In vitro behavior of fluoridated hydroxyapatite coatings in organic-containing simulated body fluid

In vitro behavior of fluoridated hydroxyapatite coatings in organic-containing simulated body fluid

Materials Science and Engineering C 27 (2007) 244 – 250 In vitro behavior of fluoridated hydroxyapatite coatings in orga...

1MB Sizes 3 Downloads 37 Views

Materials Science and Engineering C 27 (2007) 244 – 250

In vitro behavior of fluoridated hydroxyapatite coatings in organic-containing simulated body fluid Yongsheng Wang a , Sam Zhang a,⁎, Xianting Zeng b , Kui Cheng a , Min Qian b , Wenjian Weng c a

School of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798 b Singapore Institute of Manufacturing Technology, 71 Nanyang Drive, Singapore 638075 c Department of Materials Science and Engineering, Zhejiang University, Hangzhou, Zhejiang, 310027, PR China Received 8 July 2005; accepted 3 March 2006 Available online 7 July 2006

Abstract A dense and pure hydroxyapatite [HA, Ca10(PO4)6(OH)2] coating and a fluoridated HA [Ca10(PO4)6(OH)0.67F1.33] are deposited on Ti6Al4V substrates by sol-gel dip coating method. Glucose and bovine serum albumin have been added in standard simulated body fluid (SBF) to form organic-containing SBF in simulation of the physiological blood plasma. The HA and the fluoridated HA coatings are immersed in the standard and modified SBF for time periods of 2, 4, 7, 14 and 28 days at 37 ± 0.1 °C. After soaking, the coating surface is examined for nucleation and growth of apatite using SEM morphological observation. The post-soaking SBF solutions are analyzed via Inductively Coupled Plasma spectroscopy for calcium ion concentration. The results show that at concentration of 40 g/L, bovine serum albumin has significant retardation effect on apatite precipitation from SBF onto pure or fluoridated HA coatings; Fluorine-incorporation in HA has positive bio-activation effect in both standard SBF and organic-containing SBF. However, glucose addition in SBF does not generate significant influence on the bioactivity of HA and fluoridated HA. © 2006 Elsevier B.V. All rights reserved. Keywords: Fluoridated Hydroxyapatite (FHA); In vitro; Simulated body fluid; Glucose; Bovine serum albumin; Sol-gel

1. Introduction Hydroxyapatite [HA, Ca10(PO4)6(OH)2] has been developed as coatings on metallic implants in the field of orthopedics and dentistry due to its chemical and biological similarity to human hard tissues and also direct bonding capability to the surrounding tissues [1,2]. It has been established that HA coatings promote early bone apposition and fixation for HA-coated implants by encouraging chemical bonding between new bone and the surface of HA [3]. HA coating is also believed to protect the metallic substrate form corrosion in the biological environment, as well as serving as an effective barrier against the release of toxic metal ions from the metallic substrates into the living body [4]. However, the high dissolution rate of HA renders it questionable long term stability because dissolution of HA

⁎ Corresponding author. Tel.: +65 6790 4400; fax: +65 6791 1859. E-mail address: [email protected] (Sam Zhang). 0928-4931/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.msec.2006.03.012

leads to disintegration of the coatings thus hinders the fixation of implants to the surrounding host tissues. Recently, fluoridated hydroxyapatite (fluoridated HA, Ca10(PO4)6 (OH)2−xFx, x is the degree of fluoridation) has been developed that possesses lower solubility than pure HA while maintaining the comparable bioactivity and biocompatibility [5–7]. In vitro tests that employ cell cultures are performed to evaluate the biological response of the related cells, while investigations in cell-free solutions with compositions that are similar to human body fluid allow the determination of chemical and mineralogical changes of the implants under a simulate physiological environments [8]. Simulated body fluid (SBF) with inorganic ion concentrations nearly to those of human blood plasma, which was first introduced by Kokubo et al., is the most popular simulating solution used in in vitro tests [9]. However, actual body fluid contains not only the inorganic components but also various kinds of organic components (such as carbohydrates and proteins) and the organic components would exert noticeable influence on the implants. Jaou et al. [10] and Balint et al. [11] reported that although sugar and/or

Y. Wang et al. / Materials Science and Engineering C 27 (2007) 244–250


Table 1 Chemical composition of human blood plasma compared to the ion concentration of Kokubo's SBF Inorganic ion concentration (mM)

Blood plasma Kokubo's SBF

Organic composition (mg/dL)


