Toxicology Letters 181 (2008) 182189

Contents lists available at ScienceDirect

Toxicology Letters
journal homepage: www.elsevier.com/locate/toxlet

Long-term accumulation and low toxicity of single-walled carbon nanotubes in intravenously exposed mice
Sheng-Tao Yang a , Xiang Wang b , Guang Jia b , Yiqun Gu c , Tiancheng Wang d , Haiyu Nie a , Cuicui Ge e , Haifang Wang a, , Yuanfang Liu a
a Beijing National Laboratory for Molecular Sciences, Department of Chemical Biology, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, PR China b Department of Occupational and Environmental Health Sciences, School of Public Health, Peking University, Beijing 100083, PR China c Maternity Hospital of Haidian District, Beijing 100080, PR China d Department of Clinical Laboratory, Third Hospital of Peking University, Beijing 100083, PR China e Laboratory for Bio-Environmental Health Sciences of Nanoscale Materials and Nanosafety and Key Laboratory of Nuclear Analytical Techniques, Institute of High Energy Physics, Chinese Academy of Sciences, Beijing 100049, PR China

a r t i c l e

i n f o

a b s t r a c t
The biomedical application of single-walled carbon nanotubes (SWCNTs), such as drug delivery and cancer treatment, requires a clear understanding of their fate and toxicological prole after intravenous administration. In this study, the long-term accumulation and toxicity of intravenously injected SWCNTs in the main organs (such as liver, lung and spleen) in mice were carefully studied. Although SWCNTs stayed in mice over 3 months, they showed low toxicity to mice. The long-term accumulation of SWCNTs in the main organs was evidenced by using Raman spectroscopy and TEM technique. Statistically signicant changes in organ indices and serum biochemical parameters (LDH, ALT and AST) were observed. The histological observations demonstrate that slight inammation and inammatory cell inltration occurred in lung, but the serum immunological indicators (CH 50 level and TNF- level) remained unchanged. No apoptosis was induced in the main organs. The decreasing glutathione (GSH) level and increasing malondialdehyde (MDA) level suggest that the toxicity of SWCNTs might be due to the oxidative stress. 2008 Elsevier Ireland Ltd. All rights reserved.

Article history: Received 23 June 2008 Received in revised form 28 July 2008 Accepted 28 July 2008 Available online 6 August 2008 Keywords: Carbon nanotubes Intravenous administration Accumulation Toxicity Oxidative stress

1. Introduction With the development of nanotechnology, numerous nanomaterials are created for various applications. Among these nanomaterials, single-walled carbon nanotube (SWCNT) is one of the most concerned, for its unique physicochemical properties and promises in technological applications (Endo et al., 2008). SWCNTs can be used in sensor, electronic device, wastewater treatment and many other industrial applications. Importantly, broad biomedical uses of SWCNTs, such as in drug delivery systems (Lacerda et al., 2006), bone cell growth (Saito et al., 2008) and cancer treatment (Gannon et al., 2007), have been investigated. With the rapid advances in SWCNT-based new materials and technologies, there is a growing recognition that a fundamental understanding of the toxicological properties of SWCNTs is imperative (Warheit, 2006). However, the toxicity of SWCNTs is barely known when they are introduced into the blood circulation, which is especially vital for

Corresponding author. E-mail address: haifangw@pku.edu.cn (H. Wang). 0378-4274/$ see front matter 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.toxlet.2008.07.020

their biomedical applications. Previous studies have reported the cytotoxicity, pulmonary and skin toxicity of SWCNTs (Smart et al., 2006). Pristine SWCNTs have been proven to be cytotoxic, inducing the cell viability loss (Jia et al., 2005), oxidative damage (Pulskamp et al., 2007), inammation (Brown et al., 2007) and apoptosis (Cui et al., 2005). The cytotoxicity depends on the aggregation degree and pretreatment of SWCNTs samples (Wick et al., 2007). It is also suggested that the cytotoxicity of pristine SWCNTs could be reduced via chemical functionalization (Sayes et al., 2006). In the pulmonary and skin toxicity studies, pristine SWCNTs also show considerable toxicity, including animal death, inammation and other clinical signals (Smart et al., 2006). Only a very recent pilot study shows polyethylene glycol (PEG) modied SWCNTs are non-toxic after intravenous (i.v.) injection by using very limited animals (Schipper et al., 2008). In fact, most of i.v. exposure studies just focus on the biodistribution and pharmacokinetics of SWCNTs (Lacerda et al., 2006). We have reported that the surfactant suspended SWCNTs accumulate in liver, lung, and spleen and retain for at least one month without any sign of degradation after mice are i.v. exposed (Yang et al., 2007). Except for lung, the clearance of SWCNTs in body is rather

