Antibacterial activities of TiO₂ nanotubes on Porphyromonas gingivalis

Abstract

Titanium-based nanomaterials have been widely used as dental implants because of their beneficial antiseptic and nano-interfacial effects. In particular, their application as surface decontamination biocompatible materials attracts increasing attention. In this study, a TiO₂ nanotube-based antibacterial system has been fabricated by an anodic oxidation method, and its morphology, crystalline phase and hydrophilic property have been characterized. The effects of TiO₂ nanotubes on bacterial growth inhibition, as well as bacterial cell fate, were also investigated. The results indicated that mixed-phase TiO₂ nanotubes show excellent antibacterial performance under ultraviolet light irradiation, and their antibacterial ability could be attributed to the oxidative stress induced by the TiO₂ nanotubes. The antibacterial performance of the TiO₂ nanotube coating can be manipulated by photocatalytic activity as well as geometrical characteristics. Our study is the first to reveal the interactions of TiO₂ surface that may increase the potential survival chances of Porphyromonas gingivalis when exposed to an antibacterial drug.

---

1. Introduction

Dental implants are one of the most important methods for treating missing teeth because of their excellent stability and biocompatibility.^1,2 However, dental implants have low mechanical strength and may cause peri-implantitis. In particular, peri-implantitis occurs in the functional bone joint surface, leading to inflammation and transplant failure. Thus, such adverse effects need to be considered to realize the complete application of dental implant materials in biomedical fields. Several factors could contribute to peri-implantitis. First, the low biocompatibility of the implant leads to the formation of small blood vessels at the implant and bone layer interface, which inhibits the antibodies from reaching the surface of the implant and eventually cause reduced resistance around the local host.³ In addition, microbial adhesions and their toxic products could lead to epithelial separation and formation of periodontal pockets when the implant is exposed to an oral environment without the protection of a periodontal membrane and vascular plexus, resulting in a loss of supporting bone and reduction in mechanical stability at the bone–implant interface.⁴ Moreover, bacteria and other microorganisms could adhere to the implant surface, forming a bacterial plaque biofilm, which disables the immune system and antibacterial agent,^5–7 leading to implant failure. Thus, the development of dental implant materials with enhanced biocompatibility and antibacterial performance is urgently needed to fully realize their application in biomedical fields.^7,8

Generally, dental implant materials include metals and alloys and ceramic, carbon, polymer, and composite materials.⁹ Compared with other implant materials, metal alloys, such as ZrO₂, Ti and TiO₂, have been widely used in the field of stomatology and orthopaedics because of their excellent corrosion resistance and appropriate mechanical properties. Among these alloy materials, TiO₂ nanomaterials attract considerable attention due to their great stability and biocompatibility. However, few articles investigate the potential of such functional nanomaterials for use as dental implants. Thus, evaluating the potential application of TiO₂ as dental implant material is urgently needed.^10,11 Because of the absence of antibacterial properties, titanium-based materials need to be modified to achieve an excellent surface decontamination activity. Nanotechnology has been demonstrated as an efficient method for fabricating functional materials with specific properties, including self-cleaning properties.^11–16 In particular, TiO₂-coated titanium structures can reduce bacterial adhesion and display excellent antibacterial properties, which could be attributed to nanoscale surface roughness and high surface energy.^17–19 TiO₂ has the ability to absorb light having a wavelength lower than 400 nm and generate electron (e⁻) and hole (h⁺) pairs, which can produce active hydroxyl free radicals (˙OH) and superoxide anion radicals (O²⁻). The reactive oxygen species can interact with cell walls, leading to the rupturing of bacterial cell membrane or killing of the germs.^20,21 In addition, such reactive oxygen species can break the basic group of DNA strands, disrupt the duplication process of DNA in microbial cells and ultimately cause cell death. Thus, TiO₂ nanotubes (TNTs) have been applied as a potential candidate in the field of antibacterial coating^22–24 and bone transplantation^16,25 because of their highly ordered tubular structure, high specific surface area, high surface roughness and high capacity for drug loading.²²

