Materials Science and Engineering C 58 (2016) 467–477

Contents lists available at ScienceDirect

Materials Science and Engineering C

j ourna l homepage: www.e lsev ie r .com/ locate /msec

Strontium incorporation to optimize the antibacterial and biologicalcharacteristics of silver-substituted hydroxyapatite coating

Zhen Geng a, Zhenduo Cui a, Zhaoyang Li a,b,⁎, Shengli Zhu a,b, Yanqin Liang a, Yunde Liu c, Xue Li c, Xin He c,Xiaoxu Yu c, Renfeng Wang c, Xianjin Yang a,b,⁎a School of Materials Science and Engineering, Tianjin University, Tianjin 300072, Chinab Tianjin Key Laboratory of Composite and Functional Materials, Tianjin 300072, Chinac School of Laboratory Medicine, Tianjin Medical University, Tianjin 300072, China

⁎ Corresponding authors at: School of Materials SciUniversity, 92 Weijin Road, Tianjin 300072, China.

E-mail addresses: zyli@tju.edu.cn (Z. Li), xjyang@tju.ed

http://dx.doi.org/10.1016/j.msec.2015.08.0610928-4931/© 2015 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 24 April 2015Received in revised form 21 July 2015Accepted 27 August 2015Available online 3 September 2015

Keywords:Ion-substituted hydroxyapatiteAntibacterialTitaniumSilverStrontium

Infection in primary total joint prostheses is attracting considerable attention. In this study, silver (Ag)was incor-porated into hydroxyapatite (HA) using a hydrothermalmethod in order to improve its antimicrobial properties.Strontium (Sr) was added as a second binary element to improve the biocompatibility. The substituted HAsamples were fixed on titanium (Ti) substrates by dopamine-assisted immobilization in order to evaluate theirantibacterial and biological properties. The results showed that Ag and Sr were successfully incorporated intoHAwithout affecting their crystallinity. Further, the antibacterial tests showed that all the Ag-substituted sampleshad good anti-bacterial properties against Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus). Despitetheir good antibacterial ability, the Ag-substituted samples showed evidence of cytotoxicity on MG63 cells,characterized by low cell density and poor spreadability. The addition of Sr to the Ag-substituted samples consid-erably reduced the cytotoxicity of Ag. Although the viability of the cells grown on the surfaces of co-substitutedHA was not as high as that of the cells grown on the HA surfaces, it is believed that excellent antibacterialproperties and good biological activity can be achieved by balancing the dosage of Sr and Ag.

© 2015 Elsevier B.V. All rights reserved.

1. Introduction

Hydroxyapatite (HA) coatings are widely used in hip and kneesurgery as an alternative to cemented implants or uncoated press-fitimplants, because of the good bioactivity, biocompatibility, and osteo-conductivity [1–2]. Although the estimated risk is fairly low (0.5%–5%),the postoperative infections are major complications that often causesevere pain and loss of bone tissue; moreover, they may lead to theremoval of implants, and thus, a second operation [3]. To reduce theimplant-derived infections, the localized delivery of antibiotics by theHA coatings has been widely investigated; and at present, polymethylmethacrylate beads loaded with antibiotics such as vancomycin or gen-tamicin are regarded as the gold-standard treatment for the preventionof bone infections [4–6]. Nevertheless, antibiotic resistance is rapidlyemerging.

As an excellent antimicrobial agent, silver (Ag) exhibits potentantibacterial activity against several types of Gramme-positive andGramme-negative bacteria, and its rate of bacterial resistance is mini-mal [7–8]. The mechanisms are that Ag+ ions are able to penetrate the

ence and Engineering, Tianjin

u.cn (X. Yang).

bacterial cell wall and cause the DNA to transform into a condensedform that reacts with thiol group proteins, resulting in cell death, andthat they are able to inhibit the bacterial replication process [9–11].Thus, Ag has been incorporated intoHA to achieve the high antibacterialpotential by the methods of plasma spraying [12], ion-beam-assisteddeposition [13], ball milling [14], electrochemical deposition [15],magnetron sputtering [16] and pulsed laser deposition [17]. However,despite its excellent antibacterial properties, Ag is somewhat toxic tohuman cells when its concentration exceeds a certain threshold [12].Incorporating a secondary chemical or bioactive compound is a promis-ing approach to alleviating the toxicity of Ag andmaintaining its antimi-crobial properties optimally. A recent study yielded a finding that co-substitution of magnesium in Ag–HA can mitigate the negative effectsof Ag and improve the cellular viability [18]. Similarly, zinc (Zn) andAg co-implanted into titanium led to excellent osteogenic and antibac-terial activity [19]. Hence, incorporation of a secondary chemical shouldbe an encouraging approach to optimize the Ag-substituted HA withexcellent antibacterial activity and acceptable biocompatibility in bonetissues.

One of the secondary chemicals suitable for Ag-substituted HA inbone-related implants is strontium (Sr). Sr is a non-essential elementwith bone-targeting properties and it can inhibit bone resorption andincrease bone formation; thus, it is an effective anti-osteoporotic

468 Z. Geng et al. / Materials Science and Engineering C 58 (2016) 467–477

element [20–21]. Additionally, Sr not only enhances preosteoblasticcell proliferation and bone matrix synthesis but also inhibits osteo-clast activity [22–25]. Similar results have been obtained in animalstudies [26–27]. Thus, Sr is a promising secondary chemical for alle-viating the potential negative effects of Ag-substituted HA. Afterincorporation, this new composition may exhibit interesting anduseful properties.

