In vitro cytotoxicity evaluation of porous TiO2–Ag antibacterial coatings for human fetal osteoblasts

Acta Biomaterialia 8 (2012) 4191–4197

Contents lists available at SciVerse ScienceDirect

Acta Biomaterialia journal homepage: www.elsevier.com/locate/actabiomat

In vitro cytotoxicity evaluation of porous TiO2–Ag antibacterial coatings for human fetal osteoblasts B.S. Necula a,⇑, J.P.T.M. van Leeuwen b, L.E. Fratila-Apachitei a, S.A.J. Zaat c, I. Apachitei a, J. Duszczyk a a

Delft University of Technology, Department of Biomechanical Engineering, Group of Biomaterials Technology, Mekelweg 2, 2628 CD Delft, The Netherlands Erasmus Medical Center, Department of Internal Medicine, Dr. Molewaterplein 50, 3015 GE Rotterdam, The Netherlands c Academic Medical Center, Department of Medical Microbiology and Center for Infection and Immunity Amsterdam (CINIMA), Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands b

a r t i c l e

i n f o

Article history: Received 16 March 2012 Received in revised form 11 June 2012 Accepted 10 July 2012 Available online 17 July 2012 Keywords: Cytotoxicity Ag nanoparticles Osteoblasts Porous coatings Antibacterial coatings

a b s t r a c t Implant-associated infections (IAIs) may be prevented by providing antibacterial properties to the implant surface prior to implantation. Using a plasma electrolytic oxidation (PEO) technique, we produced porous TiO2 coatings bearing various concentrations of Ag nanoparticles (Ag NPs) (designated as 0 Ag, 0.3 Ag and 3.0 Ag) on a Ti–6Al–7Nb biomedical alloy. This study investigates the cytotoxicity of these coatings using a human osteoblastic cell line (SV-HFO) and evaluates their bactericidal activity against methicillin-resistant Staphylococcus aureus (MRSA). The release of Ag and the total amount of Ag in the coatings were determined using a graphite furnace atomic absorption spectrometry technique (GF-AAS) and flame-AAS, respectively. Cytotoxicity was evaluated using the AlamarBlue assay coupled with the scanning electron microscopy (SEM) observation of seeded cells and by fluorescence microscopy examination of the actin cytoskeleton and nuclei after 48 h of incubation. Antibacterial activity was assessed quantitatively using a direct contact assay. AlamarBlue viability assay, SEM and fluorescence microscopy observation of the SV-HFO cells showed no toxicity for 0 Ag and 0.3 Ag specimens, after 2, 5 and 7 days of culture, while 3.0 Ag surfaces appeared to be extremely cytotoxic. All Ag-bearing surfaces had good antibacterial activity, whereas Ag-free coatings showed an increase in bacterial numbers. Our results show that the 0.3 Ag coatings offer conditions for optimum cell growth next to antibacterial properties, which makes them extremely useful for the development of new antibacterial dental and orthopedic implants. Ó 2012 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

1. Introduction Biomedical devices (e.g. artificial joints, dental implants and fracture fixation plates) are commonly used in total joint arthroplasties, dental implantation or bone trauma surgeries, improving the quality of life by relieving pain and restoring mobility and function. Despite advanced sterilization techniques, strict surgery rules and systemic antibiotic prophylaxis, implant-associated infections (IAIs) remain a great risk in such surgeries. IAI is a result of bacteria adhesion to the implant, colonization of its surface and subsequent biofilm formation at the implantation site. The biofilm has an extraordinary resistance to antibiotics and furthermore it can promote detaching of individual bacteria into the surrounding tissue and circulatory system, leading to further complications [1]. Once the IAI occurs, it is often impossible to heal without revision surgery, which most of the time requires the replacement of the implant. This devastating complication may lead to large skeletal defects, member amputation and even death. Besides patient trau⇑ Corresponding author. Tel.: +31 152782414; fax: +31 152786730. E-mail address: [email protected] (B.S. Necula).

ma, the treatment of such infections incurs huge costs for the healthcare system [2–4]. Preventing bacterial adhesion on the biomedical devices or providing bactericidal activity to the biomedical device itself can be essential strategies to prevent IAI [5]. Therefore, research on surface modification of biomedical alloys to apply/form coatings/layers that kill any adherent and/or surrounding bacteria has garnered significant interest [6]. The unique advantage of these coatings/layers is the ability to provide locally, at the site of implantation, a controlled amount of the antibacterial agent that will prevent bacteria colonization. Furthermore, the local delivery of the antibacterial agents will reduce the risk of toxicity caused by conventional systemic delivery of antibiotics. Ideally, these coatings/layers should not change the structural integrity of the device and maintain the surface biocompatibility with the host tissue. Previous studies demonstrated the potential of the plasma electrolytic oxidation (PEO) process to produce porous TiO2–Ag antibacterial coatings on Mg and Ti biomedical alloys using electrolytes bearing Ag nanoparticles (Ag NPs) [7,8]. The coatings showed excellent in vitro antibacterial activity against methicillin-resistant