With regards to orthopedic applications, TiO₂ nanotubes have the ability to selectively adsorb vitronectin and fibronectin, and therefore promote the association of bone reconstruction involving vitro cell adhesion, proliferation and differentiation.^26,27 TiO₂ nanotubes with smaller diameters (30 nm) can improve the adhesion of osteoblasts at the greatest extent; however, such effects were not observed for TiO₂ with large diameters (70 nm). Moreover, bacterial behaviour is also dependent on size and topography when they are cultured on the surfaces of TiO₂ nanotubes. It was observed that the number of bacteria cultured on relatively rough TiO₂ nanotubes surface are significantly reduced, compared with those cultured on the smooth surface of the Ti. In particular, bacteria cultured on nanotubes with diameters of 40–60 nm (ref. 28) and 200 nm (ref. 29) showed the greatest reduction in number. The stress response of germs to the nanotubes could lead to the rupturing of bacterial cell membranes³⁰ and cell death. The nano-interfacial properties of TiO₂ can inhibit the adhesion and proliferation of certain bacteria. However, it was noted that the interaction between nanomaterials and bacteria will increase some bacterial drug resistance via an increase in the horizontal transfer between the resistant genes of bacteria, which may induce a potential threat to global public health. For example, nano-alumina can promote RP4 plasmid from Escherichia coli and Salmonella joints and lead to the horizontal transfer of drug-resistant genes.³¹ Therefore, the potential risk assessment of the nano-effect should arouse significant concerns. Porphyromonas gingivalis can generate virulence factors, which could result in the loss of periodontal attachment levels. Thus, in this study, we use P. gingivalis 381 to investigate the bacterial responses to the change of calibers and related the drug resistance under various reagent concentrations. Consequently, we could evaluate the antibacterial property and potential threat of TiO₂-based functional nanomaterials. Our work could provide guidelines for the development of novel dental implant materials.

2. Experiments

2.1. Preparation and characterization of TiO₂ nanotubes

TiO₂ nanotubes were fabricated using commercially available pure titanium sheets (Φ 32 × 1.2 mm²). The titanium sheets were mechanically polished and then cleaned with acetone, ethanol and deionized water, followed by drying in air; these are defined as “conventional Ti”. A two-step anodic oxidation method was used to prepare TNT coatings. The cleaned titanium sheets were oxidized in glycol containing 0.3 wt% ammonium fluoride at 60 V for 5 h. A platinum sheet was used as the cathode. Both platinum and Ti were connected to a DC power supply (SKD-1105A, SAKO) through copper wires. The samples were cleaned with ethanol and deionized water and dried in air. Furthermore, the samples were oxidized in glycol containing 0.3 wt% ammonium fluoride at various voltages (20–60 V) and time (6–30 h). Finally, the cleaned Ti sheet and TNT were annealed at 450 °C for 4 h (Brother vacuum tube furnace, China) with a heating/cooling rate of 2 °C min⁻¹. The surface morphology of the prepared TNT was observed by field emission scanning electron microscopy (FE-SEM, ULTRA PLUS, Zeiss, Germany). In particular, the samples in which bacteria were cultured were washed with phosphate buffered saline (PBS) and immersed in glutaraldehyde for 12 h at 4 °C. The samples were then dehydrated with ethanol and dried in air. Finally, the morphological bacteria on the treated TNT were observed using FE-SEM. The crystalline characteristics and phase formation of the prepared TNT were analyzed by X-ray diffraction analysis (XRD, Philips X'Pert PRO, USA) with Cu-Kα as the radiation source, and the scan range was set from 20° to 90°. The water contact angles of the TNT with different diameters were measured by image collection and analysis systems (Dataphysics OCA20, Germany). Water contact angle measurements were performed at room temperature with distilled water as the determining medium. Each experiment was repeated five times, and the data are expressed as mean ± standard deviation.

2.2. Bacterial culture

P. gingivalis 381 (China Medical University Hospital of Stomatology) were inoculated onto BHI blood agar plates (Brain Heart Infusion, OXOID, UK) and revived (80% N₂, 10% H₂, and 10% CO₂, anaerobic condition) in a constant temperature incubator (Shanghai Jing macro laboratory equipment Co., Ltd) for 5 days. The bacteria were then inoculated onto BHI broth (OXOID, UK) and cultured under anaerobic conditions for 48 h. The samples were collected after centrifugation at 3000 rpm for 5 min at 4 °C (1-15 k, SIGMA, Germany). Collected samples were then suspended in a BHI liquid medium, and bacterial solutions of 1 × 10⁹ CFU L⁻¹ were prepared by McFarland nephelometry.

