Introduction

Dental implants are widely used as an effective tool to address tooth deficiencies. However, the successful retention of dental implants is affected by various factors. Bacterial infection is one of the leading causes of implant failure1,2. Bacterial infections are primarily associated with an imbalance in the oral microbiome, which can damage the bone tissue surrounding the implant, ultimately leading to implant failure3,4. Peri-implantitis is the inflammation of the tissue surrounding an implant and is often caused by various pathogenic bacteria. This inflammation occurs as bacteria adhere to the implant surface and can lead to loss of alveolar bone around the implant5,6. Studies have shown that the inflammatory response is modulated continuously by oral bacteria, which emphasizes the need for continuous antibiotic treatment7. Additionally, the ongoing struggle against pathogenic bacteria and the threat of antibiotic resistance have resulted in a strong academic interest in alternative antibacterial strategies.

Photocatalysis, particularly involving titanium dioxide (TiO₂), has shown great promise for continuous water treatment and organic compound removal owing to its oxidative solid abilities and long-term functionality. However, its activation under ultraviolet (UV) light limits its practical application in the medical field because of the potential health hazards associated with UV exposure in the human body. Consequently, extending the photocatalytic response of TiO₂ to the visible light spectrum has become an important goal for utilizing Ti-based dental implant materials8,9.

It was previously reported that a combination of noble metal nanoparticles and titania nanotubes prepared on Ti implant surfaces showed antibacterial activity that was triggered by visible light irradiation10,11. However, a lack of cell adhesion owing to the hydrophobic nature of the noble metal did not improve the osteogenic performance. In addition, using noble metal nanoparticles as stepping stones between the valence and conduction bands of TiO2 to increase the photocatalytic efficacy under visible light can also improve the visible light-mediated photocatalytic efficacy. However, it is difficult to expect fundamental improvements in the photocatalytic efficacy owing to the limited space utilization of the stepping stones. Therefore, it is necessary to discover new materials that can overcome the hydrophobicity of noble metals and exhibit robust photocatalytic activity under visible light when combined with titania.

Reduced graphene oxide (rGO) is a material in which the oxygen content is reduced in graphene oxide (GO), resulting in high conductivity and a variety of physicochemical properties12. rGO is a promising photocatalyst for water treatment and antibiotic degradation, owing to its semiconductive properties13,14. In addition, it is partly hydrophilic because of the presence of residual hydroxyl groups after the reduction process. The degree of hydrophilicity can be controlled by attaching various hydrophilic end groups15,16. The presence of aminated and carboxylated surfaces in reduced graphene oxide (rGO) enhances hydrophilicity, thereby resulting in increased chemical reactivity. These functional groups can enhance adhesion and interaction between cells, thus facilitating cellular attachment and spreading. Further, at physiological pH, an aminated surface carries a positive charge that readily attracts negatively charged molecules such as most proteins, resulting in the promotion of cell attachment, proliferation, and osteogenic differentiation compared to a carboxylated surface, which has a negative charge that interacts with only some positively charged molecules17,18. Therefore, we selected an aminated rGO surface over a carboxylated rGO surface in this study.

In addition, the antimicrobial activity of rGO is mainly attributed to its unique nanostructure and physicochemical properties. rGOs can cause physical damage to bacterial cell membranes, disrupting their physiological functions, and eventually leading to their death19,20. Many studies have also reported the visible-light-mediated photocatalytic activity of a combination of rGO and TiO2 for dye sensitization and organic compound degradation21,22,23. Therefore, the authors predicted that rGO, with its tunable hydrophilicity and inherent photocatalytic potency, is the most appropriate candidate for maximizing visible light-mediated photocatalytic potency and biocompatibility through its combination with TiO2.

In this study, the structural features and antibacterial efficacy of this novel rGO-TiO2 NTs combination were evaluated under visible light conditions. The methods by which different functional groups of rGO influence the photocatalytic mechanisms and antibacterial outcomes were investigated, and the dynamics between the unique structural attributes of the rGO-TiO2 NTs combination and their performance were analyzed.

