The response effect of pheochromocytoma (PC12) cell lines to oxidized multi-walled carbon nanotubes (o-MWCMTs)

Background: The applications of oxidized carbon nanotubes (o-CNTs) have shown potentials in novel drug delivery including the brain which is usually a challenge. This underscores the importance to study its potential toxic effect in animals. Despite being a promising tool for biomedical applications little is known about the safety of drugs in treating brain diseases. The toxicity of oxidized multi-walled carbon nanotubes (o-MWCNTs) are of utmost concern and in most in-vitro studies conducted so far are on dendritic cell (DC) lines with limited data on PC12 cell lines. Objectives: We focused on the effect of o-MWCNTs in PC12 cells in vitro: a common model cell for neurotoxicity. Methods: The pristine multi-walled carbon nanotubes (p-MWCNTs) were produced by the swirled floating catalytic chemical vapour deposition method (SFCCVD). The p-MWCNTs were then oxidized using purified H 2 SO 4 /HNO 3 (3:1v/v) and 30% HNO 3 acids to produce o-MWCNTs. The Brunauer-Emmett-Teller (BET), transmission electron microscopy (TEM), Scanning electron microscopy (SEM), thermogravimetric analyser (TGA) and Raman spectroscopy techniques were used to characterize the MWCNTs. The PC12 cells were cultured in RPMI medium containing concentrations of o-MWCNTs ranging from 50 to 200 μg/ml. Results: The o-MWCNTs demonstrated slight cytotoxicity at short time period to PC12 neuronal cells whilst at longer time period, no significant (p > 0.05) toxicity was observed due to cell recovery. Conclusion: In conclusion, the o-MWCNTs did not affect the growth rate and viability of the PC12 cells due to lack of considerable toxicity in the cells during the observed time period but further investigations are required to determine cell recovery mechanism.


Introduction
Carbon nanotubes (CNTs) are molecular-scale tubes of graphitic carbon that possess excellent mechanical, magnetic and electrical properties due to their unique topology and structure. 1,2 . The versatility has attracted intense research in biomedical applications such as drug delivery 3,4 , medical imaging 5 , cancer treatment 6,7 and the treatment of brain diseases 8,9 . However, toxicological information on functionalized carbon-containing particles is limited and published data on the toxic effects of CNTs are often contradictory 10,11,12 . The toxicity studies of both p-MWCNTs (un-functionalized carbon nanotubes) and o-MWCNTs (functionalized carbon nanotubes versatile for drug and other bioconjugates attachment) conducted so far on pheochromocytoma (PC12) cells are limited. Most of the studies are done in dendritic cells with very limited studies on PC12 cells. The PC12 cells are derived from rat adrenal medullary tumor and are widely used to study responses of differentiated neuronal cells 13 . The aim of this study is to study the effect on o-MWCNTs in PC12 cells in vitro: a common model cell for neurotoxicity. Reports by Xu et al 14 have showed that carboxyl-terminated MWCNTs can suppress potassium channel activities in PC12 cells in a time dependent and irreversible manner. Other reports by Wang et al 15 found a decrease in cell viability of PC12 cells due to single walled CNTs (SWCNTs) expressing oxidative stress. Using other cell lines, Zhang et al 16 showed that acid  treated MWCNTs and SWCNTs have an increased  cytotoxic effect in human cervical carcinoma HeLa  cells when compared to the untreated CNTs (pristine  CNTs). Also Bottini and colleagues 17 compared the toxicity of pristine MWCNTs to o-MWCNTs on human T lymphocyte cells. They found that o-MWCNTs were more toxic than p-MWCNTs, with the formal inducing loss of cell viability through apoptosis at dosses of 400 µg/ml.

Chemicals and reagents
The ferrocene, sulphuric and nitric acid, methanol, diethyl ether and ethanol were obtained from Merck Chemicals (Pty) Ltd, South Africa. The argon and acetylene gases were obtained from Afrox Ltd South Africa. The RPMI medium, fetal bovine serum, and horse serum (Sigma-Aldrich, St Louis, MO, USA) and other reagents used were all of analytical grade.

