young's angle) of water on a surface, and r is the roughness factor that is defined as the ratio of actual surface area over the projected area. It could be stated that a high surface area would have a beneficial effect on the promotion of the surface wettability of a HA coating. As mentioned above, the introduction of a mesoporous structure increased the surface area of the HA coating from 9.17 m2/g to 211.7 m2/g, and decreased the water contact angles on the HA coating from 67.8° to 30.1° as a consequence [9]. Therefore, sample M-HA with higher surface hydrophilicity is more suitable for the formation of a water molecule layer on its surface than sample N-HA. Subsequently, hydrogen bonding formed between the water molecule layer and the residues of proteins, in favor of adsorbing the proteins onto the surface of sample M-HA. Arima et al. [19] prepared the self-assembled monolayers (SAMs) functionalized with CH3/COOH groups, and found that the amounts of adsorbed serum proteins increased as the water contact angle on SAMs decreased. It could be indicated that the protein adsorption would be improved with the promotion of surface wettability for sample M-HA due to the mesoporous structure. Because the cells would recognize the motifs within the proteins and adhere to the surface, the enhanced protein adsorption of sample M-HA is beneficial for the attachment of osteoblast on the mesoporous HA coating. Therefore, sample M-HA showed better in vitro cytocompatibility than sample N-
Fig. 9. Schematic illustration of the formation of mesoporous HA particles. Reaction (1) shows the formation of F127-phosphate micelles by hydrogen bonding. Reaction (2) reveals the formation of mesoporous HA precursor micelles by electrostatic force. The spherical precursor micelles induced the formation of spherical mesoporous HA particles after the calcination.
X. Ye et al. / Materials Science and Engineering C 33 (2013) 5001–5007
HA with the increase in the incubation period, which was confirmed by the cell experiment. For sample N-HA, the cells on its surface exhibited a relatively round shape with sphere-like surface evaginations and the microvilli (Fig. 6a), revealing a low degree of cell spreading. In contrast, more cells were well attached to sample M-HA and spread in an elongated shape (Fig. 6b). The mesoporous HA coating provided a surface with higher availability of cell-anchoring domains, stimulating the cell spreading and proliferation. As anchorage-dependent cells, the attachment of MC3T3-E1 cells would also determine the synthesis and secretion of bone-related proteins in cells, such as ALP and BSP. As shown in Figs. 7 and 8, the results of the protein production assay recorded significant differences on protein expression between the non-mesoporous and mesoporous HA groups. The production of ALP and BSP was higher in the cells cultured on sample M-HA than on sample N-HA at all the designed incubation periods, which is good for the osteogenic development, due to the important role of ALP and BSP in cell differentiation or de novo mineralization. Because the mesoporous HA coating provided a suitable surface for cell growth, the cells on sample M-HA displayed a more elongated shape and a higher extent of proliferation than those on sample N-HA as the incubation time increased. The cells on sample M-HA connected with each other and formed the cell multilayer on the mesoporous HA coating, being beneficial for the cell function including the synthesis and secretion of ALP and BSP. Therefore, the cells cultured on the mesoporous HA coating exhibited higher protein expression than those on the non-mesoporous HA coating at the corresponding incubation times. In addition to that, it could be also found that the ALP value of sample M-HA was markedly higher than that of sample N-HA at day 3 and day 7 (⁎p b 0.05), whereas the BSP value of the mesoporous HA coated scaffold was markedly higher at day 14 (⁎p b 0.05). This phenomenon is consistent with the study on the bone-related gene expression in osteoblasts in which ALP was highly expressed in the early stage of osteogenesis, whereas BSP was expressed late in bone maturation [20]. As bone-related proteins, ALP and BSP must be activated in a correct sequence to ensure the correct temporo-spatial expression in osteogenesis, otherwise the overexpression or underexpression of certain proteins would lead to cellular growth disorders or bone diseases [21]. Thus, in the present study, the presence of the mesoporous structure in the HA coating would improve the production of bonerelated proteins on the premise of maintaining the protein expression in a correct sequence. Based upon these analyses, the influence of mesoporous HA coating on the behavior of MC3T3-E1 cell could be summarized as follows: (1) the introduction of mesopores improved the surface area and surface wettability of the HA coating; (2) the improvement of surface wettability promoted the cell growth on the HA coating, which affected the subsequent protein production in cells; (3) the enhancement of protein expression would improve the bone maturation as a consequence. Therefore, it could be speculated that a mesoporous structure would influence the biological properties of the HA coating and that the mesoporous HA coating with a high surface area might be a candidate for the surface modification of the tissue engineering scaffolds. 5. Conclusions Cell growth and bone-related protein expression in MC3T3-E1 cells on the mesoporous HA coated β-TCP porous scaffold were investigated in this study. The variation of surface topography and properties led to the discrepancies in the behavior of the cells grown on HA coating.