Therefore, to investigate the antibacterial activity and biocompatibil-ity of Sr-optimized Ag-substituted HA, in this study, a series of Ag-and Sr-substituted HA samples were synthesized by a hydrothermalmethod. And the antibacterial activity of HAwas studied with the pres-ence of different concentrations of silver. Further, Sr was incorporatedinto Ag-substituted HA samples to investigate whether the addition ofSr could alleviate the potential negative effects of Ag while maintainingits antimicrobial properties optimally (Scheme 1).

2. Materials and methods

2.1. Sample preparation

2.1.1. Preparation of ion-substituted HAAll the samples were synthesized using a one-step hydrothermal

method. In brief, Ca(NO3)2·4H2O, AgNO3, Sr(NO3)2, and Na3PO4 wereused as the Ca, Ag, Sr, and P sources, respectively. For pure HAnanocrystallite, Ca(NO3)2·4H2O and Na3PO4 with a molar ratio of10:6 were each dissolved in 80 mL of deionized water. Then, theCa(NO3)2·4H2O solution was added dropwise into the Na3PO4 solution.After rigorous agitation for 10 min, the suspension was hydrothermallytreated at 150 °C for 5 h. Then the precipitates werewashedwith deion-ized water and dehydrated absolute ethanol twice. The product wasdried by placing it in an oven and heating it at 50 °C for 24 h for drying.Subsequently, the dried powder was manually ground using a corun-dum mortar. HA nanocrystallite doped with different concentrationsof Ag or Sr was synthesized by replacing Ca(NO3)2·4H2O with AgNO3

or Sr(NO3)2, respectively. Specifically, six samples (Ca10 − xAg2x(PO4)6(OH)2, x=0, 0.05, 0.1, 0.3; Ca9 − ySrAg2y(PO4)6(OH)2, y=0, 0.3; denot-ed by HA, Ag0.05, Ag0.1, Ag0.3, 10Sr, and 10SrAg0.3, respectively),distinguishable by the substituted ion concentration, were synthesized.

2.1.2. Preparation of HA coatings on titaniumDopamine-assisted immobilization of the synthesized samples

onto Ti substrates was carried out as reported previously [28]. In brief,3-hydroxytyramine hydrochloride (dopamine hydrochloride, Sigma-Aldrich) was dissolved in 10 mM Tris buffer (pH 8.5) to 2 mg/mL,while each sample was suspended in the same buffer to 2 mg/mL.Then, the two solutions were mixed (1:1), and 100 μL of the mixturewas dropped onto polished Ti plates (10 mm × 10 mm × 1 mm) for12 h. These modified Ti plates with different ion-substituted HA coat-ings were used for cell and bacterial culture.

Scheme 1. Schematic of the hydrothermal process and imm

2.2. Characterization of ion-substituted HA

Phase analysis of the synthesized powders was conducted usingX-ray powder diffraction (XRD) and X-ray photoelectron spectroscopy(XPS). The XRD patterns were obtained using a Bruker D8 AdvanceX-ray diffractometer equipped with graphite-monochromatized Kαradiation (λ = 1.5418 Å). The diffractometer was operated at 40.0 kVand 30.0 mA at a 2θ range of 10°–90° with a step size of 0.02 and expo-sure of 50 s. XPSwas conducted in a vacuum chamber at a base pressureof ~3.5 × 10−8 Pa with a beam spot size of ~500 μm (Escalab250, VGThermo, England).

The lattice parameters a and c were calculated from peaks (0 0 2)and (2 1 1), respectively, using the standard HCP unit cell plane spacingrelationship [29]:

1=d2 ¼ 4 h2 þ hkþ k2� �

=3a2 þ l2=c2 ð1Þ

where d is the distance between two adjacent planes in the set of Millerindices (h k l).

The degree of crystallinity, corresponding to the fraction of crystal-line phase present in the examined volume, was evaluated by thefollowing equation [30]:

Xc≈ 1− V112=300=I300� � ð2Þ

where I300 is the intensity of the (300) reflection and V112/300 is theintensity of the hollow between the (112) and the (300) reflections,which completely disappears in non-crystalline samples.

In order to examine the functional groups of the obtained powder,infrared spectra of all samples were obtained using a Fourier-transforminfrared spectrometer (FT-IR, Bruker Tensor 27, Germany) in the rangeof 4000–400 cm−1. The powder was ground with KBr in the proportionof 1/150 (by weight) and pressed to a wafer with a diameter of 13 mmusing a hand press.

The powder morphology was observed via scanning and transmis-sion electron microscopy (SEM, TEM). An accelerating voltage of 5 kVwas selected for SEM analysis, and the micrographs were capturedusing secondary electrons collected with an in-lens detector. In orderto investigate the distribution of dopant within the HA coating, energydispersive X-ray spectrometer (EDS) analysis and elemental mappingwere performed with a SEM equipped with an EDS analyser, workedat 15 kV. TEM images were acquired using a JEM-2100F microscope at200 kV. In addition, high-resolution TEM (HRTEM) and selected-areaelectron diffraction (SAED) patterns were obtained.

2.3. Ag+ and Sr2+ release

To examine the release behaviour of Ag+ and Sr2+ from theaforementioned Ti plates, each specimen was immersed in 10 mL ofphosphate buffered saline (PBS) at 37 °C. The solution was refreshed

obilization of Sr–Ag–HA coatings on titanium substrate.

