Ag-doping regulates the cytotoxicity of TiO2 nanoparticles via oxidative stress in human cancer cells

We investigated the anticancer potential of Ag-doped (0.5–5%) anatase TiO2 NPs. Characterization study showed that dopant Ag was well-distributed on the surface of host TiO2 NPs. Size (15 nm to 9 nm) and band gap energy (3.32 eV to 3.15 eV) of TiO2 NPs were decreases with increasing the concentration of Ag dopant. Biological studies demonstrated that Ag-doped TiO2 NP-induced cytotoxicity and apoptosis in human liver cancer (HepG2) cells. The toxic intensity of TiO2 NPs was increases with increasing the amount of Ag-doping. The Ag-doped TiO2 NPs further found to provoke reactive oxygen species (ROS) generation and antioxidants depletion. Toxicity induced by Ag-doped TiO2 NPs in HepG2 cells was efficiently abrogated by antioxidant N-acetyl-cysteine (ROS scavenger). We also found that Ag-doped TiO2 NPs induced cytotoxicity and oxidative stress in human lung (A549) and breast (MCF-7) cancer cells. Interestingly, Ag-doped TiO2 NPs did not cause much toxicity to normal cells such as primary rat hepatocytes and human lung fibroblasts. Overall, we found that Ag-doped TiO2 NPs have potential to selectively kill cancer cells while sparing normal cells. This study warranted further research on anticancer potential of Ag-doped TiO2 NPs in various types of cancer cells and in vivo models.

Wide-spread application of TiO 2 nanoparticles (NPs) have been increasing due to their chemical stability, photocatalytic efficiency and low cast 1 . The TiO 2 NPs are being utilized in daily life products such as sunscreens, paints and plastics 2 . Due to ever increasing market demand the annual production of TiO 2 NPs is predicted to reach around 2.5 million tons by 2025 3 . There is also growing interest of TiO 2 NPs in biomedical fields including drug delivery, cell imaging, photodynamic therapy and biosensor [4][5][6] . However, investigations have shown the conflicting results regarding the biological response of TiO 2 NPs. Several studies found that TiO 2 NPs induce inflammation, cytotoxicity and genotoxicity [7][8][9] . Contrary, several reports showed that TiO 2 NPs were not toxic or least toxic to several cell lines [10][11][12] . Conflicting reports on toxicological response of TiO 2 NPs could be due to utilization of different physical and chemical properties of this material 2,13 . In general, anatase and rutile are two crystalline forms of TiO 2 . Anatase TiO 2 NPs have high photocatalytic activity and more biologically active than those of rutile one 14,15 .
Photocatalytic activity of TiO 2 NPs is thoroughly investigated because of their applications in solar energy, environmental remediation and photodynamic therapy (PDT) 16,17 since its breakthrough in 1980 s 18 . Under light irradiation, the valence band electrons (e − ) of TiO 2 become excited and moved to conduction band leaving positive charge holes (h + ). The electrons (e − ) in conduction band and holes (h + ) in valence band have the capability to generated cellular reactive oxygen species (ROS) 19,20 . Light induced ROS generation by a photosensitizer has been applied in treatment of several diseases called PDT 21,22 . Potential of TiO 2 NPs to be applied in PDT for different types of cancers, such as leukemia, cervical, liver and lung cancers is already reported 23,24 . Still, there are some drawbacks in the application of TiO 2 NPs for PDT. The major drawbacks of TiO 2 are wide band gap (3.2 eV for anatase) that can activate only in the ultraviolet (UV) region and high rate of electrons-holes (e − /h + ) recombination that reduce considerably the photocatalytic efficiency of TiO 2 NPs 25,26 .
Recent studies have now focused on the improvement of photocatalytic activity of TiO 2 NPs. Attempts to achieve this goal is depends on doping of TiO 2 NPs with metallic or non-metallic elements 27,28 . Doping can reduce the band gap of TiO 2 NPs that extend their spectral response in visible wavelengths 29 . For example, doping of TiO 2 NPs with noble metals such as Ag, Au or Pt can efficiently decrease the e − /h + pair's recombination to enhance the photocatalytic activity and simultaneously extend their light response towards the visible region because of their d electron configuration 30 . Among these Ag-doped TiO 2 NPs has been thoroughly studied because of the dual function of Ag sites. First, Ag serves as an electron scavenging center to separate e − /h + pairs because its' Fermi level is below the conduction band of TiO 2 30,31 . Second, Ag NPs have the ability to create surface plasmon resonance (SPR) effect of TiO 2 NPs, thus leading to the distinctly enhanced photocatalytic activity of TiO 2 NPs in visible region. However, application of Ag-doped TiO 2 NPs in cancer therapy is not explored yet.