2.3. Bacteria cultured on different substrates with metronidazole

Prior to inoculation, the conventional Ti and TNT were sterilized for 6 h under an UV light with a power of 20 W. Different concentrations of metronidazole (groups A and B) were used to study the effect of TNT on P. gingivalis. The metronidazole (China Pharmaceutical and Biological Products) solution was diluted with BHI broth and mixed with bacterial solution for preparing bacterial samples. Prior to the experiment, the minimum inhibitory concentration (MIC) of metronidazole for P. gingivalis was first determined using the classical drug susceptibility testing method (agar dilution method). MIC₉₀ was about 10 mg L⁻¹, which was set as the lower limit of the drug concentration in the experiment. The final concentration of metronidazole in groups A and B were 10 mg L⁻¹ and 15 mg L⁻¹, respectively. TNT samples with different diameters and titanium sheets (a total of six) were placed into 6-well plates (Corning, USA) for each group. The bacterial solution with metronidazole concentrations of 10 mg L⁻¹ (group A) or 15 mg L⁻¹ (group B) was added to each 6-well plate (GIISON, French). The bacteria were cultured (80% N₂, 10% H₂, 10% CO₂, anaerobic environment) at a constant temperature (37 °C) for 48 h. 2 ml bacterial solution was withdrawn from each 6-well plate and measured under a UV spectrophotometer (UV765, Shanghai Precision and Scientific Instrument Co., Ltd) to determine the corresponding bacteria density. Each experiment was repeated three times, and the data are expressed as mean ± standard deviation.

3. Results

The obtained morphology of TNT (annealed) is shown in Fig. 1. The nanotubes with different calibers could be obtained by altering parameters such as oxidation time and anodizing voltage. TNTs with different diameters (namely, 30 nm, 50 nm, 60 nm, 80 nm, and 100 nm) were fabricated. Annealing treatment made little difference to the diameter of the nanotube.

Fig. 1 FE-SEM images of TNT with different diameters: (a) 30 nm, (b) 50 nm, (c) 60 nm, (d) 80 nm, (e) 100 nm, and (f) titanium sheet.

The XRD pattern of TNT is shown in Fig. 2. Without heat treatment, the nanotubes exhibited an amorphous structure, and only the diffraction peak of Ti could be observed in the XRD spectrum (as shown in Fig. 2a), whereas in the XRD pattern of annealed TNT, the characteristic peaks of 101 and 110 revealed the existence of anatase and rutile structures (as shown in Fig. 2b).

Fig. 2 XRD patterns of (a) unannealed TNT and (b) annealed TNT.

The water contact angle data of TNT and titanium sheet are shown in Fig. 3. Water contact angle is an important indicator for assessing surface hydrophobicity and hydrophilicity. The titanium sheet exhibited hydrophobicity compared with the TiO₂ nanotubes. In addition, nanotubes exhibited obvious hydrophilic properties with the increase of caliber, and the lowest contact angle was obtained for the nanotubes with a diameter of 100 nm.

Fig. 3 The water contact angles of titanium sheet and TiO₂ nanotubes with different diameters.

Bacterial densities on various culture substrates of groups A and B are shown in Fig. 4. The bacteria cultured on the Ti sheet or TiO₂ nanotubes showed reduced numbers compared with the blank sample for both the groups. In particular, for the bacteria cultured on the TNT, the number of bacteria first decreased with the increase of the diameter of TNTs, and the lowest density was achieved when its diameter was 60 nm. In addition, it is interesting to note that TNT with a diameter of 60 nm appeared to be an inflexion. The number of surviving bacteria increased when the cell was cultured on TNT with a diameter of 60–100 nm. Fig. 5 and 6 show the SEM images of bacteria in groups A and B (drug group). For group A, which had lower concentrations of metronidazole, greater bacterial adhesion and fusion was found on the nanotubes. However, for group B, which contained higher concentrations of metronidazole, the bacterial growth was significantly inhibited. Moreover, it is interesting to note that bacteria cultured on TNTs with a diameter of 60 nm showed shrinking morphology, which was significantly different from the others.

