In vitro evaluation of Ag-containing calcium

Materials Science and Engineering C 75 (2017) 926–933

Contents lists available at ScienceDirect

Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec

In vitro evaluation of Ag-containing calcium phosphates: Effectiveness of Ag-incorporated β-tricalcium phosphate Ozkan Gokcekaya a,⁎, Kyosuke Ueda a, Kouetsu Ogasawara b, Hiroyasu Kanetaka c, Takayuki Narushima a a b c

Department of Materials Processing, Tohoku University, 6-6-02 Aza Aoba, Aramaki, Aoba-ku, Sendai, Miyagi 980-8579, Japan Department of Immunobiology, Institute of Development, Aging and Cancer, Tohoku University, 4-1 Seiryo-machi, Aoba-ku, Sendai, Miyagi 980-8575, Japan Liaison Center for Innovative Dentistry, Graduate School of Dentistry, Tohoku University, 4-1 Seiryo-machi, Aoba-ku, Sendai, Miyagi 980-8575, Japan

a r t i c l e

i n f o

Article history: Received 2 May 2016 Received in revised form 30 November 2016 Accepted 14 February 2017 Available online 16 February 2017 Keywords: Calcium phosphate Silver In vitro dissolution Antibacterial activity Cytotoxicity

a b s t r a c t Development of bioceramics with antibacterial activity and without cytotoxicity would be beneficial for preventing infection associated with implants. This study aimed to capitalize on the antibacterial properties of silver (Ag) incorporated in or coexisting in metallic form with calcium phosphates (CaPs). The in vitro dissolution behavior, antibacterial activity, and cytotoxicity of Ag-containing CaPs with different phase fractions of hydroxyapatite (HA) and β-tricalcium phosphate (β-TCP) were evaluated. The antibacterial activity of Ag-containing CaPs depended on the main phase of CaP, the chemical state of Ag, and the amount of incorporated Ag. Superior antibacterial activity was obtained from sustained release of Ag ions through continuous dissolution of Ag-incorporated β-TCP compared to that obtained for HA coexisting with metallic Ag particles. Ag-containing CaPs did not exhibit any toxic effect on V79 fibroblasts. Thus, these results demonstrated the effectiveness of Ag-incorporated β-TCP in preventing infection, with respect to long-term applications. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Calcium phosphate (CaP) bioceramics have received much attention as bone filling materials because of their excellent biocompatibility, bioactivity, and osteoconduction characteristics [1–6]. Postoperative infections associated with implants are serious problems that can lead to removal of the implants [7–10]. Metal ions such as Ag+, Cu2 +, and Zn2 + are widely considered as antimicrobial agents in CaPs [11–13]. Among these ions, Ag+ is well known for its antibacterial properties against both Gram-positive and Gram-negative bacteria, fungi, protozoa, and certain viruses, including antibiotic-resistant strains [14]. Ag has demonstrated high antimicrobial activity while maintaining relatively low cytotoxicity [9,15,16]. In the human body, there are three main mechanisms for the antibacterial effect of dissolved Ag ions. First, Ag ions bind to electron donor groups containing sulfur, oxygen or nitrogen such as thio, amino, imidazole, carboxylate, and phosphate groups and deactivate them [17]; later, these complexes interact with the membrane of bacteria and finally penetrate the bacterial cell wall. Once they enter bacterial cells, they subsequently generate reactive oxygen species (ROS) and transform DNA molecules into a condensed form without replication ability [18]. These processes lead to damage or even death of the bacteria. Thus, addition of Ag to CaPs has been ⁎ Corresponding author. E-mail addresses: [email protected], [email protected] (O. Gokcekaya).

http://dx.doi.org/10.1016/j.msec.2017.02.059 0928-4931/© 2017 Elsevier B.V. All rights reserved.

investigated in several studies [13,19–21]. However, high concentrations of Ag ions in body fluids can cause toxic effects in some tissues [22]. The National Institute for Occupational Safety and Health (NIOSH) established a Recommended Exposure Limit (REL) of 0.01 mg m−3 for both soluble Ag compounds and Ag metal dust to prevent toxic effect of Ag-related products [23]. The antibacterial activity and cytotoxicity of Ag-containing CaPs strongly depend on the dissolution of Ag ions [24]. Ewald et al. examined the antibacterial activity of Ag-incorporated brushite and HA and reported that Ag-incorporated brushite had high antibacterial activity because of high Ag-ion release (1 μg mL− 1) over 7 days in Luria-Bertani medium [16]. Honda et al. synthesized Ag nanoparticlecontaining HA for bone tissue engineering with antibacterial properties and observed an increase in the amount of Ag ions released in HEPES buffer (approximately 1.5 μg mL−1) with a resulting high level of antibacterial activity [25]. Piccirillo et al. reported that Ag-containing calcium phosphates prepared from cod fish bones that contained HA and β-TCP showed antibacterial activity toward Gram-positive and Gram-negative bacteria [26]. However, no studies have evaluated the relationship between dissolution behavior and antibacterial activity while considering CaP phases with differing bioresorbability, varying Ag content, and altered chemical states of Ag in CaPs. In our previous study, Ag-containing CaP powders were synthesized using a precipitation method; after sintering, Ag-containing CaPs with various phases were obtained with incorporated Ag in β-TCP structure or where the HA coexisted with metallic Ag particles [27]. In this

