Titanium specimens were machined from grade 4 titanium bars (Seoul Titanium Co., LTD., Siheung, Korea) into disc shapes 20 mm in diameter and 0.5 mm in thickness. Three percent yttria-stabilized tetragonal zirconia polycrystalline discs that were 15 mm in diameter and 2 mm in thickness were prepared through cold isostatic pressing of the powder mixture (Tosoh, Tokyo, Japan) at 200 MPa, followed by sintering at 1500 °C for 2 h. The titanium and zirconia discs were polished following a previously published method [25] until the surface roughness was lower than 0.01 μm. The polished discs were coated with silver using the aerosol deposition method. First, the discs as substrates were sequentially washed using acetone, ethanol and water and mounted in the deposition chamber. Commercially available silver nanoparticles (ENB KOREA Co., LTD., Daejeon, Korea) mixed with alumina powder (Showa Denko, Tokyo, Japan) at a concentration of 1 wt% were preheated and aerosolized by vibration with a carrier gas in the aerosol chamber. The silver aerosol was accelerated by the pressure difference between the two chambers and deposited onto the substrates through a slit nozzle at room temperature. The thickness of the coating was confirmed by scanning electron microscopy to be approximately 1 μm.

The coated surfaces were observed using field emission scanning electron microscopy (SEM, S-4700, Hitachi High-Tech Corporation, Tokyo, Japan), and elemental analysis of the surfaces was performed using energy dispersive X-ray spectroscopy. The arithmetic mean heights of the surfaces of uncoated and coated specimens were measured using a three-dimensional confocal laser microscope (LSM 800, Zeiss, Jena, Germany). The amount of silver released from each coated disc was measured by incubating it in a tube containing phosphate-buffered saline (PBS) at 37 °C. After 24 h of incubation, the disc was removed from the tube, and the amount of silver in the solution was measured using inductively coupled plasma–mass spectrometry (ICP-MS, NexION 350D, Perkin-Elmer, Waltham, MA, USA).

Based on previously published data [25], we determined the optimum growth conditions for each bacterium in our experiments. Streptococcus gordonii ATCC 10558 cells cultivated in sterilized brain heart infusion broth (BHI broth; Bacto™ Brain Heart Infusion, BD, Franklin Lakes, NJ, USA) were harvested through centrifugation, resuspended in fresh media and adjusted to a bacterial concentration of approximately 4.8 × 104 colony forming units (CFU)/mL, corresponding to an optical density of 0.03 at 600 nm. The silver-coated and uncoated discs were sterilized by ethylene oxide and incubated with 1 mL of the bacterial suspension under aerobic conditions at 37 °C in 12-well plates, and bacterial growth was assessed after 24 h. Cultured Fusobacterium nucleatum ATCC 25586 cells in sterilized BHI broth under anaerobic conditions (80% N2, 10% H2, 10% CO2) were harvested and washed with fresh medium. Then, the cells were adjusted to an optical density of 0.064 at 600 nm, corresponding to a bacterial concentration of approximately 2 × 108 CFU/mL. The sterilized specimens were inoculated with 1 mL of the bacterial suspension and 3 mL of fresh media in 12-well plates and incubated in an anaerobic chamber at 37 °C for 72 h. Porphyromonas gingivalis ATCC 33277 cells cultivated in sterile BHI broth supplemented with 0.5 mg/mL hemin and 5 mg/mL vitamin K under anaerobic conditions were harvested, washed and adjusted to a concentration of 5.1 × 107 CFU/mL; the suspension had an optical density of 0.01 at 600 nm. The sterilized specimens were inoculated with 2 mL of the bacterial suspension in 12-well plates and then incubated in an anaerobic chamber at 37 °C for 48 h.

After incubation, each specimen was gently rinsed using PBS to remove unadhered bacteria from the surface. Thereafter, the remaining adherent bacteria were stained with 1% crystal violet solution for 10 min, washed gently with PBS and destained using 400 μL of a destaining solution (a mixture of 80% ethanol and 20% acetone). The absorbance of each solution was then measured using a microplate spectrophotometer (Epoch 2, BioTek, Winooski, VT, USA) at 590 nm. The absorbance values were calculated per unit area of each disc.

Silver solutions were prepared with commercially available silver nanoparticles (ENB KOREA Co., LTD.) in PBS at concentrations of 0.00001, 0.0001, 0.01 and 0.1 wt%. After sterilizing 5 mL of the prepared silver solutions at 121 °C for 15 min, S. gordonii, F. nucleatum and P. gingivalis were cultured in the solutions. The absorbance of each bacterial suspension was measured over time.

Human gingival fibroblasts and mouse preosteoblasts were cultured on the silver-coated and uncoated discs, and the cellular proliferation was evaluated using a 3-[4,5-dimethyl-thiazol-2-yl]-2,5-diphenyl-tetrazolium bromide (MTT) kit (TOX-1, Sigma-Aldrich, St. Louis, MO, USA). The specimens were sterilized by ethylene oxide and placed in 12-well plates. Human gingival fibroblasts (HGFs; PCS-201-018, ATCC, Manassas, VA, USA) cultured in fibroblast basal medium (PCS-201-030, ATCC, Manassas, VA, USA) using growth kit-low serum (PCS-201-041, ATCC, Manassas, VA, USA) were collected and seeded onto the specimens at a density of 105 cells/mL. After 24 h of incubation, the medium was replaced with another freshly prepared medium containing MTT in each well, and the 12-well plates were incubated at 37 °C for 4 h. After removing the medium containing MTT in each well, an MTT solubilization solution was added, and the absorbance at 570 nm was measured using a microplate reader (Epoch 2, BioTek). The background absorbance at 690 nm was also measured and subtracted from the 570 nm measurement values.

Mouse preosteoblast cells (MC3T3-E1, ATCC, Manassas, VA, USA) cultured in α-minimal essential medium containing 10% fetal bovine serum and 1% penicillin/streptomycin were also seeded on the sterilized specimens in the 12-well plates. After 4 and 7 days of culture, MTT assays were performed as described above. The negative controls were prepared by culturing the cells without specimens and the absorbance values were expressed as a percentage of the controls.

The Shapiro–Wilk test was used to assess the normality of variables and the Levene’s test was performed to assess the equality of variances. Since the results showed non-normal distribution of the measurement values, the comparison of the biofilm formation and cell proliferation between silver-coated and uncoated surfaces of each material was carried out using the Mann–Whitney U test. Version 26.0 of the IBM SPSS Statistics (IBM Corp., Armonk, NY, USA) software was used for all statistical analyses, and the level of significance was set at 0.05.