2.1

Specimen preparation and biofilm formation

Demineralized bovine enamel specimens (n = 48) were allocated into four groups: (1) resin-infiltrated enamel with biofilm, (2) enamel with biofilm, (3) resin-infiltrated enamel, (4) no treatment (control). Enamel specimens were produced using bovine incisors. In brief, crowns were cut off from the roots and stored in a 0.1% thymol solution (VWR International, Dietikon, Switzerland) for no longer than 6 month. Cylindrical specimens (4 mm in diameter) were punched out from each crown and ground stepwise from 1200 to 4000 Fepa P (1200, 2400, 4000 grit, Water Proof silicon carbide Paper, Struers, Erkath, Germany).

Initial carious lesions with intact surface layer were created in all enamel specimens by the following demineralization procedure. Briefly, specimens were immersed for 14 days in an acidic buffer with traces of thymol at a pH of 5 and 37 °C . The solution was renewed every second day to maintain a constant pH. After demineralization, control specimens (n = 6) were cut in 100 μm slices and demineralization depth and intact surface layer were controlled using transverse microradiography. In this study, lesions of about 200 μm were used. Resin infiltration with Icon (DMG, Hamburg, Germany), a caries infiltrant system, was performed for specimens of group (1) and (3) according to the manufacturer’s instructions. In brief, 15% hydrochloric acid was used to etch the surfaces for 2 min (Icon Etch, DMG) and removed by 30 s water rinsing and subsequent air-drying. Specimens were then rinsed with 99% ethanol (Icon Dry, DMG) for 30 s and air-dried. 0.75 μL resin (Icon Infiltrant, DMG) was applied for 3 min. Excess material was removed with a cotton roll prior to light-curing for 40 s. This was followed by a second application of 0.75 μL of resin for 60 s and gentle cotton roll application. Light curing followed again for 40 s (800 W/cm ² bluephase, Ivoclar Vivadent, Schaan, Liechtenstein). Subsequently, specimens were fine grinded using a holder with an integrated spring and silicon carbide discs (4000 Fepa P) for 10 s to remove the oxygen inhibition layer under constant pressure. Specimens of all groups underwent gas sterilization for 16 h (Ethylene oxide, 3 M™ Steri-Gas™ Cartridges, Healthcare, Rüschlikon, Switzerland) and were stored sterile in a moist chamber until further treatment.

For biofilm formation in groups (1) and (2), three-species biofilms were grown on specimens for 24 h. The bacterial strains ( Streptococcus mutans OMZ 918, Streptococcus oralis OMZ 60, Actinomyces oris OMZ 745) were provided by the Institute for Oral Biology, Section for Oral Microbiology and General Immunology, University of Zurich, Zurich, Switzerland. Prior to bacterial incubation, specimens were immersed in diluted saliva supernatant to form a pellicle. One healthy subject donated fresh whole mouth saliva, which was used in all experiments. Donated non-stimulated saliva was centrifuged twice for 30 min (12′100 g). The supernatant was diluted 1:2 in 0.9% NaCl (Braun, Melsungen, Germany) and subsequently sterile filtrated (TPP syringe filters with 0.2 μm pores, Faust, Schaffhausen, Switzerland). For pellicle formation, specimens were incubated with gentle agitation in 800 μL diluted saliva supernatant for 4 h. Bacterial strains in a mixture of 30% diluted saliva supernatant and 70% modified fluid universal medium (mFUM) were adjusted to an OD ₅₅₀ of 1 and mixed as inoculum. Pellicle-coated specimens were incubated during gentle agitation in 2 mL inoculum. Incubation was performed anaerobically in jars using gas-paks for 24 h at 37 °C (GENbox anaer and GENbag anaer, bioMérieux, Marcy l’Etoile, France). After 6 h, specimens were transferred to new wells with fresh medium and analyzed after 24 h incubation, whereas expended media (after 6 and 24 h incubation) were subjected to high performance liquid chromatography (HPLC).

