Interaction of carbohydrate modified boron nitride nanotubes with living cells

Colloids and Surfaces B: Biointerfaces 134 (2015) 440–446

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces journal homepage: www.elsevier.com/locate/colsurfb

Interaction of carbohydrate modified boron nitride nanotubes with living cells Melis Emanet 1 , Özlem S¸en 1 , Zehra C¸obandede, Mustafa C¸ulha ∗ Department of Genetics and Bioengineering, Faculty of Engineering, Yeditepe University, Atas¸ehir, Istanbul 34755, Turkey

a r t i c l e

i n f o

Article history: Received 30 April 2015 Received in revised form 26 June 2015 Accepted 14 July 2015 Available online 19 July 2015 Keywords: Boron nitride nanotubes Carbohydrate modification Cellular uptake Cytotoxicity Genotoxicity Comet assay

a b s t r a c t Boron nitride nanotubes (BNNTs) are composed of boron and nitrogen atoms and they show significantly different properties from their carbon analogues (carbon nanotubes, CNTs). Due to their unique properties including low electrical conductivity, and imaging contrast and neutron capture properties; they can be used in biomedical applications. When their use in biological fields is considered, the route of their toxic effect should be clarified. Therefore, the study of interactions between BNNTs and living systems is important in envisaging biological applications at both cellular and sub-cellular levels to fully gain insights of their potential adverse effects. In this study, BNNTs were modified with lactose, glucose and starch and tested for their cytotoxicity. First, the interactions and the behavior of BNNTs with bovine serum albumin (BSA), Dulbecco’s Modified Eagle’s Medium (DMEM) and DMEM/Nutrient Mixture F-12Ham were investigated. Thereafter, their cellular uptake and the cyto- and genotoxicity on human dermal fibroblasts (HDFs) and adenocarcinoma human alveolar basal epithelial cells (A549) were evaluated. HDFs and A549 cells internalized the modified and unmodified BNNTs, and BNNTs were found to not cause significant viability change and DNA damage. A higher uptake rate of BNNTs by A549 cells compared to HDFs was observed. Moreover, a concentration-dependent cytotoxicity was observed on A549 cells while they were safer for HDFs in the same concentration range. Based on these findings, it can be concluded that BNNTs and their derivatives made with biomacromolecules might be good candidates for several applications in medicine and biomedical applications. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Boron nitride nanotubes (BNNTs) were first synthesized in 1995 by Chopra et al. [1] after the discovery of carbon nanotubes (CNTs) in 1991 by Iijima [2]. Although BNNTs are structural analogues of CNTs, they are reported to be more stable than the CNTs [3]. In fact, BNNTs are gaining increasing attention as novel nanomaterials because of their high oxidative properties, and mechanical and chemical resistance [4]. In addition, BNNTs are considered as good candidates for a wide range of biomedical and related applications such as orthopedic implants [5], biosensing [6], and drug and gene delivery [7]. The high chemical stability and hydrophobicity of BNNTs result with their poor dispersibility in aqueous media [8,9], which hinders their cellular uptake studies and reliable assess-

∗ Corresponding author. Fax: +90 216 578 1587. E-mail address: [email protected] (M. C¸ulha). 1 These authors have equally contributed to this work http://dx.doi.org/10.1016/j.colsurfb.2015.07.036 0927-7765/© 2015 Elsevier B.V. All rights reserved.

ment of their adverse effects on living systems. Therefore, BNNTs were functionalized with several biomolecules such as gum Arabic, glycine, mesoporous silica and europium doped sodium gadolinium fluoride (NaGdF4:Eu) for cellular uptake studies for biomedical applications including drug delivery and tissue engineering [10–13]. The functionalization provides dispersibility in aqueous media and functional groups to visualize them by binding fluorescent dyes or quantum dots in fluorescence imaging studies [14,15]. The cytocompatibility of derivatized BNNTs and their interactions with the living systems were first evaluated by Ciofani et al. [14]. In another study, the glycol–chitosan (GC) non-covalently coated BNNTs exposed human neuroblastoma SH-SY5Y cells were viable up to100 ␮g/mL of GC-BNNTs [16]. In another study, the BNNTs were functionalized with organic hydrophilic agents including glucosamine (GA), poly(ethylene glycol)1000 (PEG1000 ) and chitosan (CH), and their toxicity was investigated [17]. The study found that the PEG1000 -BNNT and CH-BNNT were cytotoxic at high concentrations (100 ␮g/mL) while the GA-BNNT was not cytotoxic. Moreover, the hemocompatibility study of the unmodified BNNTs