HPO2− 4






SO2− 4





2.5 2.5

1 1

142 142

103 148

1.5 1.5

5 5

27 4.2

0.5 0.5

3300–4000 /

880–3530 /

340–430 /

100 /

glucose have a minor influence on crystallization of HA, they significantly inhibit the crystallization process of Fluoridated Apatite (FA). This effect was attributed to the formation of nonstoichiometric apatite in the presence of sugar. The inhibition effects of carbohydrates on the bone mineralization also reported by other researchers [11,12]. Dorozhkin et al. [13,14]concluded that glucose exhibited negligible influence on crystallization of calcium phosphate based on in vitro tests with the glucose modified SBF solution. On effects of proteins, extensively investigations have been done on CaP biomaterials, especially on HA [15,16]. It has been reported that plasma

proteins would adsorb immediately on the surface of HA after it was implanted in vivo, and the initial cellular response was partly dependent on the proteins adsorbed by the implant surfaces [17]. The first protein layer adsorbed on the implant surface affects the cellular adhesion [18,19], differentiation and production of extracellular matrix production. It also affects dissolution [17], nucleation and crystal growth of HA [15] as well as the final fixation between the implant and surrounding tissues. Albumin is usually selected for this kind of study due to its high concentration in blood plasma, favorable diffusion coefficient and ability to bind other ions and molecules [20]. Therefore, it is unwise to neglect the influences of organic components in the in vitro tests. The objective of the current work is to study the effect of proteins and glucose on the apatite deposition on an HA and an FHA coating.

Fig. 1. XRD patterns of the coating fired at 600 °C: a) HA, b) fluoridated HA.

Fig. 2. XPS patterns of the fluoridated HA coating fired at 600 °C.

Fig. 3. Surface morphology of the sol-gel derived a) HA coating and b) fluoridated HA coating.


Y. Wang et al. / Materials Science and Engineering C 27 (2007) 244–250

2. Experimental 2.1. Coating preparation and characterization The process of preparation for dip-sols and deposition of FHA coatings are detailed in our previous work [6,7]. Based on our previous work, Ca10(PO4)6(OH)0.67F1.33 has much better bioactivity than other composition FHA coatings [21]. Therefore, current work chose to focus on this specific FHA for its bioactivity in organic-containing SBF. An HA coating was used as control. Titanium alloy (Ti6Al4V) of 20 × 30 × 1.2 mm with a final finish of polishing by silicon carbide sandpaper (1200#) was used as substrates. The deposition run was repeated 4 times for a final coating thickness of ∼ 1.5 μm. The phase characterization of the prepared coatings was conducted by Xray diffraction analysis (XRD, Philips PW 1830) using monochromatic CuKα radiation with a step size of 0.02°. The fluorine concentrations were determined by X-ray Photoelectron Spectroscopy (XPS, Kratos-Axis Ultra System) using monochromatic Al Kα X-ray source (1486.7 eV). The surface morphology of FHA coatings was characterized using scanning electron microscopy (SEM, LEICA S360). 2.2. In vitro tests in standard SBF and modified SBF solutions Three kinds of solutions were used in the immersion tests: (1) the standard SBF solution, (2) glucose modified SBF

solution (G-SBF) and (3) bovine serum albumin modified SBF solution (A-SBF). The standard SBF solution was prepared according to Kokubo's protocol by dissolving appropriate quantities of the relevant reagent-grade chemicals in deionized water [9]: NaCl, NaHCO3, KCl, K2HPO4U3H2O, MgCl2U6H2O, CaCl2, HCl (1 M), Na2SO4 and NH2C(CH2OH)3. After all the reagents were dissolved, the solution was then heated to 37 °C and maintained at this temperature while titrating the solution to a pH of 7.4 with 1 M HCl or NH2C(CH2OH)3. The inorganic ion concentrations in the standard SBF solution are almost the same in human blood plasma Table 1. However, as shown in the table, besides the inorganic components, there are many organic compounds in human blood plasma, which are not included in the Kukobo's SBF solution. Bovine serum albumin (BSA, Merck) and glucose (D(+)-Glucose anhydrous, Merck) were selected respectively for the preparation of protein and carbohydrate-containing SBF solutions. The glucose modified SBF solution (G-SBF) and BSA modified SBF solution (ASBF) were obtained by dissolving 1 g/L for glucose and 40 g/L for BSA respectively in the prepared SBF solution. The coatings were placed in sterilized bottle containing solution with a liquid/area ratio of 50 ml/cm2. Before soaking in the solution, the samples were washed ultrasonically in acetone for 10 min, and then sterilized in ethanol. The soaking took place in a temperature-controlled shaking water bath for various periods of 2, 4, 7, 14, 21 and 28 days at 37 °C ± 0.1 °C. After the desired immersion, the samples were taken out, gently washed

Fig. 4. SEM micrographs of the fluoridated HA coating after the in vitro tests in SBF and G-SBF: a) 2 days in SBF, b) 7 days in SBF, c) 2 days in G-SBF, d) 7 day in G-SBF.