1742-7061/$ - see front matter Ó 2012 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.actbio.2012.07.005

4192

B.S. Necula et al. / Acta Biomaterialia 8 (2012) 4191–4197

Staphylococcus aureus (MRSA) [9]. The mechanism of incorporation of the Ag NPs, their distribution throughout the coatings and their chemical composition in the layer were assessed using high resolution transmission electron microscopy (HRTEM) and scanning electron microscopy (SEM), combined with energy dispersive X-ray spectroscopy (EDX) [7]. However, the toxicity of TiO2–Ag coatings for osteoblast cells still remains an open question. The simian virus 40 (SV40)-immortalized human fetal osteoblast (SV-HFO) cell line is valuable for studying bone metabolism via osteoblast differentiation [10–12]. Proliferating osteoblasts (days 2–7) differentiate to extracellular matrix producing cells (days 7–14), which will mineralize after 14–21 days. In this study, the human osteoblast cells were used to assess the toxicity of TiO2–Ag coatings bearing various amounts of Ag NPs during the proliferating stage (days 2–7). This study focuses on the investigation of the relationship between various Ag NPs contents in the PEO coatings and their effects on human osteoblast viability and antibacterial activity against MRSA. 2. Materials and methods 2.1. Coatings synthesis and characterization 2.1.1. Preparation of Ag-bearing TiO2 coatings Ti–6Al–7Nb biomedical alloy disks (22 mm diameter, 8 mm height) were used in this study. They were ground with 320, 800 and 1200# abrasive papers and then ultrasonically cleaned with acetone, ethanol and distilled water in an ultrasonic bath prior to PEO treatment. The PEO process was performed on a custom-made laboratory scale set-up consisting of an AC power supply, type ACS 1500 (ET Power Systems Ltd., UK) and a double wall glass electrolytic cell cooled by a mixture of water and glycerol using a Haake V15 thermostatic bath. The electrolytic cell was filled with the electrolyte, which comprised 0.02 M calcium acetate (P99%, Sigma–Aldrich) and 0.15 M calcium glycerophosphate (Dr. Paul Lohmann, Germany) solutions to which 0, 0.3 and 3.0 g l1 Ag nanoparticles (99.5%, Sigma–Aldrich) were added. To maintain particle homogeneity the electrolyte was stirred at 500 rpm during the PEO process. The PEO process was performed under galvanostatic conditions using a current density of 20 A dm2 for 300 s. To remove any unattached Ag NPs, after the PEO process the treated samples were kept in running tap water for 5 min, ultrasonically cleaned in 70% ethanol and rinsed for 5 min in deionized water. All samples were sterilized for 1 h at 110 °C using a Nabertherm TR60 oven. The sample code, electrolyte components and process parameters are listed in Table 1. The coating morphology was investigated using the SEM, coating thickness was measured using the Elcometer 456 Coating Thickness Gauge, and elemental composition was determined by energy dispersive X-ray analyses (EDX). 2.1.2. Ag release Ag release experiments were performed in triplicates for the 0.3 Ag and 3.0 Ag samples. The oxidized disks were placed in dark Wheaton bottles containing 30 ml of double distilled water and

