Functional Biomaterials Surfaces

Investigating the effect of silver coating on the solubility, antibacterial properties, and cytocompatibility of glass microspheres

Journal of Biomaterials Applications 0(0) 1–13 ! The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0885328215591902 jba.sagepub.com

LM Haas, CM Smith, LM Placek, MM Hall, Y Gong, NP Mellott and Anthony W Wren

Abstract Silver (Ag) coatings have been incorporated into many medical materials due to its ability to eradicate harmful microbes. In this study, glass microspheres (SiO2–Na2O–CaO–Al2O3) were synthesized and employed as substrates to investigate the effect Ag coating has on glass solubility and the subsequent biological effects. Initially, glasses were amorphous with a glass transition point (Tg) of 605 C and microspheres were spherical with a mean particle diameter of 120 mm (27). The Ag coating was determined to be crystalline in nature and its presence was confirmed using scanning electron microscopy and X-ray photoelectron spectroscopy. Ion release determined that Ag-coated (Ag-S) microspheres increased the Naþ release rate but slightly reduced the Ca2þ and Si4þ release compared to an uncoated control (UC-S). Additionally, the Ag-S reduced the pH to just above neutral (7.3–8.5) compared to the UC-S (7.7–9.1). Antibacterial testing determined significant reductions in planktonic Escherichia coli (p ¼ 0.000), Staphylococcus epidermidis (p ¼ 0.000) and Staphylococcus aureus (p ¼ 0.000) growth as a function of the presence of Ag and with respect to maturation (1, 7, and 30 days). Testing for toxicity levels using L929 Fibroblasts determined higher cell viability for the Ag-S at lower concentrations (5 mg/ml); in addition, no significant reduction in cell viability was observed with higher concentrations (15, 30 mg/ml). Keywords Glass microspheres, silver, coatings, S. aureus, fibroblasts

Introduction Interest in glass-based biomaterials has increased in recent years since the original 45S5 Bioglass formulation developed by Hench in the late 1960s.1 These glasses were originally designed for orthopaedic applications, to restore damaged or diseased bone, which is due to their ability to form a close bond to bone tissue.1,2 The therapeutic advantages attributed to these glasses (SiO2–Na2O–CaO–P2O5) are related to both ion release (Ca2þ, Naþ PO3-4) from the glass and the subsequent precipitation of a calcium/phosphate (CaP) surface layer to the negatively charged Si-OH groups present on the glass surface. This reaction is regarded as a precursor to bone bonding in vivo.3 Today, many medical materials are being developed from bioactive glasses as different compositions and structures can be designed for specific surgical applications. Currently these include bone void fillers,1,4

composite materials with polymers to improve bioactivity, composite bone cements,5,6 glass–ceramic scaffolds to facilitate cellular and tissue ingrowth,7–9 and yttrium containing glass microspheres to treat tumor growths with radiotherapy.10–12 A highly desirable property that is currently being investigated in relation to medical materials is their antimicrobial capabilities. In particular, specific ions (Agþ, Cu2þ, Zn2þ) when eluted from medical materials can act as disinfectants. More recently, ions such as Ga3þ have been incorporated into bioactive glasses in order to eliminate harmful bacteria for both dental and Inamori School of Engineering, Alfred University, Alfred, NY, USA Corresponding author: Anthony W Wren, Inamori School of Engineering, Alfred University, Alfred, NY 14802, USA. Email: [email protected]

Downloaded from jba.sagepub.com by guest on July 24, 2015

2

Journal of Biomaterials Applications 0(0)

skeletal applications.13–15 Ion release from these materials is superior in terms of safety, durability and heat resistance than conventional organic compounds such as antibiotics.16 Some concerns with antibiotic use include allergic reactions, microbial flora depletion and bacterial resistance.16 In particular, the evolution of antibiotic resistant strains of bacteria such as Methicillin-resistant S. aureus (MRSA), Vancomycinresistant S. aureus and Methicillin-resistant S. epidermidis (MRSE) are known to pose significant concerns to elderly or immune-compromised patients in hospitals.17–20 As such, alternative antibiotic-free methods have been investigated to impart antibacterial properties to implantable materials. Agþ in particular is known to have a wide antibacterial spectrum (including Escherichia coli and Staphylococcus aureus) and to be relatively safe to humans21,22 as the cytotoxic concentration to humans is cited to be 1.6 ppm.23 Silver-based antimicrobial agents receive much attention because of the low toxicity of the active Agþ ion to human cells, as well as it being a long-lasting biocide with high thermal stability and low volatility.22 In addition, microbes such as S. epidermidis can form a thick multilayered biofilm which is composed of an extracellular polysaccharide known as polysaccharide intercellular adhesion (PIA) which can be very resilient to antimicrobial compounds,24 hence it is highly desirable to synthesize antimicrobial surfaces on medical materials that can reduce the possibility of biofilm colonization. Previous studies have demonstrated that silver aids in skin wound healing in rat models where it reduced the inflammatory and granulation tissue phases of healing, and encouraged epidermal repair.25 Also, Agþ, a positive cation, can bond to negative charges on the surface of glass which form an antimicrobial surface layer. There are numerous reports in the scientific literature on the antibacterial nature of medical materials containing Agþ.22,26 It has previously been used as a bactericidal agent added to materials in the form of nanoparticles,27 and it has also been investigated for incorporation into non-implantable medical products and health care/ hygienic products27,28 and as coatings on titanium surfaces to impart antibacterial properties to metallic implants.22 With respect to previous studies comparable to this work, a number of glass/ceramic materials have been doped with Agþ using different processing methods. Some of these materials and methods include Ag-glasses for dental restoration,26 Ag-doped silica coatings on glass substrates,29 and sol–gel-derived Ag-incorporated Bioglass (SiO2–CaO–P2O5–Ag2O).30 Verne et al. doped Ag into SiO2–CaO–Na2O glass plates (10  10  1 mm3) by ion exchange process in molten salts as well as in aqueous solution and achieved positive SBF (CaP deposition) and cytotoxicity results.16,31 Ag-doped SiO2–B2O3–Na2O glasses also