M. Emanet et al. / Colloids and Surfaces B: Biointerfaces 134 (2015) 440–446

showed that they were hemocompatible on the malignant U87 (wild type p53) and T98 (mutant p53) glioblastoma, MCF-7 adenocarcinoma mammary gland cells and normal MRC-5 fibroblast lung cells [18]. In another study, the cytotoxicity of Tween 80 coated BNNTs was investigated with several cell lines including A549, murine alveolar macrophage cells (RAW 264.7), murine embryonic fibroblast cells (3T3-L1) and human embryonic kidney cells (HEK293) [19]. The results of that study demonstrated that BNNTs were cytotoxic at variable BNNT concentrations (0.2–20 ␮g/mL) in a cell-typedependent manner. They were enormously cytotoxic especially on macrophage cells due to the high endocytosis capacity of these cells. Furthermore, the biocompatibility of the glycol–chitosan modified BNNTs was investigated in vivo. The BNNTs were intravenously injected in rabbits at 1 mg/mL dose in a first experiment [20]; then, 5 and 10 mg/kg doses were adopted in a second experiment [21]. All the collected results indicated that the BNNTs have no significant adverse effect on the white blood cells, red blood cells and many other blood parameters of the rabbits upon their injection in the blood. In a recent study, the gum Arabic functionalized BNNTs was injected into planarian in order to investigate the influence of the modified BNNTs on the regeneration of planarians. The animals amputated below their heads showed no important morphological and progressive differences [22]. The inconsistent results in the literature cause confusions about the biocompatibility of BNNTs. It is clear that more data is needed to understand their behavior in biological systems. Therefore, the toxicity of the BNNTs should be further investigated by using many other cell models before concluding about their biocompatibility. In this study, the cellular uptake and the cyto- and genotoxicity of unmodified and carbohydrate-modified BNNTs were investigated. Their modification with carbohydrates was performed to increase their dispersibility in aqueous media. Then, their interactions with culture media contents, bovine serum albumin (BSA), DMEM, and DMEM/Ham’s F-12, were investigated using Bradford assay. Finally their cellular uptake and toxicity were evaluated using HDFs and A549 cell lines.

2. Materials and methods 2.1. Hydroxylation and glucose, lactose and starch modification of hydroxylated BNNTs BNNTs were synthesized from colemanite based on a chemical vapor deposition (CVD) method, and purified as reported earlier [23]. For hydroxylation, 100 mg of pure BNNTs were added into 10 mL of 30% H2 O2 solution, and the mixture was sonicated at 25 ◦ C for 1 h. Then, the mixture was refluxed for 48 h at 110 ◦ C while stirring. The obtained hydroxylated BNNTs (h-BNNTs) were precipitated by centrifugation (15 min, 10,000 rpm) and washed with ddH2 O five times and dried at 60 ◦ C. Glucose, lactose and starch-modified BNNTs(m-BNNTs) were synthesized using a previously reported method [24]. Briefly, a suspension was prepared by dispersing 100 mg of h-BNNTs in 10 mL of deoxygenated H2 O, and sonicated for 30 min. 10 mL of 5% w/v glucose, 5% w/v lactose or 2% w/v starch solution were added to the suspension. This suspension was then incubated for 48 h on a magnetic stirrer at 37 ◦ C after adding 500 ␮L of 10% v/v glutaraldehyde. The obtained m-BNNTs were precipitated by centrifugation (30 min, 14,000 rpm) and washed with ddH2 O five times and dried in a vacuum oven at 30 ◦ C. The h-BNNTs and m-BNNTs were analyzed with Fourier transform infrared spectroscopy (FT-IR) and thermogravimetric analysis (TGA).