S.-T. Yang et al. / Toxicology Letters 181 (2008) 182189

183

tron microscopy (TEM). Organ indices, serum biochemical parameters, histological observations, apoptosis, immunological indicators, and oxidative stress indicators were recorded to evaluate the toxicity.
2. Materials and methods 2.1. Synthesis, purication and characterization of SWCNTs SWCNTs were synthesized by arc-discharge method and puried to a carbonaceous purity of 95% following the reported method with minor modications (Wang et al., 2004). Briey, graphite powder was mixed with FeS and Ni2 Y at a mass ratio of 1:5:8.6. The obtained raw SWCNTs were reuxed in 15% H2 O2 for 2 h, then in 6 mol/L HCl for 12 h and nally in 2.6 mol/L HNO3 for another 12 h. The resulting residues were heated under air at 400 C for 1 h and annealed under N2 at 1000 C for 2 h. The puried SWCNTs were characterized by TEM (JEM-200CX, Japan), BrunauerEmmettTeller (BET) technique (ASAP2010, Micromeritics, USA), thermogravimetric analysis (TGA) (SDT 2900, Thermal, USA), Raman spectroscopy (Renishaw micro-Raman instrument, England), inductively coupled plasma-mass spectrometry (ICP-MS) (Thermo Elemental X7, Thermo Electron Co., USA) and infrared spectroscopy (IR) (Magna-IR 750, Nicolet, USA). The puried SWCNTs were suspended in 1.0 wt% Tween 80 aqueous solution by sonication for 30 min for the following animal exposure. 2.2. Animal administration and sampling All animal experiments were performed in compliance with the institutional ethics committee regulations and guidelines on animal welfare (Animal Care and Use Program Guidelines of Peking University). Male CD-ICR mice (25 g) were obtained from Peking University Animal Centre, Beijing, China. They were housed in plastic cages (5 mice/cage) and kept on a 12 h light/dark cycle. Food and water were provided ad libitum. Following acclimation, mice were randomly divided into four groups (5 mice/group).

Fig. 1. A representative TEM image of the puried SWCNTs sample.

low, manifesting that the long-term fate and bio-effect of SWCNTs should be concerned. The experience of pulmonary toxicity studies, in which SWCNTs retain in lungs for 90 days and lead to the formation of granuloma (Lam et al., 2004; Warheit et al., 2004), also suggests that the study of long-term consequence is of prime importance. Herein, we focus on the toxicity of SWCNTs to main organs, including liver, lung and spleen, to provide a general toxicological prole. Liver, lung and spleen are selected here mainly because of the high accumulation of SWCNTs in these organs (Yang et al., 2007). After i.v. administration, the long-term accumulation of SWCNTs was proved by Raman spectroscopy and transmission elec-

Fig. 2. Representative Raman spectra of SWCNTs suspended in Tween 80 aqueous solution (a) and Raman spectra of tissue homogenates of liver (b), lung (c) and spleen (d) of both SWCNTs exposed mice at 90 days post-exposure and control mice.

184

S.-T. Yang et al. / Toxicology Letters 181 (2008) 182189

Fig. 3. TEM images of SWCNTs (some indicated by the black arrows) in the residue from the digested liver (a) and lung (b) of mice at 90 days after i.v. exposure to SWCNTs.

Mice of each group were single i.v. (tail vein) injected with SWCNTs in 1.0 wt% Tween 80 aqueous solution at a dose of 40 g/mouse, 200 g/mouse and 1.0 mg/mouse, respectively. For 1.0 mg/mouse group, each mouse was exposed to SWCNTs by three times of i.v. injection of 1.0 ml solution once every 4 h (330 g SWCNTs/0.33 mL). Mice i.v. exposed to 1.0 wt% Tween 80 aqueous solution were taken as the control group. Body weight and behaviors were recorded once per 3 days post-exposure. At 90 days post-exposure, mice were sacriced and blood/organ samples were collected for accumulation determination and toxicological assays. Serum samples were obtained from blood by centrifugation (3000 rpm for 10 min). Liver, lung and spleen were collected and weighted for organ indices (organ weight/body weight) calculation. A piece of each organ sample was cut off and xed in 4% formaldehyde solution. All the rest were stored at 80 C before assays. 2.3. Raman spectra of SWCNTs in organ homogenates To determine the accumulation of SWCNTs in mice, Raman spectra of homogenates of liver, lung and spleen were recorded. Organs of 1.0 mg/mouse group and the control group were homogenized in water (0.1 g tissue in 1.0 mL water) with homogenizers. The homogenate was dropped on glass slip and dried under infrared lamp. Raman spectra of these samples were recorded upon a Renishaw micro-Raman instrument (laser excitation wavelength 785 nm, 50 mW power, 50 objective, laser spot size 2 m 2 m, 10 s collection time). The Raman spectra of SWCNTs suspended in Tween 80 were also recorded. 2.4. TEM investigation of SWCNTs in digested organs For TEM analyses, liver and lung samples of 1.0 mg/mouse group were digested with 10 mL mixture of 65% HClO4 and 30% H2 O2 (1:1 in volume) at 90 C for 30 min. After centrifugation (12,000 rpm for 15 min), the residue was collected and washed with deionized water twice and alcohol once. The nal residue was dispersed in alcohol for TEM specimen preparation. 2.5. Serum biochemical parameters