ROS generating potential of Ag-doped TiO 2 NPs under visible light have been recently investigated in killing of microbial communities 32,33 . However, some studies have shown that Ag-doped TiO 2 can kill bacteria without any light illumination 34,35 . This could be possible because Ag-doping tunes band gap (e − /h + recombination) of TiO 2 NPs that enhances the catalytic activity to generate ROS within bacterial cells without light illumination. Therefore, ROS generating potential of Ag-doped TiO 2 NPs can be applied in treatment of cancer without the illumination of any light. Manipulating intracellular ROS level by redox modulators is a possible way to harm cancer cells selectively without affecting the normal cells [36][37][38][39] . Therefore, we explored the anticancer potential of Ag-doped TiO 2 NPs via ROS pathway. Using Ag-doped TiO 2 NPs without light in the treatment of cancer have some advantages over PDT 40 . For example, visible light used in PDT cannot travel very far through body tissue. Therefore, PDT is used to treat to the problem on or just under the skin on the lining of some internal organs or cavities. Metastasized cancer also cannot treat with PDT due to the inability of the light source to penetrate large tumors or reach areas where cancer may have spread. Hence, Ag-doped TiO 2 NPs can have advantage of other exposure routes such as oral or intravenous injection. In this study, we investigated the cytotoxicity mechanisms of Ag-doped TiO 2 NPs in human liver cancer (HepG2) cells. To avoid cell type-specific response we have also employed human lung (A549) and breast cancer (MCF-7) cells to assess the anticancer effect of Ag-doped TiO 2 NPs. We have chosen these cancer cell lines because of the lung, liver and breast cancers are life menacing disease and the occurrence of these types of cancer are increasing rapidly worldwide [41][42][43] . These cell lines are also well-known in vitro models and have been widely utilized in toxicology and pharmacology studies [44][45][46] . We have also examined the benign nature of Ag-doped TiO 2 NPs on two non-cancerous normal cells; human lung fibroblasts (IMR-90) and primary rat hepatocytes. We observed that Ag-doped TiO 2 NPs selectively kill the cancer cells (HepG2, A549 & MCF-7) without much affecting the normal cells.

Materials and Methods
Preparation of nanoparticles. Pure and Ag-doped TiO 2 NPs were synthesized by sol-gel procedure.
Titanium (IV) isopropoxide Ti[OCH(CH 3 ) 2 ] 4 and silver nitrate (AgNO 3 ) were utilized as precursors. In brief, 0.1 M solution of titanium (IV) isopropoxide was prepared in absolute ethanol. Then, solution was mixed with distilled water and stirred for 2 h to get a clear and transparent TiO 2 solution. The solution was further dried at 100 °C for 48 h to obtain TiO 2 gel. After aging 24 h the TiO 2 gel was filtered and dried. Then, prepared TiO 2 samples were calcined at 400-600 °C for 24 h to get TiO 2 nanopowder. The Ag-doped TiO 2 nanopowder was synthesized by the same method as described above. The only difference was the addition of AgNO 3 into the TiO 2 solution. The dopant Ag concentrations were varied to 0.5, 2.5 and 5.0%, respectively. Characterization of nanoparticles. Crystal structure and phase purity of pure and Ag-doped TiO 2 NPs were assessed by X-ray diffraction (XRD) (PanAnalytic X'Pert Pro) using Cu-K α radiation (λ = 0.15405 nm, at 45 kV and 40 mA). Morphology was examined by field emmission transmission electron microscopy (FE-TEM) (JEM-2100F, JEOL Inc. Japan). Energy dispersive X-ray spectroscopy (EDS) was used to determine the elemental composition. Prepared NPs were also characterized micro-Raman spectroscopy through Horiba Raman system (IY-Horiba-T64000). UV-visible absorption spectra were obtained using a spectrometer (Shimadzu-2550, Japan). Surface composition and oxygen vacancies of the Ag-doped TiO 2 NPs were determined by X-ray photoelectron spectroscopy (XPS) (PHI-5300 ESCA PerkinElmer, Boston, MA). The peak positions were internally referenced to the C 1 s peak at 284.6 eV. Aqueous behaviour (hydrodynamic size and zeta potential) of prepared NPs was assessed in a ZetaSizer Nano-HT (Malvern Instruments, UK).