441

2.2. Interaction of the BNNTs with media and their contents A 1 mg of unmodified or modified BNNTs was added into 1 mg/mL of BSA solution, DMEM or DMEM/Nutrient Mixture F12Ham (Sigma–Aldrich, Germany), and the mixture was sonicated for 2 h in an ice bath. After the sonication, they were shaken in a cold room (4 ◦ C) overnight. Thereafter, they were centrifuged at 14,000 rpm for 30 min at 4 ◦ C, and the supernatant was discarded. This washing procedure was repeated three times. After the washing steps, 1 mL of ddH2 O was added into the BNNTs–BSA, BNNTs–DMEM and BNNTs–DMEM/nutrient mixture F-12Ham, respectively. A 100 ␮L of samples were added into 400 ␮L of Bradford reagent in a 24-well plate, and the mixture was allowed to incubate for 30 min at room temperature. 2.3. Cell culture HDF and A549 cell lines were utilized to assess the cytotoxicity and the genotoxicity of BNNTs. The HDFs were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM), supplemented with 10% Fetal Bovine Serum (FBS) and 1% penicillin streptomycin ampicillin (PSA) antibiotics. The A549 cells were cultured in (DMEM F-12), supplemented with 1% L-glutamine in addition to 10% Fetal Bovine Serum (FBS) and 1% Penicillin Streptomycin Ampicillin (PSA) . The cells were incubated in water-jacketed incubator in a 5% CO2 , 95% air atmosphere at 37 ◦ C. 2.4. Cellular uptake assays The cells were seeded on cover slips in a 6-well plate and incubated for 24 h. A 5-␮L of diamidino-2-phenylindole (DAPI) was added to 1 mL of suspension containing a 50 ␮g/mL of BNNTs, h-BNNTs and m-BNNTs, respectively. They were kept at room temperature for 30 min, and then centrifuged at 14,000 rpm for 30 min at 25 ◦ C. The supernatants were discarded and the precipitates were washed for three more times. 1 mL of cell medium was used to resuspend the BNNT–DAPI, h-BNNT–DAPI or m-BNNT–DAPI pellets at the end of the washing steps. A 5 ␮L DAPI was added into 995 ␮L of medium for the control experiments. The BNNT-DAPI, h-BNNTDAPI or m-BNNT-DAPI included medium were added on the cells that were seeded on the cover slips and incubated for 24 h. After removing the medium from the cells, 1 mL of PBS was added onto the cells and shaken for 5 min in a shaker at room temperature. A 2.5% glutaraldehyde solution was prepared in PBS. After discarding PBS from the cells, 2.5% glutaraldehyde solution was added and incubated for 30 min at 4 ◦ C. After the fixation, the cells were dehydrated through ethanol washing series (50, 70, 90, and 100%) for 5 min per solution. Samples were examined with a Zeiss LSM 700 confocal laser-scanning microscope. 2.5. Biocompatibility assays 2.5.1. Cell viability assay The cytotoxic effects of the modified and unmodified BNNTs on cells were quantified with 2-(2-methoxy-4-nitrophenyl)3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium (WST-1) colorimetric assay. First, HDFs were seeded at 5 × 103 cells/well and A549 cells were seeded at 1 × 104 cells/well in 96-well plates, and incubated for 24 h. The cells were then treated with the BNNTs, h-BNNTs and m-BNNTs at increasing concentrations (5 ␮g/mL, 10 ␮g/mL, 20 ␮g/mL, 50 ␮g/mL, 100 ␮g/mL and 200 ␮g/mL). The concentration of the modified BNNTs was adjusted based on the TGA data (see Fig. 2c). After 1–3 days of incubation the culture medium in 96-well plates was replaced with fresh culture medium containing WST-1 reagent with 1:10 ratio and incubated for a further 1 h. The percentage of living cells is calculated by measuring

442

M. Emanet et al. / Colloids and Surfaces B: Biointerfaces 134 (2015) 440–446

Fig. 1. A schematic view of chemical reaction of hydroxylated BNNTs with a lactose molecule.