experienced positive antibacterial properties when tested in E. coli.32 Glass microsphere substrates were employed for this study as the surfaces of the resulting spheres are relatively smooth and each has an analogous morphology as opposed to using glass frit. In addition, imaging the Ag coating on the glass sphere surface is more accurate and repeatable, for techniques such as AFM, and the exposed surface areas can be better described and are more evenly distributed.33 In the past, glass microspheres have been used for medical applications to treat patients with colorectal liver metastases, liver and hepatic cancer by incorporating Yttrium 90 into the glass phase.10–12 From an industrial perspective, glass microspheres have been employed for applications including hydrogen storage and lead acid batteries.34–36 Previous studies related to applying coatings on glass microspheres include coating with hydroxyapatite to improve bioactivity and biocompatibility,33 and studies that apply SnO2 nanoparticle coatings on glass microspheres for applications in photo-catalysis.37 Employing microspheres for this study has a number of advantages over commercial Bioglass as these particles for instance are distributed over a broad range (90–710 mm) and are irregular in morphology. Glass microspheres have smooth surfaces that can limit damage to sensitive tissues if implanted in vivo; in addition, particle distribution can be controlled better using less processing steps, i.e. grinding and sieving. In relation to this study, glass microspheres were synthesized and used as substrates for Ag coatings to investigate any changes in the material solubility and the subsequent therapeutic response to prokaryotic and eukaryotic cells. For this work, the application of Ag was performed using spin coating with AgNO3, as Ag-doped silicate-based glasses are difficult to prepare by the conventional melt-quench method.26 This study initially aims to characterize the Ag coating and monitor any changes on glass solubility and also the Ag coatings influence on opportunistic bacteria such as E. coli, S. epidermidis and S. aureus in addition to determining toxicity levels for mammalian cells (L929 Fibroblasts).

Materials and methods Material synthesis Microsphere synthesis. Amber glass with a basic starting composition of 79SiO2–13CaO–7Na2O–1Al2O3 (mol%) was used for this study. The glass was prepared by grinding glass and ball milling (1 h). The resulting frit was dried, ground and sieved to retrieve a glass particle sizes 100:1. Analysis area for each sample is 2–3 mm in diameter using a 100 mm beam. Spectra analysis was performed on CasaXPS (Casa Software Ltd.). Peak positions were calibrated through normalization of the C1s peak to 284.6 eV.

Microsphere solubility and ion release Ion release profiles. Microspheres (UC-S and Ag-S where n ¼ 3) were immersed in sterile Millipore de-ionized H2O for 1, 7, and 30 days. Approximately 0.1 m2 surface area of microspheres were submerged in 10 ml of de-ionized H2O and rotated slowly on an oscillating platform at 37 C. The ion release profile of each glass was measured using Inductively Coupled Plasma– Optical Emission Spectroscopy (ICP–OES) on a Perkin-Elmer Optima 8000 (Perkin Elmer, MA, USA). ICP–OES calibration standards for Ca, Si, Na,

Downloaded from jba.sagepub.com by guest on July 24, 2015

4

Journal of Biomaterials Applications 0(0)

and Ag ions were prepared from a stock solution on a gravimetric basis. pH Measurements. Changes in pH of solutions were monitored using a Corning 430 pH meter. Prior to testing, the pH meter was calibrated using pH buffer solution 4.00  0.02 and 7.00  0.02 (Fisher Scientific, Pittsburgh, PA, USA). Sample solutions were prepared by exposing 0.05 m2 surface area of each glass (UC-S and Ag-S, where n ¼ 3) in 5 ml de-ionized water. Measurements were recorded over t ¼ 0, 1, 7, and 30 days. De-ionized water was used as a control and was measured at each time period.

Antimicrobial testing The antibacterial activity of the UC-S and Ag-S was evaluated against a number of bacteria including E. coli (ATCC 8739, LB broth and agar), S. epidermidis (ATCC 14990, BHI broth and agar), and S. aureus (strain UAMS-1, TSB broth and agar). Agar diffusion analysis. LB agar and broth were used for the culturing E. coli, which were grown aerobically in an incubator at 37 C. Preparation of the agar discdiffusion plates for E. coli involved seeding agar plates with a sterile swab dipped in a 1/50 dilution of the appropriate 16 h culture of bacteria. UC-S and AgS samples were spread across the surface of the agar (0.05 m2) and the plates were then stored in an incubator for 24 h at 37 C. After 24 h, the plates were imaged with an Olympus IX20-UCB Optical Fluorescent Microscope. Extract preparation and liquid broth analysis. Liquid sample extracts were prepared (n ¼ 3, 0.05 m2 surface area) by placing the UC-S and Ag-S in 5 ml sterile de-ionized H2O and stored for 1, 7, and 30 days. After each incubation period, 100 ml of each extract was removed and added as 10% by volume to growing bacteria cultures. Then, 100 ml of sterile de-ionized water was used for addition to the growing bacteria controls. Each sample, in addition to a growing control and sterile broth, were incubated for 24 h in an incubator at 37 C. After 24 h, the samples were seeded into a 96well-plate and read at a wavelength of 590 nm.

