Biological Interactions of Functionalized Single-Wall Carbon Nanotubes in Human Epidermal Keratinocytes

Abstract

Carbon nanotube–based nanovectors, especially functionalized nanotubes, have shown potential for therapeutic drug delivery. 6-Aminohexanoic acid–derivatized single-wall carbon nanotubes (AHA-SWNTs) are soluble in aqueous stock solutions over a wide range of physiologically relevant conditions; however, their interactions with cells and their biological compatibility has not been explored. Human epidermal keratinocytes (HEKs) were dosed with AHA-SWNTs ranging in concentration from 0.00000005 to 0.05 mg/ml. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) cell viability decreased significantly (p < .05) from 0.00005 to 0.05 mg/ml after 24 h. The proinflammatory mediators of inflammation cytokines interleukin (IL)-6, IL-8, tumor necrosis factor (TNF)-α, IL-10, and IL-1β were also assessed. Cytokine analysis did not show a significant increase in IL-6 and IL-8 in the medium containing 0.000005 mg/ml of AHA-SWNTs from 1 to 48 h. IL-6 increased in cells treated with 0.05 mg/ml of AHA-SWNTs from 1 to 48 h, whereas IL-8 showed a significant increase at 24 and 48 h. No significant difference (p < .05) was noted with TNF-α, IL-10, and IL-1β expression at any time point. Transmission electron microscopy of HEKs treated with 0.05 mg/ml AHA-SWNTs for 24 h depicted AHA-SWNTs localized within intracytoplasmic vacuoles in HEKs. Treatment with the surfactant 1% Pluronic F127 caused dispersion of the AHA-SWNT aggregates in the culture medium and less toxicity. These data showed that the lower concentration of 0.000005 mg/ml of AHA-SWNTs maintains cell viability and induces a mild cytotoxicity, but 0.05 mg/ml of AHA-SWNTs demonstrated an irritation response by the increase in IL-8.

Carbon nanotubes are a unique class of nanomaterials due to their physical, chemical, thermal, and optical properties (Endo et al. 2004). Recently, there has been an interest in exploring their novel properties for biological applications. Carbon nanotubes can serve as translocators for therapeutic molecules in drug delivery (Bianco et al. 2005). Single-wall carbon nanotubes (SWNTs) have also been used in medical imaging (Cherukuri et al. 2004). Selective cell destruction has been achieved by functionalizing SWNTs with a folate moiety thereby, causing selective internalization of the SWNTs into cells labeled with a folate receptor tumor marker. After near-infrared (NIR) radiation, extensive HeLa cell death was observed (Kam et al. 2005).

SWNTs have a diameter of 1 nm and the length can range from several hundred nanometers to more than 1 μm, making them ideal for intracellular interactions. Unfortunately, pristine SWNTs cannot readily disperse or dissolve in water, and thus must be solubilized through either wrapping (surfacting) the SWNTs, end-group functionalization, or sidewall functionalization. Sidewall functionalization has also been used to promote their action as nanovectors for the delivery of therapeutics (Klumpp et al. 2006). An additional advantage of functionalization of carbon-based nanomaterials is their reduction in toxicity. For example, when hydroxyl groups are added to the surface of fullerenes, they are less cytotoxic (Sayes et al. 2004). Recently, SWNTs functionalized with an additional carbon/-phenyl-SO₃H has been shown to be less cytotoxic than nonfunctionalized SWNTs in human dermal fibroblasts (Sayes et al. 2006).

The poor water solubility of many functionalized SWNTs in aqueous media restricts their development to be considered as nanovectors in a biocompatible system. If SWNTs are to be used as vectors for drug delivery, they must be soluble so that they can have the ability to penetrate cell membranes and distribute to specific target cells. Recently, we have reported that 6-aminohexanoic acid–functionalized SWNTs (AHA-SWNTs; Figure 1) can be synthesized to be aqueous soluble at 0.5 mg/ml across the pH ranges of 4.0 to 11.0 (Zeng et al. 2005).