absorbance of formed formazan salts at 450 nm with an ELISA Plate Reader. 2.5.2. Reactive oxygen species (ROS) detection assay The ROS detection test was applied to the BNNTs, h-BNNTs, and m-BNNTs exposed A549 cells. The cells were seeded in 24-well plates (the number of the cells was 1 × 105 for each well) and incubated for 24 h at 37 ◦ C. In the following day, the cells were treated with ROS detection reagent (2 ,7 dichlorofluorescindiacetate, DCFDA) and incubated for 30 min at 37 ◦ C. After diffusion into the cells, the DCFDA is transformed to a non-fluorescent compound by deacetilation. Then, the ROS detection reagent containing medium was removed and the cells were incubated for 4 h at 37 ◦ C with the increasing concentrations (5 ␮g/mL, 10 ␮g/mL, 20 ␮g/mL, 50 ␮g/mL, 100 ␮g/mL and 200 ␮g/mL) of BNNTs, h-BNNTs and mBNNTs including medium. The non-fluorescent compound in the cells was oxidized into the fluorescent 2 ,7 -dichlorofluorescein (DCF) by the presence of ROS in the cells. Finally, the cells were analyzed with a flow cytometer. 2.5.3. Genotoxicity assay The genotoxicity of BNNTs, h-BNNTs, and m-BNNTs on A549 cells were analyzed by comet assay. The cells were treated with all the samples at a 100 ␮g/mL and incubated for 24 h. Following the BNNT exposure, the cells were collected into 1 mL PBS and embedded in 1% low melting agarose onto high melting agarose coated slides, which were kept into refrigerator at 4 ◦ C for 30 min. Then, the slides were put into lysis solution (1% Triton X, 10 mL DMSO, 2.5 M NaCl, 0.1 EDTA, 10 mM Tris base, pH 10) for 1 h at 4 ◦ C. The cells were denaturated with alkaline buffer for 40 min; then, electrophoresis was performed at 25 V and 300 mA for 20 min with alkaline buffer. The slides were immersed in neutralization buffer (0.5 M Tris–HCl, pH 7.5) for 15 min at 4 ◦ C. Finally, the samples were fixed into 70% ethanol for 10 min. The DNA was stained with SYBR green dye at the 4 ◦ C for 30 min. After all these procedures, the comet images were taken by florescence microscopy and the tail length of the DNA was measured by comet image analysis (Comet IV). Fifty comets were analyzed for BNNTs, h-BNNTs and m-BNNTs sample. 3. Results and discussions 3.1. HYdroxylation and glucose, lactose and starch modification of the hydroxylated BNNTs High hydrophobicity of BNNTs hinders their use in medical and biomedical applications. A strategy to make BNNTs dispersible in aqueous media is the covalent modification. In this study, BNNTs were first hydroxylated with a hydrogen peroxide treatment procedure. This procedure generates OH groups at the defects and edges of the BNNTs, which can be used for further modification. Glutaraldehyde was used as a cross-linker between the OH groups

Fig. 2. (a) TEM images, (b) comparative FT-IR spectra and (c) TGA analysis of BNNTs, h-BNNTs and m-BNNTs.

of hydroxylated BNNTs and carbohydrates such as glucose, lactose and starch. Fig. 1 shows a representative reaction schema between a h-BNNT and a lactose molecule. The TEM images of the BNNTs before (1) and after (2) hydroxylation procedure are given in Fig. 2a. On the TEM images, some damage to BNNTs, shown with a black arrow on the image, during hydroxylation step, which contains continuous stirring in hydrogen peroxide for 48 h, is observed. The covalent modification was eval-

M. Emanet et al. / Colloids and Surfaces B: Biointerfaces 134 (2015) 440–446

443

Fig. 3. Image of Bradford assay performed wells. (a) Only Bradford reagent, (b) control BNNTs, (c) control DMEM, (d) control DMEM/Nutrient Mixture F-12 Ham, (e) control BSA, (f) BNNTs-BSA, (g) BNNTs-DMEM and (h) BNNTs-DMEM/Nutrient Mixture F-12 Ham.