kept at 37 °C under gentle stirring conditions up to 7 days. The liquid was collected at days 1, 2, 5 and 7 and replaced with fresh double distilled water. All collected samples were acidified with HNO3 to a final concentration of 0.5 M and were analyzed with the graphite furnace atomic absorption spectrometry (GF-AAS) technique using the PerkinElmer 4100ZL instrument. The time points were selected according to the cytotoxicity tests. 2.1.3. Total Ag content in the coatings The total amount of silver incorporated in the coatings was determined for the 0.3 Ag and 3.0 Ag samples. The TiO2 matrix, bearing Ag NPs, was dissolved in 98% H2SO4 at 90 °C in an ultrasonic bath for 2 h. After cooling, 69% HNO3 was added to further dissolve the loose Ag NPs, and the liquid samples were tested for Ag content with the flame atomic absorption spectroscopy (flame-AAS) technique using the PerkinElmer AAnalyst 100 instrument. To estimate the total amount of Ag available for release, a new set of PEO coated samples were immersed in 15 ml HNO3 with a concentration of 69%. Only the exposed Ag NPs (i.e. embedded on the surface and inside the open porosity) will be dissolved by the nitric acid. After 24 h, the immersion liquid was collected and analyzed by flame AAS. 2.2. In vitro cytotoxicity evaluation 2.2.1. Cell culture SV-HFO cells [13] were precultured for 1 week in a-minimum essential medium (aMEM) (GIBCO, Paisley, UK) supplemented with 20 mM 4-(2-hydroxyethyl)piperazine-1-ethanesulfonic acid (HEPES) (Sigma–Aldrich), 2% streptomycin/penicillin, 1.8 mM CaCl2 (Sigma–Aldrich) and 10% heat-inactivated fetal calf serum (HI-FCS) (GIBCO) at pH 7.5, 37 °C and 5% CO2 in a humidified atmosphere. During this preculture, the SV-HFO cells remained in an undifferentiated stage. After preculturing, the cells were washed with a phosphate buffered saline solution (PBS), detached with trypsin, counted with a hemocytometer and seeded on the samples in a density of 3.5  104 cells per disk. Osteogenic medium (aMEM supplemented with fresh b-glycerophosphate (Sigma–Aldrich) (10 mM) and dexamethasone (Sigma–Aldrich) (100 nM)) was added, starting 2 days after seeding and replaced every 2–3 days. The coated disks fit precisely in the 12-well plates, avoiding the growth of cells on the polystyrene. 2.2.2. Cell viability assay AlamarBlue assay (Invitrogen) was used to assess SV-HFO cell viability on the Ag-bearing and Ag-free coatings. At days 2, 5 and 7 of culture, AlamarBlue reagent was added to each well, in an amount equal to 10% of the well volume. The plates were incubated for a further 3 h at 37 °C to allow the resazurin, a blue non-fluorescent indicator dye, to be converted to highly red fluorescent resorufin via reduction reactions of metabolically active cells. Fluorescence measurements were made on a Wallac microplate fluorometer using an excitation length of 535 nm and emission of 590 nm. The assay was performed at least in triplicate.

Table 1 Sample code, electrolyte components and process parameters used. Sample code

Electrolyte components Calcium acetate, mol L1

Calcium glycero-phosphate, mol L1

Ag nanoparticles, g L1

Current density, A dm2

Process parameters Final voltage, V

Duration, min

Temperature, °C Start

End

0 Ag 0.3 Ag 3.0 Ag

0.02 0.02 0.02

0.15 0.15 0.15

0 0.3 3.0

20 20 20

239 ± 2 234 ± 3 237 ± 2

5 5 5

11 ± 1 11 ± 1 11 ± 1

34 ± 0.3 33 ± 0.9 32 ± 1

4193

B.S. Necula et al. / Acta Biomaterialia 8 (2012) 4191–4197

2.2.3. SV-HFO cell morphology and spreading assay The SV-HFO cell morphology and spreading were observed using SEM. After 2, 5 and 7 days of SV-HFO culture on the coated disks, the cells were fixed with 4% formaldehyde and 2% glutaraldehyde in 0.1 M sodium cacodylate solution at 4 °C for 1 h. The samples were rinsed in 0.1 M sodium cacodylate solution and dehydrated in a series of graded ethanol and then 100% tetramethylsilane. All samples were gold coated and examined on a Jeol 6500F-D scanning electron microscope operated at 5–10 kV. 2.2.4. Fluorescence microscopy of actin cytoskeletal organization and nucleus After culturing for 48 h, the cells seeded on the coated disks were washed with PBS and fixed for 15 min at room temperature using 4% paraformaldehyde in PBS. After three rinses with PBS the cells were permeabilized with PBS/Triton X-100 for 10 min and stained with rhodamine–phalloidin (Invitrogen) (1:100 in PBS) for 20 min at room temperature and in the dark. The disks were further rinsed three times with PBS and stained with a 40 ,6-diamidino-2phenylindole (DAPI) solution (Invitrogen) (1:50000 in PBS) for 5 min. The actin cytoskeleton and cell nuclei were examined using an epifluorescence microscope (Nikon Eclipse LV 100D-U, Japan). 2.3. In vitro antibacterial activity The antibacterial activity of the Ag-bearing layers against MRSA was assessed in vitro using the Japanese Industrial Standard JIS Z 2801:2000 [14] modified to better reproduce the scenario of an implant infection during a primary surgery [9]. In short, a fresh liquid culture of MRSA, strain AMC201, was prepared by adding 1–5 colonies from a streak plate into 5 ml of trypticase soy broth (TSB). The suspension was incubated at 37 °C under rotation for 2 h and subsequently diluted with TSB to an optical density (OD620) of 0.03, corresponding to 107 colony-forming units (CFU) per ml. Sterilized nitrocellulose filter disks were placed on a blood agar plate and 20 ll of the diluted MRSA culture, containing 2  105

