This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy Environment International 40 (2012) 244–256 Contents lists available at SciVerse ScienceDirect Environment International journal homepage: www.elsevier.com/locate/envint Review Recent progress and perspectives on the toxicity of carbon nanotubes at organism, organ, cell, and biomacromolecule levels Xingchen Zhao, Rutao Liu ⁎ Shandong Key Laboratory of Water Pollution Control and Resource Reuse, School of Environmental Science and Engineering, Shandong University, China–America CRC for Environment & Health, Shandong Province, 27# Shanda South Road, Jinan 250100, PR China a r t i c l e i n f o Article history: Received 2 July 2011 Accepted 19 December 2011 Available online 14 January 2012 Keywords: Carbon nanotubes Toxicity In vivo In vitro Characterization Mechanism a b s t r a c t A wide application of carbon nanotubes (CNTs) is on the way owing to their unique structural, optical, mechanical and electronic properties, high specific surface area, and facile functionalization. As a result, human beings will inevitably be exposed to CNTs, especially when the tubes are utilized as diagnostic and therapeutic tools to better understand, detect, and treat human diseases. Therefore the new subject of nanotoxicology, which is the study of the toxicity of nanomaterials, is now gaining public concern. This review provides an overview and comments on recent advances (mostly within the last 3 years) in the toxicology of CNTs, including their toxicity targeted to cells, organs, tissues and the whole organism, including mammals and other species (e.g. aquatic species, plants, and bacteria). Not only these traditional subjects of toxicological study but the interaction of CNTs and biomacromolecules is also covered so that the mechanism of their toxicity may be understood and their undesirable properties are more likely to be avoided. © 2011 Elsevier Ltd. All rights reserved. Contents 1. 2. Introduction . . . . . . . . . . . . . . . . . . . Surface modification of CNTs used in toxicology studies 2.1. Covalent modification . . . . . . . . . . . 2.2. Noncovalent modification . . . . . . . . . 3. In vivo toxicity of CNTs . . . . . . . . . . . . . . 4. In vitro toxicity of CNTs . . . . . . . . . . . . . . 5. Toxicity studies on species other than mammals . . 6. Biomacromolecules and CNT interactions . . . . . . 7. Outlook and suggestions for future research . . . . 8. Conclusions . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introduction Nanomaterial applications have expanded not only into materials science but also into biomedical, chemical and electronics fields because of their high specific surface area, conductivity, magnetic susceptibility and catalytic activity. More than five hundred consumer products currently on the market claim to contain elements of nanoscience and nanotechnology with new entries coming daily ⁎ Corresponding author at: School of Environmental Science and Engineering, Shandong University, Jinan 250100, PR China. Tel./fax: + 86 531 88364868. E-mail address: rutaoliu@sdu.edu.cn (R. Liu). 0160-4120/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.envint.2011.12.003 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 245 245 247 247 249 249 251 252 253 253 253 (Jones and Grainger, 2009). The market requires metric tons of raw nanomaterials annually, including nanometal particles, metal oxide particles and CNTs. Demands for nanoproducts especially in medicine and the pharmaceutical industry is expected to grow by over 17% each year (Jones and Grainger, 2009). As a very important member in the nanomaterial family, CNTs exhibit unique chemical, physical and electronic properties, which make them potentially useful in many applications in nanotechnology, electronics, optics, and other fields of material science, as well as potential uses in architecture (Gomez et al., 2009; Sheng et al., 2009; Tsai et al., 2009). They may also have applications in the construction of body armor (Mylvaganam, 2008). CNTs have excellent characteristics as Author's personal copy X. Zhao, R. Liu / Environment International 40 (2012) 244–256 245 Fig. 1. Carbon nanotube release to human beings and ecosystems. the probes for scanning probe microscopes and their application to cell manipulation technology seems very promising (Buchoux et al., 2009; Narui et al., 2009; Stevens, 2009; Suga et al., 2009; Zhang and Li, 2009). Expectations are also centering on such areas as the development of drug delivery systems that use CNTs' interior spaces (Im et al., 2010; Liu et al., 2009a, b; Oh et al., 2010; Tran et al., 2009). Since CNTs are so attractive in both basic science and applied technology, uncontrolled inadvertent exposure of CNTs to human beings and ecosystems will unavoidably increase through direct or indirect routes (Brumfiel, 2003; Kostarelos et al., 2009; Lee et al., 2010; Service, 2003) (Fig. 1). Therefore, the focus of this review will be to briefly summarize the important and new results of nanotoxicity studies conducted to date, with an emphasis on contamination methods for living organisms and cells that are suspected to contribute to their toxic potential, as well as debates in the literature that may be caused by inconsistency of raw materials or experimental conditions. A description of the current in vitro and in vivo toxicity analysis and results will be discussed within the context of the ongoing challenges and recommendations that will be made for achieving more reliable measures of CNT safety. Besides, we hope this review can give rise to more attention to the advances of fundamental researches concerning basic properties and criteria of CNTs, from synthesis to comprehensive characterization. 2. Surface modification of CNTs used in toxicology studies CNTs are artificial carbon-structured substances that were first discovered by Iijima (1991). Ideal CNTs are formed by graphene layers rolled into cylindrical tubes and the number of graphene layers can range from one to more than one hundred. CNTs with one graphene layer are called single-walled carbon nanotubes (SWCNTs), while tubes containing more than one layer are multi-walled carbon nanotubes (MWCNTs). Since their discovery, CNTs have been the focus of extensive research worldwide because of their unique structure and fascinating physical and chemical properties including low density, high ductility, high mechanical strength, and excellent conductivity. CNTs are macromolecules consisting of tens of thousands of carbon atoms which are located in a delocalized aromatic system. They are nearly insoluble in any solvent because they form bundles. Their insolubility not only severely hinders research on their chemical properties but significantly restrict the application in every field. So scientists have explored many methods to increase the solubility of CNTs for both practical use and toxicological research. Approaches of increasing solubility can be roughly divided into two categories: one is to covalently modify functional groups on the surface by chemically decorating their side-walls and tips; the other is to physically adsorb small molecules or polymers onto the surface of CNTs through hydrophobic interactions, π–π interactions or supermolecular inclusions (Hersam, 2008) (Fig. 2). 2.1. Covalent modification Fig. 2. (A) Covalent sidewall chemistry. (B) Covalent chemistry at defects or open ends. (C) Non-covalent surfactant encapsulation. (D) Non-covalent polymer wrapping. Reproduced with permission. Copyright © 2008 Macmillan Publishers Limited. Covalent modification is to attach functional groups onto CNTs by forming covalent bonds, and is categorized as end group reaction and side wall reaction. The end caps of CNTs are typically hemispheres formed by hexatomic rings or pentatomic rings which are therefore highly reactive compared to the sidewalls. But the sidewalls may contain certain defects, such as sp3 hybridized defects. These intrinsic defects are usually oxidized by strong acids such as nitric acid, sulfuric acid, their mixture, or with other strong oxidative agents such as KMnO4/H2SO4 and oxygen gas. These agents open the tubes and subsequently allow attachment of chemical moieties onto the ends and Author's personal copy 246 X. Zhao, R. Liu / Environment International 40 (2012) 244–256 Table 1 In vivo toxic effects of CNTs to mammals. Organisms CNTs Size Dosage of CNTs Contamination Findings References SKH-1 mice Unpurified SWCNTs Not mentioned 5 days, with daily doses of 40 μg/mouse, 80 μg/mouse, or 160 μg/mouse Skin contact Murray et al., 2009 BALA/c mice Length: 10–20 μm, diameter: 100–150 nm 1 mg CNTs implanted into 1 cm incision Skin contact C57BL/6 mice Purified/ unpurified MWCNTs Unpurified MWCNTs Diameter: 10–50 nm 1 or 30 mg/m3 for 6 h per mouse Lung inhalation Wistar rats MWCNTs with defects Length: 0.7 μm, diameter: 20–50 nm 2 mg/rat a day Lung inhalation Pulmonary toxicology and genotoxicity C57BL/6 mice Raw MWCNTs Diameter: 10–20 nm, surface area: 100 m2/g 0.3, 1 or 5 mg/m3 for 6 h per day Lung inhalation Suppression of systemic immune function Kunming mice Raw MWCNTs C57BL/6 mice Nonpurified SWCNTs 50 nm outer diameter× 10 μm length Diameter: 0.8–1.2 nm, length: 100–1000 nm 90 min a time for 4 times a Lung day with initial inhalation concentration 80 mg/m3 5 mg/m3, 5 h/day for 4 days Lung inhalation Wistar rats Raw MWCNTs 0.1, 0.5, and 2 mg/m3, 