Materials and methods

Preparation and characterization of rGO-coated TiO2 NTs (rGO-TiO2 NTs)

TiO2 NTs (diameter: 100 nm) were prepared by anodizing (voltage: 20 V, duration: 30 min) a pure Ti sheet (99.5%; size: 5 × 5 cm2, thickness: 250 μm; Hyundai Titanium Co., Incheon, South Korea) in a hydrofluoric acid (0.5 w/v%, Merck & Co., Chicago, IL, USA) solution. The anodized specimen was heat-treated (temperature: 500 °C, soaking time: 2 h) to crystalize the specimen. Two rGO powders including (1) a non-functional rGO (NON-rGO) and (2) an aminated rGO (NH2-rGO) (Sigma-Aldrich Co., St. Louis, MO, USA) were used in this study. The rGO powder coating on the surface of the TiO2 NTs specimen was fabricated using electrophoretic deposition (EPD)24. The rGO powder was dispersed in deionized water (concentration: 0.5 mg/mL) by ultrasonication (VC 505, Sonics & Materials Inc., Newtown, CT, USA) with magnesium nitrate hydrate (concentration; 0.05 mg/mL: Mg(NO3)2-6H2O, Sigma-Aldrich Co., St Louis, MO, USA) to activate the charge on the rGO surface. rGO dispersed in deionized water was coated onto the TiO2 NTs via EPD for 1 min (EPD voltage: 50 V). After the EPD process, the rGO-TiO2 NTs specimens were washed and dried at 60 °C for 24 h. The rGO-TiO2 NTs were characterized using field-emission scanning electron microscopy (FE-SEM; S-4800; Hitachi Co., Tokyo, Japan), contact angle measurements, diffuse reflectance ultraviolet–visible-near infrared spectrophotometry (DRS; SolidSpec-3700; Shimadzu Co., Kyoto, Japan), and Photoluminescence spectrophotometry (PL; LabRAM HR-800, Horiba Co., Kyoto, Japan). The optical band gap energy of the rGO-TiO2 NTs was determined using the Tauc plot calculated from the DRS spectra as shown in Eq. (1)25.

$$\left( {\alpha h\nu } \right)^{1/n} = \, A \, \left( {h\nu - E_{g} } \right),$$
(1)

where α, hν, A, and Eg are the absorption coefficient, incident light frequency, proportionality constant and band gap, respectively. The value of the exponent ‘n’ determines the nature of the electronic transition; further, the anatase TiO2 and rGO typically exhibit an indirect band gap, and hence, the exponent value corresponds to 226,27. From the Tauc plot, the linear extrapolation of (αhν)1/n to the zero of the Y-axis provides the value of the optical band gap energy of the specimen.

Methylene blue degradation test

Methylene blue (MB) degradation tests were performed to evaluate the photocatalytic activity of the rGO-TiO2 NTs under visible light at 470 nm. The experimental specimen was placed in a 12-well plate and 1 mL of methylene blue (MB) solution (1 ppm) was added to each well. Lab-fabricated 470 nm LED lights with a power density of 5.5 mW/cm2 were irradiated on the specimens for 15 and 30 min. The distance between the specimen and the LED was 4 cm. The absorbance of the MB solution in the experimental groups was determined using a UV − vis spectrometer (wavelength of maximum absorbance peak = 665 nm; UVmini-1240, Shimadzu Co., Kyoto, Japan).

In vitro biocompatibility test

Live/dead and MTT assays were performed to test the in vitro biocompatibility of the rGO-TiO2 NTs. Human mesenchymal stem cells (hMSCs Lonza Co., Basel, Switzerland) were cultured in an α-modified Eagle’s minimum essential medium (α-MEM; Invitrogen Co., Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; Invitrogen) and 1% antibiotics (Invitrogen). To perform the live/dead assay, hMSCs were dispensed into a 24-well plate (concentration = 1 × 104 cells/well) including the specimen. After 24 h of incubation, 470 nm visible light was irradiated to the specimen for 15 min (Fig. 1). After an additional 24 h of incubation, 500 μL of phosphate buffer solution (PBS; Invitrogen) including 2 µM of calcein AM (live cell staining agent; Invitrogen) and 4 µM of ethidium homodimer-1 (dead cell staining agent; Invitrogen) was added to the well. After staining for 30 min, live (green) and dead (red) cells were visually inspected using an inverted fluorescence microscope (CKX41; Olympus Co., Tokyo, Japan).

Fig. 1
figure 1

Schematic diagram of 470 nm light emitting diode (LED) irradiation system for in vitro test.