A

2.4. Statistical analysis Statistical analyses were performed using the one-way Anova test and Tukey–Kramer method to make multiple unplanned comparisons of means based on unequal sample sizes. In the Tukey– Kramer method, the minimum significant difference (MSD) was calculated for each pair of means. If the observed absolute difference between a pair of means was greater than the MSD, the pair of means was declared significantly different. 3. Results 3.1. Coating appearance and morphology Fig. 1 shows the visual appearance and the surface morphology of the uncoated, 0 Ag, 0.3 Ag and 3.0 Ag specimens. It can be seen that Ag-bearing coatings exhibit a brown color as compared with the white–grey Ag-free coatings. The brown color is attributed to the Ag content and increases in intensity with increasing the

0.3Ag

0Ag

Ti6Al7Nb

CFU, was pipetted onto the filters. The medium was absorbed by the agar while the MRSA bacteria were retained on the filter. 20 ll of 1% TSB in 10 mM phosphate was pipetted centrally on the surface of each titanium disk and an inoculated filter disk was carefully placed on top, with the bacteria contacting the coated surface. All disks with bacterial filters were placed individually in petri dishes and incubated at 37 °C for 24 h in a humid atmosphere. After incubation, each disk and the corresponding filter were placed in 5 ml of TSB, sonicated for 30 s and vortexed for 1 min to dislodge adherent bacteria. This procedure does not affect bacterial viability. Seven ten-fold serial dilutions were made in a 96-well plate. Subsequently, triplicate 10 ll aliquots of the undiluted suspension and of the seven dilutions were pipetted onto blood agar plates. The blood agar plates were incubated overnight at 37 °C and the number of colonies was counted the following day. All experiments were performed at least in triplicate.

3.0Ag

B

10 µm

10 µm

10 µm

10 µm

C Ag

1 µm

Ag 1 µm

1 µm

Fig. 1. General sample appearance before and after the PEO process (A), surface morphology of the Ti6Al7Nb samples before and after the PEO process (B) and high magnification SEM images of the surface of the coatings showing the presence of Ag nanoparticles (C).

4194

B.S. Necula et al. / Acta Biomaterialia 8 (2012) 4191–4197

0.1

0.3Ag

Ag released, µg ml-1

0.09

3.0Ag

0.08 0.07 0.06 0.05 0.04 0.03 0.02 0.01 0

1

2

3

4

5

6

7

Immersion time, days Fig. 2. Cumulative release of Ag from TiO2–Ag coatings immersed in deionized water at 37 °C for 7 days.

no cytotoxic effect for the 0.3 Ag surfaces after 2 days of culture. After culturing for 5 and 7 days, the cells on the 0.3 Ag showed similar viability with those on 0 Ag, indicating that the presence of Ag had no adverse effect on osteoblast viability. However, the 3.0 Ag specimens appeared to be toxic for the SV-HFO cells showing low viability after 2, 5 and 7 days in culture. These observations were further confirmed by the SEM images of the SV-HFO cells cultured on the 0 Ag, 0.3 Ag and 3.0 Ag disks for 2, 5 and 7 days (Fig. 4). Epifluorescence microscopy of the SV-HFO cells (Fig. 5) revealed a well-defined actin cytoskeletal organization and numerous stress fibers throughout the cells for the 0 Ag and 0.3 Ag surfaces. The cells on 0 Ag and 0.3 Ag exhibited a stellate polygonal morphology with multidirectional spreading while on the 3 Ag the cells were round and lower in number. 3.4. In vitro antibacterial activity