Cell culture and exposure of nanoparticles. The HepG2, A549, MCF-7 & IMR-90 cell lines were bought from American Type Culture Collection (ATCC) (Manassas, VA). Primary hepatocytes were isolated from rat using collagenase perfusion method 47 .
The DMEM medium supplemented with 10% fetal bovine serum (FBS) and 100 U/ml penicillin-streptomycin was used to culture the MCF-7 cells at 5% CO 2 and 37 °C. At 80-90% confluence, cells were harvested sub-cultured for nanotoxicity parameters. Cells were allowed to attach on the surface of culture flask for 24 h prior to exposure of NPs. Pure and Ag-doped TiO 2 NPs were suspended in DMEM medium and diluted to different concentrations (0.5-200 µg/ml). The NPs suspensions were then sonicated at room temperature for 10 min at 40 W to avoid agglomeration of NPs before exposure to cells. In some parameters, cells were pre-exposed for 1 h with N-acetyl-cysteine (NAC) (10 mM) before co-exposure with or without NPs. Hydrogen peroxide (H 2 O 2 ) (2 mM), buthionine sulphoximine (BSO) (200 µM) or ZnO NPs (50 µg/ml) were also used as positive controls.
Assay of cytotoxicity endpoints. Cell viability against NPs exposure was assessed by MTT and NRU assays. MTT assay was performed according to the protocol of Mossman 48 with some modifications 49 . MTT assay assesses the function of mitochondrial by measuring the potential of living cells to reduce colorless MTT into blue formazon. The formazan was dissolved in acidified isopropanol and absorbance was recorded at 570 nm SCIeNtIfIC REPORTS | 7: 17662 | DOI:10.1038/s41598-017-17559-9 using a microplate reader (Synergy-HT, BioTek). Lysosomal activity (NRU cell viability assay) was performed according to the method of Borenfreund and Puerner 50 with some modifications 51 . Cell membrane damage after NPs exposure was examined by lactate dehydrogenase (LDH) assay. LDH is an enzyme extensively found in the cytosol that converts lactate to pyruvate. Upon cell membrane damage, LDH leaks into extracellular matrix (culture medium). LDH level in culture medium was examined using a BioVision kit (Milpitas, CA). Morphology of cells after exposure to NPs was determined by phase-contrast inverted microscope (Leica) at 10X magnification.
Assay of apoptotic markers. Mitochondrial membrane potential (MMP) was measured using Rh-123 fluorescent dye according to Siddiqui et al. 46 . MMP level was determined by two methods; cell imaging by fluorescent microscopy (OLYMPUS CKX 41) and quantitative assay by microplate reader (Synergy-HT, BioTek). Caspase-3 enzyme activity was determined by BioVision kit (Milpitas, CA). This assay is based on the principle that activated caspases in apoptotic cells cleave the synthetic substrates to release free chromophore p-nitroanilide (pNA) 52 . Cell cycle phases were measured by a Beckman Coulter Flow cytometer (Coulter Epics XL/Xl-MCL) through a FL-4 filter (585 nm) using propiodium iodide (PI) probe 53 . The data were analyzed by Coulter Epics XL/XL-MCL, System II Software.
Assay of oxidative stress markers. Intracellular reactive oxygen species (ROS) generation was assessed utilizing 2,7-dichlorofluorescin diacetate (DCFH-DA) probe as reported elsewhere 54 with few changes 46 . ROS level was determined by two methods; quantitative assay by a microplate reader (Synergy-HT, BioTek, USA) and cell imaging by fluorescent microscopy (OLYMPUS CKX 41). For the measurement of glutathione (GSH) level and superoxide dismutase (SOD) enzyme activity, cell extracts were prepared from the control and treated cells as described earlier 51 . Intracellular GSH level was quantified by Ellman' method 55 using 5,5-dithio-bis-2-nitrobenzoic acid (DTNB). SOD enzyme activity was measured by a kit (Cayman Chemical Company, Michigan, OH).
Protein estimation. Protein level was estimated by Bradford method 56 using bovine serum albumin as standard.

Statistics.
One-way analysis of variance followed by Dunnett's multiple comparison tests were performed for statistical analysis. Significance was ascribed at p < 0.05.