2.2

HPLC

Analysis of uncured Icon Infiltrant (DMG) was carried out using HPLC on an Agilent 1100 LC/MS (Agilent Technologies, Basel, Switzerland). An Agilent column (ZORBAX Eclipse XDB-C8, 4.6 × 150 mm, 5 μm) with a 50/50 mixture of acetonitrile and water at a flowrate of 0.75 mL/min and a run time of 10 min was used. Identification of uncured Icon Infiltrant, namely TEGDMA, was performed with single ion detection at 309 m / z after diluting the experimental media (n = 12) with 50 vol% of a methanol-water (Millipore) mixture (80/20). To quantify the amount, the chromatogram peak area was compared with a calibration curve (linear range, R ² = 0.998) of mere TEGDMA (95%). All chemicals for HPLC analysis were purchased from Sigma Aldrich (Buchs, Switzerland) and were of HPLC analytical grade, except otherwise stated.

2.3

Bacteria counts and metabolic activity

Plate counts were performed for specimens w/wo resin infiltration and subjected to biofilm (groups 1 and 2, n = 8) after 24 h. Specimens were sonified in 1 mL 0.9% NaCl and vortexed. 50 μL of different bacterial dilutions (in 0.9% NaCl) were plated out on Columbia sheep blood agar plates (CSBA, bioMérieux, Geneve, Switzerland) and incubated under anaerobic conditions using gas-paks. Plate counting followed after 2 days of incubation using a light microscope with 10-fold magnification (Wild Stereoskop, Heerbrugg, Switzerland).

Bacterial metabolic activity of groups (1) and (2) was measured after 24 h (n = 10). Specimens were transferred into 96-well plates and incubated in 300 μL resazurin solution consisting of 10 vol% alamarBlue Cell Viability Assay Reagent (Life Technologies, Zug, Switzerland) and fresh media (30% saliva solution + 70% mFUM) under anaerobic conditions. Two wells were additionally filled with blank resazurin solution (without specimens for background detection) and one well with bacteria from the planktonic inocula. After 15 min, 200 μL of each solution was pipetted into new 96-well plates and fluorescence was measured in a spectrophotometer with plate-reader at 560 nm excitation/585 nm emission at 37 °C (Spectramax M2, Molecular Devices, Bucher Biotec, Basel, Switzerland). Results were presented as relative fluorescence units (rfu) after background subtraction.

2.4

Bacteria imaging

Two specimens of each group were analyzed by scanning electron microscopy (SEM) after 24 h of incubation (SUPRA 50VP and Genesis, Carl Zeiss, Oberkochen, Germany). Briefly, specimens were washed with 0.9% NaCl solution and fixed for at least 24 h in 4% glutaraldehyde solution (in 0.1 M sodium potassium phosphate buffer, pH 7). A dehydration procedure followed and subsequent critical point drying was performed. Specimens were coated by gold sputtering for 60 s. Images were taken to show the detailed surface characteristics’.

Biofilms and resin-infiltrated surfaces were examined using confocal laser scanning microscopy (CLSM) after 24 h of incubation. Two specimens of each group were washed in 0.9% NaCl to remove loosely bound bacteria, fixed with 4% paraformaldehyde in the dark at room temperature for 60 min and washed again with 0.9% NaCl. Fluorescence staining was performed using Syto 59 (Life Technologies) to visualize all bacteria. The detailed staining procedure has been described elsewhere . In brief, each specimen was incubated in 500 μL staining solution consisting of 5 μM Syto 59 for 15 min at room temperature in the dark. After staining, specimens were washed in phosphate buffered saline (PBS, Invitrogen, Carlsbad, CA, USA) and mounted onto chamber slides (Fisher Scientific, Schwerte, Germany) using Mowiol 4-88 (Sigma Aldrich). Images were taken with a CLSM (SP5, Leica Microsystems, Heidelberg, Germany) using a 20× objective (numerical aperture: 1.25) and a helium laser (561 nm). Emission was detected with a photomultiplier between 630 and 660 nm. Three random areas of each specimen were examined with a z step size of 1 μm (512 × 512 pixels). Image processing was performed using Imaris Software 7.7.2 (Bitplane, Zurich, Switzerland).

2.5

Data presentation analysis

Leakage of TEGDMA monomers was compared between the resin-infiltrated specimens with biofilm (group 1) and without biofilm (group 3) after 6 h and after 24 h. Plate counts and relative fluorescence units were separately analyzed after 24 h between group (1) and group (2). All values were not normally distributed and Wilcoxon/Kruskal-Wallis test was applied for non-parametrical analysis. JMP (version 10, SAS, Cary, NC, USA) was used for statistical analysis. The overall level of significance was set at p ≤ 0.05.