Fig. 4 Bacterial concentrations of different samples: (a) group A (metronidazole: 10 mg L⁻¹); and (b) group B (metronidazole: 15 mg L⁻¹).

Fig. 5 FE-SEM images of bacteria on different substrates in group A. (a and b): titanium sheet, (c and d): 30 nm, (e and f): 60 nm, (g and h): 100 nm.

Fig. 6 FE-SEM images of bacteria on different substrates in group B. (a and b): titanium sheet, (c and d): 30 nm, (e and f): 60 nm, (g and h): 100 nm.

4. Discussion

TNTs increase the surface area and roughness of the titanium surface, promoting the adhesion of certain cells.³² However, for bacteria, it was proven that TNTs can induce a significant reduction in the number of cells adhering to its surface.¹⁸ It is known that TiO₂ nanotubes have a greater surface energy than the unmodified Ti.¹⁷ By analyzing the wetting angle experimental results, we can conclude that the water contact angle of TiO₂ nanotubes increased, indicating the enhanced hydrophilicity of the samples, which corresponds with the increased diameters of the TiO₂ nanotubes. Currently, opinions on how materials' hydrophilic/free nature regulates the cellular behavior of bacteria are inconsistent. Some previous reports demonstrated that the enhanced hydrophilicity of the surface can promote bacterial adhesion and proliferation, and the bacteria are expected to grow with the increase in the diameter of TiO₂ nanotubes.^33,34 However, in our experiment, the number of bacteria first decreased then increased with increasing diameter of TiO₂ nanotubes, which indicated that there are complex factors that regulate bacterial behavior at the biocompatible interface. In this study, the TNTs exhibited obviously different antibacterial performance when compared with the blank and titanium sheet samples, which could be attributed to sterilization and their nanoscale effects. When the nanotubes are sterilized by ultraviolet irradiation, they produce highly oxidative photo-generated holes (h⁺), which can react with H₂O and O₂ to further generate active groups with strong oxidization ability such as hydroxyl radicals (˙OH) and superoxide free radicals (O²⁻).³⁵ These active groups can interact with proteins, nucleic acids and bacterial cell membranes, causing damage to their molecular structures.^36,37 In addition, the reactive oxygen generated by TiO₂ can increase lipid peroxidation and the amount of lactate dehydrogenase, causing DNA damage, leading to cell death.³⁸ In this process, the unsaturated fatty acids in the cell membrane can also be oxidized by ˙OH group, which would affect the fluidity and structure of the cell membrane. Subsequently, the thickness of the cell membrane would decrease and lead to cell apoptosis.^39,40 Thus, the interactions between TNTs and bacteria would eventually lead to cell death with the generation of reactive oxidation groups under UV light excitation, and it is worth noting that the mixed crystallites (anatase and rutile) exhibit an enhanced ability to produce photo-generated holes (h⁺), resulting in a stronger antibacterial effect than that of amorphous TiO₂. Thus, the generation of oxidative stress by the UV light irradiation of TiO₂ has contributed to their antibacterial effect and nanotoxicity.

