The effect of nitrogen-doped multiwall carbon nanotubes (CNx) on the proliferation of NIH-3T3 murine fibroblasts is presented. CNTs were dispersed in distillated water and incubated with mammalian cells in order to evaluate their toxicity. Also, the influence of factors such as dosage (7 and 70 µg/mL), exposure time (24 to 96 h), and the exposure route (before and after cell liftoff) on the cell proliferation was evaluated. When the CNx were simultaneously incubated with the cells, the control culture reached a maximum cell concentration of 1.3 × 105  ±  3.4 × 104 cells per well at 96 h, whereas cultures with 7 µg/mL reached a concentration of 2.6 × 104  ±  5.3 × 103 cells. In the case of 70 µg/mL of CNx most of the cells were dead. The CNx that were added 24 h after cell dissociation showed that live cells decreased, with a cell concentration of 9.6 × 104  ±  9 × 103 for 7 µg/mL and 5.5 × 104  ±  9.5 × 103 for 70 µg/mL, in contrast to control cultures with 1.1 × 106  ±  1.5 × 104. The results showed that the CNx had cytotoxic effects depending on the concentration and exposure route.

1. Introduction

A wide range of nanomaterials has been developed for several applications over the past few years. Due to their physical, chemical, electrical, and thermal properties, and since their discovery in 1991 [1], carbon nanotubes (CNTs) have shown a potential for use in drug delivery, biosensor, antimicrobial nanocomposite film, and cellular scaffolding. CNTs are tiny hollow cylinders, made from a single, double, or several layers of graphene that are concentrically arranged and capped by fullerene hemispheres. They have diameters ranging from 0.4 to 2 nm for single walled carbon nanotubes (SWCNTs) and from 2 to 200 nm for multiwall carbon nanotubes (MWCNTs), and lengths ranging from hundreds of nanometers to micrometers [13]. Since CNTs have an asbestos-like shape, research into their toxicity and potential risks to human health has been intensified [47].

Studies on cellular response in nonfunctionalized or functionalized (addition of functional groups on a graphite surface) MWCNT have been extensive. Chemical doping (carbon atoms substitution) with nitrogen of CNTs (CNx) was suggested to have positive effects on mice survival [8] and showed an improvement in cell-adhesion strength, viability, and proliferation of mammalian cells [3, 9], in contrast with the MWCNT. However, cytotoxic effects of CNx have also been reported, where long length CNx were more toxic than other functionalized CNTs [10]. Researches have demonstrated that cells exhibited variable responses to CNTs depending on different factors such as the method of synthesis, impurities, length and diameter, type (pristine, functionalized, and doped), degree of dispersion/agglomeration, dispersant, CNT concentration, time exposure, cellular type, and protein adsorption [2, 5, 6, 8, 11, 12]. Due to the inconsistency in CNx biocompatibility, more studies regarding cell response to these nanomaterials are necessary.

In the body, cell motility and wound healing are carried out by cell detachment, which is generated by proteolytic processes using endogenous proteases [13]. One of the most common enzymatic methods used for cell detachment in adherent-cell-subculture is trypsinization; trypsin cuts adhesion proteins to yield disaggregated cells with a rounded appearance. Although many cells are able to tolerate trypsin digestion during a short period of time, trypsinization causes cell stress affecting cytoskeleton proteins that are involved in regulating cell adhesion, stability, and elasticity [1416].

CNTs have the capacity to adsorb a wide range of proteins, especially those rich in histidine, tryptophan, and phenylalanine [12], and also adhesion proteins from extracellular matrix (fibronectins, collagen) and transmembrane-proteins (integrins) [17]. Since enzymatic cell detachment can produce residual fragments of adhesion proteins, these fragments could interact with CNTs altering the extracellular matrix metabolism which is regulated by a complex mechanism including cell-cell and cell-matrix interactions [13, 17]. For this reason, the knowledge of cell-CNTs interactions is essential for cell scaffold development that is used in tissue regeneration.