amount of Ag in the electrolyte. No obvious differences in the morphology, thickness and chemical composition between the Agbearing and Ag-free coatings were observed. SEM images showed well separated pores, ranging from a few nanometers up to 5 lm in size, homogeneously distributed over the coating surfaces. The high magnification SEM pictures (Fig. 1) demonstrated the presence of Ag NPs fused in the TiO2 matrix surface with increased concentration of particles on the 3.0 Ag surface. The thickness of the Ag-bearing and Ag-free coatings was 18 lm, regardless of the content of Ag NPs. EDX analyses of the layers showed the presence of Ti, Al and Nb from the metallic substrate next to Ca, P and Ag incorporated from the electrolyte. Except for the Ag NPs concentration in the layers, the other surface characteristics were not changed. 3.2. Ag content and release The amount of Ag released from the 0.3 Ag and 3.0 Ag coatings was measured by GFAAS and the results are shown in Fig. 2. The cumulative Ag concentrations in the solutions after immersion of the 0.3 Ag and 3.0 Ag surfaces for 2, 5 and 7 days are shown in Table 2. The total amount of Ag incorporated in the 0.3 Ag and 3.0 Ag coatings was determined by flame AAS after total dissolution of the porous layers in hot H2SO4 and HNO3. The results showed that the total amount of Ag embedded in the 0.3 Ag and 3.0 Ag coatings is 27.53 ± 0.82 and 213.9 ± 3.13 lg per disk, respectively. However, only the Ag NPs fused into the open pore walls and on the surface of the coatings can actually contribute to the antibacterial activity of the coatings, and this amount was determined to be 20.82 ± 0.88 and 127.75 ± 5.28 lg per disk for the 0.3 Ag and 3.0 Ag, respectively. 3.3. SV-HFO cell response The viability of SV-HFO cells cultured on the Ag-free and Agbearing surfaces was determined by using the AlamarBlue assay and the results are shown in Fig. 3. There was no significant difference in the fluorescence signal between 0 Ag and 0.3 Ag, showing

The antibacterial activity of 0 Ag, 0.3 Ag and 3.0 Ag coatings was assessed using a modified version of JIS Z 2801 standard. Fig. 6 shows the percentage reduction of MRSA colonies on 0 Ag, 0.3 Ag and 3.0 Ag coatings, calculated based on the quantitative cultures of the solutions resulted after sonication of the disks, inoculated and incubated for 24 h. The percentage killing of MRSA was found to be 98.0 ± 2% in the case of 0.3 Ag and 99.75 ± 0% for the 3.0 Ag. The Ag-free coatings (0 Ag) showed no bactericidal properties, as evidenced by the 1000-fold increase in MRSA CFU. 4. Discussion In this work, Ag-bearing TiO2 coatings with different Ag NPs contents were produced by PEO. Incorporation of various amounts of Ag NPs in the TiO2 matrix, apparently, did not change the morphology of the coatings. All layers produced were porous, with well-defined pores ranging from a few nanometers up to 5 lm in size (Fig. 1). Moreover, the coatings showed the same thickness and elemental composition, regardless of Ag NPs content. Therefore, the effect of Ag NPs, embedded in the TiO2 coatings in various amounts, on the viability of the SV-HFO cells and MRSA killing ability could be evaluated without the interference of the other factors. Many studies have been performed to evaluate the toxicity of Ag NPs using different cell lines, yet little is known about the mechanisms of Ag NPs toxicity [15]. Most of the studies conclude that under specific concentration levels Ag NPs are not cytotoxic [16–18]. Kawata et al. [16] found that under a concentration of 0.5 lg ml1 Ag NPs suspended in the culture medium, the proliferation of HepG2 human hepatoma cells was accelerated while higher doses (1 lg ml1) induced abnormal cellular morphology and cellular shrinkage. Kim et al. [18] studied the oxidative stressdependent toxicity of Ag in human hepatoma cells and reported good cell viability at dose levels smaller than 0.7 lg ml1 Ag NPs in the medium. Cao et al. [19] reported the amount of Ag+ released from titanium samples embedded with Ag NPs to be less than 0.01 lg ml1 after 60 days, promoting proliferation of the osteoblast-like cell line MG63. However, there is no indication of the total amount of Ag NPs incorporated that may be in contact with the MG63 cells. Also, Hsu et al. [20] showed enhanced fibroblast

Table 2 Total Ag content and release from Ag-bearing TiO2 coatings. Sample

Ag nanoparticles in electrolyte, g L1

Total Ag incorporated in the coatings, lg per disk

Total Ag contributing to bactericidal activity, lg per disk

Cumulative Ag release, lg ml1 2 days

5 days

7 days

0.3 Ag 3.0 Ag

0.3 3.0

27.53 ± 0.82 213.88 ± 3.13

20.82 ± 0.88 127.75 ± 5.28

0.0077 ± 0.0022 0.0492 ± 0.0129

0.0096 ± 0.0027 0.0633 ± 0.0132

0.0120 ± 0.0032 0.0894 ± 0.0154

4195

B.S. Necula et al. / Acta Biomaterialia 8 (2012) 4191–4197

4000

Fluorescence, a.u.