5007
After 7 days of culture, the cells with an elongated and flat shape were observed on the surface of the mesoporous HA coated sample. Additionally, there are significant differences between the non-mesoporous and mesoporous HA coating groups in the expression of ALP at 3 and 7 days, and BSP at 14 days (⁎p b 0.05). The expression of ALP and BSP for the mesoporous HA coated scaffold was within a correct sequence of the bone-related protein expression. It could be concluded that the introduction of the mesoporous structure changed the surface properties and enhanced the biological properties of the HA coating. Acknowledgments The authors acknowledge the financial support by the National Nature Science Foundation of China (Grant Nos. 51072129, 81271954), and the Natural Science Foundation of Tianjin (Grant No. 11 JCYBJC02600). References [1] Y.P. Guo, Y. Zhou, D.C. Jia, C.Q. Ning, Y.J. Guo, Mesoporous structure and evolution mechanism of hydroxycarbonate apatite microspheres, Mater. Sci. Eng. C 30 (2010) 472. [2] M. Vallet-Regi, C.V. Rages, A.J. Salinas, Glasses with medical applications, Eur. J. Inorg. Chem. 6 (2003) 1029. [3] V.L. Chavez, L. Song, S. Barua, X. Li, Q. Wu, D. Zhao, K. Rege, B.D. Vogt, Impact of nanopore morphology on cell viability on mesoporous polymer and carbon surfaces, Acta Biomater. 6 (2010) 3035. [4] C. Wu, W. Fan, M. Gelinsky, Y. Xiao, P. Simon, R. Schulze, T. Doert, Y. Luo, G. Cuniberti, Bioactive SrO–SiO2 glass with well-ordered mesopores: characterization, physiochemistry and biological properties, Acta Biomater. 7 (2011) 1797. [5] C. Wu, Y. Zhang, Y. Zhou, W. Fan, Y. Xiao, A comparative study of mesoporous glass/silk and non-mesoporous glass/silk scaffolds: physiochemistry and in vivo osteogenesis, Acta Biomater. 7 (2011) 2229. [6] T. Dey, P. Roy, B. Fabry, P. Schmuki, Anodic mesoporous TiO2 layer on Ti for enhanced formation of biomimetic hydroxyapatite, Acta Biomater. 7 (2011) 1873. [7] H. Tang, Y. Guo, D. Jia, Y. Zhou, Preparation and in vitro characterization of crack-free mesoporous titania films, Surf. Coat. Technol. 206 (2011) 8. [8] A.A. Ribeiro, R.F.C. Marques, A.C. Guastaldi, J.S. de Carvalho Campos, Hydroxyapatite deposition study through polymeric process on commercially pure Ti surfaces modified by laser beam irradiation, J. Mater. Sci. 44 (2009) 4056. [9] X. Ye, S. Cai, G. Xu, Y. Dou, H. Hu, Synthesis of mesoporous hydroxyapatite thin films using F127 as templates for biomedical applications, Mater. Lett. 85 (2012) 64. [10] S. Zhang, J. Li, Y. Song, C. Zhao, X. Zhang, C. Xie, Y. Zhang, H. Tao, Y. He, Y. Jiang, Y. Bian, In vitro degradation, hemolysis and MC3T3-E1 cell adhesion of biodegradable Mg-Zn alloy, Mater. Sci. Eng. C 29 (2009) 1907. [11] L. Song, D. Zhu, X. Sun, S. Wang, D. Sun, Hierarchical assembly and synthesis of mesoporous spherical hydroapatite nanoparticles, Acta Chim. Sin. 67 (2009) 2697. [12] K.W. Lee, C.M. Bae, J.Y. Jung, G.B. Sim, T.R. Rautray, H.J. Lee, T.Y. Kwon, K.H. Kim, Surface characteristics and biological studies of hydroxyapatite coating by a new method, J. Biomed. Mater. Res. B Appl. Biomater. 98B (2011) 395. [13] L. Malaval, F. Liu, P. Roche, J.E. Aubin, Kinetics of osteopro-genitor proliferation and osteoblast differentiation in vitro, J. Cell. Biochem. 74 (1999) 16. [14] Y.F. Zhao, J. Ma, Triblock co-polymer templating synthesis of mesostructured hydroxyapatite, Microporous Mesoporous Mater. 87 (2005) 110. [15] A.K. Lynn, W. Bonfield, A novel method for the simultaneous, titrant-free control of pH and calcium phosphate mass yield, Acc. Chem. Res. 38 (2005) 202. [16] H. Zeng, K.K. Chittur, W.R. Lacefield, Analysis of bovine serum albumin adsorption on calcium phosphate and titanium surface, Biomaterials 20 (1999) 377. [17] J.W. Shen, T. Wu, Q. Wang, H.H. Pan, Molecular simulation of protein adsorption and desorption on hydroxyapatite surfaces, Biomaterials 29 (2008) 513. [18] E. Martines, K. Seunarine, H. Morgan, N. Gadegaard, C.D.W. Wilkinson, M.O. Riehle, Superhydrophobicity and superhydrophilicity of regular nanopatterns, Nano Lett. 5 (2005) 2097. [19] Y. Arima, H. Iwata, Effect of wettability and surface functional groups on protein adsorption and cell adhesion using well-defined mixed self-assembled monolayers, Biomaterials 28 (2007) 3074. [20] C. Wang, Y. Duan, B. Markovic, J. Barbara, C.R. Howlett, X. Zhang, H. Zreiqat, Phenotypic expression of bone-related genes in osteoblasts grown on calcium phosphate ceramics with different phase compositions, Biomaterials 25 (2004) 2507. [21] M.M. Stevens, J.H. George, Exploring and engineering the cell surface interface, Science 310 (2005) 1135.