The response behaviours of bacteria to TNT with different geometrical characteristics were also investigated. Previous studies demonstrated that the antibacterial property of TNT was mainly affected by the number of active groups, which are correlated with the specific surface areas of the functional materials. The antibacterial materials with increased specific surface areas displayed enhanced decontamination ability.⁴¹ In particular, the specific surface area of the nanotubes is correlated with its diameter;⁴¹ thus, nanotubes with increased pipe diameters are expected to have improved bacterial growth inhibition effects. It was noteworthy that the bacterial survival on TNT decreased corresponding to the increase in the diameter ranging from 40 nm to 60 nm. In particular, TNT with a diameter of 60 nm has a higher density and thinner nanotube wall, which is beneficial for inhibiting hole electron recombination and displaying improved photocatalytic ability than that of TNT with nanotube diameters of 30 nm and 50 nm;²⁹ thus, the most significant antibacterial effect was expected to be observed.⁴² However, for bacteria cultured on TNT with diameters from 60 nm to 100 nm, the number of bacteria began to increase. The observed improved growth behaviour of bacteria suggested that bacterial survival rate on TNT is determined by multiple factors. Previous studies indicated that materials at nanoscale have specific effects that can trigger a series of changes in biological behavior on cells.^42–44 In this study, P. gingivalis on TNT also exhibited a nanoscale-dependent behavior. TNT with diameters ranging from 30 nm to 60 nm could cause physical impact on cells via mechanical contact to the cell membrane because of the cylindrical shape and high aspect ratio.^45,46 The physical interaction could induce the rupturing of the cell membrane, causing cytoplasm outflow, and eventually leading to cell apoptosis. However, the continuous increase in the diameter of TNTs (i.e. 60–100 nm) could directly reduce the mechanical contact effects of cells and show decreased inhibition on bacterial growth.⁴⁷ Thus, bacterial survival rates are dominated by the photocatalytic oxidation ability of TNT and the dependence of bacteria on nano-geometrical characteristics. It was interesting to note that the concentration of bacterial culture on TNT with a diameter of 80 nm in group A was slightly higher than that of the control group. This could be attributed to the fact that the drug inhibition effect on bacteria in group A was weak due to the relatively low concentration of the drug (close to MIC₉₀) and the potential impact of TNT on metronidazole (discussed in the subsequent Section). In addition, TNT with larger diameters (80 nm) could weaken the physical impact effect and result in bacterial increase. The exact mechanism will be investigated in subsequent experiments.

The potential disadvantage on the drug resistance caused by nanotubes is also discussed in this study. Bacteria can obtain exogenously resistant genes mainly via three ways: transformation, transduction and joint. The joint is the most effective and common route of horizontal transfer, which refers to DNA transfer between cells with the aid of plasmids.⁴⁸ Previous studies have found that nano-alumina can promote the conjugational transfer of multi-resistant plasmid RP4 from Escherichia coli to Salmonella joint (200 times of the untreated cells). Under some conditions, nanomaterials can promote the horizontal transmission of resistant genes between different strains, and lead to an increase in bacterial drug resistance.³¹ Thus, to evaluate the effect of TNT on bacterial drug resistance properties, we tested the antibacterial properties of TNTs towards bacteria treated with drugs at various concentrations. The results suggested that when exposed to low concentrations of drugs, the antibacterial effects of TNTs were significantly lower than the blank group, and most of them were even lower than that of the control. Under the exposure of relatively high concentration of drug, the antibacterial effects of all the samples exhibited a certain improvement. However, the antibacterial effects of TNT coatings were still lower than the blank and control groups. The good antibacterial effects of the TNT coatings with a diameter of 50 nm might be attributed to the effect of nanometer size. Oxidative stress induced by the TNT coating can change the fluidity and structure of the cell membrane of P. gingivalis,⁴⁰ thus promoting the transfer and transmission of plasmid containing resistance genes between the cells,^49,50 eventually resulting in an increased performance of bacterial drug resistance. There might be two reasons for the relatively low antibacterial effect of metronidazole on the TNT coating. First, it had been reported that the photocatalytic ability of TiO₂ could be used to degrade the metronidazole under the visible light.⁵¹ This can reduce the concentration of metronidazole to a certain extent, which will weaken the drug effect. Second, the oxidative stress of the nanotubes can restrain the sterilization effect of metronidazole. Metronidazole enters into the bacteria cell, and its nitro group can be easily reduced to amino group by electron transfer protein under an anaerobic environment. This can prevent the synthesis of bacterial DNA and degrade the synthesized DNA to achieve the antibacterial effect.⁵² However, the TNT-produced photo-generated holes (h⁺) have strong oxidation ability under illumination, which could restrain the reduction process by the production of an immobile amino group. Therefore, the TNT coating might weaken the inhibition effect of metronidazole on bacterial growth. However, for the further validation of the hypothesis, related investigations should be implemented in the future studies.

5. Conclusions

In this study, a two-step method was used to prepare TNT coatings with different diameters, and the antibacterial properties of different TNT coatings and their effects on bacterial resistance were studied. The biological behaviours of bacteria on the surface of the TNT coating appear to be affected by both oxidative stress and nanometer size effect. In addition, the nanotubes could be inclined to attenuate the antibacterial effect of metronidazole. The argument would be bolstered through further investigations.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (50872019, 51002027), post-doctoral foundation of China (2013M530930), scientific foundation of the educational department of Liaoning Province (L2012084), the state key laboratory of metal matrix composite, Shanghai Jiaotong University (no. mmckf1411), and the Basic Scientific Research Foundation of Central College (N130402001).