Y. Wang et al. / Materials Science and Engineering C 27 (2007) 244–250

with deionized water and dried at room temperature before the effect characterization. Scanning electron microscopy (SEM, LEICA S360) was used to examine the chemical and surface morphology changes that occurred after the immersion in simulated solutions. The Ca2+ ion concentration of the solutions were measured with inductively coupled plasma atomic emission spectrometer (ICPAES, PerkinElmer Optima 2000). An average of three measurements was taken for each sample. 3. Results and discussion 3.1. Phase, composition and surface morphologies of the prepared coatings The XRD profiles of the Fluoridated HA coating and the HA coating are shown in Fig 1. The coatings have similar diffraction profile. The peaks (002), (211), (112), (300) are those of the HA structure (JCPDS file card #9-432). No tricalcium phosphate (TCP), CaO, CaF2 or other phases are detected in these coatings. The concentration of Ca, F, and P in the coating is determined by ratio of the area under the respective elemental peak in the XPS narrow scan spectrum [7]. As such, the Ca/P molar ratio in the coating is 1.65 in HA and 1.63 in Fluoridated HA, which is very close to the stoichiometric value of 1.67. Fig 2 shows the XPS spectra of fluoridated HA. F1s peak is evident in the wide scan. The narrow scan analysis around 684 eV (the inset) reveals only one peak at ∼ 684.3 eV belonging to F1s. That peak is the fingerprint for fluorine in FHA structures [22]. This indicates that the fluorine ions have been successfully incorporated into the coating. Fig 3 shows the surface morphologies of the sol-gel derived HA and fluoridated HA coatings. Two observations are noticed: (1). both coatings have complete coverage of the substrate; the coatings look uniform and generally smooth. (2). the fluoridated coating appears rougher (Fig 3b) since the Fluorine agent (HPF6) promotes gelation in FHA deposition process [23].


observed after 2 days in G-SBF, but after 7 days the new apatite seems to be smoother and without the apatite granules (Fig 4d). These results indicate that glucose does not seem to have a significant influence on the nucleation and growth of apatite on fluoridated HA coating. This is also observed on pure HA coating [13,14]. 3.3. Effect of albumin After soaking in the protein-containing SBF or A-SBF for 2 days (Fig 5a), there is only sporadic nucleation of new apatite on the Fluoridated HA surface. Compare Fig. 5a with Fig 4a, one can easily see that albumin addition drastically slowed down nucleation of apatite from SBF. This retardation effect is likely carried to the growth stage of the newly nucleated apatite crystals. Only after 28 days can the new apatite layer becomes continuous (Fig 5b, the cracks are drying artifacts), which only needs 7 days in standard SBF (Fig 4b) or glucose modified SBF. The newly deposited apatite layer is usually considered gel-like. After drying, the gel-like structure shrinks and thus cracks form [24]. Higher magnification (inset in Fig 5b) also reveals porosity in the newly precipitated apatite layer. Fig 6 compares the bioactivity of pure HA coatings in SBF and A-SBF. After 2 days in standard SBF (Fig 6a), obvious apatite nucleation is

3.2. Effect of glucose Fig 4 shows the characteristic surface morphologies on the fluoridated HA coating after in vitro tests in SBF and G-SBF. After 2 days in standard SBF (Fig 4a), a new apatite layer precipitated from the solution completely covers the surface of the FHA coating deposited from the sol-gel process. After 7 days, the new apatite layer becomes smooth and uniform (unlike the rough appearance after 2 days), as is seen in Fig 4b. At low magnification, the surface appears dense and contains spherical apatite particles. At higher magnification (Fig 4b inset), however, porosity can still be seen in the surface. Further increase of soaking time has no influence on the morphology of the newly grown apatite layer except increase in thickness. Fig. 4c and d are the same coating in glucose-modified SBF for 2 days and 7 days respectively. Comparing Fig 4a with Fig 4c and b with Fig 4d, one does not find too much difference except that more pin-holes are

Fig. 5. SEM micrographs of the fluoridated HA coating after soaking in A-SBF for: a) 2 days, b) 28 days.


