| Sign In to gain access to subscriptions and/or personal tools. |
Antimicrobial Effect of Nanometric Bioactive Glass 45S5
1 Institute of Oral Microbiology and Preventive Dentistry, University of Basel Center of Dental Medicine, Switzerland; Correspondence: * corresponding author, matthias.zehnder{at}zzmk.uzh.ch
Most recent advances in nanomaterials fabrication have given access to complex materials such as SiO2-Na2O-CaO-P2O5 bioactive glasses in the form of amorphous nanoparticles of 20- to 60-nm size. The clinically interesting antimicrobial properties of commercially available, micron-sized bioactive glass 45S5 have been attributed to the continuous liberation of alkaline species during application. Here, we tested the hypothesis that, based on its more than ten-fold higher specific surface area, nanometric bioactive glass releases more alkaline species, and consequently displays a stronger antimicrobial effect, than the currently applied micron-sized material. Ionic dissolution profiles were monitored in simulated body fluid. Antimicrobial efficacy was assessed against clinical isolates of enterococci from persisting root canal infections. The shift from micron- to nano-sized treatment materials afforded a ten-fold increase in silica release and solution pH elevation by more than three units. Furthermore, the killing efficacy was substantially higher with the new material against all tested strains.
Key Words: bioactive glass • nanotechnology • enterococci
Bioactive glasses of the SiO2-Na2O-CaO-P2O5 system have some antimicrobial activity when suspended in aqueous solutions via the release of their ionic compounds over time (Zehnder et al., 2006a). The release of Na+ and Ca2+ ions from, and the incorporation of H3O+ protons into, the corroding glass result in a high-pH environment in closed systems (Sepulveda et al., 2002), which is not well-tolerated by microbiota (Allan et al., 2001). In addition, the release of silica has been linked to the antibacterial bioactive glass effect (Zehnder et al., 2006a). In contrast to commonly used disinfectants in dentistry, silica-containing bioactive glasses induce dentin mineralization (Forsback et al., 2004), and thus are potentially interesting materials for the treatment of demineralized and infected dentin found in deep caries lesions and necrotic root canals. Conventional bioactive glasses show some promise as dentin disinfectants (Zehnder et al., 2004); however, their antibacterial efficacy in human teeth is still inferior to that of calcium hydroxide, the gold standard material (Zehnder et al., 2006b). Attempts have been made to spike bioactive glass with silver to increase its antimicrobial efficacy (Bellantone et al., 2002). The consequences of this procedure on glass biocompatibility, however, remain questionable. An alternative approach may be simply to decrease glass particle size, and thus increase the active exchange surface of glass and surrounding liquid. This substantially increases ionic release in suspension (Sepulveda et al., 2002), and may thus result in enhanced antimicrobial efficacy. Recently, flame spray synthesis has been extended from metal oxides to more complex materials such as salts (Grass and Stark, 2005; Loher et al., 2005), and even more complex systems such as the five-element-containing bioactive glasses mentioned above (Brunner et al., 2006). Glasses prepared by flame spray synthesis consist of x-ray diffraction amorphous nanoparticles with a primary particle size of 20–60 nm (specific surface area: 70 m2/g). In contrast, the currently marketed bioactive glasses for dental applications are manufactured by the melting of corresponding oxides followed by grinding of the cooled product. Depending on the manufacturer, the resulting shards are sieved to obtain particles in the range of 10, 50, or several hundred 3m, resulting in a surface area in the range of a few m2/g (Jones et al., 2001). The aim of the current study was to test the hypothesis that, based on its high specific surface area, nanometric bioactive glass 45S5 releases more alkaline species, and consequently has a better antimicrobial effect, than the currently available melt-derived, micron-sized 45S5 bioactive glass. Ionic dissolution of nano- and micron-sized bioactive glass in simulated body fluid was monitored. Furthermore, the efficacy of these materials on bacteria associated with failed root canal treatments (Molander et al., 1998) was tested.
Preparation of Test and Reference Materials Conventional bioactive glass 45S5 (PerioglassTM, US Biomaterials Corp., Alachua, FL, USA) was obtained from a commercial source. According to the manufacturer, this material consists of 45 wt% SiO2, 6 wt% P2O5, 24.5 wt% CaO, and 24.5% Na2O. Bioactive glass 45S5 was prepared from metal organic precursors as described previously (Brunner et al., 2006). Nanoparticulate zirconium oxide was produced accordingly as a reference material, for assessment of the impact of an inert nanometric substance on bacterial viability. The particle size of the nanometric materials under investigation was determined in a transmission electron microscope (CM30 ST, LaB6 cathode, Philips Electron Optics, Eindhofen, the Netherlands) operated at 300 kV at a point resolution of 0.4 nm. Mean particle size for the bioactive glass and zirconium oxide was calculated based on the diameters of 200 randomly selected particles per material. To study the appearance of the 2 different materials, we carried out scanning electron microscopic analysis on a LEO 1530 Gemini (Zeiss, Oberkochen, Germany) after sputtering the samples with ~ 4 nm of platinum. Chemical composition of the nanoparticulate bioglass was assessed by laser ablation inductively coupled plasma mass spectrometry.
Chemical Assessment of Glass Dissolution
Direct Exposure Tests
As observed by scanning electron microscopy, the nanoparticulate bioglass was spherically shaped and highly agglomerated, while the conventional melt-derived glass was in the form of sharp-edged shards (Fig. 1  ). According to laser ablation inductively coupled plasma mass spectrometry, the composition of the nanometric bioactive glass was 44.7 wt% SiO2, 4.9 wt% P2O5, 27.6 wt% CaO, and 22.8 % Na2O. Transmission electron microscopy confirmed particle size of the nanometric materials to be 30 ± 7.8 nm and 8.0 ± 5.8 nm for bioactive glass and zirconium oxide, respectively (N = 200).