Notes and references

1. T. Albrektsson, G. Zarb, P. Worthington and A. Eriksson, Int. J. Oral. Maxillofac. Implants, 1986, 1, 11.
2. M. Esposito, J. M. Hirsch, U. Lekholm and P. Thomsen, Int. J. Oral. Maxillofac. Implants, 1999, 4, 473.
3. K. S. Brammer, S. Oh, C. J. Cobb, L. M. Bjursten, H. van der Heyde and S. Jin, Acta Biomater., 2009, 5, 3215.
4. T. Nakamura, A. Amano, I. Nakagawa and S. Hamada, FEMS Microbiol. Lett., 1999, 175, 267.
5. M. M. Danser, A. J. vanWinjelhoff and U. vanderVelden, J. Periodontol., 1997, 68, 209.
6. S. Kotsovilis, I. K. Karoussis, M. Trianti and I. Fourmousis, J. Clin. Periodontol., 2008, 35, 621.
7. G. Tawil, R. Younan, P. Azar and G. Sleilati, Int. J. Oral. Maxillofac. Implants, 2008, 23, 744.
8. B. H. Lee, Y. D. Kim, J. H. Shin and K. H. Lee, J. Biomed. Mater. Res., 2002, 61, 466.
9. Z. Wang, Y. Shen and M. Haapasalo, Dent. Mater., 2014, 30, E1–E16.
10. X. Liu, P. K. Chu and C. Ding, Mater. Sci. Eng., R, 2004, 47, 49.
11. T. N. Kim, A. Balakrishnan, B. Lee, W. Kim, B. Dvorankova and K. Smetana, et al., J. Mater. Sci.: Mater. Med., 2008, 19, 553.
12. C. B. Dong, Z. J. Yan, J. Kokx, D. B. Chrisey and C. Z. Dinu, Appl. Surf. Sci., 2012, 258, 9218.
13. A. S. Campbell, C. B. Dong, F. K. Meng, J. Hardinger, G. Perhinschi, N. Q. Wu and C. Z. Dinu, ACS Appl. Mater. Interfaces, 2014, 6, 5393.
14. Q. Xu, M. T. Li, J. B. Niu and Z. H. Xia, Langmuir, 2013, 29, 13743.
15. C. Liu, A. Tian, H. Yang, Q. Xu and X. Xue, Appl. Surf. Sci., 2013, 287, 218–222.
16. A. Tian, X. Qin, A. Wu, H. Zhang, Q. Xu, D. Xing, H. Yang, B. Qiu, X. Xue, D. Zhang and C. Dong, Int. J. Nanomed., 2015, 10, 2423–2439.
17. S. D. Puckett, E. Taylor, T. Raimondo and T. J. Webster, Biomaterials, 2010, 31, 706.
18. G. Colon, B. C. Ward and T. J. Webster, J. Biomed. Mater. Res., Part A, 2006, 78, 595.
19. V. K. Truong, R. Lapovok, Y. S. Estrin, S. Rundell, J. Y. Wang and C. J. Fluke, et al., Biomaterials, 2010, 31, 3674.
20. T. Saito, T. Iwase, J. Horie and T. Morioka, J. Photochem. Photobiol., B, 1992, 14, 369.
21. H. Aita, N. Hori, M. Takeuchi, T. Suzuki, M. Yamada and M. Anpo, et al., Biomaterials, 2009, 30, 1015.
22. K. M. Kummer, E. N. Taylor, N. G. Durmas, K. M. Tarquinio, B. Ercan and T. J. Webster, J. Biomed. Mater. Res., Part B, 2013, 101, 677.
23. A. Roguska, A. Belcarz, T. Piersiak, M. Pisarek, G. Ginalska and M. Lewandowska, Eur. J. Inorg. Chem., 2012, 32, 5199.
24. L. Zhao, H. Wang, K. Huo, L. Cui, W. Zhang and H. Ni, et al., Antibacterial nano-structured titania coating incorporated with silver nanoparticles, Biomaterials, 2011, 32, 5706–5716.
25. T. J. Webster and J. U. Ejiofor, Biomaterials, 2004, 25, 4731.