Cell culture analysis The established cell line L-929 (American Type Culture collection CCL 1 fibroblast, NCTC clone 929) was used in this study as required by ISO10993 part 5.38 Cells were maintained on a regular feeding regime in a cell culture incubator at 37 C/5% CO2/95% air atmosphere. Cells were seeded into 24 well plates at a density

of 10,000 cells per well and incubated for 24 h prior to testing with liquid extracts. The culture media used was M199 media (Fisher Scientific, PA, USA) supplemented with 10% fetal bovine serum (Fisher Scientific, PA, USA) and 1% (2 mM) L-glutamine (Fisher Scientific, PA, USA). The cytotoxicity of liquid extracts was evaluated using the Methyl Tetrazolium (MTT) assay in 24 well plates. Liquid extracts (5, 15, 30 ml/ml) of sample were added into wells containing L-929 cells in culture medium (1 ml) in triplicate after 1, 7, and 30 days of incubation. Sample extracts (n ¼ 3) were placed in the plate wells and were tested after 24 h. Each of the prepared plates was incubated for 24 h at 37 C/5% CO2. The MTT assay was then added in an amount equal to 10% of the culture medium volume/well. The cultures were then re-incubated for a further 2 h (37 C/5% CO2). Next, the cultures were removed from the incubator and the resultant formazan crystals were dissolved by adding an amount of MTT Solubilization Solution (10% Triton x-100 in acidic isopropyl alcohol. (0.1 n HCI)) equal to the original culture medium volume. Once the crystals were fully dissolved, the absorbance was measured at a wavelength of 570 nm. Extracts (100 ml) of tissue culture water were used as controls, and cells were assumed to have metabolic activities of 100%.

Statistical analysis One-way analysis of variance (ANOVA) was employed to compare the antibacterial and cell culture results with respect to the control cell populations, time and concentration. Experimental comparisons were made in relation to (1) comparing the control bacteria cells to UC-S and Ag-S at 1, 7, and 30 days; (2) comparing the control fibroblast cells to UC-S and Ag-S as a function of concentration at each time period; and (c) comparing the cell viability in relation to liquid extract concentrations (5–15 ml/ml and 5–30 ml/ml) at each time period. Comparison of relevant means was performed using the post hoc Bonferroni test. Differences between groups was deemed significant when p  0.05.

Results Glass microspheres were coated with silver using the spin coating and heat treatment method presented in Figure 1. Glasses were initially characterized to investigate the changes in surface chemistry and morphology as a function of Ag deposition. X-ray diffraction (XRD) was employed to determine the presence of crystalline Ag on the surface of the microspheres, and is presented in Figure 2. Figure 2(a) determines the presence of Ag (Ref: 00-001-1167) on the surface of

Downloaded from jba.sagepub.com by guest on July 24, 2015

Haas et al.

5 Glass Frit -< 10µm

30

17.82

25 20

17.79

Weight (mg)

Collect in IPA Bath Dissolve AgNO3 in Ethyl Alcohol - 1 hr Add 2g spheres to Solution - 3 hrs

15 10

17.76

5 0

17.73

Dry for 24 hrs

Tg - 605°C

Heat to 300°C for hrs

–5

17.70 0

Figure 1. Microsphere synthesis and Ag coating.

* Ref: Silver (Ag) - 00-001-1167

200

400 600 Temperature (°C)

Temperature difference (µV)

Feed frit through O/A Flame

800

–10 1000

Figure 3. Differential thermal analysis and weight changes of initial glass frit.

(a)

Counts (a.u.)

Glass Microspheres 60X

(b)

*

Mean Min. Value 81 µm Mean Max. Value 187 µm. Mean 120 µm SD. 27 µm

* * * 10

20

30

40

50

60

70

80

Position [2° Theta]

Figure 2. X-ray diffraction of microspheres. (a) Ag-S and (b) UC-S.

the Ag-coated microspheres (Ag-S) with peaks present at 38, 44, 68 and 77 2 theta. XRD was also conducted on the uncoated spheres (UC-S) and was confirmed to be amorphous (Figure 2b). Additional peaks evident in figure 2 can be attributed to the sample holder. Differential thermal analysis/thermal gravimetric analysis (DTA/TGA) profiles are presented for the UC-S in Figure 3. DTA determined that the glass transition temperature (Tg) of UC-S was determined to be 605 C, while an additional thermal transition occurs at approximately 800 C. Weight changes, determined by TGA over the temperature range 20–1000 C, were minimal at approximately 0.4%. Optical microscopy was used to determine the morphology of the microspheres and to calculate the mean sphere diameter. Figure 4 presents images of UC-S at 4 and 60. In each case, spherical particles are produced within a size distribution of 81–187 mm in diameter. The mean sphere size was calculated at

100 mm

4X

Figure 4. Optical Microscopy of microspheres showing relevant statistics, single microsphere at 60 and distribution of microsphere size.

120  27 mm. Scanning electron microscopy (SEM) and energy dispersive X-ray (EDS) were used to further characterize the presence of the Ag coatings and is presented in Figure 5. Figure 5(a) shows the starting glass particles prior to forming microspheres. Particle size ranges from approximately 10 to 100 mm and the composition was determined to be predominantly Ca, Na, and Si, with traces of O, C, and Al from the sample holder. Figure 5(b) presents SEM and EDS of UC-S which presented the same composition as the starting glass; however, no small glass particles were present. Sphere size ranged from approximately 100 to 400 mm. Figure 5(c) presents the Ag-S microsphere which, in addition to the original elements in the glass composition, determined the presence of Ag. Higher resolution images present the Ag coating on the surface of the spheres. X-ray photoelectron spectroscopy was employed to confirm the presence of Ag on the Ag-S surface as well as investigate the valency of silver. Survey scans are presented in Figure 6(a) and (b) for UC-S and Ag-S, respectively, and high-resolution Ag

Downloaded from jba.sagepub.com by guest on July 24, 2015

6

Journal of Biomaterials Applications 0(0) (c)

Si

C O

Na

Ag Ag

AI

Ca

(b) Counts (a.u.)