Results and Discussion
TEM analysis. Morphology and structural characterization of pure and Ag-doped TiO 2 NPs were assessed by field emission transmission electron microscopy (FETEM) (Fig. 1). Upper and middle panels of Fig. 1 show low magnification images of pure and Ag-doped TiO 2 NPs. The average particle size of pure TiO 2 NPs was around 15 nm while particle size of Ag-doped (5%) TiO 2 NPs was approximately 9 nm. These results indicated that Ag-doping reduces the size of host TiO 2 NPs. Generally, metal ions doping at optimal level hinders the particles growth. Effect of Ag dopant on TiO 2 NPs size reduction has been attributed to grain-boundary pinning caused by dopant ions, which limits the grain growth by the symmetry-breaking effects of the dopant at the boundary, resulting in smaller size of particles 57 . Reduction in size of NPs after doping was also reported in other studies 49,57 .
High resolution TEM images (lower panel of Fig. 1) clearly shows that dopant Ag was well distributed and decorated on the surface of host TiO 2 NPs. High resolution TEM images also demonstrated that TiO 2 NPs has a high crystalline nature with the plane spacing of 0.353 nm, 0.350 nm, 0.532 nm 0.351 nm, which matches well with (101) plane of anatase TiO 2 . After the combination with different amount of Ag (0.5, 2.5 & 5%), spacing between two adjacent lattice places is about 0.20, 0.21 and 0.22 nm, which corresponds to the (200) lattice distance of Ag (JCPDS: 04-0783). These lattice parameters were in agreement with the X-ray diffraction (XRD) spectra as shown in Fig. 2A.
XRD analysis. XRD measurements were carried out to examine the crystallographic structure of prepared NPs. Figure Figure 2B represents the Raman spectra of pure and Ag-doped (0.5-5%) TiO 2 NPs in the range 100-1200 cm −1 at room temperature. Three peaks with strong intensities are observed around 397 (B1g), 515(A1g), and 637 (Eg) cm −1 , which indicates that all samples were mostly dominated by anatase phase of TiO 2 NPs 58,59 . An interesting observation was that the peak intensities increased with the deposition of Ag, while the position of the Raman signal remained the same, indicating the crystallinity becomes better, which also corresponds to the results of XRD and high resolution TEM. XPS analysis. X-ray photoelectron spectroscopy (XPS) was performed to further characterize chemical composition and elemental status of pure and Ag-doped TiO 2 NPs. Figure 3A shows the typical XPS survey spectra of Ag-doped (5%) TiO 2 NPs. Results showed that Ti, O and Ag elements exist in Ag-doped TiO 2 NPs. Peak located at binding energy of 463.75 eV corresponds to the Ti (2p1/2) and another one located at 460.12 eV is assigned to the Ti (2p3/2) (Fig. 3C). In the O1s region, highest intense peak at 529.8 eV is attributed to the lattice oxygen (Ti-O-Ti) in anatase (Fig. 3D). The Ag3d3 and Ag3d5 peaks indicated the presence of Ag in Ag-doped TiO 2 NPs (Fig. 3B). The binding energies of Ag3d3 and Ag3d5 peaks are 369.5 eV and 371.4 eV, respectively. Our results have a strong agreement with the previous studies 60, 61 . The EDS data also showed that Ti and O were the main elemental species in pure TiO 2 NPs while additional Ag peaks were observed in Ag-doped TiO 2 NPs supporting XPS results (Supplementary Fig. S1).    other studies 49,57 . Tauc Model was employed to determine the optical band gap energy of the aggregates, according to the following equation 49 :

Raman analysis.
where hν is the photon energy, E g is the optical band gap, A is a constant, m is equal to 1/2 for allowed direct optical transitions and α is the absorption coefficient. The band gap values were determined by extrapolating the linear region of the plot to hν = 0. From the Tauc plots of (αhν) 2  Hydrodynamic size and zeta potential. It is essential to characterize the behavior of NPs in aqueous state before their biological studies. We have assessed the zeta potential and particle size of pure and Ag-doped TiO 2 NPs in water and DMEM to get a realistic overview of NPs interaction with cells. We found that hydrodynamic  size of pure and Ag-doped (0.5-5%) TiO 2 NPs was 10-15 time higher than those of sizes calculated from TEM and XRD (primary particle size) ( Table 1). We further noticed that hydrodynamic size of TiO 2 NPs was slightly increases with the incremental of Ag-doping. Higher hydrodynamic size than primary particle size was also reported in other studies 63,64 . In ZetaSizer measurements higher size of NPs was because of tendency of NPs to agglomerate. We further observed little variation in hydrodynamic size of NPs dispersed in DMEM than those of deionized water. This could be due to presence of serum in the culture medium. It is known that serum could bind to NPs and form a protein corona 65 . This protein corona might be responsible for size variation in water and cell culture medium 66 . Protein corona presents on the surface of NPs also influences the interaction of NPs with cells. Zeta potential study suggested that pure and Ag-doped TiO 2 NPs suspended in water had positive charge on the surface, whereas in culture medium NPs had negative surface (Table 1). Differences in surface charge could be due to adsorption of negative charged proteins on the surface of NPs.