standard histopathological techniques. The mounted sections were stained with hematoxylineosin (H&E) and examined by light microscopy. 2.7. TNF- assay To determine the plasma tumor necrosis factor (TNF- ) levels, enzyme linked immunosorbent assay (ELISA) was performed by using commercial kits that are selective for mouse TNF- (BD Biosciences, USA). Manufacturers instruction was followed. The absorbance was measured on a microplate reader at 450 nm (Varioskan Flash, Thermo Electron, Finland) and the TNF- concentration of samples was calculated from a standard curve. 2.8. Cell apoptosis assay Cell apoptosis was evaluated by using terminal deoxynucleotidyl transferasemediated dUTP nick-end-labeling (TUNEL) staining on the organ sections. All the reagents used were purchased from Dingguo Biotechnology Co., Beijing, China and the instruction was followed. Briey, the slides were predated to get rid of parafn. Then, the slides were digested in 20 g/mL of proteinase K at 37 C for 10 min and rinsed in PBS. After that the slides were incubated in 0.3% H2 O2 methanol solution and rinsed twice for 2 min in PBS. After adding 0.1% Triton X-100, the slides were incubated at 20 C for 10 min. The slides were washed twice with PBS and dried at room temperature. Then, 10 L terminal deoxynucleotidyl transferase (TDT) solution was placed on the slides, followed by incubation at 37 C for 2 h. The reaction was stopped by washing slides twice with buffer A (pH 9.0, containing 0.85% NaCl and 1.211% Tris). After blocking with 3% BSA solution for 30 min, the slides were incubated for 60 min with streptavidinalkaline phosphatase conjugate. The slides were washed twice with buffer B (pH 9.0, containing 0.85% NaCl, 1.0165% MgCl2 and 1.211% Tris). And the slides were further developed with 40 L nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (BCIP/NBT) solution for 20 min. The resulting slides were washed twice with buffer B and stained with nuclear fast red solution. Finally the slides were dehydrated and observed under the light microscopy. 2.9. Oxidative stress assay

All biochemical assays were performed using a Hitachi 7170A clinical automatic chemistry analyzer. Lactate dehydrogenase (LDH), total bilirubin (TBIL), alanine aminotransferase (ALT), total bile acid (TBA), alkaline phosphatase (ALP), aspartate aminotransferase (AST) and total hemolytic complement levels (CH50) were measured using the commercial kits (Bhlmann Laboratories, Switzerland). 2.6. Histological observations For histological observations, the formalin-xed tissue samples were embedded in parafn, thin-sectioned, and then mounted on glass microscope slides using the

For the assays of reduced GSH level and MDA level, each organ sample was minced and homogenized in 4 C water for three times (10 s/time, intermittent for 30 s) to yield 10% (w/v) homogenate. The homogenates were centrifuged at 2000 rpm for 10 min to obtain the supernatants. Protein concentrations in the supernatants were determined according to the method of Bradford, using bovine serum albumin as the standard (Bradford, 1976). The reduced GSH level of the supernatant was examined by using spectrophotometric diagnostic kits (Nanjing Jiancheng Biotechnology Institute, China) based on the method of Jollow et al. (1974). Results of GSH were expressed as mg GSH/g protein. The lipid peroxidation indicator malondialdehyde (MDA) was determined using the

Table 1 Organ indices of the control mice and the SWCNTs exposed mice at 90 days post-i.v. exposure. Data represent mean S.D. (n = 5) Control Liver Lung Spleen
*

40 g/mouse 0.051 0.004 0.0052 0.0007 0.0047 0.0016*

200 g/mouse 0.051 0.005 0.0056 0.0007 0.0043 0.0008*

1.0 mg/mouse 0.049 0.005 0.0064 0.0006* 0.0033 0.0005

0.048 0.005 0.0052 0.0004 0.0030 0.0006

p < 0.05 compared with control group.