Si O

Na C

Ca

AI

(a)

Si

O

C

Na

Ca AI

0

1.00

2.00

3.00

4.00

5.00

keV

Figure 5. Energy dispersive X-ray of (a) initial glass frit, (b) UC-S and (c) Ag-S and corresponding scanning electron microscopy images, frit and UC-S at 300 mm and Ag-S at 50 mm.

spectra is presented in Figure 6(c). Survey scans determined that UC-S contains all expected elements (Na, O, Ca, and Si) while Ag-S contains all the same elements in addition to Ag. Figure 6(c) shows slightly asymmetric 3 d peaks. In particular, the 3d5/2 peak is centered at approximately 367.9 eV while the 3d3/2 peak is centered at approximately 373.9 eV. Solubility studies include measuring the changes in ion release as a function of incubation time in an aqueous environment and determining the influence of Ag coating on ion release. In addition, changes in pH were recorded to monitor the influence of the microspheres solubility in a hydrated environment. Ion release profiles were determined using inductively coupled plasma optical emission spectroscopy (ICP-OES). Both the UC-S and Ag-S were evaluated after 1, 7, and 30 days in sterile DI-water and are presented in Figure 7(a) and (b). Figure 7(a) presents the UC-S where Naþ release was found to increase from 2 to 6 mg/l over 1–30 days. Si4þ release ranged from 5 to 16 mg/l over 1–30 days and Ca2þ release was low at 1–1.5 mg/l over the same time period. The Ag-S microspheres are presented in Figure 7(b) where Naþ release was much higher than the UC-S which ranged from 22 to 36 mg/l, Si4þ release ranged from 3 to 9 mg/l over 1–

30 days, Ca2þ ranged from 0.5 to 1 mg/l. Agþ release was also low and ranged from 0.2 to 1 mg/l over 1–30 days. The pH of the ion release extracts was recorded at each time period, i.e. 1, 7, and 30 days and is presented in Figure 8. With respect to UC-S the pH was found to increase from 8.9 to 9.0 over 1–7 days, but reduced to 7.8 after 30 days. The Ag-S extracts presented a similar trend where the pH increased from 7.9 to 8.5, but reduced to 7.3 after 30 days. Initial antibacterial testing in E. coli is presented in Figure 9 for both the UC-S and Ag-S. It is evident from Figure 9(a) that UC-S did not present any antibacterial effects as E. coli inhibition was not evident; however, Figure 9(b) shows clear inhibition of E. coli with the addition of Ag-S to the culture. Higher resolution optical microscope imaging revealed that no bacterial colonies were in contact with any of the Ag-S samples. In order to quantitatively determine the antibacterial efficacy of the Ag coating, UC-S, and Ag-S were tested in a number of relevant bacteria (E. coli, S. epidermidis, and S. aureus) using liquid extracts in broth culture. Liquid broth testing for UC-S and Ag-S (Figure 10(a) to (c)) was determined over 1, 7, and 30 days and were compared to a control healthy growing bacterial culture. E. coli testing is presented in Figure 10(a), and with

Downloaded from jba.sagepub.com by guest on July 24, 2015

7

Si2p Na2s O2s

Si2s

C1s

Na KLL Ca2s

O KLL O KLL

O1s

Na1s Na1s

1200

1000

800 600 400 Binding energy (eV)

200

Si2p Ag4d O2s Ag4p

Si2s

C1s

Ag3d Ca2p

Na KLL Ca2s

(b) Ag3p1 Ag3p2

Intensity (a.u.)

(a)

Ca2p

O1s

Haas et al.

0

Intensity (a.u.)

(c)

367.9 eV 373.9 eV

376

374

372 370 368 Binding energy (eV)

366

364

Figure 6. X-Ray photoelectron spectroscopy survey scan of (a) UC-S and (b) Ag-S and (c) high-resolution Ag scan.

respect to Ag-S, the bacterial populations reduced to 69%, 53%, and 9% after 1, 7, and 30 days, respectively. The UC-S was found to increase the bacteria populations to 159%, 156%, and 141% after 1, 7, and 30 days. Regarding S. epidermidis (Figure 10b), there was a clear and rapid decrease in bacteria cell viability which reduced to 0.5%, 1.5%, and 7.4% after 1, 7, and 30 days, respectively. The UC-S experienced a similar trend to the E. coli testing, where an increase in bacterial viability was experience at 141%, 132%, and 125% after 1, 7, and 30 days. Testing in S. aureus is presented in Figure 10(c) and shows a less prominent reduction in bacteria cell viability than either E. coli or S. epidermidis for Ag-S at 37%, 56%, and 40% after 1, 7, and 30 days, respectively. In this particular instance, UC-S resulted in a reduction in bacterial viability which reduced to 86%, 88%, and 83%, respectively. Cell culture testing was conducted using L-929 Fibroblasts to determine if the ionic dissolution products and concentrations positively or negatively influence cell growth and metabolism. Ion extracts for UC-S and Ag-S were added at concentrations of 5 mg/ml, 15 mg/ml, and 30 mg/ml for 1-, 7-, and 30-day samples.

Cell viability data is presented in Figure 11(a) for the UC-S samples. At 5 mg/ml concentrations, the UC-S produced relatively little changes over 1, 7, and 30 days at 104%, 112%, and 103%, respectively. At 15 mg/ml, the cell viability began to reduce to 94%, 87%, and 104% over the same time period, while at 30 mg/ml the cell viability further reduced to 86%, 74%, and 82% at 1, 7, and 30 days. The Ag-S samples are presented in Figure 11(b) and presented cell viability of 116%, 104%, and 115% for the 5 mg/ml after 1, 7, and 30 days. Concentrations of 15 mg/ml presented viability of 86%, 88%, and 94% over the same time period and when concentrations reached 30 mg/ml the cell viability reduced to 89%, 79%, and 83% after 1, 7, and 30 days, respectively.