6 h a Lung inhalation day, from Monday to Friday, for 13 weeks Wistar rats Raw MWCNTs Diameters: 5–15 nm, length: 0.1–10 μm, specific surface area: 250–300 m2/g Diameter: 10–16 nm, surface area: 253 m2/g 11 and 241 mg/m3 for 6 h followed by a postexposure period of 3 months Lung inhalation Wistar rats Raw MWCNTs (Baytubes) Diameter: 10–15 nm, length: 200–1000 nm surface area: 257 m2/g Lung inhalation C57BL/6J mice Raw MWCNTs with dispersant Mean length 3.9 μm long × 49 nm diameter Acute inhalation: 11 and 241 mg/m3, repeated inhalation: 0.1, 0.4, 1.62 and 5.98 mg/m3 10, 20, 40 and 80 μg/rat C57BL/6 mice Purified SWCNTs B6C3F1 mice Crl: CD(SD) IGS BR rats Sprague– Dawley rats Raw and purified SWCNTs Raw SWCNTs dispersed with Tween 80 Mean diameter and surface area: 1–4 nm and 1040 m2/g Diameter mostly below 1 μm 1.4 nm diameter× >1 μm length ICR mice Sprague– Dawley rats BALB/C mice Kunming mouse Nude mice C57BL/6 mice Mice Purified MWCNTs Length and diameter: 5.9 or 0.7 μm, or 9.7 and 11.3 nm, respectively Not mentioned SWCNTs dispersed with Pluronic F-68 MWCNTs Diameter:20 to 50 nm; dispersed with length: 0.5 to 2 μm albumin 40 nm by 0.5–5 μm Purified carboxylated MWCNTs dispersed with tween 80 MWCNT-COCl 12.6±3.2 nm by 269±160 nm 1–5 nm in diameter SWCNT-PEG and 100–300 mm in (polyethylene length glycol) and SWCNT-O-PEG Dispersed in BSA/ Four different samples saline solution of MWCNTs MWCNTs coated Diameter 10–30 nm with average length with Pluronic 2 μm F127 (PF127) Pharyngeal aspiration Oxidative stress, depletion of glutathione, oxidation of protein thiols and carbonyls, elevated myeloperoxidase activity, an increase of dermal cell number, and skin thickening resulting from the accumulation of polymorphonuclear leukocytes (PMNs) and mast cells Immunological toxicity and localized alopecia for unpurified material and goof biocompatibility for purified material Migration of MWCNTs to the subpleura, accumulation of pleural mononuclear cells and subpleural fibrosis Proliferation and thickening of alveolar walls SWCNT inhalation was more effective than aspiration in causing inflammatory response, oxidative stress, collagen deposition, and fibrosis as well as mutations of K-ras gene locus Inflammation and granuloma formation in the lung and associated lymph nodes, little substance-related systemic toxicity Koyama et al., 2009 RymanRasmussen et al., 2009 Fenoglio et al., 2008; Muller et al., 2008 Mitchell et al., 2007, 2009 Li et al., 2007 Shvedova et al., 2008 Ma-Hock et al., 2009 EllingerZiegelbauer and Pauluhn, 2009 An OEL of 0.05 mg Baytubes/m3 (time weighted Pauluhn, average) is considered to be protective to prevent lung 2010 in the workplace environment Pulmonary inflammogenicity was concentrationdependent and catalyst-independent MWCNT penetrations of alveolar macrophages, the alveolar wall, and visceral pleura are both frequent and sustained A robust but acute inflammation with early onset yet progressive fibrosis and granulomas Mercer et al., 2010 Lam et al., 2004 Warheit et al., 2004 Shvedova et al., 2005 A dose of 0, 10, 20, 40 μg/ mouse Intratracheal instillation 0, 0.1, or 0.5 mg/mouse Intratracheal instillation Intratracheal instillation Dose-dependent epithelioid granulomas, interstitial inflammation, peribronchial inflammation or death Transient inflammatory and cell injury effects, nondose-dependent series of multifocal granulomas or death 0.5, 2 or 5 mg/mouse Intratracheal instillation Pulmonary lesions, inflammatory and fibrotic reactions Muller et al., 2005 0.5 mg/mouse Intratracheal instillation Chou et al., 2008 0, 1, 10 or 100 μg/rat Intratracheal instillation Alveolar macrophage activation, various chronic inflammatory responses, and severe pulmonary granuloma formation Only evidence of apoptosis of alveolar macrophages 0.1 mg/mouse Injection Agglomerated CNTs in the lungs caused inflammatory responses while the well-suspended ones did not Qu et al., 2009 60 or 100 mg/kg Injection No observable sign of damage in spleen, however transference of CNTs from the red pulp to the white pulp No acute or chronic toxicity but some changes in red blood cells Deng et al., 2009 Schipper et al., 2008 Asbestos-like, length dependent inflammation and the formation of lesions known as granulomas CNTs can be used to avoid PF127-induced apoptosis in brain Poland et al., 2008 Bardi et al., 2009 1 or 5 mg/kg mouse 151 mg of SWCNT-PEG and Injection 47 mg of SWCNT-O-PEG for total mass 50 mg/mouse Injection 0.035 μg/mouse Injection Elgrabli et al., 2008a Author's personal copy X. Zhao, R. Liu / Environment International 40 (2012) 244–256 sidewall defects of the tubes. The attached moieties include small molecules, particles, bio-inspired moieties and coordination compounds (Banerjee et al., 2005; Karousis et al., 2010; Wildgoose et al., 2009). 2.2. Noncovalent modification Noncovalent modification refers to physical adsorption or wrapping of molecules or polymers onto the surface of CNTs, thus the structure and desired properties can be preserved while improving their solubility quite remarkably. Aromatic molecules, such as pyrene, porphyrin, their derivatives and surfactants, can interact with the sidewalls of CNTs by means of π–π stacking interactions, thus opening up the way for the noncovalent functionalization of CNTs (Zhao and Stoddart, 2009). To increase solubility of CNTs by noncovalent attachment of biomacromolecules in both aqueous and organic solutions has been researched recently at considerable length. Saccharides, peptides, proteins, DNA and other biomacromolecules are often employed in the noncovalent functionalization of CNTs through hydrophobic forces, π–π stacking interactions or electrostatic forces. Polymers, especially conjugated polymers, are also excellent candidate wrapping materials for the noncovalent functionalization of CNTs as a result of π–π stacking and van der Waals interactions between the conjugated polymers and the surfaces of CNTs (Karousis et al., 2010; Murakami and Nakashima, 2006; Zhao and Stoddart, 2009). 3. In vivo toxicity of CNTs In vivo testing is often employed over in vitro because it is better suited for observing the overall effects of an experiment on a living subject (Dhawan and Sharma, 2010; Mutlu et al., 2010). CNT in vivo toxicology has been researched recently in considerable depth and it has been shown that CNTs can induce toxic responses in multiple organ systems (Table 1). For humans, skin contact is a very likely route that may cause an inflammatory response via dermal toxicity. Exposure of mice to unpurified CNTs generally causes oxidative stress, depletion of glutathione, an increase of dermal cell number, localized alopecia and skin thickening (Koyama et al., 2009; Murray et al., 2009). CNT toxicity may be dependent upon the metal (particularly iron) content. Metals may interact with the skin, initiate oxidative stress, and induce redoxsensitive transcription factors thereby affecting/leading to inflammation. However, mice implanted with highly pure and clean tubes did 247 not experience any skin hair loss, which suggests that purification is an effective way to improve the biocompatibility of CNTs. The lungs are the most likely route of exposure to CNTs. Thus inhalation is the most relevant method for determining the toxicity of CNTs and there are several reasons for this. First, airborne CNTs are subject to the physics of impaction, sedimentation and diffusion, whereas instilled CNTs in aqueous suspension are not. Second, inhaled CNTs reach the distal regions of the lung. Third, inhaled CNTs are more dispersed and less agglomerated when compared to an instilled bolus dose in aqueous liquid. Hence, inhalation studies more accurately model deposition and pathologic responses to “real world” exposure scenarios. Observed results include defect-governed acute pulmonary toxicity, subpleural fibrosis, immune suppression, etc. CNTs have a high aspect ratio (length to width), a property shared with asbestos fibers, which has led to concern that inhalation of CNTs may cause similar lung pathologies (Crouzier et al., 2010; Elgrabli et al., 2008b; Tantra and Cumpson, 2007). However, the pathology of MWCNTs is different from the pleural pathology induced by asbestos fibers. MWCNTs caused focal subpleural fibrosis and mononuclear cell aggregation (Ryman-Rasmussen et al., 2009), whereas asbestos causes pleural inflammation (granulomas) and diffuse pleural fibrosis (Choe et al., 1997; Kane, 2006). Nevertheless, some CNT inhalation studies showed different results in mice, where the induction of lung damage like tissue damage or the presence of inflammation was not observed, which is contrary to what had been previously described in ordinary MWCNT inhalation studies. Large concentrations of CNTs may be present in occupational environments, which deserve particular attention from the standpoint of exposure. Limited data and guidelines are available for handling CNTs in occupational settings as well as research laboratories. Thus the study by Pauluhn (2010) is of great value for setting guidelines for occupational exposure in workplace environment and deriving a reasonable occupational exposure limit (OEL) to help the government draw up corresponding rules. While inhalation is the most physiologically relevant pathway to study the effect of lung exposure to pollutants, intratracheal instillation may still be a suitable assay to investigate