Commercial pure titanium (cp-Ti, grade 2, Baoji Titanium Industry Co., Ltd., Baoji, China) was used to prepare titanium disks with a diameter of 12 mm and a thickness of 1.5 mm. All disks were uniformly polished with silicon carbide sandpapers ranging from 400 to 1,200 grit in sequence and then cleaned by ultrasonic bathing in deionized water, acetone, 75% ethanol, and deionized water for 10 min, respectively. After drying overnight at 60°C, these disks were sterilized by UV light for 6 h on each side and placed in sterile 24-well plates.

The elemental analysis of titanium disks was detected by EDS (Helios Nanolab 650, FEI, Eindhoven, Netherlands) to inspect their purity. Three titanium disks were randomly selected, and three random areas on each disk were recorded and measured by EDS at a voltage of 10 kV.

As described in previous studies (Yang et al., 2019; Hu et al., 2023), peptides DJK-5 and 1018 were synthesized in the method of solid-phase 9-fluorenylmethoxy carbonyl (Fmoc) chemistry by CPC Scientific (Sunnyvale, United States). The synthesized peptides were identified by amino acid analysis, and reverse-phase high-performance liquid chromatography was utilized to achieve at least 95% purity. One hundred μg/mL stock solutions of both peptides were then prepared by dissolving lyophilized powders in sterile deionized water and keeping them sterile over time.

S. mutans (ATCC 25175) was grown anaerobically on plates with 32 g/L of brain-heart infusion (BHI) agar (Becton-Dickinson, Sparks, United States) at 37°C and subcultured on a new agar plate for 48–72 h. After that, the bacteria were picked up and transferred into BHI broth (Becton-Dickinson, Sparks, United States) for 24 h of incubation at 37°C. Amplified bacteria were harvested by centrifugation (5,000 rpm) at 4°C for 5 min and washed twice with PBS. The collected S. mutans were suspended in BHI to achieve an optical density (OD) at 0.05 as determined by a microplate reader (ELx808, BioTek, Winooski, United States) at the wavelength of 405 nm.

Titanium disks were placed in 24-well plates (one disk in each well) and randomly distributed into different groups in accordance with the following treatments: 1) the positive control (untreated biofilm) group: bacterial suspension (200 μL) and BHI (1,800 μL) containing sucrose (222.22 g/L) were added to each well; 2) 5 μg/mL of peptide treatment groups (DJK-5 or 1018): peptide stock solution (100 μL), bacterial suspension (200 μL), and BHI (1,700 μL) containing sucrose (235.29 g/L) were added to each well; 3) 10 μg/mL of peptide treatment groups (DJK-5 or 1018): peptide stock solution (200 μL), bacterial suspension (200 μL), and BHI (1,600 μL) containing sucrose (250 g/L) were added to each well; and 4) the negative control group: each well only added with BHI (2,000 μL) containing sucrose (200 g/L). Each group of bacteria received the same amount of sucrose (200 g/L as the final concentration (Souza et al., 2010; Souza et al., 2013).

Two different incubation times were tested separately for each experiment condition. In the 48 h group, titanium disks were removed from each well after 48 h and rinsed three times with sterile deionized water, followed by a 5-min ultrasonic bath in 100% ethanol and air drying. In the two 48-h treatments group, the consumed solution was removed after incubation for the first 48 h, and the corresponding fresh peptide solution and BHI medium were supplied for another 48 h. Following that, titanium disks were rinsed and dried as previously stated.

To monitor the dynamic growth of S. mutans biofilms on titanium disks treated with or without peptides DJK-5 and 1018, biofilms were detached at designated time intervals (0, 2, 4, 6, 8, 10, 12, 18, 24, 30, 36, 48, 72, and 96 h) and dispersed by pipetting to obtain an even bacterial suspension. Subsequently, 150 μL of the suspension was aspirated into 96-well plates and read by the microplate reader at OD_405nm. Three parallel wells were replicated at each time interval for each group. Then, the pH of the remaining suspension in each well was measured by a pH Meter (Accumet Basic AB 15 Plus, Thermo Fisher Scientific, Waltham, United States). For each group, three parallel wells were tested and recorded.