Discussion Regarding this study, Ag coatings were applied to glass microsphere substrates to determine the effect that Ag coatings have on the glass solubility and the subsequent therapeutic effects. XRD was initially used to confirm the presence of the crystalline Ag coating on the surface of the microspheres, and that the surface is free of any additional crystal phases such as AgNO3. Optical microscopy and scanning electron microscopy (SEM)/ energy dispersive microscopy (EDX) were employed to examine the morphology and composition of UC-S and Ag-S. It is evident that there is no change in glass composition as microspheres are produced from the glass frit, which occurs due to agglomeration of multiple frit particles (90 mm). However, after Ag coating, the addition of Ag was detected with EDX in addition to the presence of an irregular granular surface as presented by SEM imaging. XPS also confirmed the presence of the Ag on the glass surface of Ag-S. The 3d5/2 approximate peak position (367.9 eV) as well as the approximate difference in binding energy between the Ag3d3/2 and Ag3d5/2 (4.0 eV) suggests that the silver on the surface of the microspheres are predominantly metallic silver.39–41 However, the relatively small difference in binding energies of the various common Ag states (e.g. ionic vs. metallic), as well as the slightly asymmetric peak shape, suggests that there is likely some contribution from ionic silver.42–45 Ion release studies were conducted to determine the effect that Ag coating has on the solubility of the glass microspheres. The Ag coating greatly increased the rate of Naþ release at 1, 7, and 30 days (23–45 mg/l) compared to UC-S where levels increased from 2–6 mg/l over the same time period. Naþ is known to be an important ion in the dissolution of glass-based biomaterials as it promotes de-polymerization of Si–O–Si bonds within the glass network, which in turn promotes

Downloaded from jba.sagepub.com by guest on July 24, 2015

8

Journal of Biomaterials Applications 0(0) (a) 16 Ion release (mg/L)

14 12

Na Si Ca

10 8 6 4 2 0 Day 1

Day 7

(b) 40

Day 30 10

Na

Si Ca

Ion release (mg/L)

Ion release (mg/L)

8 30

20

10

Ag

6

4

2

0

0 Day 1

Day 7

Day 30

Day 1

Day 7

Day 30

Figure 7. Ion Release profiles of (a) uncoated spheres and (b) Ag-coated spheres over 1, 7, and 30 days.

9.5

UC-S Ag-S

9.13

8.99

9.0 8.5 8.57

pH

the ion exchange process.46 An additional observation is that the Si4þ levels presented a slight decrease with the application of Ag coating. Si4þ levels for UC-S ranged from 4 to 14 mg/l and presented an increase with respect to time. Post Ag coating, the rate of Si4þ release slightly reduced and ranged from 2 to 8 mg/l, but did also present a time-dependant increase in release. This is likely due to the presence of Ag restricting the diffusion of soluble Si4þ species from the glass surface. Si4þ release and solubility are known to be essential for the precipitation of Ca and P from physiological fluids on glass and ceramic biomaterials. Aqueous Si(OH)4 is also known to stimulate collagen I formation and osteoblastic differentiation in vivo.47 Si4þ release from commercial Bioglass ranges from 5 to 45 mg/l over 1–30 days,48 and the levels presented here are within this distribution. Ca2þ levels did not present any significant change with the Ag coating and were less than 1.5 mg/l for UC-S and Ag-S. Ca2þ release from Bioglass is also quite low and ranges from 7.5 to 16 mg/l (1–30 days).48 Ca2þ is also known to be a critical component of glassy biomaterials as it is cited to favor cellular processes such as osteoblast proliferation, differentiation and extracellular (ECM) mineralization, in addition to activating Ca-sensing receptors in osteoblast cells which increases the expression of growth factors.47 A review by Hoppe et al. on ionic dissolution products from glasses cites that low (3–7 mg/l) and

8.0

7.78 7.95

7.5 7.34

7.0 1 Day

7 Day

30 Day

Figure 8. pH of UC-S and Ag-S tested over 1, 7, and 30 days.

medium (10–14 mg/l) Ca2þ concentrations are suitable for osteoblast proliferation, differentiation, and extracellular matrix formation, whereas higher Ca2þ concentrations (18 mg/l) can be cytotoxic.47 In addition, Ca2þ is also essential for encouraging precipitation of bioactive calcium phosphate surface layer in vivo.[3, 46] Ag release from Ag-S was lower than each of the other ions tested (0.2–1 mg/l). The Ag release rates were relatively low despite the Ag being deposited as a surface coating on the microspheres. This suggests that Ag is tightly bound to the glassy surface, which may be incorporated, by ion exchange processes during heat treatment, into sites occupied by Naþ. This may

Downloaded from jba.sagepub.com by guest on July 24, 2015

Haas et al.

9

Figure 9. Preliminary optical evaluation of (a) UC-S and (b) Ag-S in E. coli at 4 (above image) and 10 (below image).

(b)

E. coli cell viability (%)

160 120 80 40 0 Culture 1D

7D 30D 1D Ag-S

7D 30D UC-S

(c) 160 160

S. aureus cell viability (%)

S. epidermidis cell viability (%)

(a)

120 80 40 0 Culture 1D

120

80

40

0 7D 30D 1D Ag-S

7D 30D UC-S

Culture 1D

7D 30D 1D Ag-S

7D 30D UC-S

Figure 10. Antibacterial testing of UC-S and Ag-S in (a) E. coli, (b) S. epidermidis and (c) S. aureus over 1, 7, and 30 days.