the in vivo toxicity of fibers and particles. The first studies of CNT toxicity were performed by intratracheal instillation and showed that high doses of CNTs induced dose-dependent granulomas, fibrosis or animal death. The mechanism of lethality is probably associated with the impact of agglomerating the major airways in the rat while not the inherent toxicity of CNTs or metal catalyst. One finding which is worth noting is the study by Mercer et al. (2010), which shows rapid, frequent and Fig. 3. Lengths of fibers can affect clearance by macrophages. In this figure, (A) shows an incomplete, “frustrated” phagocytosis of long-fiber amosite and (B) shows a successful phagocytosis of short-fiber amosite; (D) shows “frustrated” phagocytosis of long CNTs by macrophages similar to the case of long-fiber amosite (A) (E is erythrocytes); and (E) demonstrates a successful engulfment of short CNTs by macrophages (successful phagocytosis). Foreign body giant cells were present after injection of long-fiber amosite (C) or long CNTs (F) (PMN is polymorphonuclear leukocyte). Reproduced with permission. Copyright © 2008 Macmillan Publishers Limited. Author's personal copy 248 X. Zhao, R. Liu / Environment International 40 (2012) 244–256 persistent CNT penetrations of alveolar macrophages and pleura. The significant and rapid transportation of CNTs to the intrapleural space is distinctly different from the decades-long process identified in asbestos exposed humans, the latter of which is considered to be the reason of mesothelioma development in terms of fibers. In addition to these findings, some attempts are made to explain the mechanism of inflammatory responses at the molecular level. It is presumed that CNTs with good biodistribution are not toxic enough to cause inflammation and formation of pathological structures in the lung because of the dispersibility and surface. Aside from the above ways of exposure, in vivo injections of CNTs are far from rare as protocols in recent research. The aggregation of CNTs still influences their mobility and bioavailability and thus their toxicity through injection. For well dispersed CNTs they may exhibit no acute general toxicity, no or temporary organ injury and delayed clearance. MWCNTs could behave like asbestos in that they are deposited at the alveolar level and reach the subpleura after injection. Formation of granulomas that surround MWCNTs or asbestos fibers was only found in the long-fiber treated samples, which was not associated with metal residuals (Fig. 3). Table 2 In vitro toxic effects of CNTs to mammal cell lines. Cell lines CNTs Size Dosage of CNTs Detection Findings References Wick et al., 2007 SWCNTs purified by heat and acid or suspended in polyoxyethylene sorbitan monooleate and tween 80 solution A549 human pneumocytes MWCNTs suspended in Arabic gum solution Diameter: 20 nm for well- 7.5, 15 and dispersed bundles, more 30 μg for 3000 than 100 nm for agglom- cells erated tubes DNA quantity Toxicity was found to increase from by Hoechst well-dispersed CNTs to asbestos and 33258, cell then to agglomerated CNTs activity by MTT 0.1–12 μm or 0.1–3.5 μm by 10–160 nm MTT, XTT, LDH CNTs were more toxic than metal oxide SimonDeckers et assay, etc. nanoparticles and they did not have al., 2008 length-dependent cytotoxicity Human lung epithelial H460 cells SWCNTs with or without purification Not mentioned Mouse peritoneal macrophage-like cells SWCNTs in Pluronic F80 surfactant 1.0 nm average diameter, 1 μm average length 3.8 μg/mL treatment Guinea pig alveolar macrophages SWCNTs and MWCNTs were freshly suspended in culture media for use 0.38 μg/cm2, 0.76 μg/cm2 or 3.06 μg/cm2 RAW 264.7 macrophages Purified SWCNTs Diameter of 1.4 nm and length of 1 μm for SWCNTs, diameter of 10– 20 nm and length of 0.5– 40 μm for MWCNTs Mean diameter and surface area: 1–4 nm and 1040 m2/g 0.1 mg/mL treatment Modified Bradford assay, etc. RAW 264.7 macrophages Purified and non-purified SWCNTs 0.12–0.5 mg/ mL treatment DHE oxidation assay, etc. A mouse macrophage cell line, J774.1, and CHO-K1 cells MWCNTs suspended in 10% endotoxin-free Pluronic F68 Diameter:1–4 nm, length: 950 m2/g (non-purified) or 1040 m2/g (purified) Average diameter, 67 nm; surface area, 26 m2/g 0 to 100 μg/mL treatment Modified MTT essay, SDSPAGE, etc. Different types of neuronal and glial cells from Leghorn chicken embryos SWCNT-agglomerates (Sa) and SWCNT-bundles (S-b) Diameter: 100 nm (S-a) or 20 nm (S-b) 0 to 30 μg/mL treatment Hela cells Purified MWCNTs coated with serum proteins 600–800 nm long with diameter of 40–100 nm 100 μg/mL treatment SAOS-2, a human osteoblast-like cell line Purified or nonpurified SWCNTs Not mentioned DNA content by Hoechst 33258, cytotoxicity by cell-based ELISA Cytotoxicity by MTT method MTS method 6.25 or 7.8 μg/ cm2 CNTs for 25,000 to 63,000 cells/ cm2 1000, 200, or MTT,WST-1 40 μg/mL for assays 4 9 × 10 per well Human MSTO-211H cells A diameter of 10–30 nm Clonal pheochromocytoma Sodium dodecyl sulfate dispersed CNTs or CNT-PC and a length of 5–15 μm cells and human colon (phosphoryl choline) carcinoma cell lines Cells were seeded and exposed to 0.25 to 100 μg/mL CNTs after 24 h 1.0 mg/mL treatment The uptake of Ni inside the cells was assessed Near-infrared fluorescence imaging MTT reduction experiment MTT: 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazoliumbromide. XTT: 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino)carbonyl]. LDH: lactate dehydrogenase. DHE: dihydroethidium. SDS-PAGE: sodium dodecyl sulfate polyacrylamide gel electrophoresis. ELISA: enzyme-linked immunosorbent assay. MTS: 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium inner salt. WST-1: 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium. Ni encapsuled in the CNTs is released and cause cytotoxicity Liu et al., 2007 Macrophage cells actively ingest significant quantities of SWCNTs without showing toxic effects Cytotoxicity: SWCNTs > MWCNTs > quartz > C60 Cherukuri et al., 2004 Jia et al., 2005 Transforming growth factor-β1 (TGF-β1) production, less tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β), and no superoxide or nitric oxide production or apoptosis Inflammatory reactions and oxidative stress Shvedova et al., 2005 Cytotoxic effects in phagocytotic cells by reacting with collagenous structure (MARCO) on the plasma membrane and rupturing the plasma membrane Cytotoxicity: S-a > S-b Hirano et al., 2008 The extent of toxicity attenuation increased with increasing amounts of serum proteins adsorbed on CNTs SWCNTs films are not toxic for human osteoblasts Zhu et al., 2009 Kalbacova et al., 2007 Both CNTs and CNT-PC induce no cytotoxicity Zhang et al., 2008 Kagan et al., 2006 Belyanskaya et al., 2009 Author's personal copy X. Zhao, R. Liu / Environment International 40 (2012) 244–256 While we know from the above that pristine CNTs can induce negative effects to organisms; interestingly, the in vivo injection of MWCNTs can protect against apoptosis induced by Pluronic F127 (PF127), another toxic material. Pluronic-coated CNTs may be biocompatible and have potential applications in nanomedicine (Bardi et al., 2009). Obviously, conflicting results were found in the above research. Do CNTs really cause toxicity? Toxicity is the degree to which a substance can damage an object that has a metabolic function. Toxicity can refer to the effect on a whole organism, such as an animal, bacterium, or plant, or the effect on a substructure of the organism, such as a cell (cytotoxicity) or an organ (organotoxicity) such as the liver (hepatotoxicity). Therefore, some changes in organisms cannot be considered to be toxic responses. Organisms will have a natural response when stimulated by foreign objects. Even flour powder can lead to pulmonary changes when taken in (Ren et al., 2010). Besides, the toxicity of CNTs seems to depend on many factors including dosage, impurities, pretreatment processes, physical form, surface chemistry, aggregation degree, etc. However, no uniform criteria have been found in these studies, and the authors used different techniques under different conditions. 