Full size image

An MTT assay kit (Sigma-Aldrich, St Louis, MO, USA) was used to evaluate initial cell metabolic activity, including survival and growth. The cell culture, light irradiation conditions, and procedure in the MTT assay were the same as those in the live-dead assay. After an additional 24 h of incubation, 100 μg/mL of MTT solution was added to each well, and the specimens were cultured at 37 °C in a 5% CO2 incubator. After 4 h of MTT treatment, DMSO (Sigma-Aldrich, St Louis, MO, USA) was added to each well to dissolve the formazan produced by the MTT solution. The absorbance of the final solution was measured at 570 nm using a microplate reader (SpectraMax Mini; Molecular Devices Co., Sunnyvale, CA, USA).

Colony forming unit test

Colony forming unit (CFU) tests against Staphylococcus aureus (S. aureus; ATCC 25,923, Manassas, VA, USA) and Pseudomonas aeruginosa (P. aeruginosa; PA1428) were conducted to estimate the photocatalysis-based antimicrobial activities of the two rGO TiO2 NTs. After 24 h of initial incubation of the bacterial strains in fresh media, the bacterial solution (500 µL) was adjusted to a concentration of 1 × 105 CFU/mL. The solution with the adjusted concentration was inoculated onto specimens (1 × 1 cm). After an additional 24 h of incubation, the specimen was exposed to 470 nm visible light for 15 min. Then, the bacteria on the specimen were collected and diluted in 100 μL of PBS solution. The PBS solution, including bacteria, was placed on an agar plate (100 mm in diameter) and incubated for 24 h at 37 °C. After 24 h of incubation, a visual inspection was performed to analyze the resulting CFU values.

Enzymatic activity analysis

The enzymatic activities of superoxide dismutase (SOD) and catalase (CAT) in S. aureus and P. aeruginosa were measured to evaluate the antimicrobial capabilities of the rGO-TiO2 NTs under visible light irradiation at 470 nm. The detailed procedure for extracting these enzymes from bacteria is described in previous studies29. The cultivation and inoculation concentrations of the bacteria as well as other experimental conditions were the same as those used for the CFU test. To elaborate on the experimental procedures for the SOD and CAT enzymatic activity assays, after visible light irradiation, the bacteria collected from each specimen were centrifuged (at 5,000 rpm for 10 min) and washed with PBS. The washed bacteria were then sonicated (Sonics VCX-130, USA) for 20 s (with 5-s treatment and 3-s intervals). The supernatant was collected from the sonicated solution including bacteria by further centrifugation (at 12,000 rpm for 20 min at 4 °C) and was then utilized for SOD (EIASODC, Invitrogen Co., Carlsbad, CA, USA) and CAT (CAT 100, Sigma-Aldrich, St. Louis, MO, USA) enzymatic activity assays. The absorbance at 450 and 570 nm was measured using a microplate reader (SpectraMax Mini), and the results were calculated according to a standard calibration curve. The Bradford reagent (Sigma-Aldrich, St. Louis, MO, USA) was used to measure the protein concentrations. Owing to the limited number of bacteria cultured on one specimen, bacteria extracted from two specimens were used to obtain one sample.

Statistical analysis

One-way analysis of variance (IBM SPSS Statistics 24.0; IBM, USA) and post hoc Games-Howell tests were used to perform the statistical analysis. All data are expressed as the mean ± standard deviation. Differences were considered statistically significant at p < 0.05.

Results

Figure 2 shows plain and oblique FE-SEM images of the NON-rGO and NH2-rGO groups. Both rGO-TiO2 groups showed a uniform thin coating of rGO layer onto the surface of TiO2 NTs, and no significant differences were observed in surface geometry between the two experimental groups. In Fig. 3, the results of the contact angle measurements indicate that all rGO-TiO2 NTs specimens were hydrophilic, but the contact angle of the non-rGO specimens was higher than that of the other groups.

Fig. 2
figure 2

(A) FE-SEM images (plain and oblique views: × 100,000) of NON-rGO and NH2-rGO for 1 min coating.

Full size image
Fig. 3
figure 3

(A) Contact angle measurement images and (B) results of TiO2 NTs, NON-rGO, and NH2-rGO groups for 1 min coating.