At the same solid-to-liquid ratio (wt/vol), the nanoparticulate 45S5 released approximately 10-fold more silica into the simulated body fluid than the conventional bioglass (Fig. 2 ). In contrast, 10-minute conventional bioglass supernatants contained substantially more calcium, 158.5 ± 8.4 ppm vs. 31.2 ± 5.8 ppm. Nanometric bioglass supernatants had a pH of 11.7, compared with 8.3 measured in counterparts obtained from conventional bioglass suspensions. Nanometric zirconium oxide supernatants were identical to the untreated simulated body fluid with respect to all of the above parameters, i.e., calcium and silicon concentrations and pH.
Bacterial viability was not affected by the presence of nanoparticulate zirconium oxide (Fig. 3 ). At the relatively low solid-to-liquid ratio used in the current study, the conventional bioglass did not show any antibacterial effect either. In contrast, nanoparticulate 45S5 substantially decreased the viability of the enterococci under investigation over time. Clinical isolates behaved similarly to the type strain; their total counts were all reduced by more than 99% after 100 min of incubation (Table).
In confirmation of the hypothesis of this investigation, the current study showed a substantially higher release of alkaline species and better antibacterial efficacy of bioactive glass 45S5 in nanoparticulate form, compared with the conventional preparation. To our knowledge, this is the first report on the antibacterial effects of a nanometric bioactive glass. The current in vitro conditions involved solution saturation without replenishment of the liquid phase. They thus differed from in vivo conditions in areas reached by blood circulation, but may reflect the entombed situation in the necrotic and infected root canal space or the infected area in dentin left behind during an indirect pulp-capping procedure (Langeland, 1987). Enterococci were used for this study, since they are more resistant against alkaline biocides than most other microbiota (Evans et al., 2002). The experiments were performed in simulated body fluid solution, because this solution reflects the ionic composition of blood plasma and is commonly used for the study of corrosion of bioactive materials. However, under the current study conditions, it actually did not matter whether the bacteria were suspended in simulated body fluid or unbuffered saline solution (data not shown). Consequently, specific interactions of the nanometric bioactive glass with components of the simulated body fluid did not influence the results. Rather, the effects appeared to be transmitted by the species liberated from the glass. The antibacterial effect of nanoparticulate bioactive glass appears to be directly linked to its high surface area and thus the resulting release of ionic components in solution. However, the low concentration of calcium in nanometric 45S5 supernatants suggests that calcium and phosphate released from this glass preparation precipitate immediately (Brunner et al., 2006). The release of silica should eventually result in the transformation of glass nanoparticles into inert Ca-P shells (Radin et al., 2000). Chemistry dictates that the total silica content is linked to the pH of the solution over a series of chemical acid-base equilibria involving the successive de/re-protonation of the silica ion series (SiO4)4–, (HSiO4)3–, (H2SiO4)2–, etc. Therefore, the silica concentration is linked to basicity. Since silica solutions themselves have been suggested for food disinfection (Weber et al., 2004), an ideal antimicrobial agent would combine high silica delivery, high pH, and a high alkaline buffer capacity. The gold standard Ca(OH)2 is a potent topical dentin disinfectant, because of a combination of basicity and capacity (typical use as a > 10 wt% slurry; this capacity is over 10 times higher than that of a corresponding NaOH solution). As measured in pilot experiments, the pH of 11.7 measured in supernatants of 1:10 supernatants of nanometric bioactive glass 45S5 in simulated body fluid after 10 min was comparable with a corresponding value of 12.6 obtained with calcium hydroxide, just as calcium hydroxide bioactive glass slowly but steadily releases alkaline species, which interact with its environment (in our case, the microbiota). Hence, there is a cumulative effect over time. This is also the reason why a saturated calcium hydroxide solution is not very antiseptic, which is in stark contrast to a calcium hydroxide suspension (Stevens and Grossman, 1983). The physical behavior of bioactive glass follows the same mechanism. However, apart from the antibacterial effect and unlike calcium hydroxide, the nanometric 45S5 bioactive glass described here promotes Ca/P precipitation in its environment (Brunner et al., 2006), and has a substantial potential to (re)mineralize dentin (unpublished observations). In theory, there are two possibilities for increasing the surface area of bioactive glasses, and thus rendering them more active: the direct preparation of glass nanoparticles in a high-temperature environment, as used in this study; or glass production via a sol-gel process (Jones et al., 2001). The direct approach via flame spray synthesis offers clear advantages in terms of homogeneity and particle size of the resulting material. Sol-gel-derived glasses have a high specific surface area, but are not nanometric. Consequently, they are unable to enter dentinal tubules, which have a diameter in the range of 0.5 µm to 2.5 µm (Garberoglio and Brannström, 1976), a feature that would appear advantageous for a material to be used as a dentin disinfectant. Further studies should focus on the efficacy of the nanometric bioglass material in situ, i.e., in carious teeth or counterparts with infected root canals.
This research was supported by funds of the Department of Preventive Dentistry, Periodontology, and Cariology, University of Zürich, and by Grant number GRS-048/04 from the Gebert Rüf Foundation. We thank the Oral Microbiological Service Laboratory at the Institute of Dentistry in Helsinki for providing the clinical isolates. Furthermore, we express our gratitude to the following ETH Zürich staff members: Prof. D. Günther and K. Birbaum for the element analyses, F. Krumeich for providing TEM images, R.N. Grass for assistance in scanning electron microscopy, and Prof. L. Gauckler for providing measurement time. Received for publication August 25, 2006. Revision received March 6, 2007. Accepted for publication March 29, 2007.
Journal of Dental Research, Vol. 86, No. 8,
754-757 (2007) This article has been cited by other articles:
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