4. In vitro toxicity of CNTs The influence or possible toxic effects of CNTs on cells are an essential focus in evaluating and understanding CNT compatibility versus toxicity (Cheng et al., 2009a; Davoren et al., 2007; Gellein et al., 2009; Thurnherr et al., 2009; Tutak et al., 2009). Cell and CNT interactions of focus include cellular uptake and processing of CNTs by different routes, effects on cell signaling, membrane perturbations, production of cytokines, chemokines and reactive oxygen species (ROS), overt toxic reactivity, cell apoptosis, and no obvious toxicity (Hillegass et al., 2010). Generally, in vitro culture of cell lines or primary cells (tissue-harvested) on plastic plates, with/without serum, with bolus dosing of CNTs and subsequent cell activity characterization are the most common methods. Several types of cells have been selected for testing including neural, phagocytic and various cancer cell lines, etc. (Table 2). The toxicity was found to increase from well-dispersed CNTs to asbestos and then to agglomerated CNTs for human lung cancer cell lines (Wick et al., 2007). CNTs are more toxic than metal oxide nanoparticles and do not possess length-dependent cytotoxicity, even in the presence of metal catalyst impurities (Simon-Deckers et al., 2008). Probably because the proportions of CNTs that are too long for the cells to engulf in long and short samples are almost the same, and because CNTs that cannot be engulfed by cells cause similar effects no matter how long they are, the length made no significant difference. Besides, the average diameters of the two types are not unified and may cause chaos in criteria. Also, the catalyst content of the tubes was comparatively low (b15%, wt). Results were seen where functionalized CNTs showed that less cellular toxicity and purification could increase their biocompatibility (Liu et al., 2007). When the catalyst (mostly nickel) content is high, its cytotoxicity emerges. Importantly, excessive catalyst is bound to influence the chemical and physical properties of CNTs, and those nanotubes could be called “metal compounded CNTs”. Macrophages play a key role in removing inhaled, instilled or injected particulate substances from tissues or cellular surface, therefore their biological response to CNTs should be addressed in depth. Relevant toxicity data can be found in the early literature. Murine peritoneal macrophages actively ingest a large quantity of SWCNTs without any sign of cytotoxicity even at a dose of 7.3 μg/mL (Cherukuri et al., 2004). The cytotoxicity of SWCNTs seems to be higher than that of MWCNTs, while CNTs are more cytotoxic than quartz or fullerenes (Jia et al., 2005). SWCNTs were found to trigger abnormality of cytokine, and did not cause superoxide or nitric oxide production or apoptosis (Shvedova et al., 2005). However, 249 iron-rich SWCNTs produced hydroxyl radicals efficiently in zymosan- or phorbol ester-stimulated cells, and induced oxidative stress (Kagan et al., 2006). Moreover, it has been shown that in vitro exposure to MWCNTs could trigger cytotoxic effects in phagocytotic cells by reacting with macrophage receptors with collagenous structure (MARCO) on the plasma membrane and rupturing the plasma membrane (Hirano et al., 2008). For other types of cell lines, Belyanskaya et al. (2009) measured the toxicity to neurons and glial cells and found SWCNTs significantly decreased the overall DNA content, which is agglomeration related. Serum proteins adsorbed on CNTs were found to attenuate the inherent cytotoxicity of CNTs, and the extent of toxicity attenuation increased with increasing amounts of serum proteins adsorbed on CNTs (Zhu et al., 2009). These results are generally in accordance with the in vivo studies. Similar to the in vivo studies, although it is generally believed that well-dispersed CNTs (especially using biodispersant) are less toxic than agglomerated ones, some observed toxicity of CNTs differs from one in vitro study to another. CNTs have unique properties which have been elucidated in the above sections and the modification conditions and surface chemistry may also influence the evaluation of their toxicity. Particularly, it has been found that the results of the MTT assay can be influenced by the surfactants used to disperse the CNTs. Moreover, CNTs are able to reduce MTT to the MTTformazan form in the absence of cells or enzymes, and the reduction ability is associated with the purification procedure of CNTs (Belyanskaya et al., 2007). Therefore careful validation and extreme caution are needed in experiments where CNTs are one of the constituents so as to avoid a potential bias in concluding and interpreting results of cytotoxicity studies. Diverse assessment methods which are commonly used were also compared and considerable variation was found depending on the dye employed. Interactions of various degrees between the indicator dyes with CNTs may occur. This phenomenon undoubtedly raises questions about the reliability of the cytotoxicity data based on the absorption/fluorescence emission of these dyes. Moreover, as shown in Fig. 4 the toxic dosage for the same cell varies according to the dye used. Thus the indicator dyes are not appropriate for the quantitative assessment of CNT toxicity and new screening technologies are required which do not involve interferences with CNTs (Casey et al., 2007). 5. Toxicity studies on species other than mammals CNTs may be accidentally or incidentally released to the environment at different stages of their life cycle, whether through use or disposal (Lee et al., 2010). Leaching of other hazardous materials from various products has been previously reported, including Ni/Cd from batteries (Jennings et al., 2009), lead from paint (Turner, 2010), organo-tin from ship hull paint (Oliveira and Santelli, 2010), asbestos from tiles (Murbach et al., 2008), and bisphenol A from food containers (Carwile et al., 2009). CNTs are considered to have potential mobility and therefore impacts in and across air, water, soil, and biota. Once their environmental release occurs, they may induce unforeseen environmental impacts not only to humans and but also to ecosystem health. Thus, understanding the ecotoxicity of CNTs and assessing the environmental exposure of a wider range of species to CNTs should provide valuable insight into likely exposure scenarios. The aquatic environment is particularly at risk of exposure to CNTs, as it acts as a sink for most environmental contaminants. Research on aquatic species mainly has focused on Daphnia magna, Tetrahymena thermophila and selected fish species. SWCNTs have been shown to be bioavailable to aquatic organisms, as both watersoluble (wrapped with synthetic peptide) and insoluble SWCNTs were detected in the fecal material collected from the digestive tract of the exposed fathead minnow. For the fish exposed to water- Author's personal copy 250 X. Zhao, R. Liu / Environment International 40 (2012) 244–256 Fig. 4. (A)–(D) Cytotoxicity of SWCNT to A549 cells after 24 h exposure determined by the Commassie, Alamar Blue, Neutral Red and MTT assay, respectively. Data are expressed as percent of control mean ± SD of six independent experiments. *Denotes a significant difference from the control (P ≤ 0.05). Reproduced with permission. Copyright © 2007 Elsevier B.V. soluble SWCNTs, clumps of SWCNTs were also found on the gill, but similar clusters were not visible in fish exposed to insoluble SWCNTs (Helland et al., 2007). Raw CNTs associated with contaminants were shown to induce hatching delay in zebrafish embryos (Cheng et al., 2007), while the BSA-coated MWCNTs could cause immune responses at early stages and were distinctively excluded from the yolk cell. They were found to move easily in the compartments and finally were cleaned out by the body at 96 h after the loading. However, the larvae of the second generation had an obviously lower survival rate, which shows the reproductive toxicity of CNTs (Cheng et al., 2009b). D. magna (water flea, up to 5 mm in length) are very sensitive to aquatic environmental changes. When lysophophatidylcholinecoated SWCNTs are released to water, D. magna are able to ingest the tubes and utilize lysophophatidylcholine as a food source. Different from fish, accumulation of SWCNTs was found on the external surface of D. magna and acute toxicity was observed only at concentrations more than 1 mg/L (Roberts et al., 2007). T. thermophila (50 μm in length), an important organism in wastewater treatment and an indicator of sewage effluent quality, has also been studied as a model organism. SWCNTs were internalized by T. thermophila, which interestingly caused the protozoa to aggregate and impeded their ability to ingest and digest their prey bacteria species (Ghafari et al., 2008). Worryingly, CNTs do stay in the digestive tract of some bottom consumers in the ecological pyramid, thus this material could move up through the food chain as these worms and other organisms are consumed by benthivores. Recent years have seen an increasing number of reports on the influence of CNTs to plants or plant cells. Indeed, several important lessons can be learned from the careful investigation of the in vivo and in vitro toxicity of CNTs. Nonfunctionalized nanotubes inhibited root elongation in tomato but enhanced that of onion and cucumber. Functionalized nanotubes inhibited root elongation in lettuce while cabbage and carrots were not affected by either form of nanotubes. Fig. 5. Effect of CNTs on tomato seed germination and seedling growth. (A) Phenotype of tomato seeds incubated during 3 days with and without CNTs in medium. (B) Phenotypes of 27-day-old tomato seedlings growing on medium with different concentrations of CNTs. Reproduced with permission. Copyright © 2009 American Chemical Society. Author's personal copy X. Zhao, R. Liu / Environment International 40 (2012) 244–256 To date, studies have shown that functionalized CNTs are generally less toxic than nonfunctionalized ones (Canas et al., 2008). In one recent work by Khodakovskaya et al., they did not find any toxic effects of CNTs on root development and root elongation of tomato seedlings up to a CNT concentration of 40 mg/L (Khodakovskaya et al., 2009). In fact, they demonstrated that exposing tomato seeds to CNTs can increase the germination percentage and support and enhance the growth of seedlings (see Fig. 5). It was hypothesized that the mechanism is that the CNTs penetrate the seed coat and enhance the amount of water uptake for seeds during the germination period, which is essential to seed metabolism and growth at early stages. However the mechanism by which nanoparticles can support water uptake inside seeds is not clear yet. Besides, the biodistribution of CNTs and their influence on the next stages of the plant's growth (e.g. anthesis and bearing) has not yet been determined. For in vitro studies, experimental evidence shows that nonfunctionalized MWCNTs are toxic to Arabidopsis T87 cells, depending on the size of agglomerates (Lin et al., 2009). Similar to the toxicity for mammalian cells, agglomerates of smaller size induce stronger toxicity than those of larger size, which indicates surface-dependent characteristics. Another study showed that reactive oxygen species (ROS) in rice cells increased and the cell viability decreased when they were cultured in suspension with MWCNTs. When antioxidant was introduced into the culture medium, the ROS content and cell viability returned to baseline levels (Tan et al., 2009). Both SWCNTs and MWCNTs manifest antibacterial properties. The mechanism of microbial toxicity of SWCNTs appears to be direct damage to cell walls (Kang et al., 2007; Kang et al., 2009), while MWCNTs cause toxicity via oxidative stress (Kang et al., 2008a, b). Toxicity of MWCNTs is generally greater than that of SWCNTs; however it strongly depends on their physicochemical properties, aqueous solution chemistry, and the microbial species. These results call for the standardization of protocols for analyzing the potential impacts of CNTs in ecosystems. 