uated with FT-IR. As seen on Fig. 2b, the FT-IR spectra demonstrate a broad peak in the range of 3000–3600 cm−1 , which is attributed to OH group vibrations of the carbohydrates. While this band is absent on the pristine BNNT spectrum, the band becomes clear after the hydroxylation and the carbohydrate attachment. This band is more evident in starch-modified BNNTs, because of the large size of the starch molecule with respect to glucose and lactose. A band attributed to the asymmetric C H stretching vibrations originating from carbohydrates was also observed at around 2930 cm−1 . The amount of carbohydrates bound to the BNNTs was determined with TGA. The BNNTs are, by nature, highly resistant against heat degradation. Therefore, as seen in Fig. 2c, a decrease in the weight of pristine BNNTs was not appreciable with the temperature increase, because the structure of BNNTs does not degrade up to 800 ◦ C. On the contrary, the weight loss in glucose modified BNNTs was about 9%, and about 4% weight for lactose and 5% for starch modified BNNTs. This observation suggests that a higher amount of glucose is bound to the BNNT surfaces with respect to the other investigated carbohydrates. This finding can be explained with a more efficient binding of glucose molecules to the OH groups on the BNNT surfaces, most probably because of its small size. 3.2. Interaction of BNNTs with medium proteins Since their behavior can change upon their addition into the cell culture media, the interaction of BNNTs with the cell culture medium was investigated. As the BNNTs were added into the cell culture media, they tend to form aggregates due to their high hydrophobic nature. They may also interact with peptides and proteins, which may help their dispersion in the media. However, it is not obvious which one of these events, i.e., forming aggregates or non-covalent interaction with peptides and proteins, is taking place upon addition of the BNNTs into the media. Thus, an investigation of the status of the unmodified and modified BNNTs was first undertaken. BNNTs were first treated with BSA, DMEM, or DMEM/Nutrient Mixture Ham’s F-12. BSA was selected because medium, supplemented with FBS, contains a high amount of BSA. In addition, the BNNTs were prepared in medium to observe their interaction with the medium amino acids, peptides and proteins. Bradford reagent was added to the BNNTs-protein/medium mixture after the washing process. Bradford assay is a protein determination method that is based on the observation of the absorbance shifts from 465 to 595 nm when Coomassie Brilliant Blue G-250 dye binds to the proteins under acidic conditions. The negative charge of the dye interacts with the positive amine groups of the proteins by electrostatic interactions, causing a visible color change. In this assay, the two different colors, red and blue, are exploited. When the proteins

Fig. 4. Confocal images of cells treated with BNNTs, h-BNNTs and m-BNNTs.

444

M. Emanet et al. / Colloids and Surfaces B: Biointerfaces 134 (2015) 440–446

Fig. 5. Cell viability assessments of BNNTs, h-BNNTs and m-BNNTs at increasing concentrations.

M. Emanet et al. / Colloids and Surfaces B: Biointerfaces 134 (2015) 440–446

445

cellular uptake of BNNTs. It was found that DAPI non-covalently interacts with BNNTs, h-BNNTs and m-BNNTs, this allowing their tracking under confocal microscopy. As the chemical structure of DAPI presents several amino groups and aromatic rings, it is not surprising to see some DAPI molecules binding the BNNT surfaces. While only DAPI-stained cell nuclei were detectable in control cells, the cells incubated with DAPI-labeled BNNTs clearly showed fluorescence not only in cell nuclei, but also in the cytosols (Fig. 4). The two reasons can be the origin of the fluorescence in the treatedcell nuclei. Since the BNNTs–DAPI complexes are formed through non-covalent interactions, DAPI molecules might have separated upon their up-take into the cells. On the other hand, BNNTs–DAPI complexes might have moved into the cell nuclei directly after being internalized by the cells. A549, lung cancer cells, have higher uptake capacity compared to healthy cells, because of their higher nutrition requirement for proliferation and growth. Thus, they are likely to internalize more material with respect to healthy cells. Additionally, it is known that the cellular uptake of hydrophobic materials shows lower efficiency with respect to hydrophilic materials [27]. Therefore, it is assumed that the interaction between BNNTs and medium proteins in cell culture media, and their covalent modification with hydroxyl groups and carbohydrates enhances its cellular uptake by increasing its hydrophilicity. 3.4. Biocompatibility studies of the BNNTs, h-BNNTs and m-BNNTs

Fig. 6. (a) ROS detection and (b) genotoxicity assessment using A549 cells at increasing concentrations of BNNTs, h-BNNT and m-BNNTs.