O. Gokcekaya et al. / Materials Science and Engineering C 75 (2017) 926–933

927

Table 1 Abbreviations, measured atomic ratios, relative densities, and phase fractions of sintered compacts used in this study. Abbreviations

TCP series

HA series

0AgHA/TCP 0.4AgHA/TCP 1AgHA/TCP 2.1AgTCP 0AgHA 1.6AgHA

Measured (Ca + Ag)/P atomic ratio

1.54 1.50 1.55 1.46 1.70 1.71

Measured Ag/(Ca + Ag) atomic ratio

0 0.0043 0.0103 0.0214 0 0.0160

study, we examined the in vitro dissolution behavior, antibacterial activity, and cytotoxicity of sintered CaP compacts with emphasis on the phase of the CaPs and the chemical state of Ag in the CaPs. 2. Experimental 2.1. Preparation of Ag-containing CaPs The CaP powders were synthesized by a precipitation method as described elsewhere [27]. Briefly, calcium nitrate [Ca(NO3)2·4H2O], ammonium phosphate [(NH4) 3 PO 4 ], and silver nitrate (AgNO 3 ) were mixed as an aqueous solution. The pH was adjusted to 11 by addition of aqueous ammonia, and the solution was stirred for 24 h. After precipitation, powders were filtered and dried at 393 K for 24 h. The chemical composition of the powders after precipitation was measured by inductively coupled plasma atomic emission spectrometry (ICP-AES). The abbreviations for the CaP sintered compacts, measured (Ca + Ag)/P and Ag/(Ca + Ag) atomic ratios, and relative densities are listed in Table 1. The 0AgHA/TCP, 0.4AgHA/TCP, 1AgHA/TCP, 2.1AgTCP, 0AgHA, and 1.6AgHA specimens in this study correspond to the 0AgCaP1.33, 2AgCaP1.33, 5AgCaP1.33, 10AgCaP1.33, 0AgCaP1.67, and 10AgCaP1.67 specimens, respectively, in our previous study [27]. The CaP powders (0.5 g) were compressed at 200 MPa to form pellets with a diameter of 10 mm and sintered at 1373 K for 24 h in air with heating and cooling rates of 0.05 K s− 1 to form Ag-containing CaP sintered compacts with various phase fractions of HA and β-TCP. To determine the phase fractions of sintered compacts precisely for this study (Table 1), Rietveld analysis was performed with X'Pert HighScore Plus on X-ray diffraction (XRD) patterns for 2θ from 20° to 80°. The TCP series of sintered compacts exhibited biphasic β-TCP/HA or single β-TCP, in which the β-TCP phase fraction increased with increasing Ag content. No peak related to Ag was detected in the TCP series in the XRD measurement; additionally, a decrease in lattice parameters and homogenous distribution of Ag in the β-TCP phase were observed, which indicated incorporation of Ag [27]. However, the phase of the 0AgHA sintered compact was HA, and that of 1.6AgHA was HA coexisting with metallic Ag particles (ϕ1–3 μm). The sintered compacts were ground with a mortar and pestle to prepare powders for antibacterial tests; hereafter, these powders are referred to as sintered-compact powders.

Relative density (%)

97.6 97.4 90.6 88.3 91.1 96.3

Phase fractions β-TCP

HA

Metallic Ag

CaO

0.48 0.73 0.85 1.00 0.00 0.00

0.52 0.27 0.15 0 0.92 0.90

0 0 0 0 0.00 0.08

0 0 0 0 0.08 0.02

60 rpm and 310 K for up to 30 days. At 1, 3, 5, 7, and 15 days, 10 mL of the supernatant of the Tris-HCl was removed from each bottle to measure the amount of ions, and 10 mL of fresh Tris-HCl was poured into the bottle to maintain a volume of 50 mL for the immersion solution. Next, 1 mL of 5% HNO3 was added to the collected Tris-HCl (10 mL) to maintain the optimum pH for ICP-AES measurement. The amounts of Ca, P-related, and Ag ions in the solutions were determined by ICP-AES. The dissolution test was conducted in triplicate. The surface of the sintered compacts before and after the dissolution tests was analyzed using scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM-EDS).