Production of carbon nanotubes
The modification of the chemical vapour deposition (CVD), termed the swirled floating catalyst (SFCCVD) 18 was used to produce the p-MWCNTs. It consists of a vertical silica plug flow reactor that is immersed in a furnace connected to a temperature regulator with valves, rotameter and a pressure controller. The CNTs were produced by the catalytic decomposition of acetylene in argon gas flow using ferrocene as the catalyst. The production was carried out at a reactor temperature range of 700 -900 ºC and ferrocene heated at 150ÚC for 1 h reaction time using acetylene with flow rate of 100 ml/min. The transmission electron microscopy(TEM, JEOL JEM-100S), scanning electron microscopy (SEM), Brunauer-Emmett-Teller (BET, Micromeritics Tristar), Raman spectroscopy, and Thermogravimetric analysis (TGA, Perkin Elmer TGA 4000 Thermogravimetric Analyser) were used to characterise the synthesized CNTs.
Functionalization of carbon nanotubes. The pristine MWCNTs (p-MWCNTs) were oxidised using H2SO 4 /HNO3 in a 3:1 (v/v) and 30% HNO 3 acids to create the carboxylic groups (-COOH) by sonication. The solution was then filtered and washed several times with distilled water to remove residual acids (pH neutal) and then dried at 50 o C in an oven. The oxidation procedure modifies the p-MWCNTs surfaces with carboxylic groups resulting into MWCNT-COOH (figure. 1), which are required for further coupling 4 . The o-MWCNTs were then sterilely (filtration) and serially (10-fold dilution) diluted in RPMI medium in a biosafety cabinet and sonicated as stock solution.
PC12 neuronal cell culture PC12 cells (obtained from the Department of Pharmacy and Pharmacology, University of the Witwatersrand, South Africa) were maintained in RPM1 medium supplemented with 10% v/v donor horse serum, 5% v/v fetal bovine serum and 1% v/ v PenStrap (penicillin and streptomycin). All cells were cultured in plastic culture flasks at 37ÚC in a 5% CO 2 humidified incubator (HERA Cell, Thermo Electron Corporation).

Cell-viability
A CytoTox-Glo assay was used to measure cell viability. All samples were sterilized for 24 h under UV irradiation. The PC12 cells were seeded in each well containing 0, 50, 100 and 200 µg/ml of o-MWCNTs. They were then incubated at 0, 2, 4, 22 and 24 hr.

Cytotoxicity of carbon nanotubes towards PC-12 neuronal cells
To determine the cytotoxicity of CNTs in PC-12 neuronal cells, the CytoTox-Glo™ Promega assay (Madison, USA) was used according to the manufacturer's instructions (Niles et al 2007). In a 96-well plate, cells were prepared by adding 12µl of PC12 cell culture to each well in triplicates. Twelve microliter of freshly prepared sterile media containing CNTs (containing 2, 50, 100 and 200 µg/ml of o-MWCNTs) were then added to the 96well plates containing cells. The cells were then added and cultured for not more than 24 hr, as recommended by the manufacturer. After 2, 4, 22 and 24hr of incubation 12.5µl of CytoTox-Glo Cytotoxicity Reagent was added to all wells. The 96-well plate was then taken to the Victor X3 and mixed by orbital shaking for about 30 seconds and incubated at room temperature for 15 min. The luminescence was then measured onVictor X3 (PerkinElmer) and recorded (representing signal obtained from dead cells). 12.5µl of lysis reagent (Assay kit) was then added to each and mixed again by orbital shaking for 30 seconds and incubated at room temperature for 15 minutes. The Luminescence was recorded again on Victor X3 (Signal obtained from total cell population). This procedure was repeated at 4, 22 and 24 hr incubation, to obtain the luminescent contribution to viable cells.

Morphometric analysis of PC12 cells
Cells were placed onto a glass slide using a micropipette and covered with a cover slip before viewing. The morphology of PC12 cells after treatment was studied under a bright field microscope (Olympus Optical CO. Ltd, Tokyo, Japan).

Statistical Analyses
Statistical analyses were performed using the analysis of variance (One-Way ANOVA) to compare between the triplicates and Student T-test to compared between p-MWCNTs and o-MWCNTs using the Origin 6.0 Professional software and was considered statistically significant p <0.05. All the experiments were done in triplicates and data shown as mean ± SD.

Ethics Clearance
The proposed study was submitted for ethical consideration and approval by the University of the Witwatersrand Research Ethics Committee.   The Brunauer-Emmett-Teller (BET) analysis of the various MWCNT are shown in table 1. The surface areas (SA) ranged from 29.5-36.0m 2 /g, average pore volume (0.1-0.16 cm 2 /g) and average pore diameter of pristine (13.4-20.6nm).

