The antimicrobial and physicochemical properties of experimental light curing composites prepared with fillers made of mechanically activated alkali-substituted calcium phosphates like CaKPO 4 , CaNaPO 4 or Ca 2 KNa(PO 4 ) 2 were compared with a commercial silane-modified cristobalite filler.
The antimicrobial properties were tested using Streptococcus mutans , Staphylococcus aureus and a clinically isolated plaque mixture. The potential for reducing bacteria growth on modified composites was determined using the proliferation reagent WST-1, which enables the measurement of metabolic activity and therefore the colonization with living bacteria. Investigated material properties included the degree of conversion and a test of flexural strength.
All alkali-substituted composites provide a changed, mainly basic micro-milieu leading to a reduction of bacteria population with respect to the non-modified composite of about 25–70% with a flexural strength of cured composites in the range of 55–77 MPa complying with the clinical standard and a degree of conversion of 44–66%.
This study suggests that the modified composites increase antimicrobial properties while basic composite characteristics are not influenced by the filler.
Photo-curable resin-based composites have become very popular in operative dentistry as they provide excellent esthetics and at the same time minimize the need to sacrifice sound dental hard tissue as required for retaining conventional restorative materials. Disadvantages are the polymerization shrinkage, which can lead to a deformation of the dental hard tissue and in the worst case to a gap between filling and cavity as well as the adhesion of bacteria on composite surfaces . The biocompatibility of composites as filling material is still in discussion, but generally they are considered to be systemically non-toxic. Aims for the further improvement of filling composites include besides the reduction of polymerization shrinkage and the improvement of the mechanical properties and abrasion resistance , the reduction of microbial colonization of the materials . An approach to obtain antimicrobial composites is the modification of the polymeric matrix with either additives like chlorhexidine or methacryloyloxydodecylpyridinium bromide (MDPB) . While the first is rapidly released during several days showing only good antimicrobial properties in short term experiments, the latter is chemically bonded into the matrix by copolymerization such that a long lasting antimicrobial effect on the composite surface is obtained.
A second possibility to achieve antimicrobial properties of dental composites is the use of antibacterial active fillers, such as fluoride , quaternary ammonium polyethyleneimine nanoparticles , silver containing zeolites , silver-apatite or silver supported silica gel . Although these approaches generate sufficient antimicrobial properties, the poor color stability of silver containing materials prevents a clinical application. Recently, it could be demonstrated that other mineral compounds such as mechanically activated calcium alkali phosphate cements also provide strong antimicrobial activity against various dental microbia, which was a result of the strong basic setting reaction due to the release of tertiary alkali phosphates . In this study calcium alkali phosphates such as CaKPO 4 , CaNaPO 4 or Ca 2 KNa(PO 4 ) 2 were used as fillers in dental composite formulations. The hypothesis was that the matrix exhibited antimicrobial properties while basic composite properties such as mechanical strength or the degree of conversion were not influenced by the filler.
Materials and methods
Calcium potassium phosphate (CaKPO 4 , CPP) and calcium sodium phosphate (CaNaPO 4 , CSP) were synthesized by heating a mixture of monetite (DCPA; Mallinckrodt Baker, Griesham, Germany) and potassium or sodium carbonate (both Merck, Darmstadt, Germany) in a 2:1 molar ratio to 1050 °C for 7 h followed by quenching to room temperature in a desiccator. Calcium potassium sodium phosphate (Ca 2 KNa(PO 4 ) 2 , CPSP) was produced similarly from a mixture of DCPA, potassium carbonate and sodium carbonate in a 4:1:1 molar ratio. The sintered cakes were crushed with pestle and mortar and passed through a 355 μm sieve. Mechanical activation of CPP, CSP and CPSP was performed by grinding in ethanol for 24 h in a planetary ball mill (PM400 Retsch, Germany, diameter: 400 mm unidirectional) at 250 rpm with 500 ml agate jars, 200 agate balls (10 mm) and a load of 125 g powder and 125 ml ethanol (99.9%), per jar. The ground powders were dried at 60 °C and surface modified with methacryloxypropyltrimethoxysilane (Merck, Darmstadt, Germany) by dispersing 10 g powder in a solution of 1 ml of the silane and 0.05 ml water in 50 ml isopropanol for 1 h, following filtering the suspension and drying in air at 40 °C. Particle size distributions were determined using laser particle size analysis (L300, Horiba, Kyoto, Japan). 100 mg of the powder particles were suspended in 200 ml isopropanol and dispersed by applying ultrasound for 15 min. X-ray diffraction patterns of the sintered and mechanically activated compounds were recorded on a diffractometer D5005 (Siemens, Karlsruhe, Germany). Data were collected from 2 = 20–40° with a step size of 0.02°.