Full size image

Figure 4 shows the diffuse reflectance UV − Vis − NIR spectra and band gap energies of the TiO2 NTs, NON-rGO, and NH2-rGO groups. The band gap energies of the TiO2 NTs, NON-rGO-TiO2 NTs, and NH2-rGO-TiO2 NTs groups were 3.20, 2.40, and 2.21 eV, respectively. These band gap energies correspond to light wavelengths of 387.5, 516.7, and 561 nm. Therefore, both rGO-TiO2 NTs groups were confirmed to exhibit photocatalytic reactions to visible light irradiation at 470 nm, which is the wavelength used in this study. The results of the PL spectrophotometry (Fig. 5) indicated that the PL intensities of both the NON-rGO and NH2-rGO groups were lower than those of the TiO2 NTs (control). The PL data represents the excited state of the photocatalytic semiconductor material affected by the recombination of photogenerated charges (e- and h +), so a lower PL intensity indicates better charge separation of e- and h + and more powerful photocatalytic effect30.

Fig. 4
figure 4

(A) Diffuse reflectance UV–Vis-NIR spectra, and (B) Tauc plot of TiO2 NTs, NON-rGO, and NH2-rGO groups.

Full size image
Fig. 5
figure 5

Photoluminescence (PL) spectra of TiO2 NTs, NON-rGO, and NH2-rGO groups.

Full size image

Figure 6 shows the degradation values of MB dye by the TiO2 NTs, NON-rGO, and NH2-rGO groups after 15 and 30 min of visible light irradiation at 470 nm. The MB degradation in the NON-rGO and NH2-rGO groups decreased faster than that in the TiO2 NTs, and there was no significant difference in MB degradation between the two rGO groups.

Fig. 6
figure 6

Degradation plot of methylene blue (MB) dye of TiO2 NTs, NON-rGO, and NH2-rGO groups. under 470 nm visible light irradiation.

Full size image

Figure 7 shows the calcein AM- and EthD-1-stained fluorescent images and the results of the MTT assay for the experimental groups. There were no damaged or dead cells (red fluorescence) in any of the experimental groups under 470 nm visible light irradiation. From the results of the MTT assays (Fig. 7B), the cell viability of hMSCs cultured in NH2-rGO specimens tended to increase after visible light irradiation, whereas the TiO2 NTs and NON-rGO groups showed the opposite results. Although the cell viability value of the NON-rGO specimens was significantly lower than that of TiO2 NTs and NH2-rGO specimens, regardless of visible light irradiation, it was still above the minimum standard of 70% for biocompatibility specified in ISO 10,993–517. Therefore, the results of the MTT assay confirmed that all the experimental groups were biocompatible.

Fig. 7
figure 7

(A) Calcein AM and EthD-1 (Live-Dead) stained fluorescent images and (B) MTT assay results of hMSCs on TiO2 NTs, NON-rGO, and NH2-rGO groups with 470 nm visible light irradiation (Different alphabetical letter means that there is significant difference between groups, P < 0.05).

Full size image

Figure 8 shows the images and CFU values of S. aureus and P. aeruginosa tested in this study. Visible light irradiation reduced the CFU values of all experimental groups, especially the NH2-rGO group, which showed a statistically significant reduction in the CFU values for both bacteria (p < 0.05).

Fig. 8
figure 8

(A) The photo images and (B) the results of CFU test of S. aureus and P. aeruginosa cultured on TiO2 NTs, NON-rGO, and NH2-rGO groups with 470 nm visible light irradiation (Different alphabetical letter means that there is significant difference between groups, P < 0.05).

Full size image

Figure 9 shows the superoxide dismutase (SOD) enzymatic activities of S. aureus and P. aeruginosa. Under the experimental conditions of S aureus, the SOD activity values of the NON-rGO and NH2-rGO groups after visible light irradiation were significantly reduced (p < 0.05). In addition, under the experimental conditions for P. aeruginosa, the SOD activity values of all experimental groups tended to decrease after visible light irradiation. Also, the value of NH2-rGO was reduced significantly (p < 0.05).

Fig. 9
figure 9

Superoxide dismutase (SOD) enzyme activity of S. aureus and P. aeruginosa cultured on TiO2 NTs, NON-rGO, and NH2-rGO groups with 470 nm visible light irradiation (Different alphabetical letter means that there is significant difference between groups, P < 0.05).