2.1

Specimen preparation and biofilm formation

Demineralized bovine enamel specimens (n = 48) were allocated into four groups: (1) resin-infiltrated enamel with biofilm, (2) enamel with biofilm, (3) resin-infiltrated enamel, (4) no treatment (control). Enamel specimens were produced using bovine incisors. In brief, crowns were cut off from the roots and stored in a 0.1% thymol solution (VWR International, Dietikon, Switzerland) for no longer than 6 month. Cylindrical specimens (4 mm in diameter) were punched out from each crown and ground stepwise from 1200 to 4000 Fepa P (1200, 2400, 4000 grit, Water Proof silicon carbide Paper, Struers, Erkath, Germany).

Initial carious lesions with intact surface layer were created in all enamel specimens by the following demineralization procedure. Briefly, specimens were immersed for 14 days in an acidic buffer with traces of thymol at a pH of 5 and 37 °C . The solution was renewed every second day to maintain a constant pH. After demineralization, control specimens (n = 6) were cut in 100 μm slices and demineralization depth and intact surface layer were controlled using transverse microradiography. In this study, lesions of about 200 μm were used. Resin infiltration with Icon (DMG, Hamburg, Germany), a caries infiltrant system, was performed for specimens of group (1) and (3) according to the manufacturer’s instructions. In brief, 15% hydrochloric acid was used to etch the surfaces for 2 min (Icon Etch, DMG) and removed by 30 s water rinsing and subsequent air-drying. Specimens were then rinsed with 99% ethanol (Icon Dry, DMG) for 30 s and air-dried. 0.75 μL resin (Icon Infiltrant, DMG) was applied for 3 min. Excess material was removed with a cotton roll prior to light-curing for 40 s. This was followed by a second application of 0.75 μL of resin for 60 s and gentle cotton roll application. Light curing followed again for 40 s (800 W/cm ² bluephase, Ivoclar Vivadent, Schaan, Liechtenstein). Subsequently, specimens were fine grinded using a holder with an integrated spring and silicon carbide discs (4000 Fepa P) for 10 s to remove the oxygen inhibition layer under constant pressure. Specimens of all groups underwent gas sterilization for 16 h (Ethylene oxide, 3 M™ Steri-Gas™ Cartridges, Healthcare, Rüschlikon, Switzerland) and were stored sterile in a moist chamber until further treatment.

For biofilm formation in groups (1) and (2), three-species biofilms were grown on specimens for 24 h. The bacterial strains ( Streptococcus mutans OMZ 918, Streptococcus oralis OMZ 60, Actinomyces oris OMZ 745) were provided by the Institute for Oral Biology, Section for Oral Microbiology and General Immunology, University of Zurich, Zurich, Switzerland. Prior to bacterial incubation, specimens were immersed in diluted saliva supernatant to form a pellicle. One healthy subject donated fresh whole mouth saliva, which was used in all experiments. Donated non-stimulated saliva was centrifuged twice for 30 min (12′100 g). The supernatant was diluted 1:2 in 0.9% NaCl (Braun, Melsungen, Germany) and subsequently sterile filtrated (TPP syringe filters with 0.2 μm pores, Faust, Schaffhausen, Switzerland). For pellicle formation, specimens were incubated with gentle agitation in 800 μL diluted saliva supernatant for 4 h. Bacterial strains in a mixture of 30% diluted saliva supernatant and 70% modified fluid universal medium (mFUM) were adjusted to an OD ₅₅₀ of 1 and mixed as inoculum. Pellicle-coated specimens were incubated during gentle agitation in 2 mL inoculum. Incubation was performed anaerobically in jars using gas-paks for 24 h at 37 °C (GENbox anaer and GENbag anaer, bioMérieux, Marcy l’Etoile, France). After 6 h, specimens were transferred to new wells with fresh medium and analyzed after 24 h incubation, whereas expended media (after 6 and 24 h incubation) were subjected to high performance liquid chromatography (HPLC).