2.2. In vitro dissolution test Dissolution tests were conducted for the sintered compacts to investigate the in vitro dissolution behaviors of Ag-containing CaPs with different phase fractions of CaPs and chemical states of Ag in Tris(hydroxymethyl)aminomethane (Tris)-buffered solution (Tris-HCl, Tris-hydrochloric acid (HCl)). Tris-HCl was prepared as previously reported [28]. The initial pH of the Tris-HCl was maintained at 7.4 using 1 M HCl. Each sintered compact was placed in a bottle with 50 mL of Tris-HCl. The bottles with sintered compacts were shaken in a water bath at

Fig. 1. Amounts of (a) Ca, (b) P-related, and (c) Ag ions in Tris-HCl after immersion of TCP and HA series sintered compacts.

928


2.3. Antibacterial test The antibacterial activities of sintered-compact powders and sintered compacts were evaluated using the shake-flask method for bacterial solutions. Gram-negative Escherichia coli (E. coli) cells were used for evaluation of antibacterial activity. A bacterial solution with an initial concentration of 109 colony-forming units (CFU)·mL−1 was diluted to 106 and 107 CFU·mL−1 using 1/500 nutrient broth (NB) solution (ISO 27447) [29]. The sintered-compact powders (5 mg) and sintered compacts were autoclaved at 394 K for 21 min in a test tube. Next, 5 mL of bacterial solution was poured into the test tube and incubated at 310 K and 200 rpm for 0.5, 2, 6, and 24 h. The bacterial solution without powders and compacts was used as a control. After shaking in an incubator, 0.5 mL of the supernatant bacterial solution was collected, placed into a microfuge tube, and diluted using 1/500 NB solution. The diluted bacterial solution (100 μL) was then spread on NB agar plates. The NB agar plates were incubated for 24 h at 310 K, and the colonies were then counted. The numbers of colonies were then converted to the concentration of viable bacteria (CFU·mL− 1). The antibacterial test was conducted four times, and the mean values and standard deviations were calculated. Statistical analysis was performed using one-way

ANOVA. Multiple comparisons of the means of independent groups were performed using Tukey tests. Statistical significance was defined as p b 0.05 and p b 0.01. 2.4. Cytotoxicity test Cytotoxicity tests were conducted based on ISO 10993–5 [30]. Chinese hamster lung fibroblasts (V79 cells) were cultured in Eagle's minimal essential medium (Eagle's MEM) supplemented with 10% fetal bovine serum (FBS) in a CO2 incubator (310 K, 5% CO2 atmosphere). The Ag-containing CaP sintered compacts were washed in acetone/MilliQ water and autoclaved at 394 K for 21 min. For direct cytotoxicity tests, five specimens of each sintered compact type were placed in a 24-well plate separately. V79 cells (50 cells mL−1) in 1 mL of MEM 10 medium (Eagle's MEM containing 10% FBS and 100 nM sodium pyruvate) were seeded on sintered compacts. Wells without specimen were used as controls. Cells in the 24-well plates were incubated for 7 days at 310 K under 5% CO2. For indirect cytotoxicity tests, all of five sintered compacts for each type of specimen were immersed in 20 mL of MEM 10 medium without V79 cells and incubated for 24 h at 310 K under 5% CO2. After incubation, 1 mL of extract solution (100% extract) or diluted extract solution

Fig. 2. Surface of sintered compacts of (a–c) 0AgHA/TCP, (d–f) 0.4AgHA/TCP, and (g–i) 2.1AgTCP (a, d, g) before and after immersion for (b, e, h) 15 and (c, f, i) 30 days; EDS spectra of the area (i1) with and (i2) without small white precipitates.