469Z. Geng et al. / Materials Science and Engineering C 58 (2016) 467–477

at each time point for 7 days. The PBS containing the released Ag+ andSr2+ was analysed by inductively coupled plasma mass spectroscopy(ICP-MS, VISTA-MPX).

2.4. Antibacterial activity

The antibacterial activity of the samples, tested using Escherichiacoli and Staphylococcus aureus, was evaluated using the agar disc dif-fusion method. The bacterial stock solution was prepared via over-night growth of E. coli and S. aureus lysogeny broth (LB) at roomtemperature with constant stirring. The bacterial cell density wasmeasured using optical methods at 600 nm with concentrations of105, 107, and 109 cells/mL. Further, 50 μL of each bacterium solutionwas inoculated onto an agar plate, and then, a wafer of each sample(diameter, 10 mm; cold pressing, 600 MPa) was placed onto theseeded agar plate. After incubation for 24 h at 37 °C, the diametersof the inhibition zones were measured and optical images were cap-tured using an ordinary camera.

Changes in the bacterial morphology were investigated via SEM.E. coli and S. aureus cells were dispersed in lysogeny broth with afinal concentration of 105 cells/mL, and then, 50 μL of the dispersionswas seeded onto the aforementioned Ti plates. After incubation inambient air (37 ± 1 °C) for 24 h, each plate was washed with PBSbuffer thrice to remove non-adherent bacteria and medium residue.Subsequently, the plate was fixed in 2.5% glutaraldehyde for 2 h.After fixation, the plate was rinsed with PBS buffer twice and sub-jected to dehydration by replacing the buffer with a graded seriesof ethanol (30%, 50%, 60%, 70%, 80%, 90%, and 100%, for 10 mineach). The Ti plate was critical-point dried and coated with goldsputter for conductivity. Then, the sample was examined via fieldemission SEM.

Fig. 1. (a) XRD patterns of the as-synthesized samples. (b) XPS spectra of the as-synthesized(d) high-resolution Sr3d and Sr3p spectra of Ag0.3 and 10SrAg0.3.

2.5. Evaluation of biocompatibility

2.5.1. Cell seeding and culture conditionsAll the samples (the aforementioned Ti plates) used in the cell

culture experiments were first sterilized in an autoclave (at 121 °Cfor 1 h) and then inserted into 12-well polystyrene cell culture plates(TPP, Switzerland; internal well diameter 22.0 mm). Then they wereseeded with human osteoblast-like MG 63 cells (European Collection ofCell Cultures, Salisbury, UK) and suspended inDulbecco'smodified Eagle'sminimum essential medium (DMEM; Sigma, USA, Cat. No. D5648) with10% foetal bovine serum (FBS; Sebak GmbH, Aidenbach, Germany) andgentamicin (40 μg/mL, LEK, Ljubljana, Slovenia). Each well contained36,000 cells (i.e., approximately 10,000 cells/cm2) and 2 mL of the medi-um. The cellswere cultured for 1, 3, and 7 days at 37 °C in a humidified airatmosphere containing 5% CO2.

2.5.2. Evaluation of cell number and viabilityAfter seeding for 1 and 3 days, the samples were rinsed with

phosphate-buffered saline (PBS; Sigma, USA) and stained with Hoechst#33342, which stains the cell nuclei (excitation max. 346 nm, emissionmax. 460 nm; Sigma, USA; 5 μg/mL of PBS). This dye was applied for 2 hat room temperature. The number of cells was evaluated using micro-photographs acquired by an IX-51 microscope equipped with a DP-70digital camera (Olympus, Japan).

On day 7 after seeding,when the cells grew inmultilayers, theyweredetached using trypsin–EDTA solution (Sigma, U.S.A., Cat. No. T4174) inPBS for 10 min at room temperature, and the number of cells was eval-uatedusing a Vi-CELLXRanalyser (BeckmanCoulter, USA). Formorpho-logical analysis, the samples were rinsed with phosphate-bufferedsaline (PBS; Sigma, USA) and stained with 500 μL of acridine orangesolution (100 μg/mL acridine orange in PBS); then, they were visualized

samples. (c) High-resolution Ag3d spectra of Ag0.05, Ag0.1, Ag0.3, and 10SrAg0.3; and

Table 1Lattice parameters and crystallinity of the prepared samples.

Designed samples Crystallinity 2θ values of (300)Lattice parameters

a axis (Å) c axis (Å)

HA 56.5% 33.055 9.4035 6.8692Ag0.05 53.9% 32.870 9.4211 6.8875Ag0.1 58.5% 32.849 9.4287 6.8943Ag0.3 59.8% 32.849 9.4287 6.894310Sr 55.5% 32.788 9.4522 6.910310SrAg0.3 52.1% 32.685 9.4758 6.9283

Fig. 2. FT-IR spectra of all synthesized samples.

470 Z. Geng et al. / Materials Science and Engineering C 58 (2016) 467–477

under blue-green fluorescence using an Olympus microscope system(Japan).

2.5.3. Cell–surface interactions (cell morphology)The surface of the cell-adhered experimental samples was observed

via SEM. For SEM observation, after 1, 3, and 7 days of incubation, thecells were gently rinsed with PBS thrice to remove unattached cellsand then fixed in 2.5% glutaraldehyde for 60 min at room temperature.After dehydration in a graded series of alcohol (50%, 60%, 70%, 80%, 90%

Fig. 3. SEM morphology of synthesized samples. (a) HA, (b) Ag0.05, (c

and 100%) for 10min each and drying in hexamethyldisilazane (HMDS)solution, the samples were sputter-coated with gold.