SWNTs can enter the body by inhalational or dermal routes. Studies have focused on the inhalational toxicity of SWNTs in mice (Lam et al. 2004; Shvedova et al. 2005) and in rats (Warheit et al. 2004) after intratracheal instillation. Studies have also been conducted with SWNTs in immortalized HaCaT cells, which depicted an increase in oxidative stress, a decrease in glutathione, and a depletion of vitamin E (Shvedova et al. 2003). There are several reports on cytokine release when cells were dosed with nanovectors. Malonic acid–C₆₀ derivatives have been shown to increase the release of interleukin (IL)-1β in activated microglial cell cultures (Tzeng et al. 2002). Our data have shown that human epidermal keratinocytes (HEKs) exposed to multi-walled carbon nanotubes (MWCNTs) led to a proinflammatory response depicting IL-8 release in a dose- and time-dependent manner (Monteiro-Riviere et al. 2005a). Other studies using carbon-based nanomaterials such as fullerene-based amino acids in HEKs also depicted a similar increase in IL-6, IL-8, and IL- 1β (Rouse et al. 2006). In pharyngeal aspiration studies utilizing SWNTs, tumor necrosis factor (TNF)-α and IL-1β was induced followed by transforming growth factor (TGF)-β 1 (Shvedova et al. 2005).

Several carbon-based nanomaterials, such as different types of carbon black, fullerenes, and SWNTs, have been shown to adsorb dyes and proteins (Monteiro-Riviere and Inman 2006; Marcorin et al. 2000; Meng et al. 2005). It is possible that derivatized SWNTs may also adsorb proteins and dyes that are used for assessing viability and cytokine activity in toxicology studies.

As a consequence of strong van der Waal interactions, nanotubes readily aggregate in aqueous culture medium. Previously, we have reported that several nonionic surfactants may be used to disperse MWCNTs without being cytotoxic to HEKs. A 1% Pluronic F127 surfactant did not effect HEK viability, whereas MWCNTs dosed alone caused a significant decrease in viability and an increase in IL-8 expression (Monteiro-Riviere et al. 2005b). The effects of different vehicles such as DMSO and an anionic surfactant such as Pluronic F127 was also investigated in 6-aminohexanoic acid–derivatized SWNTs (AHA-SWNTs).

By creating a water-soluble SWNTs, SWNTs become more biocompatible for physiological systems; however, despite the broad potential application in drug delivery, the viability and toxicity of these functionalized SWNTs has not been demonstrated. Furthermore, it has been generally assumed that the solution properties of a particular functionalized SWNTs in aqueous media (i.e., individual tubes versus bundles) would be indicative of their properties within a cell. The purpose of this study was to evaluate the effects of the functionalized AHA-SWNTs in HEKs, to assess cell viability and cytokine activity, and to determine the biological effects of the cells with different vehicles such as DMSO and Pluronic F127.

MATERIALS AND METHODS

Synthesis of AHA-SWNTs

High-pressure carbon monoxide (HiPco) SWNTs were purified to remove iron and other impurities (Chiang et al. 2001). They were fluorinated to a C/F ratio of approximately 2.4:1 by direct fluorination at 150°C by a previously reported procedure (Mickelson et al. 1998). The fluoronanotubes were functionalized by the reaction with 6-aminohexanoic acid (AHA) in the presence of pyridine in dimethylformamide (DMF). Fluoronanotubes (ca. 30 mg) were sonicated in 100 ml of DMF for 10 min, resulting in complete dispersion to form a dark solution. Then 600 mg of 6-aminohexanoic acid (600 mg) was well dispersed in 200 ml DMF and 4 to 5 drops of pyridine (catalyst) was added. The reaction mixture was stirred under N₂ for 5 days at 90°C, and then filtered through a 0.2-μm Cole Palmer Teflon membrane, and washed with water and acetone to ensure complete removal of the unreacted 6-aminohexanoic acid, reaction byproducts, and the solvent. The AHA-SWNTs were washed with nanopure water (Zeng et al. 2005). It is difficult to estimate the molecular weight of a single AHA-SWNT due to its variable length and width.