929

and 2.1AgTCP, which had higher β-TCP/(β-TCP + HA) fractions than 0AgHA/TCP and 0.4AgHA/TCP, was significantly lower than that for 0AgHA/TCP and 0.4AgHA/TCP. The amount of Ag ions released from Ag-incorporated 1AgHA/TCP and 2.1AgTCP sintered compacts into Tris-HCl was also much higher than that released from 0.4AgHA/TCP (see Fig. 1(c)), which had a lower amount of Ag incorporation. The release of Ag ions differed according to the chemical state of Ag; 2.1AgTCP (Ag-incorporated β-TCP, Ag/(Ca + Ag) atomic ratio = 0.0214) released a higher amount of Ag ions into Tris-HCl than 1.6AgHA (HA coexisting with metallic Ag particles, Ag/(Ca + Ag) atomic ratio = 0.0160). The amount of Ag ions in Tris-HCl increased and then decreased for all specimens. The surfaces of the TCP series of sintered compacts before and after dissolution tests are shown in Fig. 2. The grain sizes of CaP matrix of 0AgHA/TCP, 0.4AgHA/TCP, and 2.1AgTCP were 2.4 μm, 2.5 μm, and 2.5 μm, respectively, and were identical. For 0AgHA/TCP and 0.4AgHA/TCP sintered compacts, which were found to be biphasic for HA and β-TCP, non-uniform dissolution was observed after 15 days of immersion in Tris-HCl, and plate-like precipitates were detected after 30 days of immersion. The phase of the plate-like precipitates was considered to be HA, similarly to Lin et al., who reported plate-like precipitation of HA in Tris-HCl at pH 7.4 [31]. Additionally, overall dissolution occurred for the single β-TCP 2.1AgTCP sintered compacts, and small white precipitates were observed on their surface after 30 days of immersion. Ag and Cl signals were detected from areas with small white precipitates, whereas no Ag or Cl signals were detected from areas without small white precipitates on performing EDS analysis, as shown in Figs. 2(i1) and (i2), respectively. These results indicate that the small white precipitates are AgCl. The decrease in the amount of Ag ions during immersion occurred because of the precipitation of AgCl as described in Eq. (1). −

Agþ þ Cl ⇄AgClðsÞ Fig. 3. Changes in the concentrations of viable bacteria with shaking time in the control and after exposure to TCP series and 1.6AgHA sintered-compact powders at initial bacterial concentrations of (a) 106 CFU·mL−1 and (b) 107 CFU·mL−1. * and ** represent significant differences at p b 0.05 and p b 0.01, respectively, when compared with the control at the same shaking time.

(10−1, 10−2, 10−3, or 10−4) was added to each well of five separate 24well plates. V79 cells were seeded at 50 cells mL− 1 in all plates, and wells without extract solution were used as controls. The cells were incubated in the extracts for 7 days at 310 K under 5% CO2. After incubation, the cell culture medium was aspirated from each well. Wells were then washed with phosphate-buffered saline (PBS), and the viable cell colonies were fixed with methanol and stained with 5% Giemsa solution. For direct cytotoxicity tests, the cell colonies were counted on compacts and control well plates. For indirect cytotoxicity tests, the cell colonies on plates that contained extract solution or on control plates were counted. Because of the difference in the diameter of 24-well plates (ϕ15 mm) and sintered compacts (ϕ14 mm), the area correlation factor (1.23) was used to determine the number of colonies in the control. Cytotoxicity tests were statistically analyzed using one-way ANOVA and Tukey's multiple comparison tests, as described for the antibacterial tests.

ð1Þ

AgCl is well known for its low solubility product ([Ag+][Cl−]) value of 1.8 × 10−10 mol2 L−2 at room temperature [32]. 3.2. Antibacterial activity of Ag-containing CaPs Changes in the concentrations of viable bacteria with shaking time for the TCP series and 1.6AgHA sintered-compact powders are shown in Fig. 3 for the initial bacterial concentrations of 106 CFU·mL−1 and

3. Results 3.1. In vitro dissolution of Ag-containing CaPs Fig. 1 shows the amounts of Ca, P-related, and Ag ions in Tris-HCl after immersion of sintered compacts. The HA series of sintered compacts, in which HA was the main phase, released lower amounts of Ca and P-related ions than the TCP series, in which β-TCP was contained. The total amount of Ca and P-related ions released from 1AgHA/TCP

Fig. 4. Changes in the concentrations of viable bacteria with shaking time in the control and after exposure to 0AgHA/TCP, 2.1AgTCP, and 1.6AgHA sintered compacts at an initial bacterial concentration of 107 CFU·mL−1. * and ** represent significant differences at p b 0.05 and p b 0.01, respectively, when compared with the control at the same shaking time.

930


107 CFU·mL−1 used to evaluate the effects of the amount of incorporated Ag and the chemical state of Ag on antibacterial activity. When no viable bacteria were observed after antibacterial activity testing, the concentration of viable bacteria was plotted as 10−1 CFU·mL−1 [26]. The 0AgHA/ TCP sintered-compact powders exhibited no decrease in the concentration of viable bacteria, similar to the control. In contrast, the concentration of viable bacteria for the Ag-containing TCP series significantly decreased when compared with the control at shaking times ≥0.5 h. No viable bacteria were observed for 0.4AgHA/TCP, 1AgHA/TCP, and 2.1AgTCP, which contained Ag-incorporated β-TCP, after 6, 2, and 0.5 h of shaking, respectively. The antibacterial activity of the TCP series depended on the amount of incorporated Ag, which was related to the amount of released Ag ions in Tris-HCl (Fig. 1(c)). In the case of 1.6AgHA sintered-compact powders (HA coexisting with metallic Ag particles), the concentration of viable bacteria did not decrease and showed no significant difference at shaking times ≥0.5 h, whereas it significantly decreased at shaking times ≥2 h when compared with the control at an initial bacterial concentration of 106 CFU·mL−1. However, no significant antibacterial activity was observed at an initial bacterial concentration of 107 CFU·mL−1. In contrast,