26. S. Oh, C. Daraio, L. H. Chen, T. R. Pisanic, R. R. Finones and S. Jin, J. Biomed. Mater. Res., Part A, 2006, 78, 97.
27. L. M. Bjursten, L. Rasmusson, S. Oh, G. C. Smith, K. S. Brammer and S. Jin, Titanium dioxide nanotubes enhance bone bonding in vivo, J. Biomed. Mater. Res., Part A, 2010, 92, 1218–1224.
28. B. Ercan, E. Taylor, E. Alpaslan and T. J. Webster, Nanotechnology, 2011, 22, 295102.
29. H. R. Li, Q. Cui, B. Feng, J. Wang, X. Lu and J. Weng, Appl. Surf. Sci., 2013, 284, 179.
30. S. Kang, M. Herzberg, D. F. Rodrigues and M. Elimelech, Langmuir, 2008, 24, 6409.
31. Z. Qiu, Y. Yu, Z. Chen, M. Jin, D. Yang and Z. Zhao, et al., Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 4944.
32. H. Zhou, X. Gan, J. Wang, X. L. Zhu and G. X. Li, Anal. Chem., 2005, 77, 6102.
33. J. Ji and W. Zhang, J. Biomed. Mater. Res., 2009, 88, 448–453.
34. J. H. Lee, S. J. Lee, G. Khang and H. B. Lee, J. Biomater. Sci., Polym. Ed., 1999, 10, 283–294.
35. G. Storz and J. A. Imlayt, Curr. Opin. Microbiol., 1999, 2, 188.
36. R. Kohen and A. Nyska, Toxicol. Pathol., 2002, 30, 620.
37. A. M. Gallardo-Moreno, M. A. Pacha-Olivenza, M. C. Fernández-Calderón, C. Perez-Giraldo, J. M. Bruque and M. L. González-martin, Biomaterials, 2010, 31, 5159.
38. K. J. Davies, M. Delsignore and S. Lin, J. Biol. Chem., 1987, 262, 9902.
39. Q. Xiao and L. Ouyang, J. Phys. Chem. Solids, 2011, 72, 39.
40. S. S. Soni, G. S. Dave, M. J. Henderson and A. Gibaud, Thin Solid Films, 2013, 531, 559.
41. Q. Cui, B. Feng, W. Chen, J. X. Wang, X. Lu and J. Weng, J. Inorg. Mater., 2010, 25, 916.
42. I. Demetrescu, C. Pirvu and V. Mitran, Bioelectrochemistry, 2010, 79, 122.
43. K. K. Sethi, I. V. Yannas, V. Mudera, M. Eastwood, C. Mcfarland and R. A. Brown, Wound Repair Regen., 2002, 10, 397.
44. K. M. Woo, V. J. Chen and P. X. Ma, J. Biomed. Mater. Res., Part A, 2003, 67, 531.
45. S. Kang, M. Pinault, L. D. Pfefferle and M. Elimelech, Langmuir, 2007, 23, 8670.
46. Y. J. Tang, J. M. Ashcroft, D. Chen, G. Min, C. H. Kim and B. Murkhejee, et al., Nano Lett., 2007, 7, 754.
47. S. Grigorescu, C. Ungureanu, R. Kirchgeorg, P. Schmuki and I. Demetrescu, Appl. Surf. Sci., 2013, 270, 190.
48. S. H. So, K. Miyahara and Y. Ohshima, Genes Cells, 2011, 16, 639.
49. J. W. Beaber, B. Hochhut and M. K. Waldor, Nature, 2004, 427, 472.
50. E. Maiques, C. Ubeda, S. Campoy, N. Salvador, I. Lasa and R. P. Novick, et al., J. Bacteriol., 2006, 188, 2726.
51. S. S. Boxi and S. Paria, RSC Adv., 2014, 4, 37752.
52. P. Poulet, D. Duffaut and J. P. Lodter, J. Clin. Periodontol., 1999, 26, 261.

---

Footnote

† The authors contributed equally to this work.

---