2.1

Preparation of indomethacin and triclosan-loaded NCs

Indomethacin and triclosan-loaded NCs were prepared by the interfacial deposition of preformed polymer in a miniemulsion. The reagents were obtained from Sigma Chemical (St. Louis, USA). An organic phase was composed by polymer (MMA-co-MAA), Eudragit ^® S100 (0.50 g), indomethacin (0.025 g), triclosan (0.025 g), medium chain triglycerides (0.81 mL), sorbitan monostearate (0.19 g) and acetone (125 mL). An aqueous phase contained polysorbate 80 (0.385 g) and water (250 mL). The organic phase was added through a funnel to aqueous phase under magnetic stirring at 25 °C. Acetone and water excess were eliminated using a rotary evaporator (Rotavapor II, Buchi, Flawi, Switzerland), a B-740 recirculating chiller (Buchi, Flawi, Switzerland) and a U-700 vacuum pump (Buchi, Flawi, Switzerland). The suspension containing NCs was spray dried (B-290, Buchi, Flawi, Switzerland) using hydrophilic fumed silicon dioxide (Aerosil ^® 200) in amount of 1.5% of the suspension content as an adjuvante to avoid the aggregation on internal wall of equipment. The inlet temperature at the drying chamber was approximately 150 ± 4 °C, and the outlet temperature was 107 ± 4 °C.

2.2

Characterization of indomethacin and triclosan-loaded NCs

The mean size (d _4.3 ) of indomethacin and triclosan-loaded NCs were measured by laser diffractometry (Mastersizer 2000, Malvern, Worcestershire, United Kingdom), in wet and dried states for NCs in aqueous suspension and dried NCs respectively. The distribution of the particle size (span) values was calculated by (d _0.9 − d _0.1 )/d _0.5 , where d _0.9 , d _0.5 , and d _0.1 were the particle diameters at the 90th, 50th, and 10th percentile of particles. Zeta potential of the suspension was determined using a Zetasizer nano-ZS ZEN 3600 model (Malvern Instruments, Malvern, Worcestershire, United Kingdom). The samples were then diluted with 1 mM NaCl aqueous solution. The measurements were made in triplicate to assure accuracy.

The morphological analysis was realized using a transmission electron microscopy (TEM, JEM 1200 Exll, Jeol, Tokyo, Japan) at 80 kV. An amount of 20 μL of NCs suspension at a dilution of 1:10 was deposited in a Formvar-Carbon support film on a specimen grid and negatively stained with uranyl acetate solution (2% m/v). Dried NCs (0.01 g) were processed using gold-sputter-coating and submitted to scanning electron microscopy (SEM, JSM 6060, Jeol, Tokyo, Japan) at an accelerating voltage of 10 kV.

2.3

Determination of drugs content for dried indomethacin and triclosan-loaded NCs

The dried NCs (20 mg) were dissolved in acetonitrile (10 mL) under 30 min of ultrasound stirring. The solution was filtered using a 0.45-μm (Millipore) filter, and free drugs was measured using high performance liquid chromatography (HPLC, Shimadzu LC 10-A Shimadzu, Kyoto, Japan), a UV/vis detector ( λ = 280 nm) and Nova-Pak ^® C18 3.9 × 150 mm (4 μm) Waters column. A mobile phase (60:40 acetonitrile:water solution in volume, pH 4.5, adjusted with acetic acid) was pumped at a constant flow rate. The method was previously validated with linearity (y = 159387x + 62555 and r ² = 0.99822 for indomethacin, y = 73937 + 31383 and r ² = 0.98672 for triclosan) and a precision of 1.02% for indomethacin and 1.00% for triclosan.

2.4

Cytotoxicity

Cell viability was tested by direct contact using fibroblast cells (L929, BCR, batch no. 000604, Rio de Janeiro, Brazil) according to ISO 10993-5. The MTT test was performed in sextuplicate. Cells were incubated at 37 °C and atmosphere containing 5% CO ₂ . Cells (L929) were seeded in a 96-well tissue culture plate at a concentration of 10 ⁴ cells/well (100 μL). After 24 h, the Indomethacin and triclosan-loaded NCs suspension (5 μL of nanocapsules/mL in well) and its 1:10 dilutions were added to wells and incubated for 24 h at 37 °C. The medium alone was used as a negative control, and, as positive control, cells were cultured with dimethylsulfoxide (DMSO). A volume of 50 μL of MTT (1 mg/mL) was added to each well. Formazan salts were dissolved in DMSO (100 μL), and the absorbance was measured at 570 nm (Multiskan EX Microplate Reader, MTX Lab Systems, Vienna, USA).