After incubation for one or two 48-h periods, titanium disks coated with biofilms were gently washed with PBS and stained with a fresh mixture of SYTO-9 and propidium iodide (BacLight live/dead bacterial viability kit, Molecular Probes, Eugene, United States) based on the manufacturer’s instruction. And observed by confocal laser scanning microscopy (CLSM, FV10i-LIV, Olympus, Canada) (Yu et al., 2021; Liu et al., 2022). The wavelength of excitation/emission light detected for SYTO-9 and propidium iodide was set 480/500 nm and 490/635 nm, respectively. For each group, three disks were selected, and five areas for each disk were scanned from the bottom to the top of the biofilm with a 2 μm step. An Imaris 7.2 software (Bitplane, Switzerland) was employed to convert the two-dimension images into three-dimensional volume stacks. The obtained green and red fluorescence intensity was analyzed as the total volume of live and dead bacteria, and the dead bacteria proportion was calculated using red fluorescence in relation to total fluorescence (green + red) (Guo et al., 2021).

The characteristics of corrosive titanium surfaces exposed to S. mutans (treated with or without peptides DJK-5 and 1018) were inspected through SEM (SU3500, Hitachi, Toronto, Canada). After one or two 48-h periods of incubation, titanium disks were taken out from the wells and washed three times using sterile deionized water to detach the biofilms before being ultrasonically cleaned for 5 min in sterile deionized water and 100% ethanol, respectively. After drying in the air, randomly selected areas from SEM images (at ×3,000) were captured at 3 kV for each disk.

The electrochemical experiments were carried out to investigate the effects of DJK-5 and 1018 on the corrosion resistance of titanium after exposure to S. mutans. A Princeton Applied Research (PAR) VersaSTAT 4 potentiostat (AMETEK, Berwyn, United States) was employed in this study. The electrochemical setup consisted of a conventional three-electrode cell with a platinum net as a counter electrode (CE) and an Ag/AgCl [(KCl) = 4 M] electrode as a reference electrode (RE) introduced into the cell through a Luggin probe. All potential values are used according to the Ag/AgCl reference electrode (0.197 V vs. SHE). The working electrode (WE) was set in a special mould with an exposed area of 1 cm² as the working area. The schematic diagram of the mould and the electrochemical setup are presented in Figures 1A, B. The biofilm-coated disks were connected to the PAR via a copper sheet. The solutions were purged with nitrogen for at least 30 min before introducing the working electrode, and the purging lasted throughout the experiment. Each test was repeated at least three times to ensure reproducibility.

The electrochemical tests were performed in a modified Fusayama’s artificial saliva solution (0.4 g/L NaCl, 0.4 g/L KCl, 0.795 g/L CaCl₂·2H₂O, 0.005 g/L Na₂S·9H₂O, 0.69 g/L NaH₂PO4·H₂O, 1.0 g/L urea, pH = 6.0) at 37°C ± 0.5°C as previously reported (Souza et al., 2013; Fukushima et al., 2018). The potential dynamic polarization (PDP) curves and electrochemical impedance spectroscopy (EIS) tests were determined to evaluate the corrosion resistance of titanium in the Fusayama’s artificial saliva after different treatments (Zhang et al., 2007). Prior to the tests, the open circuit potential (OCP) measurements were carried out for at least 1 h until it reached a quasi-steady state value (deviation in OCP from its 5 min value within 2%). PDP measurements were conducted from −0.5 V vs. OCP to 1.5 V vs. OCP at a scan rate of 1 mV/s. The electrochemical parameters, including the corrosion potential (E _corr), corrosion current density (I _corr), pitting potential (E _pit), and passive current density (I _passive), were determined. EIS measurements were conducted at OCP with a sinusoidal potential wave as a disturbing signal. The amplitude of sinusoidal potential was 10 mV, and the scan frequency ranged from 0.01 to 105 Hz. The EIS results were analyzed with ZSimpWin software (AMETEK, Berwyn, United States). All experiments were performed with at least three independent tests.

All data obtained were presented as means ± standard deviations for statistical analysis. One-way analysis of variance (ANOVA) was performed by SPSS 22.0 software (IBM, Armonk, United States), and post hoc Fisher’s LSD multiple comparison tests were then examined. The significance level was set at p < 0.05.