are added to the dye, the red color is converted into the blue color due to binding of the dye to the proteins [25,26]. If the proteins or peptides bind to the BNNTs non-specifically, a color change should be observed, since the assay dye will bind to the protein or peptide adsorbed onto BNNTs. Fig. 3 shows the results of Bradford assay. Fig. 3a and b shows plain Bradford reagent and the control BNNTs, respectively. The controls containing only proteins are shown in Fig. 3c–e. The BNNTs–BSA, BNNTs–DMEM and BNNTs–DMEM/nutrient mixture Ham’s F-12 are shown in Fig. 3f–h, respectively. After three washing steps of the BNNTs–protein/media mixture, only amino acids, peptides or proteins adsorbed onto the BNNTs remain in the suspension. When Bradford reagent is added into these suspensions, a color change from blue to red is expected. Interestingly, the color change was not of the same amount in all wells, because of the different interaction strength of BNNTs with proteins and peptides. As it can be seen from the color of the suspensions in the wells, the BNNTs interact better with amino acids, peptides and proteins in the medium compared to the BSA case. Therefore, the blue color is clearer in Fig. 3g and with respect to the suspension in the well in Fig. 3f. In addition, a blue precipitate was formed, indicating that peptides and proteins were retained on the BNNT walls. The obtained results demonstrate that BNNTs interacted with BSA and the medium amino acids, peptides and proteins. This suggests that the formation of a protein corona takes place on the BNNT surfaces upon their addition into the culture medium. 3.3. Cellular uptake of the BNNTs, h-BNNTs and m-BNNTs The confocal microscopy images were acquired to analyze the internalization of BNNTs in the cells. Since BNNTs show no fluorescent properties, a fluorescent dye, DAPI, was selected to track the

The comparative cytotoxicity assessment of BNNTs, h-BNNTs, and m-BNNTs on HDFs and A549 cells are given on Fig. 5. The HDFs and A549 cells were treated with the BNNTs, h-BNNTs and m-BNNTs at increasing concentrations (5–200 ␮g/mL) for long incubation times (1–3 days). The left column of the images shows the BNNT, h-BNNT and m-BNNT-exposed HDFs, while the right column shows BNNT, h-BNNT, and m-BNNT-exposed A549 cells. Each row of the image shows the cells exposed to the same type of the BNNTs. As it can be seen, whereas BNNTs and h-BNNTs non-significantly affect the viability and proliferation of the HDFs, they considerably decrease the viability of A549 cells at higher concentrations (100 and 200 ␮g/mL, analysed with student t-test p < 0.05). The A549 cell viability decreased to 40 and 60%, while the HDF cell viability decreased to 90% at 2nd and 3rd days of the incubation with BNNTs and h-BNNTs. In addition, each type of m-BNNTs non-significantly affected viability of HDFs and A549 cells (80 and 110%, p > 0.05 in all cases). According to the cellular uptake outcomes depicted in Fig. 4, the lower viability of A549 cells exposed to BNNTs and h-BNNTs can be explained with the fact that the two cell lines do not have the same uptake capacity. The A549, lung cancer cells, take up more BNNTs because of their higher nutrition requirement for their fast proliferation. Therefore, the BNNTs show toxic effects on A549 cells because of a final higher content inside the cells. In addition, the decreased viability of the h-BNNT exposed HDF and A549 cells at increased incubation times could be due to an increased uptake of the h-BNNTs because of their higher interactions with the cells since they carry hydroxyl groups. [28]. It is also known that the hydrophobicity of BNNTs induces toxicity and decreases cell viability [27]; the reduced m-BNNT hydrophobicity could thus explain the high cell viability, for both HDF and A549 cells, treated with extremely high concentrations of m-BNNTs. When cells are exposed to the foreign substances, they tend to increase ROS production as a defense mechanism. The increased ROS levels cause cellular stress that stimulates further ROS production. The high ROS levels in the cells causes damages in the structure of many cell components such as proteins, membrane lipids, and DNA. These alterations might result in many impor-