Y. Wang et al. / Materials Science and Engineering C 27 (2007) 244–250

Fig. 6. SEM micrographs of pure HA coating after soaking for: a) 2 days in SBF, b) 14 days in SBF, c) 2 days in A-SBF, d) 28 days in A-SBF.

observed, and after 14 days, a continuous layer is formed (Fig. 6b). But 2 days in A-SBF (Fig 6c) results in much less new apatite than that in SBF. Even after 28 days, the apatite formation is still sporadic (Fig. 6c). Therefore, protein addition in SBF adversely affect apatite nucleation and growth on both HA and FHA coatings. According to Combes and Rey [25], the slow deposition and growth comes from the fact that the adsorption of albumin reduces the interfacial energy of apatite nuclei with the solution. That deprives the driving force for the growth of the new apatite. 3.4. Effect of fluorine The effect of fluorine inclusion in HA has been extensively studied recently [7], and it has been reported that the degree of fluoridation of x ∼ 1.33 improves biomineralization in HA [21]. Comparing this fluoridated HA (where fluoridation extent is 1.33) with pure HA in standard SBF for 2 days (Fig 4a vs Fig 6a), one sees a drastic increase in nucleation rate as fluorine is incorporated in HA structure. To have a continuous apatite layer formed, it requires 14 days on a pure HA coating (Fig 6b) but only 7 days on fluoridated HA (Fig 4b). This bio-activation effect of fluorine is also effective in the presence of protein as is manifested in the in vitro tests in A-SBF: without fluorine in the coating, 2 days soaking in A-SBF results in almost no apatite nucleation (Fig. 6c), but with fluorine, obvious nucleation is observed (Fig. 5a). In the case of pure HA, 28 days in A-SBF gives rise to only sporadic apatite (Fig. 6d), however, on the fluoridated HA, a continuous thick layer of apatite is formed (Fig. 5b).

3.5. Variation of Ca2+ concentration during in vitro tests The variation of Ca2+ concentration in SBF, G-SBF and ASBF solutions are shown in Fig 7 as a function of soaking time. The ups and downs of the Ca2+ concentration respectively indicate dissolution and precipitation of apatite from the solution to the surface of the coatings. In Fig 7 a), during the first 4 days the pure HA coating experienced more dissolution than reprecipitation that resulted in the increase of Ca2+ concentration in the solution. The Ca2+ released from the coating may result in a supersaturation of Ca2+, a situation more favorable for nucleation [15,26]. Therefore, spontaneous growth of apatite layer takes place which consumes calcium and phosphate ions in SBF causing the gradually decrease of Ca2+ concentration. With the incorporation of F− ion into the HA lattice, no obvious dissolution process is observed. It seems that during the whole test process, the Ca2+ ion concentration maintains a gradual reduction, indicating a continuous precipitation of apatite. This agrees well with SEM morphological observations. Fig 7b describes the Ca2+ concentration in G-SBF. As glucose has no obvious influence on the precipitation process on either HA or FHA coating, it is not surprising that the Ca ion concentration profile does looks very much the same as that in Fig. 7a. Fig 7c shows the Ca2+ concentration in A-SBF after the in vitro test. It is interesting to note that the curves show a sudden drop in the first 2 days for both HA and Fluoridated HA coatings. After that, calcium concentration increases and then slowly decreases. The sudden drop of Ca2+ concentration is resulted from the adsorption of BSA on the coating surface,

Y. Wang et al. / Materials Science and Engineering C 27 (2007) 244–250


other hand, the isoelectric point of BSA is 4.7, therefore, BSA will be negatively charged in the physiological solution with a pH of 7.4, thus tends to bind positive ions like Ca2+ in the solution [28]. This strongly affects the available Ca2+ ions for nucleation and growth of apatite. 4. Conclusions This work has examined the biological response of hydroxyapatite (HA) and a fluoridated HA with a fluoridation degree of 1.33, i.e., Ca10(PO4)6(OH)0.67F1.33, in standard Simulated Body Fluid (SBF) and organic-containing, i.e., glucose and protein-modified, SBF. This work concludes that 1. At concentration of 40 g/L, bovine serum albumin has significant retardation effect on apatite precipitation from SBF onto pure or the fluoridated HA coatings; 2. Fluorine-incorporation in HA has positive bio-activation effect in both standard SBF and organic-containing SBF. 3. Glucose has negligible influence on the bioactivity of HA and the fluoridated HA. Acknowledgements This work is supported by the Agency for Science Technology and Research, Singapore (A⁎Star) through project 032101 0005 and the SIMTech-NTU collaboration project U03-S-389B.