2.5

Formulation of the adhesive resin and primer containing indomethacin and triclosan-loaded NCs

Experimental dental adhesives were formulated using 66/33 wt% bisphenol A glycol dimethacrylate (BisGMA)/2-hydroxyethyl methacrylate (HEMA). Camphoroquinone (CQ) and ethyl 4-dimethylaminobenzoate (EDAB) were added at a concentration of 1 mol% as a photoactivation system. The dried indomethacin and triclosan-loaded NCs were added at 1, 2, 5 and 10 wt%. One group had no addition of particles, as control. An amount of 2 wt% of indomethacin and triclosan-loaded NCs were incorporated into a comercial primer (Scotchbond MP, 3M-ESPE, St Paul, MN, USA). All formulations were mixed and ultrasonicated (CBU 100/1 LDG, Plana, São Paulo, Brazil) for thirty minutes.

2.6

Morphological characterization

The adhesive containing 10% of dried indomethacin and triclosan-loaded NCs was analyzed by transmission electron microscopy at 80 kV. The monomeric adhesive was diluted (1:10) and prepared as described above, in Section 2.2 .

2.7

Degree of conversion (DC)

DC was evaluated with an ATR-FTIR spectrometer . A disk (5.0 mm diameter and 1.5 mm thick) from each sample (n = 5) was photoactivated for 20 s by a light-emitting diode with an irradiance value of 1200 mW/cm ² (Radii cal, SDI, Bayswater, Australia). Absorbance spectra were obtained before and immediately after light polymerization. DC was calculated for the intensity (peak height) of the aliphatic carbon–carbon double bond stretching vibration at 1635 cm ⁻¹ and aromatic ring at 1610 cm ⁻¹ from the polymerized and unpolymerized samples. DC measurements were repeated after one month of adhesive resin storage at room temperature, in an eppendorf protected from light.

2.8

Softening in ethanol

Disk specimens (5.0 mm diameter and 1.5 mm thick; n = 5) prepared as described in Section 2.7 were embedded in acrylic resin and polished through a series of silicon carbide (SiC) papers (400-, 600-, 800- and 1200-grit) during 2 min each. Surface microhardness was measured using a microhardness tester (HMV-2, Shimadzu, Tokyo, Japan) and Knoop indenter at a load of 25 g for 15 s, before and after immersion in absolute ethanol for two hours, and percent reduction was calculated. Three indentations were performed on each sample surface.

2.9

Indomethacin and triclosan release

Discs of adhesive (10 × 1.5 mm) containing 1, 2, 5 and 10% of indomethacin and triclosan-loaded NCs, prepared as described in Section 2.7 , and 1 mL of primer containing 2% of NCs into a dialysis bag were used to measured drugs release, both in triplicate. The specimens were immersed in 10 mL of simulated body fluid (SBF) under magnetic stirring at 37 °C. After 6, 12, 24, 48, 72, 96, and 120 h, 1 mL of released medium was collected and fresh SBF was replaced. The aliquots were filtered using a 0.45-μm (Millipore) filter and analyzed using HPLC method previously validated and described above, in Section 2.3 .

2.10

Drug diffusion through dentin

A total of 24 healthy upper premolar teeth (n = 3), which were extracted for orthodontic purpose, were used in this study for eight groups, combining primer with or with no NCs and adhesive with 1, 2, 5, and 10% of NCs. Prior to the study, patients were informed and consented about the extracted teeth would be used in an in vitro study. After the removal of the soft tissues, a plane parallel dentin section, with a thickness of 0.75 (±0.05) mm right after the end of pulp horns, was obtained from each tooth using a low-speed diamond saw with water coolant. A side of the dentin slabs was ground with 600-grit silicon carbide (SiC) abrasive paper under water for 30 s to create a standardized smear layer. Each dentin slab was fixed in a Teflon disc (outer diameter of 27 mm, inner diameter of 4 mm, and thickness of 2 mm) using cyanoacrylate adhesive. The role (diameter of 4 mm) in the center of Teflon disc exposed a part of dentin that was etched with phosphoric acid for 15 s, washed for 15 s, and dried. Primer (Scotchbond multi-purpose, 3 M ESPE, St. Paul, USA) with or with no NCs was vigorously applied, and the solvent was evaporated for 10 s. The adhesive containing nanocapsules was applied in an amount of 0.08 ± 0.01 g and photoactivated for 20 s using a light-emitting diode (Radii cal, SDI, Bayswater, Australia). A composite build-up was performed (Z350, 3M ESPE, St Paul, USA) in an increment of 2 mm and photoactivated.