446

M. Emanet et al. / Colloids and Surfaces B: Biointerfaces 134 (2015) 440–446

tant degenerative diseases [29]. The ROS detection assays were performed to analyze whether BNNTs, h-BNNTs and m-BNNTs cause any cellular stress to stimulate death mechanism of A549 cells. The cells were treated with DCFDA, ROS detection reagent, and they were exposed to BNNTs, h-BNNTs and m-BNNTs at increasing concentrations (5–200 ␮g/mL) for 4 h. As seen in the Fig. 6a, the ROS production significantly increased in BNNT and h-BNNT exposed cells up to 60 and 70% (p < 0.05), while the ROS production was not significant (increased of 20 and 30%, p > 0.05) in m-BNNT exposed cells, with respect to the control cultures. As already provided by the cell viability assays, the ROS detection assay results indicate that the cellular uptake of the BNNTs and h-BNNTs at high doses decrease cell viability, and this decrement is correlated to an increased oxidative stress in A549 cells. The comet assay, a single cell gel electrophoresis method used for the genotoxicity assessment, provides detection of DNA damage in single cells. The DNA damage in the cells can be detected from the length of the smear structure of the whole cell DNA that means tail length occurred after single cell gel electrophoresis. The different size of DNA fragments occurring following DNA damages lead to stretch of smear structure. The smear structure of the cells was analyzed with the Comet IV software. In Fig. 6b, BNNT and h-BNNT treated cells tail lengths were approximately 38%, while the m-BNNT exposed cell tail lengths were 20 and 30% if compared to the positive control cells, which were exposed to hydrogen peroxide: these results indicate that the increased ROS levels in the cells are responsible of the DNA damage. 4. Conclusions In this study, BNNTs were hydroxylated and modified with carbohydrates to increase their dispersibility in aqueous media for cellular uptake and biocompatibility studies. Thereafter, BNNTs were first treated with BSA and different types of cell culture media to investigate the degree of their interaction with medium components by using Bradford assay. The findings revealed that BNNTs interacted with the medium proteins and peptides possibly forming a protein/peptide corona layer on the BNNT surfaces similar to other nanomaterials. The cellular uptake of BNNTs, h-BNNTs and m-BNNTs was investigated by using confocal microscopy after DAPI labeling. The confocal microscopy images indicated that unmodified BNNTs, hBNNTs and m-BNNTs were efficiently internalized. In addition, BNNTs, h-BNNTs and m-BNNTs were internalized into A549 cells at a higher rate compared to HDF cells. The influence of BNNTs, h-BNNTs and m-BNNTs on viability, ROS generations and genotoxicity was evaluated using A549 cells and HDF cell lines. The results indicated that the unmodified BNNTs, h-BNNTs and m-BNNTs had no negative effects on viability of HDFs, whereas BNNTs and h-BNNTs were highly cytotoxic on A549 cells at high concentrations (100–200 ␮g/mL). The ROS detection assays showed that the BNNTs and h-BNNTs increased the cellular stress on A549 cells. The BNNTs and h-BNNTs also caused genotoxic effect on this cell line at higher concentrations (100–200 ␮g/mL) confirming the cell viability results. The difference in the toxicity results of the BNNT derivatives on HDFs and A549 cells can be attributed to the higher cellular uptake capacity of the cancer cells as compared to the healthy cells since HDF cell are healthy and A549 cells are cancerous. Moreover, the hydrophobic nature of BNNTs is a cause of their toxic effects on

