Specimen preparation

Based on the vertical height requirements of overdenture attachments in clinical implant-supported overdenture systems, several standard models were designed using the Solidworks software package (Dassault Systems SA, Conker County, MA, America) (Table 1). The model was designed as a 28 mm * 8 mm * 8 mm box that has a surface with an attachment space for a denture fitting surface; for one sample, the surface was left unpolished, and for the other five samples the surface was polished [10]. PMMA model resin (Bio-pink, Aditie, Qinhuangdao, China) was milled using a CAD/CAM milling unit (K5 impression, VHF, Ammerbuch, Germany) according to the model STL files. All samples were sonicated with ultra-filtered water and then immersed in ultra-filtered water for 48 h to remove residual monomers. For each group, the surfaces were polished with sandpaper from 120 to 2000 mesh (Struers, Copenhagen, Denmark) and inspected using a digital caliper. The overall size of each sample and the thickness of the material removed by polishing were controlled to between 0.2–0.3 μm.

Table 1 Five group denture fitting surface model by SolidworksProfilometry

The surfaces of each sample were examined using scanning electronic microscopy (JSM-IT500LA, JEOL, Tokyo, Japan). Mean roughness (Sa) was measured using a 3D optical profiler system (Rtec instrument, San Jose, CA, USA) with a confocal objective BF-5X and calculated using Gwydion.

Bacterial strains, and media

Streptococcus mutans UA 159 WT strain and its derivatives were maintained in a sterile brain heart infusion (BHI) broth (Oxoid, Hampshire, England, UK), Staphylococcus aureus was maintained in BHI broth at 37 °C, and Candida albicans was cultured in a yeast extract peptone glucose (YPGA) medium (Oxoid) overnight at 35 °C in an anaerobic incubator. The resulting bacterial suspensions were respectively diluted to concentrations of 1 × 106 for Streptococcus mutans, 1 × 106 for Staphylococcus aureus, and 1 × 105 for Candida albicans colony forming units (CFUs)/mL, with fresh medium available for further use.

Surface cleaning

Before cleaning, all specimens were gently washed three times with phosphate buffered saline (PBS).

Immersion cleaning protocol (ICP): specimens were soaked in distilled water at 40 °C for 15 min.

Chemical cleaning protocol (CCP): specimens were soaked in denture cleaning tablet solution (Y-kelin, Beijing, China) for 15 min at 40 °C.

Mechanical cleaning protocol (MCP): specimens were washed in an ultrasonic cleaner (40,000 vibration/sec, Guanshibo, Shenzhen, China) with distilled water for 15 min at 40 °C.

Chemical and mechanical combined cleaning protocol (C + MCP): specimens were washed in an ultrasonic cleaner with denture cleaning tablet solution (Y-kelin, Beijing, China) for 15 min at 40 °C.

Scanning electron microscopy

Scanning electron microscopy was performed for all samples (JSM-IT500LA, JEOL). Biofilms from each sample were fixed with 2.5% glutaraldehyde for 12 h, serially dehydrated in ethanol, and sputter-coated with gold.

Biofilm formation and biomass quantification

Before being co-cultured with the microorganisms, the samples were sterilised with 75% ethanol solution and ultraviolet light, after which they were incubated for 2 h with artificial saliva and rinsed with PBS. Next, 8 ml of mixed bacteria liquid was incubated with a resin sample in 12-well polystyrene microtiter plates. After 4 h, the medium was removed and fresh BHI culture containing 1% sucrose was placed in an aerobic environment at 37 °C for 24 h for biofilm formation. Thereafter, the broth was removed and the specimens were washed with PBS to remove the non-adherent cells. All adhesions were collected and stained with crystal violet solution at room temperature for 15 min, and the biomass was quantified using a plate reader at 570 nm [12].

Live cell counting

The biofilm that formed in the fitting surfaces of the specimens was collected, added to a PBS solution, and fully vortexed. This bacterial solution was then serially diluted and spread on BHI, Baird-Parker, and Chromogenic Candida agar plates. After 48 h incubation, the numbers of colony-forming units were counted.

Biofilm analysis and structural imaging

Mixed biofilms were generated as described above. Live bacteria with intact cell membranes were labeled with SYTO9, and dead bacteria with compromised membranes were labeled with propidium iodide (Invitrogen, Carlsbad, CA, USA). Images of the biofilm were captured using confocal laser scanning microscopy (Olympus FV3000, Tokyo, Japan). The imaging gates were set to 488 nm for SYTO9 and 561 nm for propidium iodide. Each biofilm was scanned at three random positions, optical sections were used for 3D reconstruction of live/dead cell biofilms with Imaris 7.0.0, and the live-/dead-cell biomass of the biofilms was quantified using ImageJ.

Statistical analysis

All statistical data were tested to establish whether they met the normal distribution and chi-square criteria; if so, a one-way ANOVA or Tukey’s post-hoc test was performed; if not, the Kruskal–Wallis test or Dunn’s multiple comparison test was performed. All statistical tests were two-tailed with a significance level of 0.05. The statistical analyses were conducted using GraphPad Prism 8.2.0.