Cytotoxicity. Human liver cancer (HepG2) cells were treated with different concentrations (0.5-200 µg/ml) of pure and Ag-doped TiO 2 NPs for 24 h and cell viability was measured by MTT and NRU assays. Both parameters serve as sensitive and integrated tools to measure the cell integrity and cell proliferation inhibition 49,67 . The MTT assay was used to evaluate the mitochondrial function while NRU assay represents the lysosomal activity. Both MTT and NRU data showed that Ag-doped TiO 2 NPs reduced the viable number of cells dose-dependently in the concentration range of 25-200 µg/ml. Besides, cell viability decreases with increasing concentrations of Ag dopant ( Fig. 5A and B). On the other hand, pure TiO 2 NPs did not reduce the viability of HepG2 cells. LDH enzyme leakage in culture medium from cells is also an indicator of NPs penetration into cells 68 . A plenty of studies have shown that LDH level increases in culture medium after exposure to NPs 37,69 . Our results also demonstrated that Ag-doped TiO 2 NPs induced LDH leakage and incremental Ag-doping resulted in higher leakage of LDH enzyme (Fig. 5C). However, pure TiO 2 NPs did induce LDH leakage in HepG2 cells. To support LDH data we further studied the cellular uptake of pure and Ag-doped TiO 2 in HepG2 cells by IC-MS. After exposure of 100 µg/ml pure and Ag-doped (0.5-5%) TiO 2 NPs for 24 h, ICP-MS analysis showed the presence of Ti and Ag elements in HepG2 cells (Supplementary Fig. S2).
We further examined the morphology of HepG2 cells after exposure to Ag-doped (0.5-5%) TiO 2 NPs at a concentration of 100 µg/ml for 24 h. Results demonstrated low cell density and rounding of cells after exposure to Ag-doped TiO 2 NPs as compared to the controls (Fig. 5D). Similar to cell viability and LDH leakage results, morphology data showed that cytotoxic response of TiO 2 NPs increases with increasing the amount of Ag-doping. Our previous study also reported that Zn-doped TiO 2 NPs induced cytotoxicity in human breast cancer (MCF-7) cells 49 . Other studies have also shown that metal ions doping tunes the cytotoxic response of semiconductor metal oxide NPs 62,70,71 . These results were according to other reports demonstrating that pure TiO 2 NPs did not induce cytotoxicity in different types of human cells 72,73 . Apoptosis. Apoptosis is known as a distinct mode of programmed cell death that involves the elimination of genetically damaged cells. Apoptotic cell death occurs as a defense mechanism when cellular DNA is damaged beyond the repair 74 . We studied the MMP level, caspase-3 enzyme activity and cell cycle as markers of apoptosis in HepG2 cells against pure and Ag-doped TiO 2 NPs exposure. MMP level in HepG2 cells were measured after exposure to pure and Ag-doped TiO 2 NPs at the concentration of 25-100 µg/ml for 6 h. MMP level was assayed using Rh-123 fluorescent probe. Quantitative data indicated that Ag-doped TiO 2 NPs caused MMP loss in a dose-dependent manner (Fig. 6A). Fluorescence microscopy images also showed that the brightness of red intensity was decreases with increasing the concentration of Ag-doping (Fig. 6B). Caspase genes are activated during the process of cell death and are known to play critical roles in apoptotic pathway. Studies have shown that caspase-3 gene is imperative for genetic damage and programmed cell death 75 . Our results demonstrated that Ag-doped TiO 2 NPs induced caspase-3 enzyme activity dose-dependently. Besides, caspase-3 enzyme activity was increases with increasing the level of Ag-doping (Fig. 6C). We further studied the cell cycle progression against pure and Ag-doped TiO 2 NPs exposure. It is known that cells with damaged DNA accumulated in gap1 (G1), DNA synthesis (S) or in gap2/mitosis (G2/M) phase. Cells with irreversible damage undergo apoptosis, giving rise to accumulation of cells in sub-G1 phase. Flow-cytometric data demonstrated the induction of apoptosis in HepG2 cells upon exposure to Ag-doped TiO 2 NPs exposure (Fig. 6D). The Ag-doped (5%) TiO 2 NPs (100 µg/ ml for 24 h) resulted in the appearance of a significant 12.8% cells in the sub-G1 phase than those of 6.1% of untreated control cells. A significant decline in G2/M phase was also evident in Ag-doped TiO 2 NPs treated cells. Similar to cytotoxicity results, pure TiO 2 NPs did not induce apoptosis in HepG2 cells.