explain the relatively high Naþ release attributed to Ag-S compared to UC-S. It has also been suggested that the Ag-O bonds are more covalent than Ca-O; however, in relation to this study, there was no significant changes evident in Ca release in the UC-S glass compared to Ag-S.49 Additional studies by Balamurugan et al. on the dissolution of Ag from SiO2–P2O5–CaO–Ag2O glasses also report slow dissolution rates compared to other glass components which was attributed to Ag being chelated by the silicate network.30 Regarding the influence that the Ag coating has

on pH, it can be observed that Ag-S experienced a lower pH at each time period (1, 7, and 30 days) compared to UC-S. However, after 30 days, both the Ag-S and UC-S approached neutral. Antibacterial testing was conducted to determine if the Ag coating on the glass microspheres presented bactericidal properties amongst a range of bacteria that represents both gram ve (E. coli) and gram þve (S. epidermidis, S. aureus) species and are common opportunistic microbes responsible for septic complications. Regarding this study, Ag-S did not significantly

Downloaded from jba.sagepub.com by guest on July 24, 2015

10

Journal of Biomaterials Applications 0(0) (b) 140

120

1 Day 7 Day 30 Day

120

Cell viability (%)

Control

Cell viability (%)

(a) 140

100

80

60

100

80

60 0

5

10

15

20

25

30

35

0

5

10

Concentration (µg/mL)

15

20

25

30

35

Concentration (µg/mL)

Figure 11. Cell viability analysis of (a) UC-S and (b) Ag-S glass liquid extracts tested in L929 Mouse Fibroblasts.

decrease E. coli numbers after 1 and 7 days (p ¼ 0.848, 0.152, Table 1); however, after 30 days the reduction proved significant (p ¼ 0.000). An additional unexpected result related to the UC-S was presented where it caused a significant increase in E. coli numbers compared to the control E. coli population (p ¼ 0.000). This may be due to E. coli using the ions released from the glass to metabolize and increase their cell numbers. Regarding S. epidermidis, there was a significant decrease in bacteria numbers at 1, 7, and 30 days (p ¼ 0.000) with the addition of Ag-S. There was also a significant increase in S. epidermidis numbers at 1, 7, and 30 days (p ¼ 0.000) with the addition of the UC-S extracts. Testing in S. aureus also revealed a significant decrease in cell numbers (p ¼ 0.000) with the addition of Ag-S; however, in this instance the UC-S also showed a slight reduction in cell numbers that proved to be significant (p ¼ 0.000) compared to the growing control. The difference in antibacterial response to the Ag coating is likely due to differences in the cell walls of each bacterium. Gramve E. coli present a relatively complex cell wall, and an outer lipopolysaccharide (LPS) membrane, which acts as a selectively permeable barrier and a thin peptidoglycan layer (2 nm).50 In contrast, gramþve (S. epidermidis, S. aureus) bacteria have a simple cell wall consisting of a thick peptidoglycan layer (20–80 nm) and lack an outer membrane. In general, due to this thick peptidoglycan layer, the walls of gram þve bacteria are more resistant to antimicrobial compounds than that of gram ve bacteria.50 This was demonstrated in studies conducted by Yang et al. on the effect of Ag ion on the cell walls of E. coli and S. epidermidis. This study determined that the difference in cell wall structure resulted in differences in bacterial response to Ag, where the thick peptidoglycan walls of S. epidermidis proved more difficult to breakdown, in contrast to what was observed in this study.50 The mechanism of action of Ag is cited to include a number of processes such as uncoupling of respiratory electron

Table 1. Means comparison between the control bacteria populations and each bacteria (E. coli, S. epidermidis, and S. aureus) at each time period (1, 7, and 30 days) where p  0.05. Control

Sample Time

Control UC-S bacteria cells Ag-S

1 Day 7 Days 30 Days 1 Day 7 Days 30 Days

E. coli S. epidermidis S. aureus 0.848 0.152 0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000 0.000 0.000

Table 2. Means comparison between the control L929 Fibroblast populations and liquid extract concentration (5%, 15% and 30%) at each time period (1, 7, and 30 days) where p  0.05. Control

Sample

Concentration

1 Day

7 Days

30 Days

Control cells

UC-S

5 mg/ml 15 mg/ml 30 mg/ml 5 mg/ml 15 mg/ml 30 mg/ml

1.000 1.000 1.000 1.000 1.000 1.000

1.000 0.842 0.157 1.000 1.000 0.157

1.000 1.000 0.589 1.000 1.000 0.386

Ag-S

Table 3. Means comparison between cell culture liquid extract concentration (5%, 15%, and 30%) at each time period (1, 7, and 30 days) where p  0.05. Sample

Concentration

UC-S

5 5 5 5

Ag-S

v v v v

15 mg/ml 30 mg/ml 15 mg/ml 30 mg/ml

Downloaded from jba.sagepub.com by guest on July 24, 2015

1 Day

7 Days

30 Days

0.908 0.307 0.032 0.050

0.416 0.125 1.000 0.467

1.000 0.392 0.576 0.179

Haas et al.