The elemental content of titanium disks used in this study was examined by EDS. As shown in Figure 1C, almost all of the typical titanium element peaks were detected. The results of the EDS analysis revealed that titanium had a purity of more than 99.9%, as measured by the atomic percentage of the Ti element.

The dynamic growth of S. mutans biofilm on titanium disks in each group was plotted in Figure 2A. In the first 2 h, there was no discernible increase in bacteria. Then, S. mutans proliferation in the positive control group accelerated and reached ∼9-fold the initial level after 12 h, followed by a gradual increase up to 48 h and then levels plateaued. S. mutans exposed to 5 μg/mL of peptide 1018 or DJK-5 showed similar but delayed kinetics with a longer lag phase, beginning to thrive after 4 h and lower growth by 6–7 fold in the first 8 h. Bacterial growth was stabilized between 12 and 48 h and reached a plateau within 96 h. Overall, there were more bacteria in the 1018 group (5 μg/mL) than in the DJK-5 group (5 μg/mL) at each time interval. The kinetics of growth was much more dramatically affected at 10 μg/mL of peptide. Treatment with 10 μg/mL of 1018 led to much slower proliferation of S. mutans in the group between 12 and 48 h, but its eventual optical density after 48 h value was higher than in the 5 μg/mL of DJK-5 group. The biofilm development of S. mutans exposed to 10 μg/mL of DJK-5 was substantially and effectively inhibited during the first 72-h, but increased slightly by 96 h.

S. mutans is known acidify the medium. The pH of the medium in the untreated S. mutans control decreased to around 4 within 18 h, due to biofilm growth (Figure 2B). The decrease in pH values in each group at the designated time intervals corresponded to the increase in OD values. Specifically, in the positive control group, pH values dropped rapidly from 7.0 at 0 h to 4.48 within 10 h, then to 4.02 by 18 h, and then remained stable. The pH values of the medium with 5 μg/mL of 1018 and DJK-5 added revealed the same declining trend. They maintained a slower decline in pH than the positive control group but still reached a pH of around 4 by 18 h. Ten μg/mL of 1018 delayed the pH drop for the first 12 h, but it still went down to 4.17 at 24 h. Remarkably, 10 μg/mL of DJK-5 was able to maintain the pH around neutrality with only a 0.5 unit decrease during the 96-h incubation period.

The dead bacterial proportions and live/dead bacteria CLSM images of S. mutans biofilms on titanium surfaces after incubation with or without peptides for one or two 48-h periods are shown in Figures 3A, B. Both peptides 1018 and DJK-5 (at 5 and 10 μg/mL) killed the majority of bacteria over 48 h (from 53.95% to 84.82%) with significant differences compared to the positive control group (p < 0.01), and the most bacterial killing was found in the 10 μg/mL of DJK-5 group. Peptide DJK-5 was more effective in killing S. mutans than peptide 1018 at the same concentration, and higher concentrations (10 μg/mL) of peptide led to a higher proportion of dead bacteria than lower concentrations (5 μg/mL) did (p < 0.05). Except for DJK-5 at 10 μg/mL, more S. mutans were killed in the peptide-treated groups after two 48-h treatments than a single 48-h treatment. Ten μg/mL of DJK-5 (48 + 48 h) induced approximately 73.88% of bacterial cell death and had a higher killing proportion than the 5 μg/mL of DJK-5 and the positive control groups (p < 0.05).

The SEM surface microstructures of titanium disks (after biofilm removal) from each group incubated with S. mutans for two 48-h treatments are shown in Figure 4. In the negative control group (no bacteria), the disks retained their smooth and even surface morphology, with the presence of longitudinal grooves as a consequence of the polishing process. In the positive control group, the titanium surface colonized by S. mutans from the beginning of immersion was rich in irregular micro-pits of various sizes and shapes along the longitudinal grooves, as indicated by red arrows in SEM images. In the case of titanium disks exposed to bacteria and 5 μg/mL of peptides 1018 and DJK-5, corrosion pits on their surfaces appeared to decrease. Ten μg/mL of peptide DJK-5 had the greatest ability to prevent material degradation, followed by 10 μg/mL of peptide 1018, with only a few visible randomly-scattered micro-pits.