S.-T. Yang et al. / Toxicology Letters 181 (2008) 182189

185

3. Results 3.1. Characterization of the puried SWCNTs The puried SWCNTs were characterized by TEM, BET, TGA, ICPMS and IR. A representative TEM image is shown in Fig. 1. The SWCNTs are generally pure, with limited carbon particles and catalyzer. The SWCNTs form bundles with 1030 nm in diameter and lengths of 23 m in length in the suspension. The specic surface area (SSA) of SWCNTs is 267.8 m2 /g according to the BET analysis. TGA analysis suggests that the carbonaceous purity of SWCNTs is higher than 95.4%. ICP-MS further conrms the high purity. The metal impurities are 0.4 wt% Fe, 3.0 wt% Ni and 1.3 wt% Y. The remaining metals can be mainly attributed to the encapsulation by carbon, which prevents the solubilization of the metals during the purication process and also makes them non-bioavailable (Liu et al., 2008a). An IR was recorded to show the surface properties of SWCNTs. Only a tiny peak is observed at around 1744 cm1 , suggesting few carbonyls or carboxyls are on the surface of sidewall. There is no signal of carbonhydrogen bond observed. The nearly intact surface of SWCNTs should be due to the nal anneal process. The puried SWCNTs were nally suspended in 1.0 wt% Tween 80 aqueous solution for animal exposure. 3.2. Raman spectra of SWCNTs in organ homogenates To determine the long-term accumulation of SWCNTs in main organs (liver, lung and spleen), Raman spectra of the SWCNTs Tween 80 suspension and the homogenated tissue samples were recorded. Fig. 2a shows the typical G-band of SWCNTs at around

Fig. 4. Serum biochemical parameters of the control mice and the SWCNTs exposed mice at 90 days post-i.v. exposure. Data represent mean S.D. (n = 5). *p < 0.05 compared with control group.

method of thiobarbituric acid reactive species (Nanjing Jiancheng Biotechnology Institute, China). The level of MDA was expressed as nmol MDA/mg protein, using 1,1,3,3-tetraethoxypropane (TEP) as the standard (Dahle et al., 1962). The measurements of GSH and MDA were performed following the manufacturers instruction. 2.10. Statistical analysis All data are presented as the mean of ve individual observations with the standard deviation. Signicance has been calculated using Students t-test. Difference was considered signicant if p < 0.05.

Fig. 5. Histopathological observation of the control mice and the SWCNTs exposed mice at 90 days post-i.v. exposure (400). White arrows point to SWCNT aggregates.

186

S.-T. Yang et al. / Toxicology Letters 181 (2008) 182189

The body weight increase is similar among the four groups, except 1.0 mg/mouse group, in which the body weight increase is a little lower than that of the control group at initial 3 days (data not shown). The organ indices were recorded to provide a general impression of the toxicity. As shown in Table 1, no signicant difference is observed in the liver indices. The lung index of 1.0 mg/mouse group is signicantly higher than that of the control group, indicating the potential pulmonary injury. The spleen indices of 40 g/mouse group and 200 g/mouse group increase, too. This suggests that SWCNTs may induce the immunological response. 3.5. Serum biochemical parameters The serum biochemical parameters were assayed to further evaluate the toxicity of SWCNTs. Fig. 4 shows that TBIL, TBA and ALP levels of all groups are similar after mice were exposed to SWCNTs. But ALT and AST levels of 200 g/mouse group and 1.0 mg/mouse group are higher than those of the control group. The increase of ALT and AST, which are important indicators of the hepatic injury, demonstrates that SWCNTs induced hepatic injury still remains even at 90 days post-exposure. Furthermore, the increase of ALT and AST is dose-dependent, inferring that the induced hepatic injury is dose-dependent. The LDH levels of all exposed group are nearly two folds of those of the control group. LDH is a cytoplasmic enzyme and is often used as an indicator of alveolar macrophage (AM) injury in the pulmonary toxicity study. LDH is also a general indicator of hepatic injury. The high level of LDH activity manifests that serious pulmonary or hepatic injury is aroused by SWCNTs. 3.6. Histopathological observations The histopathological observations were carried out to determine the organic damage induced by SWCNTs (Fig. 5). The black or brown aggregates observed in liver, lung and spleen sections are SWCNTs, which is consistent with the Raman spectra measurements. No obvious hepatic damage is observed in the histopathological study among all the exposed groups, though serum biochemistry indicates a considerable hepatic toxicity. There is no histopathological change in spleen either. The SWCNTs are generally trapped in capillary and aggregate to different sizes in lung. Some inammatory cells are observed surrounding the trapped SWCNTs in lung. When the dose increases to 1.0 mg/mouse, the inammation becomes more serious and inammatory cell inltration is observed. This is much different with that of intratracheally instilled SWCNTs, which induce granulomas within 90 days post-exposure (Lam et al., 2004; Warheit et al., 2004). 3.7. Serum immunological indicators The inammation happened in lung according to the histopathological observations. To evaluate the immune response, CH 50 level and TNF- level in serum were measured. The CH 50 level is a typical indicator of complement activation. Complement activation could regulate the phagocytosis ability and the immunological activity. As shown in Fig. 6a, the CH 50 level is quite similar among all groups. Although SWCNTs were reported to activate the complement system via the classical pathway (Salvador-Morales et al., 2006), in this study the complement level may increase and then decrease to the normal level within 90 days exposure. Besides being an important indicator of immunological activity, TNF- level is also an indicator of inammation and brosis. Likewise, SWCNTs do not induce the change of TNF- level (Fig. 6b), which is coincident with that no brosis is observed in the exposed mice. The slight inammation is unable to induce an increase of TNF- level.