3500

3000

0Ag 0.3Ag

3.0Ag

2500 2000 1500

1000 500 0 2

5

7

Culture time, days Fig. 3. Simian virus 40 (SV40)-immortalized human fetal osteoblast cells viability on TiO2 porous coatings bearing different concentrations of Ag NPs.

0.3Ag

0Ag

2D

3.0Ag

300 µm

5D

7D Fig. 4. SEM observation of the SV-HFO cells cultured on the 0 Ag, 0.3 Ag and 3.0 Ag for 2, 5 and 7 days.

0Ag

0.3Ag

3.0Ag

Fig. 5. Fluorescence microscopy images of actin-stained (phalloidin) (red) and nuclei-stained (DAPI) (blue) human fetal osteoblasts cultured for 48 h on Ag-free and Agbearing coatings. Scale bar is 100 lm.

4196

B.S. Necula et al. / Acta Biomaterialia 8 (2012) 4191–4197

*

Percent killing of MRSA

105

*

100

95

90 10 0

0Ag

0.3Ag

1000-fold increase in bacteria CFU

98 ± 2 % killing

3.0Ag

100 10-1 10-2 10-3 10-4 10-5 10-6 10-7 >99.75 % killing

Fig. 6. Bactericidal activity of TiO2 coatings bearing different concentrations of Ag NPs against methicillin-resistant Staphylococcus aureus.

attachment on the polyurethane–Ag nanocomposites bearing 30 ppm AgNPs and reported a release of 0.00035 lg ml1 of both ionic and nanoparticulate Ag. Summarizing, the experimental conditions differ and there is a clear distinction between testing the Ag NPs suspended in the culture media and Ag NPs incorporated/attached (in)to a solid substrate. When suspended in the culture media, depending on the dispersion stability, the Ag NPs might settle on the surface of the cells. Thus, Ag NPs toxicity might be due to the direct contact after settlement, or Ag NPs entering the cell membrane by different uptake routes (i.e. endocytosis or diffusion) causing cytoskeleton damage, oxidation of DNA, mitochondrial dysfunction and chromosomal aberrations leading to cell apoptosis [21]. However, there is no control on the amount of Ag NPs in contact with the cells and it is difficult to estimate the toxic concentration levels. When incorporated and fixed into a substrate, the mechanism of Ag NPs toxicity might be due to direct contact with the seeded cells disrupting normal cell behavior. However, Ag NPs have a high surface area per unit mass and always release a continuous level of ionic silver (Ag+) into their environment. Furthermore, according to a study of Liu and Hurt [22], Ag NPs surfaces can adsorb Ag+. It is generally agreed that both Ag+ (free or adsorbed) as well as the Ag nanoparticle itself may contribute to the cytotoxic effects of Ag NPs [16,17]. However, Navarro et al. [23] suggested that Ag NPs contribute to toxicity only by serving as sources of Ag+ and alone are not a direct source of toxicity. On the other hand, Kim et al. [18] proposed that Ag NPs cytotoxicity is primarily the result of oxidative stress and is independent of the toxicity of Ag+ ions. In the present work, various amounts of Ag NPs are fused in the TiO2 matrix and SV-HFO cells are seeded on top of the surfaces. Thus, both direct contact of the Ag-bearing surfaces with the cells