Cell Culture and Viability Assay

Neonatal HEKs (Cambrex, Walkersville, MO) were plated in 96-well culture plates (0.32 cm² growth area) at approximately 7000 cells per well and grown in a humidified environment of 5% CO₂. Upon reaching 70% confluency, the cells were exposed to AHA-SWNTs in KGM-2, as well as KGM-2 alone. Immediately prior to treating the cells, a 0.5 mg/ml stock solution of AHA-SWNTs was mixed with KGM-2 to provide serial dilutions of 0.05 to 0.00000005 mg/ml of nanotubes (log interval). In another experiment, 1% Pluronic F127 (BASF, New Milford, CT) or 1% DMSO was added to the aqueous culture medium containing 0.000005 mg/ml (the highest concentration that maintained viability at 24 h) and 0.05 mg/ml of AHA-SWNTs. The medium was harvested at 1, 2, 4, 8, 12, 24, and 48 h (8 wells/plate/time point), and stored at −80°C. HEKs at the 24- and 48-h wells were assayed using MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) for viability as described previously (Mosmann 1983). Because the residual nanotubes can affect the absorbance values, the solution in each well was pipetted into new 96-well plates (Monteiro-Riviere and Inman 2006). The absorbance, directly proportional to cell viability, was determined spectrophotometrically at 550 nm in a plate reader (Multiskan RC; Labsystems, Helsinki, Finland). The results from this study determined the range of AHA-SWNT concentrations that would not be toxic to the HEKs.

Cytokine Assay

Human cytokines IL-6, IL-8, IL-10, TNF-α, and IL-1β were quantitated with the Bio-Plex suspension array system (Bio-Rad Laboratories, Hercules, CA). This system utilizes multiplexing to simultaneously assay for cytokines. Customized beads (5.6 μm diameter) conjugated to a capture antibody that is specific to each cytokine and possessing a unique spectral address were mixed with culture media and incubated in a 96-well filter plate. The beads were rinsed and then incubated with a fluorescent-labeled reporter molecule that specifically binds to the analyte. The contents of each well were analyzed in the Bio-Plex array reader. As the beads flow into the reader, one laser identifies the spectral address of each cytokine and the other laser excites the reporter molecule to quantify the specific cytokine relative to the standard curve. The average concentration (pg/ml) of each cytokine for each treatment and time point were calculated in order to determine the time dependency of the toxicity on HEKs.

Cytokine Adsorption by AHA-SWNTs

Cytokine adsorption studies for AHA-SWNTs were conducted with Bio-Plex standard cytokine stock solution. AHA-SWNTs of the final concentration of 0.05, 0.005, and 0.0005 mg/ml were incubated with the cytokine standard (800 pg/ml). The medium was incubated in 96-well plates for 24 h, with the cytokine concentration assayed by the Bio-Plex suspension array system.

Transmission Electron Microscopy of AHA-SWNTs in HEKs

Transmission electron microscopy (TEM) was conducted on HEKs to determine the location of AHA-SWNTs. HEKs were seeded in cell culture flasks (25 cm²; ∼100,000 cells) and grown to 70% confluency at 37°C in a humidified 5% CO₂ environment. HEKs were treated with AHA-SWNTs at a concentration of 0.000005 mg/ml (at the highest concentration that maintains viability), and at 0.05 mg/ml (the highest dose). The HEKs were treated with AHA-SWNTs for 24 h, rinsed in Hanks’ buffered salt solution (HBSS), and fixed in Trump’s fixative at 4°C. Cells were then rinsed in 0.1 M phosphate buffer (pH 7.2), pelleted in a microcentrifuge tube with 3000 rpm for 30 s, and embedded in 3% molten agar. The cells were postfixed in 1% osmium tetroxide (Polysciences, Warrington, PA) in 0.1 M sodium phosphate buffer (pH 7.2), dehydrated through graded ethanol solutions, cleared in acetone, and infiltrated and embedded in Spurr’s resin. Unstained thin sections (approximately 800 to 1000 μm) were mounted on copper grids and then examined on a Philips EM208S transmission electron microscope.