Full size image

Figure 10 illustrates the catalase (CAT) enzymatic activities of S. aureus and P. aeruginosa tested in this study. Under the experimental conditions for S. aureus, the CAT enzyme activity of all experimental groups did not show any significant reduction after visible light irradiation (P > 0.05). However, under the experimental conditions for P. aeruginosa, only the SOD activity value of NH2-rGO group was significantly reduced after visible light irradiation compared to the other experimental groups (p < 0.05).

Fig. 10
figure 10

Catalase (CAT) enzyme activity of S. aureus and P. aeruginosa cultured on TiO2 NTs, NON-rGO, and NH2-rGO groups with 470 nm visible light irradiation (Different alphabetical letters means that there is significant difference between groups, P < 0.05).

Full size image

Discussion

The main objectives of this study were to (1) characterize the surface of rGO-coated TiO2 NTs to evaluate whether the presence of rGO affects the visible light-mediated photocatalytic activity of TiO2 NTs and (2) investigate the antibacterial activity of both NON-rGO-coated TiO2 NTs and NH2-rGO-coated TiO2 NTs under visible light at 470 nm, to analyze whether the presence of the terminal functional group of rGO affects the antibacterial activity and biocompatibility of TiO2 NTs.

The antibacterial mechanism used in this study was a visible light-mediated photocatalytic effect based on the combination of rGO and TiO2 oxide as a photocatalyst. TiO2 only exhibits photocatalytic effects under UV light owing to its semiconducting nature and large bandgap energy between the conduction band and valence band of TiO2. For TiO2 to exhibit photocatalytic effects under visible light irradiation, the large bandgap energy distance between the conduction and valence bands must be reduced. The visible light-mediated photocatalytic activity can be achieved by coating a material such as a noble metal, heavy metal, or graphene oxide that acts as a bridge31,32,33. Generally, to exhibit photocatalytic effects, electrons in the valence band must be excited to the conduction band by absorbing a certain amount of energy in the form of light. In this case, the noble metal nanoparticles, which do not have their own photocatalytic activity, act as a stepping stone between the valence band and the conduction band, thereby enabling the photocatalytic effect to be activated by visible light, which has lower energy than ultraviolet light. However, because the area of this stepping stone is small, only a limited number of electrons can be excited, and this results in a constrained photocatalytic effect. In contrast, rGO exerts its own photocatalytic activity13. Therefore, a combination of rGO and TiO2, both of which possess their own photocatalytic activities, will result in an extensive photocatalytic activity compared to a combination of noble metal nanoparticles and TiO2. In addition, the correlation between the photocatalytic and antimicrobial effects is driven by the role of reactive oxygen species (ROS) groups, such as hydroxyl radicals, superoxide anions, and hydrogen peroxide, generated by the photocatalytic effect, which result in damage to bacteria through lipid peroxidation, protein oxidation, and DNA damage34,35. Therefore, the first objective of this study was to evaluate the photocatalytic effect of rGO-coated TiO2 NTs under visible light through surface characterization.

The FE-SEM images shown in Fig. 2 show that the rGO nanosheets were deposited uniformly on the surface of the TiO2 NTs. There were no significant differences in the surface geometry between the two groups. The DRS spectra and Tauc plot results demonstrate that both rGO TiO2 NTs groups exhibited sufficient photocatalytic activity to generate ROS groups under 470 nm visible light to exhibit antimicrobial activity. Pilot tests were conducted to optimize the coating time of rGO on the surface of TiO2 NTs for coating periods of 30 s, 1 min, 3 min, and 5 min. The FE-SEM images (Figure S1) show that the rGO sheet layer was barely observed on the TiO2 NTs surface after a coating time of 30 s. In addition, the sheet-like rGO nanoparticles began to be stacked in multiple layers after 3 min of coating, and the rGO nanosheets were coated so thickly that the nanotube structure was not visible after 5 min of coating.