Each Teflon disc containing a dentin slab was fixed in the effective diffusion area of a Franz diffusion cell apparatus, equivalent to a device with donor and receiver compartments divided by a membrane (slab of dentin) . The donor and receptor compartments were filled with 1 and 2.5 mL of SBF, respectively. After seven days, an amount of 1 mL of the content of receptor compartment was col- lected, filtered, and analyzed by HPLC as described above, in Section 2.3 . The results were calculated in percentage of diffused drug and drug diffusion in gram per square millimeters of area of restoration.

2.11

Antimicrobial test

For antibacterial activity evaluation of adhesive, six specimens (3.0 × 2.0 mm) of each group were fixed on teflon matrixes (two samples on each side of the matrixes) on the lid of a 48-well plate and sterilized by hydrogen peroxide gas plasma. Antimicrobial test was performed according . To evaluate primer and adhesive together, disks were made using one third portion in weigth of primer and two thirds of adhesive, also polymerized during 20s, fixed on teflon matrixes and sterilezed. In the sterile 48-well plate, 900 μL of brain heart infusion (BHI) broth (Sigma–Aldrich, St. Louis, MO, USA) with 0.5% sucrose and 90 μL of a suspension of an overnight broth culture of S. mutans UA 159, adjusted to optical density of 0.3 (550 nm) were added to each one of the wells. The plate was closed and incubated at 37 °C for 24, 48, and 96 h. The samples from each group were then removed from the lid’s teflon matrixes and placed inside a micro-tube containing 900 μL of saline and vortexed. Dilutions were made up to 10 ⁻⁶ . Two 25 μL-drops of each dilution were platted in BHI agar Petri dishes and incubated for 48 h at 37 °C. The number of colony forming units (CFU) was visually counted by optical microscopy and transformed to logCFU/mL.

2.12

Statistical analysis

Statistical analysis was performed using two-way ANOVA (nanocapsule concentration and time) and Tukey’s post hoc test for DC. One-way ANOVA (groups) was performed for cytotoxicity, softening in ethanol, drug diffusion through dentin, and antibacterial test in each time. The paired t-test was used for comparison between the initial and final Knoop microhardness. All tests were performed at α = 0.05.

2.1

Preparation of indomethacin and triclosan-loaded NCs

Indomethacin and triclosan-loaded NCs were prepared by the interfacial deposition of preformed polymer in a miniemulsion. The reagents were obtained from Sigma Chemical (St. Louis, USA). An organic phase was composed by polymer (MMA-co-MAA), Eudragit ^® S100 (0.50 g), indomethacin (0.025 g), triclosan (0.025 g), medium chain triglycerides (0.81 mL), sorbitan monostearate (0.19 g) and acetone (125 mL). An aqueous phase contained polysorbate 80 (0.385 g) and water (250 mL). The organic phase was added through a funnel to aqueous phase under magnetic stirring at 25 °C. Acetone and water excess were eliminated using a rotary evaporator (Rotavapor II, Buchi, Flawi, Switzerland), a B-740 recirculating chiller (Buchi, Flawi, Switzerland) and a U-700 vacuum pump (Buchi, Flawi, Switzerland). The suspension containing NCs was spray dried (B-290, Buchi, Flawi, Switzerland) using hydrophilic fumed silicon dioxide (Aerosil ^® 200) in amount of 1.5% of the suspension content as an adjuvante to avoid the aggregation on internal wall of equipment. The inlet temperature at the drying chamber was approximately 150 ± 4 °C, and the outlet temperature was 107 ± 4 °C.

2.2

Characterization of indomethacin and triclosan-loaded NCs

The mean size (d _4.3 ) of indomethacin and triclosan-loaded NCs were measured by laser diffractometry (Mastersizer 2000, Malvern, Worcestershire, United Kingdom), in wet and dried states for NCs in aqueous suspension and dried NCs respectively. The distribution of the particle size (span) values was calculated by (d _0.9 − d _0.1 )/d _0.5 , where d _0.9 , d _0.5 , and d _0.1 were the particle diameters at the 90th, 50th, and 10th percentile of particles. Zeta potential of the suspension was determined using a Zetasizer nano-ZS ZEN 3600 model (Malvern Instruments, Malvern, Worcestershire, United Kingdom). The samples were then diluted with 1 mM NaCl aqueous solution. The measurements were made in triplicate to assure accuracy.