The aim of this study was to evaluate the possible toxic effect of CNx on NIH-3T3 murine fibroblast stressed by enzymatic detachment and nonstressed cells, in which a natural cell detachment stress was simulated by a trypsin incubation during a short period of time. Exposure route was defined in this work as the way to add nanomaterials to cell cultures (stressed and nonstressed cells). Besides several parameters are required to determine if new materials are safe for biomedical use; the effects of CNx concentration and exposure time were also evaluated.

2. Materials and Methods

2.1. Synthesis, Purification, and Characterization of CNx

In this way, CNx were synthesized by using the chemical vapor deposition (CVD) method. As a chemical precursor 2.5 wt% ferrocene in benzylamine was used; the solution was placed into a reservoir and atomized. The aerosol was carried by an Argon flow at 2.5 L/min into a quartz tube 100 cm in length, placed inside of two tubular furnaces heated at 850°C. After 30 min of synthesis, the quartz tube was then cooled at room temperature and the CNx were collected by internal scraping. Then, the pristine CNx were purified and dispersed by using a pulsed probe sonicator in water under reflux, followed by a reflux in 6 M HCl and filtration.

Consequently, the purified CNx were analyzed by scanning electron microscopy (SEM) as follows: first, the nanomaterials were pounded into a whole powder and separated into equal portions. Then, each portion was loaded into pins and visualized by SEM (Philips-XL 30 SFEG; Dual Beam (FIB/SEM) FEI-Helios Nanolab 600 equipped with an EDX detector) to determine lengths, diameters, and chemical composition of CNx. Raman characterization was performed using a laser of 633 nm in Raman Renishaw Micro-Raman equipment.

2.2. Preparation of Dispersion of Purified CNx

Stocks of purified CNx were dispersed in distillated water at 1 mg/mL. Then, the samples were sonicated by an ultrasonic bath at 42 kHz and 100 W (Branson 2510 Ultrasonic Cleaner), at 40°C for 8 h, having as a result stable dispersions; these conditions were strong enough to obtain no visible agglomerates of purified CNx. Finally, all the stocks were stored at 4°C until further use.

2.3. Cytotoxicity Assays

The effects of purified CNx on the viability of NIH-3T3 murine fibroblast were evaluated by using the Trypan-blue exclusion method. Briefly, the cells were defrosted and cultured in a basal-IMDM (Iscove’s Modified Dulbecco’s Medium, SIGMA) at pH 7.2 and supplemented with 10% fetal bovine serum (GIBCO), 100 U/mL penicillin, 100 μg/mL streptomycin, and 0.25 μg/mL amphotericin B (SIGMA), using 24-well plates (Corning) over a period of 72 h in a humidity chamber at 37°C and CO2 5% (Shell-Lab). After three passes, when 80% of cellular confluence was reached, the cells were washed twice gently with PBS (pH 7.2) and then harvested by incubation with trypsin-EDTA (0.25%-1X, GIBCO) for 10 min. Cell suspensions with a density of 2 × 103 cells per well were added into 96-well plates in absence or presence of purified CNx at final concentrations of 7 and 70 μg/mL. For exposure route experiments, purified CNx were added (1) immediately after cell dissociation (stressed cells) or (2) fibroblasts were firstly incubated for 24 h and then purified CNx were added into each well (nonstressed cells). Samples were washed twice gently with PBS, incubated with trypsin for 6 min, and cells were counted by using the Trypan-blue method. During the 96 h of exposure with the nanomaterial, samples were taken each 24 h. NIH-3T3 cell cultures without nanomaterials were used as control. To avoid variation on purified CNx concentration in cell cultures when medium was changed, kinetics were carried out using a working volume of 250 µL without medium replacement.

2.4. Statistics

The data is presented as the mean ± standard deviation, with a statistical comparison of one- and two-way ANOVA. We used Dunnett’s posttests to compare treatments with control groups, and values <0.05 were considered significant. All experiments were done in triplicate.