Fig. 6. Serum immunological indicator levels of the control mice and the SWCNTs exposed mice at 90 days post-i.v. exposure. (a) CH50 levels. (b) TNF- levels. Data represent mean S.D. (n = 5).

1590 cm1 , which is the characteristic of graphite carbon. No Gband was observed in all control tissue samples (Fig. 2bd). In the exposed group, although rather weak, G-band in liver and spleen is observed, suggesting there are still a few of SWCNTs at 90 days post-exposure. But in lung, the Raman intensity is still pretty high, which refers to the high concentration of SWCNTs in lung. Although a remarkable clearance in lung was observed previously during the rst 28 days post-exposure, the Raman spectra demonstrate that a remarkable amount of SWCNTs are still accumulated in lung even at 90 days post-exposure (a lot of black spots formed in lung in the light microscopic images). The long-term accumulation in lung is consistent with the results of pulmonary intratracheal instillation of SWCNTs, in which SWCNTs also retain in lung over 90 days (Lam et al., 2004; Warheit et al., 2004). 3.3. TEM images of SWCNTs in digested organs The high accumulation level in liver and lung allow the direct electron microscopic imaging of residue from the digested tissues. Fig. 3 shows the characteristic structures of SWCNTs could be clearly identied in the digested liver and lung samples. The intact nanotube structures observed in TEM images indicate that SWCNTs are stable against the biotransformation during the 90 days exposure as well as the subsequent chemical digestion process. 3.4. Clinical symptoms and organ indices All mice showed no symptoms of abnormality, such as lethargy, anorexia, vomiting, or diarrhea, during the 90 days post-exposure.

S.-T. Yang et al. / Toxicology Letters 181 (2008) 182189

187

Fig. 7. Apoptosis analysis of the control mice and the SWCNTs exposed mice at 90 days post-i.v. exposure by TUNEL method (400). White arrows point to the accumulated SWCNTs in the main organs.

3.8. Cell apoptosis Apoptosis is regarded as a general adaptive response when cells are exposed to SWCNTs in vitro. Here, the apoptosis of the cells in organ sections was assayed by TUNEL method. As shown in Fig. 7, there are few cells labeled positive (indigo nucleolus) by TUNEL assay at 90 days post-exposure, indicating little apoptosis occurred. But the apoptosis levels are similar among all exposed and control groups. Similar to the histopathological observations, black or brown aggregates of SWCNTs are observed in the exposed groups. Interestingly, the dye molecules are adsorbed onto SWCNTs slightly during the staining process, resulting in the blue-black SWCNTs. This adsorption should be due to the stacking interaction between SWCNTs and the tetrazolium structure of the dye molecules. 3.9. Oxidative stress The oxidative stress aroused by SWCNTs in main organs was measured to reveal the possible toxicological pathway. As shown in Fig. 8, the reduced GSH level in liver and lung of all exposed groups decreases statistically, and 1.0 mg/mouse group shows the most serious oxidative damage. Contrast to the GSH results, MDA level of liver and lung in 1.0 mg/mouse group shows a statistical increase, indicating that SWCNTs induce oxidative damage. The reduced GSH level and MDA level in spleen remain unchanged in all groups. Namely, there is no observed oxidative damage to spleen.