The band gap energy calculations from the Tauc plot (Figure S2 and S3) showed that the 1 min rGO-coated rGO-TiO2 NTs groups exhibited the lowest bandgap energy and were expected to show the best photocatalytic activity among all coating periods. Unlike the single or double peaks detected in the Tauc plot of common semiconducting materials, multiple peaks were observed in the Tauc plot of this study. Previous studies related to the photocatalytic activity of TiO2 nanotubes also observed multiple peaks in the Tauc plots, but specific explanations were not provided36,37,38. We speculate that this phenomenon is caused by a unique combination of the crystalline structure of TiO2 nanotubes. When titania nanotubes are heat-treated at 500 °C, the titania nanotubes and the base substrate change to anatase and rutile crystalline phases, respectively39. Therefore, the heterojunction systems of coupled semiconductors (anatase and rutile TiO2) may have various heterojunction structures, allowing heat-treated TiO2 nanotubes to exhibit multiple energy levels, thus resulting in multiple peaks in the Tauc plot40,41. Additionally, among the multiple peaks in the Tauc plot, the main peak at the higher energy level (shortest wavelength) corresponds to the optical gap, while the other peaks at lower energy levels (longer wavelengths) correspond to the fundamental gaps of the semiconductor42,43. Consequently, in this study, we obtained the optical band gap of the experimental specimens by extrapolating the main line at a high energy level to the X-axis.

The PL spectra results show that lower PL spectral intensities were observed for the NON-rGO-TIO2 NTs and NH2-rGO-TiO2 NTs groups than the results of the TiO2 NTs control. Effective photocatalysts must be able to separate the charge carriers (electrons and holes) to facilitate redox reactions. The PL intensity measurements can reveal how well a material maintains these separated charge carriers. When a photocatalyst absorbs light, it generates electron–hole pairs; these charges need to be effectively separated and transported to the surface of the material to participate in redox reactions. The phenomenon of PL arises from the recombination of excited electrons and holes. Therefore, a high PL intensity signifies a high rate of recombination, while a low PL intensity indicates that the charge carriers are efficiently separated and not readily recombining44,45. Therefore, the results of PL spectra could be due to graphene acting as an electron transport material for TiO2, limiting the recombination of e- and h + . It can be concluded that graphene increased the photocatalytic activity efficiency of all the rGO-TiO2 NTs composites under visible light.

The MTT assay results demonstrate that the group cell viability values were above 70%, regardless of visible light irradiation. However, the NON-rGO groups exhibited reduced cell viability after 48 h of incubation with and without visible-light irradiation. This decrease in cell viability in the non-rGO group was also observed in the live/dead stained images, and the number of cell attachments was lower than that in the NH2-rGO and TiO2 NTs groups, both with and without visible-light irradiation. In general, rGO is cytotoxic to cells, and the mechanisms of its toxicity have been described in many different ways, including interactions with cell membranes, reactive oxygen species, DNA damage, apoptosis, and autophagy, but remain incompletely understood46,47,48. When the process of coating rGO onto TiO2 NTs is considered, numerous factors such as the particle size, surface area, concentration, cell type, the degree of reduction (amount of remaining oxygen functional groups), and the presence of sharp edges are expected to determine its toxicity49,50,51. Therefore, optimizing the coating time of rGO to achieve maximum photocatalytic effect with minimal thickness for good biocompatibility is the focus of this study. The optimal coating period of 1 min was selected, as it showed a uniform coating of rGO nanosheets and the best bandgap energy among all coating conditions (Figure S1, S2, and S3). The rGO nanosheets thinly and uniformly coated onto the TiO2 NT surface exhibited excellent biocompatibility (Figure S4). In terms of the lower cell viability of the NON-rGO group, it is expected that the difference in surface energy as shown the results of contact angle measurement (Fig. 3) between the terminal functional groups of the rGO nanosheets was one of the main factors determining the initial attachment of cells, as shown by contact angle measurements. In addition, the NH2-rGO group showed increased cell viability with increasing incubation period and visible light. The results of the live/dead and MTT assays performed in this study were insufficient for assessing the biological response of stem cells to the NH2-rGO surface and the resulting osteogenic performance. Although further investigations are required to elucidate the reasons for the increase in the cell growth rate under visible light irradiation, within the limited conditions of this study, a time window of 1 min for coating of rGO exhibited the best photocatalytic effect and biocompatibility as illustrated in Figures S2, S3, and S4.