251 Because of the complexity of ecosystems, the microenvironment of CNTs that are released into the environment may be more complicated than lab studies. Temperature, pH, and interactions with natural substances may alter the surface chemistry significantly so that the initial surface chemistry of the nanotubes upon entrance into the environment becomes less important. Some CNTs with certain surfaces may be toxic in the lab and inert in natural environment, it is also likely that other CNTs will become more lethal as a result of long abiotic and biotic processing in complex ecosystems. Besides, the release of toxic coatings and metals from the catalyst in CNTs should be assessed by measuring the toxicity of the filtrate after nanotube removal. These are all important topics for future research. 6. Biomacromolecules and CNT interactions When CNTs enter a biological fluid, the first step would be association with biomacromolecules, mostly proteins (Cedervall et al., 2007b). When we think of the interactions of CNTs with a living system, we are really speaking of protein-coated CNTs (Klein, 2007), thus probing the association of protein and CNTs would be of great value. Not only this, some functional proteins (e.g. enzyme, serum albumin) may be damaged by CNTs and lose their functions (Zhang et al., 2009), further influencing the performance of the organism. However the information in this field is very limited. In previous work of our laboratory, we systematically studied the interaction between BSA and MWCNTs using spectroscopic methods. We found that BSA noncovalently attached to the surfaces of carboxylated MWCNTs at the expense of partly losing its secondary and tertiary structure through electrostatic interaction (Zhao et al., 2010). Mu and colleagues found that electrostatic and stereochemical properties of both nanotubes and proteins govern nanotube-protein binding. The curvature of nanoparticles plays a key role in determining the protein binding affinity (Mu et al., 2008). However, on drawing the Stern–Volmer plots they failed to consider the inner filter effect Fig. 6. Toxicology of CNTs: an interdisciplinary and newly-emerging field. An overview of the principal components of CNT toxicology, from pretreatment of CNTs to the pending mechanisms and the aim of the subject. Author's personal copy 252 X. Zhao, R. Liu / Environment International 40 (2012) 244–256 which was caused by the absorption of fluorescents and quenchers. Researchers must pay attention to the inner filter effect when using fluorescence spectroscopy in case of false quenching effects, for the mixture of nanomaterials and biomacromolecules usually absorbs a lot at both excitation and emission wavelengths. Wijaya et al. made a study concerning the role carboxylic groups play in the interaction of BSA and SWCNTs. They compared pristine and carboxylated SWCNTs and found that functionalized CNTs cause deeper changes to the protein conformation than their pristine counterparts (Wijaya et al., 2009). CNTs with larger diameters and more functional groups are known to be less toxic to organisms; however, the contrary seems to be true for proteins. It is interesting to find these differences for CNT bio-interactions macroscopically and microscopically. Concerning the impacts that CNTs exert on the function of protein, it is reported that the interaction of carboxylic groups on CNTs and RNase A may force the protein to lose its native secondary structure and greatly reduce its enzymatic activity (Yi et al., 2008). Another effort showed that the interaction with CNTs disrupts and blocks the active sites of human yes-associated protein from binding to the corresponding ligands, thus leading to loss of the original function of the protein. One key to note is the small diameter of the CNTs (8 Å), which makes it possible for them to insert into the hydrophobic core of the protein (Zuo et al., 2010). When CNTs enter a biological fluid, proteins and other biomolecules rapidly compete for binding to the CNT surfaces, leading to the formation of a dynamic protein corona that critically defines the biological fate of CNTs. Therefore, to study a single class of protein interacting with CNTs seems to be both lop-sided and limited. Proteomic analysis holds the promise to be a powerful tool to reflect early biological responses. To fill the gap between theoretical studies and a real world scenario, other advanced techniques are encouraged to be applied to establish connections between this area and in vitro investigation. 7. Outlook and suggestions for future research The wide variety of types of CNTs has brought them strength in innovation but also presents challenges of safe use because of their potential toxicity. As described above and in the literature, advances in nanotoxicity research are imperative and come from reliable toxicity tests (Brain et al., 2009; Dhawan et al., 2009; Jones and Grainger, 2009; Oberdorster et al., 2007; Teeguarden et al., 2007; Warheit et al., 2007). After reviewing the literature, we noticed that different pretreatment methods of CNTs are used and additionally there are no uniform criteria of physical properties and chemical content, therefore this area in particular requires urgent attention. One of the solutions to this discrepancy is metrology and developing tools to characterize and measure the relevant attributes of CNTs, including diameter, length, dose, surface chemistry, element analysis and surface area, etc. (Fig. 6). Toxicological investigations of other nanoparticles have demonstrated the importance of size and two examples are cited to suggest that the small size effect plays a role at least for certain types of particles, which may lead to progress in the study of CNTs. First, CNTs may enter the body via novel routes: translocation of inhaled nanoparticles to the brain through the olfactory neuronal pathway has been reported and was not predicted from previous studies of larger particles (Oberdorster et al., 2004). Second, CNTs may interact in unanticipated ways with cellular components: intercalation of nano-sized Au55 gold clusters with DNA was shown and this appears to be related to the specific size of these nanostructures (Tsoli et al., 2005). Disparity in functionalization of CNTs is also important. When CNTs are functionalized or tethered with organic groups, it becomes important to quantify the amount of functional groups which reside at the surface of CNTs, their binding strength (e.g. covalent or ionic) to the tubes, and their availability to perform their anticipated function. Spectroscopic techniques such as infrared spectroscopy, UV–vis spectroscopy and Raman spectroscopy are invaluable both for the identification and locating the presence and position of functional groups on CNTs. Moreover, the state of dispersion of CNTs is of obvious importance for nanotoxicological studies. Indeed, some engineered CNTs, which usually lack hydrophilic functional groups, tend to aggregate and agglomerate rapidly to form larger particles. The dimensions of the aggregates often exceed the nanometric range, which may introduce an additional level of complexity to characterizing nanotube behavior in cell culture. Commonly used methods of CNT sizing currently include transmission electron microscopy (TEM), scanning electron microscopy (SEM), UV–vis spectroscopy, and fluorescence polarization. Each method possesses its own inherent uncertainties, making corroboration of results with one or more additional methods a desirable means of sizing or aggregation determination. All in all, consistency in the reporting of physico-chemical characteristics of CNTs is essential to allow for comparison of toxicity data across different studies. The crux of the experimental method is the testing process, but what implications will it have on our current understanding of CNT toxicity if some results arise from internally invalid experiments? Data obtained from in vitro experiments could be misleading for a variety of reasons. CNTs are known to exhibit a high surface