as well as the release of Ag+ in the osteogenic medium may contribute to the toxicity of these surfaces. The release of Ag from the 0.3 Ag coatings was assessed by GFAAS and the results suggest that the cumulative concentrations within 2, 5 and 7 days of immersion are below the cytotoxic level found in the literature i.e., 0.0077, 0.0096 and 0.012 lg ml1, respectively. These surfaces showed no toxicity for SV-HFO cells as confirmed by AlamarBlue assay, SEM and fluorescence microscopy. The 3.0 Ag coatings are very toxic to SV-HFO cells despite still low Ag concentrations, i.e., 0.05, 0.063 and 0.09 lg ml1 released within 2, 5 and 7 days, respectively. This cytotoxic effect may be due to the higher concentration of Ag NPs present in the TiO2 matrix comparative with 0.3 Ag surfaces. Thus, disruption of the normal SV-HFO cell behavior on the 3.0 Ag coatings may be because of the direct contact with the Ag-bearing surface rather than the Ag released in the osteogenic medium. The bactericidal activity of 0 Ag, 0.3 Ag and 3.0 Ag was assessed using one of the most prevalent and virulent pathogen responsible for IAI (i.e. MRSA) through a direct contact assay. The results show a clear difference between the effects of Ag-bearing surfaces on osteoblasts and MRSA. If on the 0.3 Ag surfaces the SV-HFO cells viability was not affected, the growth of bacteria is strongly repressed. The percentage reduction of MRSA seeded on 0.3 Ag coatings was 98.0 ± 2% (Fig. 6), indicating that incorporation of a small amount of Ag NPs can provide antibacterial activity of the TiO2 coatings. As expected, this activity is enhanced with increasing the concentration of Ag NPs in the coatings. Incorporation of a larger concentration of Ag nanoparticles in the 3.0 Ag coatings showed an enhanced killing of the MRSA inoculum (>99.75%) after 24 h, while Ag-free samples (0 Ag) allowed bacteria growth up to 1000-fold. It is generally agreed that both ionic silver and the nanoparticles per se contribute to the bactericidal activity of Ag NPs. Using the direct contact assay the Ag NPs embedded in the TiO2 matrix were in close contact with the MRSA. Thus, Ag NPs can directly interact with the microbial cells, causing degradation of the lipopolysaccharide molecules and forming ‘‘pits’’ leading to large increases in membrane permeability [24]. Ag NPs may also produce secondary products (such as reactive oxygen species) that can penetrate the bacterial cell envelope and cause DNA damage and protein oxidation [25]. Furthermore, Ag+ released from Ag NPs will have also a contribution to the bactericidal effect of the TiO2–Ag coatings by interacting with the thiol group proteins, turning the DNA into condensed form and disabling its replication ability [26]. PEO is a highly promising surface modification technology that enables the control of the antibacterial agent loaded in the porous coatings in order to minimize cytotoxicity as well as maintain good ability to kill bacteria. The coatings produced in electrolytes bearing 0.3 g l1 Ag NPs were showing good osteoblast cell viability after 7 days’ exposure as well as effective bactericidal activity against MRSA, which makes them suitable in the fight against IAIs. Future research will focus on the long-run testing of the antibacterial activity and cytotoxicity of Ag-bearing surfaces.

5. Conclusions TiO2–Ag coatings with various contents of Ag NPs were successfully formed by PEO on Ti–6Al–7Nb medical alloys, in electrolytes bearing different concentrations of nanoparticles. Except for the Ag NPs concentration in the layers, the other surface characteristics were not changed. The Ag-free coatings and 0.3 Ag coatings showed no significant cytotoxicity for SV-HFO cells after 2, 5 and 7 days of culture as evidenced by AlamarBlue viability assay, SEM and fluorescence

B.S. Necula et al. / Acta Biomaterialia 8 (2012) 4191–4197

microscopy examination. The 3.0 Ag surfaces were toxic to SHHVO cells, probably because of the higher concentration of Ag in the TiO2 matrix. The direct contact with Ag-bearing surfaces may have strongly contributed to the cytotoxic effect of 3.0 Ag coatings rather than the Ag released in the osteogenic medium. All Ag-bearing TiO2 coatings showed antibacterial activity against MRSA while a 1000-fold increase in bacterial CFU was recorded on Ag-free coatings. The findings of this in vitro study indicate that the 0.3 Ag coatings can provide good osteoblast viability after 2, 5 and 7 days as well as effective bactericidal activity against MRSA, which make them extremely useful for the development of new antibacterial dental and orthopedic implants. Appendix A. Figures with essential colour discrimination Certain figures in this article, particularly Figs. 1, 5 and 6, are difficult to interpret in black and white. The full colour images can be found in the on-line version, at http://dx.doi.org/10.1016/ j.actbio.2012.07.005. References [1] Davies D. Understanding biofilm resistance to antibacterial agents. Nat Rev Drug Discov 2003;2:114–22. [2] Esposito S, Leone S. Prosthetic joint infections: microbiology, diagnosis, management and prevention. Int J Antimicrob Agents 2008;32:287–93. [3] Zimmerli W. Infection and musculoskeletal conditions: prosthetic-jointassociated infections. Best Pract Res Clin Rheumatol 2006;20:1045–63. [4] Lidgren L, Knutson K, Stefánsdóttir A. Infection of prosthetic joints. Best Pract Res Clin Rheumatol 2003;17:209–18. [5] Francolini I, Donelli G. Prevention and control of biofilm-based medicaldevice-related infections. FEMS Immunol Med Microbiol 2010;59:227–38. [6] Simchi A, Tamjid E, Pishbin F, Boccaccini AR. Recent progress in inorganic and composite coatings with bactericidal capability for orthopaedic applications. Nanomedicine-UK 2011;7:22–39. [7] Necula BS, Apachitei I, Tichelaar FD, Fratila-Apachitei LE, Duszczyk J. An electron microscopical study on the growth of TiO(2)–Ag antibacterial coatings on Ti6Al7Nb biomedical alloy. Acta Biomater 2011;7:2751–7. [8] Necula BS, Fratila-Apachitei LE, Berkani A, Apachitei I, Duszczyk J. Enrichment of anodic MgO layers with Ag nanoparticles for biomedical applications. J Mater Sci Mater Med 2009;20:339–45.