S. aureus and P. aeruginosa, representative gram-positive and gram-negative bacteria that cause a wide range of clinical diseases, are involved in biofilm formation and are associated with biomaterial-associated infections52. Therefore, these bacteria were selected for antibacterial testing. Based on the results of CFU antibacterial tests against S. aureus and P. aeruginosa and two enzyme activity tests (SOD and CAT), a reduction in the colony-forming ability of the two bacteria and inhibition of SOD/CAT enzyme activity were observed in the NH2-rGO groups after visible light irradiation, except for the CAT enzyme activity against S. aureus. In general, rGO itself expresses antimicrobial activity by forming ROS through photocatalytic effects owing to its semiconductor properties, and it expresses a more substantial photocatalytic function under visible light irradiation in combination with TiO2. To protect cells from toxic oxygen products, SOD catalyzes the formation of hydrogen peroxide from superoxide radicals, and CAT catalyzes the decomposition of hydrogen peroxide into water and oxygen48,49. Therefore, although we were unable to identify a decrease in CAT enzyme activity against S. aureus within our limited experimental conditions, excess superoxide anions from the photocatalytic effect of aminated rGO-TiO2 NTs attack the bacterial cell wall, resulting in the ability of SOD to scavenge superoxide anions and the ability of CAT to decompose hydrogen peroxide to decrease based on the overall results of the antimicrobial tests. Consequently, a reduction in SOD and CAT enzymatic activity results in the disruption of respiratory metabolism, leading to cell death.

The most common antimicrobial technologies that use light is UV light-based germicidal irradiation, wherein the DNA replication of bacteria is directly damaged, and this results in cell death53. However, previous studies have reported that bacteria with multi-layered cell walls exhibit a stiffer and more rigid cell wall upon UV irradiation, but complete cell wall disruption and bacterial death were only observed with photocatalysis based on the combination of UV light and TiO254. Photocatalysis based antimicrobial technologies exert their antimicrobial effects by generating ROS and super-hydroxyl ions from light irradiation to disrupt cell walls and inhibit the normal functionality of the bacteria. Therefore, practical antimicrobial effects can be predicted for surface-adherent bacteria, but are less effective for suspended planktonic bacteria. In addition, structural differences between gram-positive and gram-negative bacteria may cause changes to their antimicrobial effects based on their photocatalytic efficacy. In general, the bacterial cell wall is composed of peptidoglycans, and gram-negative bacteria have a single-layer cell wall. By contrast, gram-positive bacteria have multilayer cell walls.

Previous findings predicted that structural differences between gram-positive and gram-negative bacteria were the primary reason for the less effective antibacterial activity against S. aureus than against P. aeruginosa55. However, the antimicrobial activities observed in the present study against both bacteria were excellent. This is because the combination of rGO and TiO2 NTs, which individually express photocatalytic effects, showed enhanced antibacterial activity based on a more robust photocatalytic efficacy compared to previous studies.

On considering the clinical applications of this study in the field of dentistry, it has been observed that a combination of visible light and rGO-TiO2 NTs has the potential for use in various dental prosthetic materials made of titanium such as implants, guided bone regeneration mesh, and orthodontic mini-screws56,57. Additionally, visible light has a short penetration distance into the skin and cannot reach the deep areas of the implant body; however, it can readily penetrate the thin layer of the oral mucosa58. Therefore, a combination of visible light and rGO-TiO2 NTs is expected to be feasible for immediate antimicrobial treatment of the surgical site after dental surgery.

Overall, all antimicrobial activity evaluations comprehensively demonstrated that the bacteria used in this study exhibited a precise extinction mechanism influenced by the combination of visible light irradiation and aminated rGO-TiO2 NT properties. Therefore, the combination of TiO NTs, with a high surface area and exceptional photocatalytic activity, and aminated rGO, with beneficial electronic properties and excellent biocompatibility, is expected to make the composite particularly effective for biomedical applications, including infection prevention in medical implants.

Conclusions

Within the limited range of bacterial strains and rGO types examined in this study, the combination of rGO-coated TiO2 NTs and visible light demonstrated an excellent antimicrobial performance. Unlike previous studies that coated precious metal nanoparticles as stepping stones to help express the photocatalytic effect lacking under visible light, this approach leveraged the superior photocatalytic properties of each component, resulting in a more pronounced photocatalytic-based antimicrobial effect under visible light. Furthermore, the excellent biocompatibility of aminated rGO-coated TiO2 NTs is anticipated to provide a foundation for developing novel surface treatment technologies for implantable devices, offering enhanced antimicrobial efficacy and osteogenic performance.