these cell lines at higher doses indicating that the modification of BNNT by carbohydrates prevents their toxicity even at extreme doses. This study suggests that BNNTs, h-BNNTs, but especially mBNNTs, can be safely used as nanovectors for cell therapy, as drug delivery agents, and for other biomedical applications without significant damages to healthy cells. Due to their cytotoxic effect on cancer cells at higher concentrations (especially BNNTs and h-BNNTs), they could be moreover exploited to develop effective cancer treatment in combination with other therapeutic approaches. Acknowledgements We acknowledge the support from the Scientific and Technological Research Council of Turkey (TUBITAK) (Project No: 112M480) and from Yeditepe University. References [1] N.G. Chopra, R.J. Luyken, K. Cherrey, V.H. Crepsi, M.L. Cohen, S.G. Louie, A. Zettl, Science 269 (1995) 966. [2] S. Iijima, Nature 354 (1991) 56. [3] X. Chen, P. Wu, M. Rousseas, D. Okawa, Z. Gartner, C.R. Zettl, A. Bertozzi, J. Am. Chem. Soc. 131 (2009) 890. [4] S. Kalay, Z. Yilmaz, O. Sen, M. Emanet, M. Kazanc, E. C¸ulha Beilstein, J. Nanotechnol. 6 (2015) 84. [5] D. Lahiri, V. Singh, A.P. Benaduce, S. Seal, A. Kos, L. Agarwala, J. Mech. Behav. Biomed. Mater. 4 (2011) 44. [6] M.S. Islam, A.Z. Kouzani, X.J. Dai, W.P. Michalski, Design and analysis of a cantilever biosensor based on a boron nitride nanotube, in: TENCON 2010: Proceedings of the 2010 IEEE Region 10 Conference, IEEE, Piscataway, N.J., 1951. [7] V. Raffa, A. Ciofani, G. Cuschieri, Nanotechnology 7 (2009) 18. [8] Y. Li, L.H. Chen, Langmuir (26 2010) 5135. [9] L.H. Li, A.M. Chen, Y. Glushenkova, J. Mater. Chem. (20 2010) 9679. [10] G. Ciofani, S. Del Turco, A. Rocca, G. Vito, V. Cappello, M. Yamaguchi, X. Li, B. Mazzolai, G. Basta, M. Gemmi, V. Piazza, V. Golberg, D. Mattoli, Nanomedicine 9 (2014) 773. [11] Y.-T.R. Lau, M. Yamaguchi, X. Li, Y. Bando, D. Golberg, F.M. Winnik, J. Phys. Chem. C 117 (2013) 19568. [12] X. Li, C. Zhi, N. Hanagata, M. Yamaguchi, D. Bando, Y. Golberg, Chem. Commun. 49 (2013) 7337. [13] X. Li, N. Hanagata, X. Wang, M. Yamaguchi, W. Yi, D. Bando, Y. Golberg, Chem. Commun. 50 (2014) 4371. [14] G. Ciofani, V. Raffa, A. Menciassi, A. Cuschieri, Biotechnol. Bioeng. 101 (2008) 850. [15] G. Ciofani, S. Del Turco, G.G. Genchi, D. D’Alessandro, V. Basta, G. Mattoli, Int. J. Pharm. 436 (2012) 444. [16] G. Ciofani, S. Danti, D. D’Alessandro, A. Moscatoc, S. Menciassi, Biochem. Biophys. Res. Commun. 394 (2010) 405. [17] T.H. Ferreira, D.C.F. Soares, L.M.C. Moreira, P.R.O. Silva, E.M.B. Santos, R.G. Sousa, Mater. Sci. Eng. C Mater. Biol. Appl. (33 2013) 4616. [18] T. Ferreira, P.R.O. Silva, E.M.B. Santos, R.G. Sousa, J. Biomater. Nanobiotechnol. 2 (2011) 426. [19] L. Horvath, A. Magrez, D. Golberg, Z. Chunyi, Y. Bando, R. Samajda, E. Horvath, L. Forro, B. Schwaller, ACS Nano 5 (2011) 3800. [20] G. Ciofani, S. Danti, G.G. Genchi, D. D’Alessandro, J.L. Pellequer, M. Odorico, V. Mattoli, M. Giorgi, Int. J. Nanomed. 7 (2012) 19. [21] G. Ciofani, S. Danti, S. Nitti, B. Mazzolai, V. Mattoli, M. Giorgi, Int. J. Pharm. 444 (2013) 85. [22] A. Salwetti, L. Rossi, P. Iacopetti, X. Li, S. Nitti, T. Pellegrino, V. Mattoli, D. Golberg, G. Ciofani, Nanomedicine 2 (2015) 1. [23] S. Kalay, Z. Yilmaz, M. C¸ulha, Beilstein J. Nanotechnol. 4 (2013) 843. [24] M. Gupta, A.K. Gupta, Biomaterials (26 2005) 1565. [25] M.M. Bradford, Anal. Biochem. 7 (1976) 248. [26] C.G. Compto, S.J. n Jones, Anal. Biochem. 151 (1985) 369. [27] L. Li, L.H. Li, S. Ramakrishnan, X.J. Dai, K. Nicholas, Y. Chen, Z. Chen, X. Liu, J. Phys. Chem. C 116 (2012) 18334. [28] X. Zhang, Y. Zhu, J. Li, Z. Zhu, J. Li, Q. Li, W. Huang, J. Nanopart. Res. 13 (2011) 6941. [29] Y. Christen, Am. J. Clin. Nutr. 71 (2000) 621.