The morphological analysis was realized using a transmission electron microscopy (TEM, JEM 1200 Exll, Jeol, Tokyo, Japan) at 80 kV. An amount of 20 μL of NCs suspension at a dilution of 1:10 was deposited in a Formvar-Carbon support film on a specimen grid and negatively stained with uranyl acetate solution (2% m/v). Dried NCs (0.01 g) were processed using gold-sputter-coating and submitted to scanning electron microscopy (SEM, JSM 6060, Jeol, Tokyo, Japan) at an accelerating voltage of 10 kV.

2.3

Determination of drugs content for dried indomethacin and triclosan-loaded NCs

The dried NCs (20 mg) were dissolved in acetonitrile (10 mL) under 30 min of ultrasound stirring. The solution was filtered using a 0.45-μm (Millipore) filter, and free drugs was measured using high performance liquid chromatography (HPLC, Shimadzu LC 10-A Shimadzu, Kyoto, Japan), a UV/vis detector ( λ = 280 nm) and Nova-Pak ^® C18 3.9 × 150 mm (4 μm) Waters column. A mobile phase (60:40 acetonitrile:water solution in volume, pH 4.5, adjusted with acetic acid) was pumped at a constant flow rate. The method was previously validated with linearity (y = 159387x + 62555 and r ² = 0.99822 for indomethacin, y = 73937 + 31383 and r ² = 0.98672 for triclosan) and a precision of 1.02% for indomethacin and 1.00% for triclosan.

2.4

Cytotoxicity

Cell viability was tested by direct contact using fibroblast cells (L929, BCR, batch no. 000604, Rio de Janeiro, Brazil) according to ISO 10993-5. The MTT test was performed in sextuplicate. Cells were incubated at 37 °C and atmosphere containing 5% CO ₂ . Cells (L929) were seeded in a 96-well tissue culture plate at a concentration of 10 ⁴ cells/well (100 μL). After 24 h, the Indomethacin and triclosan-loaded NCs suspension (5 μL of nanocapsules/mL in well) and its 1:10 dilutions were added to wells and incubated for 24 h at 37 °C. The medium alone was used as a negative control, and, as positive control, cells were cultured with dimethylsulfoxide (DMSO). A volume of 50 μL of MTT (1 mg/mL) was added to each well. Formazan salts were dissolved in DMSO (100 μL), and the absorbance was measured at 570 nm (Multiskan EX Microplate Reader, MTX Lab Systems, Vienna, USA).

2.5

Formulation of the adhesive resin and primer containing indomethacin and triclosan-loaded NCs

Experimental dental adhesives were formulated using 66/33 wt% bisphenol A glycol dimethacrylate (BisGMA)/2-hydroxyethyl methacrylate (HEMA). Camphoroquinone (CQ) and ethyl 4-dimethylaminobenzoate (EDAB) were added at a concentration of 1 mol% as a photoactivation system. The dried indomethacin and triclosan-loaded NCs were added at 1, 2, 5 and 10 wt%. One group had no addition of particles, as control. An amount of 2 wt% of indomethacin and triclosan-loaded NCs were incorporated into a comercial primer (Scotchbond MP, 3M-ESPE, St Paul, MN, USA). All formulations were mixed and ultrasonicated (CBU 100/1 LDG, Plana, São Paulo, Brazil) for thirty minutes.

2.6

Morphological characterization

The adhesive containing 10% of dried indomethacin and triclosan-loaded NCs was analyzed by transmission electron microscopy at 80 kV. The monomeric adhesive was diluted (1:10) and prepared as described above, in Section 2.2 .

2.7

Degree of conversion (DC)

DC was evaluated with an ATR-FTIR spectrometer . A disk (5.0 mm diameter and 1.5 mm thick) from each sample (n = 5) was photoactivated for 20 s by a light-emitting diode with an irradiance value of 1200 mW/cm ² (Radii cal, SDI, Bayswater, Australia). Absorbance spectra were obtained before and immediately after light polymerization. DC was calculated for the intensity (peak height) of the aliphatic carbon–carbon double bond stretching vibration at 1635 cm ⁻¹ and aromatic ring at 1610 cm ⁻¹ from the polymerized and unpolymerized samples. DC measurements were repeated after one month of adhesive resin storage at room temperature, in an eppendorf protected from light.

2.8

Softening in ethanol

Disk specimens (5.0 mm diameter and 1.5 mm thick; n = 5) prepared as described in Section 2.7 were embedded in acrylic resin and polished through a series of silicon carbide (SiC) papers (400-, 600-, 800- and 1200-grit) during 2 min each. Surface microhardness was measured using a microhardness tester (HMV-2, Shimadzu, Tokyo, Japan) and Knoop indenter at a load of 25 g for 15 s, before and after immersion in absolute ethanol for two hours, and percent reduction was calculated. Three indentations were performed on each sample surface.