4197

[9] Necula BS, Fratila-Apachitei LE, Zaat SA, Apachitei I, Duszczyk J. In vitro antibacterial activity of porous TiO2–Ag composite layers against methicillinresistant Staphylococcus aureus. Acta Biomater 2009;5:3573–80. [10] Seriwatanachai D, Krishnamra N, van Leeuwen J. Evidence for direct effects of prolactin on human osteoblasts: inhibition of cell growth and mineralization. J Cell Biochem 2009;107:677–85. [11] van Driel M, Koedam M, Buurman CJ, Roelse M, Weyts F, Chiba H, et al. Evidence that both 1 alpha, 25-dihydroxyvitamin D-3 and 24-hydroxylated D3 enhance human osteoblast differentiation and mineralization. J Cell Biochem 2006;99:922–35. [12] Chiba H, Sawada N, Ono T, Ishii S, Mori M. Establishment and characterization of a simian virus 40-immortalized osteoblastic cell-line from normal human bone. Japanese Journal of Cancer Research 1993;84:290–7. [13] Eijken M, Swagemakers S, Koedam M, Steenbergen C, Derkx P, Uitterlinden AG, et al. The activin A-follistatin system: potent regulator of human extracellular matrix mineralization. FASEB J 2007;21:2949–60. [14] Anon. Antimicrobial products – test for antimicrobial activity and efficacy. Japanese Industrial Standard JIS Z 2000;2801:2000. [15] Cao H, Liu X. Silver nanoparticles-modified films versus biomedical deviceassociated infections. Wiley Interdiscip Rev Nanomed Nanobiotechnol 2010;2:670–84. [16] Kawata K, Osawa M, Okabe S. In vitro toxicity of silver nanoparticles at noncytotoxic doses to HepG2 human hepatoma cells. Environ Sci Technol 2009;43:6046–51. [17] AshaRani PV, Low Kah Mun G, Hande MP, Valiyaveettil S. Cytotoxicity and genotoxicity of silver nanoparticles in human cells. ACS Nano 2009;3:279–90. [18] Kim S, Choi JE, Choi J, Chung KH, Park K, Yi J, et al. Oxidative stress-dependent toxicity of silver nanoparticles in human hepatoma cells. Toxicol in Vitro 2009;23:1076–84. [19] Cao HL, Liu XY, Meng FH, Chu PK. Biological actions of silver nanoparticles embedded in titanium controlled by micro-galvanic effects. Biomaterials 2011;32:693–705. [20] Hsu SH, Tseng HJ, Lin YC. The biocompatibility and antibacterial properties of waterborne polyurethane-silver nanocomposites. Biomaterials 2010;31:6796–808. [21] Asharani PV, Hande MP, Valiyaveettil S. Anti-proliferative activity of silver nanoparticles. BMC Cell Biol 2009;10:65. [22] Liu JY, Hurt RH. Ion release kinetics and particle persistence in aqueous nanosilver colloids. Environmental Science & Technology 2010;44:2169–75. [23] Navarro E, Piccapietra F, Wagner B, Marconi F, Kaegi R, Odzak N, et al. Toxicity of silver nanoparticles to Chlamydomonas reinhardtii. Environ Sci Technol 2008;42:8959–64. [24] Sondi I, Salopek-Sondi B. Silver nanoparticles as antimicrobial agent: a case study on E. coli as a model for Gram-negative bacteria. J Colloid Interf Sci 2004;275:177–82. [25] Morones JR, Elechiguerra JL, Camacho A, Holt K, Kouri JB, Ramirez JT, et al. The bactericidal effect of silver nanoparticles. Nanotechnology 2005;16:2346–53. [26] Feng QL, Wu J, Chen GQ, Cui FZ, Kim TN, Kim JO. A mechanistic study of the antibacterial effect of silver ions on Escherichia coli and Staphylococcus aureus. J Biomed Mater Res 2000;52:662–8.