4. Discussion In this study, we nd that pristine SWCNTs retain in mice for 3 months after i.v. exposure and induce low toxicity. The long-term accumulation seems to be a bio-property of most SWCNTs. Except for DTPA-SWCNTs and hydroxylated SWCNTs (Singh et al., 2006; Wang et al., 2004), other SWCNTs retain in body for more than one month (Yang et al., 2007, 2008; Liu et al., 2008b; Schipper et al., 2008). All the available data imply that the accumulation of SWCNTs depends on the functional groups and the functionalization degree. Lacerda et al. also reported that functionalized multi-walled carbon nanotubes (MWCNTs) possessed a much lower accumulation than that of pristine MWCNTs and that was attributed to the high individualization of functionalized MWCNTs (Lacerda et al., 2008b). The long-term accumulation of SWCNTs suggests that the study of the corresponding long-term toxicity is valuable and vital. The accumulation also demands the proper chemical functionalization of SWCNTs to achieve a lower accumulation in reticuloendothelial system (RES) for the sake of the drug delivery and related biomedical applications. Compared with PEGylated SWCNTs, pristine SWCNTs show higher toxicity in the serum biochemical parameters assessments. The ALT and AST levels of mice exposed to pristine SWCNTs showed a dose-dependent increase, while those of mice exposed to PEGylated SWCNTs remained unchanged (Schipper et al., 2008). Since the toxicity of SWCNTs is dependent on their functionalization degree (Sayes et al., 2006), such hepatic toxicity might be reduced when proper chemical functionalization is adopted to obtain a high

188

S.-T. Yang et al. / Toxicology Letters 181 (2008) 182189

studies, oxidative stress is regarded as the cause of the cytotoxicity (Pulskamp et al., 2007). In fact, oxidative stress is taken as an important pathway of toxicity of SWCNTs and other nanomaterials (Stern and McNeil, 2008). It is very hard to eliminate the metal residues from SWCNT samples, even though most of the metals can be removed by reuxing SWCNTs sample in H2 O2 and acid. Traditional purication is helpful to remove the majority of metals loaded on the outside and internal surface of tubes, but this process is in vain for metals packed inside the surrounding carbon fragments. A few studies reported that metal contents in SWCNTs were partly responsible for the serious oxidative stress (Guo et al., 2007; Liu et al., 2007, 2008a; Pulskamp et al., 2007). But a very recent study by Liu et al. focused on the bioavailability of metals in SWCNTs suggest that the encapsulated metals are non-bioavailable for at least two months (Liu et al., 2008a). Our purication procedure is similar (even stricter) to Liu et al.s study (Liu et al., 2008a). The metal impurities left in our SWCNT samples should be inside SWCNTs or encapsulated in carbon fragments, therefore they are hardly attributed to the toxicity of SWCNTs. In other words, the oxidative stress comes from SWCNTs per se. In summary, the long-term accumulation and low toxicity in mice i.v. exposed to SWCNTs are reported. Only serum biochemical changes and pulmonary inammation were observed. No obvious cell apoptosis or changes of immunological indicators were induced. The proposed main toxicological mechanism is oxidative stress aroused in liver and lungs. The low toxicity of pristine SWCNTs implicates that SWCNTs could be used as the safe materials for biomedical applications. It also suggests that further chemical functionalization should be taken to improve the dispersion and excretion of SWCNTs. Conict of interest
Fig. 8. The oxidative stress of the control mice and the SWCNT exposed mice at 90 days post-i.v. exposure. (a) GSH level of main organs. (b) MDA level of main organs. *p < 0.05 compared with control group. Data represent mean S.D. (n = 5).

The authors declare that there are no conicts of interest. Acknowledgments

functionalization degree of SWCNTs with a higher in vivo stability. In our previous study of taurine functionalized MWCNTs, we did not nd the increase of the ALT and AST levels and the hepatic toxicity (Deng et al., 2007). Lacerda et al. also found that serum suspended MWCNTs showed higher acute toxicity than ammonium functionalized MWCNTs (Lacerda et al., 2008a). To clarify the role of functionalization in the in vivo toxicity of SWCNTs, more efforts are required and parallel comparison is preferred. The low toxicity of intravenously exposed SWCNTs might be due to the active protecting mechanism (restriction of the SWCNTs by certain cells). RES uptake of nanoparticles post-i.v. injection is the main adaptive mechanism here. The opsonin regulated RES capture avoids the interaction of SWCNTs with parenchyma cells and restricts the toxicity to certain cells, such as Kupffer cells (KC) and AM (Owens and Peppas, 2006). SWCNTs could be secreted by AM and leave the lung via mucus through mucociliary transport (Yang et al., 2007). SWCNTs trapped in KC could be excreted via bile, though the process is very slow (Liu et al., 2008b). As for intratracheally instillated SWCNTs, they interacted with all cells they encountered, so a more serious injury was induced (Lam et al., 2004; Warheit et al., 2004). When SWCNTs were injected to peritoneal cavity of mice, inammation and the formation of granuloma occurred (Poland et al., 2008). According to our results, the dominant toxicological mechanism of the intravenously exposed SWCNTs would be the oxidative stress. Oxidative stress is a broadly existent phenomenon when cells are exposed to SWCNTs (Lewinski et al., 2008). Although the cytotoxicity of SWCNTs is not always observed in the cell culture