Statistical Analysis

The mean values for HEK viability and cytokine concentration (normalized to viability) for each treatment were calculated and the significant differences (p < .05) determined using the least significance differences in the analysis of variance procedure of SAS (SAS 9.1 for Windows; SAS Institute, Cary, NC). Multiple comparisons among different treatments were conducted within each exposure and sampling time using the student’s t-test at p < .05.

RESULTS

HEK viability assessed by MTT in treatments ranging from 0.00000005 to 0.05 mg/ml of AHA-SWNTs showed a significant decrease (p < .05) that was dose dependent from 0.00005 to 0.05 mg/ml at 24 h (Figure 2). AHA-SWNTs at 0.000005 mg/ml was the highest concentration that did not show a significant decrease in viability at 24 h. At 0.0000005 mg/ml AHA-SWNTs, viability continued to decrease significantly at 48 h (Figure 2). Elevated IL-8 expression (Figure 3) significantly increased at AHA-SWNT concentration of 0.0000005 to 0.005 mg/ml by 24 h and AHA-SWNT concentration of 0.00005 to 0.005 mg/ml by 48 h. IL-8 expression at AHA-SWNT concentration of 0.05 mg/ml was the exception, showing a decrease at 24 h but remained constant from AHA-SWNT concentration of 0.005 to 0.05 mg/ml at 48 h.

The expression of cytokines IL-6, IL-8, IL-10, TNF-α, and IL-1β was assessed at AHA-SWNT concentration of 0.000005 mg/ml (the highest concentration that did not decrease the viability by 24 h) and at 0.05 mg/ml (the highest concentration dosed in the medium) at several time points. IL-6 expression increased from 4 to 12 h at both concentrations, but was not significant compared to controls. At 24 h, IL-6 slightly decreased and then increased by 48 h at 0.05 mg/ml AHA-SWNTs compared to controls (Figure 4). IL-8 expression (Figure 5) showed a time-dependent increase through 12 h at both AHA-SWNT concentrations and its level was significantly greater than control at 0.05 mg/ml AHA-SWNTs at 24 and 48 h. IL-10, TNF-α, and IL-1β were not detected at any time point.

At 24 h, TEM of HEKs treated with 0.000005 mg/ml AHA-SWNTs showed minimal uptake of AHA-SWNTs (Figure 6). In contrast, TEM of HEKs dosed with the highest AHA-SWNT concentration, 0.05 mg/ml, depicted several cells containing large aggregates of the AHA-SWNTs within intracytoplasmic vacuoles (Figure 7). At 0.05 mg/ml, at 24 h, AHA-SWNTs can readily aggregate within cytoplasmic vacuoles (Figure 8a). It is of interest that not only were aggregated AHA-SWNTs noted, but structures smaller than 10 nm were also present (Figure 8b).

When AHA-SWNTs were incubated in KGM-2 medium containing a cytokine standard without cells, the AHA-SWNTs were capable of adsorbing cytokines. After 24 h, IL-6, IL-8, and IL-10 were adsorbed at 0.0005, 0.05, and 0.05 mg/ml AHA-SWNTs. At the 0.05 mg/ml concentration, AHA-SWNTs are capable of adsorbing 52% of IL-6, 38% of IL-8, and 63% of IL-10. The adsorption for TNF-α by 0.05 mg/ml AHA-SWNTs was at 35%, whereas IL-1β showed no significant adsorption compared to the controls (Figure 9).