Fig. 7. Post-soaking ICP concentration of Ca2+ in a) standard SBF, b) G-SBF, c) A-SBF solution.

which captures large amount of Ca2+ from the solution. The interactions between BSA and the coatings are complex. Generally, after the coating is immersed into the A-SBF solution, BSA will be immediately adsorbed onto the surface and thus inhibit both the dissolution of the coating into the solution and the precipitation of apatite from the solution on to the coating surface [27]. It has been suggested that the proteins compete with ions, e.g. Ca2+ and PO43− etc., in the solution for the same surface binding sites [24]. Therefore the adsorption of BSA to the surface of coating surface reduces the number of nucleation and growth sites for apatite. On the

[1] M. Sivakumar, Mater. Lett. 50 (2001) 199. [2] R. Roop Kumar, M. Wang, Mater. Lett. 55 (2002) 133. [3] A. Aoki, Medical Applications of Hydroxyapatite, Ishiyaku Euroamerica Press, Toyoko, 1994. [4] M. Cavalli, G. Gnappi, D. Montenero, P. Bersani, P. Lottici, S. Kaciulis, G. Mattogno, M. Fini, J. Mater. Sci. 36 (2001) 3253. [5] H.W. Kim, H.E. Kim, J.C. Knowles, Biomaterials 25 (2004) 3351. [6] W. Weng, S. Zhang, K. Cheng, H. Qu, P. Du, G. Shen, J. Yuan, G. Han, Surf. Coat. Technol. 167 (2003) 292. [7] S. Zhang, X. Zeng, Y. Wang, K. Cheng, W. Weng, Surf. Coat. Tech. 200 (2006) 6350. [8] S. Ha, R. Reber, K. Eckert, M. Petitmermet, J. Mayer, E. Wintermantel, J. Am. Ceram. Soc. 81 (1998) 81. [9] A. Oyane, K. Nakanishi, H. Kim, F. Miyaji, T. Kokubo, N. Soga, T. Nakamura, Biomaterials 20 (1999) 79. [10] W. Jauo, M.B. Hachimi, T. Koutit, J.L. Lacout, M. Ferhat, Mater. Res. Bull. 35 (2000) 1419. [11] E. Balint, P. Szabo, C.F. Marshall, S.M. Sprague, Bone 28 (2001) 21. [12] E.I.F. Pearce, E.M. Hancock, I.H.C. Gallagher, Arch. Oral. Biol. 29 (1984) 521. [13] S.V. Dorozhkin, E.I. Dorozhkina, S. Agathopoulos, J.M.F. Ferreira, Key. Eng. Mater. 254–256 (2004) 327. [14] S.V. Dorozhkin, E.I. Dorozhkina, M. Epple, J. Appl. Biomater. Biomech. 1 (2003) 200. [15] J. Xie, C. Riley, K. Chittur, J. Biomed. Mater. Res. 57 (2001) 357. [16] Q. Luo, J.D. Andrade, J. Colloid Interface Sci. 200 (1998) 104. [17] S.A. Bender, J.D. Bumgardner, M.D. Roach, K. Bessho, J.L. Ong, Biomaterials 21 (2000) 299. [18] C. Combes, C. Rey, Biomateials 23 (2002) 2817. [19] P. Ducheyne, Q. Qiu, Biomaterials 20 (1999) 2287.


Y. Wang et al. / Materials Science and Engineering C 27 (2007) 244–250

[20] C.R. Jenney, J.M. Aderson, J. Biomed. Mater. Res. 49 (2000) 435. [21] K. Cheng, W. Weng, H. Qu, P. Du, G. Shen, G. Han, J. Yang, J.M.F. Ferreira, J. Biomed. Mater. Res. 69B (2003) 33. [22] M.A. Stranick, M.J. Root, Colloids Surf. 55 (1991) 137. [23] K. Cheng, S. Zhang, W. Weng, J. Sol-Gel Sci. Techn. 38 (2006) 13. [24] S. Ha, R. Reber, K. Ecker, M. Petitmermet, J. Mayer, E. Winermantel, J. Am. Ceram. Soc. 81 (1998) 81.

[25] C. Combes, C. Rey, J. Mater. Sci., Mater. Med. 10 (1999) 153. [26] Q. Zhang, J. Chen, J. Feng, Y. Cao, C. Deng, X. Zhang, Biomaterials 24 (2003) 4741. [27] H. Zeng, K.K. Chittur, W.R. Lacefield, Biomaterials 20 (1999) 443. [28] T. Asaoka, T. Ando, T. Meguro, I. Yanato, Chem-Bio Inform. J. 3 (2003) 96.