3. Results and discussion

3.1. Characterization of synthesized ion-substituted HA

The typical XRDpatterns of all the synthesized samples are shown inFig. 1(a). It is evident that the diffraction peaks shifted to lower 2θvalueswith the addition of Ag andSr, indicating an increase in the latticeparameters, which can be attributed to the higher ionic radius of Sr(1.13 Å) and Ag (1.15 Å), as compared to Ca (0.99 Å). Many previousstudies have shown that the incorporation of ions, such as Mg2+ andZn2+, which are smaller than the ionic radius of Ca2+, will decreasethe lattice parameters [31–32]. In contrast, bigger ions, like Sr2+ andBa2+, will increase the lattice parameters [33–34]. The XRD patternsalso indicate that the Ag and Sr are incorporated into HA, rather thanjust a surface modification. Moreover, no changes were observed inthe peak positions with further addition of Ag, indicating that the max-imumamount of Ag substitutionwas less than 0.1mol% inHA. Thisfind-ing is consistentwith the previous study [35]. In addition, the diffractionpeaks of 10SrAg0.5 had the lowest 2θ values, suggesting that Sr and Agcould successfully co-substitute for Ca in HA. Fig. 1(b)–(d) shows theXPS spectra of the as-prepared samples. The high-resolution Ag spectra(Fig. 1(c)) indicate that the intensities of all the Ag peaks increasedconsiderably with the Ag content. It is worth noting that the intensitiesof both Ag and Sr peaks declined significantly when Ag and Sr co-substituted for HA (Fig. 1(c)–(d)), indicating an interplay of effects forAg and Sr (10SrAg0.3).

The cell parameters and crystallinity of the samples are summarizedin Table 1. The lattice parameters a and c evidently increased with theaddition of Sr andAg. Further, 10SrAg0.5 had the largest lattice constant,indicating that Sr and Ag could successfully co-substitute for Ca in HA.The addition of Ag and Sr did not affect the crystallinity of HA.

Fig. 2 shows the FT-IR spectra of the prepared samples. The ab-sorption bands at 1000–1100(υ3), 963(υ1), and 577–603(υ4) detect-ed in the spectrum, which were attributed to the phosphate (PO4

3−)characteristic absorption, were present in all the spectra for the syn-thesized samples. The absorption band at 1640 cm−1 was attributedto hydrogen phosphate, HPO4

2−. The absorption band at 1390 cm−1,whichwas derived from the vibration of theCO3

2− group,was attributedto CO2 from the air. The wide peak observed at 3300–3600 cm−1 was

) Ag0.1, (d) Ag0.3, (e) 10Sr and (f) 10SrAg0.3. Scale bars, 500 nm.

471Z. Geng et al. / Materials Science and Engineering C 58 (2016) 467–477

attributed to the water adsorbed at the surface of the synthesizedpowders and the characteristic stretching vibrationalmode of the struc-tural –OH groups [36]. The FTIR spectra for different concentrations ofAg or Sr were similar to those of HA. No significant changes in theshapes and intensities of the peaks were observed for any of the sam-ples (the carbonate absorption intensity was slightly weakened withthe addition of Ag).

Fig. 3 shows the SEM morphology of the synthesized samples. Allthe samples had nanoparticle morphology and did not differ from oneanother significantly.

TEM, SAED, and HRTEM patterns were obtained to observe thesamples in greater detail, as shown in Fig. 4. Pure HA had regular nano-particle morphology with good crystallinity (Fig. 4(a)). The hexagonalstructure was not destroyed by the addition of Ag (Fig. 4(b–d)), butit was affected by the addition of Sr (slightly irregular, Fig. 4(e–f)).The SAED patterns, which were in agreement with the results of XRD,indicated that all samples had good crystallinity. The research of Predoi

Fig. 4. TEM, SAED and HRTEM patterns of each sample. (a) HA, (b) Ag0.05

et al. showed that the Ag-doped HA could still have good crystallinitywith Ag content of 2 mol% [35]. It is worth noting that HRTEMshowed many Ag nanoparticles attached to the surface of all theAg-substituted HA samples. Unsurprisingly, the size of these Agnanoparticles increased with the Ag content and attained a maxi-mum value of approximately 10 nm for 10SrAg0.3 (Fig. 4). The XRDresults had shown that Ag+ were partially incorporated into HA.Thus, the non-incorporated Ag+ would be reduced to Ag nanoparti-cles and adsorbed on the surface of HA under high temperature andhigh pressure. The size of these Ag nanoparticles naturally increasedwith the addition of Ag.

The results ofmeasurements of elemental composition and distribu-tion of dopant within the HA coating by EDS are presented in Fig. 5. Theelementalmaps showhomogenous distribution of elements in the coat-ings. However, the dot densities of Ca, P, O, and C are higher than thoseof other elementswhichwere due to the high concentration and atomicnumber of these elements [37].

, (c) Ag0.1, (d) Ag0.3, (e) 10Sr and (f) 10SrAg0.3. Scale bars, 100 nm.

Fig. 5. Elemental composition and distribution of dopant within the HA coating. (a) HA, (b) Ag0.05, (c) Ag0.1, (d) Ag0.3, (e) 10Sr and (f) 10SrAg0.3.

Fig. 6. Cumulative Ag+ and Sr2+ ion release as a function of time.