In the above experiments, AHA-SWNTs readily aggregated in the KGM-2 medium, although they were soluble at the 0.5 mg/ml concentration. DMSO and Pluronic F127 vehicles were used to test the dispersion ability and penetration of the aggregated AHA-SWNTs. AHA-SWNTs had formed large aggregates within the KGM-2 medium at 24 h (Figure 10a). In contrast, 1% Pluronic F127 dispersed the nanotubes within the medium, forming much smaller aggregates (Figure 10b). The 1% DMSO solution still showed large aggregates and did not disperse the AHA-SWNTs as much as the 1% Pluronic F127 treatment (Figure 10c).

Viability was also assessed with HEKs exposed to 1% Pluronic F127, which showed a significant decrease compared to the controls. AHA-SWNTs alone significantly reduced viability compared to the controls. Pluronic F127 + AHA-SWNTs did not decrease viability compared to the AHA-SWNT treatment alone. One percent DMSO did not affect viability compared to the controls, and the viability was not different between the AHA-SWNT + 1% DMSO treatment and AHA-SWNT–alone treatment (Figure 11).

We further investigated the effect of the surfactant Pluronic F127 on AHA-SWNT–induced cytokine release from 1 to 48 h. IL-6 expression was significantly greater than controls at 4, 8, 12, and 48 h treated with 0.05 mg/ml of AHA-SWNTs alone. The 0.05 mg/ml AHA-SWNT + 1% Pluronic F127 treatment caused a significant decrease in IL-6 expression compared to the AHA-SWNT–alone treatment from 4 to 48 h (Figure 12). Normalized IL-8 release by HEKs exposed to 0.05 mg/ml of AHA-SWNTs showed an increase compared to controls by 24 and 48 h. The 1% Pluronic F127 + AHA-SWNT treatment caused a decrease in IL-8 expression compared to the AHA-SWNT–alone treatment at 24 and 48 h (Figure 13). The release of TNF-α was greater than control at 1, 4, 12, and 48 h with AHA-SWNTs alone. The AHA-SWNT + 1% Pluronic F127 treatment markedly caused a decrease in TNF-α expression compared to the AHA-SWNT–alone treatment (Figure 14). The IL-1β data showed no difference between the AHA-SWNT–alone treatment and the combined AHA-SWNT + 1% Pluronic F127 treatmnet at any time point (data not shown).

DISCUSSION

Most cytotoxicity studies of SWNTs have focused on the respiratory responses. Pulmonary toxicity of SWNTs was studied in mice (Lam et al. 2004) and in rats (Warheit et al. 2004), which showed an induction of epithelioid granulomas that was similar to a foreign body tissue reaction. Pharyngeal aspiration of SWNTs in C57BL/6 mice led to neutrophil accumulation accompanied by an early elevation of the proinflammatory cytokines such as TNF-α and IL-1β (Shvedova et al. 2005). Research has also been conducted on SWNTs in immortalized HEKs but not on normal HEKs (Shvedova et al. 2003). Our study has shown that AHA-SWNTs did not show cytotoxicity until the 0.00005 mg/ml concentration, which led to a decrease in viability at 24 h compared to controls. The increase in IL-6 and IL-8 expression indicated that AHA-SWNTs could initiate an early inflammatory response. IL-6 is a classical proinflammatory cytokine that is secreted by keratinocytes when cells are exposed to irritants (Sugawara et al. 2001). Keratinocytes have also been shown to release TNF-α prior to IL-8 and IL-1β as an activator of keratinocyte injury (Allen et al. 2000; Freedberg et al. 2001). In this study, IL-1β release was not detected. At 0.05 mg/ml AHA-SWNTs induced the release of TNF-α at the 12-, 24-, and 48-h time points that were above the detection limit (data not shown). TNF-α probably was expressed earlier and is a potential initiator that induced the proinflammatory response, resulting in an IL-6 and IL-8 release.