3. Results and Discussion

3.1. Determination of Length and Diameter of Purified CNx

Figure 1 shows SEM micrographs and size distribution of purified CNx. Micrographs by the XL30 and Helios are shown in Figures 1(a) and 1(c), respectively. A few bundles were found in purified CNx samples to determine the lengths of nanomaterials (Figure 1(a)). The length range was 10 to 130 µm, being the most abundant lengths of 40–50 µm (Figure 1(b)). In micrographs of purified CNx, the nanomaterials seemed to have similar diameters (Figure 1(c)), but, after an analysis with the Helios microscopy, the diameter sizes were around 10–80 nm, with a diameter predominance of 20–40 nm (Figure 1(d)).

Nanomaterial sizes were obtained from their own bundles and short “fibrous” structures (Figure 2). The number of bundles among different samples of purified CNx was small (around 38 bundles) (Figures 2(a)2(c), yellow arrows), as well as in between the same sample (Figures 2(d)2(f)). Since purified CNx samples were mostly agglomerated in big structures by dry process, the determination of their total lengths was difficult; thus only the bundles lengths were reported. The presence of amino groups in the CNx could be the reason to find less bundles in samples, due to their weaker van der Waals interactions, resulting in lower formation of bundles [8].

3.2. Raman and EDX Characterization of Pristine and Purified CNx

Figure 3 shows the Raman spectra of pristine and purified CNx plotted between 100 and 3000 cm−1. The bands D (defect mode), G (graphite mode), and G′ (second order mode) situated at 1340, 1592, and 2686 cm−1, respectively, are the typical peaks corresponding to carbonaceous materials. In the case of purified CNx, the shifting to higher frequencies of G band suggests that nitrogen doping decreased. The values were 1.1488 and 1.2815 for pristine and purified CNx, respectively. This increasing in ratio has been suggested as an evidence for sidewall functionalization of CNTs [18, 19].

With respect to chemical composition of our CNx (pristine and purified), EDX analysis was carried out. Figures S1 and S2 (see Supplementary Material available online at http://dx.doi.org/10.1155/2015/801606) show the SEM images and their respective EDX graphs. The average quantity of iron in pristine samples was of 2.22 wt% (Figure S1), which decreased after purification process to 0.61 wt% (Figure S2), indicating the elimination of this contaminant (see Supplementary Material for the quantities of carbon, nitrogen, and oxygen elements).

3.3. Effect of Purified CNx on Murine Fibroblasts Nonstressed and Stressed

3T3 murine fibroblasts were used as a model for stromal cells, which can be found in matrix and connective tissue throughout the body. Figure 4 shows the kinetics of fibroblasts growth with water-dispersed purified CNx. When the cells were incubated simultaneously with the purified CNx, the control culture (without purified CNx) reached a maximum live cell concentration of 1.3 × 105  ±  3.4 × 104 cells per well at 96 h; meanwhile cultures with 7 µg/mL reached 2.6 × 104  ±  5.3 × 103 cells, and a drastic no cell survival was at 70 µg/mL of purified CNx (Figure 4(a)). Purified CNx added after 24 h of cell dissociation showed a decreased live cell, with a cell concentration of 9.6 × 104  ±  9 × 103 for 7 µg/mL and 5.5 × 104  ±  9.5 × 103 for 70 µg/mL, compared to control culture with 1.1 × 106  ±  1.5 × 104, at 96 h of exposure (Figure 4(b)). Results suggest that nanomaterials exhibited toxic effects, in concentration and exposure route-dependent manner. No effects concerning time exposure were observed.