area per mass. Therefore, adsorption of chemicals used for toxicology testing onto CNTs could represent problems for in vitro investigation of CNT cytotoxicity, which has been discussed above. Additionally, common amino acids and vitamins (e.g. phenylalanine and folate) may passively adsorb onto CNTs, giving artifactual cell death (Guo et al., 2008). Such adsorption of nutrients and assay reagents can be avoided by using relatively low concentrations (μg/mL) of nanoparticles for in vitro assays (Geraci and Castranova, 2010). Thereby it is recommended to use in vitro doses (μg/surface area of cultured cells) representative of in vivo lung burdens (μg/alveolar epithelial cell surface area) and future research should take caution from such findings. Additionally, results from imaging assays and biochemical tests also require skepticism. A chronic toxicity study showed that images of SWCNTs in the spleen can be obscured by other compounds endogenous to cells like the hemosiderin in splenic macrophages (Firme and Bandaru, 2010). Importantly, CNTs may also bind to proteins in biological fluids, which in turn could affect their biological performance. Recent landmark publications have thus indicated that adsorbed proteins could play an important role in modulating the uptake and toxicity of nanoparticles (Cedervall et al., 2007a; Dutta et al., 2007). Based on these findings we caution investigators to be wary of forming conclusions based on single biological assays, isolated cell lines, or protein-free media, and to strengthen observations by correlating measurements from multiple assays (Fig. 6). In the past decade much progress has been made in understanding what influences CNTs have on a given type of cell or species. However, the details of the proposed mechanisms are still debated (Donaldson et al., 2010; Johnston et al., 2010). To date, two major mechanisms have been widely considered as the uptake mechanisms of nanoscale materials: (1) endocytosis/phagocytosis (Cherukuri et al., 2004) and (2) nanopenetration (Firme and Bandaru, 2010). The former mechanisms are common physiological ones representing the engulfing of an extracellular particle (e.g. viruses and bacteria) by the cell. They are energy dependent and are hindered at low temperatures and in low adenosine-triphosphate (ATP) environments. Nanopenetration is an energy-independent, passive process, where CNTs diffuse across the cellular membrane (Singh et al., 2005). On the issue of determining the precise mechanism at the molecular biological level, however, few efforts have been made. The emerging field of nanotoxicogenomics (Ding et al., 2005; Zhang et al., 2006) which attempts to correlate global gene expression profiles of cells or tissues exposed to CNTs with biological/toxicological responses using cDNA microarray technologies may provide useful information in this regard. Potentially, the application of two-dimensional electrophoresis and mass-spectrometry methods (proteomics) could also be Author's personal copy X. Zhao, R. Liu / Environment International 40 (2012) 244–256 helpful in interpreting the mechanism and enhancing our understanding of the biological responses induced by CNTs (Sheehan, 2007). For this reason, one of the most significant tasks for nanotoxicologists in the second decade of this century should be to conduct mechanism-driven research in order to provide a solid scientific basis and comprehensive understanding for safety and risk assessment of CNTs. Furthermore, recent studies show a lack of correlation between in vivo and in vitro effects of CNTs. Indeed, while in vitro assays may provide useful mechanistic information, revealing cell-type-specific responses, such assays may fail to capture intercellular effects such as cross-talk between inflammatory cells in culture media and signals that occur in the natural state. In vitro tests integrated with in vivo ones are advocated. Future testing would then benefit by using mutually exclusive assays in order to reconcile variations in toxicity. A major complication is the chance that in vivo toxicity may just be a pronounced inflammatory response. Other reactions like white blood cell buildup and fibrinogenesis may also occur, and tests must be configured to consider these alterations. Another suggestion which is worth proposing is that while the current tests for evaluation of drugs and devices may be appropriate to detect many risks associated with CNTs, it cannot be assumed that these assays will detect all potential risks. Therefore, additional, specific assays may be needed to measure unknown effects in the future. CNTs are possibly one of the least biodegradable man-made materials ever devised. Once they are released into environmental compartments, one can therefore not exclude the possibility that CNTs may accumulate and their toxicity may biomagnify as they travel up food chains, due to their biopersistent and lipophilic nature. In work on quantum dots (QDs) made from cadmium, selenium, zinc and sulfate, whose fluorescing properties make them easy to detect microscopically, David Holbrook and colleagues reported that carboxylated and biotinylated QDs may not accumulate in organisms at higher trophic levels, but the researchers were quick to add that much more work is required before any generalizations can be made regarding environmental and human safety of nanomaterials (Burton, 2008; Holbrook et al., 2008). This result is evidently not proof that CNTs will pose no environmental threat; consequently, such a scenario should also be evaluated with CNTs. For these reasons we emphasize a further need to study CNT ecotoxicity as well as longer time-scale impacts covering bioaccumulation, biopersistence, and negative effects on reproduction. 8. Conclusions Although many efforts have been made to carefully investigate the in vitro and in vivo toxicity of CNTs, researchers still fail to reach consensus on the toxicity of CNTs and the mechanism, as the core of nanotoxicology, remains out of reach. Fortunately, several experimental issues which may lead to these uncertainties have been progressively unveiled. Modification and characterization of CNTs are not unified. The methodology can also be problematic as interference by CNTs during the evaluation cannot be neglected. In view of the issues above, multiple measurement techniques are required to assess CNT toxicity, because no single analysis can provide sufficient information. Characterization of CNTs and assessment of their effect on the cells, organ, or entire organism should also be standardized systematically so that nanotoxicity mechanisms can be uncovered and the safe use of CNTs can be achieved. Furthermore, the ecotoxicology of CNTs urgently needs to be studied, as it is essential for proper development of government regulations for the use of CNTs. Acknowledgements This work is supported by NSFC (20875055, 81161120413), the Cultivation Fund of the Key Scientific and Technical Innovation Project, 253 Ministry of Education of China (708058), and Key Science–Technology Project in Shandong Province (2008GG10006012) is also acknowledged. The authors thank Dr. Pamela Holt for editing the manuscript. References Banerjee S, Hemraj-Benny T, Wong SS. Covalent surface chemistry of single-walled carbon nanotubes. Adv Mater 2005;17:17–29. Bardi G, Tognini P, Ciofani G, Raffa V, Costa M, Pizzorusso T. Pluronic-coated carbon nanotubes do not induce degeneration of cortical neurons in vivo and in vitro. Nanomed Nanotechnol Biol Med 2009;5:96-104. Belyanskaya L, Manser P, Spohn P, Bruinink A, Wick P. The reliability and limits of the MTT reduction assay for carbon nanotubes–cell interaction. Carbon 2007;45:2643–8. Belyanskaya L, Weigel S, Hirsch C, Tobler U, Krug HF, Wick P. Effects of carbon nanotubes on primary neurons and glial cells. Neurotoxicology 2009;30:702–11. Brain JD, Curran MA, Donaghey T, Molina RM. Biologic responses to nanomaterials depend on exposure, clearance, and material characteristics. Nanotoxicology 2009;3: 174–80. Brumfiel G. Nanotechnology: a little knowledge. Nature 2003;424:246–8. Buchoux J, Aime JP, Boisgard R, Nguyen CV, Buchaillot L, Marsaudon S. Investigation of the carbon nanotube AFM tip contacts: free sliding versus pinned contact. Nanotechnology 2009;20. Burton A. Nanotechnology: nano-food chain link examined. Environ Health Perspect 2008;116:A336. Canas JE, Long MQ, Nations S, Vadan R, Dai L, Luo MX, et al. Effects of functionalized and nonfunctionalized single-walled carbon nanotubes on root elongation of select crop species. Environ Toxicol Chem 2008;27:1922–31. Carwile JL, Luu HT, Bassett LS, Driscoll DA, Yuan C, Chang JY, et al. Polycarbonate bottle use and urinary bisphenol a concentrations. Environ Health Perspect 2009;117: 1368–72. Casey A, Herzog E, Davoren M, Lyng FM, Byrne HJ, Chambers G. Spectroscopic analysis confirms the interactions between single walled carbon nanotubes and various dyes commonly used to assess cytotoxicity. Carbon 2007;45:1425–32. Cedervall T, Lynch I, Lindman S, Berggård T, Thulin E, Nilsson H, et al. Understanding the nanoparticle–protein corona using methods to quantify exchange rates and affinities of proteins for nanoparticles. Proc Natl Acad Sci 2007a;104:2050–5. Cedervall T, Lynch I, Lindman S, Berggard T, Thulin E, Nilsson H, et al. Understanding the nanoparticle–protein corona using methods to quantify exchange rates and affinities of