11

transport processes, collapse of the proton-motive forces across the cytoplasmic membrane, and interference with membrane permeability, membrane-bound enzymes and proteins.30,50 In addition to testing the effect of Ag coating on the antibacterial properties, UC-S and Ag-S were tested using L-929 Fibroblasts in order to determine the concentration at which Ag-levels become toxic to the cells. UC-S and Ag-S liquid extracts were tested at 5, 15, and 30 mg/ml concentrations using 1, 7, and 30 days samples. Regarding UC-S testing in L929 Fibroblasts, there was no significant difference (p ¼ 0.157–1.000, Table 2) in cell viability with respect to concentration (5, 15, 30 mg/ml) or with respect to maturation. However, cell numbers were above the control cell population (100%) at each time period with the 5 mg/ml and reduced as the concentration increased to 15 and 30 mg/ml by 15–25% below the control cells. The Ag-S particles also did not experience any significant change (p ¼ 0.157–1.000) in cell viability with respect to both extract concentration or maturation time. However, the Ag-S was observed to increase cell viability by 16% compared to the control cells at the lower 5 mg/ml concentration. This may be due to the fact that the incorporation of Ag into the glass surface results in an overall increase in Naþ release. Naþ is known to be a critical component for cell metabolism and growth, and the increased release rate is likely facilitating greater cell proliferation. The release rate of Ag is also quite low compared to other ions from the glass which limits its toxicity. Additionally, when comparing the effect of cell viability based on concentration at each time period, there was no significant difference attributed to UC-S (p ¼ 0.125– 1.000, Table 3) at any time period. Ag-S experienced a significant change between 5 and 15 mg/ml (p ¼ 0.032) at 1 day; however, the remaining comparisons did not prove to be significant (p ¼ 0.050–1.000). In summary, glass microspheres were used as substrates for investigating the influence that Ag coatings have on the glass solubility and the resulting cell compatibility and antibacterial activity. Ag coatings were found to increase the ion release rate, particularly with respect to Naþ, with an associated reduction in solution pH. The Ag coating also imparted much greater antibacterial efficacy when tested in E. coli, S. epidermidis and S. aureus compared to the uncoated glasses (UC-S), which in some cases increased bacterial cell numbers. Cell culture testing in L929 Fibroblasts with differing concentrations resulted in no significant difference in cell viability. This suggests that opportunistic microbes, such as the ones tested as part of this study, are far more susceptible to Ag than are mammalian cells. This is a positive result as Ag can be easily incorporated into medical materials in place of antibiotics. Future work on Ag coatings on glass will include evaluating the effect Ag coatings have

on the precipitation of calcium phosphate in Simulated Body Fluid (SBF) and the subsequent crystallization of hydroxyapatite in addition to determining the toxicity limits with MC3T3 Osteoblast cells. Declaration of conflicting interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References 1. Hench LL. The story of bioglass. J Mat SciMat Med 2006; 17: 967–978. 2. Hench L, Peitl O and Zanotto ED. Highly bioactive P2O5-Na2O-CaO-SiO2 glass ceramics. J Non-Crys Sol 2001; 292: 115–126. 3. Kokubo T and Takadama H. How useful is SBF in predicting in vivo bone bioactivity. Biomat 2006; 27: 2907–2915. 4. Jones JR. Review of bioactive glass: from Hench to hybrids. Acta Biomat 2013; 9: 4457–4486. 5. Wren AW, Boyd D and Towler MR. The processing, mechanical properties and bioactivity of strontium based glass polyalkenoate cements. J Mat Sci Mat Med 2005; 19: 1737–1743. 6. Wren AW, Cummins NM, Laffir FR, et al. The bioactivity and ion release of titanium-containing glass polyalkenoate cements for medical applications. J Mat Sci Mat Med 2011; 22: 19–28. 7. Chen QZ, Thompson ID and Boccaccini AR. 45S5 BioglassÕ -derived glass-ceramic scaffolds for bone tissue engineering. Biomat 2006; 27: 2414–2425. 8. Haimi S, Gorianc G, Moimas L, et al. Characterization of zinc-releasing three-dimensional bioactive glass scaffolds and their effect on human adipose stem cell proliferation and osteogenic differentiation. Acta Biomater 2009; 5: 3122–3131. 9. Vargas GE, Mesones RV, Bretcanu O, et al. Biocompatibility and bone mineralization potential of 45S5 BioglassÕ -derived glass-ceramic scaffolds in chick embryos. Acta Biomater 2009; 5: 374–380. 10. Anderson JH, Goldberg JA, Bessent RG, et al. Glass yttrium-90 microspheres for patients with colorectal liver metastases. Radiother Oncol 1992; 25: 137–139. 11. da Costa Guimara˜es C, Moralles M and Roberto Martinelli J. Monte Carlo simulation of liver cancer treatment with 166Ho-loaded glass microspheres. Rad Phys Chem 2014; 95: 185–187. 12. Bortot MB, Prastalo S and Prado M. Production and characterization of glass microspheres for hepatic cancer treatment. Proc Mat Sci 2012; 1: 351–358. 13. Valappil S, Ready D, Abou Neel EA, et al. Antimicrobial gallium-doped phosphate-based glasses. Adv Func Mat 2008; 18: 732–741.

Downloaded from jba.sagepub.com by guest on July 24, 2015

12

Journal of Biomaterials Applications 0(0)