Immortalized HaCaT cells exposed to nonfunctionalized SWNTs showed oxidative stress, a decrease in glutathione levels, and a depletion of vitamin E. Higher concentrations of the SWNTs caused ultrastructural changes, including altered cytoplasmic organelles (Shvedova et al. 2003). However, this study was conducted in immortalized cells and did not evaluate for inflammation. Recently, it has been shown that SWNTs can activate nuclear factor (NF)-κB in a dose-dependent manner in HEKs (Manna et al. 2005) and that NF-κB is capable of activating IL-6 expression (Xiao et al. 2004) in autocrine human myeloma cells. As a result, it is possible that AHA-SWNTs induced IL-6 expression in part from the activation NF-κB initiated by TNF-α. IL-1β also induces IL-8 in bronchial cells via NF-κB (Edwards et al. 2005). AHA-SWNT–induced IL-8 secretion may be initiated from the activation of IL-1β, although it was expressed slightly and under the detection limit or occurred rapidly and then ceased.

The synthesis of carbon-based materials requires a catalyst, such as Fe and Ni. Nonpurified iron-rich SWNTs converted superoxide radicals into hydroxyl radicals more effectively than purified SWNTs, and caused the loss of glutathione (GSH) in cells and the accumulation of lipid hydroperoxides (Kagan et al. 2006). It is necessary to use pure SWNTs with little or no residual catalysts. In this study, the inflammatory response was probably not due to the catalysts, because AHA-SWNTs were purified by the fluorination and subsequent functionalization process, and the purity was greater than 99.8% (Chiang et al. 2001).

Apoptosis occurred in T lymphocytes showing annexin V–positive cells following treatments with MWCNTs (Bottini et al. 2006). Our MWCNTs studied in HEKs did not show any apoptotic cells by TEM (Monteiro-Riviere et al. 2005a). This difference could be due to the method of preparation of the nanotubes or that the T lymphocytes exposed to the nanotubes may be more sensitive to apoptosis than HEKs. In the current study, no apoptotic HEKs were found by TEM following treatment with derivatized single-wall nanotubes.

Intratracheal exposure of SWNTs at high concentrations caused death and granulomas in lungs with agglomerates of 10 to 100 nanotubes; the aggregation properties may be due to its electrostatic character (Lam et al. 2004; Warheit et al. 2004). TEM also revealed MWCNTs in the intracytoplasmic vacuoles of HEKs (Monteiro-Riviere et al. 2005a). Our present study not only depicted the presence of aggregated AHA-SWNTs, but less aggregation of the AHA-SWNTs within the intracytoplasmic vacuoles of HEKs. A study with hat-stacked carbon nanofibers (H-CNFs) implanted in the subcutaneous tissue of rats also showed a decrease in H-CNF aggregation, but after 4 weeks, and appeared shorter in size over time (Yokoyama et al. 2005). Our study showed higher concentrations of AHA-SWNT aggregates in HEKs, which is different from the H-CNFs, possibly due to our time differences and the structure of the nanotubes. The structural difference in H-CNFs and SWNTs may be explained by the fact that the edges of the stacked graphene layers are on the H-CNF surface that provides additional sites for hydrophilic groups to attach. This arrangement could provide water-soluble SWNTs by functionalization that can mimic the H-CNF structure.

In addition, different types of carbon black have been shown to adsorb dyes such as neutral red and MTT, which can also interfere with the absorption spectra. Carbon black may also adsorb constituents of grow media as well as proteins (Monteiro-Riviere and Inman 2006). Adsorption of albumin and fibrinogen by SWNT membranes may also exist (Meng et al. 2005). Derivatized fullerenes are capable of associating with the protein of human immnunodeficiency virus (HIV) protease (Marcorin et al. 2000). In our viability study, IL-8 in the medium was assessed with log-interval concentrations of AHA-SWNTs (Figure 3). IL-8 release was greater with an increase in AHA-SWNT concentrations, but decreased at 0.05 mg/ml of AHA-SWNTs as compared to the 0.005 mg/ml concentration. At 0.05 mg/ml concentration, AHA-SWNTs alone are capable of adsorbing cytokines, especially IL-6, IL-8, and IL-10 (Figure 9). As a result, the cytokine concentrations that were observed may be even greater due to this adsorption, and therefore the amount of cytokine release for AHA-SWNTs may be underestimated.