2.2

HPLC

Analysis of uncured Icon Infiltrant (DMG) was carried out using HPLC on an Agilent 1100 LC/MS (Agilent Technologies, Basel, Switzerland). An Agilent column (ZORBAX Eclipse XDB-C8, 4.6 × 150 mm, 5 μm) with a 50/50 mixture of acetonitrile and water at a flowrate of 0.75 mL/min and a run time of 10 min was used. Identification of uncured Icon Infiltrant, namely TEGDMA, was performed with single ion detection at 309 m / z after diluting the experimental media (n = 12) with 50 vol% of a methanol-water (Millipore) mixture (80/20). To quantify the amount, the chromatogram peak area was compared with a calibration curve (linear range, R ² = 0.998) of mere TEGDMA (95%). All chemicals for HPLC analysis were purchased from Sigma Aldrich (Buchs, Switzerland) and were of HPLC analytical grade, except otherwise stated.

2.3

Bacteria counts and metabolic activity

Plate counts were performed for specimens w/wo resin infiltration and subjected to biofilm (groups 1 and 2, n = 8) after 24 h. Specimens were sonified in 1 mL 0.9% NaCl and vortexed. 50 μL of different bacterial dilutions (in 0.9% NaCl) were plated out on Columbia sheep blood agar plates (CSBA, bioMérieux, Geneve, Switzerland) and incubated under anaerobic conditions using gas-paks. Plate counting followed after 2 days of incubation using a light microscope with 10-fold magnification (Wild Stereoskop, Heerbrugg, Switzerland).

Bacterial metabolic activity of groups (1) and (2) was measured after 24 h (n = 10). Specimens were transferred into 96-well plates and incubated in 300 μL resazurin solution consisting of 10 vol% alamarBlue Cell Viability Assay Reagent (Life Technologies, Zug, Switzerland) and fresh media (30% saliva solution + 70% mFUM) under anaerobic conditions. Two wells were additionally filled with blank resazurin solution (without specimens for background detection) and one well with bacteria from the planktonic inocula. After 15 min, 200 μL of each solution was pipetted into new 96-well plates and fluorescence was measured in a spectrophotometer with plate-reader at 560 nm excitation/585 nm emission at 37 °C (Spectramax M2, Molecular Devices, Bucher Biotec, Basel, Switzerland). Results were presented as relative fluorescence units (rfu) after background subtraction.

2.4

Bacteria imaging

Two specimens of each group were analyzed by scanning electron microscopy (SEM) after 24 h of incubation (SUPRA 50VP and Genesis, Carl Zeiss, Oberkochen, Germany). Briefly, specimens were washed with 0.9% NaCl solution and fixed for at least 24 h in 4% glutaraldehyde solution (in 0.1 M sodium potassium phosphate buffer, pH 7). A dehydration procedure followed and subsequent critical point drying was performed. Specimens were coated by gold sputtering for 60 s. Images were taken to show the detailed surface characteristics’.

Biofilms and resin-infiltrated surfaces were examined using confocal laser scanning microscopy (CLSM) after 24 h of incubation. Two specimens of each group were washed in 0.9% NaCl to remove loosely bound bacteria, fixed with 4% paraformaldehyde in the dark at room temperature for 60 min and washed again with 0.9% NaCl. Fluorescence staining was performed using Syto 59 (Life Technologies) to visualize all bacteria. The detailed staining procedure has been described elsewhere . In brief, each specimen was incubated in 500 μL staining solution consisting of 5 μM Syto 59 for 15 min at room temperature in the dark. After staining, specimens were washed in phosphate buffered saline (PBS, Invitrogen, Carlsbad, CA, USA) and mounted onto chamber slides (Fisher Scientific, Schwerte, Germany) using Mowiol 4-88 (Sigma Aldrich). Images were taken with a CLSM (SP5, Leica Microsystems, Heidelberg, Germany) using a 20× objective (numerical aperture: 1.25) and a helium laser (561 nm). Emission was detected with a photomultiplier between 630 and 660 nm. Three random areas of each specimen were examined with a z step size of 1 μm (512 × 512 pixels). Image processing was performed using Imaris Software 7.7.2 (Bitplane, Zurich, Switzerland).

2.5

Data presentation analysis

Leakage of TEGDMA monomers was compared between the resin-infiltrated specimens with biofilm (group 1) and without biofilm (group 3) after 6 h and after 24 h. Plate counts and relative fluorescence units were separately analyzed after 24 h between group (1) and group (2). All values were not normally distributed and Wilcoxon/Kruskal-Wallis test was applied for non-parametrical analysis. JMP (version 10, SAS, Cary, NC, USA) was used for statistical analysis. The overall level of significance was set at p ≤ 0.05.