Oxidative stress. Oxidative stress has been played a critical role in the cytotoxic response of a number of NPs whether by the extreme generation of oxidants (e.g. ROS) or by reduction of antioxidants (e.g. GSH) 52,64,76 . Evidence are rapidly increasing that manipulation of intracellular ROS production can be utilized in killing of cancer cells without much affecting the normal cells 36,39 . Our earlier studies have shown that semiconductor nanoparticles such ZnO have potential to selectively kill cancer cells via ROS generation while sparing the normal cells 37,38,52 . In the present study, we further investigated the regulation of oxidative stress markers (ROS, SOD & GSH) in HepG2 cells upon exposure to 25, 50 and 100 μg/ml of pure and Ag-doped TiO 2 NPs for 6 h. Intracellular ROS generation was assessed by DCFDA fluorescent probe. ROS such as superoxide anion (O 2 •− ), hydroxyl radical (HO • ) and hydrogen peroxide (H 2 O 2 ) elicit a variety of physiological and cellular events including DNA damage and apoptosis 76,77 . Quantitative results demonstrated that ROS level was increases dose-dependently and proportional to the amount of Ag-doping in TiO 2 NPs (Fig. 7A). Fluorescence microscopy images also supporting that the brightness of green probe was higher in Ag-doped TiO 2 NPs in comparison to controls (Fig. 7B). Although, pure TiO 2 NPs did not induce ROS production in HepG2 cells. Superoxide dismutase (SOD) enzyme is acting as front liner in antioxidant defense system. This enzyme catalyses the dismutation of highly reactive superoxide (O 2

•−
) anion into hydrogen peroxides (H 2 O 2 ). We observed dose-dependent reduction in SOD enzyme activity and proportional to the Ag-doping (Fig. 7C). On the other hand, pure TiO 2 NPs did not affect the activity of SOD enzyme. Higher production of intracellular ROS leads to oxidize the cellular biomolecules such as glutathione (GSH), which plays a critical role in maintaining the redox homeostasis through its antioxidant activity. We also found that Ag-doped TiO 2 NPs induced GSH depletion in dose-dependent manner and proportional to the amount of Ag-doping (Fig. 7D).

Ag-doped TiO2 NPs induced cytotoxicity in HepG2 cells via oxidative stress.
In this section, we explored the role of ROS and oxidative stress in cytotoxic response of Ag-doped (5%) TiO 2 NPs in HepG2 cells (Fig. 8). The HepG2 cells were treated with Ag-doped (5%) TiO 2 NPs with or without N-acetyl-cysteine (NAC) or buthionine sulphoximine (BSO). We have also used ZnO NPs or H 2 O 2 as positive controls. Results demonstrated that NAC efficiently averted the ROS generation and SOD depletion caused by Ag-doped TiO 2 NPs, or ZnO NPs ( Fig. 8A and B). BSO was used as positive control for GSH depletion. Besides, NAC exposure restored the GSH in cells treated with Ag-doped TiO 2 NPs or BSO (Fig. 8C). At last, we also found that co-exposure of NAC, effectively abolished the cytotoxicity induced Ag-doped TiO 2 NPs, ZnO NPs or H 2 O 2 (Fig. 8D). Altogether, these results suggested that oxidative stress could be one of the potential mechanisms of toxicity induced by Ag-doped TiO 2 NPs in human liver cancer (HepG2) cells.   . Amount of Ag present in Ag-doped TiO 2 nano-composite did not induced toxicity alone to HepG2, A549 and MCF-7 cancer cells. Cytotoxicity was measured by MTT cell viability assay. Cells were exposed to 0.5, 2.5 and 5 µg/ml of Ag NPs. This is the amount of Ag present in the 100 µg/ml of Ag-doped TiO 2 NPs. Data represented are mean ± SD of three identical experiments made in three replicate.