DMSO, a stratum corneum permeation enhancer, did not effectively disperse the aggregates (Figure 10c) and did not significantly influence the AHA-SWNT viability, (Figure 11). Nonionic surfactants such as Pluronic F127 are relatively mild and nonirritating. It has been shown that 1% Pluronic F127 was nontoxic to HEKs in culture and did not increase MWCNT cytotoxicity (Monteiro-Riviere et al. 2005b). These studies were conducted by assessing viability with the neutral red assay. In our current studies, the MTT assay showed a decrease in viability to some extent by Pluronic F127. The viability difference may be explained by the method used to assess viability, different stages of the cell cycle, and differences in cell lots from the manufacturer. However, 1% Pluronic F127 dispersed the AHA-SWNT aggregates into finer particles that was similar to the MWCNT study and did not cause a decrease in viability. At 0.05 mg/ml AHA-SWNTs decreased HEK viability by 24 h, whereas the addition of 1% Pluronic F127 addition was capable of shielding AHS-SWNT cytotoxicity. Compared to the AHA-SWNT–alone treatment, the cytokine response decreased when Pluronic F127 was added to the culture medium containing the AHA-SWNTs. These studies have demonstrated that 1% Pluronic F127 can decrease the cytotoxicity of AHA-SWNTs. Our previous data also showed less irritation assessed by IL-8 release when MWCNTs were dosed with this surfactant (Monteiro-Riviere et al. 2005b). This may be explained by the nonionic surfactant causes a suspension of AHA-SWNTs by coating the surface and by forming micelles in the medium (Matthew et al. 2002; Moore et al. 2003; Monteiro-Riviere et al. 2005b). Also, it is possible that this micelle formation altered the AHA-SWNT surface properties. Pluronic F127 may be capable of altering the inflammatory response by shielding some of the cell membrane receptors and thereby decreasing the AHA-SWNT adsorption to the cell membrane, leading to less cytokine release.

Previously, we have reported that the solubility of AHA-SWNTs is dependent on the ionic strength of the solution (Zeng et al. 2005). Addition of salts to an aqueous solution at a pH 7.0 can result in aggregation of the SWNTs and their precipitation from the solution. A solution of AHA-SWNTs (0.5 mg/ml) incubated at 37°C showed aggregation after 48 h and after 4 days the nanotubes precipitated out. Aggregation can also occur when the AHA-SWNT stock solution is kept at room temperature for 2 months. Aggregation of AHA-SWNTs was also seen 24 h after nanotubes were introduced into the medium. We propose that simple aqueous solubility is an insufficient parameter by which to guide the potential application of functionalized AHA-SWNTs for biological applications.

In conclusion, our data show that AHA-SWNTs are capable of localizing within intracytoplasmic vacuoles of HEKs. In addition, derivatized AHA-SWNTs can initiate an early inflammatory response, as demonstrated by an increase in IL-6 and IL-8. Also, the increase in cytokines may even be greater due to the fact that adsorption of the AHA-SWNTs can occur. The toxicity of SWNTs, as depicted by a decrease in viability, can be dependent on several physiochemical factors, such as their physical shape, diameter, length, and functionalization. Our studies investigated the dispersion ability of Pluronic F127 on AHA-SWNT aggregation, showing that it was capable of dispersing AHA-SWNTSs into smaller particles.

It should be noted that AHA-SWNTs are anionic and the effects of the surface charge in controlling aggregation are important. We propose that in order to understand the interplay between functionalized SWNTs and their interactions in cells, a systematic study of functionalized SWNTs with different charges and aggregation properties needs to be conducted.