2.1AgTCP sintered-compact powders (Ag-incorporated β-TCP) exhibited significantly increased antibacterial activity when compared with the control at shaking times ≥0.5 h and killed all viable bacteria at shaking times ≥6 h, even with an initial bacterial concentration of 107 CFU·mL−1. The results for the antibacterial activity of 0AgHA/TCP, 2.1AgTCP, and 1.6AgHA sintered compacts are shown in Fig. 4 at an initial bacterial concentration of 107 CFU·mL− 1 . In the case of 0AgHA/ TCP, the concentration of viable bacteria was almost constant and no significant differences were observed when compared with the control. Although the concentration of viable bacteria for 1.6AgHA slightly decreased with up to 6 h of shaking, no significant difference was observed at 24 h of shaking when compared with the control. In contrast, the concentration of viable bacteria for 2.1AgTCP sintered compacts significantly decreased from 2 h to 24 h of shaking when compared with the control, and no viable bacteria were observed after 24 h of shaking. These results indicate that 2.1AgTCP, which was characterized as Ag-incorporated β-TCP, showed noticeably stronger antibacterial activity than 1.6AgHA, in which HA coexisted with metallic Ag particles.

Fig. 5. Number of viable V79 cell colonies (a) on sintered compacts in direct cytotoxicity tests and (b) with undiluted and diluted extracts from sintered compacts in indirect cytotoxicity tests. *– represents p N 0.05 with respect to 0AgHA/TCP.


931

17.2, 16.6, and 15.2, respectively. The statistical analysis revealed no significant differences in the number of cell colonies between the CaP sintered compacts and controls. Fig. 5(b) shows the numbers of viable V79 cell colonies after incubation with undiluted or diluted extracts from sintered compacts as an indirect measure of cytotoxicity. Similar to the direct cytotoxicity assay, no significant differences were observed between the CaP sintered compacts and controls in the indirect cytotoxicity tests. These results indicate that all specimens evaluated in this study had no significant cytotoxicity.

4. Discussion 4.1. Bioresorbability of Ag-containing CaP sintered compacts

Fig. 6. Changes in the amount of Ag ions and total amounts of Ca and P-related ions in TrisHCl after 30 days of immersion and the β-TCP/(β-TCP + HA) fraction of the TCP series of sintered compacts as a function of the Ag/(Ca + Ag) atomic ratio.

3.3. Cytotoxicity of Ag-containing CaP sintered compacts The numbers of viable V79 cell colonies on 0AgHA/TCP, 2.1AgTCP, and 1.6AgHA sintered compacts in direct cytotoxicity tests are shown in Fig. 5(a). In the control, 16.8 cell colonies were observed. The numbers of the cell colonies for 0AgHA/TCP, 2.1AgTCP, and 1.6AgHA were

For the HA and β-TCP biphasic CaPs, the bioresorbability of CaPs is expected to increase as the β-TCP fraction increases [33,34]. However, 1AgHA/TCP and 2.1AgTCP, which had higher β-TCP fractions and Ag incorporation than 0.4AgHA/TCP, exhibited lower total amounts of Ca + P-related ions in Tris-HCl than 0.4AgHA/TCP as summarized in Fig. 6 for the TCP series of sintered compacts after 30 days of immersion. This behavior indicated that extensive Ag incorporation suppressed the dissolution of the β-TCP in Tris-HCl, which is important because the effect of Ag incorporation on the dissolution behavior of β-TCP has not been previously reported. Matsumoto et al. reported that β-TCP was thermally stabilized by monovalent ion incorporation. In particular, Li, Na, and K ions were incorporated at Ca sites, resulting in a decrease in lattice parameters and crystal stabilization [35]. In our previous study, we found that Ag incorporation decreased the lattice parameters of βTCP and suppressed the β-to-α transformation of TCP [27]. These results indicated that Ag incorporation stabilizes β-TCP. The dissolution behavior of TCP with incorporation of Zn [36] and Mg [37] in simulated body fluids has been reported. Incorporation of Zn and Mg stabilized the βTCP phase and inhibited the dissolution of β-TCP. From these results, it was considered that Ag incorporation stabilized the β-TCP, resulting in decreased bioresorbability. Nevertheless, 1AgHA/TCP and 2.1AgTCP released higher amounts of Ag ions into the Tris-HCl than 0.4AgHA/ TCP because of the higher amounts of Ag incorporated in their β-TCP

Fig. 7. Schematic illustration of the in vitro dissolution and antibacterial activity of 2.1AgTCP (Ag-incorporated β-TCP) and 1.6AgHA (HA coexisting with metallic Ag particles).