To assess electrochemical corrosion resistance, PDP curves of titanium were examined in artificial saliva at 37°C after incubation with bacteria and peptides. As shown in Figures 5A, B, titanium disks demonstrated moderate corrosion resistance in the test environment because a typical passive plateau of 10⁻⁷ A·cm⁻² was observed, followed by the I _passive ranging from 10⁻⁶ to 10⁻⁷ A·cm⁻². This phenomenon reflected that a relatively protective passive film of titanium dioxide remained on the disks. The E _corr and I _corr in the active region were measured using the Tafel extension method. The passive film was broken down as the applied potential was increased above 10⁻⁶ A·cm⁻², leading to a sharp increase in the current density. The specific electrochemical data of PDP are depicted in Figures 5C–F. The E _corr of the positive control group was lower than that of the other groups (p < 0.05), whereas the I _corr of the positive control group was higher than that of the other groups (p < 0.05). With the addition of peptides DJK-5 and 1018, the E _corr shifted to a less negative value while the E _pit showed significantly higher values than the positive control group (p < 0.05), and the I _corr and I _passive were significantly lower (p < 0.05).

After 48 h, the I _corr and I _passive had increased in the following order: negative control <10 μg/mL DJK-5 < 10 μg/mL 1018 < 5 μg/mL 1018 < 5 μg/mL DJK-5 < positive control. The OCP and E _pit decreased in the following order: negative control >10 μg/mL DJK-5 > 10 μg/mL 1018 > 5 μg/mL DJK-5 > 5 μg/mL 1018 > positive control. Treatment with 10 μg/mL of peptides 1018 and DJK-5 showed improved corrosion resistance of titanium when compared to 5 μg/mL of these peptides, with 10 μg/mL of DJK-5 achieving the greatest effect. A similar trend was observed for titanium disks after two treatments for 48 h of immersion, and the I _corr in each group after two 48-h treatments was lower than those after one treatment, except in the case of the positive control.

The EIS tests were performed to characterize the stability of the passive film on titanium after different treatments, with the Nyquist and Bode plots for each group in artificial saliva at 37°C are shown in Figures 6A–D. The Nyquist plots (Figures 6A, C) revealed a capacitive loop, whereas the Bode plots (Figures 6B, D) revealed only a one-time constant, suggesting that the corrosion process of titanium was controlled by charge transfer. The impedance values were in the order of 10⁶ Ω·cm⁻², indicating that the titanium was covered by a protective passive film. The most commonly used electrical equivalent circuit (EEC) of titanium’s passive film is R _s {Q _ou [R _ou (R _in Q _in)]}. R _s denotes the solution resistance, Q _ou and Q _in refer to the constant phase element (Q) of the outer porous film and inner barrier film, respectively, and R _ou and R _in represent the outer and inner film resistance. The resistance of the passive film (R _T) equals R _ou + R _in. The corresponding fitting data are listed in Table 1, with chi-square (χ²) values in the order of 10⁻⁴. The fitting curves, fitting results, and experimental data showed great agreement. Thus, the R _s {Q _ou [R _ou (R _in Q _in)]} of EEC was appropriate to describe the passive film in this study.

Figure 6E exhibits the equivalent circuit fitting the experimental impedance data. The Q represents the non-ideal behavior of the capacitor, and there is a positive correlation between Q and capacitance (C). The capacity was related to the property of the passive film’s metal/solution interface based on the following equation: C = εε₀A/d, where C denotes the capacitance of a film, ε₀ (8.854 × 10⁻¹⁴ F·cm⁻¹) is the vacuum permittivity, ε is the dielectric constant, and A and d are the area and thickness of the film, respectively. According to Table 1, Q _ou is one order of magnitude greater than Q _in, and R _ou is two orders of magnitude less than R _in. The ε of the outer layer was estimated to be greater than that of the inner layer, indicating that the inner layer is more compact and protective. The corrosion resistance of the passive film (R _T) for each group is manifested in Figure 6F. The R _T values of titanium treated with peptides (in 48 h) were higher than those of the positive control group (p < 0.05), demonstrating that 1018 and DJK-5 could prevent S. mutans from deteriorating the passive film. Furthermore, the RT values of titanium treated with 10 μg/mL of DJK-5 were close to the negative control group, denoting a significant anticorrosion efficacy. In the case of two 48 h treatments, the R _T of titanium treated with peptides increased slightly. This trend was consistent with the results of PDP experiments.