We acknowledge nancial support from the China Natural Science Foundation (Signicant Project No. 10490180) and the China Ministry of Science and Technology (973 Project No. 2006CB705604). References
Bradford, M.M., 1976. Rapid and sensitive method for quantitation of microgram quantities of protein utilizing principle of proteindye binding. Anal. Biochem. 72, 248254. Brown, D.M., Kinloch, I.A., Bangert, U., Windle, A.H., Walter, D.M., Walker, G.S., Scotchford, C.A., Donaldson, K., Stone, V., 2007. An in vitro study of the potential of carbon nanotubes and nanobres to induce inammatory mediators and frustrated phagocytosis. Carbon 45, 17431756. Cui, D.X., Tian, F.R., Ozkan, C.S., Wang, M., Gao, H.J., 2005. Effect of single wall carbon nanotubes on human HEK293 cells. Toxicol. Lett. 155, 7385. Dahle, L.K., Hill, E.G., Holman, R.T., 1962. Thiobarbituric acid reaction and autoxidations of polyunsaturated fatty acid methyl esters. Arch. Biochem. Biophys. 98, 253261. Deng, X., Jia, G., Wang, H., Sun, H., Wang, X., Yang, S., Wang, T., Liu, Y., 2007. Translocation and fate of multi-walled carbon nanotubes in vivo. Carbon 45, 14191424. Endo, M., Strano, M.S., Ajayan, P.M., 2008. Potential applications of carbon nanotubes. Carbon Nanotubes 111, 1361. Gannon, C.J., Cherukuri, P., Yakobson, B.I., Cognet, L., Kanzius, J.S., Kittrell, C., Weisman, R.B., Pasquali, M., Schmidt, H.K., Smalley, R.E., Curley, S.A., 2007. Carbon nanotube-enhanced thermal destruction of cancer cells in a noninvasive radiofrequency eld. Cancer 110, 26542665. Guo, L., Morris, D.G., Liu, X.Y., Vaslet, C., Hurt, R.H., Kane, A.B., 2007. Iron bioavailability and redox activity in diverse carbon nanotube samples. Chem. Mater. 19, 34723478. Jia, G., Wang, H.F., Yan, L., Wang, X., Pei, R.J., Yan, T., Zhao, Y.L., Guo, X.B., 2005. Cytotoxicity of carbon nanomaterials: single-wall nanotube, multi-wall nanotube, and fullerene. Environ. Sci. Technol. 39, 13781383.

S.-T. Yang et al. / Toxicology Letters 181 (2008) 182189 Jollow, D.J., Mitchell, J.R., Zampaglione, N., Gillette, J.R., 1974. Bromobenzeneinduced liver necrosisprotective role of glutathione and evidence for 3,4-bromobenzene oxide as hepatotoxic metabolite. Pharmacology 11, 151169. Lacerda, L., Ali-Boucetta, H., Herrero, M., Pastorin, G., Bianco, A., Prato, M., Kostarelos, K., 2008a. Tissue histology and physiology following intravenous administration of different types functionalizaed multiwalled carbon nanotubes. Nanomedicine 3, 149161. Lacerda, L., Bianco, A., Prato, M., Kostarelos, K., 2006. Carbon nanotubes as nanomedicines: from toxicology to pharmacology. Adv. Drug Deliv. Rev. 58, 14601470. Lacerda, L., Soundararajan, A., Singh, R., Pastorin, G., Al-Jamal, K.T., Turton, J., Frederik, P., Herrero, M.A., Li, S., Bao, A., Emetzoglou, D., Mather, S., Phillips, W.T., Prato, M., Bianco, A., Goins, B., Kostarelos, K., 2008b. Dynamic imaging of functionalized multi-walled carbon nanotube systemic circulation and urinary excretion. Adv. Mater. 20, 225230. Lam, C.W., James, J.T., McCluskey, R., Hunter, R.L., 2004. Pulmonary toxicity of singlewall carbon nanotubes in mice 7 and 90 days after intratracheal instillation. Toxicol. Sci. 77, 126134. Lewinski, N., Colvin, V., Drezek, R., 2008. Cytotoxicity of nanoparticles. Small 4, 2649. Liu, X., Guo, L., Morris, D., Kane, A.B., Hurt, R.H., 2008a. Targeted removal of bioavailable metal as a detoxication strategy for carbon nanotubes. Carbon 46, 489500. Liu, X.Y., Gurel, V., Morris, D., Murray, D.W., Zhitkovich, A., Kane, A.B., Hurt, R.H., 2007. Bioavailability of nickel in single-wall carbon nanotubes. Adv. Mater. 19, 27902796. Liu, Z., Davis, C., Cai, W.B., He, L., Chen, X.Y., Dai, H.J., 2008b. Circulation and longterm fate of functionalized, biocompatible single-walled carbon nanotubes in mice probed by Raman spectroscopy. Proc. Natl. Acad. Sci. U.S.A. 105, 14101415. Owens, D.E., Peppas, N.A., 2006. Opsonization, biodistribution, and pharmacokinetics of polymeric nanoparticles. Int. J. Pharm. 307, 93102. Poland, C.A., Dufn, R., Kinloch, I., Maynard, A., Wallace, W.A.H., Seaton, A., Stone, V., Brown, S., Macnee, W., Donaldson, K., 2008. Carbon nanotubes introduced into the abdominal cavity of mice show asbestoslike pathogenicity in a pilot study. Nat. Nanotechnol. 3, 423428. Pulskamp, K., Diabate, S., Krug, H.F., 2007. Carbon nanotubes show no sign of acute toxicity but induce intracellular reactive oxygen species in dependence on contaminants. Toxicol. Lett. 168, 5874.