14. Valappil SP, Owens GJ, Miles EJ, et al. Effect of gallium on growth of streptococcus mutans NCTC 10449 and dental tissues. Caries Res 2014; 48: 137–146. 15. Sahdev R, Ansari T, Higham SM, et al. Potential use of gallium-doped phosphate-based glass material for periodontitis treatment. J Biomat App 2015; DOI: 10.1177/ 0885328215571952. 16. Verne E, Di Nunzio S, Bosetti M, et al. Surface characterization of silver-doped bioactive glass. Biomaterials 2005; 26: 5111–5119. 17. Witte W, Cuny C, Klare I, et al. Emergence and spread of antibiotic resistant gram-positive bacterial pathogens. Int J Med Microbio 2008; 298(5–6): 365–377. 18. Squire MW, Ludwig BJ, Thompson JR, et al. Premixed antibiotic bone cement. J Arthro 2008; 23: 110–113. 19. Rafiq I, Gambhir AK, Wroblewski BM, et al. The microbiology of infected hip arthroplasty. Int Orthopaed 2006; 30: 532–535. 20. Zilberman M and Elsner JJ. Antibiotic-eluting medical devices for various applications. J Contrl Rel 2008; 130: 202–215. 21. Husheng J, Wensheng H, Liqiao W, et al. The structures and antibacterial properties of nano-SiO2 supported silver/zinc–silver materials. Dent Mater 2008; 24: 244–249. 22. Lee S-B, Otgonbayar U, Lee J-H, et al. Silver ion-exchanged sodium titanate and resulting effect on antibacterial efficacy. Sur Coat Tech 2010; 205: S172–S176. 23. Valappil SP, Knowles JC and Wilson M. Effect of silverdoped phosphate-based glasses on bacterial biofilm growth. App Environ Micro 2008; 74: 5228–5230. 24. Harris LG and Richards RG. Staphlococci and implant surfaces: a review. Injury Int J Care Injured 2006; 37: S3–S14. 25. Blaker JJ, Nazhat SN and Boccaccini AR. Development and characterization of silver-doped bioactive glass coated sutures for tissue engineering and wound healing applications. Biomaterials 2004; 25: 1319–1329. 26. Kawashita M, Tsuneyama S, Miyaji F, et al. Antibacterial silver-containing silica glass prepared by solgel method. Biomaterials 2000; 21: 393–398. 27. Ravindra S, Murali Mohan Y, Narayana Reddy N, et al. Fabrication of antibacterial cotton fibres loaded with silver nanoparticles via ‘‘Green Approach’’. Coll Sur A: Physicochem Eng Asp 2010; 367: 31–40. 28. Furr JR, Russell AD, Turner TD, et al. Antibacterial activity of actisorb plus, actisorb and silver nitrate. J Hosp Infect 1994; 27: 201–208. 29. Lee SM, Lee BS, Byun TG, et al. Preparation and antibacterial activity of silver-doped organic–inorganic hybrid coatings on glass substrates. Coll Sur A: Physicochem Eng Asp 2010; 355: 167–171. 30. Balamurugan A, Balossier G, Laurent-Maquin D, et al. An in vitro biological and anti-bacterial study on a sol–gel derived silver-incorporated bioglass system. Dent Mater 2008; 24: 1343–1351. 31. Di Nunzio S, Vitale Brovarone C, Spriano S, et al. Silver containing bioactive glasses prepared by molten salt ionexchange. J Euro Cer Soc 2004; 24: 2935–2942.

32. Xu Y, Cheng J, Zheng W, et al. Study on the preparation and properties of silver-doped borosilicate antibacterial glass. J Non-Cryst Solid 2008; 354: 1341–1345. 33. Shrivastava P, Dalai S, Sudera P, et al. Hollow glass microspheres as potential adjunct with orthopaedic metal implants. Microelectronic Eng 2014; 126: 103–106. 34. Dalai S, Vijayalakshmi S, Sharma P, et al. Magnesium and iron loaded hollow glass microspheres (HGMs) for hydrogen storage. Int J Hyd Energy 2014; 39: 16451–16458. 35. Dalai S, Vijayalakshmi S, Shrivastava P, et al. Preparation and characterization of hollow glass microspheres (HGMs) for hydrogen storage using urea as a blowing agent. Microelectronic Eng 2014; 126: 65–70. 36. Sorge M, Bean T, Woodland T, et al. Investigating the use of porous, hollow glass microspheres in positive lead acid battery plates. J Pow Sour 2014; 266: 496–511. 37. Xu J, Yang H, Yu Q, et al. Synthesis and characterization of hollow glass microspheres coated by SnO2 nanoparticles. Mat Lett 2007; 61: 1424–1428. 38. International Standard 10993-5. Biological evaluation of medical devices Part 5: tests for in vitro cytotoxicity. Geneve, Switzerland: International Standard Organisation, Case Postale 56: CH-1211, 1999. 39. Bai J, Li M, Wang S, et al. Electrospinning method for the preparation of silver chloride nanoparticles in PVP nanofiber. App Surf Sci 2008; 254: 4520–4523. 40. Gao XY, Wang SY, Li J, et al. Study of structure and optical properties of silver oxide films by ellipsometry, XRD and XPS Methods. Thin Sol Films 2004; 455: 438–442. 41. Kuo YL, Chen HW and Ku Y. Analysis of silver particles incorporated on TiO2 coatings for the photodecomposition of O-Cresol. Thin Solid Film 2007; 515: 3461–3468. 42. Bastidas DM, Cano E, Gonzalez AG, et al. An XPS study of tarnishing of a gold mask from a pre-columbian culture. Corrosion Sci 2008; 50: 1785–1788. 43. Boronin AI, Koscheev SV, Murzakhmetov KT, et al. Associative oxygen species on the oxidized silver surface formed under O2 microwave excitation. Appl Surf Sci 2000; 165: 9–14. 44. Arabatzis IM, Stergiopoulos T, Bernard MC, et al. Silver-modified titanium dioxide thin films for efficient photodegredation of methyl orange. Appl Catalysis B: Environment 2003; 42: 187–201. 45. Tousley ME, Wren AW, Towler MR, et al. Processing, characterization and bactericidal activity of undoped and silver doped vanadium oxides. Mater Chem Phys 2012; 137: 596–603. 46. Serra J, Gonzalez P, Liste S, et al. Influence of the nonbridging oxygen groups on the bioactivity of silicate glasses. J Mater Sci Mater Med 2002; 13: 1221–1225. 47. Hoppe A, Galdal NS and Boccaccini AR. A review of the biological response to ionic dissolution products from bioactive glasses and glass-ceramics. Biomaterials 32: 2757–2774. 48. Murphy S, Wren AW, Towler MR, et al. The effect of ionic dissolution products of Ca–Sr–Na–Zn–Si bioactive glass on in vitro cytocompatibility. J Mater SciMater Med 2010; 10: 2827–2834.

Downloaded from jba.sagepub.com by guest on July 24, 2015

Haas et al.

13

49. El-Kady AM, Ali AF, Rizk RA, et al. Synthesis, characterization and microbiological response of silver doped bioactive glass nanoparticles. Cer Inter 2012; 38: 177–188.

50. Yang X, Yang W, Wang Q, et al. Atomic force microscopy investigation of the characteristic effects of silver ions on Escherichia coli and Staphylococcus epidermidis. Talanta 2010; 81: 1508–1512.

Downloaded from jba.sagepub.com by guest on July 24, 2015