proteins for nanoparticles. Proc Natl Acad Sci U S A 2007b;104:2050–5. Cheng JP, Flahaut E, Cheng SH. Effect of carbon nanotubes on developing zebrafish (Danio rerio) embryos. Environ Toxicol Chem 2007;26:708–16. Cheng C, Muller KH, Koziol KKK, Skepper JN, Midgley PA, Welland ME, et al. Toxicity and imaging of multi-walled carbon nanotubes in human macrophage cells. Biomaterials 2009a;30:4152–60. Cheng JP, Chan CM, Veca LM, Poon WL, Chan PK, Qu LW, et al. Acute and long-term effects after single loading of functionalized multi-walled carbon nanotubes into zebrafish (Danio rerio). Toxicol Appl Pharmacol 2009b;235:216–25. Cherukuri P, Bachilo SM, Litovsky SH, Weisman RB. Near-infrared fluorescence microscopy of single-walled carbon nanotubes in phagocytic cells. J Am Chem Soc 2004;126:15638–9. Choe N, Tanaka S, Xia W, Hemenway DR, Roggli VL, Kagan E. Pleural macrophage recruitment and activation in asbestos-induced pleural injury. Environ Health Perspect 1997;105:1257–60. Chou CC, Hsiao HY, Hong QS, Chen CH, Peng YW, Chen HW, et al. Single-walled carbon nanotubes can induce pulmonary injury in mouse model. Nano Lett 2008;8: 437–45. Crouzier D, Follot S, Gentilhomme E, Flahaut E, Arnaud R, Dabouis V, et al. Carbon nanotubes induce inflammation but decrease the production of reactive oxygen species in lung. Toxicology 2010;272:39–45. Davoren M, Herzog E, Casey A, Cottineau B, Chambers G, Byrne HJ, et al. In vitro toxicity evaluation of single walled carbon nanotubes on human A549 lung cells. Toxicol in Vitro 2007;21:438–48. Deng XY, Wu F, Liu Z, Luo M, Li L, Ni QS, et al. The splenic toxicity of water soluble multi-walled carbon nanotubes in mice. Carbon 2009;47:1421–8. Dhawan A, Sharma V. Toxicity assessment of nanomaterials: methods and challenges. Anal Bioanal Chem 2010;398:589–605. Dhawan A, Sharma V, Parmar D. Nanomaterials: a challenge for toxicologists. Nanotoxicology 2009;3:1–9. Ding L, Stilwell J, Zhang T, Elboudwarej O, Jiang H, Selegue JP, et al. Molecular characterization of the cytotoxic mechanism of multiwall carbon nanotubes and nanoonions on human skin fibroblast. Nano Lett 2005;5:2448–64. Donaldson K, Poland CA, Schins RPF. Possible genotoxic mechanisms of nanoparticles: criteria for improved test strategies. Nanotoxicology 2010;4:414–20. Dutta D, Sundaram SK, Teeguarden JG, Riley BJ, Fifield LS, Jacobs JM, et al. Adsorbed proteins influence the biological activity and molecular targeting of nanomaterials. Toxicol Sci 2007;100:303–15. Elgrabli D, Abella-Gallart S, Robidel F, Rogerieux F, Boczkowski J, Lacroix G. Induction of apoptosis and absence of inflammation in rat lung after intratracheal instillation of multiwalled carbon nanotubes. Toxicology 2008a;253:131–6. Elgrabli D, Floriani M, Abella-Gallart S, Meunier L, Gamez C, Delalain P, et al. Biodistribution and clearance of instilled carbon nanotubes in rat lung. Part Fibre Toxicol 2008b;5. Ellinger-Ziegelbauer H, Pauluhn J. Pulmonary toxicity of multi-walled carbon nanotubes (Baytubes (R)) relative to alpha-quartz following a single 6 h inhalation exposure of rats and a 3 months post-exposure period. Toxicology 2009;266:16–29. Author's personal copy 254 X. Zhao, R. Liu / Environment International 40 (2012) 244–256 Fenoglio I, Greco G, Tornatis M, Muller J, Rayrnundo-Pinero E, Beguin F, et al. Structural defects play a major role in the acute lung toxicity of multiwall carbon nanotubes: physicochemical aspects. Chem Res Toxicol 2008;21:1690–7. Firme III CP, Bandaru PR. Toxicity issues in the application of carbon nanotubes to biological systems. Nanomedicine 2010;6:245–56. Gellein K, Hoel S, Evje L, Syversen T. The colony formation assay as an indicator of carbon nanotube toxicity examined in three cell lines. Nanotoxicology 2009;3: 215–21. Geraci CL, Castranova V. Challenges in assessing nanomaterial toxicology: a personal perspective. Wiley Interdiscip Rev Nanomed Nanobiotechnol 2010;2:569–77. Ghafari P, St-Denis CH, Power ME, Jin X, Tsou V, Mandal HS, et al. Impact of carbon nanotubes on the ingestion and digestion of bacteria by ciliated protozoa. Nat Nanotechnol 2008;3:347–51. Gomez LM, Kumar A, Zhang Y, Ryu K, Badmaev A, Zhou CW. Scalable light-induced metal to semiconductor conversion of carbon nanotubes. Nano Lett 2009;9: 3592–8. Guo L, Von Dem Bussche A, Buechner M, Yan A, Kane AB, Hurt RH. Adsorption of essential micronutrients by carbon nanotubes and the implications for nanotoxicity testing. Small 2008;4:721–7. Helland A, Wick P, Koehler A, Schmid K, Som C. Reviewing the environmental and human health knowledge base of carbon nanotubes. Environ Health Perspect 2007;115:1125–31. Hersam MC. Progress towards monodisperse single-walled carbon nanotubes. Nat Nanotechnol 2008;3:387–94. Hillegass JM, Shukla A, Lathrop SA, MacPherson MB, Fukagawa NK, Mossman BT. Assessing nanotoxicity in cells in vitro. Wiley Interdiscip Rev Nanomed Nanobiotechnol 2010;2:219–31. Hirano S, Kanno S, Furuyama A. Multi-walled carbon nanotubes injure the plasma membrane of macrophages. Toxicol Appl Pharmacol 2008;232:244–51. Holbrook RD, Murphy KE, Morrow JB, Cole KD. Trophic transfer of nanoparticles in a simplified invertebrate food web. Nat Nanotechnol 2008;3:352–5. Iijima S. Helical microtubules of graphitic carbon. Nature 1991;354:56–8. Im JS, Bai BC, Lee YS. The effect of carbon nanotubes on drug delivery in an electrosensitive transdermal drug delivery system. Biomaterials 2010;31:1414–9. Jennings AA, Hise S, Kiedrowski B, Krouse C. Urban battery litter. J Environ Eng-Asce 2009;135:46–57. Jia G, Wang HF, Yan L, Wang X, Pei RJ, Yan T, et al. Cytotoxicity of carbon nanomaterials: single-wall nanotube, multi-wall nanotube, and fullerene. Environ Sci Technol 2005;39:1378–83. Johnston HJ, Hutchison GR, Christensen FM, Peters S, Hankin S, Aschberger K, et al. A critical review of the biological mechanisms underlying the in vivo and in vitro toxicity of carbon nanotubes: the contribution of physico-chemical characteristics. Nanotoxicology 2010;4:207–46. Jones CF, Grainger DW. In vitro assessments of nanomaterial toxicity. Adv Drug Deliv Rev 2009;61:438–56. Kagan VE, Tyurina YY, Tyurin VA, Konduru NV, Potapovich AI, Osipov AN, et al. Direct and indirect effects of single walled carbon nanotubes on RAW 264.7 macrophages: role of iron. Toxicol Lett 2006;165:88-100. Kalbacova M, Kalbac M, Dunsch L, Hempel U. Influence of single-walled carbon nanotube films on metabolic activity and adherence of human osteoblasts. Carbon 2007;45:2266–72. Kane AB. Animal models of malignant mesothelioma. Inhal Toxicol 2006;18:1001–4. Kang S, Pinault M, Pfefferle LD, Elimelech M. Single-walled carbon nanotubes exhibit strong antimicrobial activity. Langmuir 2007;23:8670–3. Kang S, Herzberg M, Rodrigues DF, Elimelech M. Antibacterial effects of carbon nanotubes: size does matter. Langmuir 2008a;24:6409–13. Kang S, Mauter MS, Elimelech M. Physicochemical determinants of multiwalled carbon nanotube bacterial cytotoxicity. Environ Sci Technol 2008b;42:7528–34. Kang S, Mauter MS, Elimelech M. Microbial cytotoxicity of carbon-based nanomaterials: implications for river water and wastewater effluent. Environ Sci Technol 2009;43:2648–53. Karousis N, Tagmatarchis N, Tasis D. Current progress on the chemical modification of carbon nanotubes. Chem Rev 2010;110:5366–97. Khodakovskaya M, Dervishi E, Mahmood M, Xu Y, Li ZR, Watanabe F, et al. Carbon nanotubes are able to penetrate plant seed coat and dramatically affect seed germination and plant growth. ACS Nano 2009;3:3221–7. Klein J. Probing the interactions of proteins and nanoparticles. Proc Natl Acad Sci U S A 2007;104:2029–30. Kostarelos K, Bianco A, Prato M. Promises, facts and challenges for carbon nanotubes in imaging and therapeutics. Nat Nanotechnol 2009;4:627–33. Koyama S, Kim YA, Hayashi T, Takeuchi K, Fujii C, Kuroiwa N, et al. In vivo immunological toxicity in mice of carbon nanotubes with impurities. Carbon 2009;47: 1365–72. Lam CW, James JT, McCluskey R, Hunter RL. Pulmonary toxicity of single-wall carbon nanotubes in mice 7 and 90 days after intratracheal instillation. Toxicol Sci 2004;77:126–34. Lee J, Mahendra S, Alvarez PJJ. Nanomaterials in the construction industry: a review of their applications and environmental health and safety considerations. ACS Nano 2010;4:3580–90. Li JG, Li WX, Xu JY, Cai XQ, Liu RL, Li YJ, et al. Comparative study of pathological lesions induced by multiwalled carbon nanotubes in lungs of mice by intratracheal instillation and inhalation. Environ Toxicol 2007;22:415–21. Lin C, Fugetsu B, Su YB, Watari F. Studies on toxicity of multi-walled carbon nanotubes on Arabidopsis T87 suspension cells. J Hazard Mater 2009;170:578–83. Liu XY, Gurel V, Morris D, Murray DW, Zhitkovich A, Kane AB, et al. Bioavailability of nickel in single-wall carbon nanotubes. Adv Mater 2007;19. 