472 Z. Geng et al. / Materials Science and Engineering C 58 (2016) 467–477

473Z. Geng et al. / Materials Science and Engineering C 58 (2016) 467–477

3.2. Ion release property

Fig. 6 shows the cumulative Ag+ and Sr2+ ion release over thecourse of 7 days. All Ag-substituted samples showed the same Ag+ ionconcentration over the course of the first 6 h. However, the Ag+ ionconcentration increased with the Ag content over time. In addition,the Ag+ release rate of Ag0.3 was higher than that of 10SrAg0.3 (albeit,insignificantly), indicating that the addition of Sr had a slight effect onthe degradation of the sample. For Sr2+, there was no significant differ-ence between 10SrAg0.3 and Ag0.3 at any time point. Fielding et al. [38]prepared SrO–Ag2O–HA coatings on Ti by plasma spraying method.Their results showed that the addition of Sr to the Ag doped coatingdid not have a significant effect on the degradation kinetics of the coat-ing, which is consistent with the results of this experiment.

3.3. Antifungal activity

The results of antifungal activities of all samples were shown inFig. 7. After incubation for 24 h at 37 °C, both in E. coli and S. aureus,no inhibition zone could be found under contact with pure HA and10Sr. The inhibition zones in the Ag-substituted samples were clearlyvisible and varied with the concentration of bacteria. It is worth notingthat the antibacterial ability did not increase linearly with the additionof Ag. Instead, it seemed that the antibacterial effect of Ag0.05 andAg0.1 was better than that of Ag0.3. Previous studies have shown thatthe bactericidal effect of Ag nanoparticles was shape-, size- and dose-

Fig. 7. Photograph of the inhibition zone of each sample to E. coli an

dependent [39–41]. There is a general consensus that the antibacterialability is determined by total Ag+ release. The antibacterial abilityincreases with the increase of total Ag+ release. In this experiment, allthe Ag-substituted samples showed the same Ag+ ion concentrationover the course of the first 6 h, which indicated that the Ag+ releasingrate increased with the decreasing of Ag nanoparticle size (Fig. 6).Recently, Mukherji et al. prepared Ag nanoparticles of various sizesand found that for Ag nanoparticles less than 10 nm in size, the antibac-terial efficacy was enhanced significantly, which could be attributed tothe increase in the ratio of surface area to volume [42]. These resultsand findings serve as a guideline for resistance against bacteria, i.e., aneffective way to decrease Ag content without affecting its antibacterialactivity is to reduce the Ag nanoparticle size. Authors studying themechanism of the antibacterial action of Ag+ ions have suggested thatthe thickness of the peptidoglycan layer of Gramme-positive bacteriamay protect the bacterial cells from the influx of Ag+ ions [8,43]. There-fore, the diameters of the inhibition zones of S. aureuswere smaller thanthose of E. coli under the same Ag concentration.

SEMmicroscopy was adopted to evaluate themorphological chang-es in E. coli and S. aureus treated with the prepared samples (Fig. 8).After treatment with pure HA and 10Sr for 24 h, E. coli grew well withhigher density, and its flagella could be observed easily (Fig. 8(a), (g),(e), and (k)). However, after contact with the Ag-substituted samples,the bacterial morphology changed significantly; major damage wasobserved in the E. coli cytoplasmicmembrane (Fig. 8(h)–(j) and (l)). Fur-ther, Ag inhibited the growth of E. coli (low density, Fig. 8(b)–(d) and

d S. aureus by the disc diffusion assay. Disc diameter: 10.00 mm.

Fig. 8. SEM images of E. coli and S. aureus incubated for 24 h on each sample. Scale bars: (a–f) and (m–r), 2 μm; (g–l) and (s–x), 400 nm.

474 Z. Geng et al. / Materials Science and Engineering C 58 (2016) 467–477

(f)). Similar phenomena were observed in the case of S. aureus. S. aureusgrew well with higher density on samples without Ag (Fig. 8(m), (s),(q), and (w)). However, for the Ag-substituted samples, the growthof S. aureus was inhibited and distinct concave could be observed onthe bacterial membranes, as shown in Fig. 8((n)–(p), (r), (t)–(v) and(x)). Numerous studies have shown that the bacterial membrane couldshrink and separate from the bacterial wall when Ag interacts with thebacteria [8,44].

3.4. Cell attachment, morphology, proliferation, and apoptosis on testedsamples

SEM micrographs illustrating the morphology of the attachedcells on each sample are shown in Fig. 9. The images clearly revealthe differences in cell density and morphology on different surfacesat different time points. On day 1 of cell culture, the cells attachedand started spreading very well on both HA and 10Sr (Fig. 9(a) and(e)). MG cells exhibited a typical osteoblast phenotype, which ap-peared flattened and three-dimensional, with many lamellipodia

and filopodia extensions (Fig. 9(a1) and (e1)). Filopodia enable thecells to make contact with the substrate and neighbouring cells. Incontrast to the cells cultured on HA and 10Sr, those cultured on theAg-substituted samples were less spread and exhibited an elongatedmorphology with only a few filopodia (Fig. 9(b)–(d)). On day 3 ofcell culture, the surface coverage was greater on both samples thanthat on day 1 (Fig. 9(g)–(l)). Cells on HA and 10Sr started to grow toconfluence (Fig. 9(g1)–(k1)). Cells on the Ag-substituted samplesstarted to appear flattened, but some cells continued to exhibit an elon-gated morphology, indicating a weak interaction between the cells andthe samples. Previous studies have shown that 6 wt.% Ag significantlyinhibits the growth of osteoblasts and leads to the death of somecells [38]. After 7 days of culture, the cells had undergone significantspreading and proliferation. A greater number of cells could be observedon all the samples (Fig. 9(m)–(r)). The surfaces of HA and 10Sr werecompletely covered by a dense and confluent cellular multilayer(Fig. 9(m) and (q)). The density of the MG63 cells decreased with Agaddition (Fig. 9(n)–(p)). Further, the cells were more spread on10SrAg0.3 than on Ag0.3 (Fig. 9(p) and (r)).