SCIeNtIfIC REPORTS | 7: 17662 | DOI:10.1038/s41598-017-17559-9 Cytotoxicity and oxidative response of Ag-doped TiO2 NPs in human lung and breast cancer cells. To avoid cell type specific response we have also employed human breast (MCF-7) and lung (549) cancer cells to see the effect of Ag-doped (5%) TiO 2 NPs. Cytotoxicity endpoints (MTT & LDH assays) and oxidative stress markers (ROS & GSH levels) were assessed. We observed that like HepG2 cells, Ag-doped TiO 2 NPs causes reduction in cell viability (Fig. 9A), LDH leakage (Fig. 9B), higher level of ROS (Fig. 9C) and depletion of GSH (Fig. 9D) in MCF-7 and A549 cells. However, pure TiO 2 NPs did not cause toxicity to both types of ells. These results are suggesting that the potential mechanism of toxicity induced by Ag-doped TiO 2 NPs in A549 and MCFcells was comparable to HepG2 cells.
Amount of Ag present in Ag-doped TiO 2 NPs did not cause cytotoxicity alone to human cancer cells. We found that pure TiO 2 NPs did not cause toxic effects to selected human cancer cell lines (HepG2, A549 & MCF-7). However, Ag-doped TiO 2 nano-complex induced toxicity to these cells. To make clear that observed toxic effect was due to exposure Ag-TiO 2 nanocomplex not by Ag alone, we examine the effect of Ag NPs alone in these cell lines. We selected the 0.5, 2.5 & 5 µg/ml of Ag NPs for cytotxicity assays. These amounts of Ag present in 100 µg/ml solution of Ag-doped (0.5, 2.5 & 5%) TiO 2 NPs. We exposed HepG2, A549 and MCF-7 cells with Ag NPs at the concentration of 0.5, 2.5 and 5 µg/ml for time period of 24 h. After the completion of exposure time, cell viability was measured by MTT assay. Results have shown that selected concentration of Ag NPs were not able to exert cytotoxicity to all three types of cancer cells (Fig. 10). These results indicated that Ag-TiO 2 nanocomplex was responsible for cytotoxicity, apoptosis and oxidative stress in cancer cells neither Ag nor TiO 2 alone.
Ag-doped TiO 2 NPs were benign to normal cells. To see the benign nature of Ag-doped TiO 2 NPs toward normal cells, we have examined the effect of Ag-doped (5%) TiO 2 NPs on human lung fibroblasts (IMR-90) and primary rat hepatocytes. Results demonstrated that Ag-doped TiO 2 NPs did induce cytotoxicity and oxidative stress in both types of normal cells (Fig. 11A and B). Other studies have also reported the benign nature of TiO 2 NPs 78-80 . These results suggested that Ag-doped TiO 2 NPs have inherent selective toxicity nature towards cancer cells while posing no effect to normal cells. In previous studies, we also found that ZnO and Al-doped ZnO NPs have the inherent selective killing nature towards cancer cells without posing much effect to normal cells 37,38 . These results suggested that Ag-doped TiO 2 NPs has anticancer activity. Preferential cancer cells killing ability of metal-based NPs are being explored at laboratory level 38,39,71,81,82 .

Conclusions
We found that Ag-doped TiO 2 NPs induced toxicity in human liver cancer (HepG2) cells via oxidative stress. The toxic intensity of Ag-doped TiO 2 NPs was increases with the incremental of Ag level. This is possibly due to the tuning of size and band gap of TiO 2 NPs by Ag-doping. Furthermore, Ag-doped TiO 2 NPs were also induced toxicity to human lung (A549) and breast (MCF-7) cancer cells. On the other hand, Ag-doped TiO 2 NPs spare the normal human lung fibroblasts (IMR-90) and primary rat hepatocytes. Altogether, our data suggested that Ag-doped TiO 2 NPs selectively kill cancer cells while sparing the normal cells. This preliminary report on selective toxicity of Ag-doped TiO 2 nano-complex toward cancer cells warranted further extensive research on various types of cancer and normal cells along with in vivo models.