932


structure compared to 0.4AgHA/TCP (see Table 1). 2.1AgTCP exhibited a higher amount of Ag ions in Tris-HCl and superior antibacterial activity compared to 1AgHA/TCP up to 24 h (1 day), because of the high amount of Ag incorporated in 2.1AgTCP (β-TCP single phase) even though Ag incorporation stabilized the β-TCP structure and suppressed the dissolution of β-TCP. To the best of our knowledge, no reports have discussed the bioresorbability of Ag-incorporated β-TCP with respect to the incorporation amount of Ag and the β-TCP phase fraction. 4.2. Effectiveness of Ag-incorporated β-TCP Ag-incorporated β-TCP, 2.1AgTCP, released a higher amount of Ag ions and had superior antibacterial activity compared to metallic Ag coexisting with HA, 1.6AgHA, even though both of them had approximately the same Ag/(Ca + Ag) atomic ratios. We previously reported that the solubility of Ag in β-TCP was much higher than that in HA [27], which resulted in complete incorporation (dissolution) of Ag in 2.1AgTCP, in contrast to the presence of metallic Ag in 1.6AgHA. These findings clarify that the high Ag solubility and high bioresorbability of 2.1AgTCP are effective for antibacterial activity resulting from the sustained release of Ag ions through continuous dissolution of Agincorporated β-TCP. A schematic illustration of the proposed in vitro dissolution and antibacterial activities of 2.1AgTCP (Ag-incorporated β-TCP) and 1.6AgHA (HA coexisting with metallic Ag particles) is shown in Fig. 7. In the case of metallic Ag coexisting with HA (1.6AgHA), metallic Ag particles existing on the surface of the compacts were dissolved in the Tris-HCl. Owing to the low resorbability of HA, metallic Ag particles inside the sintered compact made of HA would not contribute to the release of Ag ions. Therefore, low antibacterial activity was observed for 1.6AgHA. In contrast, in the case of 2.1AgTCP, Ag was incorporated in the β-TCP phase, and sustained Ag ion release resulted from continuous dissolution of β-TCP, which has a higher dissolution rate than HA. The long-lasting and early-stage antibacterial activity of Ag-incorporated β-TCP may allow it to replace conventional antibiotics for preventing implant-related infections. One concern for the use of Ag is the possibility of accumulation of Ag ions in body fluids, resulting in toxicity. However, Ag ion levels below 0.2 μg mL−1 in the blood are considered normal [38]. In vitro cytotoxicity of AgNO3 has been reported for L929 mouse fibroblasts and MC3T3E1 mouse osteoblasts at concentrations of 0.72 μg mL−1 and 0.47 μg mL− 1, respectively [39]. The effect of Ag ions on cells is concentration-dependent; a high amount of Ag ions can induce cytotoxicity, whereas small amounts can even have positive effects [34]. In our study, the highest amounts of Ag ions for the 2.1AgTCP and 1.6AgHA sintered compacts were 9.85 μg and 2.07 μg, respectively, in 50 mL of Tris-HCl (0.197 μg mL−1and 0.042 μg mL−1); this is lower than the level needed to induce cytotoxicity [38–40]. Thus, V79 cells are viable on Ag-containing CaP sintered compacts. 5. Conclusion Ag-containing CaPs with various (Ca + Ag)/P and Ag/(Ca + Ag) atomic ratios were synthesized by the precipitation method and sintering at 1373 K. In vitro dissolution tests in Tris-HCl were conducted for Ag-containing CaP sintered compacts. The antibacterial activity and cytotoxicity were evaluated for the sintered-compact powders and sintered compacts. The antibacterial activity of Ag-incorporated β-TCP was higher than that of HA coexisting with metallic Ag particles. Dissolution tests showed higher and more continuous Ag ion release from Ag-incorporated β-TCP than from HA coexisting with metallic Ag particles. Our data also show that the Ag-containing CaP sintered compacts were not cytotoxic. This study has demonstrated the effectiveness of Ag-incorporated β-TCP compared to HA coexisting with metallic Ag particles, which also advances the state of the art for CaPs with enhancement of the antibacterial activity of β-TCP without toxic effects. These

results are promising for the application of Ag-incorporated β-TCP in further in vivo applications.