Fig. 9. SEM images of MG63 cells after culturing for 1, 3 and 7 days on each sample. Pictures with highermagnification are taken from the area enclosed by a square in pictureswith lowermagnification.

475Z. Geng et al. / Materials Science and Engineering C 58 (2016) 467–477

To analyse the cell growth state on each sample, representativefluorescence microscopy images of the MG63 cells cultured for 3 and7 days were observed, as shown in Fig. 10. On day 3 of cell culture, thecells on HA and 10Sr grew well and spread uniformly, whereas thecells on the Ag-substituted samples spread unevenly with low cell den-sity. Moreover, certain cells were in the stage of apoptosis (highlightedbywhite arrows). After 7 days of culture, the cell density increased signif-icantly and the cells grew in multilayers on HA and 10Sr (Fig. 10(g) and(k)). All the Ag-substituted samples showed lower cell density than HAand 10Sr. Further, Ag0.3 had the lowest cell density. All the observationswere consistent with the SEM observations.

Cell proliferation on the samples after 1, 3, and 7 days of incubationis shown in Fig. 11. At each time interval adopted in this study, the num-ber of cells on HA and 10Sr was considerably greater than that on theother samples. The addition of Ag had a significant negative impact onthe cell proliferation. Moreover, the higher the Ag content, the greaterwas the toxicity. Further, there was no obvious difference betweenAg0.3 and 10SrAg0.3 over the course of the first 3 days. However, after7 days of culture, the number of cells on 10SrAg0.3 was significantlygreater than that on Ag0.3.

These results indicate that Ag has a cytotoxic effect on the cells andthat Sr is able to offset this effect. A recent study has suggested thatthe interaction of Ag nanoparticles with the cell surface induces damagebrought about by the alteration of the intracellular pH, caused by theblockage of the cell membrane [45]. Other independent studies have

shown that Sr could directly interact with the Ca sensing receptorto enter into the osteoblast cells and trigger mitogenic signals [46].Fielding et al. have suggested that Ag+ ions released from the Ag–HAsamples bind to ALP high affinity metal ion sites, causing functionaldestabilization, whereas in the binary system, the same sites show agreater preference for Sr2+ [38]. Thus, Sr could compete with Ag forbinding sites specific to cellular function. Although in this study thecells grown on the surface of 10SrAg0.3 were not as viable as thosegrown on HA, we believe that it is possible to create samples with thecorrect dosage, enabling them to exhibit nontoxic properties wherebythe cells are able to grow, while bacteria growth is inhibited.

4. Conclusion

Hydrothermal method can successfully synthesize Ag- and Sr- co-substituted HA and the dopamine-assisted immobilization is a simpleway to fabricate HA coating on Ti substrate. The dopamine-assistedimmobilization retains the phase purity and crystallinity of HA and itis suitable for surface modification of bone scaffolds those used innon-bearing sites. Incorporation of Ag in the HA coating shows signifi-cant antibacterial efficacy and demonstrates sustainable release of Agions, but the cell viability decreases sharply because of the cytotoxicityof Ag. As a binary element, Sr compensates the cytotoxicity of Ag onosteoblast, however, the biocompatibility of Ag- and Sr- co-substituted

Fig. 10.Representativefluorescencemicroscopy images of theMG63 cells cultured for 3 (a–f) and 7 (g–l) days. (a andg)HA, (b andh)Ag0.05, (c and i)Ag0.1, (d and j)Ag0.3, (e and k) 10Srand (f and l) 10SrAg0.3. Scale bars, 200 μm.

Fig. 11. MG63 cell proliferation measured by counting cells under a fluorescence micro-scope after 1, 3 and 7 days of incubation. ⁎p b 0.05 compared with the HA and 10Sr,#p b 0.05 compared with the Ag0.3.

476 Z. Geng et al. / Materials Science and Engineering C 58 (2016) 467–477

HA should be further improved by optimizing the substitution degreesof both Ag and Sr.

Acknowledgements

This work was financially supported by the National NaturalScience Foundation of China (Grant No. 31200715), Research Fundfor the Doctoral Program of Higher Education of China (Grant No.20120032120022), and Tianjin Research Program of Application Foun-dation and Advanced Technology (Grant No. 13JCZDJC33300).

References

[1] S. Ruchholtz, G. Tager, D. Nast-Kolb, The periprosthetic total hip infection,Unfallchirurg 107 (2004) 307–317.

[2] D.P. Lew, F.A. Waldvogel, Osteomyelitis, Lancet 364 (2004) 369–379.[3] 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.

[4] S.B. Kim, Y.J. Kim, T.L. Yoon, S.A. Park, I.H. Cho, E.J. Kim, The characteristics of ahydroxyapatite–chitosan–PMMA bone cement, Biomaterials 25 (2004) 5715–5723.

477Z. Geng et al. / Materials Science and Engineering C 58 (2016) 467–477

[5] K. Shinsako, Y. Okui, Y. Matsuda, J. Kunimasa, M. Otsuka, Effects of bead size andpolymerization in PMMA bone cement on vancomycin release, Biomed. Mater.Eng. 18 (2008) 377–385.