2790-+. Liu Z, Fan AC, Rakhra K, Sherlock S, Goodwin A, Chen XY, et al. Supramolecular stacking of doxorubicin on carbon nanotubes for in vivo cancer therapy. Angew Chem Int Ed 2009a;48:7668–72. Liu Z, Tabakman S, Welsher K, Dai HJ. Carbon nanotubes in biology and medicine: in vitro and in vivo detection, imaging and drug delivery. Nano Res 2009b;2:85-120. Ma-Hock L, Treumann S, Strauss V, Brill S, Luizi F, Mertler M, et al. Inhalation toxicity of multiwall carbon nanotubes in rats exposed for 3 months. Toxicol Sci 2009;112: 468–81. Mercer RR, Hubbs AF, Scabilloni JF, Wang L, Battelli LA, Schwegler-Berry D, et al. Distribution and persistence of pleural penetrations by multi-walled carbon nanotubes. Part Fibre Toxicol 2010;7. Mitchell LA, Gao J, Wal RV, Gigliotti A, Burchiel SW, McDonald JD. Pulmonary and systemic immune response to inhaled multiwalled carbon nanotubes. Toxicol Sci 2007;100:203–14. Mitchell LA, Lauer FT, Burchiel SW, McDonald JD. Mechanisms for how inhaled multiwalled carbon nanotubes suppress systemic immune function in mice. Nat Nanotechnol 2009;4:451–6. Mu QX, Liu W, Xing YH, Zhou HY, Li ZW, Zhang Y, et al. Protein binding by functionalized multiwalled carbon nanotubes is governed by the surface chemistry of both parties and the nanotube diameter. J Phys Chem C 2008;112:3300–7. Muller J, Huaux F, Moreau N, Misson P, Heilier JF, Delos M, et al. Respiratory toxicity of multi-wall carbon nanotubes. Toxicol Appl Pharmacol 2005;207:221–31. Muller J, Huaux F, Fonseca A, Nagy JB, Moreau N, Delos M, et al. Structural defects play a major role in the acute lung toxicity of multiwall carbon nanotubes: toxicological aspects. Chem Res Toxicol 2008;21:1698–705. Murakami H, Nakashima N. Soluble carbon nanotubes and their applications. J Nanosci Nanotechnol 2006;6:16–27. Murbach DM, Devlin KD, Franke KS, Paustenbach DJ. A review of historical ambient airborne asbestos concentrations in cities and buildings: 1950s to the present day. Epidemiology 2008;19:S256–7. Murray AR, Kisin E, Leonard SS, Young SH, Kommineni C, Kagan VE, et al. Oxidative stress and inflammatory response in dermal toxicity of single-walled carbon nanotubes. Toxicology 2009;257:161–71. Mutlu GM, Budinger GRS, Green AA, Urich D, Soberanes S, Chiarella SE, et al. Biocompatible nanoscale dispersion of single-walled carbon nanotubes minimizes in vivo pulmonary toxicity. Nano Lett 2010;10:1664–70. Mylvaganam K. Carbon nanotubes build better protective body armor. Adv Mater Processes 2008;166:24. Narui Y, Ceres DM, Chen JY, Giapis KP, Collier CP. High aspect ratio silicon dioxidecoated single-walled carbon nanotube scanning probe nanoelectrodes. J Phys Chem C 2009;113:6815–20. Oberdorster G, Sharp Z, Atudorei V, Elder A, Gelein R, Kreyling W, et al. Translocation of inhaled ultrafine particles to the brain. Inhal Toxicol 2004;16:437–45. Oberdorster G, Stone V, Donaldson K. Toxicology of nanoparticles: a historical perspective. Nanotoxicology 2007;1:2-25. Oh WK, Yoon H, Jang J. Size control of magnetic carbon nanoparticles for drug delivery. Biomaterials 2010;31:1342–8. Oliveira RD, Santelli RE. Occurrence and chemical speciation analysis of organotin compounds in the environment: a review. Talanta 2010;82:9-24. Pauluhn J. Multi-walled carbon nanotubes (baytubes (R)): approach for derivation of occupational exposure limit. Regul Toxicol Pharmacol 2010;57:78–89. Poland CA, Duffin R, Kinloch I, Maynard A, Wallace WAH, Seaton A, et al. Carbon nanotubes introduced into the abdominal cavity of mice show asbestos-like pathogenicity in a pilot study. Nat Nanotechnol 2008;3:423–8. Qu GB, Bai YH, Zhang Y, Jia Q, Zhang WD, Yan B. The effect of multiwalled carbon nanotube agglomeration on their accumulation in and damage to organs in mice. Carbon 2009;47:2060–9. Ren HX, Chen X, Liu JH, Gu N, Huang XJ. Toxicity of single-walled carbon nanotube: how we were wrong? Mater Today 2010;13:6–8. Roberts AP, Mount AS, Seda B, Souther J, Qiao R, Lin S, et al. In vivo biomodification of lipid-coated carbon nanotubes by Daphnia magna. Environ Sci Technol 2007;41: 3025–9. Ryman-Rasmussen JP, Cesta MF, Brody AR, Shipley-Phillips JK, Everitt JI, Tewksbury EW, et al. Inhaled carbon nanotubes reach the subpleural tissue in mice. Nat Nanotechnol 2009;4:747–51. Schipper ML, Nakayama-Ratchford N, Davis CR, Kam NWS, Chu P, Liu Z, et al. A pilot toxicology study of single-walled carbon nanotubes in a small sample of mice. Nat Nanotechnol 2008;3:216–21. Service RF. American chemical society meeting: nanomaterials show signs of toxicity. Science 2003;300:243. Sheehan D. The potential of proteomics for providing new insights into environmental impacts on human health. Rev Environ Health 2007;22:175–94. Sheng XL, Liu NR, Zhai J, An LP. Application of one-dimensional nanomaterials in dyesensitized solar cells. Prog Chem 2009;21:1969–79. Shvedova AA, Kisin ER, Mercer R, Murray AR, Johnson VJ, Potapovich AI, et al. Unusual inflammatory and fibrogenic pulmonary responses to single-walled carbon nanotubes in mice. Am J Physiol Lung Cell Mol Physiol 2005;289:L698–708. Shvedova AA, Kisin E, Murray AR, Johnson VJ, Gorelik O, Arepalli S, et al. Inhalation vs. aspiration of single-walled carbon nanotubes in C57BL/6 mice: inflammation, fibrosis, oxidative stress, and mutagenesis. Am. J. Physiol. Lung Cell. Mol. Physiol. 2008;295:L552–65. Simon-Deckers A, Gouget B, Mayne-L'Hermite M, Herlin-Boime N, Reynaud C, Carriere M. In vitro investigation of oxide nanoparticle and carbon nanotube toxicity and intracellular accumulation in A549 human pneumocytes. Toxicology 2008;253:137–46. Singh R, Pantarotto D, McCarthy D, Chaloin O, Hoebeke J, Partidos CD, et al. Binding and condensation of plasmid DNA onto functionalized carbon nanotubes: toward the Author's personal copy X. Zhao, R. Liu / Environment International 40 (2012) 244–256 construction of nanotube-based gene delivery vectors. J Am Chem Soc 2005;127: 4388–96. Stevens RM. New carbon nanotube AFM probe technology. Mater Today 2009;12:42–5. Suga H, Naitoh Y, Tanaka M, Horikawa M, Kobori H, Shimizu T. Nanomanipulation of single nanoparticle using a carbon nanotube probe in a scanning electron microscope. Appl Phys Express 2009;2. Tan X, Lin C, Fugetsu B. Studies on toxicity of multi-walled carbon nanotubes on suspension rice cells. Carbon 2009;47:3479–87. Tantra R, Cumpson P. The detection of airborne carbon nanotubes in relation to toxicology and workplace safety. Nanotoxicology 2007;1:251–65. Teeguarden JG, Hinderliter PM, Orr G, Thrall BD, Pounds JG. Particokinetics in vitro: dosimetry considerations for in vitro nanoparticle toxicity assessments. Toxicol Sci 2007;95:300–12. Thurnherr T, Su DS, Diener L, Weinberg G, Manser P, Pfander N, et al. Comprehensive evaluation of in vitro toxicity of three large-scale produced carbon nanotubes on human Jurkat T cells and a comparison to crocidolite asbestos. Nanotoxicology 2009;3:319–38. Tran PA, Zhang LJ, Webster TJ. Carbon nanofibers and carbon nanotubes in regenerative medicine. Adv Drug Deliv Rev 2009;61:1097–114. Tsai TY, Lee CY, Tai NH, Tuan WH. Transfer of patterned vertically aligned carbon nanotubes onto plastic substrates for flexible electronics and field emission devices. Appl Phys Lett 2009;95. Tsoli M, Kuhn H, Brandau W, Esche H, Schmid G. Cellular uptake and toxicity of Au55 clusters. Small 2005;1:841–4. Turner A. Marine pollution from antifouling paint particles. Mar Pollut Bull 2010;60: 159–71. Tutak W, Park KH, Vasilov A, Starovoytov V, Fanchini G, Cai SQ, et al. Toxicity induced enhanced extracellular matrix production in osteoblastic cells cultured on singlewalled carbon nanotube networks. Nanotechnology 2009;20. Warheit DB, Laurence BR, Reed KL, Roach DH, Reynolds GAM, Webb TR. Comparative pulmonary toxicity assessment of single-wall carbon nanotubes in rats. Toxicol Sci 2004;77:117–25. 255 Warheit DB, Hoke RA, Finlay C, Donner EM, Reed KL, Sayes CM. Development of a base set of toxicity tests using ultrafine TiO2 particles as a component of nanoparticle risk management. Toxicol Lett 2007;171:99-110. Wick P, Manser P, Limbach LK, Dettlaff-Weglikowska U, Krumeich F, Roth S, et al. The degree and kind of agglomeration affect carbon nanotube cytotoxicity. Toxicol Lett 2007;168:121–31. Wijaya IPM, Gandhi S, Nie TJ, Wangoo N, Rodriguez I, Shekhawat G, et al. Protein/carbon nanotubes interaction: the effect of carboxylic groups on conformational and conductance changes. Appl Phys Lett 2009;95. Wildgoose GG, Abiman P, Compton RG. Characterising chemical functionality on carbon surfaces. J Mater Chem 2009;19:4875–86. Yi CQ, Fong CC, Zhang Q, Lee ST, Yang MS. The structure and function of ribonuclease A upon interacting with carbon nanotubes. Nanotechnology 2008;19. Zhang M, Li J. Carbon nanotube in different shapes. Mater Today 2009;12:12–8. Zhang T, Stilwell JL, Gerion D, Ding L, Elboudwarej O, Cooke PA, et al. Cellular effect of high doses of silica-coated quantum dot profiled with high throughput gene expression analysis and high content cellomics measurements. Nano Lett 2006;6:800–8. Zhang T, Xu M, He L, Xi K, Gu M, Jiang ZS. Synthesis, characterization and cytotoxicity of phosphoryl choline-grafted water-soluble carbon nanotubes. Carbon 2008;46:1782–91. Zhang B, Xing YH, Li ZW, Zhou HY, Mu QX, Yan B. Functionalized carbon nanotubes specifically bind to alpha-chymotrypsin's catalytic site and regulate its enzymatic function. Nano Lett 2009;9:2280–4. Zhao YL, Stoddart JF. Noncovalent functionalization of single-walled carbon nanotubes. Acc Chem Res 2009;42:1161–71. Zhao XC, Liu RT, Chi ZX, Teng Y, Qin PF. New insights into the behavior of bovine serum albumin adsorbed onto carbon nanotubes: comprehensive spectroscopic studies. J Phys Chem B 2010;114:5625–31. Zhu Y, Li WX, Li QN, Li YG, Li YF, Zhang XY, et al. Effects of serum proteins on intracellular uptake and cytotoxicity of carbon nanoparticles. Carbon 2009;47:1351–8. Zuo GH, Huang Q, Wei GH, Zhou RH, Fang HP. Plugging into proteins: poisoning protein function by a hydrophobic nanoparticle. ACS Nano 2010;4:7508–14.