Acknowledgments This study was financially supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (JSPS) under Contract No. 201500660 and the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan under Contract Nos: 25249094 and 26709049. References [1] K. de Groot, Bioceramics of Calcium Phosphate, J. Clin. Eng. 9 (1984) 52. [2] R.W. Bucholz, A. Carlton, R.E. Holmes, Hydroxyapatite and tricalcium phosphate bone graft substitutes, Orthop. Clin. North Am. 18 (1987) 323–334. [3] L.L. Hench, Bioceramics: from concept to clinic, J. Am. Ceram. Soc. 74 (1991) 1487–1510. [4] L.L. Hench, J. Wilson, An Introduction to Bioceramics, World Scientific, 1993. [5] L.L. Hench, Bioceramics, J. Am. Ceram. Soc. 81 (1998) 1705–1728. [6] T.V. Thamaraiselvi, S. Rajeswari, Biological evaluation of bioceramic materials a review, Trends Biomater. Artif. Organs 18 (2004) 9–17. [7] D. Campoccia, L. Montanaro, C.R. Arciola, The significance of infection related to orthopedic devices and issues of antibiotic resistance, Biomaterials 27 (2006) 2331–2339. [8] D.J. Kilgus, D.J. Howe, A. Strang, Results of periprosthetic hip and knee infections caused by resistant bacteria, Clin. Orthop. Relat. Res. 404 (2002) 116–124. [9] W. Chen, Y. Liu, H.S. Courtney, M. Bettenga, C.M. Agrawal, J.D. Bumgardner, J.L. Ong, In vitro anti-bacterial and biological properties of magnetron co-sputtered silver-containing hydroxyapatite coating, Biomaterials 27 (2006) 5512–5517. [10] G.A. Fielding, M. Roy, A. Bandyopadhyay, S. Bose, Antibacterial and biological characteristics of silver containing and strontium doped plasma sprayed hydroxyapatite coatings, Acta Biomater. 8 (2012) 3144–3152. [11] V. Stanić, S. Dimitrijević, J. Antić-Stanković, M. Mitrić, B. Jokić, I.B. Plećaš, S. Raičević, Synthesis, characterization and antimicrobial activity of copper and zinc-doped hydroxyapatite nanopowders, Appl. Surf. Sci. 256 (2010) 6083–6089. [12] E. Boanini, M. Gazzano, A. Bigi, Ionic substitutions in calcium phosphates synthesized at low temperature, Acta Biomater. 6 (2010) 1882–1894. [13] N. Matsumoto, K. Sato, K. Yoshida, K. Hashimoto, Y. Toda, Preparation and characterization of beta-tricalcium phosphate co-doped with monovalent and divalent antibacterial metal ions, Acta Biomater. 5 (2009) 3157–3164. [14] D.R. Monteiro, L.F. Gorup, A.S. Takamiya, A.C. Ruvollo-Filho, E.R. de Camargo, D.B. Barbosa, The growing importance of materials that prevent microbial adhesion: antimicrobial effect of medical devices containing silver, Int. J. Antimicrob. Agents 34 (2009) 103–110. [15] H.-L. Huang, Y.-Y. Chang, M.-C. Lai, C.-R. Lin, C.-H. Lai, T.-M. Shieh, Antibacterial TaN-Ag coatings on titanium dental implants, Surf. Coat. Technol. 205 (2010) 1636–1641. [16] A. Ewald, D. Hösel, S. Patel, L.M. Grover, J.E. Barralet, U. Gbureck, Silver-doped calcium phosphate cements with antimicrobial activity, Acta Biomater. 7 (2011) 4064–4070. [17] J.M. Schierholz, L.J. Lucas, A. Rump, G. Pulverer, Efficacy of silver-coated medical devices, J. Hosp. Infect. 40 (1998) 257–262. [18] Q.L. Feng, J. Wu, G.Q. Chen, F.Z. Cui, T.N. Kim, J.O. Kim, A mechanistic study of the antibacterial effect of silver ions on Escherichia coli and Staphylococcus aureus, J. Biomed. Mater. Res. 52 (2000) 662–668. [19] P.N. Lim, E.Y. Teo, B. Ho, B.Y. Tay, E.S. Thian, Effect of silver content on the antibacterial and bioactive properties of silver-substituted hydroxyapatite, J. Biomed. Mater. Res. A 101 (2013) 2456–2464. [20] Y. Zhang, Q.-S. Yin, Y. Zhang, H. Xia, F.-Z. Ai, Y.-P. Jiao, X.-Q. Chen, Determination of antibacterial properties and cytocompatibility of silver-loaded coral hydroxyapatite, J. Mater. Sci. Mater. Med. 21 (2010) 2453–2462. [21] N. Rameshbabu, T.S. Sampath Kumar, T.G. Prabhakar, V.S. Sastry, K.V.G.K. Murty, K. Prasad Rao, Antibacterial nanosized silver substituted hydroxyapatite: synthesis and characterization, J. Biomed. Mater. Res. A 80 (2007) 581–591. [22] B. Singh, A.K. Dubey, S. Kumar, N. Saha, B. Basu, R. Gupta, In vitro biocompatibility and antimicrobial activity of wet chemically prepared Ca10 −xAgx(PO4)6(OH)2 (0.0 ≤ x ≤ 0.5) hydroxyapatites, Mater. Sci. Eng. C 31 (2011) 1320–1329. [23] P.L. Drake, K.J. Hazelwood, Exposure-related health effects of silver and silver compounds: a review, Ann. Occup. Hyg. 49 (2005) 575–585. [24] V. Alt, T. Bechert, P. Steinrucke, M. Wagener, P. Seidel, E. Dingeldein, E. Domann, R. Schnettler, An in vitro assessment of the antibacterial properties and cytotoxicity of nanoparticulate silver bone cement, Biomaterials 25 (2004) 4383–4391. [25] M. Honda, Y. Kawanobe, K. Ishii, T. Konishi, M. Mizumoto, N. Kanzawa, M. Matsumoto, M. Aizawa, In vitro and in vivo antimicrobial properties of silvercontaining hydroxyapatite prepared via ultrasonic spray pyrolysis route, Mater. Sci. Eng. C 33 (2013) 5008–5018. [26] C. Piccirillo, R.C. Pullar, D.M. Tobaldi, P.M. Lima Castro, M.M. Estevez Pintado, Silvercontaining calcium phosphate materials of marine origin with antibacterial activity, Ceram. Int. 41 (2015) 10152–10159.