[6] D. Neut, H. van de Belt, I. Stokroos, J.R. van Horn, H.C. van der Mei, H.J. Busscher,Biomaterial-associated infection of gentamicin-loaded PMMA beads in orthopaedicrevision surgery, J. Antimicrob. Chemother. 47 (2001) 885–891.

[7] R.K. Kunkalekar, M.S. Prabhu, M.M. Naik, A.V. Salker, Silver-doped manganesedioxide and trioxide nanoparticles inhibit both Gram positive and Gram negativepathogenic bacteria, Colloids Surf. B: Biointerfaces 113 (2014) 429–434.

[8] Q.L. Feng, J. Wu, G.Q. Chen, F.Z. Cui, T.N. Kim, J.O. Kim, A mechanistic study of theantibacterial effect of silver ions on Escherichia coli and Staphylococcus aureus,J. Biomed. Mater. Res. 52 (2000) 662–668.

[9] A. Kazachenko, A. Legler, O. Per'yanova, Y.A. Vstavskaya, Synthesis and antimicrobialactivity of silver complexes with histidine and tryptophan, Pharm. Chem. J. 34(2000) 257–258.

[10] M. Rai, A. Yadav, A. Gade, Silver nanoparticles as a new generation of antimicrobials,Biotechnol. Adv. 27 (2009) 76–83.

[11] P. DeVasConCellos, S. Bose, H. Beyenal, A. Bandyopadhyay, L.G. Zirkle, Antimicrobialparticulate silver coatings on stainless steel implants for fracture management,Mater. Sci. Eng. C Mater. 32 (2012) 1112–1120.

[12] M. Roy, G.A. Fielding, H. Beyenal, A. Bandyopadhyay, S. Bose, Mechanical, in vitroantimicrobial, and biological properties of plasma-sprayed silver-doped hydroxyap-atite coating, ACS Appl. Mater. Interfaces 4 (2012) 1341–1349.

[13] X.A. Bai, K. More, C.M. Rouleau, A. Rabiei, Functionally graded hydroxyapatite coat-ings doped with antibacterial components, Acta Biomater. 6 (2010) 2264–2273.

[14] E. Andrea, H. Daniel, P. Sarika, L.M. Grover, J.E. Barralet, G. Uwe, Silver-dopedcalcium phosphate cements with antimicrobial activity, Acta Biomater. 7 (2011)4064–4070.

[15] C.M. Xie, X. Lu, K.F. Wang, F.Z. Meng, O. Jiang, H.P. Zhang, W. Zhi, L.M. Fang, Silvernanoparticles and growth factors incorporated hydroxyapatite coatings on metallicimplant surfaces for enhancement of osteoinductivity and antibacterial properties,ACS Appl. Mater. Interfaces 6 (2014) 8580–8589.

[16] 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.

[17] M. Jelinek, T. Kocourek, J. Remsa, M. Weiserová, K. Jurek, J. Mikšovský, J. Strnad, A.Galandáková, J. Ulrichová, Antibacterial, cytotoxicity and physical properties oflaser — silver doped hydroxyapatite layers, Mater. Sci. Eng. C Mater. 33 (2013)1242–1246.

[18] D. Gopi, E. Shinyjoy, L. Kavitha, Synthesis and spectral characterization of silver/magnesium co-substituted hydroxyapatite for biomedical applications, Spectrochim.Acta A 127 (2014) 286–291.

[19] G.D. Jin, H. Qin, H.L. Cao, S. Qian, Y.C. Zhao, X.C. Peng, X.L. Zhang, X.Y. Liu, P.K. Chu,Synergistic effects of dual Zn/Ag ion implantation in osteogenic activity and antibac-terial ability of titanium, Biomaterials 35 (2014) 7699–7713.

[20] E. Landi, S. Sprio, M. Sandri, G. Celotti, A. Tampieri, Development of Sr and CO3 co-substituted hydroxyapatites for biomedical applications, Acta Biomater. 4 (2008)656–663.

[21] E. Landi, A. Tampieri, G. Celotti, S. Sprio, M. Sandri, G. Logroscino, Sr-substituted hy-droxyapatites for osteoporotic bone replacement, Acta Biomater. 3 (2007) 961–969.

[22] E. Gentleman, Y.C. Fredholm, G. Jell, N. Lotfibakhshaiesh, M.D. O'Donnell, R.G. Hill,The effects of strontium-substituted bioactive glasses on osteoblasts and osteoclastsin vitro, Biomaterials 31 (2010) 3949–3956.

[23] M. Schumacher, A. Lode, A. Helth, M. Gelinsky, A novel strontium(II)-modified calci-um phosphate bone cement stimulates hMSC proliferation and osteogenic differen-tiation in vitro, Acta Biomater. 9 (2013) 9547–9557.

[24] J. Braux, F. Velard, C. Guillaume, S. Bouthors, E. Jallot, J.M. Nedelec, D.L. Maquin, P.Laquerrière, A new insight into the dissociating effect of strontium on bone resorp-tion and formation, Acta Biomater. 7 (2011) 2593–2603.

[25] C. Capuccini, P. Torricelli, F. Sima, E. Boanini, C. Ristoscu, B. Bracci, G. Socol, M. Fini,I.N. Mihailescu, A. Bigi, Strontium-substituted hydroxyapatite coatings synthesizedby pulsed-laser deposition: in vitro osteoblast and osteoclast response, ActaBiomater. 4 (2008) 1885–1893.