189

Saito, N., Usui, Y., Aoki, K., Narita, N., Shimizu, M., Ogiwara, N., Nakamura, K., Ishigaki, N., Kato, H., Taruta, S., Endo, M., 2008. Carbon nanotubes for biomaterials in contact with bone. Curr. Med. Chem. 15, 523527. Salvador-Morales, C., Flahaut, E., Sim, E., Sloan, J., Green, M.L.H., Sim, R.B., 2006. Complement activation and protein adsorption by carbon nanotubes. Mol. Immunol. 43, 193201. Sayes, C.M., Liang, F., Hudson, J.L., Mendez, J., Guo, W.H., Beach, J.M., Moore, V.C., Doyle, C.D., West, J.L., Billups, W.E., Ausman, K.D., Colvin, V.L., 2006. Functionalization density dependence of single-walled carbon nanotubes cytotoxicity in vitro. Toxicol. Lett. 161, 135142. Schipper, M.L., Nakayama-Ratchford, N., Davis, C.R., Kam, N.W.S., Chu, P., Liu, Z., Sun, X.M., Dai, H.J., Gambhir, S.S., 2008. A pilot toxicology study of single-walled carbon nanotubes in a small sample of mice. Nat. Nanotechnol. 3, 216221. Singh, R., Pantarotto, D., Lacerda, L., Pastorin, G., Klumpp, C., Prato, M., Bianco, A., Kostarelos, K., 2006. Tissue biodistribution and blood clearance rates of intravenously administered carbon nanotube radiotracers. Proc. Natl. Acad. Sci. U.S.A. 103, 33573362. Smart, S.K., Cassady, A.I., Lu, G.Q., Martin, D.J., 2006. The biocompatibility of carbon nanotubes. Carbon 44, 10341047. Stern, S.T., McNeil, S.E., 2008. Nanotechnology safety concerns revisited. Toxicol. Sci. 101, 421. Wang, H.F., Wang, J., Deng, X.Y., Sun, H.F., Shi, Z.J., Gu, Z.N., Liu, Y.F., Zhao, Y.L., 2004. Biodistribution of carbon single-wall carbon nanotubes in mice. J. Nanosci. Nanotechnol. 4, 10191024. Warheit, D.B., 2006. What is currently known about the health risks related to carbon nanotube exposures? Carbon 44, 10641069. Warheit, D.B., Laurence, B.R., Reed, K.L., Roach, D.H., Reynolds, G.A.M., Webb, T.R., 2004. Comparative pulmonary toxicity assessment of single-wall carbon nanotubes in rats. Toxicol. Sci. 77, 117125. Wick, P., Manser, P., Limbach, L.K., Dettlaff-Weglikowska, U., Krumeich, F., Roth, S., Stark, W.J., Bruinink, A., 2007. The degree and kind of agglomeration affect carbon nanotube cytotoxicity. Toxicol. Lett. 168, 121131. Yang, S.T., Fernando, K.A.S., Liu, J.H., Wang, J., Sun, H.F., Liu, Y.F., Chen, M., Huang, Y.P., Wang, X., Wang, H.F., Sun, Y.P., 2008. Covalently PEGylated carbon nanotubes with stealth character in vivo. Small 4, 940944. Yang, S.T., Guo, W., Lin, Y., Deng, X.Y., Wang, H.F., Sun, H.F., Liu, Y.F., Wang, X., Wang, W., Chen, M., Huang, Y.P., Sun, Y.P., 2007. Biodistribution of pristine single-walled carbon nanotubes in vivo. J. Phys. Chem. C 111, 1776117764.