Experimental light curing composites were produced by mixing 14 g bisphenol-A-glycidyl-methacrylate (Bis-GMA, Sigma–Aldrich, Steinheim, Germany) and 6 g triethylenglycol-dimethacrylate (TEGDMA, Sigma–Aldrich, Steinheim, Germany) as the organic matrix, 0.2 g campherchinon (Merck, Hohenbrunn, Germany) and 200 μl N,N-dimethyl-p-toluidine (Sigma–Aldrich, Steinheim, Germany) as the photo initiating system and 20 g of the filler. The latter was either commonly used commercial silane-modified cristobalite filler as control samples (Silbond 6000 MST, 6 μm d 50 – Bassermann minerals, Mannheim, Germany) or a 1:1 mixture of cristobalite with the alkali substituted calcium phosphate powders. The composites were cured in silicone (Dublisil, Dreve Dentamid GmbH, Unna) rubber molds (15.5 mm diameter, 2 mm height) using blue light (Bluephase G2 (380–515 nm), Ivoclar Vivadent GmbH, Ellwangen, Germany). The radiation and conversion of the specimen was performed in orbital sequence four times for 40 s (high level) at front and back side respectively. The surface inhibition layer was removed by grinding with SiC-paper (No. 4000 – Struers, Willich, Germany). For pH-value measurements the test bodies were polished with “brownie”, “greenie” Eveflex and Evebrush WT1 (EVE Ernst Vetter GmbH, Pforzheim, Germany) during watercooling with 7000–12,000 rpm.
The antimicrobial properties of the composite discs were tested using Staphylococcus aureus (Strain RN 4220), Streptococcus mutans (DSM 20 523) and a clinically isolated plaque culture. This mixture was taken from a volunteer and represents an individual mixture of plaque bacteria. The LB (lysogeny broth) culture medium for growing the different bacterial strains was prepared by dissolving 2 g yeast extract (AppliChem, Darmstadt, Germany), 4 g trypton/pepton Difco™ (Becton and Dickinson, USA) and 2 g NaCl (Merck, Darmstadt, Germany) in 400 ml water. The LB medium was autoclaved for 20 min, 121 °C. Metabolic activity of the bacteria on the composites was tested using the WST-1 test (water soluble tetrazolium, Roche Diagnostica, Mannheim, Germany). The procedure of testing the antimicrobial capacity of the composites is schematically depicted in Fig. 1 . The microorganisms were grown in LB broth at 37 °C for 24 h under shaking. To get bacteria in logarithmic growth phase, 1 ml of this over night culture was transferred to 30 ml fresh medium and incubated for 2 h at 37 °C. The composite specimens were placed in triplicate into 24 well plates (Nunc, Wiesbaden, Germany), covered with 1.0 ml bacteria suspension in logarithmic growth phase and incubated for 24 h at 37 °C. As the experiments are not done all at one time, for each set of test material control samples were run in parallel to be able to calculate the reduction for every set of samples. After discarding the bacterial culture and carefully washing with PBS (phosphate buffered saline, 137 mM NaCl, 2.7 mM KCl, 7 mM Na 2 HPO 4 ·2H 2 O, 1.5 mM KH 2 PO 4 ) adhered organisms were covered with 500 μl PBS containing 1% WST reagent and incubated at 37 °C for 1 h. The conversion of WST reagent (4-[3-(4-Iodophenyl)-2-(4-Nitrophenyl)-2H-5-Tetrazolium]-1,3-benzenedisulfonate) to formazan by succinate dehydrogenase of the bacterial respiratory chain was determined by colorimetric measurement in an ELISA-Reader (Tecan, Crailsheim, Germany) at 450 nm with the appropriate calculation software (Magellan 3, Tecan, Germany). The metabolic activity of bacteria grown on modified samples (alkali-substituted calcium phosphate filler) was compared to the one of bacteria on control (cristobalite filler) samples. This enabled to directly compare the amount of adhered living microorganisms.
Degree of conversion
The degree of conversion of the experimental composites was analyzed by means of Raman spectroscopy (Renishaw 1000 series, England) . Three spectra of the light cured specimens (40 s) were recorded from 1800 to 1300 cm −1 for each specimen in a depth of approximately 1 mm from the surface with a spectral resolution of 1 cm −1 , and the degree of conversion was calculated on the basis of the areas of vinylic (1643 cm −1 ) and aromatic (1613 cm −1 ) C C vibrational bands of the polymerized ( A P ) and non-polymerized ( A E ) composite:
Degree of conversion ( DC ) = 1 − A vinyl P A aromat P × A aromat E A vinyl E × 100 %
where A E and A P are the areas of the C C peaks in the resin-based composite before and after conversion, respectively.
Flexural strength of cured composites
Flexural strength and modulus of elasticity were evaluated according to DIN EN ISO 4049 (CEN-European Committee for Standardization, 2000). Five specimens of (2 ± 0.1) mm × (2 ± 0.1) mm × (25 ± 2) mm were prepared using a stainless steel split mold and a small steel plate covered with a microscope slide (76 mm × 26 mm × 1 mm, Langenbrinck, Emmendingen, Germany). To ease separation of the specimens from the slide, a transparent polyester foil matrix strip (8.0 mm, Dental Exports of London, Hanworth, England) was inserted between the mold and the slide. The resin composite cements were packed into the mold with slight excess and covered by matrix band and microscope slide. Slide, mold and plate were clamped together to keep them aligned. Specimens were polymerized for 40 s at the center following 40 s respectively at both ends and same procedure dorsal. Then the specimens were stored in a water bath of 37 ± 1 °C for 15 min. After removing the overlap with SiC-paper (No. 140 or 320, Struers, Willich, Germany), the specimens were stored in a water bath of 37 ± 1 °C until mechanical testing, making sure to exclude all ambient light. 24 h after manufacture, the height and width of the specimens were determined (±0.01 mm). The flexural strength was determined by a Zwick Roell Z010 (feed rate: 0.75 ± 0.25 mm/min, loading rate: 50 ± 16 N/min). The maximum load was recorded and the flexural strength σ was calculated according to:
σ = 3 ⋅ F ⋅ I 2 ⋅ b ⋅ h 2