2.9

Indomethacin and triclosan release

Discs of adhesive (10 × 1.5 mm) containing 1, 2, 5 and 10% of indomethacin and triclosan-loaded NCs, prepared as described in Section 2.7 , and 1 mL of primer containing 2% of NCs into a dialysis bag were used to measured drugs release, both in triplicate. The specimens were immersed in 10 mL of simulated body fluid (SBF) under magnetic stirring at 37 °C. After 6, 12, 24, 48, 72, 96, and 120 h, 1 mL of released medium was collected and fresh SBF was replaced. The aliquots were filtered using a 0.45-μm (Millipore) filter and analyzed using HPLC method previously validated and described above, in Section 2.3 .

2.10

Drug diffusion through dentin

A total of 24 healthy upper premolar teeth (n = 3), which were extracted for orthodontic purpose, were used in this study for eight groups, combining primer with or with no NCs and adhesive with 1, 2, 5, and 10% of NCs. Prior to the study, patients were informed and consented about the extracted teeth would be used in an in vitro study. After the removal of the soft tissues, a plane parallel dentin section, with a thickness of 0.75 (±0.05) mm right after the end of pulp horns, was obtained from each tooth using a low-speed diamond saw with water coolant. A side of the dentin slabs was ground with 600-grit silicon carbide (SiC) abrasive paper under water for 30 s to create a standardized smear layer. Each dentin slab was fixed in a Teflon disc (outer diameter of 27 mm, inner diameter of 4 mm, and thickness of 2 mm) using cyanoacrylate adhesive. The role (diameter of 4 mm) in the center of Teflon disc exposed a part of dentin that was etched with phosphoric acid for 15 s, washed for 15 s, and dried. Primer (Scotchbond multi-purpose, 3 M ESPE, St. Paul, USA) with or with no NCs was vigorously applied, and the solvent was evaporated for 10 s. The adhesive containing nanocapsules was applied in an amount of 0.08 ± 0.01 g and photoactivated for 20 s using a light-emitting diode (Radii cal, SDI, Bayswater, Australia). A composite build-up was performed (Z350, 3M ESPE, St Paul, USA) in an increment of 2 mm and photoactivated.

Each Teflon disc containing a dentin slab was fixed in the effective diffusion area of a Franz diffusion cell apparatus, equivalent to a device with donor and receiver compartments divided by a membrane (slab of dentin) . The donor and receptor compartments were filled with 1 and 2.5 mL of SBF, respectively. After seven days, an amount of 1 mL of the content of receptor compartment was col- lected, filtered, and analyzed by HPLC as described above, in Section 2.3 . The results were calculated in percentage of diffused drug and drug diffusion in gram per square millimeters of area of restoration.

2.11

Antimicrobial test

For antibacterial activity evaluation of adhesive, six specimens (3.0 × 2.0 mm) of each group were fixed on teflon matrixes (two samples on each side of the matrixes) on the lid of a 48-well plate and sterilized by hydrogen peroxide gas plasma. Antimicrobial test was performed according . To evaluate primer and adhesive together, disks were made using one third portion in weigth of primer and two thirds of adhesive, also polymerized during 20s, fixed on teflon matrixes and sterilezed. In the sterile 48-well plate, 900 μL of brain heart infusion (BHI) broth (Sigma–Aldrich, St. Louis, MO, USA) with 0.5% sucrose and 90 μL of a suspension of an overnight broth culture of S. mutans UA 159, adjusted to optical density of 0.3 (550 nm) were added to each one of the wells. The plate was closed and incubated at 37 °C for 24, 48, and 96 h. The samples from each group were then removed from the lid’s teflon matrixes and placed inside a micro-tube containing 900 μL of saline and vortexed. Dilutions were made up to 10 ⁻⁶ . Two 25 μL-drops of each dilution were platted in BHI agar Petri dishes and incubated for 48 h at 37 °C. The number of colony forming units (CFU) was visually counted by optical microscopy and transformed to logCFU/mL.

2.12

Statistical analysis

Statistical analysis was performed using two-way ANOVA (nanocapsule concentration and time) and Tukey’s post hoc test for DC. One-way ANOVA (groups) was performed for cytotoxicity, softening in ethanol, drug diffusion through dentin, and antibacterial test in each time. The paired t-test was used for comparison between the initial and final Knoop microhardness. All tests were performed at α = 0.05.