[26] K.L. Wong, C.T. Wong, W.C. Liu, H.B. Pan, M.K. Fong, W.M. Lam, W.L. Cheung, W.M.Tang, K.Y. Chiu, K.D.K. Luk, W.W. Lu, Mechanical properties and in vitro responseof strontium-containing hydroxyapatite/polyetheretherketone composites, Bioma-terials 30 (2009) 3810–3817.

[27] B. Busse, B. Jobke, M. Hahn, M. Priemel, M. Niecke, S. Seitz, J. Zustin, J. Semler, M.Amling, Effects of strontium ranelate administration on bisphosphonate-alteredhydroxyapatite: matrix incorporation of strontium is accompanied by changes inmineralization and microstructure, Acta Biomater. 6 (2010) 4513–4521.

[28] C.Y. Chien, W.B. Tsai, Poly(dopamine)-assisted immobilization of Arg-Gly-Asp pep-tides, hydroxyapatite, and bone morphogenic protein-2 on titanium to improvethe osteogenesis of bone marrow stem cells, ACS Appl. Mater. Interfaces 5 (2013)6975–6983.

[29] T.J. Webster, E.A. Massa-Schlueter, J.L. Smith, E.B. Slamovich, Osteoblast response tohydroxyapatite doped with divalent and trivalent cations, Biomaterials 25 (2004)2111–2121.

[30] E. Landi, A. Tampieri, G. Celotti, S. Sprio, Densification behaviour and mechanisms ofsynthetic hydroxyapatites, J. Eur. Ceram. Soc. 20 (2000) 2377–2387.

[31] A. Bigi, G. Falini, E. Foresti, M. Gazzano, A. Ripamonti, N. Roveri, Rietveld structurerefinements of calcium hydroxylapatite containing magnesium, Acta Crystallogr.52 (1996) 87–92.

[32] F.Z. Ren, R.L. Xin, X. Ge, Y. Leng, Characterization and structural analysis of zinc-substituted hydroxyapatites, Acta Biomater. 5 (2009) 3141–3149.

[33] L.C. Robert, Strontium-calcium hydroxyapatite solid solutions: preparation andlattice constant measurements, J. Am. Chem. Soc. 20 (1959) 5275–5278.

[34] Z.L. Xiu, M.K. Lu, S.W. Liu, G.J. Zhou, B.Y. Su, H.P. Zhang, Barium hydroxyapatitenanoparticles synthesized by citric acid sol–gel combustion method, Mater. Res.Bull. 40 (2005) 1617–1622.

[35] C.S. Ciobanu, F. Massuyeau, L.V. Constantin, D. Predoi, Structural and physicalproperties of antibacterial Ag-doped nano-hydroxyapatite synthesized at 100 °C,Nanoscale Res. Lett. 6 (2011) 613–620.

[36] M. Kheradmandfard, M.H. Fathi, Fabrication and characterization of nanocrystallineMg-substituted fluorapatite by high energy ball milling, Ceram. Int. 39 (2013)1651–1658.

[37] P.J. Goodhew, J. Humphreys, R. Beanland, Electron Microscopy and Analysis, Taylorand Francis, Third ed., 2001 180–200.

[38] G.A. Fielding,M. Roy, A. Bandyopadhyay, S. Bose, Antibacterial and biological charac-teristics of silver containing and strontium doped plasma sprayed hydroxyapatitecoatings, Acta Biomater. 8 (2012) 3144–3152.

[39] S. Pal, Y.K. Tak, J.M. Song, Does the antibacterial activity of silver nanoparticlesdepend on the shape of the nanoparticle? A study of the Gram-negative bacteriumEscherichia coli, Appl. Environ. Microbiol. 73 (2007) 1712–1720.

[40] G.A. Martinez-Castanon, N. Nino-Martinez, F. Martinez-Gutierrez, J.R. Martinez-Mendoza, F. Ruiz, Synthesis and antibacterial activity of silver nanoparticles withdifferent sizes, J. Nanoparticle Res. 10 (2008) 1343–1348.

[41] A. Gao, R.Q. Hang, X.B. Huang, L.Z. Zhao, X.Y. Zhang, L. Wang, B. Tang, S.L. Ma, P.K.Chu, The effects of titania nanotubes with embedded silver oxide nanoparticles onbacteria and osteoblasts, Biomaterials 35 (2014) 4223–4235.

[42] S. Agnihotri, S. Mukherji, S. Mukherji, Size-controlled silver nanoparticles synthe-sized over the range 5–100 nm using the same protocol and their antibacterialefficacy, RSC Adv. 4 (2014) 3974–3983.

[43] W.K. Jung, H.C. Koo, K.W. Kim, S. Shin, S.H. Kim, Y.H. Park, Antibacterial activity andmechanism of action of the silver ion in Staphylococcus aureus and Escherichia coli,Appl. Environ. Microbiol. 74 (2008) 2171–2178.

[44] P.N. Lim, E.Y. Teo, B. Ho, B.Y. Tay, E.S. Thian, Effect of silver content on the antibacte-rial and bioactive properties of silver-substituted hydroxyapatite, J. Biomed. Mater.Res. A 101 (2013) 2456–2464.

[45] A. Bankapur, S. Barkur, S. Chidangil, D. Mathur, A micro-Raman study of live, singlered blood cells (RBCs) treated with AgNO3 nanoparticles, PLoS One 9 (7) (2014),e103493. http://dx.doi.org/10.1371/journal.pone.0103493.

[46] J. Caverzasio, Strontium ranelate promotes osteoblastic cell replication through atleast two different mechanisms, Bone 42 (2008) 1131–1136.

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具