Results
The Raman spectroscopy analysis of the pristine, 30% HNO 3 and 1:3 HNO 3 :H 2 SO 4 treated MWCNTs are shown in table 2. The band positions were found to range from 1321 cm -1 -1327 cm -1 with ID/IG ration of 30% HNO 3 treated MWCNTs at 1.01 while that of pristine 1.03 and 1:3 HNO 3 :H 2 SO 4 at 1.39. The TGA analysis was employed to determine the quality 19 and chemical stability of MWCNTs and to estimate the amount of residual catalyst present in the sample before and after oxidation/purification 20 . The TGA curves of p-MWCNTs, 30% HNO 3 and 1:3 HNO 3 :H 2 SO 4 acid treated MWCNTs are represented in Figure. 4. The p-MWCNTs were found to decompose easily and reported the highest value of metal catalyst (13.20%). The 30% HNO 3 oxidation showed 9.70% while 1:3 HNO 3 :H 2 SO 4 reported catalyst impurities of 11.67%. Thus the 30% HNO 3 treatment removed more metal catalysts than the 1:3 HNO 3 :H 2 SO 4 . Thermal stability of MWCNTs depend on the side wall ''defects'' and the amount of metallic impurities. The undifferentiated PC12 cells were used to evaluate the cell viability of 30% HNO 3 as shown in Figure  5. The 30% HNO 3 o-MWCNTs was used because of the purity when compared to p-MWCNTs and 1:3 HNO 3 :H 2 SO 4 acid treated. There was a no significant difference between the p-MWCNTs and o-MWCNTs (p<0.05).
In this study the Promega assay was utilized to determine the effect of various concentrations of o-MWCNTs on the viability of cultured PC12 cells as shown in figure 5. After 2 hours of incubation, the 50 µg/ml o-MWCNTs resulted in no loss in cell viability whilst after 4hr of incubation the percentage of viable cells was reduced to 79% as compared to the control group (incubated with no MWCNTs).The 100 µg/ml and 200 µg/ml o-MWCNT showed a reduction in cell viability from 84 and 61% respectively in 2 hours when compared to the control group. After 4 hours the reduction in viability was observed in all the 3 concentrations (50, 100 and 200 µg/ml). The cells were found to recover at longer period times of incubation. At 22 hours, 87 % of the cells in 50 to 200 µg/ml o-MWCNT concentrations recovered when compared to the control groups. The recovery was found to increase up to 24 hours of incubation with no loss in cell viability. The effect of high concentrations of o-MWCNTs on the viability of PC12 cells was found suppressive only at earlier periods of incubation and this suppressive effect was concentration dependent. This indicates that at shorter (2 and 4hours) incubation in o-MWCNTs affect PC12 cells while at longer period, regeneration can occurs due to adaptation in the o-MWCNTs.
To examine the morphology of PC12 cells, a bright field microscope was used, figure 6 (a)-(b). Figure 6a shows the cells cultured without MWCNTs while Figure 6b 14 showed that carboxyl-terminated MWCNTs can antagonize the 3 types of potassium channels on undifferentiated PC12 cells. However the potassium channels cell alteration had no significant effect in the generation of reactive oxygen species (ROS) 14 . The ROS may not have been involved in the potassium suppression but the elemental suppression can give a huge impact on neurone electrical impulse transmission and excitation processes 14 .
Since the inhibition of the PC12 cells was found to be dose dependent and at longer incubation period could recover has been observed in similar earlier reports 11,14 . This phenomenon has been reported by Cheng (Figures 6a & 6b) showed that of o-MWCNTs has no effect on PC12 cells. They were found to be internalized in the PC12 cells. The bright field micrographs results were similar to those reported by Raffa et al 28 where o-MWCNTs was found to internalize in PC12 cells. Although this can be influence by the size and length of the o-MWCNTs 28 . However, the mechanism of o-MWCNT penetration into PC12 cells is yet to be elucidated and the more information needed to augment o-MWCNTs PC12 cells toxicity. The induction of PC12 cell viability according to the findings by Meng et al 27 Iron (Fe) impurities embedded in the o-MWCNTs during the production process can be responsible for the PC12 induction. Therefore, the dynamics of o-MWCNTs cellular involvement in PC12 toxicity and safe applications need further studies.

Conclusion
The quality of purified MWCNTs resulting from the 30% HNO 3 acid treatment was higher than that of 1:3 HNO 3 :H 2 SO 4 acid treated MWCNTs. The response of o-MWCNTs in the PC12 cells was both concentration and time dependent. PC12 cells inhibition was observed at shorter incubation period (2-4 hours). However, at long exposure periods (22-24 hours) there was massive cells recovery in the present of o-MWCNT doses. Generally the o-MWCNTs in-vitro toxicity studied in PC12 cells showed very minimal adverse effect but their use in biomedical applications need further safety investigation as well as mechanism of cell recovery at longer periods of exposure.