O. Gokcekaya et al. / Materials Science and Engineering C 75 (2017) 926–933 [27] O. Gokcekaya, K. Ueda, T. Narushima, C. Ergun, Synthesis and characterization of Agcontaining calcium phosphates with various Ca/P ratios, Mater. Sci. Eng. C 53 (2015) 111–119. [28] A.C. Tas, The use of physiological solutions or media in calcium phosphate synthesis and processing, Acta Biomater. 10 (2014) 1771–1792. [29] I. Standard, ISO27447 - Fine Ceramics (Advanced Ceramics, Advanced Technical Ceramics) - Test Method for Antibacterial Activity of Semiconducting Photocatalytic Materials, 2010. [30] Iso/En10993-5, Biological Evaluation of Medical Devices - Part 5: Tests for Cytotoxicity: In vitro Methods, 2009. [31] F.H. Lin, C.J. Liao, K.S. Chen, J.S. Sun, C.P. Lin, Petal-like apatite formed on the surface of tricalcium phosphate ceramic after soaking in distilled water, Biomaterials 22 (2001) 2981–2992. [32] S. Zumdahl, Chemical Principles, sixth ed. Houghton Mifflin Company, 222 Boston, USA, 2007. [33] S.-H. Kwon, Y.-K. Jun, S.-H. Hong, H.-E. Kim, Synthesis and dissolution behavior of βTCP and HA/β-TCP composite powders, J. Eur. Ceram. Soc. 23 (2003) 1039–1045.

933

[34] Y. Zhang, J. Ai, D. Wang, Z. Hong, W. Li, Y. Yokogawa, Dissolution properties of different compositions of biphasic calcium phosphate bimodal porous ceramics following immersion in simulated body fluid solution, Ceram. Int. 39 (2013) 6751–6762. [35] N. Matsumoto, K. Yoshida, K. Hashimoto, Y. Toda, Thermal stability of β-tricalcium phosphate doped with monovalent metal ions, Mater. Res. Bull. 44 (2009) 1889–1894. [36] X. Wei, O. Ugurlu, A. Ankit, H.Y. Acar, M. Akinc, Dissolution behavior of Si,Zncodoped tricalcium phosphates, Mater. Sci. Eng. C 29 (2009) 126–135. [37] X. Li, A. Ito, Y. Sogo, X. Wang, R.Z. LeGeros, Solubility of Mg-containing β-tricalcium phosphate at 25 °C, Acta Biomater. 5 (2009) 508–517. [38] A.B.G. Lansdown, Silver in Healthcare: Its Antimicrobial Efficacy and Safety in Use, Royal Society of Chemistry, 2010. [39] A. Yamamoto, R. Honma, M. Sumita, Cytotoxicity evaluation of 43 metal salts using murine fibroblasts and osteoblastic cells, J. Biomed. Mater. Res. 39 (1998) 331–340. [40] K. Kawata, M. Osawa, S. Okabe, In vitro toxicity of silver nanoparticles at noncytotoxic doses to HepG2 human hepatoma cells, Environ. Sci. Technol. 43 (2009) 6046–6051.