Murine fibroblasts were susceptible to purified CNx in concentration and exposure route-dependent manner. As previously mentioned, toxicity/biocompatibility of CNTs (SWCNT, MWCNT, and functionalized CNTs) on mammalian cells depends on different factors [6, 20, 21]. A lot of data research has shown the toxicity of CNTs [2225] in human mesenchymal stem cells [26], 3T3 L1 fibroblasts [27], 3T3 fibroblast, telomerase, immortalized human bronchiolar epithelial cells, RAW 264.7 macrophages [6], mouse fibroblast cell L929, and mouse adipose-derived stem cells [3], but, to our knowledge, no experiments about the effects of CNTs have been reported on mammalian cells stressed by enzymatic detachment, which is a natural process in the body. Treatments with water-dispersed purified CNx immediately added after cell liftoff (stressed cells by trypsin) were more toxic than purified CNx added after 24 h of cell dissociation, suggesting that exposure route factor had negative effects on cell proliferation. This could have been caused by interaction of CNTs with residual fragments of adhesion proteins generated after cell trypsinization [12], which can still adversely affect cytoskeleton proteins that are involved in regulating cell adhesion, stability, and elasticity [1416, 28]. However, in this work, only the cell proliferation was evaluated as a first approach to determine the purified CNx toxicity; therefore, more studies are required and are currently underway.

Specific growth rate (µ) was calculated from exponential growth phase of fibroblasts and used as a parameter to evaluate the effect of purified CNx on cell growth. In 7 µg/mL of dispersed-water purified CNx incubated simultaneously with cells, µ was lower (0.031 ± 0.004 h−1) than control cultivation (0.048 ± 0.004 h−1); since no cells survived at 70 µg/mL, µ was not determined. Concerning incubation of cells with nanomaterials for 24 h after cell liftoff, values of µ were 0.044 ± 0.002 h−1 for control culture and 0.037 ± 0.002 h−1 and 0.035 ± 0.001 h−1 for purified CNx at 7 and 70 µg/mL, respectively; both concentrations affected negatively µ. Results confirm a cytotoxic effect that is concentration and exposure route dependent.

3.4. Morphology Diversity of Pristine CNx

In several investigations about cytotoxic effects of CNTs, these nanomaterials are purchased from companies, which are synthesized by CVD. However, researchers have reported different patterns in the bulk growth of CNTs during their synthesis, showing that the CVD method produces a wide range of CNTs morphologies with varieties of lengths and diameters [29]. In preliminary results, different fractions from the same batch were analyzed by SEM showing a wide collection of pristine CNx sizes (Figure 5), with lengths of range between 30 and 250 µm (Figure 5(a)) and diameters of 24–60 nm (Figure 5(b)). Figure 6 shows a SEM micrograph gallery of the morphology of the different pristine CNx fractions, where the variations in lengths (Figures 6(a)6(c)) and diameters (Figures 6(d)6(f)) among three fractions are clear. The morphology variation of CNTs could be the reason behind having contradictory results regarding cytotoxicity/biocompatibility of CNTs reported in several researches, and this issue should be studied in order to understand the relationship between CNTs and mammalian cell response.

4. Conclusion

Finally, purified CNx have a cytotoxicity effect that is directly dependent on their concentration; also purified CNx showed a more toxic effect in enzymatic stressed cells than in the nonstressed ones. Since cells in the body are exposed to enzymatic processes of detachment, the present study of the effects on overstressed cells by enzymatic digestion is important for the development and potential uses of these nanomaterials in the biomedical field. On the other hand, chemical synthesis of pristine CNx yields heterogenic product with substantial differences on length and diameter size, which have distinctive cytotoxic effects on the proliferation of NIH-3T3 cells. There is still a long path that we must take in order to understand the relationship between nanomaterials and mammalian cells. However, concentrations up to 7 µg/mL of nanotubes are well tolerated by the cells, and they could be used in biomedical applications.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.


The authors thank the Marcos Moshinsky Foundation and CONACYT-Mexico Grants 107082 (JFRM), CB-2013-220744 (EMS) for partial funding. J. Gil Munguía-López thanks CONACYT for its Scholarship 250279. They also thank L. Aldana for the English review and L. Ordoñez, B. A. Rivera-Escoto, and G. J. Labrada-Delgado for their technical assistance.

Supplementary Materials

Supplementary material contains SEM images and EDX analysis in different parts of pristine and purified CNx used in this work.

  1. Supplementary Material